2 _dnl__ Copyright (c) 1988 1989 1990 1991 Free Software Foundation, Inc.
4 @setfilename _GDBP__.info
6 @c THIS MANUAL REQUIRES TEXINFO-2 macros and info-makers to format properly.
8 @c NOTE: this manual is marked up for preprocessing with a collection
9 @c of m4 macros called "pretex.m4". If you see <_if__> and <_fi__>
10 @c scattered around the source, you have the full source before
11 @c preprocessing; if you don't, you have the source configured for
12 @c _HOST__ architectures (and you can of course get the full source,
13 @c with all configurations, from wherever you got this).
16 THIS IS THE SOURCE PRIOR TO PREPROCESSING. The full source needs to
17 be run through m4 before either tex- or info- formatting: for example,
19 m4 pretex.m4 none.m4 all.m4 gdb.texinfo >gdb-all.texinfo
21 will produce (assuming your path finds either GNU m4 >= 0.84, or SysV
22 m4; Berkeley won't do) a file suitable for formatting. See the text in
23 "pretex.m4" for a fuller explanation (and the macro definitions).
28 \def\$#1${{#1}} % Kluge: collect RCS revision info without $...$
29 \xdef\manvers{\$Revision$} % For use in headers, footers too
33 @c FOR UPDATES LEADING TO THIS DRAFT, GDB CHANGELOG CONSULTED BETWEEN:
34 @c Fri Oct 11 23:27:06 1991 John Gilmore (gnu at cygnus.com)
35 @c Sat Dec 22 02:51:40 1990 John Gilmore (gnu at cygint)
37 This file documents the GNU debugger _GDBN__.
39 Copyright (C) 1988, 1989, 1990, 1991 Free Software Foundation, Inc.
41 Permission is granted to make and distribute verbatim copies of
42 this manual provided the copyright notice and this permission notice
43 are preserved on all copies.
46 Permission is granted to process this file through TeX and print the
47 results, provided the printed document carries copying permission
48 notice identical to this one except for the removal of this paragraph
49 (this paragraph not being relevant to the printed manual).
52 Permission is granted to copy and distribute modified versions of this
53 manual under the conditions for verbatim copying, provided also that the
54 section entitled ``GNU General Public License'' is included exactly as
55 in the original, and provided that the entire resulting derived work is
56 distributed under the terms of a permission notice identical to this
59 Permission is granted to copy and distribute translations of this manual
60 into another language, under the above conditions for modified versions,
61 except that the section entitled ``GNU General Public License'' may be
62 included in a translation approved by the Free Software Foundation
63 instead of in the original English.
66 @setchapternewpage odd
68 @settitle Using _GDBN__ (<v>_GDB_VN__)
71 @settitle Using _GDBN__ <v>_GDB_VN__ (_HOST__)
78 @subtitle{A Guide to the GNU Source-Level Debugger}
80 @subtitle{On _HOST__ Systems}
83 @c Maybe crank this up to "Fourth Edition" when released at FSF
84 @c @subtitle Third Edition---_GDBN__ version _GDB_VN__
85 @subtitle _GDBN__ version _GDB_VN__
87 @author{Richard M. Stallman@qquad @hfill Free Software Foundation}
88 @author{Roland H. Pesch@qquad @hfill Cygnus Support}
92 \hfill rms\@ai.mit.edu, pesch\@cygnus.com\par
93 \hfill {\it Using _GDBN__}, \manvers\par
94 \hfill \TeX{}info \texinfoversion\par
98 @vskip 0pt plus 1filll
99 Copyright @copyright{} 1988, 1989, 1990, 1991 Free Software Foundation, Inc.
101 Permission is granted to make and distribute verbatim copies of
102 this manual provided the copyright notice and this permission notice
103 are preserved on all copies.
105 Permission is granted to copy and distribute modified versions of this
106 manual under the conditions for verbatim copying, provided also that the
107 section entitled ``GNU General Public License'' is included exactly as
108 in the original, and provided that the entire resulting derived work is
109 distributed under the terms of a permission notice identical to this
112 Permission is granted to copy and distribute translations of this manual
113 into another language, under the above conditions for modified versions,
114 except that the section entitled ``GNU General Public License'' may be
115 included in a translation approved by the Free Software Foundation
116 instead of in the original English.
120 @node Top, Summary, (dir), (dir)
122 This file describes version _GDB_VN__ of GDB, the GNU symbolic debugger.
126 * Summary:: Summary of _GDBN__
127 * New Features:: New Features in _GDBN__ version _GDB_VN__
128 * Sample Session:: A Sample _GDBN__ Session
129 * Invocation:: Getting In and Out of _GDBN__
130 * Commands:: _GDBN__ Commands
131 * Running:: Running Programs Under _GDBN__
132 * Stopping:: Stopping and Continuing
133 * Stack:: Examining the Stack
134 * Source:: Examining Source Files
135 * Data:: Examining Data
136 * Languages:: Using _GDBN__ with Different Languages
137 * Symbols:: Examining the Symbol Table
138 * Altering:: Altering Execution
139 * _GDBN__ Files:: _GDBN__'s Files
140 * Targets:: Specifying a Debugging Target
141 * Controlling _GDBN__:: Controlling _GDBN__
142 * Sequences:: Canned Sequences of Commands
143 * Emacs:: Using _GDBN__ under GNU Emacs
144 * _GDBN__ Bugs:: Reporting Bugs in _GDBN__
146 * Installing _GDBN__:: Installing _GDBN__
147 * Copying:: GNU GENERAL PUBLIC LICENSE
150 --- The Detailed Node Listing ---
154 * Free Software:: Free Software
155 * Contributors:: Contributors to _GDBN__
157 Getting In and Out of _GDBN__
159 * Starting _GDBN__:: Starting _GDBN__
160 * Leaving _GDBN__:: Leaving _GDBN__
161 * Shell Commands:: Shell Commands
165 * File Options:: Choosing Files
166 * Mode Options:: Choosing Modes
170 * Command Syntax:: Command Syntax
171 * Help:: Getting Help
173 Running Programs Under _GDBN__
175 * Compilation:: Compiling for Debugging
176 * Starting:: Starting your Program
177 * Arguments:: Your Program's Arguments
178 * Environment:: Your Program's Environment
179 * Working Directory:: Your Program's Working Directory
180 * Input/Output:: Your Program's Input and Output
181 * Attach:: Debugging an Already-Running Process
182 * Kill Process:: Killing the Child Process
184 Stopping and Continuing
186 * Breakpoints:: Breakpoints, Watchpoints, and Exceptions
187 * Continuing and Stepping:: Resuming Execution
190 Breakpoints, Watchpoints, and Exceptions
192 * Set Breaks:: Setting Breakpoints
193 * Set Watchpoints:: Setting Watchpoints
194 * Exception Handling:: Breakpoints and Exceptions
195 * Delete Breaks:: Deleting Breakpoints
196 * Disabling:: Disabling Breakpoints
197 * Conditions:: Break Conditions
198 * Break Commands:: Breakpoint Command Lists
199 * Breakpoint Menus:: Breakpoint Menus
200 * Error in Breakpoints:: ``Cannot insert breakpoints''
204 * Frames:: Stack Frames
205 * Backtrace:: Backtraces
206 * Selection:: Selecting a Frame
207 * Frame Info:: Information on a Frame
209 Examining Source Files
211 * List:: Printing Source Lines
212 * Search:: Searching Source Files
213 * Source Path:: Specifying Source Directories
214 * Machine Code:: Source and Machine Code
218 * Expressions:: Expressions
219 * Variables:: Program Variables
220 * Arrays:: Artificial Arrays
221 * Output formats:: Output formats
222 * Memory:: Examining Memory
223 * Auto Display:: Automatic Display
224 * Print Settings:: Print Settings
225 * Value History:: Value History
226 * Convenience Vars:: Convenience Variables
227 * Registers:: Registers
228 * Floating Point Hardware:: Floating Point Hardware
230 Using GDB with Different Languages
232 * Setting:: Switching between source languages
233 * Show:: Displaying the language
234 * Checks:: Type and Range checks
235 * Support:: Supported languages
237 Switching between source languages
239 * Manually:: Setting the working language manually
240 * Automatically:: Having GDB infer the source language
242 Type and range Checking
244 * Type Checking:: An overview of type checking
245 * Range Checking:: An overview of range checking
250 * Modula-2:: Modula-2
254 * C Operators:: C and C++ Operators
255 * C Constants:: C and C++ Constants
256 * Cplusplus expressions:: C++ Expressions
257 * C Defaults:: Default settings for C and C++
258 * C Checks:: C and C++ Type and Range Checks
259 * Debugging C:: _GDBN__ and C
260 * Debugging C plus plus:: Special features for C++
264 * M2 Operators:: Built-in operators
265 * Builtin Func/Proc:: Built-in Functions and Procedures
266 * M2 Constants:: Modula-2 Constants
267 * M2 Defaults:: Default settings for Modula-2
268 * Deviations:: Deviations from standard Modula-2
269 * M2 Checks:: Modula-2 Type and Range Checks
270 * M2 Scope:: The scope operators @code{::} and @code{.}
271 * GDB/M2:: GDB and Modula-2
275 * Assignment:: Assignment to Variables
276 * Jumping:: Continuing at a Different Address
277 * Signaling:: Giving the Program a Signal
278 * Returning:: Returning from a Function
279 * Calling:: Calling your Program's Functions
280 * Patching:: Patching your Program
284 * Files:: Commands to Specify Files
285 * Symbol Errors:: Errors Reading Symbol Files
287 Specifying a Debugging Target
289 * Active Targets:: Active Targets
290 * Target Commands:: Commands for Managing Targets
291 * Remote:: Remote Debugging
295 * i960-Nindy Remote:: _GDBN__ with a Remote i960 (Nindy)
296 * EB29K Remote:: _GDBN__ with a Remote EB29K
297 * VxWorks Remote:: _GDBN__ and VxWorks
299 _GDBN__ with a Remote i960 (Nindy)
301 * Nindy Startup:: Startup with Nindy
302 * Nindy Options:: Options for Nindy
303 * Nindy reset:: Nindy Reset Command
305 _GDBN__ with a Remote EB29K
307 * Comms (EB29K):: Communications Setup
308 * gdb-EB29K:: EB29K cross-debugging
309 * Remote Log:: Remote Log
313 * VxWorks connection:: Connecting to VxWorks
314 * VxWorks download:: VxWorks Download
315 * VxWorks attach:: Running Tasks
320 * Editing:: Command Editing
321 * History:: Command History
322 * Screen Size:: Screen Size
324 * Messages/Warnings:: Optional Warnings and Messages
326 Canned Sequences of Commands
328 * Define:: User-Defined Commands
329 * Command Files:: Command Files
330 * Output:: Commands for Controlled Output
332 Reporting Bugs in _GDBN__
334 * Bug Criteria:: Have You Found a Bug?
335 * Bug Reporting:: How to Report Bugs
339 * Subdirectories:: Configuration subdirectories
340 * Config Names:: Specifying names for hosts and targets
341 * configure Options:: Summary of options for configure
342 * Formatting Documentation:: How to format and print GDB documentation
345 @node Summary, New Features, Top, Top
346 @unnumbered Summary of _GDBN__
348 The purpose of a debugger such as _GDBN__ is to allow you to see what is
349 going on ``inside'' another program while it executes---or what another
350 program was doing at the moment it crashed.
352 _GDBN__ can do four main kinds of things (plus other things in support of
353 these) to help you catch bugs in the act:
357 Start your program, specifying anything that might affect its behavior.
360 Make your program stop on specified conditions.
363 Examine what has happened, when your program has stopped.
366 Change things in your program, so you can experiment with correcting the
367 effects of one bug and go on to learn about another.
370 You can use _GDBN__ to debug programs written in C, C++, and Modula-2.
371 Fortran support will be added when a GNU Fortran compiler is ready.
374 * Free Software:: Free Software
375 * Contributors:: Contributors to GDB
378 @node Free Software, Contributors, Summary, Summary
379 @unnumberedsec Free Software
380 _GDBN__ is @dfn{free software}, protected by the GNU General Public License (GPL).
381 The GPL gives you the freedom to copy or adapt a licensed
382 program---but every person getting a copy also gets with it the
383 freedom to modify that copy (which means that they must get access to
384 the source code), and the freedom to distribute further copies.
385 Typical software companies use copyrights to limit your freedoms; the
386 Free Software Foundation uses the GPL to preserve these freedoms.
388 Fundamentally, the General Public License is a license which says that
389 you have these freedoms and that you can't take these freedoms away
392 @c FIXME: (passim) go through all xrefs, expanding to use text headings
393 For full details, @pxref{Copying}.
394 @node Contributors, , Free Software, Summary
395 @unnumberedsec Contributors to GDB
397 Richard Stallman was the original author of GDB, and of many other GNU
398 programs. Many others have contributed to its development. This
399 section attempts to credit major contributors. One of the virtues of
400 free software is that everyone is free to contribute to it; with
401 regret, we cannot actually acknowledge everyone here. The file
402 @file{ChangeLog} in the GDB distribution approximates a blow-by-blow
405 Changes much prior to version 2.0 are lost in the mists of time.
408 @emph{Plea:} Additions to this section are particularly welcome. If you
409 or your friends (or enemies; let's be evenhanded) have been unfairly
410 omitted from this list, we would like to add your names!
413 So that they may not regard their long labor as thankless, we
414 particularly thank those who shepherded GDB through major releases: John
415 Gilmore (releases _GDB_VN__, 4.1, 4.0); Jim Kingdon (releases 3.9, 3.5,
416 3.4, 3.3); and Randy Smith (releases 3.2, 3.1, 3.0). As major
417 maintainer of GDB for some period, each contributed significantly to the
418 structure, stability, and capabilities of the entire debugger.
420 Richard Stallman, assisted at various times by Pete TerMaat, Chris
421 Hanson, and Richard Mlynarik, handled releases through 2.8.
423 Michael Tiemann is the author of most of the GNU C++ support in GDB,
424 with significant additional contributions from Per Bothner. James
425 Clark wrote the GNU C++ demangler. Early work on C++ was by Peter
426 TerMaat (who also did much general update work leading to release 3.0).
428 GDB _GDB_VN__ uses the BFD subroutine library to examine multiple
429 object-file formats; BFD was a joint project of V. Gumby
430 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
432 David Johnson wrote the original COFF support; Pace Willison did
433 the original support for encapsulated COFF.
435 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
436 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
437 support. Jean-Daniel Fekete contributed Sun 386i support. Chris
438 Hanson improved the HP9000 support. Noboyuki Hikichi and Tomoyuki
439 Hasei contributed Sony/News OS 3 support. David Johnson contributed
440 Encore Umax support. Jyrki Kuoppala contributed Altos 3068 support.
441 Keith Packard contributed NS32K support. Doug Rabson contributed
442 Acorn Risc Machine support. Chris Smith contributed Convex support
443 (and Fortran debugging). Jonathan Stone contributed Pyramid support.
444 Michael Tiemann contributed SPARC support. Tim Tucker contributed
445 support for the Gould NP1 and Gould Powernode. Pace Willison
446 contributed Intel 386 support. Jay Vosburgh contributed Symmetry
449 Rich Schaefer and Peter Schauer helped with support of SunOS shared
452 Jay Fenlason and Roland McGrath ensured that GDB and GAS agree about
453 several machine instruction sets.
455 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped
456 develop remote debugging. Intel Corporation and Wind River Systems
457 contributed remote debugging modules for their products.
459 Brian Fox is the author of the readline libraries providing
460 command-line editing and command history.
462 Andrew Beers of SUNY Buffalo wrote the language-switching code and
463 the Modula-2 support, and contributed the Languages chapter of this
466 @node New Features, Sample Session, Summary, Top
467 @unnumbered New Features since _GDBN__ version 3.5
471 Using the new command @code{target}, you can select at runtime whether
472 you are debugging local files, local processes, standalone systems over
473 a serial port, realtime systems over a TCP/IP connection, etc. The
474 command @code{load} can download programs into a remote system. Serial
475 stubs are available for Motorola 680x0 and Intel 80386 remote systems;
476 _GDBN__ also supports debugging realtime processes running under
477 VxWorks, using SunRPC Remote Procedure Calls over TCP/IP to talk to a
478 debugger stub on the target system. Internally, _GDBN__ now uses a
479 function vector to mediate access to different targets; if you need to
480 add your own support for a remote protocol, this makes it much easier.
483 _GDBN__ now sports watchpoints as well as breakpoints. You can use a
484 watchpoint to stop execution whenever the value of an expression
485 changes, without having to predict a particular place in your program
486 where this may happen.
489 Commands that issue wide output now insert newlines at places designed
490 to make the output more readable.
492 @item Object Code Formats
493 _GDBN__ uses a new library called the Binary File Descriptor (BFD)
494 Library to permit it to switch dynamically, without reconfiguration or
495 recompilation, between different object-file formats. Formats currently
496 supported are COFF, a.out, and the Intel 960 b.out; files may be read as
497 .o's, archive libraries, or core dumps. BFD is available as a
498 subroutine library so that other programs may take advantage of it, and
499 the other GNU binary utilities are being converted to use it.
501 @item Configuration and Ports
502 Compile-time configuration (to select a particular architecture and
503 operating system) is much easier. The script @code{configure} now
504 allows you to configure _GDBN__ as either a native debugger or a
505 cross-debugger. @xref{Installing _GDBN__} for details on how to
506 configure and on what architectures are now available.
509 The user interface to _GDBN__'s control variables has been simplified
510 and consolidated in two commands, @code{set} and @code{show}. Output
511 lines are now broken at readable places, rather than overflowing onto
512 the next line. You can suppress output of machine-level addresses,
513 displaying only source language information.
517 _GDBN__ now supports C++ multiple inheritance (if used with a GCC
518 version 2 compiler), and also has limited support for C++ exception
519 handling, with the commands @code{catch} and @code{info catch}: _GDBN__
520 can break when an exception is raised, before the stack is peeled back
521 to the exception handler's context.
524 _GDBN__ now has preliminary support for the GNU Modula-2 compiler,
525 currently under development at the State University of New York at
526 Buffalo. Coordinated development of both _GDBN__ and the GNU Modula-2
527 compiler will continue through the fall of 1991 and into 1992. Other
528 Modula-2 compilers are currently not supported, and attempting to debug
529 programs compiled with them will likely result in an error as the symbol
530 table of the executable is read in.
532 @item Command Rationalization
533 Many _GDBN__ commands have been renamed to make them easier to remember
534 and use. In particular, the subcommands of @code{info} and
535 @code{show}/@code{set} are grouped to make the former refer to the state
536 of your program, and the latter refer to the state of _GDBN__ itself.
537 @xref{Renamed Commands}, for details on what commands were renamed.
539 @item Shared Libraries
540 _GDBN__ _GDB_VN__ can debug programs and core files that use SunOS shared
544 _GDBN__ _GDB_VN__ has a reference card; @xref{Formatting Documentation} for
545 instructions on printing it.
547 @item Work in Progress
548 Kernel debugging for BSD and Mach systems; Tahoe and HPPA architecture
553 @node Sample Session, Invocation, New Features, Top
554 @chapter A Sample _GDBN__ Session
556 You can use this manual at your leisure to read all about _GDBN__.
557 However, a handful of commands are enough to get started using the
558 debugger. This chapter illustrates these commands.
561 In this sample session, we emphasize user input like this: @i{input},
562 to make it easier to pick out from the surrounding output.
565 @c FIXME: this example may not be appropriate for some configs, where
566 @c FIXME...primary interest is in remote use.
568 One of the preliminary versions of GNU @code{m4} (a generic macro
569 processor) exhibits the following bug: sometimes, when we change its
570 quote strings from the default, the commands used to capture one macro's
571 definition in another stop working. In the following short @code{m4}
572 session, we define a macro @code{foo} which expands to @code{0000}; we
573 then use the @code{m4} builtin @code{defn} to define @code{bar} as the
574 same thing. However, when we change the open quote string to
575 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
576 procedure fails to define a new synonym @code{baz}:
585 @i{define(bar,defn(`foo'))}
589 @i{changequote(<QUOTE>,<UNQUOTE>)}
591 @i{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
594 m4: End of input: 0: fatal error: EOF in string
598 Let's use _GDBN__ to try to see what's going on.
602 GDB is free software and you are welcome to distribute copies of it
603 under certain conditions; type "show copying" to see the conditions.
604 There is absolutely no warranty for GDB; type "show warranty" for details.
605 GDB _GDB_VN__, Copyright 1991 Free Software Foundation, Inc...
610 _GDBN__ reads only enough symbol data to know where to find the rest
611 when needed; as a result, the first prompt comes up very quickly. We
612 then tell _GDBN__ to use a narrower display width than usual, so
613 that examples will fit in this manual.
616 (_GDBP__) @i{set width 70}
620 Let's see how the @code{m4} builtin @code{changequote} works.
621 Having looked at the source, we know the relevant subroutine is
622 @code{m4_changequote}, so we set a breakpoint there with _GDBN__'s
623 @code{break} command.
626 (_GDBP__) @i{break m4_changequote}
627 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
631 Using the @code{run} command, we start @code{m4} running under _GDBN__
632 control; as long as control does not reach the @code{m4_changequote}
633 subroutine, the program runs as usual:
637 Starting program: /work/Editorial/gdb/gnu/m4/m4
645 To trigger the breakpoint, we call @code{changequote}. _GDBN__
646 suspends execution of @code{m4}, displaying information about the
647 context where it stops.
650 @i{changequote(<QUOTE>,<UNQUOTE>)}
652 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70) at builtin.c:879
653 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]), argc, 1, 3))
657 Now we use the command @code{n} (@code{next}) to advance execution to
658 the next line of the current function.
662 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1]) : nil,
666 @code{set_quotes} looks like a promising subroutine. We can go into it
667 by using the command @code{s} (@code{step}) instead of @code{next}.
668 @code{step} goes to the next line to be executed in @emph{any}
669 subroutine, so it steps into @code{set_quotes}.
673 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
675 530 if (lquote != def_lquote)
679 The summary display showing the subroutine where @code{m4} is now
680 suspended (and its arguments) is called a stack frame display. We can
681 use the @code{backtrace} command (which can also be spelled @code{bt}),
682 to see where we are in the stack: it displays a stack frame for each
687 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
689 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70) at builtin.c:882
690 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
691 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
693 #4 0x79dc in expand_input () at macro.c:40
694 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
698 Let's step through a few more lines to see what happens. The first two
699 times, we can use @samp{s}; the next two times we use @code{n} to avoid
700 falling into the @code{xstrdup} subroutine.
703 0x3b5c 532 if (rquote != def_rquote)
705 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? def_lquote :\
708 536 rquote = (rq == nil || *rq == '\0') ? def_rquote : xstrdup\
711 538 len_lquote = strlen(rquote);
715 The last line displayed looks a little odd; let's examine the variables
716 @code{lquote} and @code{rquote} to see if they are in fact the new left
717 and right quotes we specified. We can use the command @code{p}
718 (@code{print}) to see their values.
721 (_GDBP__) @i{p lquote}
722 $1 = 0x35d40 "<QUOTE>"
723 (_GDBP__) @i{p rquote}
724 $2 = 0x35d50 "<UNQUOTE>"
728 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
729 Let's look at some context; we can display ten lines of source
730 surrounding the current line, with the @code{l} (@code{list}) command.
736 535 lquote = (lq == nil || *lq == '\0') ? def_lquote : xstrdup\
738 536 rquote = (rq == nil || *rq == '\0') ? def_rquote : xstrdup\
741 538 len_lquote = strlen(rquote);
742 539 len_rquote = strlen(lquote);
749 Let's step past the two lines that set @code{len_lquote} and
750 @code{len_rquote}, and then examine the values of those variables.
754 539 len_rquote = strlen(lquote);
757 (_GDBP__) @i{p len_lquote}
759 (_GDBP__) @i{p len_rquote}
764 That certainly looks wrong, assuming @code{len_lquote} and
765 @code{len_rquote} are meant to be the lengths of @code{lquote} and
766 @code{rquote} respectively. Let's try setting them to better values.
767 We can use the @code{p} command for this, since it'll print the value of
768 any expression---and that expression can include subroutine calls and
772 (_GDBP__) p len_lquote=strlen(lquote)
774 (_GDBP__) p len_rquote=strlen(rquote)
779 Let's see if that fixes the problem of using the new quotes with the
780 @code{m4} built-in @code{defn}. We can allow @code{m4} to continue
781 executing with the @code{c} (@code{continue}) command, and then try the
782 example that caused trouble initially:
788 @i{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
795 Success! The new quotes now work just as well as the default ones. The
796 problem seems to have been just the two typos defining the wrong
797 lengths. We'll let @code{m4} exit by giving it an EOF as input.
801 Program exited normally.
805 The message @samp{Program exited normally.} is from _GDBN__; it
806 indicates @code{m4} has finished executing. We can end our _GDBN__
807 session with the _GDBN__ @code{quit} command.
811 _1__@end smallexample
813 @node Invocation, Commands, Sample Session, Top
814 @chapter Getting In and Out of _GDBN__
817 * Starting _GDBN__:: Starting _GDBN__
818 * Leaving _GDBN__:: Leaving _GDBN__
819 * Shell Commands:: Shell Commands
822 @node Starting _GDBN__, Leaving _GDBN__, Invocation, Invocation
823 @section Starting _GDBN__
825 _GDBN__ is invoked with the shell command @code{_GDBP__}. Once started,
826 it reads commands from the terminal until you tell it to exit.
828 You can run @code{_GDBP__} with no arguments or options; but the most
829 usual way to start _GDBN__ is with one argument or two, specifying an
830 executable program as the argument:
835 You can also start with both an executable program and a core file specified:
840 You can, instead, specify a process ID as a second argument, if you want
841 to debug a running process:
846 would attach _GDBN__ to process @code{1234} (unless you also have a file
847 named @file{1234}; _GDBN__ does check for a core file first).
850 You can further control how _GDBN__ starts up by using command-line
851 options. _GDBN__ itself can remind you of the options available:
856 will display all available options and briefly describe their use
857 (@samp{_GDBP__ -h} is a shorter equivalent).
859 All options and command line arguments you give are processed
860 in sequential order. The order makes a difference when the
861 @samp{-x} option is used.
864 * File Options:: Choosing Files
865 * Mode Options:: Choosing Modes
867 _include__(gdbinv-m.m4)_dnl__
871 @node File Options, Mode Options, Starting _GDBN__, Starting _GDBN__
872 @subsection Choosing Files
874 As shown above, any arguments other than options specify an executable
875 file and core file; that is, the first argument encountered with no
876 associated option flag is equivalent to a @samp{-se} option, and the
877 second, if any, is equivalent to a @samp{-c} option. Many options have
878 both long and short forms; both are shown here. The long forms are also
879 recognized if you truncate them, so long as enough of the option is
880 present to be unambiguous. (If you prefer, you can flag option
881 arguments with @samp{+} rather than @samp{-}, though we illustrate the
882 more usual convention.)
885 @item -symbols=@var{file}
887 Read symbol table from file @var{file}.
889 @item -exec=@var{file}
891 Use file @var{file} as the executable file to execute when
892 appropriate, and for examining pure data in conjunction with a core
896 Read symbol table from file @var{file} and use it as the executable
899 @item -core=@var{file}
901 Use file @var{file} as a core dump to examine.
903 @item -command=@var{file}
905 Execute _GDBN__ commands from file @var{file}. @xref{Command Files}.
907 @item -directory=@var{directory}
908 @itemx -d @var{directory}
909 Add @var{directory} to the path to search for source files.
913 @node Mode Options, Mode Options, File Options, Starting _GDBN__
916 @node Mode Options, , File Options, Starting _GDBN__
918 @subsection Choosing Modes
923 Do not execute commands from any @file{_GDBINIT__} initialization files.
924 Normally, the commands in these files are executed after all the
925 command options and arguments have been processed.
926 @xref{Command Files}.
930 ``Quiet''. Do not print the introductory and copyright messages. These
931 messages are also suppressed in batch mode.
934 Run in batch mode. Exit with status @code{0} after processing all the command
935 files specified with @samp{-x} (and @file{_GDBINIT__}, if not inhibited).
936 Exit with nonzero status if an error occurs in executing the _GDBN__
937 commands in the command files.
939 Batch mode may be useful for running _GDBN__ as a filter, for example to
940 download and run a program on another computer; in order to make this
941 more useful, the message
943 Program exited normally.
946 (which is ordinarily issued whenever a program running under _GDBN__ control
947 terminates) is not issued when running in batch mode.
949 @item -cd=@var{directory}
950 Run _GDBN__ using @var{directory} as its working directory,
951 instead of the current directory.
955 Emacs sets this option when it runs _GDBN__ as a subprocess. It tells _GDBN__
956 to output the full file name and line number in a standard,
957 recognizable fashion each time a stack frame is displayed (which
958 includes each time the program stops). This recognizable format looks
959 like two @samp{\032} characters, followed by the file name, line number
960 and character position separated by colons, and a newline. The
961 Emacs-to-_GDBN__ interface program uses the two @samp{\032} characters as
962 a signal to display the source code for the frame.
965 Set the line speed (baud rate or bits per second) of any serial
966 interface used by _GDBN__ for remote debugging.
968 @item -tty=@var{device}
969 Run using @var{device} for your program's standard input and output.
970 @c FIXME: kingdon thinks there's more to -tty. Investigate.
974 _include__(gdbinv-s.m4)
977 @node Leaving _GDBN__, Shell Commands, Starting _GDBN__, Invocation
978 @section Leaving _GDBN__
979 @cindex exiting _GDBN__
984 To exit _GDBN__, use the @code{quit} command (abbreviated @code{q}), or type
985 an end-of-file character (usually @kbd{C-d}).
989 An interrupt (often @kbd{C-c}) will not exit from _GDBN__, but rather
990 will terminate the action of any _GDBN__ command that is in progress and
991 return to _GDBN__ command level. It is safe to type the interrupt
992 character at any time because _GDBN__ does not allow it to take effect
993 until a time when it is safe.
995 If you've been using _GDBN__ to control an attached process or device,
996 you can release it with the @code{detach} command; @pxref{Attach}.
998 @node Shell Commands, , Leaving _GDBN__, Invocation
999 @section Shell Commands
1000 If you need to execute occasional shell commands during your
1001 debugging session, there's no need to leave or suspend _GDBN__; you can
1002 just use the @code{shell} command.
1005 @item shell @var{command string}
1007 @cindex shell escape
1008 Directs _GDBN__ to invoke an inferior shell to execute @var{command
1009 string}. If it exists, the environment variable @code{SHELL} is used
1010 for the name of the shell to run. Otherwise _GDBN__ uses
1014 The utility @code{make} is often needed in development environments.
1015 You don't have to use the @code{shell} command for this purpose in _GDBN__:
1018 @item make @var{make-args}
1020 @cindex calling make
1021 Causes _GDBN__ to execute an inferior @code{make} program with the specified
1022 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1025 @node Commands, Running, Invocation, Top
1026 @chapter _GDBN__ Commands
1029 * Command Syntax:: Command Syntax
1030 * Help:: Getting Help
1033 @node Command Syntax, Help, Commands, Commands
1034 @section Command Syntax
1035 A _GDBN__ command is a single line of input. There is no limit on how long
1036 it can be. It starts with a command name, which is followed by arguments
1037 whose meaning depends on the command name. For example, the command
1038 @code{step} accepts an argument which is the number of times to step,
1039 as in @samp{step 5}. You can also use the @code{step} command with
1040 no arguments. Some command names do not allow any arguments.
1042 @cindex abbreviation
1043 _GDBN__ command names may always be truncated if that abbreviation is
1044 unambiguous. Other possible command abbreviations are listed in the
1045 documentation for individual commands. In some cases, even ambiguous
1046 abbreviations are allowed; for example, @code{s} is specially defined as
1047 equivalent to @code{step} even though there are other commands whose
1048 names start with @code{s}. You can test abbreviations by using them as
1049 arguments to the @code{help} command.
1051 @cindex repeating commands
1053 A blank line as input to _GDBN__ (typing just @key{RET}) means to
1054 repeat the previous command. Certain commands (for example, @code{run})
1055 will not repeat this way; these are commands for which unintentional
1056 repetition might cause trouble and which you are unlikely to want to
1059 The @code{list} and @code{x} commands, when you repeat them with
1060 @key{RET}, construct new arguments rather than repeating
1061 exactly as typed. This permits easy scanning of source or memory.
1063 _GDBN__ can also use @key{RET} in another way: to partition lengthy
1064 output, in a way similar to the common utility @code{more}
1065 (@pxref{Screen Size}). Since it's easy to press one @key{RET} too many
1066 in this situation, _GDBN__ disables command repetition after any command
1067 that generates this sort of display.
1071 A line of input starting with @kbd{#} is a comment; it does nothing.
1072 This is useful mainly in command files (@xref{Command Files}).
1074 @node Help, , Command Syntax, Commands
1075 @section Getting Help
1076 @cindex online documentation
1078 You can always ask _GDBN__ itself for information on its commands, using the
1079 command @code{help}.
1085 You can use @code{help} (abbreviated @code{h}) with no arguments to
1086 display a short list of named classes of commands:
1089 List of classes of commands:
1091 running -- Running the program
1092 stack -- Examining the stack
1093 data -- Examining data
1094 breakpoints -- Making program stop at certain points
1095 files -- Specifying and examining files
1096 status -- Status inquiries
1097 support -- Support facilities
1098 user-defined -- User-defined commands
1099 aliases -- Aliases of other commands
1100 obscure -- Obscure features
1102 Type "help" followed by a class name for a list of commands in that class.
1103 Type "help" followed by command name for full documentation.
1104 Command name abbreviations are allowed if unambiguous.
1108 @item help @var{class}
1109 Using one of the general help classes as an argument, you can get a
1110 list of the individual commands in that class. For example, here is the
1111 help display for the class @code{status}:
1113 (_GDBP__) help status
1118 show -- Generic command for showing things set with "set"
1119 info -- Generic command for printing status
1121 Type "help" followed by command name for full documentation.
1122 Command name abbreviations are allowed if unambiguous.
1126 @item help @var{command}
1127 With a command name as @code{help} argument, _GDBN__ will display a
1128 short paragraph on how to use that command.
1131 In addition to @code{help}, you can use the _GDBN__ commands @code{info}
1132 and @code{show} to inquire about the state of your program, or the state
1133 of _GDBN__ itself. Each command supports many topics of inquiry; this
1134 manual introduces each of them in the appropriate context. The listings
1135 under @code{info} and under @code{show} in the Index point to
1136 all the sub-commands.
1137 @c FIXME: @pxref{Index} used to be here, but even though it shows up in
1138 @c FIXME...the 'aux' file with a pageno the xref can't find it.
1145 This command (abbreviated @code{i}) is for describing the state of your
1146 program; for example, it can list the arguments given to your program
1147 (@code{info args}), the registers currently in use (@code{info
1148 registers}), or the breakpoints you've set (@code{info breakpoints}).
1149 You can get a complete list of the @code{info} sub-commands with
1150 @w{@code{help info}}.
1154 In contrast, @code{show} is for describing the state of _GDBN__ itself.
1155 You can change most of the things you can @code{show}, by using the
1156 related command @code{set}; for example, you can control what number
1157 system is used for displays with @code{set radix}, or simply inquire
1158 which is currently in use with @code{show radix}.
1161 To display all the settable parameters and their current
1162 values, you can use @code{show} with no arguments; you may also use
1163 @code{info set}. Both commands produce the same display.
1164 @c FIXME: "info set" violates the rule that "info" is for state of
1165 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1166 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1170 Here are three miscellaneous @code{show} subcommands, all of which are
1171 exceptional in lacking corresponding @code{set} commands:
1174 @kindex show version
1175 @cindex version number
1177 Show what version of _GDBN__ is running. You should include this
1178 information in _GDBN__ bug-reports. If multiple versions of _GDBN__ are
1179 in use at your site, you may occasionally want to make sure what version
1180 of _GDBN__ you're running; as _GDBN__ evolves, new commands are
1181 introduced, and old ones may wither away. The version number is also
1182 announced when you start _GDBN__ with no arguments.
1184 @kindex show copying
1186 Display information about permission for copying _GDBN__.
1188 @kindex show warranty
1190 Display the GNU ``NO WARRANTY'' statement.
1193 @node Running, Stopping, Commands, Top
1194 @chapter Running Programs Under _GDBN__
1197 * Compilation:: Compiling for Debugging
1198 * Starting:: Starting your Program
1199 * Arguments:: Your Program's Arguments
1200 * Environment:: Your Program's Environment
1201 * Working Directory:: Your Program's Working Directory
1202 * Input/Output:: Your Program's Input and Output
1203 * Attach:: Debugging an Already-Running Process
1204 * Kill Process:: Killing the Child Process
1207 @node Compilation, Starting, Running, Running
1208 @section Compiling for Debugging
1210 In order to debug a program effectively, you need to generate
1211 debugging information when you compile it. This debugging information
1212 is stored in the object file; it describes the data type of each
1213 variable or function and the correspondence between source line numbers
1214 and addresses in the executable code.
1216 To request debugging information, specify the @samp{-g} option when you run
1219 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1220 options together. Using those compilers, you cannot generate optimized
1221 executables containing debugging information.
1223 The GNU C compiler supports @samp{-g} with or without @samp{-O}, making it
1224 possible to debug optimized code. We recommend that you @emph{always} use
1225 @samp{-g} whenever you compile a program. You may think the program is
1226 correct, but there's no sense in pushing your luck.
1228 Some things do not work as well with @samp{-g -O} as with just
1229 @samp{-g}, particularly on machines with instruction scheduling. If in
1230 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1231 please report it as a bug (including a test case!).
1233 Older versions of the GNU C compiler permitted a variant option
1234 @samp{-gg} for debugging information. _GDBN__ no longer supports this
1235 format; if your GNU C compiler has this option, do not use it.
1238 @comment As far as I know, there are no cases in which _GDBN__ will
1239 @comment produce strange output in this case. (but no promises).
1240 If your program includes archives made with the @code{ar} program, and
1241 if the object files used as input to @code{ar} were compiled without the
1242 @samp{-g} option and have names longer than 15 characters, _GDBN__ will get
1243 confused reading the program's symbol table. No error message will be
1244 given, but _GDBN__ may behave strangely. The reason for this problem is a
1245 deficiency in the Unix archive file format, which cannot represent file
1246 names longer than 15 characters.
1248 To avoid this problem, compile the archive members with the @samp{-g}
1249 option or use shorter file names. Alternatively, use a version of GNU
1250 @code{ar} dated more recently than August 1989.
1254 @node Starting, Arguments, Compilation, Running
1255 @section Starting your Program
1262 Use the @code{run} command to start your program under _GDBN__. You
1263 must first specify the program name
1267 with an argument to _GDBN__
1268 (@pxref{Invocation}), or using the @code{file} or @code{exec-file}
1269 command (@pxref{Files}).
1273 @c FIXME explain or avoid "target" here?
1274 On targets that support processes, @code{run} creates an inferior
1275 process and makes that process run your program. On other targets,
1276 @code{run} jumps to the start of the program.
1278 The execution of a program is affected by certain information it
1279 receives from its superior. _GDBN__ provides ways to specify this
1280 information, which you must do @i{before} starting the program. (You
1281 can change it after starting the program, but such changes will only affect
1282 the program the next time you start it.) This information may be
1283 divided into four categories:
1286 @item The @i{arguments.}
1287 You specify the arguments to give your program as the arguments of the
1288 @code{run} command. If a shell is available on your target, the shell
1289 is used to pass the arguments, so that you may use normal conventions
1290 (such as wildcard expansion or variable substitution) in
1291 describing the arguments. In Unix systems, you can control which shell
1292 is used with the @code{SHELL} environment variable. @xref{Arguments}.@refill
1294 @item The @i{environment.}
1295 Your program normally inherits its environment from _GDBN__, but you can
1296 use the _GDBN__ commands @code{set environment} and @code{unset
1297 environment} to change parts of the environment that will be given to
1298 the program. @xref{Environment}.@refill
1300 @item The @i{working directory.}
1301 Your program inherits its working directory from _GDBN__. You can set
1302 _GDBN__'s working directory with the @code{cd} command in _GDBN__.
1303 @xref{Working Directory}.
1305 @item The @i{standard input and output.}
1306 Your program normally uses the same device for standard input and
1307 standard output as _GDBN__ is using. You can redirect input and output
1308 in the @code{run} command line, or you can use the @code{tty} command to
1309 set a different device for your program.
1310 @xref{Input/Output}.
1313 @emph{Warning:} While input and output redirection work, you can't use
1314 pipes to pass the output of the program you're debugging to another
1315 program; if you attempt this, _GDBN__ is likely to wind up debugging the
1319 When you issue the @code{run} command, your program begins to execute
1320 immediately. @xref{Stopping}, for discussion of how to arrange for your
1321 program to stop. Once your program has been started by the @code{run}
1322 command (and then stopped), you may evaluate expressions that involve
1323 calls to functions in the inferior, using the @code{print} or
1324 @code{call} commands. @xref{Data}.
1326 If the modification time of your symbol file has changed since the last
1327 time _GDBN__ read its symbols, _GDBN__ will discard its symbol table and re-read
1328 it. In this process, it tries to retain your current breakpoints.
1330 @node Arguments, Environment, Starting, Running
1331 @section Your Program's Arguments
1333 @cindex arguments (to your program)
1334 The arguments to your program can be specified by the arguments of the
1335 @code{run} command. They are passed to a shell, which expands wildcard
1336 characters and performs redirection of I/O, and thence to the program.
1337 _GDBN__ uses the shell indicated by your environment variable
1338 @code{SHELL} if it exists; otherwise, _GDBN__ uses @code{/bin/sh}.
1340 @code{run} with no arguments uses the same arguments used by the previous
1341 @code{run}, or those set by the @code{set args} command.
1346 Specify the arguments to be used the next time your program is run. If
1347 @code{set args} has no arguments, @code{run} will execute your program
1348 with no arguments. Once you have run your program with arguments,
1349 using @code{set args} before the next @code{run} is the only way to run
1350 it again without arguments.
1354 Show the arguments to give your program when it is started.
1357 @node Environment, Working Directory, Arguments, Running
1358 @section Your Program's Environment
1360 @cindex environment (of your program)
1361 The @dfn{environment} consists of a set of environment variables and
1362 their values. Environment variables conventionally record such things as
1363 your user name, your home directory, your terminal type, and your search
1364 path for programs to run. Usually you set up environment variables with
1365 the shell and they are inherited by all the other programs you run. When
1366 debugging, it can be useful to try running the program with a modified
1367 environment without having to start _GDBN__ over again.
1370 @item path @var{directory}
1372 Add @var{directory} to the front of the @code{PATH} environment variable
1373 (the search path for executables), for both _GDBN__ and your program.
1374 You may specify several directory names, separated by @samp{:} or
1375 whitespace. If @var{directory} is already in the path, it is moved to
1376 the front, so it will be searched sooner.
1378 You can use the string @samp{$cwd} to refer to whatever is the current
1379 working directory at the time _GDBN__ searches the path. If you use
1380 @samp{.} instead, it refers to the directory where you executed the
1381 @code{path} command. _GDBN__ fills in the current path where needed in
1382 the @var{directory} argument, before adding it to the search path.
1383 @c 'path' is explicitly nonrepeatable, but RMS points out it's silly to
1384 @c document that, since repeating it would be a no-op.
1388 Display the list of search paths for executables (the @code{PATH}
1389 environment variable).
1391 @item show environment @r{[}@var{varname}@r{]}
1392 @kindex show environment
1393 Print the value of environment variable @var{varname} to be given to
1394 your program when it starts. If you don't supply @var{varname},
1395 print the names and values of all environment variables to be given to
1396 your program. You can abbreviate @code{environment} as @code{env}.
1398 @item set environment @var{varname} @r{[}=@r{]} @var{value}
1399 @kindex set environment
1400 Sets environment variable @var{varname} to @var{value}. The value
1401 changes for your program only, not for _GDBN__ itself. @var{value} may
1402 be any string; the values of environment variables are just strings, and
1403 any interpretation is supplied by your program itself. The @var{value}
1404 parameter is optional; if it is eliminated, the variable is set to a
1406 @c "any string" here doesn't include leading, trailing
1407 @c blanks. Gnu asks: does anyone care?
1409 For example, this command:
1416 tells a Unix program, when subsequently run, that its user is named
1417 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1418 are not actually required.)
1420 @item unset environment @var{varname}
1421 @kindex unset environment
1422 Remove variable @var{varname} from the environment to be passed to your
1423 program. This is different from @samp{set env @var{varname} =};
1424 @code{unset environment} removes the variable from the environment,
1425 rather than assigning it an empty value.
1428 @node Working Directory, Input/Output, Environment, Running
1429 @section Your Program's Working Directory
1431 @cindex working directory (of your program)
1432 Each time you start your program with @code{run}, it inherits its
1433 working directory from the current working directory of _GDBN__. _GDBN__'s
1434 working directory is initially whatever it inherited from its parent
1435 process (typically the shell), but you can specify a new working
1436 directory in _GDBN__ with the @code{cd} command.
1438 The _GDBN__ working directory also serves as a default for the commands
1439 that specify files for _GDBN__ to operate on. @xref{Files}.
1442 @item cd @var{directory}
1444 Set _GDBN__'s working directory to @var{directory}.
1448 Print _GDBN__'s working directory.
1451 @node Input/Output, Attach, Working Directory, Running
1452 @section Your Program's Input and Output
1457 By default, the program you run under _GDBN__ does input and output to
1458 the same terminal that _GDBN__ uses. _GDBN__ switches the terminal to
1459 its own terminal modes to interact with you, but it records the terminal
1460 modes your program was using and switches back to them when you continue
1461 running your program.
1465 @kindex info terminal
1466 Displays _GDBN__'s recorded information about the terminal modes your
1470 You can redirect the program's input and/or output using shell
1471 redirection with the @code{run} command. For example,
1478 starts the program, diverting its output to the file @file{outfile}.
1481 @cindex controlling terminal
1482 Another way to specify where the program should do input and output is
1483 with the @code{tty} command. This command accepts a file name as
1484 argument, and causes this file to be the default for future @code{run}
1485 commands. It also resets the controlling terminal for the child
1486 process, for future @code{run} commands. For example,
1493 directs that processes started with subsequent @code{run} commands
1494 default to do input and output on the terminal @file{/dev/ttyb} and have
1495 that as their controlling terminal.
1497 An explicit redirection in @code{run} overrides the @code{tty} command's
1498 effect on the input/output device, but not its effect on the controlling
1501 When you use the @code{tty} command or redirect input in the @code{run}
1502 command, only the input @emph{for your program} is affected. The input
1503 for _GDBN__ still comes from your terminal.
1505 @node Attach, Kill Process, Input/Output, Running
1506 @section Debugging an Already-Running Process
1511 @item attach @var{process-id}
1513 attaches to a running process---one that was started outside _GDBN__.
1514 (@code{info files} will show your active targets.) The command takes as
1515 argument a process ID. The usual way to find out the process-id of
1516 a Unix process is with the @code{ps} utility, or with the @samp{jobs -l}
1519 @code{attach} will not repeat if you press @key{RET} a second time after
1520 executing the command.
1523 To use @code{attach}, you must be debugging in an environment which
1524 supports processes. You must also have permission to send the process a
1525 signal, and it must have the same effective user ID as the _GDBN__
1528 When using @code{attach}, you should first use the @code{file} command
1529 to specify the program running in the process and load its symbol table.
1532 The first thing _GDBN__ does after arranging to debug the specified
1533 process is to stop it. You can examine and modify an attached process
1534 with all the _GDBN__ commands that are ordinarily available when you start
1535 processes with @code{run}. You can insert breakpoints; you can step and
1536 continue; you can modify storage. If you would rather the process
1537 continue running, you may use the @code{continue} command after
1538 attaching _GDBN__ to the process.
1543 When you have finished debugging the attached process, you can use the
1544 @code{detach} command to release it from _GDBN__'s control. Detaching
1545 the process continues its execution. After the @code{detach} command,
1546 that process and _GDBN__ become completely independent once more, and you
1547 are ready to @code{attach} another process or start one with @code{run}.
1548 @code{detach} will not repeat if you press @key{RET} again after
1549 executing the command.
1552 If you exit _GDBN__ or use the @code{run} command while you have an attached
1553 process, you kill that process. By default, you will be asked for
1554 confirmation if you try to do either of these things; you can control
1555 whether or not you need to confirm by using the @code{set confirm} command
1556 (@pxref{Messages/Warnings}).
1558 @node Kill Process, , Attach, Running
1560 @section Killing the Child Process
1565 Kill the child process in which your program is running under _GDBN__.
1568 This command is useful if you wish to debug a core dump instead of a
1569 running process. _GDBN__ ignores any core dump file while your program
1573 On some operating systems, a program can't be executed outside _GDBN__
1574 while you have breakpoints set on it inside _GDBN__. You can use the
1575 @code{kill} command in this situation to permit running the program
1576 outside the debugger.
1578 The @code{kill} command is also useful if you wish to recompile and
1579 relink the program, since on many systems it is impossible to modify an
1580 executable file while it is running in a process. In this case, when you
1581 next type @code{run}, _GDBN__ will notice that the file has changed, and
1582 will re-read the symbol table (while trying to preserve your current
1583 breakpoint settings).
1585 @node Stopping, Stack, Running, Top
1586 @chapter Stopping and Continuing
1588 The principal purpose of using a debugger is so that you can stop your
1589 program before it terminates; or so that, if the program runs into
1590 trouble, you can investigate and find out why.
1592 Inside _GDBN__, your program may stop for any of several reasons, such
1593 as a signal, a breakpoint, or reaching a new line after a _GDBN__
1594 command such as @code{step}. You may then examine and change
1595 variables, set new breakpoints or remove old ones, and then continue
1596 execution. Usually, the messages shown by _GDBN__ provide ample
1597 explanation of the status of your program---but you can also explicitly
1598 request this information at any time.
1602 @kindex info program
1603 Display information about the status of your program: whether it is
1604 running or not, what process it is, and why it stopped.
1608 * Breakpoints:: Breakpoints, Watchpoints, and Exceptions
1609 * Continuing and Stepping:: Resuming Execution
1613 @node Breakpoints, Continuing and Stepping, Stopping, Stopping
1614 @section Breakpoints, Watchpoints, and Exceptions
1617 A @dfn{breakpoint} makes your program stop whenever a certain point in
1618 the program is reached. For each breakpoint, you can add various
1619 conditions to control in finer detail whether the program will stop.
1620 You can set breakpoints with the @code{break} command and its variants
1621 (@pxref{Set Breaks}), to specify the place where the program should stop
1622 by line number, function name or exact address in the program. In
1623 languages with exception handling (such as GNU C++), you can also set
1624 breakpoints where an exception is raised (@pxref{Exception Handling}).
1627 A @dfn{watchpoint} is a special breakpoint that stops your program when
1628 the value of an expression changes. You must use a different command to
1629 set watchpoints (@pxref{Set Watchpoints}), but aside from that, you can
1630 manage a watchpoint like any other breakpoint: you enable, disable, and
1631 delete both breakpoints and watchpoints using the same commands.
1633 Each breakpoint or watchpoint is assigned a number when it is created;
1634 these numbers are successive integers starting with one. In many of the
1635 commands for controlling various features of breakpoints you use the
1636 breakpoint number to say which breakpoint you want to change. Each
1637 breakpoint may be @dfn{enabled} or @dfn{disabled}; if disabled, it has
1638 no effect on the program until you enable it again.
1641 * Set Breaks:: Setting Breakpoints
1642 * Set Watchpoints:: Setting Watchpoints
1643 * Exception Handling:: Breakpoints and Exceptions
1644 * Delete Breaks:: Deleting Breakpoints
1645 * Disabling:: Disabling Breakpoints
1646 * Conditions:: Break Conditions
1647 * Break Commands:: Breakpoint Command Lists
1648 * Breakpoint Menus:: Breakpoint Menus
1649 * Error in Breakpoints::
1652 @node Set Breaks, Set Watchpoints, Breakpoints, Breakpoints
1653 @subsection Setting Breakpoints
1655 @c FIXME LMB what does GDB do if no code on line of breakpt?
1656 @c consider in particular declaration with/without initialization.
1658 @c FIXME 2 is there stuff on this already? break at fun start, already init?
1662 Breakpoints are set with the @code{break} command (abbreviated @code{b}).
1664 You have several ways to say where the breakpoint should go.
1667 @item break @var{function}
1668 Set a breakpoint at entry to function @var{function}. When using source
1669 languages that permit overloading of symbols, such as C++,
1670 @var{function} may refer to more than one possible place to break.
1671 @xref{Breakpoint Menus}, for a discussion of that situation.
1673 @item break +@var{offset}
1674 @itemx break -@var{offset}
1675 Set a breakpoint some number of lines forward or back from the position
1676 at which execution stopped in the currently selected frame.
1678 @item break @var{linenum}
1679 Set a breakpoint at line @var{linenum} in the current source file.
1680 That file is the last file whose source text was printed. This
1681 breakpoint will stop the program just before it executes any of the
1684 @item break @var{filename}:@var{linenum}
1685 Set a breakpoint at line @var{linenum} in source file @var{filename}.
1687 @item break @var{filename}:@var{function}
1688 Set a breakpoint at entry to function @var{function} found in file
1689 @var{filename}. Specifying a file name as well as a function name is
1690 superfluous except when multiple files contain similarly named
1693 @item break *@var{address}
1694 Set a breakpoint at address @var{address}. You can use this to set
1695 breakpoints in parts of the program which do not have debugging
1696 information or source files.
1699 When called without any arguments, @code{break} sets a breakpoint at the
1700 next instruction to be executed in the selected stack frame
1701 (@pxref{Stack}). In any selected frame but the innermost, this will
1702 cause the program to stop as soon as control returns to that frame.
1703 This is similar to the effect of a @code{finish} command in the frame
1704 inside the selected frame---except that @code{finish} doesn't leave an
1705 active breakpoint. If you use @code{break} without an argument in the
1706 innermost frame, _GDBN__ will stop the next time it reaches the current
1707 location; this may be useful inside loops.
1709 _GDBN__ normally ignores breakpoints when it resumes execution, until at
1710 least one instruction has been executed. If it did not do this, you
1711 would be unable to proceed past a breakpoint without first disabling the
1712 breakpoint. This rule applies whether or not the breakpoint already
1713 existed when the program stopped.
1715 @item break @dots{} if @var{cond}
1716 Set a breakpoint with condition @var{cond}; evaluate the expression
1717 @var{cond} each time the breakpoint is reached, and stop only if the
1718 value is nonzero---that is, if @var{cond} evaluates as true.
1719 @samp{@dots{}} stands for one of the possible arguments described above
1720 (or no argument) specifying where to break. @xref{Conditions}, for more
1721 information on breakpoint conditions.
1723 @item tbreak @var{args}
1725 Set a breakpoint enabled only for one stop. @var{args} are the
1726 same as for the @code{break} command, and the breakpoint is set in the same
1727 way, but the breakpoint is automatically disabled the first time it
1728 is hit. @xref{Disabling}.
1730 @item rbreak @var{regex}
1732 @cindex regular expression
1733 @c FIXME what kind of regexp?
1734 Set breakpoints on all functions matching the regular expression
1735 @var{regex}. This command
1736 sets an unconditional breakpoint on all matches, printing a list of all
1737 breakpoints it set. Once these breakpoints are set, they are treated
1738 just like the breakpoints set with the @code{break} command. They can
1739 be deleted, disabled, made conditional, etc., in the standard ways.
1741 When debugging C++ programs, @code{rbreak} is useful for setting
1742 breakpoints on overloaded functions that are not members of any special
1745 @kindex info breakpoints
1746 @cindex @code{$_} and @code{info breakpoints}
1747 @item info breakpoints @r{[}@var{n}@r{]}
1748 @item info break @r{[}@var{n}@r{]}
1749 Print a list of all breakpoints (but not watchpoints) set and not
1750 deleted, showing their numbers, where in the program they are, and any
1751 special features in use for them. Disabled breakpoints are included in
1752 the list, but marked as disabled. @code{info break} with a breakpoint
1753 number @var{n} as argument lists only that breakpoint. The convenience
1754 variable @code{$_} and the default examining-address for the @code{x}
1755 command are set to the address of the last breakpoint listed
1756 (@pxref{Memory}). The equivalent command for watchpoints is @code{info
1759 _GDBN__ allows you to set any number of breakpoints at the same place in the
1760 program. There is nothing silly or meaningless about this. When the
1761 breakpoints are conditional, this is even useful (@pxref{Conditions}).
1763 @node Set Watchpoints, Exception Handling, Set Breaks, Breakpoints
1764 @subsection Setting Watchpoints
1765 @cindex setting watchpoints
1766 You can use a watchpoint to stop execution whenever the value of an
1767 expression changes, without having to predict a particular place
1768 where this may happen.
1770 Watchpoints currently execute two orders of magnitude more slowly than
1771 other breakpoints, but this can well be worth it to catch errors where
1772 you have no clue what part of your program is the culprit. Some
1773 processors provide special hardware to support watchpoint evaluation; future
1774 releases of _GDBN__ will use such hardware if it is available.
1778 @item watch @var{expr}
1779 Set a watchpoint for an expression.
1781 @kindex info watchpoints
1782 @item info watchpoints
1783 This command prints a list of watchpoints; it is otherwise similar to
1787 @node Exception Handling, Delete Breaks, Set Watchpoints, Breakpoints
1788 @subsection Breakpoints and Exceptions
1789 @cindex exception handlers
1791 Some languages, such as GNU C++, implement exception handling. You can
1792 use _GDBN__ to examine what caused the program to raise an exception,
1793 and to list the exceptions the program is prepared to handle at a
1794 given point in time.
1797 @item catch @var{exceptions}
1799 You can set breakpoints at active exception handlers by using the
1800 @code{catch} command. @var{exceptions} is a list of names of exceptions
1804 You can use @code{info catch} to list active exception handlers;
1807 There are currently some limitations to exception handling in _GDBN__.
1808 These will be corrected in a future release.
1812 If you call a function interactively, _GDBN__ normally returns
1813 control to you when the function has finished executing. If the call
1814 raises an exception, however, the call may bypass the mechanism that
1815 returns control to the user and cause the program to simply continue
1816 running until it hits a breakpoint, catches a signal that _GDBN__ is
1817 listening for, or exits.
1819 You cannot raise an exception interactively.
1821 You cannot interactively install an exception handler.
1824 @cindex raise exceptions
1825 Sometimes @code{catch} is not the best way to debug exception handling:
1826 if you need to know exactly where an exception is raised, it's better to
1827 stop @emph{before} the exception handler is called, since that way you
1828 can see the stack before any unwinding takes place. If you set a
1829 breakpoint in an exception handler instead, it may not be easy to find
1830 out where the exception was raised.
1832 To stop just before an exception handler is called, you need some
1833 knowledge of the implementation. In the case of GNU C++, exceptions are
1834 raised by calling a library function named @code{__raise_exception}
1835 which has the following ANSI C interface:
1838 /* @var{addr} is where the exception identifier is stored.
1839 ID is the exception identifier. */
1840 void __raise_exception (void **@var{addr}, void *@var{id});
1844 To make the debugger catch all exceptions before any stack
1845 unwinding takes place, set a breakpoint on @code{__raise_exception}
1846 (@pxref{Breakpoints}).
1848 With a conditional breakpoint (@xref{Conditions}) that depends on the
1849 value of @var{id}, you can stop your program when a specific exception
1850 is raised. You can use multiple conditional breakpoints to stop the
1851 program when any of a number of exceptions are raised.
1853 @node Delete Breaks, Disabling, Exception Handling, Breakpoints
1854 @subsection Deleting Breakpoints
1856 @cindex clearing breakpoints, watchpoints
1857 @cindex deleting breakpoints, watchpoints
1858 It is often necessary to eliminate a breakpoint or watchpoint once it
1859 has done its job and you no longer want the program to stop there. This
1860 is called @dfn{deleting} the breakpoint. A breakpoint that has been
1861 deleted no longer exists; it is forgotten.
1863 With the @code{clear} command you can delete breakpoints according to
1864 where they are in the program. With the @code{delete} command you can
1865 delete individual breakpoints or watchpoints by specifying their
1868 It is not necessary to delete a breakpoint to proceed past it. _GDBN__
1869 automatically ignores breakpoints on the first instruction to be executed
1870 when you continue execution without changing the execution address.
1875 Delete any breakpoints at the next instruction to be executed in the
1876 selected stack frame (@pxref{Selection}). When the innermost frame
1877 is selected, this is a good way to delete a breakpoint that the program
1880 @item clear @var{function}
1881 @itemx clear @var{filename}:@var{function}
1882 Delete any breakpoints set at entry to the function @var{function}.
1884 @item clear @var{linenum}
1885 @itemx clear @var{filename}:@var{linenum}
1886 Delete any breakpoints set at or within the code of the specified line.
1888 @item delete @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
1889 @cindex delete breakpoints
1892 Delete the breakpoints or watchpoints of the numbers specified as
1893 arguments. If no argument is specified, delete all breakpoints (_GDBN__
1894 asks confirmation, unless you've @code{set confirm off}). You
1895 can abbreviate this command as @code{d}.
1898 @node Disabling, Conditions, Delete Breaks, Breakpoints
1899 @subsection Disabling Breakpoints
1901 @cindex disabled breakpoints
1902 @cindex enabled breakpoints
1903 Rather than deleting a breakpoint or watchpoint, you might prefer to
1904 @dfn{disable} it. This makes the breakpoint inoperative as if it had
1905 been deleted, but remembers the information on the breakpoint so that
1906 you can @dfn{enable} it again later.
1908 You disable and enable breakpoints and watchpoints with the
1909 @code{enable} and @code{disable} commands, optionally specifying one or
1910 more breakpoint numbers as arguments. Use @code{info break} or
1911 @code{info watch} to print a list of breakpoints or watchpoints if you
1912 don't know which numbers to use.
1914 A breakpoint or watchpoint can have any of four different states of
1919 Enabled. The breakpoint will stop the program. A breakpoint set
1920 with the @code{break} command starts out in this state.
1922 Disabled. The breakpoint has no effect on the program.
1924 Enabled once. The breakpoint will stop the program, but
1925 when it does so it will become disabled. A breakpoint set
1926 with the @code{tbreak} command starts out in this state.
1928 Enabled for deletion. The breakpoint will stop the program, but
1929 immediately after it does so it will be deleted permanently.
1932 You can use the following commands to enable or disable breakpoints and
1936 @item disable @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
1937 @kindex disable breakpoints
1940 Disable the specified breakpoints---or all breakpoints, if none are
1941 listed. A disabled breakpoint has no effect but is not forgotten. All
1942 options such as ignore-counts, conditions and commands are remembered in
1943 case the breakpoint is enabled again later. You may abbreviate
1944 @code{disable} as @code{dis}.
1946 @item enable @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
1947 @kindex enable breakpoints
1949 Enable the specified breakpoints (or all defined breakpoints). They
1950 become effective once again in stopping the program.
1952 @item enable @r{[}breakpoints@r{]} once @var{bnums}@dots{}
1953 Enable the specified breakpoints temporarily. Each will be disabled
1954 again the next time it stops the program.
1956 @item enable @r{[}breakpoints@r{]} delete @var{bnums}@dots{}
1957 Enable the specified breakpoints to work once and then die. Each of
1958 the breakpoints will be deleted the next time it stops the program.
1961 Save for a breakpoint set with @code{tbreak} (@pxref{Set Breaks}),
1962 breakpoints that you set are initially enabled; subsequently, they become
1963 disabled or enabled only when you use one of the commands above. (The
1964 command @code{until} can set and delete a breakpoint of its own, but it
1965 will not change the state of your other breakpoints;
1966 @pxref{Continuing and Stepping}.)
1968 @node Conditions, Break Commands, Disabling, Breakpoints
1969 @subsection Break Conditions
1970 @cindex conditional breakpoints
1971 @cindex breakpoint conditions
1973 @c FIXME what is scope of break condition expr? Context where wanted?
1974 @c in particular for a watchpoint?
1975 The simplest sort of breakpoint breaks every time the program reaches a
1976 specified place. You can also specify a @dfn{condition} for a
1977 breakpoint. A condition is just a Boolean expression in your
1978 programming language. (@xref{Expressions}). A breakpoint with a condition
1979 evaluates the expression each time the program reaches it, and the
1980 program stops only if the condition is @emph{true}.
1982 This is the converse of using assertions for program validation; in that
1983 situation, you want to stop when the assertion is violated---that is,
1984 when the condition is false. In C, if you want to test an assertion expressed
1985 by the condition @var{assert}, you should set the condition
1986 @samp{! @var{assert}} on the appropriate breakpoint.
1988 Conditions are also accepted for watchpoints; you may not need them,
1989 since a watchpoint is inspecting the value of an expression anyhow---but
1990 it might be simpler, say, to just set a watchpoint on a variable name,
1991 and specify a condition that tests whether the new value is an interesting
1994 Break conditions ca have side effects, and may even call functions in
1995 your program. This can be useful, for example, to activate functions
1996 that log program progress, or to use your own print functions to format
1997 special data structures. The effects are completely predictable unless
1998 there is another enabled breakpoint at the same address. (In that
1999 case, _GDBN__ might see the other breakpoint first and stop the program
2000 without checking the condition of this one.) Note that breakpoint
2001 commands are usually more convenient and flexible for the purpose of
2002 performing side effects when a breakpoint is reached
2003 (@pxref{Break Commands}).
2005 Break conditions can be specified when a breakpoint is set, by using
2006 @samp{if} in the arguments to the @code{break} command. @xref{Set Breaks}.
2007 They can also be changed at any time with the @code{condition} command.
2008 The @code{watch} command doesn't recognize the @code{if} keyword;
2009 @code{condition} is the only way to impose a further condition on a
2013 @item condition @var{bnum} @var{expression}
2015 Specify @var{expression} as the break condition for breakpoint or
2016 watchpoint number @var{bnum}. From now on, this breakpoint will stop
2017 the program only if the value of @var{expression} is true (nonzero, in
2018 C). When you use @code{condition}, _GDBN__ checks @var{expression}
2019 immediately for syntactic correctness, and to determine whether symbols
2020 in it have referents in the context of your breakpoint.
2021 @c FIXME so what does GDB do if there's no referent? Moreover, what
2022 @c about watchpoints?
2024 not actually evaluate @var{expression} at the time the @code{condition}
2025 command is given, however. @xref{Expressions}.
2027 @item condition @var{bnum}
2028 Remove the condition from breakpoint number @var{bnum}. It becomes
2029 an ordinary unconditional breakpoint.
2032 @cindex ignore count (of breakpoint)
2033 A special case of a breakpoint condition is to stop only when the
2034 breakpoint has been reached a certain number of times. This is so
2035 useful that there is a special way to do it, using the @dfn{ignore
2036 count} of the breakpoint. Every breakpoint has an ignore count, which
2037 is an integer. Most of the time, the ignore count is zero, and
2038 therefore has no effect. But if the program reaches a breakpoint whose
2039 ignore count is positive, then instead of stopping, it just decrements
2040 the ignore count by one and continues. As a result, if the ignore count
2041 value is @var{n}, the breakpoint will not stop the next @var{n} times it
2045 @item ignore @var{bnum} @var{count}
2047 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
2048 The next @var{count} times the breakpoint is reached, your program's
2049 execution will not stop; other than to decrement the ignore count, _GDBN__
2052 To make the breakpoint stop the next time it is reached, specify
2055 @item continue @var{count}
2056 @itemx c @var{count}
2057 @itemx fg @var{count}
2058 @kindex continue @var{count}
2059 Continue execution of the program, setting the ignore count of the
2060 breakpoint that the program stopped at to @var{count} minus one.
2061 Thus, the program will not stop at this breakpoint until the
2062 @var{count}'th time it is reached.
2064 An argument to this command is meaningful only when the program stopped
2065 due to a breakpoint. At other times, the argument to @code{continue} is
2068 The synonym @code{fg} is provided purely for convenience, and has
2069 exactly the same behavior as other forms of the command.
2072 If a breakpoint has a positive ignore count and a condition, the condition
2073 is not checked. Once the ignore count reaches zero, the condition will
2076 You could achieve the effect of the ignore count with a
2077 condition such as _0__@w{@samp{$foo-- <= 0}}_1__ using a debugger convenience
2078 variable that is decremented each time. @xref{Convenience Vars}.
2080 @node Break Commands, Breakpoint Menus, Conditions, Breakpoints
2081 @subsection Breakpoint Command Lists
2083 @cindex breakpoint commands
2084 You can give any breakpoint (or watchpoint) a series of commands to
2085 execute when the program stops due to that breakpoint. For example, you
2086 might want to print the values of certain expressions, or enable other
2090 @item commands @r{[}@var{bnum}@r{]}
2091 @itemx @dots{} @var{command-list} @dots{}
2095 Specify a list of commands for breakpoint number @var{bnum}. The commands
2096 themselves appear on the following lines. Type a line containing just
2097 @code{end} to terminate the commands.
2099 To remove all commands from a breakpoint, type @code{commands} followed
2100 immediately by @code{end}; that is, give no commands.
2102 With no @var{bnum} argument, @code{commands} refers to the last
2103 breakpoint or watchpoint set (not to the breakpoint most recently
2107 Pressing @key{RET} as a means of repeating the last _GDBN__ command is
2108 disabled within a @var{command-list}.
2110 You can use breakpoint commands to start the program up again. Simply
2111 use the @code{continue} command, or @code{step}, or any other command
2112 that resumes execution. Subsequent commands in the command list are
2116 If the first command specified is @code{silent}, the usual message about
2117 stopping at a breakpoint is not printed. This may be desirable for
2118 breakpoints that are to print a specific message and then continue.
2119 If the remaining commands too print nothing, you will see no sign that
2120 the breakpoint was reached at all. @code{silent} is meaningful only
2121 at the beginning of a breakpoint command list.
2123 The commands @code{echo} and @code{output} that allow you to print precisely
2124 controlled output are often useful in silent breakpoints. @xref{Output}.
2126 For example, here is how you could use breakpoint commands to print the
2127 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
2140 One application for breakpoint commands is to compensate for one bug so
2141 you can test for another. Put a breakpoint just after the erroneous line
2142 of code, give it a condition to detect the case in which something
2143 erroneous has been done, and give it commands to assign correct values
2144 to any variables that need them. End with the @code{continue} command
2145 so that the program does not stop, and start with the @code{silent}
2146 command so that no output is produced. Here is an example:
2158 One deficiency in the operation of automatically continuing breakpoints
2159 under Unix appears when your program uses raw mode for the terminal.
2160 _GDBN__ switches back to its own terminal modes (not raw) before executing
2161 commands, and then must switch back to raw mode when your program is
2162 continued. This causes any pending terminal input to be lost.
2163 @c FIXME: revisit below when GNU sys avail.
2164 @c In the GNU system, this will be fixed by changing the behavior of
2167 Under Unix, you can get around this problem by writing actions into
2168 the breakpoint condition rather than in commands. For example
2171 condition 5 (x = y + 4), 0
2175 specifies a condition expression (@xref{Expressions}) that will change
2176 @code{x} as needed, then always have the value zero so the program will
2177 not stop. No input is lost here, because _GDBN__ evaluates break
2178 conditions without changing the terminal modes. When you want to have
2179 nontrivial conditions for performing the side effects, the operators
2180 @samp{&&}, @samp{||} and @samp{?@dots{}:} may be useful.
2182 @node Breakpoint Menus, Error in Breakpoints, Break Commands, Breakpoints
2183 @subsection Breakpoint Menus
2185 @cindex symbol overloading
2187 Some programming languages (notably C++) permit a single function name
2188 to be defined several times, for application in different contexts.
2189 This is called @dfn{overloading}. When a function name is overloaded,
2190 @samp{break @var{function}} is not enough to tell _GDBN__ where you
2191 want a breakpoint. _GDBN__ offers you a menu of numbered choices for
2192 different possible breakpoints, and waits for your selection with the
2193 prompt @samp{>}. The first two options are always @samp{[0] cancel}
2194 and @samp{[1] all}. Typing @kbd{1} sets a breakpoint at each
2195 definition of @var{function}, and typing @kbd{0} aborts the
2196 @code{break} command without setting any new breakpoints.
2198 For example, the following session excerpt shows an attempt to set a
2199 breakpoint at the overloaded symbol @code{String::after}.
2200 We choose three particular definitions of that function name:
2203 (_GDBP__) b String::after
2206 [2] file:String.cc; line number:867
2207 [3] file:String.cc; line number:860
2208 [4] file:String.cc; line number:875
2209 [5] file:String.cc; line number:853
2210 [6] file:String.cc; line number:846
2211 [7] file:String.cc; line number:735
2213 Breakpoint 1 at 0xb26c: file String.cc, line 867.
2214 Breakpoint 2 at 0xb344: file String.cc, line 875.
2215 Breakpoint 3 at 0xafcc: file String.cc, line 846.
2216 Multiple breakpoints were set.
2217 Use the "delete" command to delete unwanted breakpoints.
2222 @node Error in Breakpoints, , Breakpoint Menus, Breakpoints
2223 @subsection ``Cannot Insert Breakpoints''
2225 @c FIXME: "cannot insert breakpoints" error, v unclear.
2226 @c Q in pending mail to Gilmore. ---pesch@cygnus.com, 26mar91
2227 @c some light may be shed by looking at instances of
2228 @c ONE_PROCESS_WRITETEXT. But error seems possible otherwise
2229 @c too. pesch, 20sep91
2230 Under some operating systems, breakpoints cannot be used in a program if
2231 any other process is running that program. In this situation,
2232 attempting to run or continue a program with a breakpoint causes _GDBN__
2233 to stop the other process.
2235 When this happens, you have three ways to proceed:
2239 Remove or disable the breakpoints, then continue.
2242 Suspend _GDBN__, and copy the file containing the program to a new name.
2243 Resume _GDBN__ and use the @code{exec-file} command to specify that _GDBN__
2244 should run the program under that name. Then start the program again.
2246 @c FIXME: RMS commented here "Show example". Maybe when someone
2247 @c explains the first FIXME: in this section...
2250 Relink the program so that the text segment is nonsharable, using the
2251 linker option @samp{-N}. The operating system limitation may not apply
2252 to nonsharable executables.
2255 @node Continuing and Stepping, Signals, Breakpoints, Stopping
2256 @section Continuing and Stepping
2260 @cindex resuming execution
2261 @dfn{Continuing} means resuming program execution until your program
2262 completes normally. In contrast, @dfn{stepping} means executing just
2263 one more ``step'' of your program, where ``step'' may mean either one
2264 line of source code, or one machine instruction (depending on what
2265 particular command you use). Either when continuing
2266 or when stepping, the program may stop even sooner, due to a breakpoint
2267 or to a signal. (If due to a signal, you may want to use @code{handle},
2268 or use @samp{signal 0} to resume execution; @pxref{Signals}.)
2271 @item continue @r{[}@var{ignore-count}@r{]}
2273 Resume program execution, at the address where the program last stopped;
2274 any breakpoints set at that address are bypassed. The optional argument
2275 @var{ignore-count} allows you to specify a further number of times to
2276 ignore a breakpoint at this location; its effect is like that of
2277 @code{ignore} (@pxref{Conditions}).
2279 To resume execution at a different place, you can use @code{return}
2280 (@pxref{Returning}) to go back to the calling function; or @code{jump}
2281 (@pxref{Jumping}) to go to an arbitrary location in your program.
2285 A typical technique for using stepping is to set a breakpoint
2286 (@pxref{Breakpoints}) at the beginning of the function or the section of
2287 the program in which a problem is believed to lie, run the program until
2288 it stops at that breakpoint, and then step through the suspect area,
2289 examining the variables that are interesting, until you see the problem
2296 Continue running the program until control reaches a different source
2297 line, then stop it and return control to _GDBN__. This command is
2298 abbreviated @code{s}.
2301 @emph{Warning:} If you use the @code{step} command while control is
2302 within a function that was compiled without debugging information,
2303 execution will proceed until control reaches another function.
2306 @item step @var{count}
2307 Continue running as in @code{step}, but do so @var{count} times. If a
2308 breakpoint is reached or a signal not related to stepping occurs before
2309 @var{count} steps, stepping stops right away.
2311 @item next @r{[}@var{count}@r{]}
2314 Continue to the next source line in the current (innermost) stack frame.
2315 Similar to @code{step}, but any function calls appearing within the line
2316 of code are executed without stopping. Execution stops when control
2317 reaches a different line of code at the stack level which was executing
2318 when the @code{next} command was given. This command is abbreviated
2321 An argument @var{count} is a repeat count, as for @code{step}.
2323 @code{next} within a function that lacks debugging information acts like
2324 @code{step}, but any function calls appearing within the code of the
2325 function are executed without stopping.
2329 Continue running until just after function in the selected stack frame
2330 returns. Print the returned value (if any).
2332 Contrast this with the @code{return} command (@pxref{Returning}).
2338 Continue running until a source line past the current line, in the
2339 current stack frame, is reached. This command is used to avoid single
2340 stepping through a loop more than once. It is like the @code{next}
2341 command, except that when @code{until} encounters a jump, it
2342 automatically continues execution until the program counter is greater
2343 than the address of the jump.
2345 This means that when you reach the end of a loop after single stepping
2346 though it, @code{until} will cause the program to continue execution
2347 until the loop is exited. In contrast, a @code{next} command at the end
2348 of a loop will simply step back to the beginning of the loop, which
2349 would force you to step through the next iteration.
2351 @code{until} always stops the program if it attempts to exit the current
2354 @code{until} may produce somewhat counterintuitive results if the order
2355 of machine code does not match the order of the source lines. For
2356 example, in the following excerpt from a debugging session, the @code{f}
2357 (@code{frame}) command shows that execution is stopped at line
2358 @code{206}; yet when we use @code{until}, we get to line @code{195}:
2362 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
2365 195 for ( ; argc > 0; NEXTARG) @{
2368 This happened because, for execution efficiency, the compiler had
2369 generated code for the loop closure test at the end, rather than the
2370 start, of the loop---even though the test in a C @code{for}-loop is
2371 written before the body of the loop. The @code{until} command appeared
2372 to step back to the beginning of the loop when it advanced to this
2373 expression; however, it has not really gone to an earlier
2374 statement---not in terms of the actual machine code.
2376 @code{until} with no argument works by means of single
2377 instruction stepping, and hence is slower than @code{until} with an
2380 @item until @var{location}
2381 @item u @var{location}
2382 Continue running the program until either the specified location is
2383 reached, or the current stack frame returns. @var{location}
2384 is any of the forms of argument acceptable to @code{break}
2385 (@pxref{Set Breaks}). This form of the command uses breakpoints, and
2386 hence is quicker than @code{until} without an argument.
2392 Execute one machine instruction, then stop and return to the debugger.
2394 It is often useful to do @samp{display/i $pc} when stepping by machine
2395 instructions. This will cause the next instruction to be executed to
2396 be displayed automatically at each stop. @xref{Auto Display}.
2398 An argument is a repeat count, as in @code{step}.
2404 Execute one machine instruction, but if it is a function call,
2405 proceed until the function returns.
2407 An argument is a repeat count, as in @code{next}.
2411 @node Signals, , Continuing and Stepping, Stopping
2415 A signal is an asynchronous event that can happen in a program. The
2416 operating system defines the possible kinds of signals, and gives each
2417 kind a name and a number. For example, in Unix @code{SIGINT} is the
2418 signal a program gets when you type an interrupt (often @kbd{C-c});
2419 @code{SIGSEGV} is the signal a program gets from referencing a place in
2420 memory far away from all the areas in use; @code{SIGALRM} occurs when
2421 the alarm clock timer goes off (which happens only if the program has
2422 requested an alarm).
2424 @cindex fatal signals
2425 Some signals, including @code{SIGALRM}, are a normal part of the
2426 functioning of the program. Others, such as @code{SIGSEGV}, indicate
2427 errors; these signals are @dfn{fatal} (kill the program immediately) if the
2428 program has not specified in advance some other way to handle the signal.
2429 @code{SIGINT} does not indicate an error in the program, but it is normally
2430 fatal so it can carry out the purpose of the interrupt: to kill the program.
2432 _GDBN__ has the ability to detect any occurrence of a signal in the program
2433 running under _GDBN__'s control. You can tell _GDBN__ in advance what to do for
2434 each kind of signal.
2436 @cindex handling signals
2437 Normally, _GDBN__ is set up to ignore non-erroneous signals like @code{SIGALRM}
2438 (so as not to interfere with their role in the functioning of the program)
2439 but to stop the program immediately whenever an error signal happens.
2440 You can change these settings with the @code{handle} command.
2444 @kindex info signals
2445 Print a table of all the kinds of signals and how _GDBN__ has been told to
2446 handle each one. You can use this to see the signal numbers of all
2447 the defined types of signals.
2449 @item handle @var{signal} @var{keywords}@dots{}
2451 Change the way _GDBN__ handles signal @var{signal}. @var{signal} can be the
2452 number of a signal or its name (with or without the @samp{SIG} at the
2453 beginning). The @var{keywords} say what change to make.
2457 The keywords allowed by the @code{handle} command can be abbreviated.
2458 Their full names are:
2462 _GDBN__ should not stop the program when this signal happens. It may
2463 still print a message telling you that the signal has come in.
2466 _GDBN__ should stop the program when this signal happens. This implies
2467 the @code{print} keyword as well.
2470 _GDBN__ should print a message when this signal happens.
2473 _GDBN__ should not mention the occurrence of the signal at all. This
2474 implies the @code{nostop} keyword as well.
2477 _GDBN__ should allow the program to see this signal; the program will be
2478 able to handle the signal, or may be terminated if the signal is fatal
2482 _GDBN__ should not allow the program to see this signal.
2486 When a signal has been set to stop the program, the program cannot see the
2487 signal until you continue. It will see the signal then, if @code{pass} is
2488 in effect for the signal in question @i{at that time}. In other words,
2489 after _GDBN__ reports a signal, you can use the @code{handle} command with
2490 @code{pass} or @code{nopass} to control whether that signal will be seen by
2491 the program when you later continue it.
2493 You can also use the @code{signal} command to prevent the program from
2494 seeing a signal, or cause it to see a signal it normally would not see,
2495 or to give it any signal at any time. For example, if the program stopped
2496 due to some sort of memory reference error, you might store correct
2497 values into the erroneous variables and continue, hoping to see more
2498 execution; but the program would probably terminate immediately as
2499 a result of the fatal signal once it sees the signal. To prevent this,
2500 you can continue with @samp{signal 0}. @xref{Signaling}.
2502 @node Stack, Source, Stopping, Top
2503 @chapter Examining the Stack
2505 When your program has stopped, the first thing you need to know is where it
2506 stopped and how it got there.
2509 Each time your program performs a function call, the information about
2510 where in the program the call was made from is saved in a block of data
2511 called a @dfn{stack frame}. The frame also contains the arguments of the
2512 call and the local variables of the function that was called. All the
2513 stack frames are allocated in a region of memory called the @dfn{call
2516 When your program stops, the _GDBN__ commands for examining the stack allow you
2517 to see all of this information.
2519 @cindex selected frame
2520 One of the stack frames is @dfn{selected} by _GDBN__ and many _GDBN__ commands
2521 refer implicitly to the selected frame. In particular, whenever you ask
2522 _GDBN__ for the value of a variable in the program, the value is found in the
2523 selected frame. There are special _GDBN__ commands to select whichever frame
2524 you are interested in.
2526 When the program stops, _GDBN__ automatically selects the currently executing
2527 frame and describes it briefly as the @code{frame} command does
2528 (@pxref{Frame Info}).
2531 * Frames:: Stack Frames
2532 * Backtrace:: Backtraces
2533 * Selection:: Selecting a Frame
2534 * Frame Info:: Information on a Frame
2537 @node Frames, Backtrace, Stack, Stack
2538 @section Stack Frames
2542 The call stack is divided up into contiguous pieces called @dfn{stack
2543 frames}, or @dfn{frames} for short; each frame is the data associated
2544 with one call to one function. The frame contains the arguments given
2545 to the function, the function's local variables, and the address at
2546 which the function is executing.
2548 @cindex initial frame
2549 @cindex outermost frame
2550 @cindex innermost frame
2551 When your program is started, the stack has only one frame, that of the
2552 function @code{main}. This is called the @dfn{initial} frame or the
2553 @dfn{outermost} frame. Each time a function is called, a new frame is
2554 made. Each time a function returns, the frame for that function invocation
2555 is eliminated. If a function is recursive, there can be many frames for
2556 the same function. The frame for the function in which execution is
2557 actually occurring is called the @dfn{innermost} frame. This is the most
2558 recently created of all the stack frames that still exist.
2560 @cindex frame pointer
2561 Inside your program, stack frames are identified by their addresses. A
2562 stack frame consists of many bytes, each of which has its own address; each
2563 kind of computer has a convention for choosing one of those bytes whose
2564 address serves as the address of the frame. Usually this address is kept
2565 in a register called the @dfn{frame pointer register} while execution is
2566 going on in that frame.
2568 @cindex frame number
2569 _GDBN__ assigns numbers to all existing stack frames, starting with
2570 zero for the innermost frame, one for the frame that called it,
2571 and so on upward. These numbers do not really exist in your program;
2572 they are assigned by _GDBN__ to give you a way of designating stack
2573 frames in _GDBN__ commands.
2575 @cindex frameless execution
2576 Some compilers allow functions to be compiled so that they operate
2577 without stack frames. (For example, the @code{_GCC__} option
2578 @samp{-fomit-frame-pointer} will generate functions without a frame.)
2579 This is occasionally done with heavily used library functions to save
2580 the frame setup time. _GDBN__ has limited facilities for dealing with
2581 these function invocations. If the innermost function invocation has no
2582 stack frame, _GDBN__ will nevertheless regard it as though it had a
2583 separate frame, which is numbered zero as usual, allowing correct
2584 tracing of the function call chain. However, _GDBN__ has no provision
2585 for frameless functions elsewhere in the stack.
2587 @node Backtrace, Selection, Frames, Stack
2590 A backtrace is a summary of how the program got where it is. It shows one
2591 line per frame, for many frames, starting with the currently executing
2592 frame (frame zero), followed by its caller (frame one), and on up the
2600 Print a backtrace of the entire stack: one line per frame for all
2601 frames in the stack.
2603 You can stop the backtrace at any time by typing the system interrupt
2604 character, normally @kbd{C-c}.
2606 @item backtrace @var{n}
2608 Similar, but print only the innermost @var{n} frames.
2610 @item backtrace -@var{n}
2612 Similar, but print only the outermost @var{n} frames.
2618 The names @code{where} and @code{info stack} (abbreviated @code{info s})
2619 are additional aliases for @code{backtrace}.
2621 Each line in the backtrace shows the frame number and the function name.
2622 The program counter value is also shown---unless you use @code{set
2623 print address off}. The backtrace also shows the source file name and
2624 line number, as well as the arguments to the function. The program
2625 counter value is omitted if it is at the beginning of the code for that
2628 Here is an example of a backtrace. It was made with the command
2629 @samp{bt 3}, so it shows the innermost three frames.
2633 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8) at builtin.c:993
2634 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
2635 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
2637 (More stack frames follow...)
2642 The display for frame zero doesn't begin with a program counter
2643 value, indicating that the program has stopped at the beginning of the
2644 code for line @code{993} of @code{builtin.c}.
2646 @node Selection, Frame Info, Backtrace, Stack
2647 @section Selecting a Frame
2649 Most commands for examining the stack and other data in the program work on
2650 whichever stack frame is selected at the moment. Here are the commands for
2651 selecting a stack frame; all of them finish by printing a brief description
2652 of the stack frame just selected.
2659 Select frame number @var{n}. Recall that frame zero is the innermost
2660 (currently executing) frame, frame one is the frame that called the
2661 innermost one, and so on. The highest-numbered frame is @code{main}'s
2664 @item frame @var{addr}
2666 Select the frame at address @var{addr}. This is useful mainly if the
2667 chaining of stack frames has been damaged by a bug, making it
2668 impossible for _GDBN__ to assign numbers properly to all frames. In
2669 addition, this can be useful when the program has multiple stacks and
2670 switches between them.
2673 On the SPARC architecture, @code{frame} needs two addresses to
2674 select an arbitrary frame: a frame pointer and a stack pointer.
2675 @c note to future updaters: this is conditioned on a flag
2676 @c FRAME_SPECIFICATION_DYADIC in the tm-*.h files, currently only used
2677 @c by SPARC, hence the specific attribution. Generalize or list all
2678 @c possibilities if more supported machines start doing this.
2683 Move @var{n} frames up the stack. For positive numbers @var{n}, this
2684 advances toward the outermost frame, to higher frame numbers, to frames
2685 that have existed longer. @var{n} defaults to one.
2690 Move @var{n} frames down the stack. For positive numbers @var{n}, this
2691 advances toward the innermost frame, to lower frame numbers, to frames
2692 that were created more recently. @var{n} defaults to one. You may
2693 abbreviate @code{down} as @code{do}.
2696 All of these commands end by printing two lines of output describing the
2697 frame. The first line shows the frame number, the function name, the
2698 arguments, and the source file and line number of execution in that
2699 frame. The second line shows the text of that source line. For
2704 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc) at env.c:10
2705 10 read_input_file (argv[i]);
2708 After such a printout, the @code{list} command with no arguments will print
2709 ten lines centered on the point of execution in the frame. @xref{List}.
2712 @item up-silently @var{n}
2713 @itemx down-silently @var{n}
2714 @kindex down-silently
2716 These two commands are variants of @code{up} and @code{down},
2717 respectively; they differ in that they do their work silently, without
2718 causing display of the new frame. They are intended primarily for use
2719 in _GDBN__ command scripts, where the output might be unnecessary and
2724 @node Frame Info, , Selection, Stack
2725 @section Information About a Frame
2727 There are several other commands to print information about the selected
2733 When used without any argument, this command does not change which frame
2734 is selected, but prints a brief description of the currently
2735 selected stack frame. It can be abbreviated @code{f}. With an
2736 argument, this command is used to select a stack frame (@pxref{Selection}).
2742 This command prints a verbose description of the selected stack frame,
2743 including the address of the frame, the addresses of the next frame down
2744 (called by this frame) and the next frame up (caller of this frame), the
2745 language that the source code corresponding to this frame was written in,
2746 the address of the frame's arguments, the program counter saved in it
2747 (the address of execution in the caller frame), and which registers
2748 were saved in the frame. The verbose description is useful when
2749 something has gone wrong that has made the stack format fail to fit
2750 the usual conventions.
2752 @item info frame @var{addr}
2753 @itemx info f @var{addr}
2754 Print a verbose description of the frame at address @var{addr},
2755 without selecting that frame. The selected frame remains unchanged by
2760 Print the arguments of the selected frame, each on a separate line.
2764 Print the local variables of the selected frame, each on a separate
2765 line. These are all variables declared static or automatic within all
2766 program blocks that execution in this frame is currently inside of.
2770 @cindex catch exceptions
2771 @cindex exception handlers
2772 Print a list of all the exception handlers that are active in the
2773 current stack frame at the current point of execution. To see other
2774 exception handlers, visit the associated frame (using the @code{up},
2775 @code{down}, or @code{frame} commands); then type @code{info catch}.
2776 @xref{Exception Handling}.
2779 @node Source, Data, Stack, Top
2780 @chapter Examining Source Files
2782 _GDBN__ can print parts of your program's source, since the debugging
2783 information recorded in your program tells _GDBN__ what source files
2784 were used to built it. When your program stops, _GDBN__ spontaneously
2785 prints the line where it stopped. Likewise, when you select a stack
2786 frame (@pxref{Selection}), _GDBN__ prints the line where execution in
2787 that frame has stopped. You can print other portions of source files by
2790 If you use _GDBN__ through its GNU Emacs interface, you may prefer to
2791 use Emacs facilities to view source; @pxref{Emacs}.
2794 * List:: Printing Source Lines
2795 * Search:: Searching Source Files
2796 * Source Path:: Specifying Source Directories
2797 * Machine Code:: Source and Machine Code
2800 @node List, Search, Source, Source
2801 @section Printing Source Lines
2805 To print lines from a source file, use the @code{list} command
2806 (abbreviated @code{l}). There are several ways to specify what part
2807 of the file you want to print.
2809 Here are the forms of the @code{list} command most commonly used:
2812 @item list @var{linenum}
2813 Print lines centered around line number @var{linenum} in the
2814 current source file.
2816 @item list @var{function}
2817 Print lines centered around the beginning of function
2821 Print more lines. If the last lines printed were printed with a
2822 @code{list} command, this prints lines following the last lines
2823 printed; however, if the last line printed was a solitary line printed
2824 as part of displaying a stack frame (@pxref{Stack}), this prints
2825 lines centered around that line.
2828 Print lines just before the lines last printed.
2831 By default, _GDBN__ prints ten source lines with any of these forms of
2832 the @code{list} command. You can change this using @code{set listsize}:
2835 @item set listsize @var{count}
2836 @kindex set listsize
2837 Make the @code{list} command display @var{count} source lines (unless
2838 the @code{list} argument explicitly specifies some other number).
2841 @kindex show listsize
2842 Display the number of lines that @code{list} will currently display by
2846 Repeating a @code{list} command with @key{RET} discards the argument,
2847 so it is equivalent to typing just @code{list}. This is more useful
2848 than listing the same lines again. An exception is made for an
2849 argument of @samp{-}; that argument is preserved in repetition so that
2850 each repetition moves up in the source file.
2853 In general, the @code{list} command expects you to supply zero, one or two
2854 @dfn{linespecs}. Linespecs specify source lines; there are several ways
2855 of writing them but the effect is always to specify some source line.
2856 Here is a complete description of the possible arguments for @code{list}:
2859 @item list @var{linespec}
2860 Print lines centered around the line specified by @var{linespec}.
2862 @item list @var{first},@var{last}
2863 Print lines from @var{first} to @var{last}. Both arguments are
2866 @item list ,@var{last}
2867 Print lines ending with @var{last}.
2869 @item list @var{first},
2870 Print lines starting with @var{first}.
2873 Print lines just after the lines last printed.
2876 Print lines just before the lines last printed.
2879 As described in the preceding table.
2882 Here are the ways of specifying a single source line---all the
2887 Specifies line @var{number} of the current source file.
2888 When a @code{list} command has two linespecs, this refers to
2889 the same source file as the first linespec.
2892 Specifies the line @var{offset} lines after the last line printed.
2893 When used as the second linespec in a @code{list} command that has
2894 two, this specifies the line @var{offset} lines down from the
2898 Specifies the line @var{offset} lines before the last line printed.
2900 @item @var{filename}:@var{number}
2901 Specifies line @var{number} in the source file @var{filename}.
2903 @item @var{function}
2904 @c FIXME: "of the open-brace" is C-centric. When we add other langs...
2905 Specifies the line of the open-brace that begins the body of the
2906 function @var{function}.
2908 @item @var{filename}:@var{function}
2909 Specifies the line of the open-brace that begins the body of the
2910 function @var{function} in the file @var{filename}. You only need the
2911 file name with a function name to avoid ambiguity when there are
2912 identically named functions in different source files.
2914 @item *@var{address}
2915 Specifies the line containing the program address @var{address}.
2916 @var{address} may be any expression.
2919 @node Search, Source Path, List, Source
2920 @section Searching Source Files
2922 @kindex reverse-search
2924 There are two commands for searching through the current source file for a
2928 @item forward-search @var{regexp}
2929 @itemx search @var{regexp}
2931 @kindex forward-search
2932 The command @samp{forward-search @var{regexp}} checks each line, starting
2933 with the one following the last line listed, for a match for @var{regexp}.
2934 It lists the line that is found. You can abbreviate the command name
2935 as @code{fo}. The synonym @samp{search @var{regexp}} is also supported.
2937 @item reverse-search @var{regexp}
2938 The command @samp{reverse-search @var{regexp}} checks each line, starting
2939 with the one before the last line listed and going backward, for a match
2940 for @var{regexp}. It lists the line that is found. You can abbreviate
2941 this command as @code{rev}.
2944 @node Source Path, Machine Code, Search, Source
2945 @section Specifying Source Directories
2948 @cindex directories for source files
2949 Executable programs sometimes do not record the directories of the source
2950 files from which they were compiled, just the names. Even when they do,
2951 the directories could be moved between the compilation and your debugging
2952 session. _GDBN__ has a list of directories to search for source files;
2953 this is called the @dfn{source path}. Each time _GDBN__ wants a source file,
2954 it tries all the directories in the list, in the order they are present
2955 in the list, until it finds a file with the desired name. Note that
2956 the executable search path is @emph{not} used for this purpose. Neither is
2957 the current working directory, unless it happens to be in the source
2960 If _GDBN__ can't find a source file in the source path, and the object
2961 program records a directory, _GDBN__ tries that directory too. If the
2962 source path is empty, and there is no record of the compilation
2963 directory, _GDBN__ will, as a last resort, look in the current
2966 Whenever you reset or rearrange the source path, _GDBN__ will clear out
2967 any information it has cached about where source files are found, where
2968 each line is in the file, etc.
2971 When you start _GDBN__, its source path is empty.
2972 To add other directories, use the @code{directory} command.
2975 @item directory @var{dirname} @dots{}
2976 Add directory @var{dirname} to the front of the source path. Several
2977 directory names may be given to this command, separated by @samp{:} or
2978 whitespace. You may specify a directory that is already in the source
2979 path; this moves it forward, so it will be searched sooner.
2981 You can use the string @samp{$cdir} to refer to the compilation
2982 directory (if one is recorded), and @samp{$cwd} to refer to the current
2983 working directory. @samp{$cwd} is not the same as @samp{.}---the former
2984 tracks the current working directory as it changes during your _GDBN__
2985 session, while the latter is immediately expanded to the current
2986 directory at the time you add an entry to the source path.
2989 Reset the source path to empty again. This requires confirmation.
2991 @c RET-repeat for @code{directory} is explicitly disabled, but since
2992 @c repeating it would be a no-op we don't say that. (thanks to RMS)
2994 @item show directories
2995 @kindex show directories
2996 Print the source path: show which directories it contains.
2999 If your source path is cluttered with directories that are no longer of
3000 interest, _GDBN__ may sometimes cause confusion by finding the wrong
3001 versions of source. You can correct the situation as follows:
3005 Use @code{directory} with no argument to reset the source path to empty.
3008 Use @code{directory} with suitable arguments to reinstall the
3009 directories you want in the source path. You can add all the
3010 directories in one command.
3013 @node Machine Code, , Source Path, Source
3014 @section Source and Machine Code
3015 You can use the command @code{info line} to map source lines to program
3016 addresses (and viceversa), and the command @code{disassemble} to display
3017 a range of addresses as machine instructions.
3020 @item info line @var{linespec}
3022 Print the starting and ending addresses of the compiled code for
3023 source line @var{linespec}. You can specify source lines in any of the
3024 ways understood by the @code{list} command (@pxref{List}).
3027 For example, we can use @code{info line} to inquire on where the object
3028 code for the first line of function @code{m4_changequote} lies:
3030 (_GDBP__) info line m4_changecom
3031 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
3035 We can also inquire (using @code{*@var{addr}} as the form for
3036 @var{linespec}) what source line covers a particular address:
3038 (_GDBP__) info line *0x63ff
3039 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
3042 @cindex @code{$_} and @code{info line}
3043 After @code{info line}, the default address for the @code{x}
3044 command is changed to the starting address of the line, so that
3045 @samp{x/i} is sufficient to begin examining the machine code
3046 (@pxref{Memory}). Also, this address is saved as the value of the
3047 convenience variable @code{$_} (@pxref{Convenience Vars}).
3052 This specialized command is provided to dump a range of memory as
3053 machine instructions. The default memory range is the function
3054 surrounding the program counter of the selected frame. A single
3055 argument to this command is a program counter value; the function
3056 surrounding this value will be dumped. Two arguments (separated by one
3057 or more spaces) specify a range of addresses (first inclusive, second
3058 exclusive) to be dumped.
3061 We can use @code{disassemble} to inspect the object code
3062 range shown in the last @code{info line} example:
3065 (_GDBP__) disas 0x63e4 0x6404
3066 Dump of assembler code from 0x63e4 to 0x6404:
3067 0x63e4 <builtin_init+5340>: ble 0x63f8 <builtin_init+5360>
3068 0x63e8 <builtin_init+5344>: sethi %hi(0x4c00), %o0
3069 0x63ec <builtin_init+5348>: ld [%i1+4], %o0
3070 0x63f0 <builtin_init+5352>: b 0x63fc <builtin_init+5364>
3071 0x63f4 <builtin_init+5356>: ld [%o0+4], %o0
3072 0x63f8 <builtin_init+5360>: or %o0, 0x1a4, %o0
3073 0x63fc <builtin_init+5364>: call 0x9288 <path_search>
3074 0x6400 <builtin_init+5368>: nop
3075 End of assembler dump.
3080 @node Data, Languages, Source, Top
3081 @chapter Examining Data
3083 @cindex printing data
3084 @cindex examining data
3087 @c "inspect" isn't quite a synonym if you're using Epoch, which we don't
3088 @c document because it's nonstandard... Under Epoch it displays in a
3089 @c different window or something like that.
3090 The usual way to examine data in your program is with the @code{print}
3091 command (abbreviated @code{p}), or its synonym @code{inspect}. It
3092 evaluates and prints the value of an expression of the language your
3093 program is written in (@pxref{Languages}).
3096 @item print @var{exp}
3097 @itemx print /@var{f} @var{exp}
3098 @var{exp} is an expression (in the source language). By default
3099 the value of @var{exp} is printed in a format appropriate to its data
3100 type; you can choose a different format by specifying @samp{/@var{f}},
3101 where @var{f} is a letter specifying the format; @pxref{Output formats}.
3104 @itemx print /@var{f}
3105 If you omit @var{exp}, _GDBN__ displays the last value again (from the
3106 @dfn{value history}; @pxref{Value History}). This allows you to
3107 conveniently inspect the same value in an alternative format.
3110 A more low-level way of examining data is with the @code{x} command.
3111 It examines data in memory at a specified address and prints it in a
3112 specified format. @xref{Memory}.
3114 If you're interested in information about types, or about how the fields
3115 of a struct or class are declared, use the @code{ptype @var{exp}}
3116 command rather than @code{print}. @xref{Symbols}.
3119 * Expressions:: Expressions
3120 * Variables:: Program Variables
3121 * Arrays:: Artificial Arrays
3122 * Output formats:: Output formats
3123 * Memory:: Examining Memory
3124 * Auto Display:: Automatic Display
3125 * Print Settings:: Print Settings
3126 * Value History:: Value History
3127 * Convenience Vars:: Convenience Variables
3128 * Registers:: Registers
3129 * Floating Point Hardware:: Floating Point Hardware
3132 @node Expressions, Variables, Data, Data
3133 @section Expressions
3136 @code{print} and many other _GDBN__ commands accept an expression and
3137 compute its value. Any kind of constant, variable or operator defined
3138 by the programming language you are using is legal in an expression in
3139 _GDBN__. This includes conditional expressions, function calls, casts
3140 and string constants. It unfortunately does not include symbols defined
3141 by preprocessor @code{#define} commands.
3143 Because C is so widespread, most of the expressions shown in examples in
3144 this manual are in C. @xref{Languages,, Using _GDBN__ with Different
3145 Languages}, for information on how to use expressions in other
3148 In this section, we discuss operators that you can use in _GDBN__
3149 expressions regardless of your programming language.
3151 Casts are supported in all languages, not just in C, because it is so
3152 useful to cast a number into a pointer so as to examine a structure
3153 at that address in memory.
3154 @c FIXME: casts supported---Mod2 true?
3156 _GDBN__ supports these operators in addition to those of programming
3161 @samp{@@} is a binary operator for treating parts of memory as arrays.
3162 @xref{Arrays}, for more information.
3165 @samp{::} allows you to specify a variable in terms of the file or
3166 function where it is defined. @xref{Variables}.
3168 @item @{@var{type}@} @var{addr}
3169 Refers to an object of type @var{type} stored at address @var{addr} in
3170 memory. @var{addr} may be any expression whose value is an integer or
3171 pointer (but parentheses are required around binary operators, just as in
3172 a cast). This construct is allowed regardless of what kind of data is
3173 normally supposed to reside at @var{addr}.@refill
3176 @node Variables, Arrays, Expressions, Data
3177 @section Program Variables
3179 The most common kind of expression to use is the name of a variable
3182 Variables in expressions are understood in the selected stack frame
3183 (@pxref{Selection}); they must either be global (or static) or be visible
3184 according to the scope rules of the programming language from the point of
3185 execution in that frame. This means that in the function
3200 the variable @code{a} is usable whenever the program is executing
3201 within the function @code{foo}, but the variable @code{b} is visible
3202 only while the program is executing inside the block in which @code{b}
3205 @cindex variable name conflict
3206 There is an exception: you can refer to a variable or function whose
3207 scope is a single source file even if the current execution point is not
3208 in this file. But it is possible to have more than one such variable or
3209 function with the same name (in different source files). If that happens,
3210 referring to that name has unpredictable effects. If you wish, you can
3211 specify a variable in a particular file, using the colon-colon notation:
3216 @var{file}::@var{variable}
3220 Here @var{file} is the name of the source file whose variable you want.
3222 @cindex C++ scope resolution
3223 This use of @samp{::} is very rarely in conflict with the very similar
3224 use of the same notation in C++. _GDBN__ also supports use of the C++
3225 scope resolution operator in _GDBN__ expressions.
3227 @cindex wrong values
3228 @cindex variable values, wrong
3230 @emph{Warning:} Occasionally, a local variable may appear to have the
3231 wrong value at certain points in a function---just after entry to the
3232 function, and just before exit. You may see this problem when you're
3233 stepping by machine instructions. This is because on most machines, it
3234 takes more than one instruction to set up a stack frame (including local
3235 variable definitions); if you're stepping by machine instructions,
3236 variables may appear to have the wrong values until the stack frame is
3237 completely built. On function exit, it usually also takes more than one
3238 machine instruction to destroy a stack frame; after you begin stepping
3239 through that group of instructions, local variable definitions may be
3243 @node Arrays, Output formats, Variables, Data
3244 @section Artificial Arrays
3246 @cindex artificial array
3248 It is often useful to print out several successive objects of the
3249 same type in memory; a section of an array, or an array of
3250 dynamically determined size for which only a pointer exists in the
3253 This can be done by constructing an @dfn{artificial array} with the
3254 binary operator @samp{@@}. The left operand of @samp{@@} should be
3255 the first element of the desired array, as an individual object.
3256 The right operand should be the desired length of the array. The result is
3257 an array value whose elements are all of the type of the left argument.
3258 The first element is actually the left argument; the second element
3259 comes from bytes of memory immediately following those that hold the
3260 first element, and so on. Here is an example. If a program says
3263 int *array = (int *) malloc (len * sizeof (int));
3267 you can print the contents of @code{array} with
3273 The left operand of @samp{@@} must reside in memory. Array values made
3274 with @samp{@@} in this way behave just like other arrays in terms of
3275 subscripting, and are coerced to pointers when used in expressions.
3276 Artificial arrays most often appear in expressions via the value history
3277 (@pxref{Value History}), after printing one out.)
3279 Sometimes the artificial array mechanism isn't quite enough; in
3280 moderately complex data structures, the elements of interest may not
3281 actually be adjacent---for example, if you're interested in the values
3282 of pointers in an array. One useful work-around in this situation is to
3283 use a convenience variable (@pxref{Convenience Vars}) as a counter in an
3284 expression that prints the first interesting value, and then repeat that
3285 expression via @key{RET}. For instance, suppose you have an array
3286 @code{dtab} of pointers to structures, and you're interested in the
3287 values of a field @code{fv} in each structure. Here's an example of
3288 what you might type:
3297 @node Output formats, Memory, Arrays, Data
3298 @section Output formats
3300 @cindex formatted output
3301 @cindex output formats
3302 By default, _GDBN__ prints a value according to its data type. Sometimes
3303 this is not what you want. For example, you might want to print a number
3304 in hex, or a pointer in decimal. Or you might want to view data in memory
3305 at a certain address as a character string or as an instruction. To do
3306 these things, specify an @dfn{output format} when you print a value.
3308 The simplest use of output formats is to say how to print a value
3309 already computed. This is done by starting the arguments of the
3310 @code{print} command with a slash and a format letter. The format
3311 letters supported are:
3315 Regard the bits of the value as an integer, and print the integer in
3319 Print as integer in signed decimal.
3322 Print as integer in unsigned decimal.
3325 Print as integer in octal.
3328 Print as integer in binary. The letter @samp{t} stands for ``two''.
3331 Print as an address, both absolute in hex and as an offset from the
3332 nearest preceding symbol. This format can be used to discover where (in
3333 what function) an unknown address is located:
3335 (_GDBP__) p/a 0x54320
3336 _0__$3 = 0x54320 <_initialize_vx+396>_1__
3341 Regard as an integer and print it as a character constant.
3344 Regard the bits of the value as a floating point number and print
3345 using typical floating point syntax.
3348 For example, to print the program counter in hex (@pxref{Registers}), type
3355 Note that no space is required before the slash; this is because command
3356 names in _GDBN__ cannot contain a slash.
3358 To reprint the last value in the value history with a different format,
3359 you can use the @code{print} command with just a format and no
3360 expression. For example, @samp{p/x} reprints the last value in hex.
3362 @node Memory, Auto Display, Output formats, Data
3363 @section Examining Memory
3365 @cindex examining memory
3368 @item x/@var{nfu} @var{addr}
3371 You can use the command @code{x} (for `examine') to examine memory in
3372 any of several formats, independently of your program's data types.
3373 @var{n}, @var{f}, and @var{u} are all optional parameters to specify how
3374 much memory to display, and how to format it; @var{addr} is an
3375 expression giving the address where you want to start displaying memory.
3376 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
3377 Several commands set convenient defaults for @var{addr}.
3380 @var{n}, the repeat count, is a decimal integer; the default is 1. It
3381 specifies how much memory (counting by units @var{u}) to display.
3382 @c This really is **decimal**; unaffected by 'set radix' as of GDB
3385 @var{f}, the display format, is one of the formats used by @code{print},
3386 or @samp{s} (null-terminated string) or @samp{i} (machine instruction).
3387 The default is @samp{x} (hexadecimal) initially, or the format from the
3388 last time you used either @code{x} or @code{print}.
3390 @var{u}, the unit size, is any of
3395 Halfwords (two bytes).
3397 Words (four bytes). This is the initial default.
3399 Giant words (eight bytes).
3403 Each time you specify a unit size with @code{x}, that size becomes the
3404 default unit the next time you use @code{x}. (For the @samp{s} and
3405 @samp{i} formats, the unit size is ignored and is normally not written.)
3407 @var{addr} is the address where you want _GDBN__ to begin displaying
3408 memory. The expression need not have a pointer value (though it may);
3409 it is always interpreted as an integer address of a byte of memory.
3410 @xref{Expressions} for more information on expressions. The default for
3411 @var{addr} is usually just after the last address examined---but several
3412 other commands also set the default address: @code{info breakpoints} (to
3413 the address of the last breakpoint listed), @code{info line} (to the
3414 starting address of a line), and @code{print} (if you use it to display
3415 a value from memory).
3417 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
3418 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
3419 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
3420 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
3421 @pxref{Registers}) in hexadecimal (@samp{x}).
3423 Since the letters indicating unit sizes are all distinct from the
3424 letters specifying output formats, you don't have to remember whether
3425 unit size or format comes first; either order will work. The output
3426 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
3427 (However, the count @var{n} must come first; @samp{wx4} will not work.)
3429 Even though the unit size @var{u} is ignored for the formats @samp{s}
3430 and @samp{i}, you might still want to use a count @var{n}; for example,
3431 @samp{3i} specifies that you want to see three machine instructions,
3432 including any operands. The command @code{disassemble} gives an
3433 alternative way of inspecting machine instructions; @pxref{Machine
3436 All the defaults for the arguments to @code{x} are designed to make it
3437 easy to continue scanning memory with minimal specifications each time
3438 you use @code{x}. For example, after you've inspected three machine
3439 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
3440 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
3441 the repeat count @var{n} is used again; the other arguments default as
3442 for successive uses of @code{x}.
3444 @cindex @code{$_}, @code{$__}, and value history
3445 The addresses and contents printed by the @code{x} command are not saved
3446 in the value history because there is often too much of them and they
3447 would get in the way. Instead, _GDBN__ makes these values available for
3448 subsequent use in expressions as values of the convenience variables
3449 @code{$_} and @code{$__}. After an @code{x} command, the last address
3450 examined is available for use in expressions in the convenience variable
3451 @code{$_}. The contents of that address, as examined, are available in
3452 the convenience variable @code{$__}.
3454 If the @code{x} command has a repeat count, the address and contents saved
3455 are from the last memory unit printed; this is not the same as the last
3456 address printed if several units were printed on the last line of output.
3458 @node Auto Display, Print Settings, Memory, Data
3459 @section Automatic Display
3460 @cindex automatic display
3461 @cindex display of expressions
3463 If you find that you want to print the value of an expression frequently
3464 (to see how it changes), you might want to add it to the @dfn{automatic
3465 display list} so that _GDBN__ will print its value each time the program stops.
3466 Each expression added to the list is given a number to identify it;
3467 to remove an expression from the list, you specify that number.
3468 The automatic display looks like this:
3472 3: bar[5] = (struct hack *) 0x3804
3476 showing item numbers, expressions and their current values. As with
3477 displays you request manually using @code{x} or @code{print}, you can
3478 specify the output format you prefer; in fact, @code{display} decides
3479 whether to use @code{print} or @code{x} depending on how elaborate your
3480 format specification is---it uses @code{x} if you specify a unit size,
3481 or one of the two formats (@samp{i} and @samp{s}) that are only
3482 supported by @code{x}; otherwise it uses @code{print}.
3485 @item display @var{exp}
3487 Add the expression @var{exp} to the list of expressions to display
3488 each time the program stops. @xref{Expressions}.
3490 @code{display} will not repeat if you press @key{RET} again after using it.
3492 @item display/@var{fmt} @var{exp}
3493 For @var{fmt} specifying only a display format and not a size or
3494 count, add the expression @var{exp} to the auto-display list but
3495 arranges to display it each time in the specified format @var{fmt}.
3496 @xref{Output formats}.
3498 @item display/@var{fmt} @var{addr}
3499 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
3500 number of units, add the expression @var{addr} as a memory address to
3501 be examined each time the program stops. Examining means in effect
3502 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory}.
3505 For example, @samp{display/i $pc} can be helpful, to see the machine
3506 instruction about to be executed each time execution stops (@samp{$pc}
3507 is a common name for the program counter; @pxref{Registers}).
3510 @item undisplay @var{dnums}@dots{}
3511 @itemx delete display @var{dnums}@dots{}
3512 @kindex delete display
3514 Remove item numbers @var{dnums} from the list of expressions to display.
3516 @code{undisplay} will not repeat if you press @key{RET} after using it.
3517 (Otherwise you would just get the error @samp{No display number @dots{}}.)
3519 @item disable display @var{dnums}@dots{}
3520 @kindex disable display
3521 Disable the display of item numbers @var{dnums}. A disabled display
3522 item is not printed automatically, but is not forgotten. It may be
3523 enabled again later.
3525 @item enable display @var{dnums}@dots{}
3526 @kindex enable display
3527 Enable display of item numbers @var{dnums}. It becomes effective once
3528 again in auto display of its expression, until you specify otherwise.
3531 Display the current values of the expressions on the list, just as is
3532 done when the program stops.
3535 @kindex info display
3536 Print the list of expressions previously set up to display
3537 automatically, each one with its item number, but without showing the
3538 values. This includes disabled expressions, which are marked as such.
3539 It also includes expressions which would not be displayed right now
3540 because they refer to automatic variables not currently available.
3543 If a display expression refers to local variables, then it does not make
3544 sense outside the lexical context for which it was set up. Such an
3545 expression is disabled when execution enters a context where one of its
3546 variables is not defined. For example, if you give the command
3547 @code{display last_char} while inside a function with an argument
3548 @code{last_char}, then this argument will be displayed while the program
3549 continues to stop inside that function. When it stops elsewhere---where
3550 there is no variable @code{last_char}---display is disabled. The next time
3551 your program stops where @code{last_char} is meaningful, you can enable the
3552 display expression once again.
3554 @node Print Settings, Value History, Auto Display, Data
3555 @section Print Settings
3557 @cindex format options
3558 @cindex print settings
3559 _GDBN__ provides the following ways to control how arrays, structures,
3560 and symbols are printed.
3563 These settings are useful for debugging programs in any language:
3566 @item set print address
3567 @item set print address on
3568 @kindex set print address
3569 _GDBN__ will print memory addresses showing the location of stack
3570 traces, structure values, pointer values, breakpoints, and so forth,
3571 even when it also displays the contents of those addresses. The default
3572 is on. For example, this is what a stack frame display looks like, with
3573 @code{set print address on}:
3576 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
3578 530 if (lquote != def_lquote)
3581 @item set print address off
3582 Do not print addresses when displaying their contents. For example,
3583 this is the same stack frame displayed with @code{set print address off}:
3585 (_GDBP__) set print addr off
3587 #0 set_quotes (lq="<<", rq=">>") at input.c:530
3588 530 if (lquote != def_lquote)
3591 @item show print address
3592 @kindex show print address
3593 Show whether or not addresses are to be printed.
3595 @item set print array
3596 @itemx set print array on
3597 @kindex set print array
3598 _GDBN__ will pretty print arrays. This format is more convenient to read,
3599 but uses more space. The default is off.
3601 @item set print array off.
3602 Return to compressed format for arrays.
3604 @item show print array
3605 @kindex show print array
3606 Show whether compressed or pretty format is selected for displaying
3609 @item set print elements @var{number-of-elements}
3610 @kindex set print elements
3611 If _GDBN__ is printing a large array, it will stop printing after it has
3612 printed the number of elements set by the @code{set print elements} command.
3613 This limit also applies to the display of strings.
3615 @item show print elements
3616 @kindex show print elements
3617 Display the number of elements of a large array that _GDBN__ will print
3618 before losing patience.
3620 @item set print pretty on
3621 @kindex set print pretty
3622 Cause _GDBN__ to print structures in an indented format with one member per
3636 @item set print pretty off
3637 Cause _GDBN__ to print structures in a compact format, like this:
3640 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, meat \
3645 This is the default format.
3647 @item show print pretty
3648 @kindex show print pretty
3649 Show which format _GDBN__ will use to print structures.
3651 @item set print sevenbit-strings on
3652 @kindex set print sevenbit-strings
3653 Print using only seven-bit characters; if this option is set,
3654 _GDBN__ will display any eight-bit characters (in strings or character
3655 values) using the notation @code{\}@var{nnn}. For example, @kbd{M-a} is
3656 displayed as @code{\341}.
3658 @item set print sevenbit-strings off
3659 Print using either seven-bit or eight-bit characters, as required. This
3662 @item show print sevenbit-strings
3663 @kindex show print sevenbit-strings
3664 Show whether or not _GDBN__ will print only seven-bit characters.
3666 @item set print union on
3667 @kindex set print union
3668 Tell _GDBN__ to print unions which are contained in structures. This is the
3671 @item set print union off
3672 Tell _GDBN__ not to print unions which are contained in structures.
3674 @item show print union
3675 @kindex show print union
3676 Ask _GDBN__ whether or not it will print unions which are contained in
3679 For example, given the declarations
3682 typedef enum @{Tree, Bug@} Species;
3683 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
3684 typedef enum @{Caterpillar, Cocoon, Butterfly@} Bug_forms;
3694 struct thing foo = @{Tree, @{Acorn@}@};
3698 with @code{set print union on} in effect @samp{p foo} would print
3701 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
3705 and with @code{set print union off} in effect it would print
3708 $1 = @{it = Tree, form = @{...@}@}
3713 These settings are of interest when debugging C++ programs:
3716 @item set print demangle
3717 @itemx set print demangle on
3718 @kindex set print demangle
3719 Print C++ names in their source form rather than in the mangled form
3720 in which they are passed to the assembler and linker for type-safe linkage.
3723 @item show print demangle
3724 @kindex show print demangle
3725 Show whether C++ names will be printed in mangled or demangled form.
3727 @item set print asm-demangle
3728 @itemx set print asm-demangle on
3729 @kindex set print asm-demangle
3730 Print C++ names in their source form rather than their mangled form, even
3731 in assembler code printouts such as instruction disassemblies.
3734 @item show print asm-demangle
3735 @kindex show print asm-demangle
3736 Show whether C++ names in assembly listings will be printed in mangled
3739 @item set print object
3740 @itemx set print object on
3741 @kindex set print object
3742 When displaying a pointer to an object, identify the @emph{actual}
3743 (derived) type of the object rather than the @emph{declared} type, using
3744 the virtual function table.
3746 @item set print object off
3747 Display only the declared type of objects, without reference to the
3748 virtual function table. This is the default setting.
3750 @item show print object
3751 @kindex show print object
3752 Show whether actual, or declared, object types will be displayed.
3754 @item set print vtbl
3755 @itemx set print vtbl on
3756 @kindex set print vtbl
3757 Pretty print C++ virtual function tables. The default is off.
3759 @item set print vtbl off
3760 Do not pretty print C++ virtual function tables.
3762 @item show print vtbl
3763 @kindex show print vtbl
3764 Show whether C++ virtual function tables are pretty printed, or not.
3768 @node Value History, Convenience Vars, Print Settings, Data
3769 @section Value History
3771 @cindex value history
3772 Values printed by the @code{print} command are saved in _GDBN__'s @dfn{value
3773 history} so that you can refer to them in other expressions. Values are
3774 kept until the symbol table is re-read or discarded (for example with
3775 the @code{file} or @code{symbol-file} commands). When the symbol table
3776 changes, the value history is discarded, since the values may contain
3777 pointers back to the types defined in the symbol table.
3781 @cindex history number
3782 The values printed are given @dfn{history numbers} for you to refer to them
3783 by. These are successive integers starting with one. @code{print} shows you
3784 the history number assigned to a value by printing @samp{$@var{num} = }
3785 before the value; here @var{num} is the history number.
3787 To refer to any previous value, use @samp{$} followed by the value's
3788 history number. The way @code{print} labels its output is designed to
3789 remind you of this. Just @code{$} refers to the most recent value in
3790 the history, and @code{$$} refers to the value before that.
3791 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
3792 is the value just prior to @code{$$}, @code{$$1} is equivalent to
3793 @code{$$}, and @code{$$0} is equivalent to @code{$}.
3795 For example, suppose you have just printed a pointer to a structure and
3796 want to see the contents of the structure. It suffices to type
3802 If you have a chain of structures where the component @code{next} points
3803 to the next one, you can print the contents of the next one with this:
3810 You can print successive links in the chain by repeating this
3811 command---which you can do by just typing @key{RET}.
3813 Note that the history records values, not expressions. If the value of
3814 @code{x} is 4 and you type these commands:
3822 then the value recorded in the value history by the @code{print} command
3823 remains 4 even though the value of @code{x} has changed.
3828 Print the last ten values in the value history, with their item numbers.
3829 This is like @samp{p@ $$9} repeated ten times, except that @code{show
3830 values} does not change the history.
3832 @item show values @var{n}
3833 Print ten history values centered on history item number @var{n}.
3836 Print ten history values just after the values last printed. If no more
3837 values are available, produces no display.
3840 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
3841 same effect as @samp{show values +}.
3843 @node Convenience Vars, Registers, Value History, Data
3844 @section Convenience Variables
3846 @cindex convenience variables
3847 _GDBN__ provides @dfn{convenience variables} that you can use within
3848 _GDBN__ to hold on to a value and refer to it later. These variables
3849 exist entirely within _GDBN__; they are not part of your program, and
3850 setting a convenience variable has no direct effect on further execution
3851 of your program. That's why you can use them freely.
3853 Convenience variables are prefixed with @samp{$}. Any name preceded by
3854 @samp{$} can be used for a convenience variable, unless it is one of
3855 the predefined machine-specific register names (@pxref{Registers}).
3856 (Value history references, in contrast, are @emph{numbers} preceded
3857 by @samp{$}. @xref{Value History}.)
3859 You can save a value in a convenience variable with an assignment
3860 expression, just as you would set a variable in your program. Example:
3863 set $foo = *object_ptr
3867 would save in @code{$foo} the value contained in the object pointed to by
3870 Using a convenience variable for the first time creates it; but its value
3871 is @code{void} until you assign a new value. You can alter the value with
3872 another assignment at any time.
3874 Convenience variables have no fixed types. You can assign a convenience
3875 variable any type of value, including structures and arrays, even if
3876 that variable already has a value of a different type. The convenience
3877 variable, when used as an expression, has the type of its current value.
3880 @item show convenience
3881 @kindex show convenience
3882 Print a list of convenience variables used so far, and their values.
3883 Abbreviated @code{show con}.
3886 One of the ways to use a convenience variable is as a counter to be
3887 incremented or a pointer to be advanced. For example, to print
3888 a field from successive elements of an array of structures:
3892 print bar[$i++]->contents
3893 @i{@dots{} repeat that command by typing @key{RET}.}
3896 Some convenience variables are created automatically by _GDBN__ and given
3897 values likely to be useful.
3902 The variable @code{$_} is automatically set by the @code{x} command to
3903 the last address examined (@pxref{Memory}). Other commands which
3904 provide a default address for @code{x} to examine also set @code{$_}
3905 to that address; these commands include @code{info line} and @code{info
3906 breakpoint}. The type of @code{$_} is @code{void *} except when set by the
3907 @code{x} command, in which case it is a pointer to the type of @code{$__}.
3911 The variable @code{$__} is automatically set by the @code{x} command
3912 to the value found in the last address examined. Its type is chosen
3913 to match the format in which the data was printed.
3916 @node Registers, Floating Point Hardware, Convenience Vars, Data
3920 You can refer to machine register contents, in expressions, as variables
3921 with names starting with @samp{$}. The names of registers are different
3922 for each machine; use @code{info registers} to see the names used on
3926 @item info registers
3927 @kindex info registers
3928 Print the names and values of all registers except floating-point
3929 registers (in the selected stack frame).
3931 @item info all-registers
3932 @kindex info all-registers
3933 @cindex floating point registers
3934 Print the names and values of all registers, including floating-point
3937 @item info registers @var{regname}
3938 Print the relativized value of register @var{regname}. @var{regname}
3939 may be any register name valid on the machine you are using, with
3940 or without the initial @samp{$}.
3943 _GDBN__ has four ``standard'' register names that are available (in
3944 expressions) on most machines---whenever they don't conflict with an
3945 architecture's canonical mnemonics for registers. The register names
3946 @code{$pc} and @code{$sp} are used for the program counter register and
3947 the stack pointer. @code{$fp} is used for a register that contains a
3948 pointer to the current stack frame, and @code{$ps} is used for a
3949 register that contains the processor status. For example,
3950 you could print the program counter in hex with
3956 or print the instruction to be executed next with
3962 or add four to the stack pointer @footnote{This is a way of removing one
3963 word from the stack, on machines where stacks grow downward in memory
3964 (most machines, nowadays). This assumes that the innermost stack frame
3965 is selected; setting @code{$sp} is not allowed when other stack frames
3966 are selected. To pop entire frames off the stack, regardless of
3967 machine architecture, use @code{return}; @pxref{Returning}.} with
3972 Whenever possible, these four standard register names are available on
3973 your machine even though the machine has different canonical mnemonics,
3974 so long as there is no conflict. The @code{info registers} command
3975 shows the canonical names. For example, on the SPARC, @code{info
3976 registers} displays the processor status register as @code{$psr} but you
3977 can also refer to it as @code{$ps}.
3979 _GDBN__ always considers the contents of an ordinary register as an
3980 integer when the register is examined in this way. Some machines have
3981 special registers which can hold nothing but floating point; these
3982 registers are considered to have floating point values. There is no way
3983 to refer to the contents of an ordinary register as floating point value
3984 (although you can @emph{print} it as a floating point value with
3985 @samp{print/f $@var{regname}}).
3987 Some registers have distinct ``raw'' and ``virtual'' data formats. This
3988 means that the data format in which the register contents are saved by
3989 the operating system is not the same one that your program normally
3990 sees. For example, the registers of the 68881 floating point
3991 coprocessor are always saved in ``extended'' (raw) format, but all C
3992 programs expect to work with ``double'' (virtual) format. In such
3993 cases, _GDBN__ normally works with the virtual format only (the format that
3994 makes sense for your program), but the @code{info registers} command
3995 prints the data in both formats.
3997 Normally, register values are relative to the selected stack frame
3998 (@pxref{Selection}). This means that you get the value that the
3999 register would contain if all stack frames farther in were exited and
4000 their saved registers restored. In order to see the true contents of
4001 hardware registers, you must select the innermost frame (with
4004 However, _GDBN__ must deduce where registers are saved, from the machine
4005 code generated by your compiler. If some registers are not saved, or if
4006 _GDBN__ is unable to locate the saved registers, the selected stack
4007 frame will make no difference.
4009 @node Floating Point Hardware, , Registers, Data
4010 @section Floating Point Hardware
4011 @cindex floating point
4012 Depending on the host machine architecture, _GDBN__ may be able to give
4013 you more information about the status of the floating point hardware.
4018 If available, provides hardware-dependent information about the floating
4019 point unit. The exact contents and layout vary depending on the
4020 floating point chip.
4022 @c FIXME: this is a cop-out. Try to get examples, explanations. Only
4023 @c FIXME...supported currently on arm's and 386's. Mark properly with
4024 @c FIXME... m4 macros to isolate general statements from hardware-dep,
4025 @c FIXME... at that point.
4027 @node Languages, Symbols, Data, Top
4028 @chapter Using _GDBN__ with Different Languages
4031 Although programming languages generally have common aspects, they are
4032 rarely expressed in the same manner. For instance, in ANSI C,
4033 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
4034 Modula-2, it is accomplished by @code{p^}. Values can also be
4035 represented (and displayed) differently. Hex numbers in C are written
4036 like @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
4038 @cindex working language
4039 Language-specific information is built into _GDBN__ for some languages,
4040 allowing you to express operations like the above in the program's
4041 native language, and allowing _GDBN__ to output values in a manner
4042 consistent with the syntax of the program's native language. The
4043 language you use to build expressions, called the @dfn{working
4044 language}, can be selected manually, or _GDBN__ can set it
4048 * Setting:: Switching between source languages
4049 * Show:: Displaying the language
4050 * Checks:: Type and Range checks
4051 * Support:: Supported languages
4054 @node Setting, Show, Languages, Languages
4055 @section Switching between source languages
4057 There are two ways to control the working language---either have _GDBN__
4058 set it automatically, or select it manually yourself. You can use the
4059 @code{set language} command for either purpose. On startup, _GDBN__
4060 defaults to setting the language automatically.
4063 * Manually:: Setting the working language manually
4064 * Automatically:: Having _GDBN__ infer the source language
4067 @node Manually, Automatically, Setting, Setting
4068 @subsection Setting the working language
4070 @kindex set language
4071 To set the language, issue the command @samp{set language @var{lang}},
4072 where @var{lang} is the name of a language: @code{c} or @code{modula-2}.
4073 For a list of the supported languages, type @samp{set language}.
4075 Setting the language manually prevents _GDBN__ from updating the working
4076 language automatically. This can lead to confusion if you try
4077 to debug a program when the working language is not the same as the
4078 source language, when an expression is acceptable to both
4079 languages---but means different things. For instance, if the current
4080 source file were written in C, and _GDBN__ was parsing Modula-2, a
4088 might not have the effect you intended. In C, this means to add
4089 @code{b} and @code{c} and place the result in @code{a}. The result
4090 printed would be the value of @code{a}. In Modula-2, this means to compare
4091 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
4093 If you allow _GDBN__ to set the language automatically, then
4094 you can count on expressions evaluating the same way in your debugging
4095 session and in your program.
4097 @node Automatically, , Manually, Setting
4098 @subsection Having _GDBN__ infer the source language
4100 To have _GDBN__ set the working language automatically, use @samp{set
4101 language local} or @samp{set language auto}. _GDBN__ then infers the
4102 language that a program was written in by looking at the name of its
4103 source files, and examining their extensions:
4107 Modula-2 source file
4111 C or C++ source file.
4114 This information is recorded for each function or procedure in a source
4115 file. When your program stops in a frame (usually by encountering a
4116 breakpoint), _GDBN__ sets the working language to the language recorded
4117 for the function in that frame. If the language for a frame is unknown
4118 (that is, if the function or block corresponding to the frame was
4119 defined in a source file that does not have a recognized extension), the
4120 current working language is not changed, and _GDBN__ issues a warning.
4122 This may not seem necessary for most programs, which are written
4123 entirely in one source language. However, program modules and libraries
4124 written in one source language can be used by a main program written in
4125 a different source language. Using @samp{set language auto} in this
4126 case frees you from having to set the working language manually.
4128 @node Show, Checks, Setting, Languages
4129 @section Displaying the language
4131 The following commands will help you find out which language is the
4132 working language, and also what language source files were written in.
4134 @kindex show language
4139 Display the current working language. This is the
4140 language you can use with commands such as @code{print} to
4141 build and compute expressions that may involve variables in the program.
4144 Among the other information listed here (@pxref{Frame Info,,Information
4145 about a Frame}) is the source language for this frame. This is the
4146 language that will become the working language if you ever use an
4147 identifier that is in this frame.
4150 Among the other information listed here (@pxref{Symbols,,Examining the
4151 Symbol Table}) is the source language of this source file.
4155 @node Checks, Support, Show, Languages
4156 @section Type and range Checking
4159 @emph{Warning:} In this release, the _GDBN__ commands for type and range
4160 checking are included, but they do not yet have any effect. This
4161 section documents the intended facilities.
4163 @c FIXME remove warning when type/range code added
4165 Some languages are designed to guard you against making seemingly common
4166 errors through a series of compile- and run-time checks. These include
4167 checking the type of arguments to functions and operators, and making
4168 sure mathematical overflows are caught at run time. Checks such as
4169 these help to ensure a program's correctness once it has been compiled
4170 by eliminating type mismatches, and providing active checks for range
4171 errors when the program is running.
4173 _GDBN__ can check for conditions like the above if you wish.
4174 Although _GDBN__ will not check the statements in your program, it
4175 can check expressions entered directly into _GDBN__ for evaluation via
4176 the @code{print} command, for example. As with the working language,
4177 _GDBN__ can also decide whether or not to check automatically based on
4178 the source language of the program being debugged.
4179 @xref{Support,,Supported Languages}, for the default settings
4180 of supported languages.
4183 * Type Checking:: An overview of type checking
4184 * Range Checking:: An overview of range checking
4187 @cindex type checking
4188 @cindex checks, type
4189 @node Type Checking, Range Checking, Checks, Checks
4190 @subsection An overview of type checking
4192 Some languages, such as Modula-2, are strongly typed, meaning that the
4193 arguments to operators and functions have to be of the correct type,
4194 otherwise an error occurs. These checks prevent type mismatch
4195 errors from ever causing any run-time problems. For example,
4202 The second example fails because the @code{CARDINAL} 1 is not
4203 type-compatible with the @code{REAL} 2.3.
4205 For expressions you use in _GDBN__ commands, you can tell the _GDBN__
4206 type checker to skip checking; to treat any mismatches as errors and
4207 abandon the expression; or only issue warnings when type mismatches
4208 occur, but evaluate the expression anyway. When you choose the last of
4209 these, _GDBN__ evaluates expressions like the second example above, but
4210 also issues a warning.
4212 Even though you may turn type checking off, other type-based reasons may
4213 prevent _GDBN__ from evaluating an expression. For instance, _GDBN__ does not
4214 know how to add an @code{int} and a @code{struct foo}. These particular
4215 type errors have nothing to do with the language in use, and usually
4216 arise from expressions, such as the one described above, which make
4217 little sense to evaluate anyway.
4219 Each language defines to what degree it is strict about type. For
4220 instance, both Modula-2 and C require the arguments to arithmetical
4221 operators to be numbers. In C, enumerated types and pointers can be
4222 represented as numbers, so that they are valid arguments to mathematical
4223 operators. @xref{Support,,Supported Languages}, for futher
4224 details on specific languages.
4226 _GDBN__ provides some additional commands for controlling the type checker:
4229 @kindex set check type
4230 @kindex show check type
4232 @item set check type auto
4233 Set type checking on or off based on the current working language.
4234 @xref{Support,,Supported Languages}, for the default settings for
4237 @item set check type on
4238 @itemx set check type off
4239 Set type checking on or off, overriding the default setting for the
4240 current working language. Issue a warning if the setting does not
4241 match the language's default. If any type mismatches occur in
4242 evaluating an expression while typechecking is on, _GDBN__ prints a
4243 message and aborts evaluation of the expression.
4245 @item set check type warn
4246 Cause the type checker to issue warnings, but to always attempt to
4247 evaluate the expression. Evaluating the expression may still
4248 be impossible for other reasons. For example, _GDBN__ cannot add
4249 numbers and structures.
4252 Show the current setting of the type checker, and whether or not _GDBN__ is
4253 setting it automatically.
4256 @cindex range checking
4257 @cindex checks, range
4258 @node Range Checking, , Type Checking, Checks
4259 @subsection An overview of Range Checking
4261 In some languages (such as Modula-2), it is an error to exceed the
4262 bounds of a type; this is enforced with run-time checks. Such range
4263 checking is meant to ensure program correctness by making sure
4264 computations do not overflow, or indices on an array element access do
4265 not exceed the bounds of the array.
4267 For expressions you use in _GDBN__ commands, you can tell _GDBN__ to
4268 ignore range errors; to always treat them as errors and abandon the
4269 expression; or to issue warnings when a range error occurs but evaluate
4270 the expression anyway.
4272 A range error can result from numerical overflow, from exceeding an
4273 array index bound, or when you type in a constant that is not a member
4274 of any type. Some languages, however, do not treat overflows as an
4275 error. In many implementations of C, mathematical overflow causes the
4276 result to ``wrap around'' to lower values---for example, if @var{m} is
4277 the largest integer value, and @var{s} is the smallest, then
4279 @var{m} + 1 @result{} @var{s}
4282 This, too, is specific to individual languages, and in some cases
4283 specific to individual compilers or machines. @xref{Support,,
4284 Supported Languages}, for further details on specific languages.
4286 _GDBN__ provides some additional commands for controlling the range checker:
4289 @kindex set check range
4290 @kindex show check range
4292 @item set check range auto
4293 Set range checking on or off based on the current working language.
4294 @xref{Support,,Supported Languages}, for the default settings for
4297 @item set check range on
4298 @itemx set check range off
4299 Set range checking on or off, overriding the default setting for the
4300 current working language. A warning is issued if the setting does not
4301 match the language's default. If a range error occurs, then a message
4302 is printed and evaluation of the expression is aborted.
4304 @item set check range warn
4305 Output messages when the _GDBN__ range checker detects a range error,
4306 but attempt to evaluate the expression anyway. Evaluating the
4307 expression may still be impossible for other reasons, such as accessing
4308 memory that the process does not own (a typical example from many UNIX
4312 Show the current setting of the range checker, and whether or not it is
4313 being set automatically by _GDBN__.
4316 @node Support, , Checks, Languages
4317 @section Supported Languages
4319 _GDBN__ _GDB_VN__ supports C, C++, and Modula-2. The syntax for C and C++ is
4320 so closely related that _GDBN__ does not distinguish the two. Some
4321 _GDBN__ features may be used in expressions regardless of the language
4322 you use: the _GDBN__ @code{@@} and @code{::} operators, and the
4323 @samp{@{type@}addr} construct (@pxref{Expressions}) can be used with the constructs of
4324 any of the supported languages.
4326 The following sections detail to what degree each of these
4327 source languages is supported by _GDBN__. These sections are
4328 not meant to be language tutorials or references, but serve only as a
4329 reference guide to what the _GDBN__ expression parser will accept, and
4330 what input and output formats should look like for different languages.
4331 There are many good books written on each of these languages; please
4332 look to these for a language reference or tutorial.
4336 * Modula-2:: Modula-2
4339 @node C, Modula-2, Support, Support
4340 @subsection C and C++
4343 @cindex expressions in C or C++
4344 Since C and C++ are so closely related, _GDBN__ does not distinguish
4345 between them when interpreting the expressions recognized in _GDBN__
4351 The C++ debugging facilities are jointly implemented by the GNU C++
4352 compiler and _GDBN__. Therefore, to debug your C++ code effectively,
4353 you must compile your C++ programs with the GNU C++ compiler,
4358 * C Operators:: C and C++ Operators
4359 * C Constants:: C and C++ Constants
4360 * Cplusplus expressions:: C++ Expressions
4361 * C Defaults:: Default settings for C and C++
4362 * C Checks:: C and C++ Type and Range Checks
4363 * Debugging C:: _GDBN__ and C
4364 * Debugging C plus plus:: Special features for C++
4367 @cindex C and C++ operators
4368 @node C Operators, C Constants, C, C
4369 @subsubsection C and C++ Operators
4371 Operators must be defined on values of specific types. For instance,
4372 @code{+} is defined on numbers, but not on structures. Operators are
4373 often defined on groups of types. For the purposes of C and C++, the
4374 following definitions hold:
4378 @emph{Integral types} include @code{int} with any of its storage-class
4379 specifiers, @code{char}, and @code{enum}s.
4382 @emph{Floating-point types} include @code{float} and @code{double}.
4385 @emph{Pointer types} include all types defined as @code{(@var{type}
4389 @emph{Scalar types} include all of the above.
4394 The following operators are supported. They are listed here
4395 in order of increasing precedence:
4400 The comma or sequencing operator. Expressions in a comma-separated list
4401 are evaluated from left to right, with the result of the entire
4402 expression being the last expression evaluated.
4405 Assignment. The value of an assignment expression is the value
4406 assigned. Defined on scalar types.
4409 Used in an expression of the form @var{a} @var{op}@code{=} @var{b}, and
4410 translated to @var{a} @code{=} @var{a op b}. @var{op}@code{=} and
4411 @code{=} have the same precendence. @var{op} is any one of the
4412 operators @code{|}, @code{^}, @code{&}, @code{<<}, @code{>>}, @code{+},
4413 @code{-}, @code{*}, @code{/}, @code{%}.
4416 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
4417 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
4421 Logical OR. Defined on integral types.
4424 Logical AND. Defined on integral types.
4427 Bitwise OR. Defined on integral types.
4430 Bitwise exclusive-OR. Defined on integral types.
4433 Bitwise AND. Defined on integral types.
4436 Equality and inequality. Defined on scalar types. The value of these
4437 expressions is 0 for false and non-zero for true.
4439 @item <@r{, }>@r{, }<=@r{, }>=
4440 Less than, greater than, less than or equal, greater than or equal.
4441 Defined on scalar types. The value of these expressions is 0 for false
4442 and non-zero for true.
4445 left shift, and right shift. Defined on integral types.
4448 The _GDBN__ ``artificial array'' operator (@pxref{Expressions}).
4451 Addition and subtraction. Defined on integral types, floating-point types and
4454 @item *@r{, }/@r{, }%
4455 Multiplication, division, and modulus. Multiplication and division are
4456 defined on integral and floating-point types. Modulus is defined on
4460 Increment and decrement. When appearing before a variable, the
4461 operation is performed before the variable is used in an expression;
4462 when appearing after it, the variable's value is used before the
4463 operation takes place.
4466 Pointer dereferencing. Defined on pointer types. Same precedence as
4470 Address operator. Defined on variables. Same precedence as @code{++}.
4473 Negative. Defined on integral and floating-point types. Same
4474 precedence as @code{++}.
4477 Logical negation. Defined on integral types. Same precedence as
4481 Bitwise complement operator. Defined on integral types. Same precedence as
4485 Structure member, and pointer-to-structure member. For convenience,
4486 _GDBN__ regards the two as equivalent, choosing whether to dereference a
4487 pointer based on the stored type information.
4488 Defined on @code{struct}s and @code{union}s.
4491 Array indexing. @code{@var{a}[@var{i}]} is defined as
4492 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
4495 Function parameter list. Same precedence as @code{->}.
4498 C++ scope resolution operator. Defined on
4499 @code{struct}, @code{union}, and @code{class} types.
4502 The _GDBN__ scope operator (@pxref{Expressions}). Same precedence as
4503 @code{::}, above. _1__
4506 @cindex C and C++ constants
4507 @node C Constants, Cplusplus expressions, C Operators, C
4508 @subsubsection C and C++ Constants
4510 _GDBN__ allows you to express the constants of C and C++ in the
4516 Integer constants are a sequence of digits. Octal constants are
4517 specified by a leading @samp{0} (ie. zero), and hexadecimal constants by
4518 a leading @samp{0x} or @samp{0X}. Constants may also end with an
4519 @samp{l}, specifying that the constant should be treated as a
4523 Floating point constants are a sequence of digits, followed by a decimal
4524 point, followed by a sequence of digits, and optionally followed by an
4525 exponent. An exponent is of the form:
4526 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
4527 sequence of digits. The @samp{+} is optional for positive exponents.
4530 Enumerated constants consist of enumerated identifiers, or their
4531 integral equivalents.
4534 Character constants are a single character surrounded by single quotes
4535 (@code{'}), or a number---the ordinal value of the corresponding character
4536 (usually its @sc{ASCII} value). Within quotes, the single character may
4537 be represented by a letter or by @dfn{escape sequences}, which are of
4538 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
4539 of the character's ordinal value; or of the form @samp{\@var{x}}, where
4540 @samp{@var{x}} is a predefined special character---for example,
4541 @samp{\n} for newline.
4544 String constants are a sequence of character constants surrounded
4545 by double quotes (@code{"}).
4548 Pointer constants are an integral value.
4553 @node Cplusplus expressions, C Defaults, C Constants, C
4554 @subsubsection C++ Expressions
4556 @cindex expressions in C++
4557 _GDBN__'s expression handling has the following extensions to
4558 interpret a significant subset of C++ expressions:
4562 @cindex member functions
4564 Member function calls are allowed; you can use expressions like
4566 count = aml->GetOriginal(x, y)
4570 @cindex namespace in C++
4572 While a member function is active (in the selected stack frame), your
4573 expressions have the same namespace available as the member function;
4574 that is, _GDBN__ allows implicit references to the class instance
4575 pointer @code{this} following the same rules as C++.
4577 @cindex call overloaded functions
4578 @cindex type conversions in C++
4580 You can call overloaded functions; _GDBN__ will resolve the function
4581 call to the right definition, with one restriction---you must use
4582 arguments of the type required by the function that you want to call.
4583 _GDBN__ will not perform conversions requiring constructors or
4584 user-defined type operators.
4586 @cindex reference declarations
4588 _GDBN__ understands variables declared as C++ references; you can use them in
4589 expressions just as you do in C++ source---they are automatically
4592 In the parameter list shown when _GDBN__ displays a frame, the values of
4593 reference variables are not displayed (unlike other variables); this
4594 avoids clutter, since references are often used for large structures.
4595 The @emph{address} of a reference variable is always shown, unless
4596 you've specified @samp{set print address off}.
4600 _GDBN__ supports the C++ name resolution operator @code{::}---your
4601 expressions can use it just as expressions in your program do. Since
4602 one scope may be defined in another, you can use @code{::} repeatedly if
4603 necessary, for example in an expression like
4604 @samp{@var{scope1}::@var{scope2}::@var{name}}. _GDBN__ also allows
4605 resolving name scope by reference to source files, in both C and C++
4606 debugging; @pxref{Variables}.
4611 @node C Defaults, C Checks, Cplusplus expressions, C
4612 @subsubsection C and C++ Defaults
4613 @cindex C and C++ defaults
4615 If you allow _GDBN__ to set type and range checking automatically, they
4616 both default to @code{off} whenever the working language changes to
4617 C/C++. This happens regardless of whether you, or _GDBN__,
4618 selected the working language.
4620 If you allow _GDBN__ to set the language automatically, it sets the
4621 working language to C/C++ on entering code compiled from a source file
4622 whose name ends with @file{.c} or @file{.cc}.
4623 @xref{Automatically,,Having _GDBN__ infer the source language}, for
4626 @node C Checks, Debugging C, C Defaults, C
4627 @subsubsection C and C++ Type and Range Checks
4628 @cindex C and C++ checks
4631 @emph{Warning:} in this release, _GDBN__ does not yet perform type or
4634 @c FIXME remove warning when type/range checks added
4636 By default, when _GDBN__ parses C or C++ expressions, type checking
4637 is not used. However, if you turn type checking on, _GDBN__ will
4638 consider two variables type equivalent if:
4642 The two variables are structured and have the same structure, union, or
4646 Two two variables have the same type name, or types that have been
4647 declared equivalent through @code{typedef}.
4650 @c leaving this out because neither J Gilmore nor R Pesch understand it.
4653 The two @code{struct}, @code{union}, or @code{enum} variables are
4654 declared in the same declaration. (Note: this may not be true for all C
4660 Range checking, if turned on, is done on mathematical operations. Array
4661 indices are not checked, since they are often used to index a pointer
4662 that is not itself an array.
4664 @node Debugging C, Debugging C plus plus, C Checks, C
4665 @subsubsection _GDBN__ and C
4667 The @code{set print union} and @code{show print union} commands apply to
4668 the @code{union} type. When set to @samp{on}, any @code{union} that is
4669 inside a @code{struct} or @code{class} will also be printed.
4670 Otherwise, it will appear as @samp{@{...@}}.
4672 The @code{@@} operator aids in the debugging of dynamic arrays, formed
4673 with pointers and a memory allocation function. (@pxref{Expressions})
4675 @node Debugging C plus plus, , Debugging C, C
4676 @subsubsection _GDBN__ Commands for C++
4678 @cindex commands for C++
4679 Some _GDBN__ commands are particularly useful with C++, and some are
4680 designed specifically for use with C++. Here is a summary:
4683 @cindex break in overloaded functions
4684 @item @r{breakpoint menus}
4685 When you want a breakpoint in a function whose name is overloaded,
4686 _GDBN__'s breakpoint menus help you specify which function definition
4687 you want. @xref{Breakpoint Menus}.
4689 @cindex overloading in C++
4690 @item rbreak @var{regex}
4691 Setting breakpoints using regular expressions is helpful for setting
4692 breakpoints on overloaded functions that are not members of any special
4696 @cindex C++ exception handling
4697 @item catch @var{exceptions}
4699 Debug C++ exception handling using these commands. @xref{Exception Handling}.
4702 @item ptype @var{typename}
4703 Print inheritance relationships as well as other information for type
4707 @cindex C++ symbol display
4708 @item set print demangle
4709 @itemx show print demangle
4710 @itemx set print asm-demangle
4711 @itemx show print asm-demangle
4712 Control whether C++ symbols display in their source form, both when
4713 displaying code as C++ source and when displaying disassemblies.
4714 @xref{Print Settings}.
4716 @item set print object
4717 @itemx show print object
4718 Choose whether to print derived (actual) or declared types of objects.
4719 @xref{Print Settings}.
4721 @item set print vtbl
4722 @itemx show print vtbl
4723 Control the format for printing virtual function tables.
4724 @xref{Print Settings}.
4729 @node Modula-2, , C, Support
4730 @subsection Modula-2
4733 The extensions made to _GDBN__ to support Modula-2 support output
4734 from the GNU Modula-2 compiler (which is currently being developed).
4735 Other Modula-2 compilers are not currently supported, and attempting to
4736 debug executables produced by them will most likely result in an error
4737 as _GDBN__ reads in the executable's symbol table.
4739 @cindex expressions in Modula-2
4741 * M2 Operators:: Built-in operators
4742 * Builtin Func/Proc:: Built-in Functions and Procedures
4743 * M2 Constants:: Modula-2 Constants
4744 * M2 Defaults:: Default settings for Modula-2
4745 * Deviations:: Deviations from standard Modula-2
4746 * M2 Checks:: Modula-2 Type and Range Checks
4747 * M2 Scope:: The scope operators @code{::} and @code{.}
4748 * GDB/M2:: _GDBN__ and Modula-2
4751 @node M2 Operators, Builtin Func/Proc, Modula-2, Modula-2
4752 @subsubsection Operators
4753 @cindex Modula-2 operators
4755 Operators must be defined on values of specific types. For instance,
4756 @code{+} is defined on numbers, but not on structures. Operators are
4757 often defined on groups of types. For the purposes of Modula-2, the
4758 following definitions hold:
4763 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
4767 @emph{Character types} consist of @code{CHAR} and its subranges.
4770 @emph{Floating-point types} consist of @code{REAL}.
4773 @emph{Pointer types} consist of anything declared as @code{POINTER TO
4777 @emph{Scalar types} consist of all of the above.
4780 @emph{Set types} consist of @code{SET}s and @code{BITSET}s.
4783 @emph{Boolean types} consist of @code{BOOLEAN}.
4788 The following operators are supported, and appear in order of
4789 increasing precedence:
4794 Function argument or array index separator.
4797 Assignment. The value of @var{var} @code{:=} @var{value} is
4801 Less than, greater than on integral, floating-point, or enumerated
4805 Less than, greater than, less than or equal to, greater than or equal to
4806 on integral, floating-point and enumerated types, or set inclusion on
4807 set types. Same precedence as @code{<}.
4809 @item =@r{, }<>@r{, }#
4810 Equality and two ways of expressing inequality, valid on scalar types.
4811 Same precedence as @code{<}. In _GDBN__ scripts, only @code{<>} is
4812 available for inequality, since @code{#} conflicts with the script
4816 Set membership. Defined on set types and the types of their members.
4817 Same precedence as @code{<}.
4820 Boolean disjunction. Defined on boolean types.
4823 Boolean conjuction. Defined on boolean types.
4826 The _GDBN__ ``artificial array'' operator (@pxref{Expressions}).
4829 Addition and subtraction on integral and floating-point types, or union
4830 and difference on set types.
4833 Multiplication on integral and floating-point types, or set intersection
4837 Division on floating-point types, or symmetric set difference on set
4838 types. Same precedence as @code{*}.
4841 Integer division and remainder. Defined on integral types. Same
4842 precedence as @code{*}.
4845 Negative. Defined on @code{INTEGER}s and @code{REAL}s.
4848 Pointer dereferencing. Defined on pointer types.
4851 Boolean negation. Defined on boolean types. Same precedence as
4855 @code{RECORD} field selector. Defined on @code{RECORD}s. Same
4856 precedence as @code{^}.
4859 Array indexing. Defined on @code{ARRAY}s. Same precedence as @code{^}.
4862 Procedure argument list. Defined on @code{PROCEDURE}s. Same precedence
4866 _GDBN__ and Modula-2 scope operators.
4871 @emph{Warning:} Sets and their operations are not yet supported, so _GDBN__
4872 will treat the use of the operator @code{IN}, or the use of operators
4873 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
4874 @code{<=}, and @code{>=} on sets as an error.
4878 @cindex Modula-2 builtins
4879 @node Builtin Func/Proc, M2 Constants, M2 Operators, Modula-2
4880 @subsubsection Built-in Functions and Procedures
4882 Modula-2 also makes available several built-in procedures and functions.
4883 In describing these, the following metavariables are used:
4888 represents an @code{ARRAY} variable.
4891 represents a @code{CHAR} constant or variable.
4894 represents a variable or constant of integral type.
4897 represents an identifier that belongs to a set. Generally used in the
4898 same function with the metavariable @var{s}. The type of @var{s} should
4899 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}.
4902 represents a variable or constant of integral or floating-point type.
4905 represents a variable or constant of floating-point type.
4911 represents a variable.
4914 represents a variable or constant of one of many types. See the
4915 explanation of the function for details.
4919 All Modula-2 built-in procedures also return a result, described below.
4923 Returns the absolute value of @var{n}.
4926 If @var{c} is a lower case letter, it returns its upper case
4927 equivalent, otherwise it returns its argument
4930 Returns the character whose ordinal value is @var{i}.
4933 Decrements the value in the variable @var{v}. Returns the new value.
4935 @item DEC(@var{v},@var{i})
4936 Decrements the value in the variable @var{v} by @var{i}. Returns the
4939 @item EXCL(@var{m},@var{s})
4940 Removes the element @var{m} from the set @var{s}. Returns the new
4943 @item FLOAT(@var{i})
4944 Returns the floating point equivalent of the integer @var{i}.
4947 Returns the index of the last member of @var{a}.
4950 Increments the value in the variable @var{v}. Returns the new value.
4952 @item INC(@var{v},@var{i})
4953 Increments the value in the variable @var{v} by @var{i}. Returns the
4956 @item INCL(@var{m},@var{s})
4957 Adds the element @var{m} to the set @var{s} if it is not already
4958 there. Returns the new set.
4961 Returns the maximum value of the type @var{t}.
4964 Returns the minimum value of the type @var{t}.
4967 Returns boolean TRUE if @var{i} is an odd number.
4970 Returns the ordinal value of its argument. For example, the ordinal
4971 value of a character is its ASCII value (on machines supporting the
4972 ASCII character set). @var{x} must be of an ordered type, which include
4973 integral, character and enumerated types.
4976 Returns the size of its argument. @var{x} can be a variable or a type.
4978 @item TRUNC(@var{r})
4979 Returns the integral part of @var{r}.
4981 @item VAL(@var{t},@var{i})
4982 Returns the member of the type @var{t} whose ordinal value is @var{i}.
4986 @emph{Warning:} Sets and their operations are not yet supported, so
4987 _GDBN__ will treat the use of procedures @code{INCL} and @code{EXCL} as
4991 @cindex Modula-2 constants
4992 @node M2 Constants, M2 Defaults, Builtin Func/Proc, Modula-2
4993 @subsubsection Constants
4995 _GDBN__ allows you to express the constants of Modula-2 in the following
5001 Integer constants are simply a sequence of digits. When used in an
5002 expression, a constant is interpreted to be type-compatible with the
5003 rest of the expression. Hexadecimal integers are specified by a
5004 trailing @samp{H}, and octal integers by a trailing @samp{B}.
5007 Floating point constants appear as a sequence of digits, followed by a
5008 decimal point and another sequence of digits. An optional exponent can
5009 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
5010 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
5011 digits of the floating point constant must be valid decimal (base 10)
5015 Character constants consist of a single character enclosed by a pair of
5016 like quotes, either single (@code{'}) or double (@code{"}). They may
5017 also be expressed by their ordinal value (their ASCII value, usually)
5018 followed by a @samp{C}.
5021 String constants consist of a sequence of characters enclosed by a pair
5022 of like quotes, either single (@code{'}) or double (@code{"}). Escape
5023 sequences in the style of C are also allowed. @xref{C Constants}, for a
5024 brief explanation of escape sequences.
5027 Enumerated constants consist of an enumerated identifier.
5030 Boolean constants consist of the identifiers @code{TRUE} and
5034 Pointer constants consist of integral values only.
5037 Set constants are not yet supported.
5041 @node M2 Defaults, Deviations, M2 Constants, Modula-2
5042 @subsubsection Modula-2 Defaults
5043 @cindex Modula-2 defaults
5045 If type and range checking are set automatically by _GDBN__, they
5046 both default to @code{on} whenever the working language changes to
5047 Modula-2. This happens regardless of whether you, or _GDBN__,
5048 selected the working language.
5050 If you allow _GDBN__ to set the language automatically, then entering
5051 code compiled from a file whose name ends with @file{.mod} will set the
5052 working language to Modula-2. @xref{Automatically,,Having _GDBN__ set
5053 the language automatically}, for further details.
5055 @node Deviations, M2 Checks, M2 Defaults, Modula-2
5056 @subsubsection Deviations from Standard Modula-2
5057 @cindex Modula-2, deviations from
5059 A few changes have been made to make Modula-2 programs easier to debug.
5060 This is done primarily via loosening its type strictness:
5064 Unlike in standard Modula-2, pointer constants can be formed by
5065 integers. This allows you to modify pointer variables during
5066 debugging. (In standard Modula-2, the actual address contained in a
5067 pointer variable is hidden from you; it can only be modified
5068 through direct assignment to another pointer variable or expression that
5069 returned a pointer.)
5072 C escape sequences can be used in strings and characters to represent
5073 non-printable characters. _GDBN__ will print out strings with these
5074 escape sequences embedded. Single non-printable characters are
5075 printed using the @samp{CHR(@var{nnn})} format.
5078 The assignment operator (@code{:=}) returns the value of its right-hand
5082 All builtin procedures both modify @emph{and} return their argument.
5086 @node M2 Checks, M2 Scope, Deviations, Modula-2
5087 @subsubsection Modula-2 Type and Range Checks
5088 @cindex Modula-2 checks
5091 @emph{Warning:} in this release, _GDBN__ does not yet perform type or
5094 @c FIXME remove warning when type/range checks added
5096 _GDBN__ considers two Modula-2 variables type equivalent if:
5100 They are of types that have been declared equivalent via a @code{TYPE
5101 @var{t1} = @var{t2}} statement
5104 They have been declared on the same line. (Note: This is true of the
5105 GNU Modula-2 compiler, but it may not be true of other compilers.)
5109 As long as type checking is enabled, any attempt to combine variables
5110 whose types are not equivalent is an error.
5112 Range checking is done on all mathematical operations, assignment, array
5113 index bounds, and all builtin functions and procedures.
5115 @node M2 Scope, GDB/M2, M2 Checks, Modula-2
5116 @subsubsection The scope operators @code{::} and @code{.}
5121 There are a few subtle differences between the Modula-2 scope operator
5122 (@code{.}) and the _GDBN__ scope operator (@code{::}). The two have
5127 @var{module} . @var{id}
5128 @var{scope} :: @var{id}
5133 where @var{scope} is the name of a module or a procedure,
5134 @var{module} the name of a module, and @var{id} is any delcared
5135 identifier within the program, except another module.
5137 Using the @code{::} operator makes _GDBN__ search the scope
5138 specified by @var{scope} for the identifier @var{id}. If it is not
5139 found in the specified scope, then _GDBN__ will search all scopes
5140 enclosing the one specified by @var{scope}.
5142 Using the @code{.} operator makes _GDBN__ search the current scope for
5143 the identifier specified by @var{id} that was imported from the
5144 definition module specified by @var{module}. With this operator, it is
5145 an error if the identifier @var{id} was not imported from definition
5146 module @var{module}, or if @var{id} is not an identifier in
5149 @node GDB/M2, , M2 Scope, Modula-2
5150 @subsubsection _GDBN__ and Modula-2
5152 Some _GDBN__ commands have little use when debugging Modula-2 programs.
5153 Five subcommands of @code{set print} and @code{show print} apply
5154 specifically to C and C++: @samp{vtbl}, @samp{demangle},
5155 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
5156 apply to C++, and the last to C's @code{union} type, which has no direct
5157 analogue in Modula-2.
5159 The @code{@@} operator (@pxref{Expressions}), while available
5160 while using any language, is not useful with Modula-2. Its
5161 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
5162 created in Modula-2 as they can in C or C++. However, because an
5163 address can be specified by an integral constant, the construct
5164 @samp{@{@var{type}@}@var{adrexp}} is still useful. (@pxref{Expressions})
5167 @cindex @code{#} in Modula-2
5168 In _GDBN__ scripts, the Modula-2 inequality operator @code{#} is
5169 interpreted as the beginning of a comment. Use @code{<>} instead.
5173 @node Symbols, Altering, Languages, Top
5174 @chapter Examining the Symbol Table
5176 The commands described in this section allow you to inquire about the
5177 symbols (names of variables, functions and types) defined in your
5178 program. This information is inherent in the text of your program and
5179 does not change as the program executes. _GDBN__ finds it in your
5180 program's symbol table, in the file indicated when you started _GDBN__
5181 (@pxref{File Options}), or by one of the file-management commands
5185 @item info address @var{symbol}
5186 @kindex info address
5187 Describe where the data for @var{symbol} is stored. For a register
5188 variable, this says which register it is kept in. For a non-register
5189 local variable, this prints the stack-frame offset at which the variable
5192 Note the contrast with @samp{print &@var{symbol}}, which does not work
5193 at all for a register variables, and for a stack local variable prints
5194 the exact address of the current instantiation of the variable.
5196 @item whatis @var{exp}
5198 Print the data type of expression @var{exp}. @var{exp} is not
5199 actually evaluated, and any side-effecting operations (such as
5200 assignments or function calls) inside it do not take place.
5204 Print the data type of @code{$}, the last value in the value history.
5206 @item ptype @var{typename}
5208 Print a description of data type @var{typename}. @var{typename} may be
5209 the name of a type, or for C code it may have the form
5210 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
5211 @samp{enum @var{enum-tag}}.@refill
5213 @item ptype @var{exp}
5215 Print a description of the type of expression @var{exp}. @code{ptype}
5216 differs from @code{whatis} by printing a detailed description, instead of just
5217 the name of the type. For example, if your program declares a variable
5220 struct complex @{double real; double imag;@} v;
5223 compare the output of the two commands:
5226 type = struct complex
5228 type = struct complex @{
5234 As with @code{whatis}, using @code{ptype} without an argument refers to
5235 the type of @code{$}, the last value in the value history.
5237 @item info types @var{regexp}
5240 Print a brief description of all types whose name matches @var{regexp}
5241 (or all types in your program, if you supply no argument). Each
5242 complete typename is matched as though it were a complete line; thus,
5243 @samp{i type value} gives information on all types in your program whose
5244 name includes the string @code{value}, but @samp{i type ^value$} gives
5245 information only on types whose complete name is @code{value}.
5247 This command differs from @code{ptype} in two ways: first, like
5248 @code{whatis}, it does not print a detailed description; second, it
5249 lists all source files where a type is defined.
5253 Show the name of the current source file---that is, the source file for
5254 the function containing the current point of execution---and the language
5258 @kindex info sources
5259 Print the names of all source files in the program for which there is
5260 debugging information, organized into two lists: files whose symbols
5261 have already been read, and files whose symbols will be read when needed.
5263 @item info functions
5264 @kindex info functions
5265 Print the names and data types of all defined functions.
5267 @item info functions @var{regexp}
5268 Print the names and data types of all defined functions
5269 whose names contain a match for regular expression @var{regexp}.
5270 Thus, @samp{info fun step} finds all functions whose names
5271 include @code{step}; @samp{info fun ^step} finds those whose names
5272 start with @code{step}.
5274 @item info variables
5275 @kindex info variables
5276 Print the names and data types of all variables that are declared
5277 outside of functions (i.e., excluding local variables).
5279 @item info variables @var{regexp}
5280 Print the names and data types of all variables (except for local
5281 variables) whose names contain a match for regular expression
5286 This was never implemented.
5288 @itemx info methods @var{regexp}
5289 @kindex info methods
5290 The @code{info methods} command permits the user to examine all defined
5291 methods within C++ program, or (with the @var{regexp} argument) a
5292 specific set of methods found in the various C++ classes. Many
5293 C++ classes provide a large number of methods. Thus, the output
5294 from the @code{ptype} command can be overwhelming and hard to use. The
5295 @code{info-methods} command filters the methods, printing only those
5296 which match the regular-expression @var{regexp}.
5299 @item printsyms @var{filename}
5300 @itemx printpsyms @var{filename}
5304 @cindex partial symbol dump
5305 Write a dump of debugging symbol data into the file @var{filename}.
5306 These commands are used to debug the _GDBN__ symbol-reading code. Only
5307 symbols with debugging data are included. If you use @code{printsyms},
5308 _GDBN__ includes all the symbols for which it has already collected full
5309 details: that is, @var{filename} reflects symbols for only those files
5310 whose symbols _GDBN__ has read. You can use the command @code{info
5311 sources} to find out which files these are. If you use
5312 @code{printpsyms}, the dump also shows information about symbols that
5313 _GDBN__ only knows partially---that is, symbols defined in files that
5314 _GDBN__ has skimmed, but not yet read completely. The description of
5315 @code{symbol-file} describes how _GDBN__ reads symbols; both commands
5316 are described under @ref{Files}.
5320 @node Altering, _GDBN__ Files, Symbols, Top
5321 @chapter Altering Execution
5323 Once you think you have found an error in the program, you might want to
5324 find out for certain whether correcting the apparent error would lead to
5325 correct results in the rest of the run. You can find the answer by
5326 experiment, using the _GDBN__ features for altering execution of the
5329 For example, you can store new values into variables or memory
5330 locations, give the program a signal, restart it at a different address,
5331 or even return prematurely from a function to its caller.
5334 * Assignment:: Assignment to Variables
5335 * Jumping:: Continuing at a Different Address
5336 * Signaling:: Giving the Program a Signal
5337 * Returning:: Returning from a Function
5338 * Calling:: Calling your Program's Functions
5339 * Patching:: Patching your Program
5342 @node Assignment, Jumping, Altering, Altering
5343 @section Assignment to Variables
5346 @cindex setting variables
5347 To alter the value of a variable, evaluate an assignment expression.
5348 @xref{Expressions}. For example,
5355 would store the value 4 into the variable @code{x}, and then print the
5356 value of the assignment expression (which is 4). @xref{Languages}, for
5357 more information on operators in supported languages.
5359 @kindex set variable
5360 @cindex variables, setting
5361 If you are not interested in seeing the value of the assignment, use the
5362 @code{set} command instead of the @code{print} command. @code{set} is
5363 really the same as @code{print} except that the expression's value is not
5364 printed and is not put in the value history (@pxref{Value History}). The
5365 expression is evaluated only for its effects.
5367 If the beginning of the argument string of the @code{set} command
5368 appears identical to a @code{set} subcommand, use the @code{set
5369 variable} command instead of just @code{set}. This command is identical
5370 to @code{set} except for its lack of subcommands. For example, a
5371 program might well have a variable @code{width}---which leads to
5372 an error if we try to set a new value with just @samp{set width=13}, as
5373 we might if @code{set width} didn't happen to be a _GDBN__ command:
5375 (_GDBP__) whatis width
5379 (_GDBP__) set width=47
5380 Invalid syntax in expression.
5383 The invalid expression, of course, is @samp{=47}. What we can do in
5384 order to actually set our program's variable @code{width} is
5386 (_GDBP__) set var width=47
5389 _GDBN__ allows more implicit conversions in assignments than C does; you can
5390 freely store an integer value into a pointer variable or vice versa, and
5391 any structure can be converted to any other structure that is the same
5393 @comment FIXME: how do structs align/pad in these conversions?
5394 @comment /pesch@cygnus.com 18dec1990
5396 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
5397 construct to generate a value of specified type at a specified address
5398 (@pxref{Expressions}). For example, @code{@{int@}0x83040} refers
5399 to memory location @code{0x83040} as an integer (which implies a certain size
5400 and representation in memory), and
5403 set @{int@}0x83040 = 4
5407 stores the value 4 into that memory location.
5409 @node Jumping, Signaling, Assignment, Altering
5410 @section Continuing at a Different Address
5412 Ordinarily, when you continue the program, you do so at the place where
5413 it stopped, with the @code{continue} command. You can instead continue at
5414 an address of your own choosing, with the following commands:
5417 @item jump @var{linespec}
5419 Resume execution at line @var{linespec}. Execution will stop
5420 immediately if there is a breakpoint there. @xref{List} for a
5421 description of the different forms of @var{linespec}.
5423 The @code{jump} command does not change the current stack frame, or
5424 the stack pointer, or the contents of any memory location or any
5425 register other than the program counter. If line @var{linespec} is in
5426 a different function from the one currently executing, the results may
5427 be bizarre if the two functions expect different patterns of arguments or
5428 of local variables. For this reason, the @code{jump} command requests
5429 confirmation if the specified line is not in the function currently
5430 executing. However, even bizarre results are predictable if you are
5431 well acquainted with the machine-language code of the program.
5433 @item jump *@var{address}
5434 Resume execution at the instruction at address @var{address}.
5437 You can get much the same effect as the @code{jump} command by storing a
5438 new value into the register @code{$pc}. The difference is that this
5439 does not start the program running; it only changes the address where it
5440 @emph{will} run when it is continued. For example,
5447 causes the next @code{continue} command or stepping command to execute at
5448 address 0x485, rather than at the address where the program stopped.
5449 @xref{Continuing and Stepping}.
5451 The most common occasion to use the @code{jump} command is to back up,
5452 perhaps with more breakpoints set, over a portion of a program that has
5453 already executed, in order to examine its execution in more detail.
5455 @node Signaling, Returning, Jumping, Altering
5457 @section Giving the Program a Signal
5460 @item signal @var{signalnum}
5462 Resume execution where the program stopped, but give it immediately the
5463 signal number @var{signalnum}.
5465 Alternatively, if @var{signalnum} is zero, continue execution without
5466 giving a signal. This is useful when the program stopped on account of
5467 a signal and would ordinary see the signal when resumed with the
5468 @code{continue} command; @samp{signal 0} causes it to resume without a
5471 @code{signal} does not repeat when you press @key{RET} a second time
5472 after executing the command.
5476 @node Returning, Calling, Signaling, Altering
5477 @section Returning from a Function
5481 @itemx return @var{expression}
5482 @cindex returning from a function
5484 You can cancel execution of a function call with the @code{return}
5485 command. If you give an
5486 @var{expression} argument, its value is used as the function's return
5490 When you use @code{return}, _GDBN__ discards the selected stack frame
5491 (and all frames within it). You can think of this as making the
5492 discarded frame return prematurely. If you wish to specify a value to
5493 be returned, give that value as the argument to @code{return}.
5495 This pops the selected stack frame (@pxref{Selection}), and any other
5496 frames inside of it, leaving its caller as the innermost remaining
5497 frame. That frame becomes selected. The specified value is stored in
5498 the registers used for returning values of functions.
5500 The @code{return} command does not resume execution; it leaves the
5501 program stopped in the state that would exist if the function had just
5502 returned. In contrast, the @code{finish} command
5503 (@pxref{Continuing and Stepping})
5504 resumes execution until the selected stack frame returns naturally.@refill
5506 @node Calling, Patching, Returning, Altering
5507 @section Calling your Program's Functions
5509 @cindex calling functions
5512 @item call @var{expr}
5513 Evaluate the expression @var{expr} without displaying @code{void}
5517 You can use this variant of the @code{print} command if you want to
5518 execute a function from your program, but without cluttering the output
5519 with @code{void} returned values. The result is printed and saved in
5520 the value history, if it is not void.
5522 @node Patching, , Calling, Altering
5523 @section Patching your Program
5524 @cindex patching binaries
5525 @cindex writing into executables
5526 @cindex writing into corefiles
5527 By default, _GDBN__ opens the file containing your program's executable
5528 code (or the corefile) read-only. This prevents accidental alterations
5529 to machine code; but it also prevents you from intentionally patching
5530 your program's binary.
5532 If you'd like to be able to patch the binary, you can specify that
5533 explicitly with the @code{set write} command. For example, you might
5534 want to turn on internal debugging flags, or even to make emergency
5539 @itemx set write off
5541 If you specify @samp{set write on}, _GDBN__ will open executable and
5542 core files for both reading and writing; if you specify @samp{set write
5543 off} (the default), _GDBN__ will open them read-only.
5545 If you've already loaded a file, you must load it
5546 again (using the @code{exec-file} or @code{core-file} command) after
5547 changing @code{set write}, for your new setting to take effect.
5551 Display whether executable files and core files will be opened for
5552 writing as well as reading.
5556 @node _GDBN__ Files, Targets, Altering, Top
5557 @chapter _GDBN__'s Files
5560 * Files:: Commands to Specify Files
5561 * Symbol Errors:: Errors Reading Symbol Files
5564 @node Files, Symbol Errors, _GDBN__ Files, _GDBN__ Files
5565 @section Commands to Specify Files
5566 @cindex core dump file
5567 @cindex symbol table
5568 _GDBN__ needs to know the file name of the program to be debugged, both in
5569 order to read its symbol table and in order to start the program. To
5570 debug a core dump of a previous run, _GDBN__ must be told the file name of
5573 The usual way to specify the executable and core dump file names is with
5574 the command arguments given when you start _GDBN__, as discussed in
5577 Occasionally it is necessary to change to a different file during a
5578 _GDBN__ session. Or you may run _GDBN__ and forget to specify the files you
5579 want to use. In these situations the _GDBN__ commands to specify new files
5583 @item file @var{filename}
5584 @cindex executable file
5586 Use @var{filename} as the program to be debugged. It is read for its
5587 symbols and for the contents of pure memory. It is also the program
5588 executed when you use the @code{run} command. If you do not specify a
5589 directory and the file is not found in _GDBN__'s working directory,
5591 _GDBN__ uses the environment variable @code{PATH} as a list of
5592 directories to search, just as the shell does when looking for a program
5593 to run. You can change the value of this variable, for both _GDBN__ and
5594 your program, using the @code{path} command.
5597 @code{file} with no argument makes _GDBN__ discard any information it
5598 has on both executable file and the symbol table.
5600 @item exec-file @r{[} @var{filename} @r{]}
5602 Specify that the program to be run (but not the symbol table) is found
5603 in @var{filename}. _GDBN__ will search the environment variable @code{PATH}
5604 if necessary to locate the program. Omitting @var{filename} means to
5605 discard information on the executable file.
5607 @item symbol-file @r{[} @var{filename} @r{]}
5609 Read symbol table information from file @var{filename}. @code{PATH} is
5610 searched when necessary. Use the @code{file} command to get both symbol
5611 table and program to run from the same file.
5613 @code{symbol-file} with no argument clears out _GDBN__'s information on your
5614 program's symbol table.
5616 The @code{symbol-file} command causes _GDBN__ to forget the contents of its
5617 convenience variables, the value history, and all breakpoints and
5618 auto-display expressions. This is because they may contain pointers to
5619 the internal data recording symbols and data types, which are part of
5620 the old symbol table data being discarded inside _GDBN__.
5622 @code{symbol-file} will not repeat if you press @key{RET} again after
5625 On some kinds of object files, the @code{symbol-file} command does not
5626 actually read the symbol table in full right away. Instead, it scans
5627 the symbol table quickly to find which source files and which symbols
5628 are present. The details are read later, one source file at a time,
5629 when they are needed.
5631 The purpose of this two-stage reading strategy is to make _GDBN__ start up
5632 faster. For the most part, it is invisible except for occasional pauses
5633 while the symbol table details for a particular source file are being
5634 read. (The @code{set verbose} command can turn these pauses into
5635 messages if desired. @xref{Messages/Warnings}).
5637 When the symbol table is stored in COFF format, @code{symbol-file} does
5638 read the symbol table data in full right away. We haven't implemented
5639 the two-stage strategy for COFF yet.
5641 When _GDBN__ is configured for a particular environment, it will
5642 understand debugging information in whatever format is the standard
5643 generated for that environment; you may use either a GNU compiler, or
5644 other compilers that adhere to the local conventions. Best results are
5645 usually obtained from GNU compilers; for example, using @code{_GCC__}
5646 you can generate debugging information for optimized code.
5648 @item core-file @r{[} @var{filename} @r{]}
5651 Specify the whereabouts of a core dump file to be used as the ``contents
5652 of memory''. Traditionally, core files contain only some parts of the
5653 address space of the process that generated them; _GDBN__ can access the
5654 executable file itself for other parts.
5656 @code{core-file} with no argument specifies that no core file is
5659 Note that the core file is ignored when your program is actually running
5660 under _GDBN__. So, if you have been running the program and you wish to
5661 debug a core file instead, you must kill the subprocess in which the
5662 program is running. To do this, use the @code{kill} command
5663 (@pxref{Kill Process}).
5665 @item load @var{filename}
5668 Depending on what remote debugging facilities are configured into
5669 _GDBN__, the @code{load} command may be available. Where it exists, it
5670 is meant to make @var{filename} (an executable) available for debugging
5671 on the remote system---by downloading, or dynamic linking, for example.
5672 @code{load} also records @var{filename}'s symbol table in _GDBN__, like
5673 the @code{add-symbol-file} command.
5675 If @code{load} is not available on your _GDBN__, attempting to execute
5676 it gets the error message ``@code{You can't do that when your target is
5681 On VxWorks, @code{load} will dynamically link @var{filename} on the
5682 current target system as well as adding its symbols in _GDBN__.
5686 @cindex download to Nindy-960
5687 With the Nindy interface to an Intel 960 board, @code{load} will
5688 download @var{filename} to the 960 as well as adding its symbols in
5692 @code{load} will not repeat if you press @key{RET} again after using it.
5694 @item add-symbol-file @var{filename} @var{address}
5695 @kindex add-symbol-file
5696 @cindex dynamic linking
5697 The @code{add-symbol-file} command reads additional symbol table information
5698 from the file @var{filename}. You would use this command when @var{filename}
5699 has been dynamically loaded (by some other means) into the program that
5700 is running. @var{address} should be the memory address at which the
5701 file has been loaded; _GDBN__ cannot figure this out for itself.
5703 The symbol table of the file @var{filename} is added to the symbol table
5704 originally read with the @code{symbol-file} command. You can use the
5705 @code{add-symbol-file} command any number of times; the new symbol data thus
5706 read keeps adding to the old. To discard all old symbol data instead,
5707 use the @code{symbol-file} command.
5709 @code{add-symbol-file} will not repeat if you press @key{RET} after using it.
5715 @code{info files} and @code{info target} are synonymous; both print the
5716 current targets (@pxref{Targets}), including the names of the executable
5717 and core dump files currently in use by _GDBN__, and the files from
5718 which symbols were loaded. The command @code{help targets} lists all
5719 possible targets rather than current ones.
5723 All file-specifying commands allow both absolute and relative file names
5724 as arguments. _GDBN__ always converts the file name to an absolute path
5725 name and remembers it that way.
5727 @cindex shared libraries
5729 _GDBN__ supports the SunOS shared library format. _GDBN__ automatically
5730 loads symbol definitions from shared libraries when you use the
5731 @code{run} command, or when you examine a core file. (Before you issue
5732 the @code{run} command, _GDBN__ won't understand references to a
5733 function in a shared library, however---unless you're debugging a core
5735 @c FIXME: next _GDBN__ release should permit some refs to undef
5736 @c FIXME...symbols---eg in a break cmd---assuming they're from a shared lib
5740 @itemx info sharedlibrary
5741 @kindex info sharedlibrary
5743 Print the names of the shared libraries which are currently loaded.
5745 @item sharedlibrary @var{regex}
5746 @itemx share @var{regex}
5747 @kindex sharedlibrary
5749 This is an obsolescent command; you can use it to explicitly
5750 load shared object library symbols for files matching a UNIX regular
5751 expression, but as with files loaded automatically, it will only load
5752 shared libraries required by your program for a core file or after
5753 typing @code{run}. If @var{regex} is omitted all shared libraries
5754 required by your program are loaded.
5757 @node Symbol Errors, , Files, _GDBN__ Files
5758 @section Errors Reading Symbol Files
5759 While reading a symbol file, _GDBN__ will occasionally encounter
5760 problems, such as symbol types it does not recognize, or known bugs in
5761 compiler output. By default, _GDBN__ does not notify you of such
5762 problems, since they're relatively common and primarily of interest to
5763 people debugging compilers. If you are interested in seeing information
5764 about ill-constructed symbol tables, you can either ask _GDBN__ to print
5765 only one message about each such type of problem, no matter how many
5766 times the problem occurs; or you can ask _GDBN__ to print more messages,
5767 to see how many times the problems occur, with the @code{set complaints}
5768 command (@xref{Messages/Warnings}).
5770 The messages currently printed, and their meanings, are:
5773 @item inner block not inside outer block in @var{symbol}
5775 The symbol information shows where symbol scopes begin and end
5776 (such as at the start of a function or a block of statements). This
5777 error indicates that an inner scope block is not fully contained
5778 in its outer scope blocks.
5780 _GDBN__ circumvents the problem by treating the inner block as if it had
5781 the same scope as the outer block. In the error message, @var{symbol}
5782 may be shown as ``@code{(don't know)}'' if the outer block is not a
5785 @item block at @var{address} out of order
5787 The symbol information for symbol scope blocks should occur in
5788 order of increasing addresses. This error indicates that it does not
5791 _GDBN__ does not circumvent this problem, and will have trouble locating
5792 symbols in the source file whose symbols being read. (You can often
5793 determine what source file is affected by specifying @code{set verbose
5794 on}. @xref{Messages/Warnings}.)
5796 @item bad block start address patched
5798 The symbol information for a symbol scope block has a start address
5799 smaller than the address of the preceding source line. This is known
5800 to occur in the SunOS 4.1.1 (and earlier) C compiler.
5802 _GDBN__ circumvents the problem by treating the symbol scope block as
5803 starting on the previous source line.
5805 @item bad string table offset in symbol @var{n}
5808 Symbol number @var{n} contains a pointer into the string table which is
5809 larger than the size of the string table.
5811 _GDBN__ circumvents the problem by considering the symbol to have the
5812 name @code{foo}, which may cause other problems if many symbols end up
5815 @item unknown symbol type @code{0x@var{nn}}
5817 The symbol information contains new data types that _GDBN__ does not yet
5818 know how to read. @code{0x@var{nn}} is the symbol type of the misunderstood
5819 information, in hexadecimal.
5821 _GDBN__ circumvents the error by ignoring this symbol information. This
5822 will usually allow the program to be debugged, though certain symbols
5823 will not be accessible. If you encounter such a problem and feel like
5824 debugging it, you can debug @code{_GDBP__} with itself, breakpoint on
5825 @code{complain}, then go up to the function @code{read_dbx_symtab} and
5826 examine @code{*bufp} to see the symbol.
5828 @item stub type has NULL name
5829 _GDBN__ could not find the full definition for a struct or class.
5831 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
5833 The symbol information for a C++ member function is missing some
5834 information that recent versions of the compiler should have output
5837 @item info mismatch between compiler and debugger
5839 _GDBN__ could not parse a type specification output by the compiler.
5843 @node Targets, Controlling _GDBN__, _GDBN__ Files, Top
5844 @chapter Specifying a Debugging Target
5845 @cindex debugging target
5847 A @dfn{target} is the execution environment occupied by your program.
5848 Often, _GDBN__ runs in the same host environment as the program you are
5849 debugging; in that case, the debugging target is specified as a side
5850 effect when you use the @code{file} or @code{core} commands. When you
5851 need more flexibility---for example, running _GDBN__ on a physically
5852 separate host, or controlling a standalone system over a serial port or
5853 a realtime system over a TCP/IP connection---you can use the
5854 @code{target} command to specify one of the target types configured for
5855 _GDBN__ (@pxref{Target Commands}).
5858 * Active Targets:: Active Targets
5859 * Target Commands:: Commands for Managing Targets
5860 * Remote:: Remote Debugging
5863 @node Active Targets, Target Commands, Targets, Targets
5864 @section Active Targets
5865 @cindex stacking targets
5866 @cindex active targets
5867 @cindex multiple targets
5869 There are three classes of targets: processes, core files, and
5870 executable files. _GDBN__ can work concurrently on up to three active
5871 targets, one in each class. This allows you to (for example) start a
5872 process and inspect its activity without abandoning your work on a core
5875 If, for example, you execute @samp{gdb a.out}, then the executable file
5876 @code{a.out} is the only active target. If you designate a core file as
5877 well---presumably from a prior run that crashed and coredumped---then
5878 _GDBN__ has two active targets and will use them in tandem, looking
5879 first in the corefile target, then in the executable file, to satisfy
5880 requests for memory addresses. (Typically, these two classes of target
5881 are complementary, since core files contain only the program's
5882 read-write memory---variables and so on---plus machine status, while
5883 executable files contain only the program text and initialized data.)
5885 When you type @code{run}, your executable file becomes an active process
5886 target as well. When a process target is active, all _GDBN__ commands
5887 requesting memory addresses refer to that target; addresses in an active
5888 core file or executable file target are obscured while the process
5891 Use the @code{core-file}, and @code{exec-file} commands to select a new
5892 core file or executable target (@pxref{Files}). To specify as a target
5893 a process that's already running, use the @code{attach} command
5896 @node Target Commands, Remote, Active Targets, Targets
5897 @section Commands for Managing Targets
5900 @item target @var{type} @var{parameters}
5901 Connects the _GDBN__ host environment to a target machine or process. A
5902 target is typically a protocol for talking to debugging facilities. You
5903 use the argument @var{type} to specify the type or protocol of the
5906 Further @var{parameters} are interpreted by the target protocol, but
5907 typically include things like device names or host names to connect
5908 with, process numbers, and baud rates.
5910 The @code{target} command will not repeat if you press @key{RET} again
5911 after executing the command.
5915 Displays the names of all targets available. To display targets
5916 currently selected, use either @code{info target} or @code{info files}
5919 @item help target @var{name}
5920 Describe a particular target, including any parameters necessary to
5924 Here are some common targets (available, or not, depending on the _GDBN__
5928 @item target exec @var{prog}
5930 An executable file. @samp{target exec @var{prog}} is the same as
5931 @samp{exec-file @var{prog}}.
5933 @item target core @var{filename}
5935 A core dump file. @samp{target core @var{filename}} is the same as
5936 @samp{core-file @var{filename}}.
5938 @item target remote @var{dev}
5939 @kindex target remote
5940 Remote serial target in _GDBN__-specific protocol. The argument @var{dev}
5941 specifies what serial device to use for the connection (e.g.
5942 @file{/dev/ttya}). @xref{Remote}.
5945 @item target amd-eb @var{dev} @var{speed} @var{PROG}
5946 @kindex target amd-eb
5948 Remote PC-resident AMD EB29K board, attached over serial lines.
5949 @var{dev} is the serial device, as for @code{target remote};
5950 @var{speed} allows you to specify the linespeed; and @var{PROG} is the
5951 name of the program to be debugged, as it appears to DOS on the PC.
5952 @xref{EB29K Remote}.
5956 @item target nindy @var{devicename}
5957 @kindex target nindy
5958 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
5959 the name of the serial device to use for the connection, e.g.
5960 @file{/dev/ttya}. @xref{i960-Nindy Remote}.
5964 @item target vxworks @var{machinename}
5965 @kindex target vxworks
5966 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
5967 is the target system's machine name or IP address.
5968 @xref{VxWorks Remote}.
5973 Different targets are available on different configurations of _GDBN__; your
5974 configuration may have more or fewer targets.
5977 @node Remote, , Target Commands, Targets
5978 @section Remote Debugging
5979 @cindex remote debugging
5983 _include__(gdbinv-m.m4)<>_dnl__
5987 If you are trying to debug a program running on a machine that can't run
5988 _GDBN__ in the usual way, it is often useful to use remote debugging. For
5989 example, you might use remote debugging on an operating system kernel, or on
5990 a small system which does not have a general purpose operating system
5991 powerful enough to run a full-featured debugger.
5993 Some configurations of _GDBN__ have special serial or TCP/IP interfaces
5994 to make this work with particular debugging targets. In addition,
5995 _GDBN__ comes with a generic serial protocol (specific to _GDBN__, but
5996 not specific to any particular target system) which you can use if you
5997 write the remote stubs---the code that will run on the remote system to
5998 communicate with _GDBN__.
6000 To use the _GDBN__ remote serial protocol, the program to be debugged on
6001 the remote machine needs to contain a debugging stub which talks to
6002 _GDBN__ over the serial line. Several working remote stubs are
6003 distributed with _GDBN__; see the @file{README} file in the _GDBN__
6004 distribution for more information.
6006 For details of this communication protocol, see the comments in the
6007 _GDBN__ source file @file{remote.c}.
6009 To start remote debugging, first run _GDBN__ and specify as an executable file
6010 the program that is running in the remote machine. This tells _GDBN__ how
6011 to find the program's symbols and the contents of its pure text. Then
6012 establish communication using the @code{target remote} command with a device
6013 name as an argument. For example:
6016 target remote /dev/ttyb
6020 if the serial line is connected to the device named @file{/dev/ttyb}. This
6021 will stop the remote machine if it is not already stopped.
6023 Now you can use all the usual commands to examine and change data and to
6024 step and continue the remote program.
6026 To resume the remote program and stop debugging it, use the @code{detach}
6029 Other remote targets may be available in your
6030 configuration of _GDBN__; use @code{help targets} to list them.
6033 @c Text on starting up GDB in various specific cases; it goes up front
6034 @c in manuals configured for any of those particular situations, here
6036 _include__(gdbinv-s.m4)
6039 @node Controlling _GDBN__, Sequences, Targets, Top
6040 @chapter Controlling _GDBN__
6042 You can alter many aspects of _GDBN__'s interaction with you by using
6043 the @code{set} command. For commands controlling how _GDBN__ displays
6044 data, @pxref{Print Settings}; other settings are described here.
6048 * Editing:: Command Editing
6049 * History:: Command History
6050 * Screen Size:: Screen Size
6052 * Messages/Warnings:: Optional Warnings and Messages
6055 @node Prompt, Editing, Controlling _GDBN__, Controlling _GDBN__
6058 _GDBN__ indicates its readiness to read a command by printing a string
6059 called the @dfn{prompt}. This string is normally @samp{(_GDBP__)}. You
6060 can change the prompt string with the @code{set prompt} command. For
6061 instance, when debugging _GDBN__ with _GDBN__, it is useful to change
6062 the prompt in one of the _GDBN__<>s so that you can always tell which
6063 one you are talking to.
6066 @item set prompt @var{newprompt}
6068 Directs _GDBN__ to use @var{newprompt} as its prompt string henceforth.
6071 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
6074 @node Editing, History, Prompt, Controlling _GDBN__
6075 @section Command Editing
6077 @cindex command line editing
6078 _GDBN__ reads its input commands via the @dfn{readline} interface. This
6079 GNU library provides consistent behavior for programs which provide a
6080 command line interface to the user. Advantages are @code{emacs}-style
6081 or @code{vi}-style inline editing of commands, @code{csh}-like history
6082 substitution, and a storage and recall of command history across
6085 You may control the behavior of command line editing in _GDBN__ with the
6092 @itemx set editing on
6093 Enable command line editing (enabled by default).
6095 @item set editing off
6096 Disable command line editing.
6098 @kindex show editing
6100 Show whether command line editing is enabled.
6103 @node History, Screen Size, Editing, Controlling _GDBN__
6104 @section Command History
6106 @cindex history substitution
6107 @cindex history file
6108 @kindex set history filename
6109 @item set history filename @var{fname}
6110 Set the name of the _GDBN__ command history file to @var{fname}. This is
6111 the file from which _GDBN__ will read an initial command history
6112 list or to which it will write this list when it exits. This list is
6113 accessed through history expansion or through the history
6114 command editing characters listed below. This file defaults to the
6115 value of the environment variable @code{GDBHISTFILE}, or to
6116 @file{./.gdb_history} if this variable is not set.
6118 @cindex history save
6119 @kindex set history save
6120 @item set history save
6121 @itemx set history save on
6122 Record command history in a file, whose name may be specified with the
6123 @code{set history filename} command. By default, this option is disabled.
6125 @item set history save off
6126 Stop recording command history in a file.
6128 @cindex history size
6129 @kindex set history size
6130 @item set history size @var{size}
6131 Set the number of commands which _GDBN__ will keep in its history list.
6132 This defaults to the value of the environment variable
6133 @code{HISTSIZE}, or to 256 if this variable is not set.
6136 @cindex history expansion
6137 History expansion assigns special meaning to the character @kbd{!}.
6139 (@xref{Event Designators}.)
6141 Since @kbd{!} is also the logical not operator in C, history expansion
6142 is off by default. If you decide to enable history expansion with the
6143 @code{set history expansion on} command, you may sometimes need to
6144 follow @kbd{!} (when it is used as logical not, in an expression) with
6145 a space or a tab to prevent it from being expanded. The readline
6146 history facilities will not attempt substitution on the strings
6147 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
6149 The commands to control history expansion are:
6153 @kindex set history expansion
6154 @item set history expansion on
6155 @itemx set history expansion
6156 Enable history expansion. History expansion is off by default.
6158 @item set history expansion off
6159 Disable history expansion.
6161 The readline code comes with more complete documentation of
6162 editing and history expansion features. Users unfamiliar with @code{emacs}
6163 or @code{vi} may wish to read it.
6165 @xref{Command Line Editing}.
6169 @kindex show history
6171 @itemx show history filename
6172 @itemx show history save
6173 @itemx show history size
6174 @itemx show history expansion
6175 These commands display the state of the _GDBN__ history parameters.
6176 @code{show history} by itself displays all four states.
6182 @kindex show commands
6184 Display the last ten commands in the command history.
6186 @item show commands @var{n}
6187 Print ten commands centered on command number @var{n}.
6189 @item show commands +
6190 Print ten commands just after the commands last printed.
6194 @node Screen Size, Numbers, History, Controlling _GDBN__
6195 @section Screen Size
6196 @cindex size of screen
6197 @cindex pauses in output
6198 Certain commands to _GDBN__ may produce large amounts of information
6199 output to the screen. To help you read all of it, _GDBN__ pauses and
6200 asks you for input at the end of each page of output. Type @key{RET}
6201 when you want to continue the output. _GDBN__ also uses the screen
6202 width setting to determine when to wrap lines of output. Depending on
6203 what is being printed, it tries to break the line at a readable place,
6204 rather than simply letting it overflow onto the following line.
6206 Normally _GDBN__ knows the size of the screen from the termcap data base
6207 together with the value of the @code{TERM} environment variable and the
6208 @code{stty rows} and @code{stty cols} settings. If this is not correct,
6209 you can override it with the @code{set height} and @code{set
6213 @item set height @var{lpp}
6215 @itemx set width @var{cpl}
6221 These @code{set} commands specify a screen height of @var{lpp} lines and
6222 a screen width of @var{cpl} characters. The associated @code{show}
6223 commands display the current settings.
6225 If you specify a height of zero lines, _GDBN__ will not pause during output
6226 no matter how long the output is. This is useful if output is to a file
6227 or to an editor buffer.
6230 @node Numbers, Messages/Warnings, Screen Size, Controlling _GDBN__
6232 @cindex number representation
6233 @cindex entering numbers
6234 You can always enter numbers in octal, decimal, or hexadecimal in _GDBN__ by
6235 the usual conventions: octal numbers begin with @samp{0}, decimal
6236 numbers end with @samp{.}, and hexadecimal numbers begin with @samp{0x}.
6237 Numbers that begin with none of these are, by default, entered in base
6238 10; likewise, the default display for numbers---when no particular
6239 format is specified---is base 10. You can change the default base for
6240 both input and output with the @code{set radix} command.
6244 @item set radix @var{base}
6245 Set the default base for numeric input and display. Supported choices
6246 for @var{base} are decimal 2, 8, 10, 16. @var{base} must itself be
6247 specified either unambiguously or using the current default radix; for
6258 will set the base to decimal. On the other hand, @samp{set radix 10}
6259 will leave the radix unchanged no matter what it was.
6263 Display the current default base for numeric input and display.
6267 @node Messages/Warnings, , Numbers, Controlling _GDBN__
6268 @section Optional Warnings and Messages
6269 By default, _GDBN__ is silent about its inner workings. If you are running
6270 on a slow machine, you may want to use the @code{set verbose} command.
6271 It will make _GDBN__ tell you when it does a lengthy internal operation, so
6272 you won't think it has crashed.
6274 Currently, the messages controlled by @code{set verbose} are those which
6275 announce that the symbol table for a source file is being read
6276 (@pxref{Files}, in the description of the command
6277 @code{symbol-file}).
6278 @c The following is the right way to do it, but emacs 18.55 doesn't support
6279 @c @ref, and neither the emacs lisp manual version of texinfmt or makeinfo
6282 see @code{symbol-file} in @ref{Files}).
6287 @item set verbose on
6288 Enables _GDBN__'s output of certain informational messages.
6290 @item set verbose off
6291 Disables _GDBN__'s output of certain informational messages.
6293 @kindex show verbose
6295 Displays whether @code{set verbose} is on or off.
6298 By default, if _GDBN__ encounters bugs in the symbol table of an object
6299 file, it is silent; but if you are debugging a compiler, you may find
6300 this information useful (@pxref{Symbol Errors}).
6303 @kindex set complaints
6304 @item set complaints @var{limit}
6305 Permits _GDBN__ to output @var{limit} complaints about each type of unusual
6306 symbols before becoming silent about the problem. Set @var{limit} to
6307 zero to suppress all complaints; set it to a large number to prevent
6308 complaints from being suppressed.
6310 @kindex show complaints
6311 @item show complaints
6312 Displays how many symbol complaints _GDBN__ is permitted to produce.
6315 By default, _GDBN__ is cautious, and asks what sometimes seem to be a
6316 lot of stupid questions to confirm certain commands. For example, if
6317 you try to run a program which is already running:
6320 The program being debugged has been started already.
6321 Start it from the beginning? (y or n)
6324 If you're willing to unflinchingly face the consequences of your own
6325 commands, you can disable this ``feature'':
6330 @cindex confirmation
6331 @cindex stupid questions
6332 @item set confirm off
6333 Disables confirmation requests.
6335 @item set confirm on
6336 Enables confirmation requests (the default).
6339 @kindex show confirm
6340 Displays state of confirmation requests.
6343 @c FIXME this doesn't really belong here. But where *does* it belong?
6344 @cindex reloading symbols
6345 Some systems allow individual object files that make up your program to
6346 be replaced without stopping and restarting your program.
6348 For example, in VxWorks you can simply recompile a defective object file
6349 and keep on running.
6351 If you're running on one of these systems, you can allow _GDBN__ to
6352 reload the symbols for automatically relinked modules:@refill
6354 @kindex set symbol-reloading
6355 @item set symbol-reloading on
6356 Replace symbol definitions for the corresponding source file when an
6357 object file with a particular name is seen again.
6359 @item set symbol-reloading off
6360 Don't replace symbol definitions when re-encountering object files of
6361 the same name. This is the default state; if you're not running on a
6362 system that permits automatically relinking modules, you should leave
6363 @code{symbol-reloading} off, since otherwise _GDBN__ may discard symbols
6364 when linking large programs, that may contain several modules (from
6365 different directories or libraries) with the same name.
6367 @item show symbol-reloading
6368 Show the current @code{on} or @code{off} setting.
6371 @node Sequences, Emacs, Controlling _GDBN__, Top
6372 @chapter Canned Sequences of Commands
6374 Aside from breakpoint commands (@pxref{Break Commands}), _GDBN__ provides two
6375 ways to store sequences of commands for execution as a unit:
6376 user-defined commands and command files.
6379 * Define:: User-Defined Commands
6380 * Command Files:: Command Files
6381 * Output:: Commands for Controlled Output
6384 @node Define, Command Files, Sequences, Sequences
6385 @section User-Defined Commands
6387 @cindex user-defined command
6388 A @dfn{user-defined command} is a sequence of _GDBN__ commands to which you
6389 assign a new name as a command. This is done with the @code{define}
6393 @item define @var{commandname}
6395 Define a command named @var{commandname}. If there is already a command
6396 by that name, you are asked to confirm that you want to redefine it.
6398 The definition of the command is made up of other _GDBN__ command lines,
6399 which are given following the @code{define} command. The end of these
6400 commands is marked by a line containing @code{end}.
6402 @item document @var{commandname}
6404 Give documentation to the user-defined command @var{commandname}. The
6405 command @var{commandname} must already be defined. This command reads
6406 lines of documentation just as @code{define} reads the lines of the
6407 command definition, ending with @code{end}. After the @code{document}
6408 command is finished, @code{help} on command @var{commandname} will print
6409 the documentation you have specified.
6411 You may use the @code{document} command again to change the
6412 documentation of a command. Redefining the command with @code{define}
6413 does not change the documentation.
6415 @item help user-defined
6416 @kindex help user-defined
6417 List all user-defined commands, with the first line of the documentation
6421 @itemx info user @var{commandname}
6423 Display the _GDBN__ commands used to define @var{commandname} (but not its
6424 documentation). If no @var{commandname} is given, display the
6425 definitions for all user-defined commands.
6428 User-defined commands do not take arguments. When they are executed, the
6429 commands of the definition are not printed. An error in any command
6430 stops execution of the user-defined command.
6432 Commands that would ask for confirmation if used interactively proceed
6433 without asking when used inside a user-defined command. Many _GDBN__ commands
6434 that normally print messages to say what they are doing omit the messages
6435 when used in a user-defined command.
6437 @node Command Files, Output, Define, Sequences
6438 @section Command Files
6440 @cindex command files
6441 A command file for _GDBN__ is a file of lines that are _GDBN__ commands. Comments
6442 (lines starting with @kbd{#}) may also be included. An empty line in a
6443 command file does nothing; it does not mean to repeat the last command, as
6444 it would from the terminal.
6447 @cindex @file{_GDBINIT__}
6448 When you start _GDBN__, it automatically executes commands from its
6449 @dfn{init files}. These are files named @file{_GDBINIT__}. _GDBN__
6450 reads the init file (if any) in your home directory and then the init
6451 file (if any) in the current working directory. (The init files are not
6452 executed if you use the @samp{-nx} option; @pxref{Mode Options}.) You
6453 can also request the execution of a command file with the @code{source}
6457 @item source @var{filename}
6459 Execute the command file @var{filename}.
6462 The lines in a command file are executed sequentially. They are not
6463 printed as they are executed. An error in any command terminates execution
6464 of the command file.
6466 Commands that would ask for confirmation if used interactively proceed
6467 without asking when used in a command file. Many _GDBN__ commands that
6468 normally print messages to say what they are doing omit the messages
6469 when called from command files.
6471 @node Output, , Command Files, Sequences
6472 @section Commands for Controlled Output
6474 During the execution of a command file or a user-defined command, normal
6475 _GDBN__ output is suppressed; the only output that appears is what is
6476 explicitly printed by the commands in the definition. This section
6477 describes three commands useful for generating exactly the output you
6481 @item echo @var{text}
6483 @c I don't consider backslash-space a standard C escape sequence
6484 @c because it's not in ANSI.
6485 Print @var{text}. Nonprinting characters can be included in @var{text}
6486 using C escape sequences, such as @samp{\n} to print a newline. @b{No
6487 newline will be printed unless you specify one.} In addition to the
6488 standard C escape sequences, a backslash followed by a space stands for a
6489 space. This is useful for outputting a string with spaces at the
6490 beginning or the end, since leading and trailing spaces are otherwise
6491 trimmed from all arguments. Thus, to print @samp{@ and foo =@ }, use the
6492 command @samp{echo \@ and foo = \@ }.
6493 @c FIXME? '@ ' works in tex and info, but confuses texi2roff[-2].
6495 A backslash at the end of @var{text} can be used, as in C, to continue
6496 the command onto subsequent lines. For example,
6499 echo This is some text\n\
6500 which is continued\n\
6501 onto several lines.\n
6504 produces the same output as
6507 echo This is some text\n
6508 echo which is continued\n
6509 echo onto several lines.\n
6512 @item output @var{expression}
6514 Print the value of @var{expression} and nothing but that value: no
6515 newlines, no @samp{$@var{nn} = }. The value is not entered in the
6516 value history either. @xref{Expressions} for more information on
6519 @item output/@var{fmt} @var{expression}
6520 Print the value of @var{expression} in format @var{fmt}. You can use
6521 the same formats as for @code{print}; @pxref{Output formats}, for more
6524 @item printf @var{string}, @var{expressions}@dots{}
6526 Print the values of the @var{expressions} under the control of
6527 @var{string}. The @var{expressions} are separated by commas and may
6528 be either numbers or pointers. Their values are printed as specified
6529 by @var{string}, exactly as if the program were to execute
6532 printf (@var{string}, @var{expressions}@dots{});
6535 For example, you can print two values in hex like this:
6538 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
6541 The only backslash-escape sequences that you can use in the format
6542 string are the simple ones that consist of backslash followed by a
6546 @node Emacs, _GDBN__ Bugs, Sequences, Top
6547 @chapter Using _GDBN__ under GNU Emacs
6550 A special interface allows you to use GNU Emacs to view (and
6551 edit) the source files for the program you are debugging with
6554 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
6555 executable file you want to debug as an argument. This command starts
6556 _GDBN__ as a subprocess of Emacs, with input and output through a newly
6557 created Emacs buffer.
6559 Using _GDBN__ under Emacs is just like using _GDBN__ normally except for two
6564 All ``terminal'' input and output goes through the Emacs buffer.
6567 This applies both to _GDBN__ commands and their output, and to the input
6568 and output done by the program you are debugging.
6570 This is useful because it means that you can copy the text of previous
6571 commands and input them again; you can even use parts of the output
6574 All the facilities of Emacs' Shell mode are available for interacting
6575 with your program. In particular, you can send signals the usual
6576 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
6581 _GDBN__ displays source code through Emacs.
6584 Each time _GDBN__ displays a stack frame, Emacs automatically finds the
6585 source file for that frame and puts an arrow (_0__@samp{=>}_1__) at the
6586 left margin of the current line. Emacs uses a separate buffer for
6587 source display, and splits the window to show both your _GDBN__ session
6590 Explicit _GDBN__ @code{list} or search commands still produce output as
6591 usual, but you probably will have no reason to use them.
6594 @emph{Warning:} If the directory where your program resides is not your
6595 current directory, it can be easy to confuse Emacs about the location of
6596 the source files, in which case the auxiliary display buffer will not
6597 appear to show your source. _GDBN__ can find programs by searching your
6598 environment's @code{PATH} variable, so the _GDBN__ input and output
6599 session will proceed normally; but Emacs doesn't get enough information
6600 back from _GDBN__ to locate the source files in this situation. To
6601 avoid this problem, either start _GDBN__ mode from the directory where
6602 your program resides, or specify a full path name when prompted for the
6603 @kbd{M-x gdb} argument.
6605 A similar confusion can result if you use the _GDBN__ @code{file} command to
6606 switch to debugging a program in some other location, from an existing
6607 _GDBN__ buffer in Emacs.
6610 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
6611 you need to call _GDBN__ by a different name (for example, if you keep
6612 several configurations around, with different names) you can set the
6613 Emacs variable @code{gdb-command-name}; for example,
6615 (setq gdb-command-name "mygdb")
6618 (preceded by @kbd{ESC ESC}, or typed in the @code{*scratch*} buffer, or
6619 in your @file{.emacs} file) will make Emacs call the program named
6620 ``@code{mygdb}'' instead.
6622 In the _GDBN__ I/O buffer, you can use these special Emacs commands in
6623 addition to the standard Shell mode commands:
6627 Describe the features of Emacs' _GDBN__ Mode.
6630 Execute to another source line, like the _GDBN__ @code{step} command; also
6631 update the display window to show the current file and location.
6634 Execute to next source line in this function, skipping all function
6635 calls, like the _GDBN__ @code{next} command. Then update the display window
6636 to show the current file and location.
6639 Execute one instruction, like the _GDBN__ @code{stepi} command; update
6640 display window accordingly.
6643 Execute to next instruction, using the _GDBN__ @code{nexti} command; update
6644 display window accordingly.
6647 Execute until exit from the selected stack frame, like the _GDBN__
6648 @code{finish} command.
6651 Continue execution of the program, like the _GDBN__ @code{continue}
6652 command. @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
6655 Go up the number of frames indicated by the numeric argument
6656 (@pxref{Arguments, , Numeric Arguments, emacs, The GNU Emacs Manual}),
6657 like the _GDBN__ @code{up} command. @emph{Warning:} In Emacs v19, this
6658 command is @kbd{C-c C-u}.@refill
6661 Go down the number of frames indicated by the numeric argument, like the
6662 _GDBN__ @code{down} command. @emph{Warning:} In Emacs v19, this command
6666 Read the number where the cursor is positioned, and insert it at the end
6667 of the _GDBN__ I/O buffer. For example, if you wish to disassemble code
6668 around an address that was displayed earlier, type @kbd{disassemble};
6669 then move the cursor to the address display, and pick up the
6670 argument for @code{disassemble} by typing @kbd{C-x &}.
6672 You can customize this further on the fly by defining elements of the list
6673 @code{gdb-print-command}; once it is defined, you can format or
6674 otherwise process numbers picked up by @kbd{C-x &} before they are
6675 inserted. A numeric argument to @kbd{C-x &} will both indicate that you
6676 wish special formatting, and act as an index to pick an element of the
6677 list. If the list element is a string, the number to be inserted is
6678 formatted using the Emacs function @code{format}; otherwise the number
6679 is passed as an argument to the corresponding list element.
6683 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
6684 tells _GDBN__ to set a breakpoint on the source line point is on.
6686 If you accidentally delete the source-display buffer, an easy way to get
6687 it back is to type the command @code{f} in the _GDBN__ buffer, to
6688 request a frame display; when you run under Emacs, this will recreate
6689 the source buffer if necessary to show you the context of the current
6692 The source files displayed in Emacs are in ordinary Emacs buffers
6693 which are visiting the source files in the usual way. You can edit
6694 the files with these buffers if you wish; but keep in mind that _GDBN__
6695 communicates with Emacs in terms of line numbers. If you add or
6696 delete lines from the text, the line numbers that _GDBN__ knows will cease
6697 to correspond properly to the code.
6699 @c The following dropped because Epoch is nonstandard. Reactivate
6700 @c if/when v19 does something similar. ---pesch@cygnus.com 19dec1990
6702 @kindex emacs epoch environment
6706 Version 18 of Emacs has a built-in window system called the @code{epoch}
6707 environment. Users of this environment can use a new command,
6708 @code{inspect} which performs identically to @code{print} except that
6709 each value is printed in its own window.
6712 @node _GDBN__ Bugs, Renamed Commands, Emacs, Top
6713 @chapter Reporting Bugs in _GDBN__
6714 @cindex Bugs in _GDBN__
6715 @cindex Reporting Bugs in _GDBN__
6717 Your bug reports play an essential role in making _GDBN__ reliable.
6719 Reporting a bug may help you by bringing a solution to your problem, or it
6720 may not. But in any case the principal function of a bug report is to help
6721 the entire community by making the next version of _GDBN__ work better. Bug
6722 reports are your contribution to the maintenance of _GDBN__.
6724 In order for a bug report to serve its purpose, you must include the
6725 information that enables us to fix the bug.
6728 * Bug Criteria:: Have You Found a Bug?
6729 * Bug Reporting:: How to Report Bugs
6732 @node Bug Criteria, Bug Reporting, _GDBN__ Bugs, _GDBN__ Bugs
6733 @section Have You Found a Bug?
6734 @cindex Bug Criteria
6736 If you are not sure whether you have found a bug, here are some guidelines:
6740 @cindex Fatal Signal
6742 If the debugger gets a fatal signal, for any input whatever, that is a
6743 _GDBN__ bug. Reliable debuggers never crash.
6746 @cindex error on Valid Input
6747 If _GDBN__ produces an error message for valid input, that is a bug.
6750 @cindex Invalid Input
6751 If _GDBN__ does not produce an error message for invalid input,
6752 that is a bug. However, you should note that your idea of
6753 ``invalid input'' might be our idea of ``an extension'' or ``support
6754 for traditional practice''.
6757 If you are an experienced user of debugging tools, your suggestions
6758 for improvement of _GDBN__ are welcome in any case.
6761 @node Bug Reporting, , Bug Criteria, _GDBN__ Bugs
6762 @section How to Report Bugs
6764 @cindex _GDBN__ Bugs, Reporting
6766 A number of companies and individuals offer support for GNU products.
6767 If you obtained _GDBN__ from a support organization, we recommend you
6768 contact that organization first.
6770 Contact information for many support companies and individuals is
6771 available in the file @file{etc/SERVICE} in the GNU Emacs distribution.
6773 In any event, we also recommend that you send bug reports for _GDBN__ to one
6777 bug-gdb@@prep.ai.mit.edu
6778 @{ucbvax|mit-eddie|uunet@}!prep.ai.mit.edu!bug-gdb
6781 @strong{Do not send bug reports to @samp{info-gdb}, or to
6782 @samp{help-gdb}, or to any newsgroups.} Most users of _GDBN__ do not want to
6783 receive bug reports. Those that do, have arranged to receive @samp{bug-gdb}.
6785 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
6786 serves as a repeater. The mailing list and the newsgroup carry exactly
6787 the same messages. Often people think of posting bug reports to the
6788 newsgroup instead of mailing them. This appears to work, but it has one
6789 problem which can be crucial: a newsgroup posting often lacks a mail
6790 path back to the sender. Thus, if we need to ask for more information,
6791 we may be unable to reach you. For this reason, it is better to send
6792 bug reports to the mailing list.
6794 As a last resort, send bug reports on paper to:
6798 Free Software Foundation
6803 The fundamental principle of reporting bugs usefully is this:
6804 @strong{report all the facts}. If you are not sure whether to state a
6805 fact or leave it out, state it!
6807 Often people omit facts because they think they know what causes the
6808 problem and assume that some details don't matter. Thus, you might
6809 assume that the name of the variable you use in an example does not matter.
6810 Well, probably it doesn't, but one cannot be sure. Perhaps the bug is a
6811 stray memory reference which happens to fetch from the location where that
6812 name is stored in memory; perhaps, if the name were different, the contents
6813 of that location would fool the debugger into doing the right thing despite
6814 the bug. Play it safe and give a specific, complete example. That is the
6815 easiest thing for you to do, and the most helpful.
6817 Keep in mind that the purpose of a bug report is to enable us to fix
6818 the bug if it is new to us. It isn't as important what happens if
6819 the bug is already known. Therefore, always write your bug reports on
6820 the assumption that the bug has not been reported previously.
6822 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6823 bell?'' Those bug reports are useless, and we urge everyone to
6824 @emph{refuse to respond to them} except to chide the sender to report
6827 To enable us to fix the bug, you should include all these things:
6831 The version of _GDBN__. _GDBN__ announces it if you start with no
6832 arguments; you can also print it at any time using @code{show version}.
6834 Without this, we won't know whether there is any point in looking for
6835 the bug in the current version of _GDBN__.
6838 A complete input script, and all necessary source files, that will
6842 What compiler (and its version) was used to compile _GDBN__---e.g.
6846 The command arguments you gave the compiler to compile your example and
6847 observe the bug. For example, did you use @samp{-O}? To guarantee
6848 you won't omit something important, list them all.
6850 If we were to try to guess the arguments, we would probably guess wrong
6851 and then we might not encounter the bug.
6854 The type of machine you are using, and the operating system name and
6858 A description of what behavior you observe that you believe is
6859 incorrect. For example, ``It gets a fatal signal.''
6861 Of course, if the bug is that _GDBN__ gets a fatal signal, then we will
6862 certainly notice it. But if the bug is incorrect output, we might not
6863 notice unless it is glaringly wrong. We are human, after all. You
6864 might as well not give us a chance to make a mistake.
6866 Even if the problem you experience is a fatal signal, you should still
6867 say so explicitly. Suppose something strange is going on, such as,
6868 your copy of _GDBN__ is out of synch, or you have encountered a
6869 bug in the C library on your system. (This has happened!) Your copy
6870 might crash and ours would not. If you told us to expect a crash,
6871 then when ours fails to crash, we would know that the bug was not
6872 happening for us. If you had not told us to expect a crash, then we
6873 would not be able to draw any conclusion from our observations.
6876 If you wish to suggest changes to the _GDBN__ source, send us context
6877 diffs. If you even discuss something in the _GDBN__ source, refer to
6878 it by context, not by line number.
6880 The line numbers in our development sources won't match those in your
6881 sources. Your line numbers would convey no useful information to us.
6885 Here are some things that are not necessary:
6889 A description of the envelope of the bug.
6891 Often people who encounter a bug spend a lot of time investigating
6892 which changes to the input file will make the bug go away and which
6893 changes will not affect it.
6895 This is often time consuming and not very useful, because the way we
6896 will find the bug is by running a single example under the debugger
6897 with breakpoints, not by pure deduction from a series of examples.
6898 We recommend that you save your time for something else.
6900 Of course, if you can find a simpler example to report @emph{instead}
6901 of the original one, that is a convenience for us. Errors in the
6902 output will be easier to spot, running under the debugger will take
6905 However, simplification is not vital; if you don't want to do this,
6906 report the bug anyway and send us the entire test case you used.
6909 A patch for the bug.
6911 A patch for the bug does help us if it is a good one. But don't omit
6912 the necessary information, such as the test case, on the assumption that
6913 a patch is all we need. We might see problems with your patch and decide
6914 to fix the problem another way, or we might not understand it at all.
6916 Sometimes with a program as complicated as _GDBN__ it is very hard to
6917 construct an example that will make the program follow a certain path
6918 through the code. If you don't send us the example, we won't be able
6919 to construct one, so we won't be able to verify that the bug is fixed.
6921 And if we can't understand what bug you are trying to fix, or why your
6922 patch should be an improvement, we won't install it. A test case will
6923 help us to understand.
6926 A guess about what the bug is or what it depends on.
6928 Such guesses are usually wrong. Even we can't guess right about such
6929 things without first using the debugger to find the facts.
6933 @include rdl-apps.texi
6936 @node Renamed Commands, Installing _GDBN__, _GDBN__ Bugs, Top
6937 @appendix Renamed Commands
6939 The following commands were renamed in _GDBN__ 4.0, in order to make the
6940 command set as a whole more consistent and easier to use and remember:
6943 @kindex delete environment
6944 @kindex info copying
6945 @kindex info convenience
6946 @kindex info directories
6947 @kindex info editing
6948 @kindex info history
6949 @kindex info targets
6951 @kindex info version
6952 @kindex info warranty
6953 @kindex set addressprint
6954 @kindex set arrayprint
6955 @kindex set prettyprint
6956 @kindex set screen-height
6957 @kindex set screen-width
6958 @kindex set unionprint
6959 @kindex set vtblprint
6960 @kindex set demangle
6961 @kindex set asm-demangle
6962 @kindex set sevenbit-strings
6963 @kindex set array-max
6965 @kindex set history write
6966 @kindex show addressprint
6967 @kindex show arrayprint
6968 @kindex show prettyprint
6969 @kindex show screen-height
6970 @kindex show screen-width
6971 @kindex show unionprint
6972 @kindex show vtblprint
6973 @kindex show demangle
6974 @kindex show asm-demangle
6975 @kindex show sevenbit-strings
6976 @kindex show array-max
6977 @kindex show caution
6978 @kindex show history write
6983 OLD COMMAND NEW COMMAND
6984 --------------- -------------------------------
6985 add-syms add-symbol-file
6986 delete environment unset environment
6987 info convenience show convenience
6988 info copying show copying
6989 info directories show directories
6990 info editing show commands
6991 info history show values
6992 info targets help target
6993 info values show values
6994 info version show version
6995 info warranty show warranty
6996 set/show addressprint set/show print address
6997 set/show array-max set/show print elements
6998 set/show arrayprint set/show print array
6999 set/show asm-demangle set/show print asm-demangle
7000 set/show caution set/show confirm
7001 set/show demangle set/show print demangle
7002 set/show history write set/show history save
7003 set/show prettyprint set/show print pretty
7004 set/show screen-height set/show height
7005 set/show screen-width set/show width
7006 set/show sevenbit-strings set/show print sevenbit-strings
7007 set/show unionprint set/show print union
7008 set/show vtblprint set/show print vtbl
7010 unset [No longer an alias for delete]
7015 \vskip \parskip\vskip \baselineskip
7016 \halign{\tt #\hfil &\qquad#&\tt #\hfil\cr
7017 {\bf Old Command} &&{\bf New Command}\cr
7018 add-syms &&add-symbol-file\cr
7019 delete environment &&unset environment\cr
7020 info convenience &&show convenience\cr
7021 info copying &&show copying\cr
7022 info directories &&show directories \cr
7023 info editing &&show commands\cr
7024 info history &&show values\cr
7025 info targets &&help target\cr
7026 info values &&show values\cr
7027 info version &&show version\cr
7028 info warranty &&show warranty\cr
7029 set{\rm / }show addressprint &&set{\rm / }show print address\cr
7030 set{\rm / }show array-max &&set{\rm / }show print elements\cr
7031 set{\rm / }show arrayprint &&set{\rm / }show print array\cr
7032 set{\rm / }show asm-demangle &&set{\rm / }show print asm-demangle\cr
7033 set{\rm / }show caution &&set{\rm / }show confirm\cr
7034 set{\rm / }show demangle &&set{\rm / }show print demangle\cr
7035 set{\rm / }show history write &&set{\rm / }show history save\cr
7036 set{\rm / }show prettyprint &&set{\rm / }show print pretty\cr
7037 set{\rm / }show screen-height &&set{\rm / }show height\cr
7038 set{\rm / }show screen-width &&set{\rm / }show width\cr
7039 set{\rm / }show sevenbit-strings &&set{\rm / }show print sevenbit-strings\cr
7040 set{\rm / }show unionprint &&set{\rm / }show print union\cr
7041 set{\rm / }show vtblprint &&set{\rm / }show print vtbl\cr
7043 unset &&\rm(No longer an alias for delete)\cr
7047 @node Installing _GDBN__, Copying, Renamed Commands, Top
7048 @appendix Installing _GDBN__
7049 @cindex configuring _GDBN__
7050 @cindex installation
7052 _GDBN__ comes with a @code{configure} script that automates the process
7053 of preparing _GDBN__ for installation; you can then use @code{make} to
7054 build the @code{_GDBP__} program.
7056 The _GDBP__ distribution includes all the source code you need for
7057 _GDBP__ in a single directory @file{gdb-_GDB_VN__}. That directory in turn
7061 @item gdb-_GDB_VN__/configure @r{(and supporting files)}
7062 script for configuring _GDBN__ and all its supporting libraries.
7064 @item gdb-_GDB_VN__/gdb
7065 the source specific to _GDBN__ itself
7067 @item gdb-_GDB_VN__/bfd
7068 source for the Binary File Descriptor Library
7070 @item gdb-_GDB_VN__/include
7073 @item gdb-_GDB_VN__/libiberty
7074 source for the @samp{-liberty} free software library
7076 @item gdb-_GDB_VN__/readline
7077 source for the GNU command-line interface
7080 It is most convenient to run @code{configure} from the @file{gdb-_GDB_VN__}
7081 directory. The simplest way to configure and build _GDBN__ is the
7085 ./configure @var{host}
7089 where @var{host} is something like @samp{sun4} or @samp{decstation}, that
7090 identifies the platform where _GDBN__ will run. This builds the three
7091 libraries @file{bfd}, @file{readline}, and @file{libiberty}, then
7092 @code{gdb} itself. The configured source files, and the binaries, are
7093 left in the corresponding source directories.
7095 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
7096 system doesn't recognize this automatically when you run a different
7097 shell, you may need to run @code{sh} on it explicitly:
7098 @samp{sh configure @var{host}}.
7100 You can @emph{run} the @code{configure} script from any of the
7101 subordinate directories in the _GDBN__ distribution (if you only want to
7102 configure that subdirectory); but be sure to specify a path to it. For
7103 example, to configure only the @code{bfd} subdirectory,
7105 cd gdb-_GDB_VN__/bfd
7106 ../configure @var{host}
7109 You can install @code{_GDBP__} anywhere; it has no hardwired paths. However,
7110 you should make sure that the shell on your path (named by the
7111 @samp{SHELL} environment variable) is publicly readable; some systems
7112 refuse to let _GDBN__ debug child processes whose programs are not
7113 readable, and _GDBN__ uses the shell to start your program.
7116 * Subdirectories:: Configuration subdirectories
7117 * Config Names:: Specifying names for hosts and targets
7118 * configure Options:: Summary of options for configure
7119 * Formatting Documentation:: How to format and print _GDBN__ documentation
7123 @node Subdirectories, Config Names, Installing _GDBN__, Installing _GDBN__
7124 @section Configuration Subdirectories
7125 If you want to run _GDBN__ versions for several host or target machines,
7126 you'll need a different _GDBP__ compiled for each combination of host
7127 and target. @code{configure} is designed to make this easy by allowing
7128 you to generate each configuration in a separate subdirectory. If your
7129 @code{make} program handles the @samp{VPATH} feature (GNU @code{make}
7130 does), running @code{make} in each of these directories then builds the
7131 _GDBP__ program specified there.
7133 @code{configure} creates these subdirectories for you when you
7134 simultaneously specify several configurations; but it's a good habit
7135 even for a single configuration. You can specify the use of
7136 subdirectories using the @samp{+subdirs} option (abbreviated
7137 @samp{+sub}). For example, you can build _GDBN__ this way on a Sun 4 as
7143 ./configure +sub sun4
7149 When @code{configure} uses subdirectories to build programs or
7150 libraries, it creates nested directories
7151 @file{H-@var{host}/T-@var{target}}. @code{configure} uses these two
7152 directory levels because _GDBN__ can be configured for cross-compiling:
7153 _GDBN__ can run on one machine (the host) while debugging programs that
7154 run on another machine (the target). You specify cross-debugging
7155 targets by giving the @samp{+target=@var{target}} option to
7156 @code{configure}. Specifying only hosts still gives you two levels of
7157 subdirectory for each host, with the same configuration suffix on both;
7158 that is, if you give any number of hosts but no targets, _GDBN__ will be
7159 configured for native debugging on each host. On the other hand,
7160 whenever you specify both hosts and targets on the same command line,
7161 @code{configure} creates all combinations of the hosts and targets you
7164 If you run @code{configure} from a directory (notably,
7165 @file{gdb-_GDB_VN__}) that contains source directories for multiple
7166 libraries or programs, @code{configure} creates the
7167 @file{H-@var{host}/T-@var{target}} subdirectories in each library or
7168 program's source directory. For example, typing:
7171 configure sun4 +target=vxworks960
7174 creates the following directories:
7176 gdb-_GDB_VN__/H-sun4/T-vxworks960
7177 gdb-_GDB_VN__/bfd/H-sun4/T-vxworks960
7178 gdb-_GDB_VN__/gdb/H-sun4/T-vxworks960
7179 gdb-_GDB_VN__/libiberty/H-sun4/T-vxworks960
7180 gdb-_GDB_VN__/readline/H-sun4/T-vxworks960
7183 When you run @code{make} to build a program or library, you must run it
7184 in a configured directory. If you made a single configuration,
7185 without subdirectories, run @code{make} in the source directory.
7186 If you have @file{H-@var{host}/T-@var{target}} subdirectories,
7187 run @code{make} in those subdirectories.
7189 The @code{Makefile} generated by @code{configure} for each source
7190 directory runs recursively, so that typing @code{make} in
7191 @file{gdb-_GDB_VN__} (or in a
7192 @file{gdb-_GDB_VN__/H-@var{host}/T-@var{target}} subdirectory) builds
7193 all the required libraries, then _GDBN__.@refill
7195 When you have multiple hosts or targets configured, you can run
7196 @code{make} on them in parallel (for example, if they are NFS-mounted on
7197 each of the hosts); they will not interfere with each other.
7199 You can also use the @samp{+objdir=@var{altroot}} option to have the
7200 configured files placed in a parallel directory structure rather than
7201 alongside the source files; @pxref{configure Options}.
7203 @node Config Names, configure Options, Subdirectories, Installing _GDBN__
7204 @section Specifying Names for Hosts and Targets
7206 The specifications used for hosts and targets in the @code{configure}
7207 script are based on a three-part naming scheme, but some short predefined
7208 aliases are also supported. The full naming scheme encodes three pieces
7209 of information in the following pattern:
7211 @var{architecture}-@var{vendor}-@var{os}
7214 For example, you can use the alias @code{sun4} as a @var{host} argument
7215 or in a @code{+target=@var{target}} option, but the equivalent full name
7216 is @samp{sparc-sun-sunos4}.
7218 The following table shows all the architectures, hosts, and OS prefixes
7219 that @code{configure} recognizes in _GDBN__ _GDB_VN__. Entries in the ``OS
7220 prefix'' column ending in a @samp{*} may be followed by a release number.
7225 ARCHITECTURE VENDOR OS prefix
7226 ------------+------------+-------------
7229 a29k | amd | amigados
7230 alliant | amdahl | aout
7235 h8300 | bout | dgux*
7236 i386 | bull | dynix*
7239 m68000 | convergent | hds
7240 m68k | convex | hpux*
7243 ns32k | encore | kern
7244 pyramid | gould | mach*
7245 romp | hitachi | msdos*
7246 rs6000 | hp | newsos*
7247 sparc | ibm | nindy*
7248 tahoe | intel | osf*
7250 vax | little | sunos*
7252 ymp | motorola | sym*
7267 @emph{Warning:} Many combinations of architecture, vendor, and OS are
7271 @c FIXME: this table is probably screwed in @smallbook. Try setting
7272 @c FIXME...smallbook fonts?
7275 \advance\baselineskip -1pt
7276 % TERRIBLE KLUGE ABOVE makes table fit on one page (large format, prob
7277 % not smallbook). FIXME Reformat table for next time!!
7278 \vskip \baselineskip
7279 \halign{\hskip\parindent\tt #\hfil &\qquad#&\tt #\hfil &\qquad#&\tt
7280 #\hfil &\qquad\qquad\it #\hfil\cr
7281 {\bf Architecture} &&{\bf Vendor} &&{\bf OS prefix}\cr
7282 \multispan5\hrulefill\cr
7283 580 && altos && aix* \cr
7284 a29k && amd && amigados \cr
7285 alliant && amdahl && aout \cr
7286 arm && aout && bout \cr
7287 c1 && apollo && bsd* \cr
7288 c2 && att && coff \cr
7289 cray2 && bcs && ctix* \cr
7290 h8300 && bout && dgux* \cr
7291 i386 && bull && dynix* \cr
7292 i860 && cbm && ebmon &Warning: \cr
7293 i960 && coff && esix* &Many combinations \cr
7294 m68000 && convergent && hds &of architecture, vendor \cr
7295 m68k && convex && hpux* &and OS are untested. \cr
7296 m88k && cray && irix* \cr
7297 mips && dec && isc* \cr
7298 ns32k && encore && kern \cr
7299 pyramid && gould && mach* \cr
7300 romp && hitachi && msdos* \cr
7301 rs6000 && hp && newsos* \cr
7302 sparc && ibm && nindy* \cr
7303 tahoe && intel && osf* \cr
7304 tron && isi && sco* \cr
7305 vax && little && sunos* \cr
7306 xmp && mips && svr4 \cr
7307 ymp && motorola && sym* \cr
7309 && next && ultrix* \cr
7310 && nyu && unicos* \cr
7312 && sequent && uts \cr
7315 && sun && vxworks* \cr
7322 The @code{configure} script accompanying _GDBN__ _GDB_VN__ does not provide
7323 any query facility to list all supported host and target names or
7324 aliases. @code{configure} calls the Bourne shell script
7325 @code{config.sub} to map abbreviations to full names; you can read the
7326 script, if you wish, or you can use it to test your guesses on
7327 abbreviations---for example:
7329 % sh config.sub sun4
7331 % sh config.sub sun3
7333 % sh config.sub decstation
7335 % sh config.sub hp300bsd
7337 % sh config.sub i386v
7339 % sh config.sub i486v
7340 *** Configuration "i486v" not recognized
7343 @code{config.sub} is also distributed in the directory @file{gdb-_GDB_VN__}.
7345 @node configure Options, Formatting Documentation, Config Names, Installing _GDBN__
7346 @section @code{configure} Options
7348 Here is a summary of all the @code{configure} options and arguments that
7349 you might use for building _GDBN__:
7352 configure @r{[}+destdir=@var{dir}@r{]} @r{[}+subdirs@r{]}
7353 @r{[}+objdir=@var{altroot}@r{]} @r{[}+norecursion@r{]} @r{[}+rm@r{]}
7354 @r{[}+target=@var{target}@dots{}@r{]} @var{host}@dots{}
7357 You may introduce options with the character @samp{-} rather than
7358 @samp{+} if you prefer; but you may abbreviate option names if you use
7362 @item +destdir=@var{dir}
7363 @var{dir} is an installation directory @emph{path prefix}. After you
7364 configure with this option, @code{make install} will install _GDBN__ as
7365 @file{@var{dir}/bin/_GDBP__}, and the libraries in @file{@var{dir}/lib}.
7366 If you specify @samp{+destdir=/usr/local}, for example, @code{make
7367 install} creates @file{/usr/local/bin/gdb}.@refill
7370 Write configuration specific files in subdirectories of the form
7372 H-@var{host}/T-@var{target}
7375 (and configure the @code{Makefile} to generate object code in
7376 subdirectories of this form as well). Without this option, if you
7377 specify only one configuration for _GDBN__, @code{configure} will use
7378 the same directory for source, configured files, and binaries. This
7379 option is used automatically if you specify more than one @var{host} or
7380 more than one @samp{+target=@var{target}} option on the @code{configure}
7384 Configure only the directory where @code{configure} is executed; do not
7385 propagate configuration to subdirectories.
7387 @item +objdir=@var{altroot}
7388 @var{altroot} is an alternative directory used as the root for
7389 configured files. @code{configure} will create directories under
7390 @var{altroot} in parallel to the source directories. If you use
7391 @samp{+objdir=@var{altroot}} with @samp{+subdirs}, @code{configure} also
7392 builds the @samp{H-@var{host}/T-@var{target}} subdirectories in the
7393 directory tree rooted in @var{altroot}.
7397 Remove the configuration that the other arguments specify.
7399 @c This doesn't work (yet if ever). FIXME.
7400 @c @item +parse=@var{lang} @dots{}
7401 @c Configure the _GDBN__ expression parser to parse the listed languages.
7402 @c @samp{all} configures _GDBN__ for all supported languages. To get a
7403 @c list of all supported languages, omit the argument. Without this
7404 @c option, _GDBN__ is configured to parse all supported languages.
7406 @item +target=@var{target} @dots{}
7407 Configure _GDBN__ for cross-debugging programs running on each specified
7408 @var{target}. You may specify as many @samp{+target} options as you
7409 wish. Without this option, _GDBN__ is configured to debug programs that
7410 run on the same machine (@var{host}) as _GDBN__ itself.
7412 There is no convenient way to generate a list of all available targets.
7414 @item @var{host} @dots{}
7415 Configure _GDBN__ to run on each specified @var{host}. You may specify as
7416 many host names as you wish.
7418 There is no convenient way to generate a list of all available hosts.
7422 @code{configure} accepts other options, for compatibility with
7423 configuring other GNU tools recursively; but these are the only
7424 options that affect _GDBN__ or its supporting libraries.
7426 @node Formatting Documentation, , configure Options, Installing _GDBN__
7427 @section Formatting the Documentation
7429 @cindex _GDBN__ reference card
7430 @cindex reference card
7431 The _GDBN__ _GDB_VN__ release includes an already-formatted reference card,
7432 ready for printing on a PostScript printer, as @file{gdb-_GDB_VN__/gdb/refcard.ps}.
7433 It uses the most common PostScript fonts: the Times family, Courier, and
7434 Symbol. If you have a PostScript printer, you can print the reference
7435 card by just sending @file{refcard.ps} to the printer.
7437 The release also includes the online Info version of this manual already
7438 formatted: the main Info file is @file{gdb-_GDB_VN__/gdb/gdb.info}, and it
7439 refers to subordinate files matching @samp{gdb.info*} in the same
7442 If you want to make these Info files yourself from the _GDBN__ manual's
7443 source, you need the GNU @code{makeinfo} program. Once you have it, you
7446 cd gdb-_GDB_VN__/gdb
7450 to make the Info file.
7452 If you want to format and print copies of the manual, you need several
7456 @TeX{}, the public domain typesetting program written by Donald Knuth,
7457 must be installed on your system and available through your execution
7460 @file{gdb-_GDB_VN__/texinfo}: @TeX{} macros defining the GNU
7461 Documentation Format.
7463 @emph{A @sc{dvi} output program.} @TeX{} doesn't actually make marks on
7464 paper; it produces output files called @sc{dvi} files. If your system
7465 has @TeX{} installed, chances are it has a program for printing out
7466 these files; one popular example is @code{dvips}, which can print
7467 @sc{dvi} files on PostScript printers.
7470 Once you have these things, you can type
7472 cd gdb-_GDB_VN__/gdb
7476 to format the text of this manual, and print it with the usual output
7477 method for @TeX{} @sc{dvi} files at your site.
7479 If you want to print the reference card, but don't have a PostScript
7480 printer, or you want to use Computer Modern fonts instead,
7481 you can still print it if you have @TeX{}. Format the reference card by typing
7483 cd gdb-_GDB_VN__/gdb
7488 The _GDBN__ reference card is designed to print in landscape mode on US
7489 ``letter'' size paper; that is, on a sheet 11 inches wide by 8.5 inches
7490 high. You will need to specify this form of printing as an option to
7491 your @sc{dvi} output program.
7494 @node Copying, Index, Installing _GDBN__, Top
7495 @unnumbered GNU GENERAL PUBLIC LICENSE
7496 @center Version 2, June 1991
7499 Copyright @copyright{} 1989, 1991 Free Software Foundation, Inc.
7500 675 Mass Ave, Cambridge, MA 02139, USA
7502 Everyone is permitted to copy and distribute verbatim copies
7503 of this license document, but changing it is not allowed.
7506 @unnumberedsec Preamble
7508 The licenses for most software are designed to take away your
7509 freedom to share and change it. By contrast, the GNU General Public
7510 License is intended to guarantee your freedom to share and change free
7511 software---to make sure the software is free for all its users. This
7512 General Public License applies to most of the Free Software
7513 Foundation's software and to any other program whose authors commit to
7514 using it. (Some other Free Software Foundation software is covered by
7515 the GNU Library General Public License instead.) You can apply it to
7518 When we speak of free software, we are referring to freedom, not
7519 price. Our General Public Licenses are designed to make sure that you
7520 have the freedom to distribute copies of free software (and charge for
7521 this service if you wish), that you receive source code or can get it
7522 if you want it, that you can change the software or use pieces of it
7523 in new free programs; and that you know you can do these things.
7525 To protect your rights, we need to make restrictions that forbid
7526 anyone to deny you these rights or to ask you to surrender the rights.
7527 These restrictions translate to certain responsibilities for you if you
7528 distribute copies of the software, or if you modify it.
7530 For example, if you distribute copies of such a program, whether
7531 gratis or for a fee, you must give the recipients all the rights that
7532 you have. You must make sure that they, too, receive or can get the
7533 source code. And you must show them these terms so they know their
7536 We protect your rights with two steps: (1) copyright the software, and
7537 (2) offer you this license which gives you legal permission to copy,
7538 distribute and/or modify the software.
7540 Also, for each author's protection and ours, we want to make certain
7541 that everyone understands that there is no warranty for this free
7542 software. If the software is modified by someone else and passed on, we
7543 want its recipients to know that what they have is not the original, so
7544 that any problems introduced by others will not reflect on the original
7545 authors' reputations.
7547 Finally, any free program is threatened constantly by software
7548 patents. We wish to avoid the danger that redistributors of a free
7549 program will individually obtain patent licenses, in effect making the
7550 program proprietary. To prevent this, we have made it clear that any
7551 patent must be licensed for everyone's free use or not licensed at all.
7553 The precise terms and conditions for copying, distribution and
7554 modification follow.
7557 @unnumberedsec TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
7560 @center TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
7565 This License applies to any program or other work which contains
7566 a notice placed by the copyright holder saying it may be distributed
7567 under the terms of this General Public License. The ``Program'', below,
7568 refers to any such program or work, and a ``work based on the Program''
7569 means either the Program or any derivative work under copyright law:
7570 that is to say, a work containing the Program or a portion of it,
7571 either verbatim or with modifications and/or translated into another
7572 language. (Hereinafter, translation is included without limitation in
7573 the term ``modification''.) Each licensee is addressed as ``you''.
7575 Activities other than copying, distribution and modification are not
7576 covered by this License; they are outside its scope. The act of
7577 running the Program is not restricted, and the output from the Program
7578 is covered only if its contents constitute a work based on the
7579 Program (independent of having been made by running the Program).
7580 Whether that is true depends on what the Program does.
7583 You may copy and distribute verbatim copies of the Program's
7584 source code as you receive it, in any medium, provided that you
7585 conspicuously and appropriately publish on each copy an appropriate
7586 copyright notice and disclaimer of warranty; keep intact all the
7587 notices that refer to this License and to the absence of any warranty;
7588 and give any other recipients of the Program a copy of this License
7589 along with the Program.
7591 You may charge a fee for the physical act of transferring a copy, and
7592 you may at your option offer warranty protection in exchange for a fee.
7595 You may modify your copy or copies of the Program or any portion
7596 of it, thus forming a work based on the Program, and copy and
7597 distribute such modifications or work under the terms of Section 1
7598 above, provided that you also meet all of these conditions:
7602 You must cause the modified files to carry prominent notices
7603 stating that you changed the files and the date of any change.
7606 You must cause any work that you distribute or publish, that in
7607 whole or in part contains or is derived from the Program or any
7608 part thereof, to be licensed as a whole at no charge to all third
7609 parties under the terms of this License.
7612 If the modified program normally reads commands interactively
7613 when run, you must cause it, when started running for such
7614 interactive use in the most ordinary way, to print or display an
7615 announcement including an appropriate copyright notice and a
7616 notice that there is no warranty (or else, saying that you provide
7617 a warranty) and that users may redistribute the program under
7618 these conditions, and telling the user how to view a copy of this
7619 License. (Exception: if the Program itself is interactive but
7620 does not normally print such an announcement, your work based on
7621 the Program is not required to print an announcement.)
7624 These requirements apply to the modified work as a whole. If
7625 identifiable sections of that work are not derived from the Program,
7626 and can be reasonably considered independent and separate works in
7627 themselves, then this License, and its terms, do not apply to those
7628 sections when you distribute them as separate works. But when you
7629 distribute the same sections as part of a whole which is a work based
7630 on the Program, the distribution of the whole must be on the terms of
7631 this License, whose permissions for other licensees extend to the
7632 entire whole, and thus to each and every part regardless of who wrote it.
7634 Thus, it is not the intent of this section to claim rights or contest
7635 your rights to work written entirely by you; rather, the intent is to
7636 exercise the right to control the distribution of derivative or
7637 collective works based on the Program.
7639 In addition, mere aggregation of another work not based on the Program
7640 with the Program (or with a work based on the Program) on a volume of
7641 a storage or distribution medium does not bring the other work under
7642 the scope of this License.
7645 You may copy and distribute the Program (or a work based on it,
7646 under Section 2) in object code or executable form under the terms of
7647 Sections 1 and 2 above provided that you also do one of the following:
7651 Accompany it with the complete corresponding machine-readable
7652 source code, which must be distributed under the terms of Sections
7653 1 and 2 above on a medium customarily used for software interchange; or,
7656 Accompany it with a written offer, valid for at least three
7657 years, to give any third party, for a charge no more than your
7658 cost of physically performing source distribution, a complete
7659 machine-readable copy of the corresponding source code, to be
7660 distributed under the terms of Sections 1 and 2 above on a medium
7661 customarily used for software interchange; or,
7664 Accompany it with the information you received as to the offer
7665 to distribute corresponding source code. (This alternative is
7666 allowed only for noncommercial distribution and only if you
7667 received the program in object code or executable form with such
7668 an offer, in accord with Subsection b above.)
7671 The source code for a work means the preferred form of the work for
7672 making modifications to it. For an executable work, complete source
7673 code means all the source code for all modules it contains, plus any
7674 associated interface definition files, plus the scripts used to
7675 control compilation and installation of the executable. However, as a
7676 special exception, the source code distributed need not include
7677 anything that is normally distributed (in either source or binary
7678 form) with the major components (compiler, kernel, and so on) of the
7679 operating system on which the executable runs, unless that component
7680 itself accompanies the executable.
7682 If distribution of executable or object code is made by offering
7683 access to copy from a designated place, then offering equivalent
7684 access to copy the source code from the same place counts as
7685 distribution of the source code, even though third parties are not
7686 compelled to copy the source along with the object code.
7689 You may not copy, modify, sublicense, or distribute the Program
7690 except as expressly provided under this License. Any attempt
7691 otherwise to copy, modify, sublicense or distribute the Program is
7692 void, and will automatically terminate your rights under this License.
7693 However, parties who have received copies, or rights, from you under
7694 this License will not have their licenses terminated so long as such
7695 parties remain in full compliance.
7698 You are not required to accept this License, since you have not
7699 signed it. However, nothing else grants you permission to modify or
7700 distribute the Program or its derivative works. These actions are
7701 prohibited by law if you do not accept this License. Therefore, by
7702 modifying or distributing the Program (or any work based on the
7703 Program), you indicate your acceptance of this License to do so, and
7704 all its terms and conditions for copying, distributing or modifying
7705 the Program or works based on it.
7708 Each time you redistribute the Program (or any work based on the
7709 Program), the recipient automatically receives a license from the
7710 original licensor to copy, distribute or modify the Program subject to
7711 these terms and conditions. You may not impose any further
7712 restrictions on the recipients' exercise of the rights granted herein.
7713 You are not responsible for enforcing compliance by third parties to
7717 If, as a consequence of a court judgment or allegation of patent
7718 infringement or for any other reason (not limited to patent issues),
7719 conditions are imposed on you (whether by court order, agreement or
7720 otherwise) that contradict the conditions of this License, they do not
7721 excuse you from the conditions of this License. If you cannot
7722 distribute so as to satisfy simultaneously your obligations under this
7723 License and any other pertinent obligations, then as a consequence you
7724 may not distribute the Program at all. For example, if a patent
7725 license would not permit royalty-free redistribution of the Program by
7726 all those who receive copies directly or indirectly through you, then
7727 the only way you could satisfy both it and this License would be to
7728 refrain entirely from distribution of the Program.
7730 If any portion of this section is held invalid or unenforceable under
7731 any particular circumstance, the balance of the section is intended to
7732 apply and the section as a whole is intended to apply in other
7735 It is not the purpose of this section to induce you to infringe any
7736 patents or other property right claims or to contest validity of any
7737 such claims; this section has the sole purpose of protecting the
7738 integrity of the free software distribution system, which is
7739 implemented by public license practices. Many people have made
7740 generous contributions to the wide range of software distributed
7741 through that system in reliance on consistent application of that
7742 system; it is up to the author/donor to decide if he or she is willing
7743 to distribute software through any other system and a licensee cannot
7746 This section is intended to make thoroughly clear what is believed to
7747 be a consequence of the rest of this License.
7750 If the distribution and/or use of the Program is restricted in
7751 certain countries either by patents or by copyrighted interfaces, the
7752 original copyright holder who places the Program under this License
7753 may add an explicit geographical distribution limitation excluding
7754 those countries, so that distribution is permitted only in or among
7755 countries not thus excluded. In such case, this License incorporates
7756 the limitation as if written in the body of this License.
7759 The Free Software Foundation may publish revised and/or new versions
7760 of the General Public License from time to time. Such new versions will
7761 be similar in spirit to the present version, but may differ in detail to
7762 address new problems or concerns.
7764 Each version is given a distinguishing version number. If the Program
7765 specifies a version number of this License which applies to it and ``any
7766 later version'', you have the option of following the terms and conditions
7767 either of that version or of any later version published by the Free
7768 Software Foundation. If the Program does not specify a version number of
7769 this License, you may choose any version ever published by the Free Software
7773 If you wish to incorporate parts of the Program into other free
7774 programs whose distribution conditions are different, write to the author
7775 to ask for permission. For software which is copyrighted by the Free
7776 Software Foundation, write to the Free Software Foundation; we sometimes
7777 make exceptions for this. Our decision will be guided by the two goals
7778 of preserving the free status of all derivatives of our free software and
7779 of promoting the sharing and reuse of software generally.
7782 @heading NO WARRANTY
7789 BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
7790 FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
7791 OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
7792 PROVIDE THE PROGRAM ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
7793 OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
7794 MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
7795 TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
7796 PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
7797 REPAIR OR CORRECTION.
7800 IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
7801 WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
7802 REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
7803 INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
7804 OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
7805 TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
7806 YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
7807 PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
7808 POSSIBILITY OF SUCH DAMAGES.
7812 @heading END OF TERMS AND CONDITIONS
7815 @center END OF TERMS AND CONDITIONS
7819 @unnumberedsec Applying These Terms to Your New Programs
7821 If you develop a new program, and you want it to be of the greatest
7822 possible use to the public, the best way to achieve this is to make it
7823 free software which everyone can redistribute and change under these terms.
7825 To do so, attach the following notices to the program. It is safest
7826 to attach them to the start of each source file to most effectively
7827 convey the exclusion of warranty; and each file should have at least
7828 the ``copyright'' line and a pointer to where the full notice is found.
7831 @var{one line to give the program's name and a brief idea of what it does.}
7832 Copyright (C) 19@var{yy} @var{name of author}
7834 This program is free software; you can redistribute it and/or modify
7835 it under the terms of the GNU General Public License as published by
7836 the Free Software Foundation; either version 2 of the License, or
7837 (at your option) any later version.
7839 This program is distributed in the hope that it will be useful,
7840 but WITHOUT ANY WARRANTY; without even the implied warranty of
7841 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
7842 GNU General Public License for more details.
7844 You should have received a copy of the GNU General Public License
7845 along with this program; if not, write to the Free Software
7846 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
7849 Also add information on how to contact you by electronic and paper mail.
7851 If the program is interactive, make it output a short notice like this
7852 when it starts in an interactive mode:
7855 Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author}
7856 Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
7857 This is free software, and you are welcome to redistribute it
7858 under certain conditions; type `show c' for details.
7861 The hypothetical commands @samp{show w} and @samp{show c} should show
7862 the appropriate parts of the General Public License. Of course, the
7863 commands you use may be called something other than @samp{show w} and
7864 @samp{show c}; they could even be mouse-clicks or menu items---whatever
7867 You should also get your employer (if you work as a programmer) or your
7868 school, if any, to sign a ``copyright disclaimer'' for the program, if
7869 necessary. Here is a sample; alter the names:
7872 Yoyodyne, Inc., hereby disclaims all copyright interest in the program
7873 `Gnomovision' (which makes passes at compilers) written by James Hacker.
7875 @var{signature of Ty Coon}, 1 April 1989
7876 Ty Coon, President of Vice
7879 This General Public License does not permit incorporating your program into
7880 proprietary programs. If your program is a subroutine library, you may
7881 consider it more useful to permit linking proprietary applications with the
7882 library. If this is what you want to do, use the GNU Library General
7883 Public License instead of this License.
7886 @node Index, , Copying, Top
7892 % I think something like @colophon should be in texinfo. In the
7894 \long\def\colophon{\hbox to0pt{}\vfill
7895 \centerline{The body of this manual is set in}
7896 \centerline{\fontname\tenrm,}
7897 \centerline{with headings in {\bf\fontname\tenbf}}
7898 \centerline{and examples in {\tt\fontname\tentt}.}
7899 \centerline{{\it\fontname\tenit\/} and}
7900 \centerline{{\sl\fontname\tensl\/}}
7901 \centerline{are used for emphasis.}\vfill}
7903 % Blame: pesch@cygnus.com, 28mar91.