1 \input texinfo @c -*-texinfo-*-
2 @c Copyright 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1998,
4 @c Free Software Foundation, Inc.
7 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
8 @c of @set vars. However, you can override filename with makeinfo -o.
13 @settitle Debugging with @value{GDBN}
14 @setchapternewpage odd
25 @c readline appendices use @vindex, @findex and @ftable,
26 @c annotate.texi and gdbmi use @findex.
30 @c !!set GDB manual's edition---not the same as GDB version!
33 @c !!set GDB manual's revision date
34 @set DATE December 2001
36 @c THIS MANUAL REQUIRES TEXINFO 3.12 OR LATER.
38 @c This is a dir.info fragment to support semi-automated addition of
39 @c manuals to an info tree.
40 @dircategory Programming & development tools.
42 * Gdb: (gdb). The @sc{gnu} debugger.
46 This file documents the @sc{gnu} debugger @value{GDBN}.
49 This is the @value{EDITION} Edition, @value{DATE},
50 of @cite{Debugging with @value{GDBN}: the @sc{gnu} Source-Level Debugger}
51 for @value{GDBN} Version @value{GDBVN}.
53 Copyright (C) 1988,1989,1990,1991,1992,1993,1994,1995,1996,1998,1999,2000,2001
54 Free Software Foundation, Inc.
56 Permission is granted to copy, distribute and/or modify this document
57 under the terms of the GNU Free Documentation License, Version 1.1 or
58 any later version published by the Free Software Foundation; with the
59 Invariant Sections being ``Free Software'' and ``Free Software Needs
60 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
61 and with the Back-Cover Texts as in (a) below.
63 (a) The FSF's Back-Cover Text is: ``You have freedom to copy and modify
64 this GNU Manual, like GNU software. Copies published by the Free
65 Software Foundation raise funds for GNU development.''
69 @title Debugging with @value{GDBN}
70 @subtitle The @sc{gnu} Source-Level Debugger
72 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
73 @subtitle @value{DATE}
74 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
78 \hfill (Send bugs and comments on @value{GDBN} to bug-gdb\@gnu.org.)\par
79 \hfill {\it Debugging with @value{GDBN}}\par
80 \hfill \TeX{}info \texinfoversion\par
84 @vskip 0pt plus 1filll
85 Copyright @copyright{} 1988,1989,1990,1991,1992,1993,1994,1995,1996,1998,1999,2000,2001
86 Free Software Foundation, Inc.
88 Published by the Free Software Foundation @*
89 59 Temple Place - Suite 330, @*
90 Boston, MA 02111-1307 USA @*
93 Permission is granted to copy, distribute and/or modify this document
94 under the terms of the GNU Free Documentation License, Version 1.1 or
95 any later version published by the Free Software Foundation; with the
96 Invariant Sections being ``Free Software'' and ``Free Software Needs
97 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
98 and with the Back-Cover Texts as in (a) below.
100 (a) The FSF's Back-Cover Text is: ``You have freedom to copy and modify
101 this GNU Manual, like GNU software. Copies published by the Free
102 Software Foundation raise funds for GNU development.''
107 @node Top, Summary, (dir), (dir)
109 @top Debugging with @value{GDBN}
111 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
113 This is the @value{EDITION} Edition, @value{DATE}, for @value{GDBN} Version
116 Copyright (C) 1988-2001 Free Software Foundation, Inc.
119 * Summary:: Summary of @value{GDBN}
120 * Sample Session:: A sample @value{GDBN} session
122 * Invocation:: Getting in and out of @value{GDBN}
123 * Commands:: @value{GDBN} commands
124 * Running:: Running programs under @value{GDBN}
125 * Stopping:: Stopping and continuing
126 * Stack:: Examining the stack
127 * Source:: Examining source files
128 * Data:: Examining data
129 * Tracepoints:: Debugging remote targets non-intrusively
130 * Overlays:: Debugging programs that use overlays
132 * Languages:: Using @value{GDBN} with different languages
134 * Symbols:: Examining the symbol table
135 * Altering:: Altering execution
136 * GDB Files:: @value{GDBN} files
137 * Targets:: Specifying a debugging target
138 * Configurations:: Configuration-specific information
139 * Controlling GDB:: Controlling @value{GDBN}
140 * Sequences:: Canned sequences of commands
141 * TUI:: @value{GDBN} Text User Interface
142 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
143 * Annotations:: @value{GDBN}'s annotation interface.
144 * GDB/MI:: @value{GDBN}'s Machine Interface.
146 * GDB Bugs:: Reporting bugs in @value{GDBN}
147 * Formatting Documentation:: How to format and print @value{GDBN} documentation
149 * Command Line Editing:: Command Line Editing
150 * Using History Interactively:: Using History Interactively
151 * Installing GDB:: Installing GDB
157 @c the replication sucks, but this avoids a texinfo 3.12 lameness
162 @top Debugging with @value{GDBN}
164 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
166 This is the @value{EDITION} Edition, @value{DATE}, for @value{GDBN} Version
169 Copyright (C) 1988-2000 Free Software Foundation, Inc.
172 * Summary:: Summary of @value{GDBN}
173 * Sample Session:: A sample @value{GDBN} session
175 * Invocation:: Getting in and out of @value{GDBN}
176 * Commands:: @value{GDBN} commands
177 * Running:: Running programs under @value{GDBN}
178 * Stopping:: Stopping and continuing
179 * Stack:: Examining the stack
180 * Source:: Examining source files
181 * Data:: Examining data
182 * Tracepoints:: Debugging remote targets non-intrusively
183 * Overlays:: Debugging programs that use overlays
185 * Languages:: Using @value{GDBN} with different languages
187 * Symbols:: Examining the symbol table
188 * Altering:: Altering execution
189 * GDB Files:: @value{GDBN} files
190 * Targets:: Specifying a debugging target
191 * Configurations:: Configuration-specific information
192 * Controlling GDB:: Controlling @value{GDBN}
193 * Sequences:: Canned sequences of commands
194 * TUI:: @value{GDBN} Text User Interface
195 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
196 * Annotations:: @value{GDBN}'s annotation interface.
197 * GDB/MI:: @value{GDBN}'s Machine Interface.
199 * GDB Bugs:: Reporting bugs in @value{GDBN}
200 * Formatting Documentation:: How to format and print @value{GDBN} documentation
202 * Command Line Editing:: Command Line Editing
203 * Using History Interactively:: Using History Interactively
204 * Installing GDB:: Installing GDB
210 @c TeX can handle the contents at the start but makeinfo 3.12 can not
216 @unnumbered Summary of @value{GDBN}
218 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
219 going on ``inside'' another program while it executes---or what another
220 program was doing at the moment it crashed.
222 @value{GDBN} can do four main kinds of things (plus other things in support of
223 these) to help you catch bugs in the act:
227 Start your program, specifying anything that might affect its behavior.
230 Make your program stop on specified conditions.
233 Examine what has happened, when your program has stopped.
236 Change things in your program, so you can experiment with correcting the
237 effects of one bug and go on to learn about another.
240 You can use @value{GDBN} to debug programs written in C and C++.
241 For more information, see @ref{Support,,Supported languages}.
242 For more information, see @ref{C,,C and C++}.
246 Support for Modula-2 and Chill is partial. For information on Modula-2,
247 see @ref{Modula-2,,Modula-2}. For information on Chill, see @ref{Chill}.
250 Debugging Pascal programs which use sets, subranges, file variables, or
251 nested functions does not currently work. @value{GDBN} does not support
252 entering expressions, printing values, or similar features using Pascal
256 @value{GDBN} can be used to debug programs written in Fortran, although
257 it may be necessary to refer to some variables with a trailing
261 * Free Software:: Freely redistributable software
262 * Contributors:: Contributors to GDB
266 @unnumberedsec Free software
268 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
269 General Public License
270 (GPL). The GPL gives you the freedom to copy or adapt a licensed
271 program---but every person getting a copy also gets with it the
272 freedom to modify that copy (which means that they must get access to
273 the source code), and the freedom to distribute further copies.
274 Typical software companies use copyrights to limit your freedoms; the
275 Free Software Foundation uses the GPL to preserve these freedoms.
277 Fundamentally, the General Public License is a license which says that
278 you have these freedoms and that you cannot take these freedoms away
281 @unnumbered Free Software Needs Free Documentation
283 The biggest deficiency in the free software community today is not in
284 the software---it is the lack of good free documentation that we can
285 include with the free software. Many of our most important
286 programs do not come with free reference manuals and free introductory
287 texts. Documentation is an essential part of any software package;
288 when an important free software package does not come with a free
289 manual and a free tutorial, that is a major gap. We have many such
292 Consider Perl, for instance. The tutorial manuals that people
293 normally use are non-free. How did this come about? Because the
294 authors of those manuals published them with restrictive terms---no
295 copying, no modification, source files not available---which exclude
296 them from the free software world.
298 That wasn't the first time this sort of thing happened, and it was far
299 from the last. Many times we have heard a GNU user eagerly describe a
300 manual that he is writing, his intended contribution to the community,
301 only to learn that he had ruined everything by signing a publication
302 contract to make it non-free.
304 Free documentation, like free software, is a matter of freedom, not
305 price. The problem with the non-free manual is not that publishers
306 charge a price for printed copies---that in itself is fine. (The Free
307 Software Foundation sells printed copies of manuals, too.) The
308 problem is the restrictions on the use of the manual. Free manuals
309 are available in source code form, and give you permission to copy and
310 modify. Non-free manuals do not allow this.
312 The criteria of freedom for a free manual are roughly the same as for
313 free software. Redistribution (including the normal kinds of
314 commercial redistribution) must be permitted, so that the manual can
315 accompany every copy of the program, both on-line and on paper.
317 Permission for modification of the technical content is crucial too.
318 When people modify the software, adding or changing features, if they
319 are conscientious they will change the manual too---so they can
320 provide accurate and clear documentation for the modified program. A
321 manual that leaves you no choice but to write a new manual to document
322 a changed version of the program is not really available to our
325 Some kinds of limits on the way modification is handled are
326 acceptable. For example, requirements to preserve the original
327 author's copyright notice, the distribution terms, or the list of
328 authors, are ok. It is also no problem to require modified versions
329 to include notice that they were modified. Even entire sections that
330 may not be deleted or changed are acceptable, as long as they deal
331 with nontechnical topics (like this one). These kinds of restrictions
332 are acceptable because they don't obstruct the community's normal use
335 However, it must be possible to modify all the @emph{technical}
336 content of the manual, and then distribute the result in all the usual
337 media, through all the usual channels. Otherwise, the restrictions
338 obstruct the use of the manual, it is not free, and we need another
339 manual to replace it.
341 Please spread the word about this issue. Our community continues to
342 lose manuals to proprietary publishing. If we spread the word that
343 free software needs free reference manuals and free tutorials, perhaps
344 the next person who wants to contribute by writing documentation will
345 realize, before it is too late, that only free manuals contribute to
346 the free software community.
348 If you are writing documentation, please insist on publishing it under
349 the GNU Free Documentation License or another free documentation
350 license. Remember that this decision requires your approval---you
351 don't have to let the publisher decide. Some commercial publishers
352 will use a free license if you insist, but they will not propose the
353 option; it is up to you to raise the issue and say firmly that this is
354 what you want. If the publisher you are dealing with refuses, please
355 try other publishers. If you're not sure whether a proposed license
356 is free, write to @email{icensing@@gnu.org}.
358 You can encourage commercial publishers to sell more free, copylefted
359 manuals and tutorials by buying them, and particularly by buying
360 copies from the publishers that paid for their writing or for major
361 improvements. Meanwhile, try to avoid buying non-free documentation
362 at all. Check the distribution terms of a manual before you buy it,
363 and insist that whoever seeks your business must respect your freedom.
364 Check the history of the book, and try reward the publishers that have
365 paid or pay the authors to work on it.
367 The Free Software Foundation maintains a list of free documentation
368 published by other publishers, at
369 @url{http://www.fsf.org/doc/other-free-books.html}.
372 @unnumberedsec Contributors to @value{GDBN}
374 Richard Stallman was the original author of @value{GDBN}, and of many
375 other @sc{gnu} programs. Many others have contributed to its
376 development. This section attempts to credit major contributors. One
377 of the virtues of free software is that everyone is free to contribute
378 to it; with regret, we cannot actually acknowledge everyone here. The
379 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
380 blow-by-blow account.
382 Changes much prior to version 2.0 are lost in the mists of time.
385 @emph{Plea:} Additions to this section are particularly welcome. If you
386 or your friends (or enemies, to be evenhanded) have been unfairly
387 omitted from this list, we would like to add your names!
390 So that they may not regard their many labors as thankless, we
391 particularly thank those who shepherded @value{GDBN} through major
393 Andrew Cagney (releases 5.0 and 5.1);
394 Jim Blandy (release 4.18);
395 Jason Molenda (release 4.17);
396 Stan Shebs (release 4.14);
397 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
398 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
399 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
400 Jim Kingdon (releases 3.5, 3.4, and 3.3);
401 and Randy Smith (releases 3.2, 3.1, and 3.0).
403 Richard Stallman, assisted at various times by Peter TerMaat, Chris
404 Hanson, and Richard Mlynarik, handled releases through 2.8.
406 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
407 in @value{GDBN}, with significant additional contributions from Per
408 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
409 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
410 much general update work leading to release 3.0).
412 @value{GDBN} uses the BFD subroutine library to examine multiple
413 object-file formats; BFD was a joint project of David V.
414 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
416 David Johnson wrote the original COFF support; Pace Willison did
417 the original support for encapsulated COFF.
419 Brent Benson of Harris Computer Systems contributed DWARF2 support.
421 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
422 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
424 Jean-Daniel Fekete contributed Sun 386i support.
425 Chris Hanson improved the HP9000 support.
426 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
427 David Johnson contributed Encore Umax support.
428 Jyrki Kuoppala contributed Altos 3068 support.
429 Jeff Law contributed HP PA and SOM support.
430 Keith Packard contributed NS32K support.
431 Doug Rabson contributed Acorn Risc Machine support.
432 Bob Rusk contributed Harris Nighthawk CX-UX support.
433 Chris Smith contributed Convex support (and Fortran debugging).
434 Jonathan Stone contributed Pyramid support.
435 Michael Tiemann contributed SPARC support.
436 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
437 Pace Willison contributed Intel 386 support.
438 Jay Vosburgh contributed Symmetry support.
440 Andreas Schwab contributed M68K Linux support.
442 Rich Schaefer and Peter Schauer helped with support of SunOS shared
445 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
446 about several machine instruction sets.
448 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
449 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
450 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
451 and RDI targets, respectively.
453 Brian Fox is the author of the readline libraries providing
454 command-line editing and command history.
456 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
457 Modula-2 support, and contributed the Languages chapter of this manual.
459 Fred Fish wrote most of the support for Unix System Vr4.
460 He also enhanced the command-completion support to cover C@t{++} overloaded
463 Hitachi America, Ltd. sponsored the support for H8/300, H8/500, and
466 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
468 Mitsubishi sponsored the support for D10V, D30V, and M32R/D processors.
470 Toshiba sponsored the support for the TX39 Mips processor.
472 Matsushita sponsored the support for the MN10200 and MN10300 processors.
474 Fujitsu sponsored the support for SPARClite and FR30 processors.
476 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
479 Michael Snyder added support for tracepoints.
481 Stu Grossman wrote gdbserver.
483 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
484 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
486 The following people at the Hewlett-Packard Company contributed
487 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
488 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
489 compiler, and the terminal user interface: Ben Krepp, Richard Title,
490 John Bishop, Susan Macchia, Kathy Mann, Satish Pai, India Paul, Steve
491 Rehrauer, and Elena Zannoni. Kim Haase provided HP-specific
492 information in this manual.
494 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
495 Robert Hoehne made significant contributions to the DJGPP port.
497 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
498 development since 1991. Cygnus engineers who have worked on @value{GDBN}
499 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
500 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
501 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
502 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
503 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
504 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
505 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
506 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
507 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
508 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
509 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
510 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
511 Zuhn have made contributions both large and small.
515 @chapter A Sample @value{GDBN} Session
517 You can use this manual at your leisure to read all about @value{GDBN}.
518 However, a handful of commands are enough to get started using the
519 debugger. This chapter illustrates those commands.
522 In this sample session, we emphasize user input like this: @b{input},
523 to make it easier to pick out from the surrounding output.
526 @c FIXME: this example may not be appropriate for some configs, where
527 @c FIXME...primary interest is in remote use.
529 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
530 processor) exhibits the following bug: sometimes, when we change its
531 quote strings from the default, the commands used to capture one macro
532 definition within another stop working. In the following short @code{m4}
533 session, we define a macro @code{foo} which expands to @code{0000}; we
534 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
535 same thing. However, when we change the open quote string to
536 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
537 procedure fails to define a new synonym @code{baz}:
546 @b{define(bar,defn(`foo'))}
550 @b{changequote(<QUOTE>,<UNQUOTE>)}
552 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
555 m4: End of input: 0: fatal error: EOF in string
559 Let us use @value{GDBN} to try to see what is going on.
562 $ @b{@value{GDBP} m4}
563 @c FIXME: this falsifies the exact text played out, to permit smallbook
564 @c FIXME... format to come out better.
565 @value{GDBN} is free software and you are welcome to distribute copies
566 of it under certain conditions; type "show copying" to see
568 There is absolutely no warranty for @value{GDBN}; type "show warranty"
571 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
576 @value{GDBN} reads only enough symbol data to know where to find the
577 rest when needed; as a result, the first prompt comes up very quickly.
578 We now tell @value{GDBN} to use a narrower display width than usual, so
579 that examples fit in this manual.
582 (@value{GDBP}) @b{set width 70}
586 We need to see how the @code{m4} built-in @code{changequote} works.
587 Having looked at the source, we know the relevant subroutine is
588 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
589 @code{break} command.
592 (@value{GDBP}) @b{break m4_changequote}
593 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
597 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
598 control; as long as control does not reach the @code{m4_changequote}
599 subroutine, the program runs as usual:
602 (@value{GDBP}) @b{run}
603 Starting program: /work/Editorial/gdb/gnu/m4/m4
611 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
612 suspends execution of @code{m4}, displaying information about the
613 context where it stops.
616 @b{changequote(<QUOTE>,<UNQUOTE>)}
618 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
620 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
624 Now we use the command @code{n} (@code{next}) to advance execution to
625 the next line of the current function.
629 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
634 @code{set_quotes} looks like a promising subroutine. We can go into it
635 by using the command @code{s} (@code{step}) instead of @code{next}.
636 @code{step} goes to the next line to be executed in @emph{any}
637 subroutine, so it steps into @code{set_quotes}.
641 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
643 530 if (lquote != def_lquote)
647 The display that shows the subroutine where @code{m4} is now
648 suspended (and its arguments) is called a stack frame display. It
649 shows a summary of the stack. We can use the @code{backtrace}
650 command (which can also be spelled @code{bt}), to see where we are
651 in the stack as a whole: the @code{backtrace} command displays a
652 stack frame for each active subroutine.
655 (@value{GDBP}) @b{bt}
656 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
658 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
660 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
661 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
663 #4 0x79dc in expand_input () at macro.c:40
664 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
668 We step through a few more lines to see what happens. The first two
669 times, we can use @samp{s}; the next two times we use @code{n} to avoid
670 falling into the @code{xstrdup} subroutine.
674 0x3b5c 532 if (rquote != def_rquote)
676 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
677 def_lquote : xstrdup(lq);
679 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
682 538 len_lquote = strlen(rquote);
686 The last line displayed looks a little odd; we can examine the variables
687 @code{lquote} and @code{rquote} to see if they are in fact the new left
688 and right quotes we specified. We use the command @code{p}
689 (@code{print}) to see their values.
692 (@value{GDBP}) @b{p lquote}
693 $1 = 0x35d40 "<QUOTE>"
694 (@value{GDBP}) @b{p rquote}
695 $2 = 0x35d50 "<UNQUOTE>"
699 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
700 To look at some context, we can display ten lines of source
701 surrounding the current line with the @code{l} (@code{list}) command.
707 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
709 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
712 538 len_lquote = strlen(rquote);
713 539 len_rquote = strlen(lquote);
720 Let us step past the two lines that set @code{len_lquote} and
721 @code{len_rquote}, and then examine the values of those variables.
725 539 len_rquote = strlen(lquote);
728 (@value{GDBP}) @b{p len_lquote}
730 (@value{GDBP}) @b{p len_rquote}
735 That certainly looks wrong, assuming @code{len_lquote} and
736 @code{len_rquote} are meant to be the lengths of @code{lquote} and
737 @code{rquote} respectively. We can set them to better values using
738 the @code{p} command, since it can print the value of
739 any expression---and that expression can include subroutine calls and
743 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
745 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
750 Is that enough to fix the problem of using the new quotes with the
751 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
752 executing with the @code{c} (@code{continue}) command, and then try the
753 example that caused trouble initially:
759 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
766 Success! The new quotes now work just as well as the default ones. The
767 problem seems to have been just the two typos defining the wrong
768 lengths. We allow @code{m4} exit by giving it an EOF as input:
772 Program exited normally.
776 The message @samp{Program exited normally.} is from @value{GDBN}; it
777 indicates @code{m4} has finished executing. We can end our @value{GDBN}
778 session with the @value{GDBN} @code{quit} command.
781 (@value{GDBP}) @b{quit}
785 @chapter Getting In and Out of @value{GDBN}
787 This chapter discusses how to start @value{GDBN}, and how to get out of it.
791 type @samp{@value{GDBP}} to start @value{GDBN}.
793 type @kbd{quit} or @kbd{C-d} to exit.
797 * Invoking GDB:: How to start @value{GDBN}
798 * Quitting GDB:: How to quit @value{GDBN}
799 * Shell Commands:: How to use shell commands inside @value{GDBN}
803 @section Invoking @value{GDBN}
805 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
806 @value{GDBN} reads commands from the terminal until you tell it to exit.
808 You can also run @code{@value{GDBP}} with a variety of arguments and options,
809 to specify more of your debugging environment at the outset.
811 The command-line options described here are designed
812 to cover a variety of situations; in some environments, some of these
813 options may effectively be unavailable.
815 The most usual way to start @value{GDBN} is with one argument,
816 specifying an executable program:
819 @value{GDBP} @var{program}
823 You can also start with both an executable program and a core file
827 @value{GDBP} @var{program} @var{core}
830 You can, instead, specify a process ID as a second argument, if you want
831 to debug a running process:
834 @value{GDBP} @var{program} 1234
838 would attach @value{GDBN} to process @code{1234} (unless you also have a file
839 named @file{1234}; @value{GDBN} does check for a core file first).
841 Taking advantage of the second command-line argument requires a fairly
842 complete operating system; when you use @value{GDBN} as a remote
843 debugger attached to a bare board, there may not be any notion of
844 ``process'', and there is often no way to get a core dump. @value{GDBN}
845 will warn you if it is unable to attach or to read core dumps.
847 You can optionally have @code{@value{GDBP}} pass any arguments after the
848 executable file to the inferior using @code{--args}. This option stops
851 gdb --args gcc -O2 -c foo.c
853 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
854 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
856 You can run @code{@value{GDBP}} without printing the front material, which describes
857 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
864 You can further control how @value{GDBN} starts up by using command-line
865 options. @value{GDBN} itself can remind you of the options available.
875 to display all available options and briefly describe their use
876 (@samp{@value{GDBP} -h} is a shorter equivalent).
878 All options and command line arguments you give are processed
879 in sequential order. The order makes a difference when the
880 @samp{-x} option is used.
884 * File Options:: Choosing files
885 * Mode Options:: Choosing modes
889 @subsection Choosing files
891 When @value{GDBN} starts, it reads any arguments other than options as
892 specifying an executable file and core file (or process ID). This is
893 the same as if the arguments were specified by the @samp{-se} and
894 @samp{-c} options respectively. (@value{GDBN} reads the first argument
895 that does not have an associated option flag as equivalent to the
896 @samp{-se} option followed by that argument; and the second argument
897 that does not have an associated option flag, if any, as equivalent to
898 the @samp{-c} option followed by that argument.)
900 If @value{GDBN} has not been configured to included core file support,
901 such as for most embedded targets, then it will complain about a second
902 argument and ignore it.
904 Many options have both long and short forms; both are shown in the
905 following list. @value{GDBN} also recognizes the long forms if you truncate
906 them, so long as enough of the option is present to be unambiguous.
907 (If you prefer, you can flag option arguments with @samp{--} rather
908 than @samp{-}, though we illustrate the more usual convention.)
910 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
911 @c way, both those who look for -foo and --foo in the index, will find
915 @item -symbols @var{file}
917 @cindex @code{--symbols}
919 Read symbol table from file @var{file}.
921 @item -exec @var{file}
923 @cindex @code{--exec}
925 Use file @var{file} as the executable file to execute when appropriate,
926 and for examining pure data in conjunction with a core dump.
930 Read symbol table from file @var{file} and use it as the executable
933 @item -core @var{file}
935 @cindex @code{--core}
937 Use file @var{file} as a core dump to examine.
939 @item -c @var{number}
940 Connect to process ID @var{number}, as with the @code{attach} command
941 (unless there is a file in core-dump format named @var{number}, in which
942 case @samp{-c} specifies that file as a core dump to read).
944 @item -command @var{file}
946 @cindex @code{--command}
948 Execute @value{GDBN} commands from file @var{file}. @xref{Command
949 Files,, Command files}.
951 @item -directory @var{directory}
952 @itemx -d @var{directory}
953 @cindex @code{--directory}
955 Add @var{directory} to the path to search for source files.
959 @cindex @code{--mapped}
961 @emph{Warning: this option depends on operating system facilities that are not
962 supported on all systems.}@*
963 If memory-mapped files are available on your system through the @code{mmap}
964 system call, you can use this option
965 to have @value{GDBN} write the symbols from your
966 program into a reusable file in the current directory. If the program you are debugging is
967 called @file{/tmp/fred}, the mapped symbol file is @file{/tmp/fred.syms}.
968 Future @value{GDBN} debugging sessions notice the presence of this file,
969 and can quickly map in symbol information from it, rather than reading
970 the symbol table from the executable program.
972 The @file{.syms} file is specific to the host machine where @value{GDBN}
973 is run. It holds an exact image of the internal @value{GDBN} symbol
974 table. It cannot be shared across multiple host platforms.
978 @cindex @code{--readnow}
980 Read each symbol file's entire symbol table immediately, rather than
981 the default, which is to read it incrementally as it is needed.
982 This makes startup slower, but makes future operations faster.
986 You typically combine the @code{-mapped} and @code{-readnow} options in
987 order to build a @file{.syms} file that contains complete symbol
988 information. (@xref{Files,,Commands to specify files}, for information
989 on @file{.syms} files.) A simple @value{GDBN} invocation to do nothing
990 but build a @file{.syms} file for future use is:
993 gdb -batch -nx -mapped -readnow programname
997 @subsection Choosing modes
999 You can run @value{GDBN} in various alternative modes---for example, in
1000 batch mode or quiet mode.
1007 Do not execute commands found in any initialization files. Normally,
1008 @value{GDBN} executes the commands in these files after all the command
1009 options and arguments have been processed. @xref{Command Files,,Command
1015 @cindex @code{--quiet}
1016 @cindex @code{--silent}
1018 ``Quiet''. Do not print the introductory and copyright messages. These
1019 messages are also suppressed in batch mode.
1022 @cindex @code{--batch}
1023 Run in batch mode. Exit with status @code{0} after processing all the
1024 command files specified with @samp{-x} (and all commands from
1025 initialization files, if not inhibited with @samp{-n}). Exit with
1026 nonzero status if an error occurs in executing the @value{GDBN} commands
1027 in the command files.
1029 Batch mode may be useful for running @value{GDBN} as a filter, for
1030 example to download and run a program on another computer; in order to
1031 make this more useful, the message
1034 Program exited normally.
1038 (which is ordinarily issued whenever a program running under
1039 @value{GDBN} control terminates) is not issued when running in batch
1044 @cindex @code{--nowindows}
1046 ``No windows''. If @value{GDBN} comes with a graphical user interface
1047 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1048 interface. If no GUI is available, this option has no effect.
1052 @cindex @code{--windows}
1054 If @value{GDBN} includes a GUI, then this option requires it to be
1057 @item -cd @var{directory}
1059 Run @value{GDBN} using @var{directory} as its working directory,
1060 instead of the current directory.
1064 @cindex @code{--fullname}
1066 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1067 subprocess. It tells @value{GDBN} to output the full file name and line
1068 number in a standard, recognizable fashion each time a stack frame is
1069 displayed (which includes each time your program stops). This
1070 recognizable format looks like two @samp{\032} characters, followed by
1071 the file name, line number and character position separated by colons,
1072 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1073 @samp{\032} characters as a signal to display the source code for the
1077 @cindex @code{--epoch}
1078 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1079 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1080 routines so as to allow Epoch to display values of expressions in a
1083 @item -annotate @var{level}
1084 @cindex @code{--annotate}
1085 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1086 effect is identical to using @samp{set annotate @var{level}}
1087 (@pxref{Annotations}).
1088 Annotation level controls how much information does @value{GDBN} print
1089 together with its prompt, values of expressions, source lines, and other
1090 types of output. Level 0 is the normal, level 1 is for use when
1091 @value{GDBN} is run as a subprocess of @sc{gnu} Emacs, level 2 is the
1092 maximum annotation suitable for programs that control @value{GDBN}.
1095 @cindex @code{--async}
1096 Use the asynchronous event loop for the command-line interface.
1097 @value{GDBN} processes all events, such as user keyboard input, via a
1098 special event loop. This allows @value{GDBN} to accept and process user
1099 commands in parallel with the debugged process being
1100 run@footnote{@value{GDBN} built with @sc{djgpp} tools for
1101 MS-DOS/MS-Windows supports this mode of operation, but the event loop is
1102 suspended when the debuggee runs.}, so you don't need to wait for
1103 control to return to @value{GDBN} before you type the next command.
1104 (@emph{Note:} as of version 5.1, the target side of the asynchronous
1105 operation is not yet in place, so @samp{-async} does not work fully
1107 @c FIXME: when the target side of the event loop is done, the above NOTE
1108 @c should be removed.
1110 When the standard input is connected to a terminal device, @value{GDBN}
1111 uses the asynchronous event loop by default, unless disabled by the
1112 @samp{-noasync} option.
1115 @cindex @code{--noasync}
1116 Disable the asynchronous event loop for the command-line interface.
1119 @cindex @code{--args}
1120 Change interpretation of command line so that arguments following the
1121 executable file are passed as command line arguments to the inferior.
1122 This option stops option processing.
1124 @item -baud @var{bps}
1126 @cindex @code{--baud}
1128 Set the line speed (baud rate or bits per second) of any serial
1129 interface used by @value{GDBN} for remote debugging.
1131 @item -tty @var{device}
1132 @itemx -t @var{device}
1133 @cindex @code{--tty}
1135 Run using @var{device} for your program's standard input and output.
1136 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1138 @c resolve the situation of these eventually
1140 @cindex @code{--tui}
1141 Activate the Terminal User Interface when starting.
1142 The Terminal User Interface manages several text windows on the terminal,
1143 showing source, assembly, registers and @value{GDBN} command outputs
1144 (@pxref{TUI, ,@value{GDBN} Text User Interface}).
1145 Do not use this option if you run @value{GDBN} from Emacs
1146 (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1149 @c @cindex @code{--xdb}
1150 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1151 @c For information, see the file @file{xdb_trans.html}, which is usually
1152 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1155 @item -interpreter @var{interp}
1156 @cindex @code{--interpreter}
1157 Use the interpreter @var{interp} for interface with the controlling
1158 program or device. This option is meant to be set by programs which
1159 communicate with @value{GDBN} using it as a back end.
1161 @samp{--interpreter=mi} (or @samp{--interpreter=mi1}) causes
1162 @value{GDBN} to use the @dfn{gdb/mi interface} (@pxref{GDB/MI, , The
1163 @sc{gdb/mi} Interface}). The older @sc{gdb/mi} interface, included in
1164 @value{GDBN} version 5.0 can be selected with @samp{--interpreter=mi0}.
1167 @cindex @code{--write}
1168 Open the executable and core files for both reading and writing. This
1169 is equivalent to the @samp{set write on} command inside @value{GDBN}
1173 @cindex @code{--statistics}
1174 This option causes @value{GDBN} to print statistics about time and
1175 memory usage after it completes each command and returns to the prompt.
1178 @cindex @code{--version}
1179 This option causes @value{GDBN} to print its version number and
1180 no-warranty blurb, and exit.
1185 @section Quitting @value{GDBN}
1186 @cindex exiting @value{GDBN}
1187 @cindex leaving @value{GDBN}
1190 @kindex quit @r{[}@var{expression}@r{]}
1191 @kindex q @r{(@code{quit})}
1192 @item quit @r{[}@var{expression}@r{]}
1194 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1195 @code{q}), or type an end-of-file character (usually @kbd{C-d}). If you
1196 do not supply @var{expression}, @value{GDBN} will terminate normally;
1197 otherwise it will terminate using the result of @var{expression} as the
1202 An interrupt (often @kbd{C-c}) does not exit from @value{GDBN}, but rather
1203 terminates the action of any @value{GDBN} command that is in progress and
1204 returns to @value{GDBN} command level. It is safe to type the interrupt
1205 character at any time because @value{GDBN} does not allow it to take effect
1206 until a time when it is safe.
1208 If you have been using @value{GDBN} to control an attached process or
1209 device, you can release it with the @code{detach} command
1210 (@pxref{Attach, ,Debugging an already-running process}).
1212 @node Shell Commands
1213 @section Shell commands
1215 If you need to execute occasional shell commands during your
1216 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1217 just use the @code{shell} command.
1221 @cindex shell escape
1222 @item shell @var{command string}
1223 Invoke a standard shell to execute @var{command string}.
1224 If it exists, the environment variable @code{SHELL} determines which
1225 shell to run. Otherwise @value{GDBN} uses the default shell
1226 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1229 The utility @code{make} is often needed in development environments.
1230 You do not have to use the @code{shell} command for this purpose in
1235 @cindex calling make
1236 @item make @var{make-args}
1237 Execute the @code{make} program with the specified
1238 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1242 @chapter @value{GDBN} Commands
1244 You can abbreviate a @value{GDBN} command to the first few letters of the command
1245 name, if that abbreviation is unambiguous; and you can repeat certain
1246 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1247 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1248 show you the alternatives available, if there is more than one possibility).
1251 * Command Syntax:: How to give commands to @value{GDBN}
1252 * Completion:: Command completion
1253 * Help:: How to ask @value{GDBN} for help
1256 @node Command Syntax
1257 @section Command syntax
1259 A @value{GDBN} command is a single line of input. There is no limit on
1260 how long it can be. It starts with a command name, which is followed by
1261 arguments whose meaning depends on the command name. For example, the
1262 command @code{step} accepts an argument which is the number of times to
1263 step, as in @samp{step 5}. You can also use the @code{step} command
1264 with no arguments. Some commands do not allow any arguments.
1266 @cindex abbreviation
1267 @value{GDBN} command names may always be truncated if that abbreviation is
1268 unambiguous. Other possible command abbreviations are listed in the
1269 documentation for individual commands. In some cases, even ambiguous
1270 abbreviations are allowed; for example, @code{s} is specially defined as
1271 equivalent to @code{step} even though there are other commands whose
1272 names start with @code{s}. You can test abbreviations by using them as
1273 arguments to the @code{help} command.
1275 @cindex repeating commands
1276 @kindex RET @r{(repeat last command)}
1277 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1278 repeat the previous command. Certain commands (for example, @code{run})
1279 will not repeat this way; these are commands whose unintentional
1280 repetition might cause trouble and which you are unlikely to want to
1283 The @code{list} and @code{x} commands, when you repeat them with
1284 @key{RET}, construct new arguments rather than repeating
1285 exactly as typed. This permits easy scanning of source or memory.
1287 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1288 output, in a way similar to the common utility @code{more}
1289 (@pxref{Screen Size,,Screen size}). Since it is easy to press one
1290 @key{RET} too many in this situation, @value{GDBN} disables command
1291 repetition after any command that generates this sort of display.
1293 @kindex # @r{(a comment)}
1295 Any text from a @kbd{#} to the end of the line is a comment; it does
1296 nothing. This is useful mainly in command files (@pxref{Command
1297 Files,,Command files}).
1299 @cindex repeating command sequences
1300 @kindex C-o @r{(operate-and-get-next)}
1301 The @kbd{C-o} binding is useful for repeating a complex sequence of
1302 commands. This command accepts the current line, like @kbd{RET}, and
1303 then fetches the next line relative to the current line from the history
1307 @section Command completion
1310 @cindex word completion
1311 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1312 only one possibility; it can also show you what the valid possibilities
1313 are for the next word in a command, at any time. This works for @value{GDBN}
1314 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1316 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1317 of a word. If there is only one possibility, @value{GDBN} fills in the
1318 word, and waits for you to finish the command (or press @key{RET} to
1319 enter it). For example, if you type
1321 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1322 @c complete accuracy in these examples; space introduced for clarity.
1323 @c If texinfo enhancements make it unnecessary, it would be nice to
1324 @c replace " @key" by "@key" in the following...
1326 (@value{GDBP}) info bre @key{TAB}
1330 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1331 the only @code{info} subcommand beginning with @samp{bre}:
1334 (@value{GDBP}) info breakpoints
1338 You can either press @key{RET} at this point, to run the @code{info
1339 breakpoints} command, or backspace and enter something else, if
1340 @samp{breakpoints} does not look like the command you expected. (If you
1341 were sure you wanted @code{info breakpoints} in the first place, you
1342 might as well just type @key{RET} immediately after @samp{info bre},
1343 to exploit command abbreviations rather than command completion).
1345 If there is more than one possibility for the next word when you press
1346 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1347 characters and try again, or just press @key{TAB} a second time;
1348 @value{GDBN} displays all the possible completions for that word. For
1349 example, you might want to set a breakpoint on a subroutine whose name
1350 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1351 just sounds the bell. Typing @key{TAB} again displays all the
1352 function names in your program that begin with those characters, for
1356 (@value{GDBP}) b make_ @key{TAB}
1357 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1358 make_a_section_from_file make_environ
1359 make_abs_section make_function_type
1360 make_blockvector make_pointer_type
1361 make_cleanup make_reference_type
1362 make_command make_symbol_completion_list
1363 (@value{GDBP}) b make_
1367 After displaying the available possibilities, @value{GDBN} copies your
1368 partial input (@samp{b make_} in the example) so you can finish the
1371 If you just want to see the list of alternatives in the first place, you
1372 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1373 means @kbd{@key{META} ?}. You can type this either by holding down a
1374 key designated as the @key{META} shift on your keyboard (if there is
1375 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1377 @cindex quotes in commands
1378 @cindex completion of quoted strings
1379 Sometimes the string you need, while logically a ``word'', may contain
1380 parentheses or other characters that @value{GDBN} normally excludes from
1381 its notion of a word. To permit word completion to work in this
1382 situation, you may enclose words in @code{'} (single quote marks) in
1383 @value{GDBN} commands.
1385 The most likely situation where you might need this is in typing the
1386 name of a C@t{++} function. This is because C@t{++} allows function
1387 overloading (multiple definitions of the same function, distinguished
1388 by argument type). For example, when you want to set a breakpoint you
1389 may need to distinguish whether you mean the version of @code{name}
1390 that takes an @code{int} parameter, @code{name(int)}, or the version
1391 that takes a @code{float} parameter, @code{name(float)}. To use the
1392 word-completion facilities in this situation, type a single quote
1393 @code{'} at the beginning of the function name. This alerts
1394 @value{GDBN} that it may need to consider more information than usual
1395 when you press @key{TAB} or @kbd{M-?} to request word completion:
1398 (@value{GDBP}) b 'bubble( @kbd{M-?}
1399 bubble(double,double) bubble(int,int)
1400 (@value{GDBP}) b 'bubble(
1403 In some cases, @value{GDBN} can tell that completing a name requires using
1404 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1405 completing as much as it can) if you do not type the quote in the first
1409 (@value{GDBP}) b bub @key{TAB}
1410 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1411 (@value{GDBP}) b 'bubble(
1415 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1416 you have not yet started typing the argument list when you ask for
1417 completion on an overloaded symbol.
1419 For more information about overloaded functions, see @ref{C plus plus
1420 expressions, ,C@t{++} expressions}. You can use the command @code{set
1421 overload-resolution off} to disable overload resolution;
1422 see @ref{Debugging C plus plus, ,@value{GDBN} features for C@t{++}}.
1426 @section Getting help
1427 @cindex online documentation
1430 You can always ask @value{GDBN} itself for information on its commands,
1431 using the command @code{help}.
1434 @kindex h @r{(@code{help})}
1437 You can use @code{help} (abbreviated @code{h}) with no arguments to
1438 display a short list of named classes of commands:
1442 List of classes of commands:
1444 aliases -- Aliases of other commands
1445 breakpoints -- Making program stop at certain points
1446 data -- Examining data
1447 files -- Specifying and examining files
1448 internals -- Maintenance commands
1449 obscure -- Obscure features
1450 running -- Running the program
1451 stack -- Examining the stack
1452 status -- Status inquiries
1453 support -- Support facilities
1454 tracepoints -- Tracing of program execution without@*
1455 stopping the program
1456 user-defined -- User-defined commands
1458 Type "help" followed by a class name for a list of
1459 commands in that class.
1460 Type "help" followed by command name for full
1462 Command name abbreviations are allowed if unambiguous.
1465 @c the above line break eliminates huge line overfull...
1467 @item help @var{class}
1468 Using one of the general help classes as an argument, you can get a
1469 list of the individual commands in that class. For example, here is the
1470 help display for the class @code{status}:
1473 (@value{GDBP}) help status
1478 @c Line break in "show" line falsifies real output, but needed
1479 @c to fit in smallbook page size.
1480 info -- Generic command for showing things
1481 about the program being debugged
1482 show -- Generic command for showing things
1485 Type "help" followed by command name for full
1487 Command name abbreviations are allowed if unambiguous.
1491 @item help @var{command}
1492 With a command name as @code{help} argument, @value{GDBN} displays a
1493 short paragraph on how to use that command.
1496 @item apropos @var{args}
1497 The @code{apropos @var{args}} command searches through all of the @value{GDBN}
1498 commands, and their documentation, for the regular expression specified in
1499 @var{args}. It prints out all matches found. For example:
1510 set symbol-reloading -- Set dynamic symbol table reloading
1511 multiple times in one run
1512 show symbol-reloading -- Show dynamic symbol table reloading
1513 multiple times in one run
1518 @item complete @var{args}
1519 The @code{complete @var{args}} command lists all the possible completions
1520 for the beginning of a command. Use @var{args} to specify the beginning of the
1521 command you want completed. For example:
1527 @noindent results in:
1538 @noindent This is intended for use by @sc{gnu} Emacs.
1541 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1542 and @code{show} to inquire about the state of your program, or the state
1543 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1544 manual introduces each of them in the appropriate context. The listings
1545 under @code{info} and under @code{show} in the Index point to
1546 all the sub-commands. @xref{Index}.
1551 @kindex i @r{(@code{info})}
1553 This command (abbreviated @code{i}) is for describing the state of your
1554 program. For example, you can list the arguments given to your program
1555 with @code{info args}, list the registers currently in use with @code{info
1556 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1557 You can get a complete list of the @code{info} sub-commands with
1558 @w{@code{help info}}.
1562 You can assign the result of an expression to an environment variable with
1563 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1564 @code{set prompt $}.
1568 In contrast to @code{info}, @code{show} is for describing the state of
1569 @value{GDBN} itself.
1570 You can change most of the things you can @code{show}, by using the
1571 related command @code{set}; for example, you can control what number
1572 system is used for displays with @code{set radix}, or simply inquire
1573 which is currently in use with @code{show radix}.
1576 To display all the settable parameters and their current
1577 values, you can use @code{show} with no arguments; you may also use
1578 @code{info set}. Both commands produce the same display.
1579 @c FIXME: "info set" violates the rule that "info" is for state of
1580 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1581 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1585 Here are three miscellaneous @code{show} subcommands, all of which are
1586 exceptional in lacking corresponding @code{set} commands:
1589 @kindex show version
1590 @cindex version number
1592 Show what version of @value{GDBN} is running. You should include this
1593 information in @value{GDBN} bug-reports. If multiple versions of
1594 @value{GDBN} are in use at your site, you may need to determine which
1595 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1596 commands are introduced, and old ones may wither away. Also, many
1597 system vendors ship variant versions of @value{GDBN}, and there are
1598 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1599 The version number is the same as the one announced when you start
1602 @kindex show copying
1604 Display information about permission for copying @value{GDBN}.
1606 @kindex show warranty
1608 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1609 if your version of @value{GDBN} comes with one.
1614 @chapter Running Programs Under @value{GDBN}
1616 When you run a program under @value{GDBN}, you must first generate
1617 debugging information when you compile it.
1619 You may start @value{GDBN} with its arguments, if any, in an environment
1620 of your choice. If you are doing native debugging, you may redirect
1621 your program's input and output, debug an already running process, or
1622 kill a child process.
1625 * Compilation:: Compiling for debugging
1626 * Starting:: Starting your program
1627 * Arguments:: Your program's arguments
1628 * Environment:: Your program's environment
1630 * Working Directory:: Your program's working directory
1631 * Input/Output:: Your program's input and output
1632 * Attach:: Debugging an already-running process
1633 * Kill Process:: Killing the child process
1635 * Threads:: Debugging programs with multiple threads
1636 * Processes:: Debugging programs with multiple processes
1640 @section Compiling for debugging
1642 In order to debug a program effectively, you need to generate
1643 debugging information when you compile it. This debugging information
1644 is stored in the object file; it describes the data type of each
1645 variable or function and the correspondence between source line numbers
1646 and addresses in the executable code.
1648 To request debugging information, specify the @samp{-g} option when you run
1651 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1652 options together. Using those compilers, you cannot generate optimized
1653 executables containing debugging information.
1655 @value{NGCC}, the @sc{gnu} C compiler, supports @samp{-g} with or
1656 without @samp{-O}, making it possible to debug optimized code. We
1657 recommend that you @emph{always} use @samp{-g} whenever you compile a
1658 program. You may think your program is correct, but there is no sense
1659 in pushing your luck.
1661 @cindex optimized code, debugging
1662 @cindex debugging optimized code
1663 When you debug a program compiled with @samp{-g -O}, remember that the
1664 optimizer is rearranging your code; the debugger shows you what is
1665 really there. Do not be too surprised when the execution path does not
1666 exactly match your source file! An extreme example: if you define a
1667 variable, but never use it, @value{GDBN} never sees that
1668 variable---because the compiler optimizes it out of existence.
1670 Some things do not work as well with @samp{-g -O} as with just
1671 @samp{-g}, particularly on machines with instruction scheduling. If in
1672 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1673 please report it to us as a bug (including a test case!).
1675 Older versions of the @sc{gnu} C compiler permitted a variant option
1676 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1677 format; if your @sc{gnu} C compiler has this option, do not use it.
1681 @section Starting your program
1687 @kindex r @r{(@code{run})}
1690 Use the @code{run} command to start your program under @value{GDBN}.
1691 You must first specify the program name (except on VxWorks) with an
1692 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1693 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1694 (@pxref{Files, ,Commands to specify files}).
1698 If you are running your program in an execution environment that
1699 supports processes, @code{run} creates an inferior process and makes
1700 that process run your program. (In environments without processes,
1701 @code{run} jumps to the start of your program.)
1703 The execution of a program is affected by certain information it
1704 receives from its superior. @value{GDBN} provides ways to specify this
1705 information, which you must do @emph{before} starting your program. (You
1706 can change it after starting your program, but such changes only affect
1707 your program the next time you start it.) This information may be
1708 divided into four categories:
1711 @item The @emph{arguments.}
1712 Specify the arguments to give your program as the arguments of the
1713 @code{run} command. If a shell is available on your target, the shell
1714 is used to pass the arguments, so that you may use normal conventions
1715 (such as wildcard expansion or variable substitution) in describing
1717 In Unix systems, you can control which shell is used with the
1718 @code{SHELL} environment variable.
1719 @xref{Arguments, ,Your program's arguments}.
1721 @item The @emph{environment.}
1722 Your program normally inherits its environment from @value{GDBN}, but you can
1723 use the @value{GDBN} commands @code{set environment} and @code{unset
1724 environment} to change parts of the environment that affect
1725 your program. @xref{Environment, ,Your program's environment}.
1727 @item The @emph{working directory.}
1728 Your program inherits its working directory from @value{GDBN}. You can set
1729 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1730 @xref{Working Directory, ,Your program's working directory}.
1732 @item The @emph{standard input and output.}
1733 Your program normally uses the same device for standard input and
1734 standard output as @value{GDBN} is using. You can redirect input and output
1735 in the @code{run} command line, or you can use the @code{tty} command to
1736 set a different device for your program.
1737 @xref{Input/Output, ,Your program's input and output}.
1740 @emph{Warning:} While input and output redirection work, you cannot use
1741 pipes to pass the output of the program you are debugging to another
1742 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1746 When you issue the @code{run} command, your program begins to execute
1747 immediately. @xref{Stopping, ,Stopping and continuing}, for discussion
1748 of how to arrange for your program to stop. Once your program has
1749 stopped, you may call functions in your program, using the @code{print}
1750 or @code{call} commands. @xref{Data, ,Examining Data}.
1752 If the modification time of your symbol file has changed since the last
1753 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1754 table, and reads it again. When it does this, @value{GDBN} tries to retain
1755 your current breakpoints.
1758 @section Your program's arguments
1760 @cindex arguments (to your program)
1761 The arguments to your program can be specified by the arguments of the
1763 They are passed to a shell, which expands wildcard characters and
1764 performs redirection of I/O, and thence to your program. Your
1765 @code{SHELL} environment variable (if it exists) specifies what shell
1766 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
1767 the default shell (@file{/bin/sh} on Unix).
1769 On non-Unix systems, the program is usually invoked directly by
1770 @value{GDBN}, which emulates I/O redirection via the appropriate system
1771 calls, and the wildcard characters are expanded by the startup code of
1772 the program, not by the shell.
1774 @code{run} with no arguments uses the same arguments used by the previous
1775 @code{run}, or those set by the @code{set args} command.
1780 Specify the arguments to be used the next time your program is run. If
1781 @code{set args} has no arguments, @code{run} executes your program
1782 with no arguments. Once you have run your program with arguments,
1783 using @code{set args} before the next @code{run} is the only way to run
1784 it again without arguments.
1788 Show the arguments to give your program when it is started.
1792 @section Your program's environment
1794 @cindex environment (of your program)
1795 The @dfn{environment} consists of a set of environment variables and
1796 their values. Environment variables conventionally record such things as
1797 your user name, your home directory, your terminal type, and your search
1798 path for programs to run. Usually you set up environment variables with
1799 the shell and they are inherited by all the other programs you run. When
1800 debugging, it can be useful to try running your program with a modified
1801 environment without having to start @value{GDBN} over again.
1805 @item path @var{directory}
1806 Add @var{directory} to the front of the @code{PATH} environment variable
1807 (the search path for executables) that will be passed to your program.
1808 The value of @code{PATH} used by @value{GDBN} does not change.
1809 You may specify several directory names, separated by whitespace or by a
1810 system-dependent separator character (@samp{:} on Unix, @samp{;} on
1811 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
1812 is moved to the front, so it is searched sooner.
1814 You can use the string @samp{$cwd} to refer to whatever is the current
1815 working directory at the time @value{GDBN} searches the path. If you
1816 use @samp{.} instead, it refers to the directory where you executed the
1817 @code{path} command. @value{GDBN} replaces @samp{.} in the
1818 @var{directory} argument (with the current path) before adding
1819 @var{directory} to the search path.
1820 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
1821 @c document that, since repeating it would be a no-op.
1825 Display the list of search paths for executables (the @code{PATH}
1826 environment variable).
1828 @kindex show environment
1829 @item show environment @r{[}@var{varname}@r{]}
1830 Print the value of environment variable @var{varname} to be given to
1831 your program when it starts. If you do not supply @var{varname},
1832 print the names and values of all environment variables to be given to
1833 your program. You can abbreviate @code{environment} as @code{env}.
1835 @kindex set environment
1836 @item set environment @var{varname} @r{[}=@var{value}@r{]}
1837 Set environment variable @var{varname} to @var{value}. The value
1838 changes for your program only, not for @value{GDBN} itself. @var{value} may
1839 be any string; the values of environment variables are just strings, and
1840 any interpretation is supplied by your program itself. The @var{value}
1841 parameter is optional; if it is eliminated, the variable is set to a
1843 @c "any string" here does not include leading, trailing
1844 @c blanks. Gnu asks: does anyone care?
1846 For example, this command:
1853 tells the debugged program, when subsequently run, that its user is named
1854 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1855 are not actually required.)
1857 @kindex unset environment
1858 @item unset environment @var{varname}
1859 Remove variable @var{varname} from the environment to be passed to your
1860 program. This is different from @samp{set env @var{varname} =};
1861 @code{unset environment} removes the variable from the environment,
1862 rather than assigning it an empty value.
1865 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
1867 by your @code{SHELL} environment variable if it exists (or
1868 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
1869 that runs an initialization file---such as @file{.cshrc} for C-shell, or
1870 @file{.bashrc} for BASH---any variables you set in that file affect
1871 your program. You may wish to move setting of environment variables to
1872 files that are only run when you sign on, such as @file{.login} or
1875 @node Working Directory
1876 @section Your program's working directory
1878 @cindex working directory (of your program)
1879 Each time you start your program with @code{run}, it inherits its
1880 working directory from the current working directory of @value{GDBN}.
1881 The @value{GDBN} working directory is initially whatever it inherited
1882 from its parent process (typically the shell), but you can specify a new
1883 working directory in @value{GDBN} with the @code{cd} command.
1885 The @value{GDBN} working directory also serves as a default for the commands
1886 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
1891 @item cd @var{directory}
1892 Set the @value{GDBN} working directory to @var{directory}.
1896 Print the @value{GDBN} working directory.
1900 @section Your program's input and output
1905 By default, the program you run under @value{GDBN} does input and output to
1906 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
1907 to its own terminal modes to interact with you, but it records the terminal
1908 modes your program was using and switches back to them when you continue
1909 running your program.
1912 @kindex info terminal
1914 Displays information recorded by @value{GDBN} about the terminal modes your
1918 You can redirect your program's input and/or output using shell
1919 redirection with the @code{run} command. For example,
1926 starts your program, diverting its output to the file @file{outfile}.
1929 @cindex controlling terminal
1930 Another way to specify where your program should do input and output is
1931 with the @code{tty} command. This command accepts a file name as
1932 argument, and causes this file to be the default for future @code{run}
1933 commands. It also resets the controlling terminal for the child
1934 process, for future @code{run} commands. For example,
1941 directs that processes started with subsequent @code{run} commands
1942 default to do input and output on the terminal @file{/dev/ttyb} and have
1943 that as their controlling terminal.
1945 An explicit redirection in @code{run} overrides the @code{tty} command's
1946 effect on the input/output device, but not its effect on the controlling
1949 When you use the @code{tty} command or redirect input in the @code{run}
1950 command, only the input @emph{for your program} is affected. The input
1951 for @value{GDBN} still comes from your terminal.
1954 @section Debugging an already-running process
1959 @item attach @var{process-id}
1960 This command attaches to a running process---one that was started
1961 outside @value{GDBN}. (@code{info files} shows your active
1962 targets.) The command takes as argument a process ID. The usual way to
1963 find out the process-id of a Unix process is with the @code{ps} utility,
1964 or with the @samp{jobs -l} shell command.
1966 @code{attach} does not repeat if you press @key{RET} a second time after
1967 executing the command.
1970 To use @code{attach}, your program must be running in an environment
1971 which supports processes; for example, @code{attach} does not work for
1972 programs on bare-board targets that lack an operating system. You must
1973 also have permission to send the process a signal.
1975 When you use @code{attach}, the debugger finds the program running in
1976 the process first by looking in the current working directory, then (if
1977 the program is not found) by using the source file search path
1978 (@pxref{Source Path, ,Specifying source directories}). You can also use
1979 the @code{file} command to load the program. @xref{Files, ,Commands to
1982 The first thing @value{GDBN} does after arranging to debug the specified
1983 process is to stop it. You can examine and modify an attached process
1984 with all the @value{GDBN} commands that are ordinarily available when
1985 you start processes with @code{run}. You can insert breakpoints; you
1986 can step and continue; you can modify storage. If you would rather the
1987 process continue running, you may use the @code{continue} command after
1988 attaching @value{GDBN} to the process.
1993 When you have finished debugging the attached process, you can use the
1994 @code{detach} command to release it from @value{GDBN} control. Detaching
1995 the process continues its execution. After the @code{detach} command,
1996 that process and @value{GDBN} become completely independent once more, and you
1997 are ready to @code{attach} another process or start one with @code{run}.
1998 @code{detach} does not repeat if you press @key{RET} again after
1999 executing the command.
2002 If you exit @value{GDBN} or use the @code{run} command while you have an
2003 attached process, you kill that process. By default, @value{GDBN} asks
2004 for confirmation if you try to do either of these things; you can
2005 control whether or not you need to confirm by using the @code{set
2006 confirm} command (@pxref{Messages/Warnings, ,Optional warnings and
2010 @section Killing the child process
2015 Kill the child process in which your program is running under @value{GDBN}.
2018 This command is useful if you wish to debug a core dump instead of a
2019 running process. @value{GDBN} ignores any core dump file while your program
2022 On some operating systems, a program cannot be executed outside @value{GDBN}
2023 while you have breakpoints set on it inside @value{GDBN}. You can use the
2024 @code{kill} command in this situation to permit running your program
2025 outside the debugger.
2027 The @code{kill} command is also useful if you wish to recompile and
2028 relink your program, since on many systems it is impossible to modify an
2029 executable file while it is running in a process. In this case, when you
2030 next type @code{run}, @value{GDBN} notices that the file has changed, and
2031 reads the symbol table again (while trying to preserve your current
2032 breakpoint settings).
2035 @section Debugging programs with multiple threads
2037 @cindex threads of execution
2038 @cindex multiple threads
2039 @cindex switching threads
2040 In some operating systems, such as HP-UX and Solaris, a single program
2041 may have more than one @dfn{thread} of execution. The precise semantics
2042 of threads differ from one operating system to another, but in general
2043 the threads of a single program are akin to multiple processes---except
2044 that they share one address space (that is, they can all examine and
2045 modify the same variables). On the other hand, each thread has its own
2046 registers and execution stack, and perhaps private memory.
2048 @value{GDBN} provides these facilities for debugging multi-thread
2052 @item automatic notification of new threads
2053 @item @samp{thread @var{threadno}}, a command to switch among threads
2054 @item @samp{info threads}, a command to inquire about existing threads
2055 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2056 a command to apply a command to a list of threads
2057 @item thread-specific breakpoints
2061 @emph{Warning:} These facilities are not yet available on every
2062 @value{GDBN} configuration where the operating system supports threads.
2063 If your @value{GDBN} does not support threads, these commands have no
2064 effect. For example, a system without thread support shows no output
2065 from @samp{info threads}, and always rejects the @code{thread} command,
2069 (@value{GDBP}) info threads
2070 (@value{GDBP}) thread 1
2071 Thread ID 1 not known. Use the "info threads" command to
2072 see the IDs of currently known threads.
2074 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2075 @c doesn't support threads"?
2078 @cindex focus of debugging
2079 @cindex current thread
2080 The @value{GDBN} thread debugging facility allows you to observe all
2081 threads while your program runs---but whenever @value{GDBN} takes
2082 control, one thread in particular is always the focus of debugging.
2083 This thread is called the @dfn{current thread}. Debugging commands show
2084 program information from the perspective of the current thread.
2086 @cindex @code{New} @var{systag} message
2087 @cindex thread identifier (system)
2088 @c FIXME-implementors!! It would be more helpful if the [New...] message
2089 @c included GDB's numeric thread handle, so you could just go to that
2090 @c thread without first checking `info threads'.
2091 Whenever @value{GDBN} detects a new thread in your program, it displays
2092 the target system's identification for the thread with a message in the
2093 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2094 whose form varies depending on the particular system. For example, on
2095 LynxOS, you might see
2098 [New process 35 thread 27]
2102 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2103 the @var{systag} is simply something like @samp{process 368}, with no
2106 @c FIXME!! (1) Does the [New...] message appear even for the very first
2107 @c thread of a program, or does it only appear for the
2108 @c second---i.e., when it becomes obvious we have a multithread
2110 @c (2) *Is* there necessarily a first thread always? Or do some
2111 @c multithread systems permit starting a program with multiple
2112 @c threads ab initio?
2114 @cindex thread number
2115 @cindex thread identifier (GDB)
2116 For debugging purposes, @value{GDBN} associates its own thread
2117 number---always a single integer---with each thread in your program.
2120 @kindex info threads
2122 Display a summary of all threads currently in your
2123 program. @value{GDBN} displays for each thread (in this order):
2126 @item the thread number assigned by @value{GDBN}
2128 @item the target system's thread identifier (@var{systag})
2130 @item the current stack frame summary for that thread
2134 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2135 indicates the current thread.
2139 @c end table here to get a little more width for example
2142 (@value{GDBP}) info threads
2143 3 process 35 thread 27 0x34e5 in sigpause ()
2144 2 process 35 thread 23 0x34e5 in sigpause ()
2145 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2151 @cindex thread number
2152 @cindex thread identifier (GDB)
2153 For debugging purposes, @value{GDBN} associates its own thread
2154 number---a small integer assigned in thread-creation order---with each
2155 thread in your program.
2157 @cindex @code{New} @var{systag} message, on HP-UX
2158 @cindex thread identifier (system), on HP-UX
2159 @c FIXME-implementors!! It would be more helpful if the [New...] message
2160 @c included GDB's numeric thread handle, so you could just go to that
2161 @c thread without first checking `info threads'.
2162 Whenever @value{GDBN} detects a new thread in your program, it displays
2163 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2164 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2165 whose form varies depending on the particular system. For example, on
2169 [New thread 2 (system thread 26594)]
2173 when @value{GDBN} notices a new thread.
2176 @kindex info threads
2178 Display a summary of all threads currently in your
2179 program. @value{GDBN} displays for each thread (in this order):
2182 @item the thread number assigned by @value{GDBN}
2184 @item the target system's thread identifier (@var{systag})
2186 @item the current stack frame summary for that thread
2190 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2191 indicates the current thread.
2195 @c end table here to get a little more width for example
2198 (@value{GDBP}) info threads
2199 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2201 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2202 from /usr/lib/libc.2
2203 1 system thread 27905 0x7b003498 in _brk () \@*
2204 from /usr/lib/libc.2
2208 @kindex thread @var{threadno}
2209 @item thread @var{threadno}
2210 Make thread number @var{threadno} the current thread. The command
2211 argument @var{threadno} is the internal @value{GDBN} thread number, as
2212 shown in the first field of the @samp{info threads} display.
2213 @value{GDBN} responds by displaying the system identifier of the thread
2214 you selected, and its current stack frame summary:
2217 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2218 (@value{GDBP}) thread 2
2219 [Switching to process 35 thread 23]
2220 0x34e5 in sigpause ()
2224 As with the @samp{[New @dots{}]} message, the form of the text after
2225 @samp{Switching to} depends on your system's conventions for identifying
2228 @kindex thread apply
2229 @item thread apply [@var{threadno}] [@var{all}] @var{args}
2230 The @code{thread apply} command allows you to apply a command to one or
2231 more threads. Specify the numbers of the threads that you want affected
2232 with the command argument @var{threadno}. @var{threadno} is the internal
2233 @value{GDBN} thread number, as shown in the first field of the @samp{info
2234 threads} display. To apply a command to all threads, use
2235 @code{thread apply all} @var{args}.
2238 @cindex automatic thread selection
2239 @cindex switching threads automatically
2240 @cindex threads, automatic switching
2241 Whenever @value{GDBN} stops your program, due to a breakpoint or a
2242 signal, it automatically selects the thread where that breakpoint or
2243 signal happened. @value{GDBN} alerts you to the context switch with a
2244 message of the form @samp{[Switching to @var{systag}]} to identify the
2247 @xref{Thread Stops,,Stopping and starting multi-thread programs}, for
2248 more information about how @value{GDBN} behaves when you stop and start
2249 programs with multiple threads.
2251 @xref{Set Watchpoints,,Setting watchpoints}, for information about
2252 watchpoints in programs with multiple threads.
2255 @section Debugging programs with multiple processes
2257 @cindex fork, debugging programs which call
2258 @cindex multiple processes
2259 @cindex processes, multiple
2260 On most systems, @value{GDBN} has no special support for debugging
2261 programs which create additional processes using the @code{fork}
2262 function. When a program forks, @value{GDBN} will continue to debug the
2263 parent process and the child process will run unimpeded. If you have
2264 set a breakpoint in any code which the child then executes, the child
2265 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2266 will cause it to terminate.
2268 However, if you want to debug the child process there is a workaround
2269 which isn't too painful. Put a call to @code{sleep} in the code which
2270 the child process executes after the fork. It may be useful to sleep
2271 only if a certain environment variable is set, or a certain file exists,
2272 so that the delay need not occur when you don't want to run @value{GDBN}
2273 on the child. While the child is sleeping, use the @code{ps} program to
2274 get its process ID. Then tell @value{GDBN} (a new invocation of
2275 @value{GDBN} if you are also debugging the parent process) to attach to
2276 the child process (@pxref{Attach}). From that point on you can debug
2277 the child process just like any other process which you attached to.
2279 On HP-UX (11.x and later only?), @value{GDBN} provides support for
2280 debugging programs that create additional processes using the
2281 @code{fork} or @code{vfork} function.
2283 By default, when a program forks, @value{GDBN} will continue to debug
2284 the parent process and the child process will run unimpeded.
2286 If you want to follow the child process instead of the parent process,
2287 use the command @w{@code{set follow-fork-mode}}.
2290 @kindex set follow-fork-mode
2291 @item set follow-fork-mode @var{mode}
2292 Set the debugger response to a program call of @code{fork} or
2293 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2294 process. The @var{mode} can be:
2298 The original process is debugged after a fork. The child process runs
2299 unimpeded. This is the default.
2302 The new process is debugged after a fork. The parent process runs
2306 The debugger will ask for one of the above choices.
2309 @item show follow-fork-mode
2310 Display the current debugger response to a @code{fork} or @code{vfork} call.
2313 If you ask to debug a child process and a @code{vfork} is followed by an
2314 @code{exec}, @value{GDBN} executes the new target up to the first
2315 breakpoint in the new target. If you have a breakpoint set on
2316 @code{main} in your original program, the breakpoint will also be set on
2317 the child process's @code{main}.
2319 When a child process is spawned by @code{vfork}, you cannot debug the
2320 child or parent until an @code{exec} call completes.
2322 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2323 call executes, the new target restarts. To restart the parent process,
2324 use the @code{file} command with the parent executable name as its
2327 You can use the @code{catch} command to make @value{GDBN} stop whenever
2328 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2329 Catchpoints, ,Setting catchpoints}.
2332 @chapter Stopping and Continuing
2334 The principal purposes of using a debugger are so that you can stop your
2335 program before it terminates; or so that, if your program runs into
2336 trouble, you can investigate and find out why.
2338 Inside @value{GDBN}, your program may stop for any of several reasons,
2339 such as a signal, a breakpoint, or reaching a new line after a
2340 @value{GDBN} command such as @code{step}. You may then examine and
2341 change variables, set new breakpoints or remove old ones, and then
2342 continue execution. Usually, the messages shown by @value{GDBN} provide
2343 ample explanation of the status of your program---but you can also
2344 explicitly request this information at any time.
2347 @kindex info program
2349 Display information about the status of your program: whether it is
2350 running or not, what process it is, and why it stopped.
2354 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2355 * Continuing and Stepping:: Resuming execution
2357 * Thread Stops:: Stopping and starting multi-thread programs
2361 @section Breakpoints, watchpoints, and catchpoints
2364 A @dfn{breakpoint} makes your program stop whenever a certain point in
2365 the program is reached. For each breakpoint, you can add conditions to
2366 control in finer detail whether your program stops. You can set
2367 breakpoints with the @code{break} command and its variants (@pxref{Set
2368 Breaks, ,Setting breakpoints}), to specify the place where your program
2369 should stop by line number, function name or exact address in the
2372 In HP-UX, SunOS 4.x, SVR4, and Alpha OSF/1 configurations, you can set
2373 breakpoints in shared libraries before the executable is run. There is
2374 a minor limitation on HP-UX systems: you must wait until the executable
2375 is run in order to set breakpoints in shared library routines that are
2376 not called directly by the program (for example, routines that are
2377 arguments in a @code{pthread_create} call).
2380 @cindex memory tracing
2381 @cindex breakpoint on memory address
2382 @cindex breakpoint on variable modification
2383 A @dfn{watchpoint} is a special breakpoint that stops your program
2384 when the value of an expression changes. You must use a different
2385 command to set watchpoints (@pxref{Set Watchpoints, ,Setting
2386 watchpoints}), but aside from that, you can manage a watchpoint like
2387 any other breakpoint: you enable, disable, and delete both breakpoints
2388 and watchpoints using the same commands.
2390 You can arrange to have values from your program displayed automatically
2391 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2395 @cindex breakpoint on events
2396 A @dfn{catchpoint} is another special breakpoint that stops your program
2397 when a certain kind of event occurs, such as the throwing of a C@t{++}
2398 exception or the loading of a library. As with watchpoints, you use a
2399 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
2400 catchpoints}), but aside from that, you can manage a catchpoint like any
2401 other breakpoint. (To stop when your program receives a signal, use the
2402 @code{handle} command; see @ref{Signals, ,Signals}.)
2404 @cindex breakpoint numbers
2405 @cindex numbers for breakpoints
2406 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
2407 catchpoint when you create it; these numbers are successive integers
2408 starting with one. In many of the commands for controlling various
2409 features of breakpoints you use the breakpoint number to say which
2410 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
2411 @dfn{disabled}; if disabled, it has no effect on your program until you
2414 @cindex breakpoint ranges
2415 @cindex ranges of breakpoints
2416 Some @value{GDBN} commands accept a range of breakpoints on which to
2417 operate. A breakpoint range is either a single breakpoint number, like
2418 @samp{5}, or two such numbers, in increasing order, separated by a
2419 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
2420 all breakpoint in that range are operated on.
2423 * Set Breaks:: Setting breakpoints
2424 * Set Watchpoints:: Setting watchpoints
2425 * Set Catchpoints:: Setting catchpoints
2426 * Delete Breaks:: Deleting breakpoints
2427 * Disabling:: Disabling breakpoints
2428 * Conditions:: Break conditions
2429 * Break Commands:: Breakpoint command lists
2430 * Breakpoint Menus:: Breakpoint menus
2431 * Error in Breakpoints:: ``Cannot insert breakpoints''
2435 @subsection Setting breakpoints
2437 @c FIXME LMB what does GDB do if no code on line of breakpt?
2438 @c consider in particular declaration with/without initialization.
2440 @c FIXME 2 is there stuff on this already? break at fun start, already init?
2443 @kindex b @r{(@code{break})}
2444 @vindex $bpnum@r{, convenience variable}
2445 @cindex latest breakpoint
2446 Breakpoints are set with the @code{break} command (abbreviated
2447 @code{b}). The debugger convenience variable @samp{$bpnum} records the
2448 number of the breakpoint you've set most recently; see @ref{Convenience
2449 Vars,, Convenience variables}, for a discussion of what you can do with
2450 convenience variables.
2452 You have several ways to say where the breakpoint should go.
2455 @item break @var{function}
2456 Set a breakpoint at entry to function @var{function}.
2457 When using source languages that permit overloading of symbols, such as
2458 C@t{++}, @var{function} may refer to more than one possible place to break.
2459 @xref{Breakpoint Menus,,Breakpoint menus}, for a discussion of that situation.
2461 @item break +@var{offset}
2462 @itemx break -@var{offset}
2463 Set a breakpoint some number of lines forward or back from the position
2464 at which execution stopped in the currently selected @dfn{stack frame}.
2465 (@xref{Frames, ,Frames}, for a description of stack frames.)
2467 @item break @var{linenum}
2468 Set a breakpoint at line @var{linenum} in the current source file.
2469 The current source file is the last file whose source text was printed.
2470 The breakpoint will stop your program just before it executes any of the
2473 @item break @var{filename}:@var{linenum}
2474 Set a breakpoint at line @var{linenum} in source file @var{filename}.
2476 @item break @var{filename}:@var{function}
2477 Set a breakpoint at entry to function @var{function} found in file
2478 @var{filename}. Specifying a file name as well as a function name is
2479 superfluous except when multiple files contain similarly named
2482 @item break *@var{address}
2483 Set a breakpoint at address @var{address}. You can use this to set
2484 breakpoints in parts of your program which do not have debugging
2485 information or source files.
2488 When called without any arguments, @code{break} sets a breakpoint at
2489 the next instruction to be executed in the selected stack frame
2490 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
2491 innermost, this makes your program stop as soon as control
2492 returns to that frame. This is similar to the effect of a
2493 @code{finish} command in the frame inside the selected frame---except
2494 that @code{finish} does not leave an active breakpoint. If you use
2495 @code{break} without an argument in the innermost frame, @value{GDBN} stops
2496 the next time it reaches the current location; this may be useful
2499 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
2500 least one instruction has been executed. If it did not do this, you
2501 would be unable to proceed past a breakpoint without first disabling the
2502 breakpoint. This rule applies whether or not the breakpoint already
2503 existed when your program stopped.
2505 @item break @dots{} if @var{cond}
2506 Set a breakpoint with condition @var{cond}; evaluate the expression
2507 @var{cond} each time the breakpoint is reached, and stop only if the
2508 value is nonzero---that is, if @var{cond} evaluates as true.
2509 @samp{@dots{}} stands for one of the possible arguments described
2510 above (or no argument) specifying where to break. @xref{Conditions,
2511 ,Break conditions}, for more information on breakpoint conditions.
2514 @item tbreak @var{args}
2515 Set a breakpoint enabled only for one stop. @var{args} are the
2516 same as for the @code{break} command, and the breakpoint is set in the same
2517 way, but the breakpoint is automatically deleted after the first time your
2518 program stops there. @xref{Disabling, ,Disabling breakpoints}.
2521 @item hbreak @var{args}
2522 Set a hardware-assisted breakpoint. @var{args} are the same as for the
2523 @code{break} command and the breakpoint is set in the same way, but the
2524 breakpoint requires hardware support and some target hardware may not
2525 have this support. The main purpose of this is EPROM/ROM code
2526 debugging, so you can set a breakpoint at an instruction without
2527 changing the instruction. This can be used with the new trap-generation
2528 provided by SPARClite DSU and some x86-based targets. These targets
2529 will generate traps when a program accesses some data or instruction
2530 address that is assigned to the debug registers. However the hardware
2531 breakpoint registers can take a limited number of breakpoints. For
2532 example, on the DSU, only two data breakpoints can be set at a time, and
2533 @value{GDBN} will reject this command if more than two are used. Delete
2534 or disable unused hardware breakpoints before setting new ones
2535 (@pxref{Disabling, ,Disabling}). @xref{Conditions, ,Break conditions}.
2538 @item thbreak @var{args}
2539 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
2540 are the same as for the @code{hbreak} command and the breakpoint is set in
2541 the same way. However, like the @code{tbreak} command,
2542 the breakpoint is automatically deleted after the
2543 first time your program stops there. Also, like the @code{hbreak}
2544 command, the breakpoint requires hardware support and some target hardware
2545 may not have this support. @xref{Disabling, ,Disabling breakpoints}.
2546 See also @ref{Conditions, ,Break conditions}.
2549 @cindex regular expression
2550 @item rbreak @var{regex}
2551 Set breakpoints on all functions matching the regular expression
2552 @var{regex}. This command sets an unconditional breakpoint on all
2553 matches, printing a list of all breakpoints it set. Once these
2554 breakpoints are set, they are treated just like the breakpoints set with
2555 the @code{break} command. You can delete them, disable them, or make
2556 them conditional the same way as any other breakpoint.
2558 The syntax of the regular expression is the standard one used with tools
2559 like @file{grep}. Note that this is different from the syntax used by
2560 shells, so for instance @code{foo*} matches all functions that include
2561 an @code{fo} followed by zero or more @code{o}s. There is an implicit
2562 @code{.*} leading and trailing the regular expression you supply, so to
2563 match only functions that begin with @code{foo}, use @code{^foo}.
2565 When debugging C@t{++} programs, @code{rbreak} is useful for setting
2566 breakpoints on overloaded functions that are not members of any special
2569 @kindex info breakpoints
2570 @cindex @code{$_} and @code{info breakpoints}
2571 @item info breakpoints @r{[}@var{n}@r{]}
2572 @itemx info break @r{[}@var{n}@r{]}
2573 @itemx info watchpoints @r{[}@var{n}@r{]}
2574 Print a table of all breakpoints, watchpoints, and catchpoints set and
2575 not deleted, with the following columns for each breakpoint:
2578 @item Breakpoint Numbers
2580 Breakpoint, watchpoint, or catchpoint.
2582 Whether the breakpoint is marked to be disabled or deleted when hit.
2583 @item Enabled or Disabled
2584 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
2585 that are not enabled.
2587 Where the breakpoint is in your program, as a memory address.
2589 Where the breakpoint is in the source for your program, as a file and
2594 If a breakpoint is conditional, @code{info break} shows the condition on
2595 the line following the affected breakpoint; breakpoint commands, if any,
2596 are listed after that.
2599 @code{info break} with a breakpoint
2600 number @var{n} as argument lists only that breakpoint. The
2601 convenience variable @code{$_} and the default examining-address for
2602 the @code{x} command are set to the address of the last breakpoint
2603 listed (@pxref{Memory, ,Examining memory}).
2606 @code{info break} displays a count of the number of times the breakpoint
2607 has been hit. This is especially useful in conjunction with the
2608 @code{ignore} command. You can ignore a large number of breakpoint
2609 hits, look at the breakpoint info to see how many times the breakpoint
2610 was hit, and then run again, ignoring one less than that number. This
2611 will get you quickly to the last hit of that breakpoint.
2614 @value{GDBN} allows you to set any number of breakpoints at the same place in
2615 your program. There is nothing silly or meaningless about this. When
2616 the breakpoints are conditional, this is even useful
2617 (@pxref{Conditions, ,Break conditions}).
2619 @cindex negative breakpoint numbers
2620 @cindex internal @value{GDBN} breakpoints
2621 @value{GDBN} itself sometimes sets breakpoints in your program for special
2622 purposes, such as proper handling of @code{longjmp} (in C programs).
2623 These internal breakpoints are assigned negative numbers, starting with
2624 @code{-1}; @samp{info breakpoints} does not display them.
2626 You can see these breakpoints with the @value{GDBN} maintenance command
2627 @samp{maint info breakpoints}.
2630 @kindex maint info breakpoints
2631 @item maint info breakpoints
2632 Using the same format as @samp{info breakpoints}, display both the
2633 breakpoints you've set explicitly, and those @value{GDBN} is using for
2634 internal purposes. Internal breakpoints are shown with negative
2635 breakpoint numbers. The type column identifies what kind of breakpoint
2640 Normal, explicitly set breakpoint.
2643 Normal, explicitly set watchpoint.
2646 Internal breakpoint, used to handle correctly stepping through
2647 @code{longjmp} calls.
2649 @item longjmp resume
2650 Internal breakpoint at the target of a @code{longjmp}.
2653 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
2656 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
2659 Shared library events.
2666 @node Set Watchpoints
2667 @subsection Setting watchpoints
2669 @cindex setting watchpoints
2670 @cindex software watchpoints
2671 @cindex hardware watchpoints
2672 You can use a watchpoint to stop execution whenever the value of an
2673 expression changes, without having to predict a particular place where
2676 Depending on your system, watchpoints may be implemented in software or
2677 hardware. @value{GDBN} does software watchpointing by single-stepping your
2678 program and testing the variable's value each time, which is hundreds of
2679 times slower than normal execution. (But this may still be worth it, to
2680 catch errors where you have no clue what part of your program is the
2683 On some systems, such as HP-UX, Linux and some other x86-based targets,
2684 @value{GDBN} includes support for
2685 hardware watchpoints, which do not slow down the running of your
2690 @item watch @var{expr}
2691 Set a watchpoint for an expression. @value{GDBN} will break when @var{expr}
2692 is written into by the program and its value changes.
2695 @item rwatch @var{expr}
2696 Set a watchpoint that will break when watch @var{expr} is read by the program.
2699 @item awatch @var{expr}
2700 Set a watchpoint that will break when @var{expr} is either read or written into
2703 @kindex info watchpoints
2704 @item info watchpoints
2705 This command prints a list of watchpoints, breakpoints, and catchpoints;
2706 it is the same as @code{info break}.
2709 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
2710 watchpoints execute very quickly, and the debugger reports a change in
2711 value at the exact instruction where the change occurs. If @value{GDBN}
2712 cannot set a hardware watchpoint, it sets a software watchpoint, which
2713 executes more slowly and reports the change in value at the next
2714 statement, not the instruction, after the change occurs.
2716 When you issue the @code{watch} command, @value{GDBN} reports
2719 Hardware watchpoint @var{num}: @var{expr}
2723 if it was able to set a hardware watchpoint.
2725 Currently, the @code{awatch} and @code{rwatch} commands can only set
2726 hardware watchpoints, because accesses to data that don't change the
2727 value of the watched expression cannot be detected without examining
2728 every instruction as it is being executed, and @value{GDBN} does not do
2729 that currently. If @value{GDBN} finds that it is unable to set a
2730 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
2731 will print a message like this:
2734 Expression cannot be implemented with read/access watchpoint.
2737 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
2738 data type of the watched expression is wider than what a hardware
2739 watchpoint on the target machine can handle. For example, some systems
2740 can only watch regions that are up to 4 bytes wide; on such systems you
2741 cannot set hardware watchpoints for an expression that yields a
2742 double-precision floating-point number (which is typically 8 bytes
2743 wide). As a work-around, it might be possible to break the large region
2744 into a series of smaller ones and watch them with separate watchpoints.
2746 If you set too many hardware watchpoints, @value{GDBN} might be unable
2747 to insert all of them when you resume the execution of your program.
2748 Since the precise number of active watchpoints is unknown until such
2749 time as the program is about to be resumed, @value{GDBN} might not be
2750 able to warn you about this when you set the watchpoints, and the
2751 warning will be printed only when the program is resumed:
2754 Hardware watchpoint @var{num}: Could not insert watchpoint
2758 If this happens, delete or disable some of the watchpoints.
2760 The SPARClite DSU will generate traps when a program accesses some data
2761 or instruction address that is assigned to the debug registers. For the
2762 data addresses, DSU facilitates the @code{watch} command. However the
2763 hardware breakpoint registers can only take two data watchpoints, and
2764 both watchpoints must be the same kind. For example, you can set two
2765 watchpoints with @code{watch} commands, two with @code{rwatch} commands,
2766 @strong{or} two with @code{awatch} commands, but you cannot set one
2767 watchpoint with one command and the other with a different command.
2768 @value{GDBN} will reject the command if you try to mix watchpoints.
2769 Delete or disable unused watchpoint commands before setting new ones.
2771 If you call a function interactively using @code{print} or @code{call},
2772 any watchpoints you have set will be inactive until @value{GDBN} reaches another
2773 kind of breakpoint or the call completes.
2775 @value{GDBN} automatically deletes watchpoints that watch local
2776 (automatic) variables, or expressions that involve such variables, when
2777 they go out of scope, that is, when the execution leaves the block in
2778 which these variables were defined. In particular, when the program
2779 being debugged terminates, @emph{all} local variables go out of scope,
2780 and so only watchpoints that watch global variables remain set. If you
2781 rerun the program, you will need to set all such watchpoints again. One
2782 way of doing that would be to set a code breakpoint at the entry to the
2783 @code{main} function and when it breaks, set all the watchpoints.
2786 @cindex watchpoints and threads
2787 @cindex threads and watchpoints
2788 @emph{Warning:} In multi-thread programs, watchpoints have only limited
2789 usefulness. With the current watchpoint implementation, @value{GDBN}
2790 can only watch the value of an expression @emph{in a single thread}. If
2791 you are confident that the expression can only change due to the current
2792 thread's activity (and if you are also confident that no other thread
2793 can become current), then you can use watchpoints as usual. However,
2794 @value{GDBN} may not notice when a non-current thread's activity changes
2797 @c FIXME: this is almost identical to the previous paragraph.
2798 @emph{HP-UX Warning:} In multi-thread programs, software watchpoints
2799 have only limited usefulness. If @value{GDBN} creates a software
2800 watchpoint, it can only watch the value of an expression @emph{in a
2801 single thread}. If you are confident that the expression can only
2802 change due to the current thread's activity (and if you are also
2803 confident that no other thread can become current), then you can use
2804 software watchpoints as usual. However, @value{GDBN} may not notice
2805 when a non-current thread's activity changes the expression. (Hardware
2806 watchpoints, in contrast, watch an expression in all threads.)
2809 @node Set Catchpoints
2810 @subsection Setting catchpoints
2811 @cindex catchpoints, setting
2812 @cindex exception handlers
2813 @cindex event handling
2815 You can use @dfn{catchpoints} to cause the debugger to stop for certain
2816 kinds of program events, such as C@t{++} exceptions or the loading of a
2817 shared library. Use the @code{catch} command to set a catchpoint.
2821 @item catch @var{event}
2822 Stop when @var{event} occurs. @var{event} can be any of the following:
2826 The throwing of a C@t{++} exception.
2830 The catching of a C@t{++} exception.
2834 A call to @code{exec}. This is currently only available for HP-UX.
2838 A call to @code{fork}. This is currently only available for HP-UX.
2842 A call to @code{vfork}. This is currently only available for HP-UX.
2845 @itemx load @var{libname}
2847 The dynamic loading of any shared library, or the loading of the library
2848 @var{libname}. This is currently only available for HP-UX.
2851 @itemx unload @var{libname}
2852 @kindex catch unload
2853 The unloading of any dynamically loaded shared library, or the unloading
2854 of the library @var{libname}. This is currently only available for HP-UX.
2857 @item tcatch @var{event}
2858 Set a catchpoint that is enabled only for one stop. The catchpoint is
2859 automatically deleted after the first time the event is caught.
2863 Use the @code{info break} command to list the current catchpoints.
2865 There are currently some limitations to C@t{++} exception handling
2866 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
2870 If you call a function interactively, @value{GDBN} normally returns
2871 control to you when the function has finished executing. If the call
2872 raises an exception, however, the call may bypass the mechanism that
2873 returns control to you and cause your program either to abort or to
2874 simply continue running until it hits a breakpoint, catches a signal
2875 that @value{GDBN} is listening for, or exits. This is the case even if
2876 you set a catchpoint for the exception; catchpoints on exceptions are
2877 disabled within interactive calls.
2880 You cannot raise an exception interactively.
2883 You cannot install an exception handler interactively.
2886 @cindex raise exceptions
2887 Sometimes @code{catch} is not the best way to debug exception handling:
2888 if you need to know exactly where an exception is raised, it is better to
2889 stop @emph{before} the exception handler is called, since that way you
2890 can see the stack before any unwinding takes place. If you set a
2891 breakpoint in an exception handler instead, it may not be easy to find
2892 out where the exception was raised.
2894 To stop just before an exception handler is called, you need some
2895 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
2896 raised by calling a library function named @code{__raise_exception}
2897 which has the following ANSI C interface:
2900 /* @var{addr} is where the exception identifier is stored.
2901 @var{id} is the exception identifier. */
2902 void __raise_exception (void **addr, void *id);
2906 To make the debugger catch all exceptions before any stack
2907 unwinding takes place, set a breakpoint on @code{__raise_exception}
2908 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions}).
2910 With a conditional breakpoint (@pxref{Conditions, ,Break conditions})
2911 that depends on the value of @var{id}, you can stop your program when
2912 a specific exception is raised. You can use multiple conditional
2913 breakpoints to stop your program when any of a number of exceptions are
2918 @subsection Deleting breakpoints
2920 @cindex clearing breakpoints, watchpoints, catchpoints
2921 @cindex deleting breakpoints, watchpoints, catchpoints
2922 It is often necessary to eliminate a breakpoint, watchpoint, or
2923 catchpoint once it has done its job and you no longer want your program
2924 to stop there. This is called @dfn{deleting} the breakpoint. A
2925 breakpoint that has been deleted no longer exists; it is forgotten.
2927 With the @code{clear} command you can delete breakpoints according to
2928 where they are in your program. With the @code{delete} command you can
2929 delete individual breakpoints, watchpoints, or catchpoints by specifying
2930 their breakpoint numbers.
2932 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
2933 automatically ignores breakpoints on the first instruction to be executed
2934 when you continue execution without changing the execution address.
2939 Delete any breakpoints at the next instruction to be executed in the
2940 selected stack frame (@pxref{Selection, ,Selecting a frame}). When
2941 the innermost frame is selected, this is a good way to delete a
2942 breakpoint where your program just stopped.
2944 @item clear @var{function}
2945 @itemx clear @var{filename}:@var{function}
2946 Delete any breakpoints set at entry to the function @var{function}.
2948 @item clear @var{linenum}
2949 @itemx clear @var{filename}:@var{linenum}
2950 Delete any breakpoints set at or within the code of the specified line.
2952 @cindex delete breakpoints
2954 @kindex d @r{(@code{delete})}
2955 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2956 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
2957 ranges specified as arguments. If no argument is specified, delete all
2958 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
2959 confirm off}). You can abbreviate this command as @code{d}.
2963 @subsection Disabling breakpoints
2965 @kindex disable breakpoints
2966 @kindex enable breakpoints
2967 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
2968 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
2969 it had been deleted, but remembers the information on the breakpoint so
2970 that you can @dfn{enable} it again later.
2972 You disable and enable breakpoints, watchpoints, and catchpoints with
2973 the @code{enable} and @code{disable} commands, optionally specifying one
2974 or more breakpoint numbers as arguments. Use @code{info break} or
2975 @code{info watch} to print a list of breakpoints, watchpoints, and
2976 catchpoints if you do not know which numbers to use.
2978 A breakpoint, watchpoint, or catchpoint can have any of four different
2979 states of enablement:
2983 Enabled. The breakpoint stops your program. A breakpoint set
2984 with the @code{break} command starts out in this state.
2986 Disabled. The breakpoint has no effect on your program.
2988 Enabled once. The breakpoint stops your program, but then becomes
2991 Enabled for deletion. The breakpoint stops your program, but
2992 immediately after it does so it is deleted permanently. A breakpoint
2993 set with the @code{tbreak} command starts out in this state.
2996 You can use the following commands to enable or disable breakpoints,
2997 watchpoints, and catchpoints:
3000 @kindex disable breakpoints
3002 @kindex dis @r{(@code{disable})}
3003 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3004 Disable the specified breakpoints---or all breakpoints, if none are
3005 listed. A disabled breakpoint has no effect but is not forgotten. All
3006 options such as ignore-counts, conditions and commands are remembered in
3007 case the breakpoint is enabled again later. You may abbreviate
3008 @code{disable} as @code{dis}.
3010 @kindex enable breakpoints
3012 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3013 Enable the specified breakpoints (or all defined breakpoints). They
3014 become effective once again in stopping your program.
3016 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
3017 Enable the specified breakpoints temporarily. @value{GDBN} disables any
3018 of these breakpoints immediately after stopping your program.
3020 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
3021 Enable the specified breakpoints to work once, then die. @value{GDBN}
3022 deletes any of these breakpoints as soon as your program stops there.
3025 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
3026 @c confusing: tbreak is also initially enabled.
3027 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
3028 ,Setting breakpoints}), breakpoints that you set are initially enabled;
3029 subsequently, they become disabled or enabled only when you use one of
3030 the commands above. (The command @code{until} can set and delete a
3031 breakpoint of its own, but it does not change the state of your other
3032 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
3036 @subsection Break conditions
3037 @cindex conditional breakpoints
3038 @cindex breakpoint conditions
3040 @c FIXME what is scope of break condition expr? Context where wanted?
3041 @c in particular for a watchpoint?
3042 The simplest sort of breakpoint breaks every time your program reaches a
3043 specified place. You can also specify a @dfn{condition} for a
3044 breakpoint. A condition is just a Boolean expression in your
3045 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
3046 a condition evaluates the expression each time your program reaches it,
3047 and your program stops only if the condition is @emph{true}.
3049 This is the converse of using assertions for program validation; in that
3050 situation, you want to stop when the assertion is violated---that is,
3051 when the condition is false. In C, if you want to test an assertion expressed
3052 by the condition @var{assert}, you should set the condition
3053 @samp{! @var{assert}} on the appropriate breakpoint.
3055 Conditions are also accepted for watchpoints; you may not need them,
3056 since a watchpoint is inspecting the value of an expression anyhow---but
3057 it might be simpler, say, to just set a watchpoint on a variable name,
3058 and specify a condition that tests whether the new value is an interesting
3061 Break conditions can have side effects, and may even call functions in
3062 your program. This can be useful, for example, to activate functions
3063 that log program progress, or to use your own print functions to
3064 format special data structures. The effects are completely predictable
3065 unless there is another enabled breakpoint at the same address. (In
3066 that case, @value{GDBN} might see the other breakpoint first and stop your
3067 program without checking the condition of this one.) Note that
3068 breakpoint commands are usually more convenient and flexible than break
3070 purpose of performing side effects when a breakpoint is reached
3071 (@pxref{Break Commands, ,Breakpoint command lists}).
3073 Break conditions can be specified when a breakpoint is set, by using
3074 @samp{if} in the arguments to the @code{break} command. @xref{Set
3075 Breaks, ,Setting breakpoints}. They can also be changed at any time
3076 with the @code{condition} command.
3078 You can also use the @code{if} keyword with the @code{watch} command.
3079 The @code{catch} command does not recognize the @code{if} keyword;
3080 @code{condition} is the only way to impose a further condition on a
3085 @item condition @var{bnum} @var{expression}
3086 Specify @var{expression} as the break condition for breakpoint,
3087 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
3088 breakpoint @var{bnum} stops your program only if the value of
3089 @var{expression} is true (nonzero, in C). When you use
3090 @code{condition}, @value{GDBN} checks @var{expression} immediately for
3091 syntactic correctness, and to determine whether symbols in it have
3092 referents in the context of your breakpoint. If @var{expression} uses
3093 symbols not referenced in the context of the breakpoint, @value{GDBN}
3094 prints an error message:
3097 No symbol "foo" in current context.
3102 not actually evaluate @var{expression} at the time the @code{condition}
3103 command (or a command that sets a breakpoint with a condition, like
3104 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
3106 @item condition @var{bnum}
3107 Remove the condition from breakpoint number @var{bnum}. It becomes
3108 an ordinary unconditional breakpoint.
3111 @cindex ignore count (of breakpoint)
3112 A special case of a breakpoint condition is to stop only when the
3113 breakpoint has been reached a certain number of times. This is so
3114 useful that there is a special way to do it, using the @dfn{ignore
3115 count} of the breakpoint. Every breakpoint has an ignore count, which
3116 is an integer. Most of the time, the ignore count is zero, and
3117 therefore has no effect. But if your program reaches a breakpoint whose
3118 ignore count is positive, then instead of stopping, it just decrements
3119 the ignore count by one and continues. As a result, if the ignore count
3120 value is @var{n}, the breakpoint does not stop the next @var{n} times
3121 your program reaches it.
3125 @item ignore @var{bnum} @var{count}
3126 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
3127 The next @var{count} times the breakpoint is reached, your program's
3128 execution does not stop; other than to decrement the ignore count, @value{GDBN}
3131 To make the breakpoint stop the next time it is reached, specify
3134 When you use @code{continue} to resume execution of your program from a
3135 breakpoint, you can specify an ignore count directly as an argument to
3136 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
3137 Stepping,,Continuing and stepping}.
3139 If a breakpoint has a positive ignore count and a condition, the
3140 condition is not checked. Once the ignore count reaches zero,
3141 @value{GDBN} resumes checking the condition.
3143 You could achieve the effect of the ignore count with a condition such
3144 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
3145 is decremented each time. @xref{Convenience Vars, ,Convenience
3149 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
3152 @node Break Commands
3153 @subsection Breakpoint command lists
3155 @cindex breakpoint commands
3156 You can give any breakpoint (or watchpoint or catchpoint) a series of
3157 commands to execute when your program stops due to that breakpoint. For
3158 example, you might want to print the values of certain expressions, or
3159 enable other breakpoints.
3164 @item commands @r{[}@var{bnum}@r{]}
3165 @itemx @dots{} @var{command-list} @dots{}
3167 Specify a list of commands for breakpoint number @var{bnum}. The commands
3168 themselves appear on the following lines. Type a line containing just
3169 @code{end} to terminate the commands.
3171 To remove all commands from a breakpoint, type @code{commands} and
3172 follow it immediately with @code{end}; that is, give no commands.
3174 With no @var{bnum} argument, @code{commands} refers to the last
3175 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
3176 recently encountered).
3179 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
3180 disabled within a @var{command-list}.
3182 You can use breakpoint commands to start your program up again. Simply
3183 use the @code{continue} command, or @code{step}, or any other command
3184 that resumes execution.
3186 Any other commands in the command list, after a command that resumes
3187 execution, are ignored. This is because any time you resume execution
3188 (even with a simple @code{next} or @code{step}), you may encounter
3189 another breakpoint---which could have its own command list, leading to
3190 ambiguities about which list to execute.
3193 If the first command you specify in a command list is @code{silent}, the
3194 usual message about stopping at a breakpoint is not printed. This may
3195 be desirable for breakpoints that are to print a specific message and
3196 then continue. If none of the remaining commands print anything, you
3197 see no sign that the breakpoint was reached. @code{silent} is
3198 meaningful only at the beginning of a breakpoint command list.
3200 The commands @code{echo}, @code{output}, and @code{printf} allow you to
3201 print precisely controlled output, and are often useful in silent
3202 breakpoints. @xref{Output, ,Commands for controlled output}.
3204 For example, here is how you could use breakpoint commands to print the
3205 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
3211 printf "x is %d\n",x
3216 One application for breakpoint commands is to compensate for one bug so
3217 you can test for another. Put a breakpoint just after the erroneous line
3218 of code, give it a condition to detect the case in which something
3219 erroneous has been done, and give it commands to assign correct values
3220 to any variables that need them. End with the @code{continue} command
3221 so that your program does not stop, and start with the @code{silent}
3222 command so that no output is produced. Here is an example:
3233 @node Breakpoint Menus
3234 @subsection Breakpoint menus
3236 @cindex symbol overloading
3238 Some programming languages (notably C@t{++}) permit a single function name
3239 to be defined several times, for application in different contexts.
3240 This is called @dfn{overloading}. When a function name is overloaded,
3241 @samp{break @var{function}} is not enough to tell @value{GDBN} where you want
3242 a breakpoint. If you realize this is a problem, you can use
3243 something like @samp{break @var{function}(@var{types})} to specify which
3244 particular version of the function you want. Otherwise, @value{GDBN} offers
3245 you a menu of numbered choices for different possible breakpoints, and
3246 waits for your selection with the prompt @samp{>}. The first two
3247 options are always @samp{[0] cancel} and @samp{[1] all}. Typing @kbd{1}
3248 sets a breakpoint at each definition of @var{function}, and typing
3249 @kbd{0} aborts the @code{break} command without setting any new
3252 For example, the following session excerpt shows an attempt to set a
3253 breakpoint at the overloaded symbol @code{String::after}.
3254 We choose three particular definitions of that function name:
3256 @c FIXME! This is likely to change to show arg type lists, at least
3259 (@value{GDBP}) b String::after
3262 [2] file:String.cc; line number:867
3263 [3] file:String.cc; line number:860
3264 [4] file:String.cc; line number:875
3265 [5] file:String.cc; line number:853
3266 [6] file:String.cc; line number:846
3267 [7] file:String.cc; line number:735
3269 Breakpoint 1 at 0xb26c: file String.cc, line 867.
3270 Breakpoint 2 at 0xb344: file String.cc, line 875.
3271 Breakpoint 3 at 0xafcc: file String.cc, line 846.
3272 Multiple breakpoints were set.
3273 Use the "delete" command to delete unwanted
3279 @c @ifclear BARETARGET
3280 @node Error in Breakpoints
3281 @subsection ``Cannot insert breakpoints''
3283 @c FIXME!! 14/6/95 Is there a real example of this? Let's use it.
3285 Under some operating systems, breakpoints cannot be used in a program if
3286 any other process is running that program. In this situation,
3287 attempting to run or continue a program with a breakpoint causes
3288 @value{GDBN} to print an error message:
3291 Cannot insert breakpoints.
3292 The same program may be running in another process.
3295 When this happens, you have three ways to proceed:
3299 Remove or disable the breakpoints, then continue.
3302 Suspend @value{GDBN}, and copy the file containing your program to a new
3303 name. Resume @value{GDBN} and use the @code{exec-file} command to specify
3304 that @value{GDBN} should run your program under that name.
3305 Then start your program again.
3308 Relink your program so that the text segment is nonsharable, using the
3309 linker option @samp{-N}. The operating system limitation may not apply
3310 to nonsharable executables.
3314 A similar message can be printed if you request too many active
3315 hardware-assisted breakpoints and watchpoints:
3317 @c FIXME: the precise wording of this message may change; the relevant
3318 @c source change is not committed yet (Sep 3, 1999).
3320 Stopped; cannot insert breakpoints.
3321 You may have requested too many hardware breakpoints and watchpoints.
3325 This message is printed when you attempt to resume the program, since
3326 only then @value{GDBN} knows exactly how many hardware breakpoints and
3327 watchpoints it needs to insert.
3329 When this message is printed, you need to disable or remove some of the
3330 hardware-assisted breakpoints and watchpoints, and then continue.
3333 @node Continuing and Stepping
3334 @section Continuing and stepping
3338 @cindex resuming execution
3339 @dfn{Continuing} means resuming program execution until your program
3340 completes normally. In contrast, @dfn{stepping} means executing just
3341 one more ``step'' of your program, where ``step'' may mean either one
3342 line of source code, or one machine instruction (depending on what
3343 particular command you use). Either when continuing or when stepping,
3344 your program may stop even sooner, due to a breakpoint or a signal. (If
3345 it stops due to a signal, you may want to use @code{handle}, or use
3346 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
3350 @kindex c @r{(@code{continue})}
3351 @kindex fg @r{(resume foreground execution)}
3352 @item continue @r{[}@var{ignore-count}@r{]}
3353 @itemx c @r{[}@var{ignore-count}@r{]}
3354 @itemx fg @r{[}@var{ignore-count}@r{]}
3355 Resume program execution, at the address where your program last stopped;
3356 any breakpoints set at that address are bypassed. The optional argument
3357 @var{ignore-count} allows you to specify a further number of times to
3358 ignore a breakpoint at this location; its effect is like that of
3359 @code{ignore} (@pxref{Conditions, ,Break conditions}).
3361 The argument @var{ignore-count} is meaningful only when your program
3362 stopped due to a breakpoint. At other times, the argument to
3363 @code{continue} is ignored.
3365 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
3366 debugged program is deemed to be the foreground program) are provided
3367 purely for convenience, and have exactly the same behavior as
3371 To resume execution at a different place, you can use @code{return}
3372 (@pxref{Returning, ,Returning from a function}) to go back to the
3373 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
3374 different address}) to go to an arbitrary location in your program.
3376 A typical technique for using stepping is to set a breakpoint
3377 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and catchpoints}) at the
3378 beginning of the function or the section of your program where a problem
3379 is believed to lie, run your program until it stops at that breakpoint,
3380 and then step through the suspect area, examining the variables that are
3381 interesting, until you see the problem happen.
3385 @kindex s @r{(@code{step})}
3387 Continue running your program until control reaches a different source
3388 line, then stop it and return control to @value{GDBN}. This command is
3389 abbreviated @code{s}.
3392 @c "without debugging information" is imprecise; actually "without line
3393 @c numbers in the debugging information". (gcc -g1 has debugging info but
3394 @c not line numbers). But it seems complex to try to make that
3395 @c distinction here.
3396 @emph{Warning:} If you use the @code{step} command while control is
3397 within a function that was compiled without debugging information,
3398 execution proceeds until control reaches a function that does have
3399 debugging information. Likewise, it will not step into a function which
3400 is compiled without debugging information. To step through functions
3401 without debugging information, use the @code{stepi} command, described
3405 The @code{step} command only stops at the first instruction of a source
3406 line. This prevents the multiple stops that could otherwise occur in
3407 @code{switch} statements, @code{for} loops, etc. @code{step} continues
3408 to stop if a function that has debugging information is called within
3409 the line. In other words, @code{step} @emph{steps inside} any functions
3410 called within the line.
3412 Also, the @code{step} command only enters a function if there is line
3413 number information for the function. Otherwise it acts like the
3414 @code{next} command. This avoids problems when using @code{cc -gl}
3415 on MIPS machines. Previously, @code{step} entered subroutines if there
3416 was any debugging information about the routine.
3418 @item step @var{count}
3419 Continue running as in @code{step}, but do so @var{count} times. If a
3420 breakpoint is reached, or a signal not related to stepping occurs before
3421 @var{count} steps, stepping stops right away.
3424 @kindex n @r{(@code{next})}
3425 @item next @r{[}@var{count}@r{]}
3426 Continue to the next source line in the current (innermost) stack frame.
3427 This is similar to @code{step}, but function calls that appear within
3428 the line of code are executed without stopping. Execution stops when
3429 control reaches a different line of code at the original stack level
3430 that was executing when you gave the @code{next} command. This command
3431 is abbreviated @code{n}.
3433 An argument @var{count} is a repeat count, as for @code{step}.
3436 @c FIX ME!! Do we delete this, or is there a way it fits in with
3437 @c the following paragraph? --- Vctoria
3439 @c @code{next} within a function that lacks debugging information acts like
3440 @c @code{step}, but any function calls appearing within the code of the
3441 @c function are executed without stopping.
3443 The @code{next} command only stops at the first instruction of a
3444 source line. This prevents multiple stops that could otherwise occur in
3445 @code{switch} statements, @code{for} loops, etc.
3447 @kindex set step-mode
3449 @cindex functions without line info, and stepping
3450 @cindex stepping into functions with no line info
3451 @itemx set step-mode on
3452 The @code{set step-mode on} command causes the @code{step} command to
3453 stop at the first instruction of a function which contains no debug line
3454 information rather than stepping over it.
3456 This is useful in cases where you may be interested in inspecting the
3457 machine instructions of a function which has no symbolic info and do not
3458 want @value{GDBN} to automatically skip over this function.
3460 @item set step-mode off
3461 Causes the @code{step} command to step over any functions which contains no
3462 debug information. This is the default.
3466 Continue running until just after function in the selected stack frame
3467 returns. Print the returned value (if any).
3469 Contrast this with the @code{return} command (@pxref{Returning,
3470 ,Returning from a function}).
3473 @kindex u @r{(@code{until})}
3476 Continue running until a source line past the current line, in the
3477 current stack frame, is reached. This command is used to avoid single
3478 stepping through a loop more than once. It is like the @code{next}
3479 command, except that when @code{until} encounters a jump, it
3480 automatically continues execution until the program counter is greater
3481 than the address of the jump.
3483 This means that when you reach the end of a loop after single stepping
3484 though it, @code{until} makes your program continue execution until it
3485 exits the loop. In contrast, a @code{next} command at the end of a loop
3486 simply steps back to the beginning of the loop, which forces you to step
3487 through the next iteration.
3489 @code{until} always stops your program if it attempts to exit the current
3492 @code{until} may produce somewhat counterintuitive results if the order
3493 of machine code does not match the order of the source lines. For
3494 example, in the following excerpt from a debugging session, the @code{f}
3495 (@code{frame}) command shows that execution is stopped at line
3496 @code{206}; yet when we use @code{until}, we get to line @code{195}:
3500 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
3502 (@value{GDBP}) until
3503 195 for ( ; argc > 0; NEXTARG) @{
3506 This happened because, for execution efficiency, the compiler had
3507 generated code for the loop closure test at the end, rather than the
3508 start, of the loop---even though the test in a C @code{for}-loop is
3509 written before the body of the loop. The @code{until} command appeared
3510 to step back to the beginning of the loop when it advanced to this
3511 expression; however, it has not really gone to an earlier
3512 statement---not in terms of the actual machine code.
3514 @code{until} with no argument works by means of single
3515 instruction stepping, and hence is slower than @code{until} with an
3518 @item until @var{location}
3519 @itemx u @var{location}
3520 Continue running your program until either the specified location is
3521 reached, or the current stack frame returns. @var{location} is any of
3522 the forms of argument acceptable to @code{break} (@pxref{Set Breaks,
3523 ,Setting breakpoints}). This form of the command uses breakpoints,
3524 and hence is quicker than @code{until} without an argument.
3527 @kindex si @r{(@code{stepi})}
3529 @itemx stepi @var{arg}
3531 Execute one machine instruction, then stop and return to the debugger.
3533 It is often useful to do @samp{display/i $pc} when stepping by machine
3534 instructions. This makes @value{GDBN} automatically display the next
3535 instruction to be executed, each time your program stops. @xref{Auto
3536 Display,, Automatic display}.
3538 An argument is a repeat count, as in @code{step}.
3542 @kindex ni @r{(@code{nexti})}
3544 @itemx nexti @var{arg}
3546 Execute one machine instruction, but if it is a function call,
3547 proceed until the function returns.
3549 An argument is a repeat count, as in @code{next}.
3556 A signal is an asynchronous event that can happen in a program. The
3557 operating system defines the possible kinds of signals, and gives each
3558 kind a name and a number. For example, in Unix @code{SIGINT} is the
3559 signal a program gets when you type an interrupt character (often @kbd{C-c});
3560 @code{SIGSEGV} is the signal a program gets from referencing a place in
3561 memory far away from all the areas in use; @code{SIGALRM} occurs when
3562 the alarm clock timer goes off (which happens only if your program has
3563 requested an alarm).
3565 @cindex fatal signals
3566 Some signals, including @code{SIGALRM}, are a normal part of the
3567 functioning of your program. Others, such as @code{SIGSEGV}, indicate
3568 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
3569 program has not specified in advance some other way to handle the signal.
3570 @code{SIGINT} does not indicate an error in your program, but it is normally
3571 fatal so it can carry out the purpose of the interrupt: to kill the program.
3573 @value{GDBN} has the ability to detect any occurrence of a signal in your
3574 program. You can tell @value{GDBN} in advance what to do for each kind of
3577 @cindex handling signals
3578 Normally, @value{GDBN} is set up to let the non-erroneous signals like
3579 @code{SIGALRM} be silently passed to your program
3580 (so as not to interfere with their role in the program's functioning)
3581 but to stop your program immediately whenever an error signal happens.
3582 You can change these settings with the @code{handle} command.
3585 @kindex info signals
3588 Print a table of all the kinds of signals and how @value{GDBN} has been told to
3589 handle each one. You can use this to see the signal numbers of all
3590 the defined types of signals.
3592 @code{info handle} is an alias for @code{info signals}.
3595 @item handle @var{signal} @var{keywords}@dots{}
3596 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
3597 can be the number of a signal or its name (with or without the
3598 @samp{SIG} at the beginning); a list of signal numbers of the form
3599 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
3600 known signals. The @var{keywords} say what change to make.
3604 The keywords allowed by the @code{handle} command can be abbreviated.
3605 Their full names are:
3609 @value{GDBN} should not stop your program when this signal happens. It may
3610 still print a message telling you that the signal has come in.
3613 @value{GDBN} should stop your program when this signal happens. This implies
3614 the @code{print} keyword as well.
3617 @value{GDBN} should print a message when this signal happens.
3620 @value{GDBN} should not mention the occurrence of the signal at all. This
3621 implies the @code{nostop} keyword as well.
3625 @value{GDBN} should allow your program to see this signal; your program
3626 can handle the signal, or else it may terminate if the signal is fatal
3627 and not handled. @code{pass} and @code{noignore} are synonyms.
3631 @value{GDBN} should not allow your program to see this signal.
3632 @code{nopass} and @code{ignore} are synonyms.
3636 When a signal stops your program, the signal is not visible to the
3638 continue. Your program sees the signal then, if @code{pass} is in
3639 effect for the signal in question @emph{at that time}. In other words,
3640 after @value{GDBN} reports a signal, you can use the @code{handle}
3641 command with @code{pass} or @code{nopass} to control whether your
3642 program sees that signal when you continue.
3644 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
3645 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
3646 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
3649 You can also use the @code{signal} command to prevent your program from
3650 seeing a signal, or cause it to see a signal it normally would not see,
3651 or to give it any signal at any time. For example, if your program stopped
3652 due to some sort of memory reference error, you might store correct
3653 values into the erroneous variables and continue, hoping to see more
3654 execution; but your program would probably terminate immediately as
3655 a result of the fatal signal once it saw the signal. To prevent this,
3656 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
3660 @section Stopping and starting multi-thread programs
3662 When your program has multiple threads (@pxref{Threads,, Debugging
3663 programs with multiple threads}), you can choose whether to set
3664 breakpoints on all threads, or on a particular thread.
3667 @cindex breakpoints and threads
3668 @cindex thread breakpoints
3669 @kindex break @dots{} thread @var{threadno}
3670 @item break @var{linespec} thread @var{threadno}
3671 @itemx break @var{linespec} thread @var{threadno} if @dots{}
3672 @var{linespec} specifies source lines; there are several ways of
3673 writing them, but the effect is always to specify some source line.
3675 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
3676 to specify that you only want @value{GDBN} to stop the program when a
3677 particular thread reaches this breakpoint. @var{threadno} is one of the
3678 numeric thread identifiers assigned by @value{GDBN}, shown in the first
3679 column of the @samp{info threads} display.
3681 If you do not specify @samp{thread @var{threadno}} when you set a
3682 breakpoint, the breakpoint applies to @emph{all} threads of your
3685 You can use the @code{thread} qualifier on conditional breakpoints as
3686 well; in this case, place @samp{thread @var{threadno}} before the
3687 breakpoint condition, like this:
3690 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
3695 @cindex stopped threads
3696 @cindex threads, stopped
3697 Whenever your program stops under @value{GDBN} for any reason,
3698 @emph{all} threads of execution stop, not just the current thread. This
3699 allows you to examine the overall state of the program, including
3700 switching between threads, without worrying that things may change
3703 @cindex continuing threads
3704 @cindex threads, continuing
3705 Conversely, whenever you restart the program, @emph{all} threads start
3706 executing. @emph{This is true even when single-stepping} with commands
3707 like @code{step} or @code{next}.
3709 In particular, @value{GDBN} cannot single-step all threads in lockstep.
3710 Since thread scheduling is up to your debugging target's operating
3711 system (not controlled by @value{GDBN}), other threads may
3712 execute more than one statement while the current thread completes a
3713 single step. Moreover, in general other threads stop in the middle of a
3714 statement, rather than at a clean statement boundary, when the program
3717 You might even find your program stopped in another thread after
3718 continuing or even single-stepping. This happens whenever some other
3719 thread runs into a breakpoint, a signal, or an exception before the
3720 first thread completes whatever you requested.
3722 On some OSes, you can lock the OS scheduler and thus allow only a single
3726 @item set scheduler-locking @var{mode}
3727 Set the scheduler locking mode. If it is @code{off}, then there is no
3728 locking and any thread may run at any time. If @code{on}, then only the
3729 current thread may run when the inferior is resumed. The @code{step}
3730 mode optimizes for single-stepping. It stops other threads from
3731 ``seizing the prompt'' by preempting the current thread while you are
3732 stepping. Other threads will only rarely (or never) get a chance to run
3733 when you step. They are more likely to run when you @samp{next} over a
3734 function call, and they are completely free to run when you use commands
3735 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
3736 thread hits a breakpoint during its timeslice, they will never steal the
3737 @value{GDBN} prompt away from the thread that you are debugging.
3739 @item show scheduler-locking
3740 Display the current scheduler locking mode.
3745 @chapter Examining the Stack
3747 When your program has stopped, the first thing you need to know is where it
3748 stopped and how it got there.
3751 Each time your program performs a function call, information about the call
3753 That information includes the location of the call in your program,
3754 the arguments of the call,
3755 and the local variables of the function being called.
3756 The information is saved in a block of data called a @dfn{stack frame}.
3757 The stack frames are allocated in a region of memory called the @dfn{call
3760 When your program stops, the @value{GDBN} commands for examining the
3761 stack allow you to see all of this information.
3763 @cindex selected frame
3764 One of the stack frames is @dfn{selected} by @value{GDBN} and many
3765 @value{GDBN} commands refer implicitly to the selected frame. In
3766 particular, whenever you ask @value{GDBN} for the value of a variable in
3767 your program, the value is found in the selected frame. There are
3768 special @value{GDBN} commands to select whichever frame you are
3769 interested in. @xref{Selection, ,Selecting a frame}.
3771 When your program stops, @value{GDBN} automatically selects the
3772 currently executing frame and describes it briefly, similar to the
3773 @code{frame} command (@pxref{Frame Info, ,Information about a frame}).
3776 * Frames:: Stack frames
3777 * Backtrace:: Backtraces
3778 * Selection:: Selecting a frame
3779 * Frame Info:: Information on a frame
3784 @section Stack frames
3786 @cindex frame, definition
3788 The call stack is divided up into contiguous pieces called @dfn{stack
3789 frames}, or @dfn{frames} for short; each frame is the data associated
3790 with one call to one function. The frame contains the arguments given
3791 to the function, the function's local variables, and the address at
3792 which the function is executing.
3794 @cindex initial frame
3795 @cindex outermost frame
3796 @cindex innermost frame
3797 When your program is started, the stack has only one frame, that of the
3798 function @code{main}. This is called the @dfn{initial} frame or the
3799 @dfn{outermost} frame. Each time a function is called, a new frame is
3800 made. Each time a function returns, the frame for that function invocation
3801 is eliminated. If a function is recursive, there can be many frames for
3802 the same function. The frame for the function in which execution is
3803 actually occurring is called the @dfn{innermost} frame. This is the most
3804 recently created of all the stack frames that still exist.
3806 @cindex frame pointer
3807 Inside your program, stack frames are identified by their addresses. A
3808 stack frame consists of many bytes, each of which has its own address; each
3809 kind of computer has a convention for choosing one byte whose
3810 address serves as the address of the frame. Usually this address is kept
3811 in a register called the @dfn{frame pointer register} while execution is
3812 going on in that frame.
3814 @cindex frame number
3815 @value{GDBN} assigns numbers to all existing stack frames, starting with
3816 zero for the innermost frame, one for the frame that called it,
3817 and so on upward. These numbers do not really exist in your program;
3818 they are assigned by @value{GDBN} to give you a way of designating stack
3819 frames in @value{GDBN} commands.
3821 @c The -fomit-frame-pointer below perennially causes hbox overflow
3822 @c underflow problems.
3823 @cindex frameless execution
3824 Some compilers provide a way to compile functions so that they operate
3825 without stack frames. (For example, the @value{GCC} option
3827 @samp{-fomit-frame-pointer}
3829 generates functions without a frame.)
3830 This is occasionally done with heavily used library functions to save
3831 the frame setup time. @value{GDBN} has limited facilities for dealing
3832 with these function invocations. If the innermost function invocation
3833 has no stack frame, @value{GDBN} nevertheless regards it as though
3834 it had a separate frame, which is numbered zero as usual, allowing
3835 correct tracing of the function call chain. However, @value{GDBN} has
3836 no provision for frameless functions elsewhere in the stack.
3839 @kindex frame@r{, command}
3840 @cindex current stack frame
3841 @item frame @var{args}
3842 The @code{frame} command allows you to move from one stack frame to another,
3843 and to print the stack frame you select. @var{args} may be either the
3844 address of the frame or the stack frame number. Without an argument,
3845 @code{frame} prints the current stack frame.
3847 @kindex select-frame
3848 @cindex selecting frame silently
3850 The @code{select-frame} command allows you to move from one stack frame
3851 to another without printing the frame. This is the silent version of
3860 @cindex stack traces
3861 A backtrace is a summary of how your program got where it is. It shows one
3862 line per frame, for many frames, starting with the currently executing
3863 frame (frame zero), followed by its caller (frame one), and on up the
3868 @kindex bt @r{(@code{backtrace})}
3871 Print a backtrace of the entire stack: one line per frame for all
3872 frames in the stack.
3874 You can stop the backtrace at any time by typing the system interrupt
3875 character, normally @kbd{C-c}.
3877 @item backtrace @var{n}
3879 Similar, but print only the innermost @var{n} frames.
3881 @item backtrace -@var{n}
3883 Similar, but print only the outermost @var{n} frames.
3888 @kindex info s @r{(@code{info stack})}
3889 The names @code{where} and @code{info stack} (abbreviated @code{info s})
3890 are additional aliases for @code{backtrace}.
3892 Each line in the backtrace shows the frame number and the function name.
3893 The program counter value is also shown---unless you use @code{set
3894 print address off}. The backtrace also shows the source file name and
3895 line number, as well as the arguments to the function. The program
3896 counter value is omitted if it is at the beginning of the code for that
3899 Here is an example of a backtrace. It was made with the command
3900 @samp{bt 3}, so it shows the innermost three frames.
3904 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
3906 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
3907 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
3909 (More stack frames follow...)
3914 The display for frame zero does not begin with a program counter
3915 value, indicating that your program has stopped at the beginning of the
3916 code for line @code{993} of @code{builtin.c}.
3919 @section Selecting a frame
3921 Most commands for examining the stack and other data in your program work on
3922 whichever stack frame is selected at the moment. Here are the commands for
3923 selecting a stack frame; all of them finish by printing a brief description
3924 of the stack frame just selected.
3927 @kindex frame@r{, selecting}
3928 @kindex f @r{(@code{frame})}
3931 Select frame number @var{n}. Recall that frame zero is the innermost
3932 (currently executing) frame, frame one is the frame that called the
3933 innermost one, and so on. The highest-numbered frame is the one for
3936 @item frame @var{addr}
3938 Select the frame at address @var{addr}. This is useful mainly if the
3939 chaining of stack frames has been damaged by a bug, making it
3940 impossible for @value{GDBN} to assign numbers properly to all frames. In
3941 addition, this can be useful when your program has multiple stacks and
3942 switches between them.
3944 On the SPARC architecture, @code{frame} needs two addresses to
3945 select an arbitrary frame: a frame pointer and a stack pointer.
3947 On the MIPS and Alpha architecture, it needs two addresses: a stack
3948 pointer and a program counter.
3950 On the 29k architecture, it needs three addresses: a register stack
3951 pointer, a program counter, and a memory stack pointer.
3952 @c note to future updaters: this is conditioned on a flag
3953 @c SETUP_ARBITRARY_FRAME in the tm-*.h files. The above is up to date
3954 @c as of 27 Jan 1994.
3958 Move @var{n} frames up the stack. For positive numbers @var{n}, this
3959 advances toward the outermost frame, to higher frame numbers, to frames
3960 that have existed longer. @var{n} defaults to one.
3963 @kindex do @r{(@code{down})}
3965 Move @var{n} frames down the stack. For positive numbers @var{n}, this
3966 advances toward the innermost frame, to lower frame numbers, to frames
3967 that were created more recently. @var{n} defaults to one. You may
3968 abbreviate @code{down} as @code{do}.
3971 All of these commands end by printing two lines of output describing the
3972 frame. The first line shows the frame number, the function name, the
3973 arguments, and the source file and line number of execution in that
3974 frame. The second line shows the text of that source line.
3982 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
3984 10 read_input_file (argv[i]);
3988 After such a printout, the @code{list} command with no arguments
3989 prints ten lines centered on the point of execution in the frame.
3990 @xref{List, ,Printing source lines}.
3993 @kindex down-silently
3995 @item up-silently @var{n}
3996 @itemx down-silently @var{n}
3997 These two commands are variants of @code{up} and @code{down},
3998 respectively; they differ in that they do their work silently, without
3999 causing display of the new frame. They are intended primarily for use
4000 in @value{GDBN} command scripts, where the output might be unnecessary and
4005 @section Information about a frame
4007 There are several other commands to print information about the selected
4013 When used without any argument, this command does not change which
4014 frame is selected, but prints a brief description of the currently
4015 selected stack frame. It can be abbreviated @code{f}. With an
4016 argument, this command is used to select a stack frame.
4017 @xref{Selection, ,Selecting a frame}.
4020 @kindex info f @r{(@code{info frame})}
4023 This command prints a verbose description of the selected stack frame,
4028 the address of the frame
4030 the address of the next frame down (called by this frame)
4032 the address of the next frame up (caller of this frame)
4034 the language in which the source code corresponding to this frame is written
4036 the address of the frame's arguments
4038 the address of the frame's local variables
4040 the program counter saved in it (the address of execution in the caller frame)
4042 which registers were saved in the frame
4045 @noindent The verbose description is useful when
4046 something has gone wrong that has made the stack format fail to fit
4047 the usual conventions.
4049 @item info frame @var{addr}
4050 @itemx info f @var{addr}
4051 Print a verbose description of the frame at address @var{addr}, without
4052 selecting that frame. The selected frame remains unchanged by this
4053 command. This requires the same kind of address (more than one for some
4054 architectures) that you specify in the @code{frame} command.
4055 @xref{Selection, ,Selecting a frame}.
4059 Print the arguments of the selected frame, each on a separate line.
4063 Print the local variables of the selected frame, each on a separate
4064 line. These are all variables (declared either static or automatic)
4065 accessible at the point of execution of the selected frame.
4068 @cindex catch exceptions, list active handlers
4069 @cindex exception handlers, how to list
4071 Print a list of all the exception handlers that are active in the
4072 current stack frame at the current point of execution. To see other
4073 exception handlers, visit the associated frame (using the @code{up},
4074 @code{down}, or @code{frame} commands); then type @code{info catch}.
4075 @xref{Set Catchpoints, , Setting catchpoints}.
4081 @chapter Examining Source Files
4083 @value{GDBN} can print parts of your program's source, since the debugging
4084 information recorded in the program tells @value{GDBN} what source files were
4085 used to build it. When your program stops, @value{GDBN} spontaneously prints
4086 the line where it stopped. Likewise, when you select a stack frame
4087 (@pxref{Selection, ,Selecting a frame}), @value{GDBN} prints the line where
4088 execution in that frame has stopped. You can print other portions of
4089 source files by explicit command.
4091 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
4092 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
4093 @value{GDBN} under @sc{gnu} Emacs}.
4096 * List:: Printing source lines
4097 * Search:: Searching source files
4098 * Source Path:: Specifying source directories
4099 * Machine Code:: Source and machine code
4103 @section Printing source lines
4106 @kindex l @r{(@code{list})}
4107 To print lines from a source file, use the @code{list} command
4108 (abbreviated @code{l}). By default, ten lines are printed.
4109 There are several ways to specify what part of the file you want to print.
4111 Here are the forms of the @code{list} command most commonly used:
4114 @item list @var{linenum}
4115 Print lines centered around line number @var{linenum} in the
4116 current source file.
4118 @item list @var{function}
4119 Print lines centered around the beginning of function
4123 Print more lines. If the last lines printed were printed with a
4124 @code{list} command, this prints lines following the last lines
4125 printed; however, if the last line printed was a solitary line printed
4126 as part of displaying a stack frame (@pxref{Stack, ,Examining the
4127 Stack}), this prints lines centered around that line.
4130 Print lines just before the lines last printed.
4133 By default, @value{GDBN} prints ten source lines with any of these forms of
4134 the @code{list} command. You can change this using @code{set listsize}:
4137 @kindex set listsize
4138 @item set listsize @var{count}
4139 Make the @code{list} command display @var{count} source lines (unless
4140 the @code{list} argument explicitly specifies some other number).
4142 @kindex show listsize
4144 Display the number of lines that @code{list} prints.
4147 Repeating a @code{list} command with @key{RET} discards the argument,
4148 so it is equivalent to typing just @code{list}. This is more useful
4149 than listing the same lines again. An exception is made for an
4150 argument of @samp{-}; that argument is preserved in repetition so that
4151 each repetition moves up in the source file.
4154 In general, the @code{list} command expects you to supply zero, one or two
4155 @dfn{linespecs}. Linespecs specify source lines; there are several ways
4156 of writing them, but the effect is always to specify some source line.
4157 Here is a complete description of the possible arguments for @code{list}:
4160 @item list @var{linespec}
4161 Print lines centered around the line specified by @var{linespec}.
4163 @item list @var{first},@var{last}
4164 Print lines from @var{first} to @var{last}. Both arguments are
4167 @item list ,@var{last}
4168 Print lines ending with @var{last}.
4170 @item list @var{first},
4171 Print lines starting with @var{first}.
4174 Print lines just after the lines last printed.
4177 Print lines just before the lines last printed.
4180 As described in the preceding table.
4183 Here are the ways of specifying a single source line---all the
4188 Specifies line @var{number} of the current source file.
4189 When a @code{list} command has two linespecs, this refers to
4190 the same source file as the first linespec.
4193 Specifies the line @var{offset} lines after the last line printed.
4194 When used as the second linespec in a @code{list} command that has
4195 two, this specifies the line @var{offset} lines down from the
4199 Specifies the line @var{offset} lines before the last line printed.
4201 @item @var{filename}:@var{number}
4202 Specifies line @var{number} in the source file @var{filename}.
4204 @item @var{function}
4205 Specifies the line that begins the body of the function @var{function}.
4206 For example: in C, this is the line with the open brace.
4208 @item @var{filename}:@var{function}
4209 Specifies the line of the open-brace that begins the body of the
4210 function @var{function} in the file @var{filename}. You only need the
4211 file name with a function name to avoid ambiguity when there are
4212 identically named functions in different source files.
4214 @item *@var{address}
4215 Specifies the line containing the program address @var{address}.
4216 @var{address} may be any expression.
4220 @section Searching source files
4222 @kindex reverse-search
4224 There are two commands for searching through the current source file for a
4229 @kindex forward-search
4230 @item forward-search @var{regexp}
4231 @itemx search @var{regexp}
4232 The command @samp{forward-search @var{regexp}} checks each line,
4233 starting with the one following the last line listed, for a match for
4234 @var{regexp}. It lists the line that is found. You can use the
4235 synonym @samp{search @var{regexp}} or abbreviate the command name as
4238 @item reverse-search @var{regexp}
4239 The command @samp{reverse-search @var{regexp}} checks each line, starting
4240 with the one before the last line listed and going backward, for a match
4241 for @var{regexp}. It lists the line that is found. You can abbreviate
4242 this command as @code{rev}.
4246 @section Specifying source directories
4249 @cindex directories for source files
4250 Executable programs sometimes do not record the directories of the source
4251 files from which they were compiled, just the names. Even when they do,
4252 the directories could be moved between the compilation and your debugging
4253 session. @value{GDBN} has a list of directories to search for source files;
4254 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
4255 it tries all the directories in the list, in the order they are present
4256 in the list, until it finds a file with the desired name. Note that
4257 the executable search path is @emph{not} used for this purpose. Neither is
4258 the current working directory, unless it happens to be in the source
4261 If @value{GDBN} cannot find a source file in the source path, and the
4262 object program records a directory, @value{GDBN} tries that directory
4263 too. If the source path is empty, and there is no record of the
4264 compilation directory, @value{GDBN} looks in the current directory as a
4267 Whenever you reset or rearrange the source path, @value{GDBN} clears out
4268 any information it has cached about where source files are found and where
4269 each line is in the file.
4273 When you start @value{GDBN}, its source path includes only @samp{cdir}
4274 and @samp{cwd}, in that order.
4275 To add other directories, use the @code{directory} command.
4278 @item directory @var{dirname} @dots{}
4279 @item dir @var{dirname} @dots{}
4280 Add directory @var{dirname} to the front of the source path. Several
4281 directory names may be given to this command, separated by @samp{:}
4282 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
4283 part of absolute file names) or
4284 whitespace. You may specify a directory that is already in the source
4285 path; this moves it forward, so @value{GDBN} searches it sooner.
4289 @vindex $cdir@r{, convenience variable}
4290 @vindex $cwdr@r{, convenience variable}
4291 @cindex compilation directory
4292 @cindex current directory
4293 @cindex working directory
4294 @cindex directory, current
4295 @cindex directory, compilation
4296 You can use the string @samp{$cdir} to refer to the compilation
4297 directory (if one is recorded), and @samp{$cwd} to refer to the current
4298 working directory. @samp{$cwd} is not the same as @samp{.}---the former
4299 tracks the current working directory as it changes during your @value{GDBN}
4300 session, while the latter is immediately expanded to the current
4301 directory at the time you add an entry to the source path.
4304 Reset the source path to empty again. This requires confirmation.
4306 @c RET-repeat for @code{directory} is explicitly disabled, but since
4307 @c repeating it would be a no-op we do not say that. (thanks to RMS)
4309 @item show directories
4310 @kindex show directories
4311 Print the source path: show which directories it contains.
4314 If your source path is cluttered with directories that are no longer of
4315 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
4316 versions of source. You can correct the situation as follows:
4320 Use @code{directory} with no argument to reset the source path to empty.
4323 Use @code{directory} with suitable arguments to reinstall the
4324 directories you want in the source path. You can add all the
4325 directories in one command.
4329 @section Source and machine code
4331 You can use the command @code{info line} to map source lines to program
4332 addresses (and vice versa), and the command @code{disassemble} to display
4333 a range of addresses as machine instructions. When run under @sc{gnu} Emacs
4334 mode, the @code{info line} command causes the arrow to point to the
4335 line specified. Also, @code{info line} prints addresses in symbolic form as
4340 @item info line @var{linespec}
4341 Print the starting and ending addresses of the compiled code for
4342 source line @var{linespec}. You can specify source lines in any of
4343 the ways understood by the @code{list} command (@pxref{List, ,Printing
4347 For example, we can use @code{info line} to discover the location of
4348 the object code for the first line of function
4349 @code{m4_changequote}:
4351 @c FIXME: I think this example should also show the addresses in
4352 @c symbolic form, as they usually would be displayed.
4354 (@value{GDBP}) info line m4_changequote
4355 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
4359 We can also inquire (using @code{*@var{addr}} as the form for
4360 @var{linespec}) what source line covers a particular address:
4362 (@value{GDBP}) info line *0x63ff
4363 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
4366 @cindex @code{$_} and @code{info line}
4367 @kindex x@r{(examine), and} info line
4368 After @code{info line}, the default address for the @code{x} command
4369 is changed to the starting address of the line, so that @samp{x/i} is
4370 sufficient to begin examining the machine code (@pxref{Memory,
4371 ,Examining memory}). Also, this address is saved as the value of the
4372 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
4377 @cindex assembly instructions
4378 @cindex instructions, assembly
4379 @cindex machine instructions
4380 @cindex listing machine instructions
4382 This specialized command dumps a range of memory as machine
4383 instructions. The default memory range is the function surrounding the
4384 program counter of the selected frame. A single argument to this
4385 command is a program counter value; @value{GDBN} dumps the function
4386 surrounding this value. Two arguments specify a range of addresses
4387 (first inclusive, second exclusive) to dump.
4390 The following example shows the disassembly of a range of addresses of
4391 HP PA-RISC 2.0 code:
4394 (@value{GDBP}) disas 0x32c4 0x32e4
4395 Dump of assembler code from 0x32c4 to 0x32e4:
4396 0x32c4 <main+204>: addil 0,dp
4397 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
4398 0x32cc <main+212>: ldil 0x3000,r31
4399 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
4400 0x32d4 <main+220>: ldo 0(r31),rp
4401 0x32d8 <main+224>: addil -0x800,dp
4402 0x32dc <main+228>: ldo 0x588(r1),r26
4403 0x32e0 <main+232>: ldil 0x3000,r31
4404 End of assembler dump.
4407 Some architectures have more than one commonly-used set of instruction
4408 mnemonics or other syntax.
4411 @kindex set disassembly-flavor
4412 @cindex assembly instructions
4413 @cindex instructions, assembly
4414 @cindex machine instructions
4415 @cindex listing machine instructions
4416 @cindex Intel disassembly flavor
4417 @cindex AT&T disassembly flavor
4418 @item set disassembly-flavor @var{instruction-set}
4419 Select the instruction set to use when disassembling the
4420 program via the @code{disassemble} or @code{x/i} commands.
4422 Currently this command is only defined for the Intel x86 family. You
4423 can set @var{instruction-set} to either @code{intel} or @code{att}.
4424 The default is @code{att}, the AT&T flavor used by default by Unix
4425 assemblers for x86-based targets.
4430 @chapter Examining Data
4432 @cindex printing data
4433 @cindex examining data
4436 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
4437 @c document because it is nonstandard... Under Epoch it displays in a
4438 @c different window or something like that.
4439 The usual way to examine data in your program is with the @code{print}
4440 command (abbreviated @code{p}), or its synonym @code{inspect}. It
4441 evaluates and prints the value of an expression of the language your
4442 program is written in (@pxref{Languages, ,Using @value{GDBN} with
4443 Different Languages}).
4446 @item print @var{expr}
4447 @itemx print /@var{f} @var{expr}
4448 @var{expr} is an expression (in the source language). By default the
4449 value of @var{expr} is printed in a format appropriate to its data type;
4450 you can choose a different format by specifying @samp{/@var{f}}, where
4451 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
4455 @itemx print /@var{f}
4456 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
4457 @dfn{value history}; @pxref{Value History, ,Value history}). This allows you to
4458 conveniently inspect the same value in an alternative format.
4461 A more low-level way of examining data is with the @code{x} command.
4462 It examines data in memory at a specified address and prints it in a
4463 specified format. @xref{Memory, ,Examining memory}.
4465 If you are interested in information about types, or about how the
4466 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
4467 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
4471 * Expressions:: Expressions
4472 * Variables:: Program variables
4473 * Arrays:: Artificial arrays
4474 * Output Formats:: Output formats
4475 * Memory:: Examining memory
4476 * Auto Display:: Automatic display
4477 * Print Settings:: Print settings
4478 * Value History:: Value history
4479 * Convenience Vars:: Convenience variables
4480 * Registers:: Registers
4481 * Floating Point Hardware:: Floating point hardware
4482 * Memory Region Attributes:: Memory region attributes
4486 @section Expressions
4489 @code{print} and many other @value{GDBN} commands accept an expression and
4490 compute its value. Any kind of constant, variable or operator defined
4491 by the programming language you are using is valid in an expression in
4492 @value{GDBN}. This includes conditional expressions, function calls, casts
4493 and string constants. It unfortunately does not include symbols defined
4494 by preprocessor @code{#define} commands.
4496 @value{GDBN} supports array constants in expressions input by
4497 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
4498 you can use the command @code{print @{1, 2, 3@}} to build up an array in
4499 memory that is @code{malloc}ed in the target program.
4501 Because C is so widespread, most of the expressions shown in examples in
4502 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
4503 Languages}, for information on how to use expressions in other
4506 In this section, we discuss operators that you can use in @value{GDBN}
4507 expressions regardless of your programming language.
4509 Casts are supported in all languages, not just in C, because it is so
4510 useful to cast a number into a pointer in order to examine a structure
4511 at that address in memory.
4512 @c FIXME: casts supported---Mod2 true?
4514 @value{GDBN} supports these operators, in addition to those common
4515 to programming languages:
4519 @samp{@@} is a binary operator for treating parts of memory as arrays.
4520 @xref{Arrays, ,Artificial arrays}, for more information.
4523 @samp{::} allows you to specify a variable in terms of the file or
4524 function where it is defined. @xref{Variables, ,Program variables}.
4526 @cindex @{@var{type}@}
4527 @cindex type casting memory
4528 @cindex memory, viewing as typed object
4529 @cindex casts, to view memory
4530 @item @{@var{type}@} @var{addr}
4531 Refers to an object of type @var{type} stored at address @var{addr} in
4532 memory. @var{addr} may be any expression whose value is an integer or
4533 pointer (but parentheses are required around binary operators, just as in
4534 a cast). This construct is allowed regardless of what kind of data is
4535 normally supposed to reside at @var{addr}.
4539 @section Program variables
4541 The most common kind of expression to use is the name of a variable
4544 Variables in expressions are understood in the selected stack frame
4545 (@pxref{Selection, ,Selecting a frame}); they must be either:
4549 global (or file-static)
4556 visible according to the scope rules of the
4557 programming language from the point of execution in that frame
4560 @noindent This means that in the function
4575 you can examine and use the variable @code{a} whenever your program is
4576 executing within the function @code{foo}, but you can only use or
4577 examine the variable @code{b} while your program is executing inside
4578 the block where @code{b} is declared.
4580 @cindex variable name conflict
4581 There is an exception: you can refer to a variable or function whose
4582 scope is a single source file even if the current execution point is not
4583 in this file. But it is possible to have more than one such variable or
4584 function with the same name (in different source files). If that
4585 happens, referring to that name has unpredictable effects. If you wish,
4586 you can specify a static variable in a particular function or file,
4587 using the colon-colon notation:
4589 @cindex colon-colon, context for variables/functions
4591 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
4592 @cindex @code{::}, context for variables/functions
4595 @var{file}::@var{variable}
4596 @var{function}::@var{variable}
4600 Here @var{file} or @var{function} is the name of the context for the
4601 static @var{variable}. In the case of file names, you can use quotes to
4602 make sure @value{GDBN} parses the file name as a single word---for example,
4603 to print a global value of @code{x} defined in @file{f2.c}:
4606 (@value{GDBP}) p 'f2.c'::x
4609 @cindex C@t{++} scope resolution
4610 This use of @samp{::} is very rarely in conflict with the very similar
4611 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
4612 scope resolution operator in @value{GDBN} expressions.
4613 @c FIXME: Um, so what happens in one of those rare cases where it's in
4616 @cindex wrong values
4617 @cindex variable values, wrong
4619 @emph{Warning:} Occasionally, a local variable may appear to have the
4620 wrong value at certain points in a function---just after entry to a new
4621 scope, and just before exit.
4623 You may see this problem when you are stepping by machine instructions.
4624 This is because, on most machines, it takes more than one instruction to
4625 set up a stack frame (including local variable definitions); if you are
4626 stepping by machine instructions, variables may appear to have the wrong
4627 values until the stack frame is completely built. On exit, it usually
4628 also takes more than one machine instruction to destroy a stack frame;
4629 after you begin stepping through that group of instructions, local
4630 variable definitions may be gone.
4632 This may also happen when the compiler does significant optimizations.
4633 To be sure of always seeing accurate values, turn off all optimization
4636 @cindex ``No symbol "foo" in current context''
4637 Another possible effect of compiler optimizations is to optimize
4638 unused variables out of existence, or assign variables to registers (as
4639 opposed to memory addresses). Depending on the support for such cases
4640 offered by the debug info format used by the compiler, @value{GDBN}
4641 might not be able to display values for such local variables. If that
4642 happens, @value{GDBN} will print a message like this:
4645 No symbol "foo" in current context.
4648 To solve such problems, either recompile without optimizations, or use a
4649 different debug info format, if the compiler supports several such
4650 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler usually
4651 supports the @samp{-gstabs} option. @samp{-gstabs} produces debug info
4652 in a format that is superior to formats such as COFF. You may be able
4653 to use DWARF2 (@samp{-gdwarf-2}), which is also an effective form for
4654 debug info. See @ref{Debugging Options,,Options for Debugging Your
4655 Program or @sc{gnu} CC, gcc.info, Using @sc{gnu} CC}, for more
4660 @section Artificial arrays
4662 @cindex artificial array
4663 @kindex @@@r{, referencing memory as an array}
4664 It is often useful to print out several successive objects of the
4665 same type in memory; a section of an array, or an array of
4666 dynamically determined size for which only a pointer exists in the
4669 You can do this by referring to a contiguous span of memory as an
4670 @dfn{artificial array}, using the binary operator @samp{@@}. The left
4671 operand of @samp{@@} should be the first element of the desired array
4672 and be an individual object. The right operand should be the desired length
4673 of the array. The result is an array value whose elements are all of
4674 the type of the left argument. The first element is actually the left
4675 argument; the second element comes from bytes of memory immediately
4676 following those that hold the first element, and so on. Here is an
4677 example. If a program says
4680 int *array = (int *) malloc (len * sizeof (int));
4684 you can print the contents of @code{array} with
4690 The left operand of @samp{@@} must reside in memory. Array values made
4691 with @samp{@@} in this way behave just like other arrays in terms of
4692 subscripting, and are coerced to pointers when used in expressions.
4693 Artificial arrays most often appear in expressions via the value history
4694 (@pxref{Value History, ,Value history}), after printing one out.
4696 Another way to create an artificial array is to use a cast.
4697 This re-interprets a value as if it were an array.
4698 The value need not be in memory:
4700 (@value{GDBP}) p/x (short[2])0x12345678
4701 $1 = @{0x1234, 0x5678@}
4704 As a convenience, if you leave the array length out (as in
4705 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
4706 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
4708 (@value{GDBP}) p/x (short[])0x12345678
4709 $2 = @{0x1234, 0x5678@}
4712 Sometimes the artificial array mechanism is not quite enough; in
4713 moderately complex data structures, the elements of interest may not
4714 actually be adjacent---for example, if you are interested in the values
4715 of pointers in an array. One useful work-around in this situation is
4716 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
4717 variables}) as a counter in an expression that prints the first
4718 interesting value, and then repeat that expression via @key{RET}. For
4719 instance, suppose you have an array @code{dtab} of pointers to
4720 structures, and you are interested in the values of a field @code{fv}
4721 in each structure. Here is an example of what you might type:
4731 @node Output Formats
4732 @section Output formats
4734 @cindex formatted output
4735 @cindex output formats
4736 By default, @value{GDBN} prints a value according to its data type. Sometimes
4737 this is not what you want. For example, you might want to print a number
4738 in hex, or a pointer in decimal. Or you might want to view data in memory
4739 at a certain address as a character string or as an instruction. To do
4740 these things, specify an @dfn{output format} when you print a value.
4742 The simplest use of output formats is to say how to print a value
4743 already computed. This is done by starting the arguments of the
4744 @code{print} command with a slash and a format letter. The format
4745 letters supported are:
4749 Regard the bits of the value as an integer, and print the integer in
4753 Print as integer in signed decimal.
4756 Print as integer in unsigned decimal.
4759 Print as integer in octal.
4762 Print as integer in binary. The letter @samp{t} stands for ``two''.
4763 @footnote{@samp{b} cannot be used because these format letters are also
4764 used with the @code{x} command, where @samp{b} stands for ``byte'';
4765 see @ref{Memory,,Examining memory}.}
4768 @cindex unknown address, locating
4769 @cindex locate address
4770 Print as an address, both absolute in hexadecimal and as an offset from
4771 the nearest preceding symbol. You can use this format used to discover
4772 where (in what function) an unknown address is located:
4775 (@value{GDBP}) p/a 0x54320
4776 $3 = 0x54320 <_initialize_vx+396>
4780 The command @code{info symbol 0x54320} yields similar results.
4781 @xref{Symbols, info symbol}.
4784 Regard as an integer and print it as a character constant.
4787 Regard the bits of the value as a floating point number and print
4788 using typical floating point syntax.
4791 For example, to print the program counter in hex (@pxref{Registers}), type
4798 Note that no space is required before the slash; this is because command
4799 names in @value{GDBN} cannot contain a slash.
4801 To reprint the last value in the value history with a different format,
4802 you can use the @code{print} command with just a format and no
4803 expression. For example, @samp{p/x} reprints the last value in hex.
4806 @section Examining memory
4808 You can use the command @code{x} (for ``examine'') to examine memory in
4809 any of several formats, independently of your program's data types.
4811 @cindex examining memory
4813 @kindex x @r{(examine memory)}
4814 @item x/@var{nfu} @var{addr}
4817 Use the @code{x} command to examine memory.
4820 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
4821 much memory to display and how to format it; @var{addr} is an
4822 expression giving the address where you want to start displaying memory.
4823 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
4824 Several commands set convenient defaults for @var{addr}.
4827 @item @var{n}, the repeat count
4828 The repeat count is a decimal integer; the default is 1. It specifies
4829 how much memory (counting by units @var{u}) to display.
4830 @c This really is **decimal**; unaffected by 'set radix' as of GDB
4833 @item @var{f}, the display format
4834 The display format is one of the formats used by @code{print},
4835 @samp{s} (null-terminated string), or @samp{i} (machine instruction).
4836 The default is @samp{x} (hexadecimal) initially.
4837 The default changes each time you use either @code{x} or @code{print}.
4839 @item @var{u}, the unit size
4840 The unit size is any of
4846 Halfwords (two bytes).
4848 Words (four bytes). This is the initial default.
4850 Giant words (eight bytes).
4853 Each time you specify a unit size with @code{x}, that size becomes the
4854 default unit the next time you use @code{x}. (For the @samp{s} and
4855 @samp{i} formats, the unit size is ignored and is normally not written.)
4857 @item @var{addr}, starting display address
4858 @var{addr} is the address where you want @value{GDBN} to begin displaying
4859 memory. The expression need not have a pointer value (though it may);
4860 it is always interpreted as an integer address of a byte of memory.
4861 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
4862 @var{addr} is usually just after the last address examined---but several
4863 other commands also set the default address: @code{info breakpoints} (to
4864 the address of the last breakpoint listed), @code{info line} (to the
4865 starting address of a line), and @code{print} (if you use it to display
4866 a value from memory).
4869 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
4870 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
4871 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
4872 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
4873 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
4875 Since the letters indicating unit sizes are all distinct from the
4876 letters specifying output formats, you do not have to remember whether
4877 unit size or format comes first; either order works. The output
4878 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
4879 (However, the count @var{n} must come first; @samp{wx4} does not work.)
4881 Even though the unit size @var{u} is ignored for the formats @samp{s}
4882 and @samp{i}, you might still want to use a count @var{n}; for example,
4883 @samp{3i} specifies that you want to see three machine instructions,
4884 including any operands. The command @code{disassemble} gives an
4885 alternative way of inspecting machine instructions; see @ref{Machine
4886 Code,,Source and machine code}.
4888 All the defaults for the arguments to @code{x} are designed to make it
4889 easy to continue scanning memory with minimal specifications each time
4890 you use @code{x}. For example, after you have inspected three machine
4891 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
4892 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
4893 the repeat count @var{n} is used again; the other arguments default as
4894 for successive uses of @code{x}.
4896 @cindex @code{$_}, @code{$__}, and value history
4897 The addresses and contents printed by the @code{x} command are not saved
4898 in the value history because there is often too much of them and they
4899 would get in the way. Instead, @value{GDBN} makes these values available for
4900 subsequent use in expressions as values of the convenience variables
4901 @code{$_} and @code{$__}. After an @code{x} command, the last address
4902 examined is available for use in expressions in the convenience variable
4903 @code{$_}. The contents of that address, as examined, are available in
4904 the convenience variable @code{$__}.
4906 If the @code{x} command has a repeat count, the address and contents saved
4907 are from the last memory unit printed; this is not the same as the last
4908 address printed if several units were printed on the last line of output.
4911 @section Automatic display
4912 @cindex automatic display
4913 @cindex display of expressions
4915 If you find that you want to print the value of an expression frequently
4916 (to see how it changes), you might want to add it to the @dfn{automatic
4917 display list} so that @value{GDBN} prints its value each time your program stops.
4918 Each expression added to the list is given a number to identify it;
4919 to remove an expression from the list, you specify that number.
4920 The automatic display looks like this:
4924 3: bar[5] = (struct hack *) 0x3804
4928 This display shows item numbers, expressions and their current values. As with
4929 displays you request manually using @code{x} or @code{print}, you can
4930 specify the output format you prefer; in fact, @code{display} decides
4931 whether to use @code{print} or @code{x} depending on how elaborate your
4932 format specification is---it uses @code{x} if you specify a unit size,
4933 or one of the two formats (@samp{i} and @samp{s}) that are only
4934 supported by @code{x}; otherwise it uses @code{print}.
4938 @item display @var{expr}
4939 Add the expression @var{expr} to the list of expressions to display
4940 each time your program stops. @xref{Expressions, ,Expressions}.
4942 @code{display} does not repeat if you press @key{RET} again after using it.
4944 @item display/@var{fmt} @var{expr}
4945 For @var{fmt} specifying only a display format and not a size or
4946 count, add the expression @var{expr} to the auto-display list but
4947 arrange to display it each time in the specified format @var{fmt}.
4948 @xref{Output Formats,,Output formats}.
4950 @item display/@var{fmt} @var{addr}
4951 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
4952 number of units, add the expression @var{addr} as a memory address to
4953 be examined each time your program stops. Examining means in effect
4954 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining memory}.
4957 For example, @samp{display/i $pc} can be helpful, to see the machine
4958 instruction about to be executed each time execution stops (@samp{$pc}
4959 is a common name for the program counter; @pxref{Registers, ,Registers}).
4962 @kindex delete display
4964 @item undisplay @var{dnums}@dots{}
4965 @itemx delete display @var{dnums}@dots{}
4966 Remove item numbers @var{dnums} from the list of expressions to display.
4968 @code{undisplay} does not repeat if you press @key{RET} after using it.
4969 (Otherwise you would just get the error @samp{No display number @dots{}}.)
4971 @kindex disable display
4972 @item disable display @var{dnums}@dots{}
4973 Disable the display of item numbers @var{dnums}. A disabled display
4974 item is not printed automatically, but is not forgotten. It may be
4975 enabled again later.
4977 @kindex enable display
4978 @item enable display @var{dnums}@dots{}
4979 Enable display of item numbers @var{dnums}. It becomes effective once
4980 again in auto display of its expression, until you specify otherwise.
4983 Display the current values of the expressions on the list, just as is
4984 done when your program stops.
4986 @kindex info display
4988 Print the list of expressions previously set up to display
4989 automatically, each one with its item number, but without showing the
4990 values. This includes disabled expressions, which are marked as such.
4991 It also includes expressions which would not be displayed right now
4992 because they refer to automatic variables not currently available.
4995 If a display expression refers to local variables, then it does not make
4996 sense outside the lexical context for which it was set up. Such an
4997 expression is disabled when execution enters a context where one of its
4998 variables is not defined. For example, if you give the command
4999 @code{display last_char} while inside a function with an argument
5000 @code{last_char}, @value{GDBN} displays this argument while your program
5001 continues to stop inside that function. When it stops elsewhere---where
5002 there is no variable @code{last_char}---the display is disabled
5003 automatically. The next time your program stops where @code{last_char}
5004 is meaningful, you can enable the display expression once again.
5006 @node Print Settings
5007 @section Print settings
5009 @cindex format options
5010 @cindex print settings
5011 @value{GDBN} provides the following ways to control how arrays, structures,
5012 and symbols are printed.
5015 These settings are useful for debugging programs in any language:
5018 @kindex set print address
5019 @item set print address
5020 @itemx set print address on
5021 @value{GDBN} prints memory addresses showing the location of stack
5022 traces, structure values, pointer values, breakpoints, and so forth,
5023 even when it also displays the contents of those addresses. The default
5024 is @code{on}. For example, this is what a stack frame display looks like with
5025 @code{set print address on}:
5030 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
5032 530 if (lquote != def_lquote)
5036 @item set print address off
5037 Do not print addresses when displaying their contents. For example,
5038 this is the same stack frame displayed with @code{set print address off}:
5042 (@value{GDBP}) set print addr off
5044 #0 set_quotes (lq="<<", rq=">>") at input.c:530
5045 530 if (lquote != def_lquote)
5049 You can use @samp{set print address off} to eliminate all machine
5050 dependent displays from the @value{GDBN} interface. For example, with
5051 @code{print address off}, you should get the same text for backtraces on
5052 all machines---whether or not they involve pointer arguments.
5054 @kindex show print address
5055 @item show print address
5056 Show whether or not addresses are to be printed.
5059 When @value{GDBN} prints a symbolic address, it normally prints the
5060 closest earlier symbol plus an offset. If that symbol does not uniquely
5061 identify the address (for example, it is a name whose scope is a single
5062 source file), you may need to clarify. One way to do this is with
5063 @code{info line}, for example @samp{info line *0x4537}. Alternately,
5064 you can set @value{GDBN} to print the source file and line number when
5065 it prints a symbolic address:
5068 @kindex set print symbol-filename
5069 @item set print symbol-filename on
5070 Tell @value{GDBN} to print the source file name and line number of a
5071 symbol in the symbolic form of an address.
5073 @item set print symbol-filename off
5074 Do not print source file name and line number of a symbol. This is the
5077 @kindex show print symbol-filename
5078 @item show print symbol-filename
5079 Show whether or not @value{GDBN} will print the source file name and
5080 line number of a symbol in the symbolic form of an address.
5083 Another situation where it is helpful to show symbol filenames and line
5084 numbers is when disassembling code; @value{GDBN} shows you the line
5085 number and source file that corresponds to each instruction.
5087 Also, you may wish to see the symbolic form only if the address being
5088 printed is reasonably close to the closest earlier symbol:
5091 @kindex set print max-symbolic-offset
5092 @item set print max-symbolic-offset @var{max-offset}
5093 Tell @value{GDBN} to only display the symbolic form of an address if the
5094 offset between the closest earlier symbol and the address is less than
5095 @var{max-offset}. The default is 0, which tells @value{GDBN}
5096 to always print the symbolic form of an address if any symbol precedes it.
5098 @kindex show print max-symbolic-offset
5099 @item show print max-symbolic-offset
5100 Ask how large the maximum offset is that @value{GDBN} prints in a
5104 @cindex wild pointer, interpreting
5105 @cindex pointer, finding referent
5106 If you have a pointer and you are not sure where it points, try
5107 @samp{set print symbol-filename on}. Then you can determine the name
5108 and source file location of the variable where it points, using
5109 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
5110 For example, here @value{GDBN} shows that a variable @code{ptt} points
5111 at another variable @code{t}, defined in @file{hi2.c}:
5114 (@value{GDBP}) set print symbol-filename on
5115 (@value{GDBP}) p/a ptt
5116 $4 = 0xe008 <t in hi2.c>
5120 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
5121 does not show the symbol name and filename of the referent, even with
5122 the appropriate @code{set print} options turned on.
5125 Other settings control how different kinds of objects are printed:
5128 @kindex set print array
5129 @item set print array
5130 @itemx set print array on
5131 Pretty print arrays. This format is more convenient to read,
5132 but uses more space. The default is off.
5134 @item set print array off
5135 Return to compressed format for arrays.
5137 @kindex show print array
5138 @item show print array
5139 Show whether compressed or pretty format is selected for displaying
5142 @kindex set print elements
5143 @item set print elements @var{number-of-elements}
5144 Set a limit on how many elements of an array @value{GDBN} will print.
5145 If @value{GDBN} is printing a large array, it stops printing after it has
5146 printed the number of elements set by the @code{set print elements} command.
5147 This limit also applies to the display of strings.
5148 When @value{GDBN} starts, this limit is set to 200.
5149 Setting @var{number-of-elements} to zero means that the printing is unlimited.
5151 @kindex show print elements
5152 @item show print elements
5153 Display the number of elements of a large array that @value{GDBN} will print.
5154 If the number is 0, then the printing is unlimited.
5156 @kindex set print null-stop
5157 @item set print null-stop
5158 Cause @value{GDBN} to stop printing the characters of an array when the first
5159 @sc{null} is encountered. This is useful when large arrays actually
5160 contain only short strings.
5163 @kindex set print pretty
5164 @item set print pretty on
5165 Cause @value{GDBN} to print structures in an indented format with one member
5166 per line, like this:
5181 @item set print pretty off
5182 Cause @value{GDBN} to print structures in a compact format, like this:
5186 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
5187 meat = 0x54 "Pork"@}
5192 This is the default format.
5194 @kindex show print pretty
5195 @item show print pretty
5196 Show which format @value{GDBN} is using to print structures.
5198 @kindex set print sevenbit-strings
5199 @item set print sevenbit-strings on
5200 Print using only seven-bit characters; if this option is set,
5201 @value{GDBN} displays any eight-bit characters (in strings or
5202 character values) using the notation @code{\}@var{nnn}. This setting is
5203 best if you are working in English (@sc{ascii}) and you use the
5204 high-order bit of characters as a marker or ``meta'' bit.
5206 @item set print sevenbit-strings off
5207 Print full eight-bit characters. This allows the use of more
5208 international character sets, and is the default.
5210 @kindex show print sevenbit-strings
5211 @item show print sevenbit-strings
5212 Show whether or not @value{GDBN} is printing only seven-bit characters.
5214 @kindex set print union
5215 @item set print union on
5216 Tell @value{GDBN} to print unions which are contained in structures. This
5217 is the default setting.
5219 @item set print union off
5220 Tell @value{GDBN} not to print unions which are contained in structures.
5222 @kindex show print union
5223 @item show print union
5224 Ask @value{GDBN} whether or not it will print unions which are contained in
5227 For example, given the declarations
5230 typedef enum @{Tree, Bug@} Species;
5231 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
5232 typedef enum @{Caterpillar, Cocoon, Butterfly@}
5243 struct thing foo = @{Tree, @{Acorn@}@};
5247 with @code{set print union on} in effect @samp{p foo} would print
5250 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
5254 and with @code{set print union off} in effect it would print
5257 $1 = @{it = Tree, form = @{...@}@}
5263 These settings are of interest when debugging C@t{++} programs:
5267 @kindex set print demangle
5268 @item set print demangle
5269 @itemx set print demangle on
5270 Print C@t{++} names in their source form rather than in the encoded
5271 (``mangled'') form passed to the assembler and linker for type-safe
5272 linkage. The default is on.
5274 @kindex show print demangle
5275 @item show print demangle
5276 Show whether C@t{++} names are printed in mangled or demangled form.
5278 @kindex set print asm-demangle
5279 @item set print asm-demangle
5280 @itemx set print asm-demangle on
5281 Print C@t{++} names in their source form rather than their mangled form, even
5282 in assembler code printouts such as instruction disassemblies.
5285 @kindex show print asm-demangle
5286 @item show print asm-demangle
5287 Show whether C@t{++} names in assembly listings are printed in mangled
5290 @kindex set demangle-style
5291 @cindex C@t{++} symbol decoding style
5292 @cindex symbol decoding style, C@t{++}
5293 @item set demangle-style @var{style}
5294 Choose among several encoding schemes used by different compilers to
5295 represent C@t{++} names. The choices for @var{style} are currently:
5299 Allow @value{GDBN} to choose a decoding style by inspecting your program.
5302 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
5303 This is the default.
5306 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
5309 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
5312 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
5313 @strong{Warning:} this setting alone is not sufficient to allow
5314 debugging @code{cfront}-generated executables. @value{GDBN} would
5315 require further enhancement to permit that.
5318 If you omit @var{style}, you will see a list of possible formats.
5320 @kindex show demangle-style
5321 @item show demangle-style
5322 Display the encoding style currently in use for decoding C@t{++} symbols.
5324 @kindex set print object
5325 @item set print object
5326 @itemx set print object on
5327 When displaying a pointer to an object, identify the @emph{actual}
5328 (derived) type of the object rather than the @emph{declared} type, using
5329 the virtual function table.
5331 @item set print object off
5332 Display only the declared type of objects, without reference to the
5333 virtual function table. This is the default setting.
5335 @kindex show print object
5336 @item show print object
5337 Show whether actual, or declared, object types are displayed.
5339 @kindex set print static-members
5340 @item set print static-members
5341 @itemx set print static-members on
5342 Print static members when displaying a C@t{++} object. The default is on.
5344 @item set print static-members off
5345 Do not print static members when displaying a C@t{++} object.
5347 @kindex show print static-members
5348 @item show print static-members
5349 Show whether C@t{++} static members are printed, or not.
5351 @c These don't work with HP ANSI C++ yet.
5352 @kindex set print vtbl
5353 @item set print vtbl
5354 @itemx set print vtbl on
5355 Pretty print C@t{++} virtual function tables. The default is off.
5356 (The @code{vtbl} commands do not work on programs compiled with the HP
5357 ANSI C@t{++} compiler (@code{aCC}).)
5359 @item set print vtbl off
5360 Do not pretty print C@t{++} virtual function tables.
5362 @kindex show print vtbl
5363 @item show print vtbl
5364 Show whether C@t{++} virtual function tables are pretty printed, or not.
5368 @section Value history
5370 @cindex value history
5371 Values printed by the @code{print} command are saved in the @value{GDBN}
5372 @dfn{value history}. This allows you to refer to them in other expressions.
5373 Values are kept until the symbol table is re-read or discarded
5374 (for example with the @code{file} or @code{symbol-file} commands).
5375 When the symbol table changes, the value history is discarded,
5376 since the values may contain pointers back to the types defined in the
5381 @cindex history number
5382 The values printed are given @dfn{history numbers} by which you can
5383 refer to them. These are successive integers starting with one.
5384 @code{print} shows you the history number assigned to a value by
5385 printing @samp{$@var{num} = } before the value; here @var{num} is the
5388 To refer to any previous value, use @samp{$} followed by the value's
5389 history number. The way @code{print} labels its output is designed to
5390 remind you of this. Just @code{$} refers to the most recent value in
5391 the history, and @code{$$} refers to the value before that.
5392 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
5393 is the value just prior to @code{$$}, @code{$$1} is equivalent to
5394 @code{$$}, and @code{$$0} is equivalent to @code{$}.
5396 For example, suppose you have just printed a pointer to a structure and
5397 want to see the contents of the structure. It suffices to type
5403 If you have a chain of structures where the component @code{next} points
5404 to the next one, you can print the contents of the next one with this:
5411 You can print successive links in the chain by repeating this
5412 command---which you can do by just typing @key{RET}.
5414 Note that the history records values, not expressions. If the value of
5415 @code{x} is 4 and you type these commands:
5423 then the value recorded in the value history by the @code{print} command
5424 remains 4 even though the value of @code{x} has changed.
5429 Print the last ten values in the value history, with their item numbers.
5430 This is like @samp{p@ $$9} repeated ten times, except that @code{show
5431 values} does not change the history.
5433 @item show values @var{n}
5434 Print ten history values centered on history item number @var{n}.
5437 Print ten history values just after the values last printed. If no more
5438 values are available, @code{show values +} produces no display.
5441 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
5442 same effect as @samp{show values +}.
5444 @node Convenience Vars
5445 @section Convenience variables
5447 @cindex convenience variables
5448 @value{GDBN} provides @dfn{convenience variables} that you can use within
5449 @value{GDBN} to hold on to a value and refer to it later. These variables
5450 exist entirely within @value{GDBN}; they are not part of your program, and
5451 setting a convenience variable has no direct effect on further execution
5452 of your program. That is why you can use them freely.
5454 Convenience variables are prefixed with @samp{$}. Any name preceded by
5455 @samp{$} can be used for a convenience variable, unless it is one of
5456 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
5457 (Value history references, in contrast, are @emph{numbers} preceded
5458 by @samp{$}. @xref{Value History, ,Value history}.)
5460 You can save a value in a convenience variable with an assignment
5461 expression, just as you would set a variable in your program.
5465 set $foo = *object_ptr
5469 would save in @code{$foo} the value contained in the object pointed to by
5472 Using a convenience variable for the first time creates it, but its
5473 value is @code{void} until you assign a new value. You can alter the
5474 value with another assignment at any time.
5476 Convenience variables have no fixed types. You can assign a convenience
5477 variable any type of value, including structures and arrays, even if
5478 that variable already has a value of a different type. The convenience
5479 variable, when used as an expression, has the type of its current value.
5482 @kindex show convenience
5483 @item show convenience
5484 Print a list of convenience variables used so far, and their values.
5485 Abbreviated @code{show conv}.
5488 One of the ways to use a convenience variable is as a counter to be
5489 incremented or a pointer to be advanced. For example, to print
5490 a field from successive elements of an array of structures:
5494 print bar[$i++]->contents
5498 Repeat that command by typing @key{RET}.
5500 Some convenience variables are created automatically by @value{GDBN} and given
5501 values likely to be useful.
5504 @vindex $_@r{, convenience variable}
5506 The variable @code{$_} is automatically set by the @code{x} command to
5507 the last address examined (@pxref{Memory, ,Examining memory}). Other
5508 commands which provide a default address for @code{x} to examine also
5509 set @code{$_} to that address; these commands include @code{info line}
5510 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
5511 except when set by the @code{x} command, in which case it is a pointer
5512 to the type of @code{$__}.
5514 @vindex $__@r{, convenience variable}
5516 The variable @code{$__} is automatically set by the @code{x} command
5517 to the value found in the last address examined. Its type is chosen
5518 to match the format in which the data was printed.
5521 @vindex $_exitcode@r{, convenience variable}
5522 The variable @code{$_exitcode} is automatically set to the exit code when
5523 the program being debugged terminates.
5526 On HP-UX systems, if you refer to a function or variable name that
5527 begins with a dollar sign, @value{GDBN} searches for a user or system
5528 name first, before it searches for a convenience variable.
5534 You can refer to machine register contents, in expressions, as variables
5535 with names starting with @samp{$}. The names of registers are different
5536 for each machine; use @code{info registers} to see the names used on
5540 @kindex info registers
5541 @item info registers
5542 Print the names and values of all registers except floating-point
5543 registers (in the selected stack frame).
5545 @kindex info all-registers
5546 @cindex floating point registers
5547 @item info all-registers
5548 Print the names and values of all registers, including floating-point
5551 @item info registers @var{regname} @dots{}
5552 Print the @dfn{relativized} value of each specified register @var{regname}.
5553 As discussed in detail below, register values are normally relative to
5554 the selected stack frame. @var{regname} may be any register name valid on
5555 the machine you are using, with or without the initial @samp{$}.
5558 @value{GDBN} has four ``standard'' register names that are available (in
5559 expressions) on most machines---whenever they do not conflict with an
5560 architecture's canonical mnemonics for registers. The register names
5561 @code{$pc} and @code{$sp} are used for the program counter register and
5562 the stack pointer. @code{$fp} is used for a register that contains a
5563 pointer to the current stack frame, and @code{$ps} is used for a
5564 register that contains the processor status. For example,
5565 you could print the program counter in hex with
5572 or print the instruction to be executed next with
5579 or add four to the stack pointer@footnote{This is a way of removing
5580 one word from the stack, on machines where stacks grow downward in
5581 memory (most machines, nowadays). This assumes that the innermost
5582 stack frame is selected; setting @code{$sp} is not allowed when other
5583 stack frames are selected. To pop entire frames off the stack,
5584 regardless of machine architecture, use @code{return};
5585 see @ref{Returning, ,Returning from a function}.} with
5591 Whenever possible, these four standard register names are available on
5592 your machine even though the machine has different canonical mnemonics,
5593 so long as there is no conflict. The @code{info registers} command
5594 shows the canonical names. For example, on the SPARC, @code{info
5595 registers} displays the processor status register as @code{$psr} but you
5596 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
5597 is an alias for the @sc{eflags} register.
5599 @value{GDBN} always considers the contents of an ordinary register as an
5600 integer when the register is examined in this way. Some machines have
5601 special registers which can hold nothing but floating point; these
5602 registers are considered to have floating point values. There is no way
5603 to refer to the contents of an ordinary register as floating point value
5604 (although you can @emph{print} it as a floating point value with
5605 @samp{print/f $@var{regname}}).
5607 Some registers have distinct ``raw'' and ``virtual'' data formats. This
5608 means that the data format in which the register contents are saved by
5609 the operating system is not the same one that your program normally
5610 sees. For example, the registers of the 68881 floating point
5611 coprocessor are always saved in ``extended'' (raw) format, but all C
5612 programs expect to work with ``double'' (virtual) format. In such
5613 cases, @value{GDBN} normally works with the virtual format only (the format
5614 that makes sense for your program), but the @code{info registers} command
5615 prints the data in both formats.
5617 Normally, register values are relative to the selected stack frame
5618 (@pxref{Selection, ,Selecting a frame}). This means that you get the
5619 value that the register would contain if all stack frames farther in
5620 were exited and their saved registers restored. In order to see the
5621 true contents of hardware registers, you must select the innermost
5622 frame (with @samp{frame 0}).
5624 However, @value{GDBN} must deduce where registers are saved, from the machine
5625 code generated by your compiler. If some registers are not saved, or if
5626 @value{GDBN} is unable to locate the saved registers, the selected stack
5627 frame makes no difference.
5629 @node Floating Point Hardware
5630 @section Floating point hardware
5631 @cindex floating point
5633 Depending on the configuration, @value{GDBN} may be able to give
5634 you more information about the status of the floating point hardware.
5639 Display hardware-dependent information about the floating
5640 point unit. The exact contents and layout vary depending on the
5641 floating point chip. Currently, @samp{info float} is supported on
5642 the ARM and x86 machines.
5645 @node Memory Region Attributes
5646 @section Memory Region Attributes
5647 @cindex memory region attributes
5649 @dfn{Memory region attributes} allow you to describe special handling
5650 required by regions of your target's memory. @value{GDBN} uses attributes
5651 to determine whether to allow certain types of memory accesses; whether to
5652 use specific width accesses; and whether to cache target memory.
5654 Defined memory regions can be individually enabled and disabled. When a
5655 memory region is disabled, @value{GDBN} uses the default attributes when
5656 accessing memory in that region. Similarly, if no memory regions have
5657 been defined, @value{GDBN} uses the default attributes when accessing
5660 When a memory region is defined, it is given a number to identify it;
5661 to enable, disable, or remove a memory region, you specify that number.
5665 @item mem @var{address1} @var{address1} @var{attributes}@dots{}
5666 Define memory region bounded by @var{address1} and @var{address2}
5667 with attributes @var{attributes}@dots{}.
5670 @item delete mem @var{nums}@dots{}
5671 Remove memory region numbers @var{nums}.
5674 @item disable mem @var{nums}@dots{}
5675 Disable memory region numbers @var{nums}.
5676 A disabled memory region is not forgotten.
5677 It may be enabled again later.
5680 @item enable mem @var{nums}@dots{}
5681 Enable memory region numbers @var{nums}.
5685 Print a table of all defined memory regions, with the following columns
5689 @item Memory Region Number
5690 @item Enabled or Disabled.
5691 Enabled memory regions are marked with @samp{y}.
5692 Disabled memory regions are marked with @samp{n}.
5695 The address defining the inclusive lower bound of the memory region.
5698 The address defining the exclusive upper bound of the memory region.
5701 The list of attributes set for this memory region.
5706 @subsection Attributes
5708 @subsubsection Memory Access Mode
5709 The access mode attributes set whether @value{GDBN} may make read or
5710 write accesses to a memory region.
5712 While these attributes prevent @value{GDBN} from performing invalid
5713 memory accesses, they do nothing to prevent the target system, I/O DMA,
5714 etc. from accessing memory.
5718 Memory is read only.
5720 Memory is write only.
5722 Memory is read/write (default).
5725 @subsubsection Memory Access Size
5726 The acccess size attributes tells @value{GDBN} to use specific sized
5727 accesses in the memory region. Often memory mapped device registers
5728 require specific sized accesses. If no access size attribute is
5729 specified, @value{GDBN} may use accesses of any size.
5733 Use 8 bit memory accesses.
5735 Use 16 bit memory accesses.
5737 Use 32 bit memory accesses.
5739 Use 64 bit memory accesses.
5742 @c @subsubsection Hardware/Software Breakpoints
5743 @c The hardware/software breakpoint attributes set whether @value{GDBN}
5744 @c will use hardware or software breakpoints for the internal breakpoints
5745 @c used by the step, next, finish, until, etc. commands.
5749 @c Always use hardware breakpoints
5750 @c @item swbreak (default)
5753 @subsubsection Data Cache
5754 The data cache attributes set whether @value{GDBN} will cache target
5755 memory. While this generally improves performance by reducing debug
5756 protocol overhead, it can lead to incorrect results because @value{GDBN}
5757 does not know about volatile variables or memory mapped device
5762 Enable @value{GDBN} to cache target memory.
5763 @item nocache (default)
5764 Disable @value{GDBN} from caching target memory.
5767 @c @subsubsection Memory Write Verification
5768 @c The memory write verification attributes set whether @value{GDBN}
5769 @c will re-reads data after each write to verify the write was successful.
5773 @c @item noverify (default)
5777 @chapter Tracepoints
5778 @c This chapter is based on the documentation written by Michael
5779 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
5782 In some applications, it is not feasible for the debugger to interrupt
5783 the program's execution long enough for the developer to learn
5784 anything helpful about its behavior. If the program's correctness
5785 depends on its real-time behavior, delays introduced by a debugger
5786 might cause the program to change its behavior drastically, or perhaps
5787 fail, even when the code itself is correct. It is useful to be able
5788 to observe the program's behavior without interrupting it.
5790 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
5791 specify locations in the program, called @dfn{tracepoints}, and
5792 arbitrary expressions to evaluate when those tracepoints are reached.
5793 Later, using the @code{tfind} command, you can examine the values
5794 those expressions had when the program hit the tracepoints. The
5795 expressions may also denote objects in memory---structures or arrays,
5796 for example---whose values @value{GDBN} should record; while visiting
5797 a particular tracepoint, you may inspect those objects as if they were
5798 in memory at that moment. However, because @value{GDBN} records these
5799 values without interacting with you, it can do so quickly and
5800 unobtrusively, hopefully not disturbing the program's behavior.
5802 The tracepoint facility is currently available only for remote
5803 targets. @xref{Targets}.
5805 This chapter describes the tracepoint commands and features.
5809 * Analyze Collected Data::
5810 * Tracepoint Variables::
5813 @node Set Tracepoints
5814 @section Commands to Set Tracepoints
5816 Before running such a @dfn{trace experiment}, an arbitrary number of
5817 tracepoints can be set. Like a breakpoint (@pxref{Set Breaks}), a
5818 tracepoint has a number assigned to it by @value{GDBN}. Like with
5819 breakpoints, tracepoint numbers are successive integers starting from
5820 one. Many of the commands associated with tracepoints take the
5821 tracepoint number as their argument, to identify which tracepoint to
5824 For each tracepoint, you can specify, in advance, some arbitrary set
5825 of data that you want the target to collect in the trace buffer when
5826 it hits that tracepoint. The collected data can include registers,
5827 local variables, or global data. Later, you can use @value{GDBN}
5828 commands to examine the values these data had at the time the
5831 This section describes commands to set tracepoints and associated
5832 conditions and actions.
5835 * Create and Delete Tracepoints::
5836 * Enable and Disable Tracepoints::
5837 * Tracepoint Passcounts::
5838 * Tracepoint Actions::
5839 * Listing Tracepoints::
5840 * Starting and Stopping Trace Experiment::
5843 @node Create and Delete Tracepoints
5844 @subsection Create and Delete Tracepoints
5847 @cindex set tracepoint
5850 The @code{trace} command is very similar to the @code{break} command.
5851 Its argument can be a source line, a function name, or an address in
5852 the target program. @xref{Set Breaks}. The @code{trace} command
5853 defines a tracepoint, which is a point in the target program where the
5854 debugger will briefly stop, collect some data, and then allow the
5855 program to continue. Setting a tracepoint or changing its commands
5856 doesn't take effect until the next @code{tstart} command; thus, you
5857 cannot change the tracepoint attributes once a trace experiment is
5860 Here are some examples of using the @code{trace} command:
5863 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
5865 (@value{GDBP}) @b{trace +2} // 2 lines forward
5867 (@value{GDBP}) @b{trace my_function} // first source line of function
5869 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
5871 (@value{GDBP}) @b{trace *0x2117c4} // an address
5875 You can abbreviate @code{trace} as @code{tr}.
5878 @cindex last tracepoint number
5879 @cindex recent tracepoint number
5880 @cindex tracepoint number
5881 The convenience variable @code{$tpnum} records the tracepoint number
5882 of the most recently set tracepoint.
5884 @kindex delete tracepoint
5885 @cindex tracepoint deletion
5886 @item delete tracepoint @r{[}@var{num}@r{]}
5887 Permanently delete one or more tracepoints. With no argument, the
5888 default is to delete all tracepoints.
5893 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
5895 (@value{GDBP}) @b{delete trace} // remove all tracepoints
5899 You can abbreviate this command as @code{del tr}.
5902 @node Enable and Disable Tracepoints
5903 @subsection Enable and Disable Tracepoints
5906 @kindex disable tracepoint
5907 @item disable tracepoint @r{[}@var{num}@r{]}
5908 Disable tracepoint @var{num}, or all tracepoints if no argument
5909 @var{num} is given. A disabled tracepoint will have no effect during
5910 the next trace experiment, but it is not forgotten. You can re-enable
5911 a disabled tracepoint using the @code{enable tracepoint} command.
5913 @kindex enable tracepoint
5914 @item enable tracepoint @r{[}@var{num}@r{]}
5915 Enable tracepoint @var{num}, or all tracepoints. The enabled
5916 tracepoints will become effective the next time a trace experiment is
5920 @node Tracepoint Passcounts
5921 @subsection Tracepoint Passcounts
5925 @cindex tracepoint pass count
5926 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
5927 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
5928 automatically stop a trace experiment. If a tracepoint's passcount is
5929 @var{n}, then the trace experiment will be automatically stopped on
5930 the @var{n}'th time that tracepoint is hit. If the tracepoint number
5931 @var{num} is not specified, the @code{passcount} command sets the
5932 passcount of the most recently defined tracepoint. If no passcount is
5933 given, the trace experiment will run until stopped explicitly by the
5939 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of tracepoint 2
5941 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
5942 // most recently defined tracepoint.
5943 (@value{GDBP}) @b{trace foo}
5944 (@value{GDBP}) @b{pass 3}
5945 (@value{GDBP}) @b{trace bar}
5946 (@value{GDBP}) @b{pass 2}
5947 (@value{GDBP}) @b{trace baz}
5948 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
5949 // executed 3 times OR when bar has
5950 // been executed 2 times
5951 // OR when baz has been executed 1 time.
5955 @node Tracepoint Actions
5956 @subsection Tracepoint Action Lists
5960 @cindex tracepoint actions
5961 @item actions @r{[}@var{num}@r{]}
5962 This command will prompt for a list of actions to be taken when the
5963 tracepoint is hit. If the tracepoint number @var{num} is not
5964 specified, this command sets the actions for the one that was most
5965 recently defined (so that you can define a tracepoint and then say
5966 @code{actions} without bothering about its number). You specify the
5967 actions themselves on the following lines, one action at a time, and
5968 terminate the actions list with a line containing just @code{end}. So
5969 far, the only defined actions are @code{collect} and
5970 @code{while-stepping}.
5972 @cindex remove actions from a tracepoint
5973 To remove all actions from a tracepoint, type @samp{actions @var{num}}
5974 and follow it immediately with @samp{end}.
5977 (@value{GDBP}) @b{collect @var{data}} // collect some data
5979 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times and collect data
5981 (@value{GDBP}) @b{end} // signals the end of actions.
5984 In the following example, the action list begins with @code{collect}
5985 commands indicating the things to be collected when the tracepoint is
5986 hit. Then, in order to single-step and collect additional data
5987 following the tracepoint, a @code{while-stepping} command is used,
5988 followed by the list of things to be collected while stepping. The
5989 @code{while-stepping} command is terminated by its own separate
5990 @code{end} command. Lastly, the action list is terminated by an
5994 (@value{GDBP}) @b{trace foo}
5995 (@value{GDBP}) @b{actions}
5996 Enter actions for tracepoint 1, one per line:
6005 @kindex collect @r{(tracepoints)}
6006 @item collect @var{expr1}, @var{expr2}, @dots{}
6007 Collect values of the given expressions when the tracepoint is hit.
6008 This command accepts a comma-separated list of any valid expressions.
6009 In addition to global, static, or local variables, the following
6010 special arguments are supported:
6014 collect all registers
6017 collect all function arguments
6020 collect all local variables.
6023 You can give several consecutive @code{collect} commands, each one
6024 with a single argument, or one @code{collect} command with several
6025 arguments separated by commas: the effect is the same.
6027 The command @code{info scope} (@pxref{Symbols, info scope}) is
6028 particularly useful for figuring out what data to collect.
6030 @kindex while-stepping @r{(tracepoints)}
6031 @item while-stepping @var{n}
6032 Perform @var{n} single-step traces after the tracepoint, collecting
6033 new data at each step. The @code{while-stepping} command is
6034 followed by the list of what to collect while stepping (followed by
6035 its own @code{end} command):
6039 > collect $regs, myglobal
6045 You may abbreviate @code{while-stepping} as @code{ws} or
6049 @node Listing Tracepoints
6050 @subsection Listing Tracepoints
6053 @kindex info tracepoints
6054 @cindex information about tracepoints
6055 @item info tracepoints @r{[}@var{num}@r{]}
6056 Display information the tracepoint @var{num}. If you don't specify a
6057 tracepoint number displays information about all the tracepoints
6058 defined so far. For each tracepoint, the following information is
6065 whether it is enabled or disabled
6069 its passcount as given by the @code{passcount @var{n}} command
6071 its step count as given by the @code{while-stepping @var{n}} command
6073 where in the source files is the tracepoint set
6075 its action list as given by the @code{actions} command
6079 (@value{GDBP}) @b{info trace}
6080 Num Enb Address PassC StepC What
6081 1 y 0x002117c4 0 0 <gdb_asm>
6082 2 y 0x0020dc64 0 0 in gdb_test at gdb_test.c:375
6083 3 y 0x0020b1f4 0 0 in collect_data at ../foo.c:1741
6088 This command can be abbreviated @code{info tp}.
6091 @node Starting and Stopping Trace Experiment
6092 @subsection Starting and Stopping Trace Experiment
6096 @cindex start a new trace experiment
6097 @cindex collected data discarded
6099 This command takes no arguments. It starts the trace experiment, and
6100 begins collecting data. This has the side effect of discarding all
6101 the data collected in the trace buffer during the previous trace
6105 @cindex stop a running trace experiment
6107 This command takes no arguments. It ends the trace experiment, and
6108 stops collecting data.
6110 @strong{Note:} a trace experiment and data collection may stop
6111 automatically if any tracepoint's passcount is reached
6112 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
6115 @cindex status of trace data collection
6116 @cindex trace experiment, status of
6118 This command displays the status of the current trace data
6122 Here is an example of the commands we described so far:
6125 (@value{GDBP}) @b{trace gdb_c_test}
6126 (@value{GDBP}) @b{actions}
6127 Enter actions for tracepoint #1, one per line.
6128 > collect $regs,$locals,$args
6133 (@value{GDBP}) @b{tstart}
6134 [time passes @dots{}]
6135 (@value{GDBP}) @b{tstop}
6139 @node Analyze Collected Data
6140 @section Using the collected data
6142 After the tracepoint experiment ends, you use @value{GDBN} commands
6143 for examining the trace data. The basic idea is that each tracepoint
6144 collects a trace @dfn{snapshot} every time it is hit and another
6145 snapshot every time it single-steps. All these snapshots are
6146 consecutively numbered from zero and go into a buffer, and you can
6147 examine them later. The way you examine them is to @dfn{focus} on a
6148 specific trace snapshot. When the remote stub is focused on a trace
6149 snapshot, it will respond to all @value{GDBN} requests for memory and
6150 registers by reading from the buffer which belongs to that snapshot,
6151 rather than from @emph{real} memory or registers of the program being
6152 debugged. This means that @strong{all} @value{GDBN} commands
6153 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
6154 behave as if we were currently debugging the program state as it was
6155 when the tracepoint occurred. Any requests for data that are not in
6156 the buffer will fail.
6159 * tfind:: How to select a trace snapshot
6160 * tdump:: How to display all data for a snapshot
6161 * save-tracepoints:: How to save tracepoints for a future run
6165 @subsection @code{tfind @var{n}}
6168 @cindex select trace snapshot
6169 @cindex find trace snapshot
6170 The basic command for selecting a trace snapshot from the buffer is
6171 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
6172 counting from zero. If no argument @var{n} is given, the next
6173 snapshot is selected.
6175 Here are the various forms of using the @code{tfind} command.
6179 Find the first snapshot in the buffer. This is a synonym for
6180 @code{tfind 0} (since 0 is the number of the first snapshot).
6183 Stop debugging trace snapshots, resume @emph{live} debugging.
6186 Same as @samp{tfind none}.
6189 No argument means find the next trace snapshot.
6192 Find the previous trace snapshot before the current one. This permits
6193 retracing earlier steps.
6195 @item tfind tracepoint @var{num}
6196 Find the next snapshot associated with tracepoint @var{num}. Search
6197 proceeds forward from the last examined trace snapshot. If no
6198 argument @var{num} is given, it means find the next snapshot collected
6199 for the same tracepoint as the current snapshot.
6201 @item tfind pc @var{addr}
6202 Find the next snapshot associated with the value @var{addr} of the
6203 program counter. Search proceeds forward from the last examined trace
6204 snapshot. If no argument @var{addr} is given, it means find the next
6205 snapshot with the same value of PC as the current snapshot.
6207 @item tfind outside @var{addr1}, @var{addr2}
6208 Find the next snapshot whose PC is outside the given range of
6211 @item tfind range @var{addr1}, @var{addr2}
6212 Find the next snapshot whose PC is between @var{addr1} and
6213 @var{addr2}. @c FIXME: Is the range inclusive or exclusive?
6215 @item tfind line @r{[}@var{file}:@r{]}@var{n}
6216 Find the next snapshot associated with the source line @var{n}. If
6217 the optional argument @var{file} is given, refer to line @var{n} in
6218 that source file. Search proceeds forward from the last examined
6219 trace snapshot. If no argument @var{n} is given, it means find the
6220 next line other than the one currently being examined; thus saying
6221 @code{tfind line} repeatedly can appear to have the same effect as
6222 stepping from line to line in a @emph{live} debugging session.
6225 The default arguments for the @code{tfind} commands are specifically
6226 designed to make it easy to scan through the trace buffer. For
6227 instance, @code{tfind} with no argument selects the next trace
6228 snapshot, and @code{tfind -} with no argument selects the previous
6229 trace snapshot. So, by giving one @code{tfind} command, and then
6230 simply hitting @key{RET} repeatedly you can examine all the trace
6231 snapshots in order. Or, by saying @code{tfind -} and then hitting
6232 @key{RET} repeatedly you can examine the snapshots in reverse order.
6233 The @code{tfind line} command with no argument selects the snapshot
6234 for the next source line executed. The @code{tfind pc} command with
6235 no argument selects the next snapshot with the same program counter
6236 (PC) as the current frame. The @code{tfind tracepoint} command with
6237 no argument selects the next trace snapshot collected by the same
6238 tracepoint as the current one.
6240 In addition to letting you scan through the trace buffer manually,
6241 these commands make it easy to construct @value{GDBN} scripts that
6242 scan through the trace buffer and print out whatever collected data
6243 you are interested in. Thus, if we want to examine the PC, FP, and SP
6244 registers from each trace frame in the buffer, we can say this:
6247 (@value{GDBP}) @b{tfind start}
6248 (@value{GDBP}) @b{while ($trace_frame != -1)}
6249 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
6250 $trace_frame, $pc, $sp, $fp
6254 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
6255 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
6256 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
6257 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
6258 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
6259 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
6260 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
6261 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
6262 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
6263 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
6264 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
6267 Or, if we want to examine the variable @code{X} at each source line in
6271 (@value{GDBP}) @b{tfind start}
6272 (@value{GDBP}) @b{while ($trace_frame != -1)}
6273 > printf "Frame %d, X == %d\n", $trace_frame, X
6283 @subsection @code{tdump}
6285 @cindex dump all data collected at tracepoint
6286 @cindex tracepoint data, display
6288 This command takes no arguments. It prints all the data collected at
6289 the current trace snapshot.
6292 (@value{GDBP}) @b{trace 444}
6293 (@value{GDBP}) @b{actions}
6294 Enter actions for tracepoint #2, one per line:
6295 > collect $regs, $locals, $args, gdb_long_test
6298 (@value{GDBP}) @b{tstart}
6300 (@value{GDBP}) @b{tfind line 444}
6301 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
6303 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
6305 (@value{GDBP}) @b{tdump}
6306 Data collected at tracepoint 2, trace frame 1:
6307 d0 0xc4aa0085 -995491707
6311 d4 0x71aea3d 119204413
6316 a1 0x3000668 50333288
6319 a4 0x3000698 50333336
6321 fp 0x30bf3c 0x30bf3c
6322 sp 0x30bf34 0x30bf34
6324 pc 0x20b2c8 0x20b2c8
6328 p = 0x20e5b4 "gdb-test"
6335 gdb_long_test = 17 '\021'
6340 @node save-tracepoints
6341 @subsection @code{save-tracepoints @var{filename}}
6342 @kindex save-tracepoints
6343 @cindex save tracepoints for future sessions
6345 This command saves all current tracepoint definitions together with
6346 their actions and passcounts, into a file @file{@var{filename}}
6347 suitable for use in a later debugging session. To read the saved
6348 tracepoint definitions, use the @code{source} command (@pxref{Command
6351 @node Tracepoint Variables
6352 @section Convenience Variables for Tracepoints
6353 @cindex tracepoint variables
6354 @cindex convenience variables for tracepoints
6357 @vindex $trace_frame
6358 @item (int) $trace_frame
6359 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
6360 snapshot is selected.
6363 @item (int) $tracepoint
6364 The tracepoint for the current trace snapshot.
6367 @item (int) $trace_line
6368 The line number for the current trace snapshot.
6371 @item (char []) $trace_file
6372 The source file for the current trace snapshot.
6375 @item (char []) $trace_func
6376 The name of the function containing @code{$tracepoint}.
6379 Note: @code{$trace_file} is not suitable for use in @code{printf},
6380 use @code{output} instead.
6382 Here's a simple example of using these convenience variables for
6383 stepping through all the trace snapshots and printing some of their
6387 (@value{GDBP}) @b{tfind start}
6389 (@value{GDBP}) @b{while $trace_frame != -1}
6390 > output $trace_file
6391 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
6397 @chapter Debugging Programs That Use Overlays
6400 If your program is too large to fit completely in your target system's
6401 memory, you can sometimes use @dfn{overlays} to work around this
6402 problem. @value{GDBN} provides some support for debugging programs that
6406 * How Overlays Work:: A general explanation of overlays.
6407 * Overlay Commands:: Managing overlays in @value{GDBN}.
6408 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
6409 mapped by asking the inferior.
6410 * Overlay Sample Program:: A sample program using overlays.
6413 @node How Overlays Work
6414 @section How Overlays Work
6415 @cindex mapped overlays
6416 @cindex unmapped overlays
6417 @cindex load address, overlay's
6418 @cindex mapped address
6419 @cindex overlay area
6421 Suppose you have a computer whose instruction address space is only 64
6422 kilobytes long, but which has much more memory which can be accessed by
6423 other means: special instructions, segment registers, or memory
6424 management hardware, for example. Suppose further that you want to
6425 adapt a program which is larger than 64 kilobytes to run on this system.
6427 One solution is to identify modules of your program which are relatively
6428 independent, and need not call each other directly; call these modules
6429 @dfn{overlays}. Separate the overlays from the main program, and place
6430 their machine code in the larger memory. Place your main program in
6431 instruction memory, but leave at least enough space there to hold the
6432 largest overlay as well.
6434 Now, to call a function located in an overlay, you must first copy that
6435 overlay's machine code from the large memory into the space set aside
6436 for it in the instruction memory, and then jump to its entry point
6441 Data Instruction Larger
6442 Address Space Address Space Address Space
6443 +-----------+ +-----------+ +-----------+
6445 +-----------+ +-----------+ +-----------+<-- overlay 1
6446 | program | | main | | | load address
6447 | variables | | program | | overlay 1 |
6448 | and heap | | | ,---| |
6449 +-----------+ | | | | |
6450 | | +-----------+ | +-----------+
6451 +-----------+ | | | | |
6452 mapped --->+-----------+ / +-----------+<-- overlay 2
6453 address | overlay | <-' | overlay 2 | load address
6455 | | <---. +-----------+
6458 | | | +-----------+<-- overlay 3
6459 +-----------+ `--| | load address
6466 To map an overlay, copy its code from the larger address space
6467 to the instruction address space. Since the overlays shown here
6468 all use the same mapped address, only one may be mapped at a time.
6472 This diagram shows a system with separate data and instruction address
6473 spaces. For a system with a single address space for data and
6474 instructions, the diagram would be similar, except that the program
6475 variables and heap would share an address space with the main program
6476 and the overlay area.
6478 An overlay loaded into instruction memory and ready for use is called a
6479 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
6480 instruction memory. An overlay not present (or only partially present)
6481 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
6482 is its address in the larger memory. The mapped address is also called
6483 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
6484 called the @dfn{load memory address}, or @dfn{LMA}.
6486 Unfortunately, overlays are not a completely transparent way to adapt a
6487 program to limited instruction memory. They introduce a new set of
6488 global constraints you must keep in mind as you design your program:
6493 Before calling or returning to a function in an overlay, your program
6494 must make sure that overlay is actually mapped. Otherwise, the call or
6495 return will transfer control to the right address, but in the wrong
6496 overlay, and your program will probably crash.
6499 If the process of mapping an overlay is expensive on your system, you
6500 will need to choose your overlays carefully to minimize their effect on
6501 your program's performance.
6504 The executable file you load onto your system must contain each
6505 overlay's instructions, appearing at the overlay's load address, not its
6506 mapped address. However, each overlay's instructions must be relocated
6507 and its symbols defined as if the overlay were at its mapped address.
6508 You can use GNU linker scripts to specify different load and relocation
6509 addresses for pieces of your program; see @ref{Overlay Description,,,
6510 ld.info, Using ld: the GNU linker}.
6513 The procedure for loading executable files onto your system must be able
6514 to load their contents into the larger address space as well as the
6515 instruction and data spaces.
6519 The overlay system described above is rather simple, and could be
6520 improved in many ways:
6525 If your system has suitable bank switch registers or memory management
6526 hardware, you could use those facilities to make an overlay's load area
6527 contents simply appear at their mapped address in instruction space.
6528 This would probably be faster than copying the overlay to its mapped
6529 area in the usual way.
6532 If your overlays are small enough, you could set aside more than one
6533 overlay area, and have more than one overlay mapped at a time.
6536 You can use overlays to manage data, as well as instructions. In
6537 general, data overlays are even less transparent to your design than
6538 code overlays: whereas code overlays only require care when you call or
6539 return to functions, data overlays require care every time you access
6540 the data. Also, if you change the contents of a data overlay, you
6541 must copy its contents back out to its load address before you can copy a
6542 different data overlay into the same mapped area.
6547 @node Overlay Commands
6548 @section Overlay Commands
6550 To use @value{GDBN}'s overlay support, each overlay in your program must
6551 correspond to a separate section of the executable file. The section's
6552 virtual memory address and load memory address must be the overlay's
6553 mapped and load addresses. Identifying overlays with sections allows
6554 @value{GDBN} to determine the appropriate address of a function or
6555 variable, depending on whether the overlay is mapped or not.
6557 @value{GDBN}'s overlay commands all start with the word @code{overlay};
6558 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
6563 Disable @value{GDBN}'s overlay support. When overlay support is
6564 disabled, @value{GDBN} assumes that all functions and variables are
6565 always present at their mapped addresses. By default, @value{GDBN}'s
6566 overlay support is disabled.
6568 @item overlay manual
6569 @kindex overlay manual
6570 @cindex manual overlay debugging
6571 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
6572 relies on you to tell it which overlays are mapped, and which are not,
6573 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
6574 commands described below.
6576 @item overlay map-overlay @var{overlay}
6577 @itemx overlay map @var{overlay}
6578 @kindex overlay map-overlay
6579 @cindex map an overlay
6580 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
6581 be the name of the object file section containing the overlay. When an
6582 overlay is mapped, @value{GDBN} assumes it can find the overlay's
6583 functions and variables at their mapped addresses. @value{GDBN} assumes
6584 that any other overlays whose mapped ranges overlap that of
6585 @var{overlay} are now unmapped.
6587 @item overlay unmap-overlay @var{overlay}
6588 @itemx overlay unmap @var{overlay}
6589 @kindex overlay unmap-overlay
6590 @cindex unmap an overlay
6591 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
6592 must be the name of the object file section containing the overlay.
6593 When an overlay is unmapped, @value{GDBN} assumes it can find the
6594 overlay's functions and variables at their load addresses.
6597 @kindex overlay auto
6598 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
6599 consults a data structure the overlay manager maintains in the inferior
6600 to see which overlays are mapped. For details, see @ref{Automatic
6603 @item overlay load-target
6605 @kindex overlay load-target
6606 @cindex reloading the overlay table
6607 Re-read the overlay table from the inferior. Normally, @value{GDBN}
6608 re-reads the table @value{GDBN} automatically each time the inferior
6609 stops, so this command should only be necessary if you have changed the
6610 overlay mapping yourself using @value{GDBN}. This command is only
6611 useful when using automatic overlay debugging.
6613 @item overlay list-overlays
6615 @cindex listing mapped overlays
6616 Display a list of the overlays currently mapped, along with their mapped
6617 addresses, load addresses, and sizes.
6621 Normally, when @value{GDBN} prints a code address, it includes the name
6622 of the function the address falls in:
6626 $3 = @{int ()@} 0x11a0 <main>
6629 When overlay debugging is enabled, @value{GDBN} recognizes code in
6630 unmapped overlays, and prints the names of unmapped functions with
6631 asterisks around them. For example, if @code{foo} is a function in an
6632 unmapped overlay, @value{GDBN} prints it this way:
6636 No sections are mapped.
6638 $5 = @{int (int)@} 0x100000 <*foo*>
6641 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
6646 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
6647 mapped at 0x1016 - 0x104a
6649 $6 = @{int (int)@} 0x1016 <foo>
6652 When overlay debugging is enabled, @value{GDBN} can find the correct
6653 address for functions and variables in an overlay, whether or not the
6654 overlay is mapped. This allows most @value{GDBN} commands, like
6655 @code{break} and @code{disassemble}, to work normally, even on unmapped
6656 code. However, @value{GDBN}'s breakpoint support has some limitations:
6660 @cindex breakpoints in overlays
6661 @cindex overlays, setting breakpoints in
6662 You can set breakpoints in functions in unmapped overlays, as long as
6663 @value{GDBN} can write to the overlay at its load address.
6665 @value{GDBN} can not set hardware or simulator-based breakpoints in
6666 unmapped overlays. However, if you set a breakpoint at the end of your
6667 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
6668 you are using manual overlay management), @value{GDBN} will re-set its
6669 breakpoints properly.
6673 @node Automatic Overlay Debugging
6674 @section Automatic Overlay Debugging
6675 @cindex automatic overlay debugging
6677 @value{GDBN} can automatically track which overlays are mapped and which
6678 are not, given some simple co-operation from the overlay manager in the
6679 inferior. If you enable automatic overlay debugging with the
6680 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
6681 looks in the inferior's memory for certain variables describing the
6682 current state of the overlays.
6684 Here are the variables your overlay manager must define to support
6685 @value{GDBN}'s automatic overlay debugging:
6689 @item @code{_ovly_table}:
6690 This variable must be an array of the following structures:
6695 /* The overlay's mapped address. */
6698 /* The size of the overlay, in bytes. */
6701 /* The overlay's load address. */
6704 /* Non-zero if the overlay is currently mapped;
6706 unsigned long mapped;
6710 @item @code{_novlys}:
6711 This variable must be a four-byte signed integer, holding the total
6712 number of elements in @code{_ovly_table}.
6716 To decide whether a particular overlay is mapped or not, @value{GDBN}
6717 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
6718 @code{lma} members equal the VMA and LMA of the overlay's section in the
6719 executable file. When @value{GDBN} finds a matching entry, it consults
6720 the entry's @code{mapped} member to determine whether the overlay is
6724 @node Overlay Sample Program
6725 @section Overlay Sample Program
6726 @cindex overlay example program
6728 When linking a program which uses overlays, you must place the overlays
6729 at their load addresses, while relocating them to run at their mapped
6730 addresses. To do this, you must write a linker script (@pxref{Overlay
6731 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
6732 since linker scripts are specific to a particular host system, target
6733 architecture, and target memory layout, this manual cannot provide
6734 portable sample code demonstrating @value{GDBN}'s overlay support.
6736 However, the @value{GDBN} source distribution does contain an overlaid
6737 program, with linker scripts for a few systems, as part of its test
6738 suite. The program consists of the following files from
6739 @file{gdb/testsuite/gdb.base}:
6743 The main program file.
6745 A simple overlay manager, used by @file{overlays.c}.
6750 Overlay modules, loaded and used by @file{overlays.c}.
6753 Linker scripts for linking the test program on the @code{d10v-elf}
6754 and @code{m32r-elf} targets.
6757 You can build the test program using the @code{d10v-elf} GCC
6758 cross-compiler like this:
6761 $ d10v-elf-gcc -g -c overlays.c
6762 $ d10v-elf-gcc -g -c ovlymgr.c
6763 $ d10v-elf-gcc -g -c foo.c
6764 $ d10v-elf-gcc -g -c bar.c
6765 $ d10v-elf-gcc -g -c baz.c
6766 $ d10v-elf-gcc -g -c grbx.c
6767 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
6768 baz.o grbx.o -Wl,-Td10v.ld -o overlays
6771 The build process is identical for any other architecture, except that
6772 you must substitute the appropriate compiler and linker script for the
6773 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
6777 @chapter Using @value{GDBN} with Different Languages
6780 Although programming languages generally have common aspects, they are
6781 rarely expressed in the same manner. For instance, in ANSI C,
6782 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
6783 Modula-2, it is accomplished by @code{p^}. Values can also be
6784 represented (and displayed) differently. Hex numbers in C appear as
6785 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
6787 @cindex working language
6788 Language-specific information is built into @value{GDBN} for some languages,
6789 allowing you to express operations like the above in your program's
6790 native language, and allowing @value{GDBN} to output values in a manner
6791 consistent with the syntax of your program's native language. The
6792 language you use to build expressions is called the @dfn{working
6796 * Setting:: Switching between source languages
6797 * Show:: Displaying the language
6798 * Checks:: Type and range checks
6799 * Support:: Supported languages
6803 @section Switching between source languages
6805 There are two ways to control the working language---either have @value{GDBN}
6806 set it automatically, or select it manually yourself. You can use the
6807 @code{set language} command for either purpose. On startup, @value{GDBN}
6808 defaults to setting the language automatically. The working language is
6809 used to determine how expressions you type are interpreted, how values
6812 In addition to the working language, every source file that
6813 @value{GDBN} knows about has its own working language. For some object
6814 file formats, the compiler might indicate which language a particular
6815 source file is in. However, most of the time @value{GDBN} infers the
6816 language from the name of the file. The language of a source file
6817 controls whether C@t{++} names are demangled---this way @code{backtrace} can
6818 show each frame appropriately for its own language. There is no way to
6819 set the language of a source file from within @value{GDBN}, but you can
6820 set the language associated with a filename extension. @xref{Show, ,
6821 Displaying the language}.
6823 This is most commonly a problem when you use a program, such
6824 as @code{cfront} or @code{f2c}, that generates C but is written in
6825 another language. In that case, make the
6826 program use @code{#line} directives in its C output; that way
6827 @value{GDBN} will know the correct language of the source code of the original
6828 program, and will display that source code, not the generated C code.
6831 * Filenames:: Filename extensions and languages.
6832 * Manually:: Setting the working language manually
6833 * Automatically:: Having @value{GDBN} infer the source language
6837 @subsection List of filename extensions and languages
6839 If a source file name ends in one of the following extensions, then
6840 @value{GDBN} infers that its language is the one indicated.
6865 Modula-2 source file
6869 Assembler source file. This actually behaves almost like C, but
6870 @value{GDBN} does not skip over function prologues when stepping.
6873 In addition, you may set the language associated with a filename
6874 extension. @xref{Show, , Displaying the language}.
6877 @subsection Setting the working language
6879 If you allow @value{GDBN} to set the language automatically,
6880 expressions are interpreted the same way in your debugging session and
6883 @kindex set language
6884 If you wish, you may set the language manually. To do this, issue the
6885 command @samp{set language @var{lang}}, where @var{lang} is the name of
6887 @code{c} or @code{modula-2}.
6888 For a list of the supported languages, type @samp{set language}.
6890 Setting the language manually prevents @value{GDBN} from updating the working
6891 language automatically. This can lead to confusion if you try
6892 to debug a program when the working language is not the same as the
6893 source language, when an expression is acceptable to both
6894 languages---but means different things. For instance, if the current
6895 source file were written in C, and @value{GDBN} was parsing Modula-2, a
6903 might not have the effect you intended. In C, this means to add
6904 @code{b} and @code{c} and place the result in @code{a}. The result
6905 printed would be the value of @code{a}. In Modula-2, this means to compare
6906 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
6909 @subsection Having @value{GDBN} infer the source language
6911 To have @value{GDBN} set the working language automatically, use
6912 @samp{set language local} or @samp{set language auto}. @value{GDBN}
6913 then infers the working language. That is, when your program stops in a
6914 frame (usually by encountering a breakpoint), @value{GDBN} sets the
6915 working language to the language recorded for the function in that
6916 frame. If the language for a frame is unknown (that is, if the function
6917 or block corresponding to the frame was defined in a source file that
6918 does not have a recognized extension), the current working language is
6919 not changed, and @value{GDBN} issues a warning.
6921 This may not seem necessary for most programs, which are written
6922 entirely in one source language. However, program modules and libraries
6923 written in one source language can be used by a main program written in
6924 a different source language. Using @samp{set language auto} in this
6925 case frees you from having to set the working language manually.
6928 @section Displaying the language
6930 The following commands help you find out which language is the
6931 working language, and also what language source files were written in.
6933 @kindex show language
6934 @kindex info frame@r{, show the source language}
6935 @kindex info source@r{, show the source language}
6938 Display the current working language. This is the
6939 language you can use with commands such as @code{print} to
6940 build and compute expressions that may involve variables in your program.
6943 Display the source language for this frame. This language becomes the
6944 working language if you use an identifier from this frame.
6945 @xref{Frame Info, ,Information about a frame}, to identify the other
6946 information listed here.
6949 Display the source language of this source file.
6950 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
6951 information listed here.
6954 In unusual circumstances, you may have source files with extensions
6955 not in the standard list. You can then set the extension associated
6956 with a language explicitly:
6958 @kindex set extension-language
6959 @kindex info extensions
6961 @item set extension-language @var{.ext} @var{language}
6962 Set source files with extension @var{.ext} to be assumed to be in
6963 the source language @var{language}.
6965 @item info extensions
6966 List all the filename extensions and the associated languages.
6970 @section Type and range checking
6973 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
6974 checking are included, but they do not yet have any effect. This
6975 section documents the intended facilities.
6977 @c FIXME remove warning when type/range code added
6979 Some languages are designed to guard you against making seemingly common
6980 errors through a series of compile- and run-time checks. These include
6981 checking the type of arguments to functions and operators, and making
6982 sure mathematical overflows are caught at run time. Checks such as
6983 these help to ensure a program's correctness once it has been compiled
6984 by eliminating type mismatches, and providing active checks for range
6985 errors when your program is running.
6987 @value{GDBN} can check for conditions like the above if you wish.
6988 Although @value{GDBN} does not check the statements in your program, it
6989 can check expressions entered directly into @value{GDBN} for evaluation via
6990 the @code{print} command, for example. As with the working language,
6991 @value{GDBN} can also decide whether or not to check automatically based on
6992 your program's source language. @xref{Support, ,Supported languages},
6993 for the default settings of supported languages.
6996 * Type Checking:: An overview of type checking
6997 * Range Checking:: An overview of range checking
7000 @cindex type checking
7001 @cindex checks, type
7003 @subsection An overview of type checking
7005 Some languages, such as Modula-2, are strongly typed, meaning that the
7006 arguments to operators and functions have to be of the correct type,
7007 otherwise an error occurs. These checks prevent type mismatch
7008 errors from ever causing any run-time problems. For example,
7016 The second example fails because the @code{CARDINAL} 1 is not
7017 type-compatible with the @code{REAL} 2.3.
7019 For the expressions you use in @value{GDBN} commands, you can tell the
7020 @value{GDBN} type checker to skip checking;
7021 to treat any mismatches as errors and abandon the expression;
7022 or to only issue warnings when type mismatches occur,
7023 but evaluate the expression anyway. When you choose the last of
7024 these, @value{GDBN} evaluates expressions like the second example above, but
7025 also issues a warning.
7027 Even if you turn type checking off, there may be other reasons
7028 related to type that prevent @value{GDBN} from evaluating an expression.
7029 For instance, @value{GDBN} does not know how to add an @code{int} and
7030 a @code{struct foo}. These particular type errors have nothing to do
7031 with the language in use, and usually arise from expressions, such as
7032 the one described above, which make little sense to evaluate anyway.
7034 Each language defines to what degree it is strict about type. For
7035 instance, both Modula-2 and C require the arguments to arithmetical
7036 operators to be numbers. In C, enumerated types and pointers can be
7037 represented as numbers, so that they are valid arguments to mathematical
7038 operators. @xref{Support, ,Supported languages}, for further
7039 details on specific languages.
7041 @value{GDBN} provides some additional commands for controlling the type checker:
7043 @kindex set check@r{, type}
7044 @kindex set check type
7045 @kindex show check type
7047 @item set check type auto
7048 Set type checking on or off based on the current working language.
7049 @xref{Support, ,Supported languages}, for the default settings for
7052 @item set check type on
7053 @itemx set check type off
7054 Set type checking on or off, overriding the default setting for the
7055 current working language. Issue a warning if the setting does not
7056 match the language default. If any type mismatches occur in
7057 evaluating an expression while type checking is on, @value{GDBN} prints a
7058 message and aborts evaluation of the expression.
7060 @item set check type warn
7061 Cause the type checker to issue warnings, but to always attempt to
7062 evaluate the expression. Evaluating the expression may still
7063 be impossible for other reasons. For example, @value{GDBN} cannot add
7064 numbers and structures.
7067 Show the current setting of the type checker, and whether or not @value{GDBN}
7068 is setting it automatically.
7071 @cindex range checking
7072 @cindex checks, range
7073 @node Range Checking
7074 @subsection An overview of range checking
7076 In some languages (such as Modula-2), it is an error to exceed the
7077 bounds of a type; this is enforced with run-time checks. Such range
7078 checking is meant to ensure program correctness by making sure
7079 computations do not overflow, or indices on an array element access do
7080 not exceed the bounds of the array.
7082 For expressions you use in @value{GDBN} commands, you can tell
7083 @value{GDBN} to treat range errors in one of three ways: ignore them,
7084 always treat them as errors and abandon the expression, or issue
7085 warnings but evaluate the expression anyway.
7087 A range error can result from numerical overflow, from exceeding an
7088 array index bound, or when you type a constant that is not a member
7089 of any type. Some languages, however, do not treat overflows as an
7090 error. In many implementations of C, mathematical overflow causes the
7091 result to ``wrap around'' to lower values---for example, if @var{m} is
7092 the largest integer value, and @var{s} is the smallest, then
7095 @var{m} + 1 @result{} @var{s}
7098 This, too, is specific to individual languages, and in some cases
7099 specific to individual compilers or machines. @xref{Support, ,
7100 Supported languages}, for further details on specific languages.
7102 @value{GDBN} provides some additional commands for controlling the range checker:
7104 @kindex set check@r{, range}
7105 @kindex set check range
7106 @kindex show check range
7108 @item set check range auto
7109 Set range checking on or off based on the current working language.
7110 @xref{Support, ,Supported languages}, for the default settings for
7113 @item set check range on
7114 @itemx set check range off
7115 Set range checking on or off, overriding the default setting for the
7116 current working language. A warning is issued if the setting does not
7117 match the language default. If a range error occurs and range checking is on,
7118 then a message is printed and evaluation of the expression is aborted.
7120 @item set check range warn
7121 Output messages when the @value{GDBN} range checker detects a range error,
7122 but attempt to evaluate the expression anyway. Evaluating the
7123 expression may still be impossible for other reasons, such as accessing
7124 memory that the process does not own (a typical example from many Unix
7128 Show the current setting of the range checker, and whether or not it is
7129 being set automatically by @value{GDBN}.
7133 @section Supported languages
7135 @value{GDBN} supports C, C@t{++}, Fortran, Java, Chill, assembly, and Modula-2.
7136 @c This is false ...
7137 Some @value{GDBN} features may be used in expressions regardless of the
7138 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
7139 and the @samp{@{type@}addr} construct (@pxref{Expressions,
7140 ,Expressions}) can be used with the constructs of any supported
7143 The following sections detail to what degree each source language is
7144 supported by @value{GDBN}. These sections are not meant to be language
7145 tutorials or references, but serve only as a reference guide to what the
7146 @value{GDBN} expression parser accepts, and what input and output
7147 formats should look like for different languages. There are many good
7148 books written on each of these languages; please look to these for a
7149 language reference or tutorial.
7153 * Modula-2:: Modula-2
7158 @subsection C and C@t{++}
7160 @cindex C and C@t{++}
7161 @cindex expressions in C or C@t{++}
7163 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
7164 to both languages. Whenever this is the case, we discuss those languages
7168 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
7169 @cindex @sc{gnu} C@t{++}
7170 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
7171 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
7172 effectively, you must compile your C@t{++} programs with a supported
7173 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
7174 compiler (@code{aCC}).
7176 For best results when using @sc{gnu} C@t{++}, use the stabs debugging
7177 format. You can select that format explicitly with the @code{g++}
7178 command-line options @samp{-gstabs} or @samp{-gstabs+}. See
7179 @ref{Debugging Options,,Options for Debugging Your Program or @sc{gnu}
7180 CC, gcc.info, Using @sc{gnu} CC}, for more information.
7183 * C Operators:: C and C@t{++} operators
7184 * C Constants:: C and C@t{++} constants
7185 * C plus plus expressions:: C@t{++} expressions
7186 * C Defaults:: Default settings for C and C@t{++}
7187 * C Checks:: C and C@t{++} type and range checks
7188 * Debugging C:: @value{GDBN} and C
7189 * Debugging C plus plus:: @value{GDBN} features for C@t{++}
7193 @subsubsection C and C@t{++} operators
7195 @cindex C and C@t{++} operators
7197 Operators must be defined on values of specific types. For instance,
7198 @code{+} is defined on numbers, but not on structures. Operators are
7199 often defined on groups of types.
7201 For the purposes of C and C@t{++}, the following definitions hold:
7206 @emph{Integral types} include @code{int} with any of its storage-class
7207 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
7210 @emph{Floating-point types} include @code{float}, @code{double}, and
7211 @code{long double} (if supported by the target platform).
7214 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
7217 @emph{Scalar types} include all of the above.
7222 The following operators are supported. They are listed here
7223 in order of increasing precedence:
7227 The comma or sequencing operator. Expressions in a comma-separated list
7228 are evaluated from left to right, with the result of the entire
7229 expression being the last expression evaluated.
7232 Assignment. The value of an assignment expression is the value
7233 assigned. Defined on scalar types.
7236 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
7237 and translated to @w{@code{@var{a} = @var{a op b}}}.
7238 @w{@code{@var{op}=}} and @code{=} have the same precedence.
7239 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
7240 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
7243 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
7244 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
7248 Logical @sc{or}. Defined on integral types.
7251 Logical @sc{and}. Defined on integral types.
7254 Bitwise @sc{or}. Defined on integral types.
7257 Bitwise exclusive-@sc{or}. Defined on integral types.
7260 Bitwise @sc{and}. Defined on integral types.
7263 Equality and inequality. Defined on scalar types. The value of these
7264 expressions is 0 for false and non-zero for true.
7266 @item <@r{, }>@r{, }<=@r{, }>=
7267 Less than, greater than, less than or equal, greater than or equal.
7268 Defined on scalar types. The value of these expressions is 0 for false
7269 and non-zero for true.
7272 left shift, and right shift. Defined on integral types.
7275 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
7278 Addition and subtraction. Defined on integral types, floating-point types and
7281 @item *@r{, }/@r{, }%
7282 Multiplication, division, and modulus. Multiplication and division are
7283 defined on integral and floating-point types. Modulus is defined on
7287 Increment and decrement. When appearing before a variable, the
7288 operation is performed before the variable is used in an expression;
7289 when appearing after it, the variable's value is used before the
7290 operation takes place.
7293 Pointer dereferencing. Defined on pointer types. Same precedence as
7297 Address operator. Defined on variables. Same precedence as @code{++}.
7299 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
7300 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
7301 (or, if you prefer, simply @samp{&&@var{ref}}) to examine the address
7302 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
7306 Negative. Defined on integral and floating-point types. Same
7307 precedence as @code{++}.
7310 Logical negation. Defined on integral types. Same precedence as
7314 Bitwise complement operator. Defined on integral types. Same precedence as
7319 Structure member, and pointer-to-structure member. For convenience,
7320 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
7321 pointer based on the stored type information.
7322 Defined on @code{struct} and @code{union} data.
7325 Dereferences of pointers to members.
7328 Array indexing. @code{@var{a}[@var{i}]} is defined as
7329 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
7332 Function parameter list. Same precedence as @code{->}.
7335 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
7336 and @code{class} types.
7339 Doubled colons also represent the @value{GDBN} scope operator
7340 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
7344 If an operator is redefined in the user code, @value{GDBN} usually
7345 attempts to invoke the redefined version instead of using the operator's
7353 @subsubsection C and C@t{++} constants
7355 @cindex C and C@t{++} constants
7357 @value{GDBN} allows you to express the constants of C and C@t{++} in the
7362 Integer constants are a sequence of digits. Octal constants are
7363 specified by a leading @samp{0} (i.e. zero), and hexadecimal constants by
7364 a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
7365 @samp{l}, specifying that the constant should be treated as a
7369 Floating point constants are a sequence of digits, followed by a decimal
7370 point, followed by a sequence of digits, and optionally followed by an
7371 exponent. An exponent is of the form:
7372 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
7373 sequence of digits. The @samp{+} is optional for positive exponents.
7374 A floating-point constant may also end with a letter @samp{f} or
7375 @samp{F}, specifying that the constant should be treated as being of
7376 the @code{float} (as opposed to the default @code{double}) type; or with
7377 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
7381 Enumerated constants consist of enumerated identifiers, or their
7382 integral equivalents.
7385 Character constants are a single character surrounded by single quotes
7386 (@code{'}), or a number---the ordinal value of the corresponding character
7387 (usually its @sc{ascii} value). Within quotes, the single character may
7388 be represented by a letter or by @dfn{escape sequences}, which are of
7389 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
7390 of the character's ordinal value; or of the form @samp{\@var{x}}, where
7391 @samp{@var{x}} is a predefined special character---for example,
7392 @samp{\n} for newline.
7395 String constants are a sequence of character constants surrounded by
7396 double quotes (@code{"}). Any valid character constant (as described
7397 above) may appear. Double quotes within the string must be preceded by
7398 a backslash, so for instance @samp{"a\"b'c"} is a string of five
7402 Pointer constants are an integral value. You can also write pointers
7403 to constants using the C operator @samp{&}.
7406 Array constants are comma-separated lists surrounded by braces @samp{@{}
7407 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
7408 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
7409 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
7413 * C plus plus expressions::
7420 @node C plus plus expressions
7421 @subsubsection C@t{++} expressions
7423 @cindex expressions in C@t{++}
7424 @value{GDBN} expression handling can interpret most C@t{++} expressions.
7426 @cindex C@t{++} support, not in @sc{coff}
7427 @cindex @sc{coff} versus C@t{++}
7428 @cindex C@t{++} and object formats
7429 @cindex object formats and C@t{++}
7430 @cindex a.out and C@t{++}
7431 @cindex @sc{ecoff} and C@t{++}
7432 @cindex @sc{xcoff} and C@t{++}
7433 @cindex @sc{elf}/stabs and C@t{++}
7434 @cindex @sc{elf}/@sc{dwarf} and C@t{++}
7435 @c FIXME!! GDB may eventually be able to debug C++ using DWARF; check
7436 @c periodically whether this has happened...
7438 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
7439 proper compiler. Typically, C@t{++} debugging depends on the use of
7440 additional debugging information in the symbol table, and thus requires
7441 special support. In particular, if your compiler generates a.out, MIPS
7442 @sc{ecoff}, RS/6000 @sc{xcoff}, or @sc{elf} with stabs extensions to the
7443 symbol table, these facilities are all available. (With @sc{gnu} CC,
7444 you can use the @samp{-gstabs} option to request stabs debugging
7445 extensions explicitly.) Where the object code format is standard
7446 @sc{coff} or @sc{dwarf} in @sc{elf}, on the other hand, most of the C@t{++}
7447 support in @value{GDBN} does @emph{not} work.
7452 @cindex member functions
7454 Member function calls are allowed; you can use expressions like
7457 count = aml->GetOriginal(x, y)
7460 @vindex this@r{, inside C@t{++} member functions}
7461 @cindex namespace in C@t{++}
7463 While a member function is active (in the selected stack frame), your
7464 expressions have the same namespace available as the member function;
7465 that is, @value{GDBN} allows implicit references to the class instance
7466 pointer @code{this} following the same rules as C@t{++}.
7468 @cindex call overloaded functions
7469 @cindex overloaded functions, calling
7470 @cindex type conversions in C@t{++}
7472 You can call overloaded functions; @value{GDBN} resolves the function
7473 call to the right definition, with some restrictions. @value{GDBN} does not
7474 perform overload resolution involving user-defined type conversions,
7475 calls to constructors, or instantiations of templates that do not exist
7476 in the program. It also cannot handle ellipsis argument lists or
7479 It does perform integral conversions and promotions, floating-point
7480 promotions, arithmetic conversions, pointer conversions, conversions of
7481 class objects to base classes, and standard conversions such as those of
7482 functions or arrays to pointers; it requires an exact match on the
7483 number of function arguments.
7485 Overload resolution is always performed, unless you have specified
7486 @code{set overload-resolution off}. @xref{Debugging C plus plus,
7487 ,@value{GDBN} features for C@t{++}}.
7489 You must specify @code{set overload-resolution off} in order to use an
7490 explicit function signature to call an overloaded function, as in
7492 p 'foo(char,int)'('x', 13)
7495 The @value{GDBN} command-completion facility can simplify this;
7496 see @ref{Completion, ,Command completion}.
7498 @cindex reference declarations
7500 @value{GDBN} understands variables declared as C@t{++} references; you can use
7501 them in expressions just as you do in C@t{++} source---they are automatically
7504 In the parameter list shown when @value{GDBN} displays a frame, the values of
7505 reference variables are not displayed (unlike other variables); this
7506 avoids clutter, since references are often used for large structures.
7507 The @emph{address} of a reference variable is always shown, unless
7508 you have specified @samp{set print address off}.
7511 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
7512 expressions can use it just as expressions in your program do. Since
7513 one scope may be defined in another, you can use @code{::} repeatedly if
7514 necessary, for example in an expression like
7515 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
7516 resolving name scope by reference to source files, in both C and C@t{++}
7517 debugging (@pxref{Variables, ,Program variables}).
7520 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
7521 calling virtual functions correctly, printing out virtual bases of
7522 objects, calling functions in a base subobject, casting objects, and
7523 invoking user-defined operators.
7526 @subsubsection C and C@t{++} defaults
7528 @cindex C and C@t{++} defaults
7530 If you allow @value{GDBN} to set type and range checking automatically, they
7531 both default to @code{off} whenever the working language changes to
7532 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
7533 selects the working language.
7535 If you allow @value{GDBN} to set the language automatically, it
7536 recognizes source files whose names end with @file{.c}, @file{.C}, or
7537 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
7538 these files, it sets the working language to C or C@t{++}.
7539 @xref{Automatically, ,Having @value{GDBN} infer the source language},
7540 for further details.
7542 @c Type checking is (a) primarily motivated by Modula-2, and (b)
7543 @c unimplemented. If (b) changes, it might make sense to let this node
7544 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
7547 @subsubsection C and C@t{++} type and range checks
7549 @cindex C and C@t{++} checks
7551 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
7552 is not used. However, if you turn type checking on, @value{GDBN}
7553 considers two variables type equivalent if:
7557 The two variables are structured and have the same structure, union, or
7561 The two variables have the same type name, or types that have been
7562 declared equivalent through @code{typedef}.
7565 @c leaving this out because neither J Gilmore nor R Pesch understand it.
7568 The two @code{struct}, @code{union}, or @code{enum} variables are
7569 declared in the same declaration. (Note: this may not be true for all C
7574 Range checking, if turned on, is done on mathematical operations. Array
7575 indices are not checked, since they are often used to index a pointer
7576 that is not itself an array.
7579 @subsubsection @value{GDBN} and C
7581 The @code{set print union} and @code{show print union} commands apply to
7582 the @code{union} type. When set to @samp{on}, any @code{union} that is
7583 inside a @code{struct} or @code{class} is also printed. Otherwise, it
7584 appears as @samp{@{...@}}.
7586 The @code{@@} operator aids in the debugging of dynamic arrays, formed
7587 with pointers and a memory allocation function. @xref{Expressions,
7591 * Debugging C plus plus::
7594 @node Debugging C plus plus
7595 @subsubsection @value{GDBN} features for C@t{++}
7597 @cindex commands for C@t{++}
7599 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
7600 designed specifically for use with C@t{++}. Here is a summary:
7603 @cindex break in overloaded functions
7604 @item @r{breakpoint menus}
7605 When you want a breakpoint in a function whose name is overloaded,
7606 @value{GDBN} breakpoint menus help you specify which function definition
7607 you want. @xref{Breakpoint Menus,,Breakpoint menus}.
7609 @cindex overloading in C@t{++}
7610 @item rbreak @var{regex}
7611 Setting breakpoints using regular expressions is helpful for setting
7612 breakpoints on overloaded functions that are not members of any special
7614 @xref{Set Breaks, ,Setting breakpoints}.
7616 @cindex C@t{++} exception handling
7619 Debug C@t{++} exception handling using these commands. @xref{Set
7620 Catchpoints, , Setting catchpoints}.
7623 @item ptype @var{typename}
7624 Print inheritance relationships as well as other information for type
7626 @xref{Symbols, ,Examining the Symbol Table}.
7628 @cindex C@t{++} symbol display
7629 @item set print demangle
7630 @itemx show print demangle
7631 @itemx set print asm-demangle
7632 @itemx show print asm-demangle
7633 Control whether C@t{++} symbols display in their source form, both when
7634 displaying code as C@t{++} source and when displaying disassemblies.
7635 @xref{Print Settings, ,Print settings}.
7637 @item set print object
7638 @itemx show print object
7639 Choose whether to print derived (actual) or declared types of objects.
7640 @xref{Print Settings, ,Print settings}.
7642 @item set print vtbl
7643 @itemx show print vtbl
7644 Control the format for printing virtual function tables.
7645 @xref{Print Settings, ,Print settings}.
7646 (The @code{vtbl} commands do not work on programs compiled with the HP
7647 ANSI C@t{++} compiler (@code{aCC}).)
7649 @kindex set overload-resolution
7650 @cindex overloaded functions, overload resolution
7651 @item set overload-resolution on
7652 Enable overload resolution for C@t{++} expression evaluation. The default
7653 is on. For overloaded functions, @value{GDBN} evaluates the arguments
7654 and searches for a function whose signature matches the argument types,
7655 using the standard C@t{++} conversion rules (see @ref{C plus plus expressions, ,C@t{++}
7656 expressions}, for details). If it cannot find a match, it emits a
7659 @item set overload-resolution off
7660 Disable overload resolution for C@t{++} expression evaluation. For
7661 overloaded functions that are not class member functions, @value{GDBN}
7662 chooses the first function of the specified name that it finds in the
7663 symbol table, whether or not its arguments are of the correct type. For
7664 overloaded functions that are class member functions, @value{GDBN}
7665 searches for a function whose signature @emph{exactly} matches the
7668 @item @r{Overloaded symbol names}
7669 You can specify a particular definition of an overloaded symbol, using
7670 the same notation that is used to declare such symbols in C@t{++}: type
7671 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
7672 also use the @value{GDBN} command-line word completion facilities to list the
7673 available choices, or to finish the type list for you.
7674 @xref{Completion,, Command completion}, for details on how to do this.
7678 @subsection Modula-2
7680 @cindex Modula-2, @value{GDBN} support
7682 The extensions made to @value{GDBN} to support Modula-2 only support
7683 output from the @sc{gnu} Modula-2 compiler (which is currently being
7684 developed). Other Modula-2 compilers are not currently supported, and
7685 attempting to debug executables produced by them is most likely
7686 to give an error as @value{GDBN} reads in the executable's symbol
7689 @cindex expressions in Modula-2
7691 * M2 Operators:: Built-in operators
7692 * Built-In Func/Proc:: Built-in functions and procedures
7693 * M2 Constants:: Modula-2 constants
7694 * M2 Defaults:: Default settings for Modula-2
7695 * Deviations:: Deviations from standard Modula-2
7696 * M2 Checks:: Modula-2 type and range checks
7697 * M2 Scope:: The scope operators @code{::} and @code{.}
7698 * GDB/M2:: @value{GDBN} and Modula-2
7702 @subsubsection Operators
7703 @cindex Modula-2 operators
7705 Operators must be defined on values of specific types. For instance,
7706 @code{+} is defined on numbers, but not on structures. Operators are
7707 often defined on groups of types. For the purposes of Modula-2, the
7708 following definitions hold:
7713 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
7717 @emph{Character types} consist of @code{CHAR} and its subranges.
7720 @emph{Floating-point types} consist of @code{REAL}.
7723 @emph{Pointer types} consist of anything declared as @code{POINTER TO
7727 @emph{Scalar types} consist of all of the above.
7730 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
7733 @emph{Boolean types} consist of @code{BOOLEAN}.
7737 The following operators are supported, and appear in order of
7738 increasing precedence:
7742 Function argument or array index separator.
7745 Assignment. The value of @var{var} @code{:=} @var{value} is
7749 Less than, greater than on integral, floating-point, or enumerated
7753 Less than or equal to, greater than or equal to
7754 on integral, floating-point and enumerated types, or set inclusion on
7755 set types. Same precedence as @code{<}.
7757 @item =@r{, }<>@r{, }#
7758 Equality and two ways of expressing inequality, valid on scalar types.
7759 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
7760 available for inequality, since @code{#} conflicts with the script
7764 Set membership. Defined on set types and the types of their members.
7765 Same precedence as @code{<}.
7768 Boolean disjunction. Defined on boolean types.
7771 Boolean conjunction. Defined on boolean types.
7774 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
7777 Addition and subtraction on integral and floating-point types, or union
7778 and difference on set types.
7781 Multiplication on integral and floating-point types, or set intersection
7785 Division on floating-point types, or symmetric set difference on set
7786 types. Same precedence as @code{*}.
7789 Integer division and remainder. Defined on integral types. Same
7790 precedence as @code{*}.
7793 Negative. Defined on @code{INTEGER} and @code{REAL} data.
7796 Pointer dereferencing. Defined on pointer types.
7799 Boolean negation. Defined on boolean types. Same precedence as
7803 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
7804 precedence as @code{^}.
7807 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
7810 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
7814 @value{GDBN} and Modula-2 scope operators.
7818 @emph{Warning:} Sets and their operations are not yet supported, so @value{GDBN}
7819 treats the use of the operator @code{IN}, or the use of operators
7820 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
7821 @code{<=}, and @code{>=} on sets as an error.
7825 @node Built-In Func/Proc
7826 @subsubsection Built-in functions and procedures
7827 @cindex Modula-2 built-ins
7829 Modula-2 also makes available several built-in procedures and functions.
7830 In describing these, the following metavariables are used:
7835 represents an @code{ARRAY} variable.
7838 represents a @code{CHAR} constant or variable.
7841 represents a variable or constant of integral type.
7844 represents an identifier that belongs to a set. Generally used in the
7845 same function with the metavariable @var{s}. The type of @var{s} should
7846 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
7849 represents a variable or constant of integral or floating-point type.
7852 represents a variable or constant of floating-point type.
7858 represents a variable.
7861 represents a variable or constant of one of many types. See the
7862 explanation of the function for details.
7865 All Modula-2 built-in procedures also return a result, described below.
7869 Returns the absolute value of @var{n}.
7872 If @var{c} is a lower case letter, it returns its upper case
7873 equivalent, otherwise it returns its argument.
7876 Returns the character whose ordinal value is @var{i}.
7879 Decrements the value in the variable @var{v} by one. Returns the new value.
7881 @item DEC(@var{v},@var{i})
7882 Decrements the value in the variable @var{v} by @var{i}. Returns the
7885 @item EXCL(@var{m},@var{s})
7886 Removes the element @var{m} from the set @var{s}. Returns the new
7889 @item FLOAT(@var{i})
7890 Returns the floating point equivalent of the integer @var{i}.
7893 Returns the index of the last member of @var{a}.
7896 Increments the value in the variable @var{v} by one. Returns the new value.
7898 @item INC(@var{v},@var{i})
7899 Increments the value in the variable @var{v} by @var{i}. Returns the
7902 @item INCL(@var{m},@var{s})
7903 Adds the element @var{m} to the set @var{s} if it is not already
7904 there. Returns the new set.
7907 Returns the maximum value of the type @var{t}.
7910 Returns the minimum value of the type @var{t}.
7913 Returns boolean TRUE if @var{i} is an odd number.
7916 Returns the ordinal value of its argument. For example, the ordinal
7917 value of a character is its @sc{ascii} value (on machines supporting the
7918 @sc{ascii} character set). @var{x} must be of an ordered type, which include
7919 integral, character and enumerated types.
7922 Returns the size of its argument. @var{x} can be a variable or a type.
7924 @item TRUNC(@var{r})
7925 Returns the integral part of @var{r}.
7927 @item VAL(@var{t},@var{i})
7928 Returns the member of the type @var{t} whose ordinal value is @var{i}.
7932 @emph{Warning:} Sets and their operations are not yet supported, so
7933 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
7937 @cindex Modula-2 constants
7939 @subsubsection Constants
7941 @value{GDBN} allows you to express the constants of Modula-2 in the following
7947 Integer constants are simply a sequence of digits. When used in an
7948 expression, a constant is interpreted to be type-compatible with the
7949 rest of the expression. Hexadecimal integers are specified by a
7950 trailing @samp{H}, and octal integers by a trailing @samp{B}.
7953 Floating point constants appear as a sequence of digits, followed by a
7954 decimal point and another sequence of digits. An optional exponent can
7955 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
7956 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
7957 digits of the floating point constant must be valid decimal (base 10)
7961 Character constants consist of a single character enclosed by a pair of
7962 like quotes, either single (@code{'}) or double (@code{"}). They may
7963 also be expressed by their ordinal value (their @sc{ascii} value, usually)
7964 followed by a @samp{C}.
7967 String constants consist of a sequence of characters enclosed by a
7968 pair of like quotes, either single (@code{'}) or double (@code{"}).
7969 Escape sequences in the style of C are also allowed. @xref{C
7970 Constants, ,C and C@t{++} constants}, for a brief explanation of escape
7974 Enumerated constants consist of an enumerated identifier.
7977 Boolean constants consist of the identifiers @code{TRUE} and
7981 Pointer constants consist of integral values only.
7984 Set constants are not yet supported.
7988 @subsubsection Modula-2 defaults
7989 @cindex Modula-2 defaults
7991 If type and range checking are set automatically by @value{GDBN}, they
7992 both default to @code{on} whenever the working language changes to
7993 Modula-2. This happens regardless of whether you or @value{GDBN}
7994 selected the working language.
7996 If you allow @value{GDBN} to set the language automatically, then entering
7997 code compiled from a file whose name ends with @file{.mod} sets the
7998 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN} set
7999 the language automatically}, for further details.
8002 @subsubsection Deviations from standard Modula-2
8003 @cindex Modula-2, deviations from
8005 A few changes have been made to make Modula-2 programs easier to debug.
8006 This is done primarily via loosening its type strictness:
8010 Unlike in standard Modula-2, pointer constants can be formed by
8011 integers. This allows you to modify pointer variables during
8012 debugging. (In standard Modula-2, the actual address contained in a
8013 pointer variable is hidden from you; it can only be modified
8014 through direct assignment to another pointer variable or expression that
8015 returned a pointer.)
8018 C escape sequences can be used in strings and characters to represent
8019 non-printable characters. @value{GDBN} prints out strings with these
8020 escape sequences embedded. Single non-printable characters are
8021 printed using the @samp{CHR(@var{nnn})} format.
8024 The assignment operator (@code{:=}) returns the value of its right-hand
8028 All built-in procedures both modify @emph{and} return their argument.
8032 @subsubsection Modula-2 type and range checks
8033 @cindex Modula-2 checks
8036 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
8039 @c FIXME remove warning when type/range checks added
8041 @value{GDBN} considers two Modula-2 variables type equivalent if:
8045 They are of types that have been declared equivalent via a @code{TYPE
8046 @var{t1} = @var{t2}} statement
8049 They have been declared on the same line. (Note: This is true of the
8050 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
8053 As long as type checking is enabled, any attempt to combine variables
8054 whose types are not equivalent is an error.
8056 Range checking is done on all mathematical operations, assignment, array
8057 index bounds, and all built-in functions and procedures.
8060 @subsubsection The scope operators @code{::} and @code{.}
8062 @cindex @code{.}, Modula-2 scope operator
8063 @cindex colon, doubled as scope operator
8065 @vindex colon-colon@r{, in Modula-2}
8066 @c Info cannot handle :: but TeX can.
8069 @vindex ::@r{, in Modula-2}
8072 There are a few subtle differences between the Modula-2 scope operator
8073 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
8078 @var{module} . @var{id}
8079 @var{scope} :: @var{id}
8083 where @var{scope} is the name of a module or a procedure,
8084 @var{module} the name of a module, and @var{id} is any declared
8085 identifier within your program, except another module.
8087 Using the @code{::} operator makes @value{GDBN} search the scope
8088 specified by @var{scope} for the identifier @var{id}. If it is not
8089 found in the specified scope, then @value{GDBN} searches all scopes
8090 enclosing the one specified by @var{scope}.
8092 Using the @code{.} operator makes @value{GDBN} search the current scope for
8093 the identifier specified by @var{id} that was imported from the
8094 definition module specified by @var{module}. With this operator, it is
8095 an error if the identifier @var{id} was not imported from definition
8096 module @var{module}, or if @var{id} is not an identifier in
8100 @subsubsection @value{GDBN} and Modula-2
8102 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
8103 Five subcommands of @code{set print} and @code{show print} apply
8104 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
8105 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
8106 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
8107 analogue in Modula-2.
8109 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
8110 with any language, is not useful with Modula-2. Its
8111 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
8112 created in Modula-2 as they can in C or C@t{++}. However, because an
8113 address can be specified by an integral constant, the construct
8114 @samp{@{@var{type}@}@var{adrexp}} is still useful.
8116 @cindex @code{#} in Modula-2
8117 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
8118 interpreted as the beginning of a comment. Use @code{<>} instead.
8123 The extensions made to @value{GDBN} to support Chill only support output
8124 from the @sc{gnu} Chill compiler. Other Chill compilers are not currently
8125 supported, and attempting to debug executables produced by them is most
8126 likely to give an error as @value{GDBN} reads in the executable's symbol
8129 @c This used to say "... following Chill related topics ...", but since
8130 @c menus are not shown in the printed manual, it would look awkward.
8131 This section covers the Chill related topics and the features
8132 of @value{GDBN} which support these topics.
8135 * How modes are displayed:: How modes are displayed
8136 * Locations:: Locations and their accesses
8137 * Values and their Operations:: Values and their Operations
8138 * Chill type and range checks::
8142 @node How modes are displayed
8143 @subsubsection How modes are displayed
8145 The Chill Datatype- (Mode) support of @value{GDBN} is directly related
8146 with the functionality of the @sc{gnu} Chill compiler, and therefore deviates
8147 slightly from the standard specification of the Chill language. The
8150 @c FIXME: this @table's contents effectively disable @code by using @r
8151 @c on every @item. So why does it need @code?
8153 @item @r{@emph{Discrete modes:}}
8156 @emph{Integer Modes} which are predefined by @code{BYTE, UBYTE, INT,
8159 @emph{Boolean Mode} which is predefined by @code{BOOL},
8161 @emph{Character Mode} which is predefined by @code{CHAR},
8163 @emph{Set Mode} which is displayed by the keyword @code{SET}.
8165 (@value{GDBP}) ptype x
8166 type = SET (karli = 10, susi = 20, fritzi = 100)
8168 If the type is an unnumbered set the set element values are omitted.
8170 @emph{Range Mode} which is displayed by
8172 @code{type = <basemode>(<lower bound> : <upper bound>)}
8174 where @code{<lower bound>, <upper bound>} can be of any discrete literal
8175 expression (e.g. set element names).
8178 @item @r{@emph{Powerset Mode:}}
8179 A Powerset Mode is displayed by the keyword @code{POWERSET} followed by
8180 the member mode of the powerset. The member mode can be any discrete mode.
8182 (@value{GDBP}) ptype x
8183 type = POWERSET SET (egon, hugo, otto)
8186 @item @r{@emph{Reference Modes:}}
8189 @emph{Bound Reference Mode} which is displayed by the keyword @code{REF}
8190 followed by the mode name to which the reference is bound.
8192 @emph{Free Reference Mode} which is displayed by the keyword @code{PTR}.
8195 @item @r{@emph{Procedure mode}}
8196 The procedure mode is displayed by @code{type = PROC(<parameter list>)
8197 <return mode> EXCEPTIONS (<exception list>)}. The @code{<parameter
8198 list>} is a list of the parameter modes. @code{<return mode>} indicates
8199 the mode of the result of the procedure if any. The exceptionlist lists
8200 all possible exceptions which can be raised by the procedure.
8203 @item @r{@emph{Instance mode}}
8204 The instance mode is represented by a structure, which has a static
8205 type, and is therefore not really of interest.
8208 @item @r{@emph{Synchronization Modes:}}
8211 @emph{Event Mode} which is displayed by
8213 @code{EVENT (<event length>)}
8215 where @code{(<event length>)} is optional.
8217 @emph{Buffer Mode} which is displayed by
8219 @code{BUFFER (<buffer length>)<buffer element mode>}
8221 where @code{(<buffer length>)} is optional.
8224 @item @r{@emph{Timing Modes:}}
8227 @emph{Duration Mode} which is predefined by @code{DURATION}
8229 @emph{Absolute Time Mode} which is predefined by @code{TIME}
8232 @item @r{@emph{Real Modes:}}
8233 Real Modes are predefined with @code{REAL} and @code{LONG_REAL}.
8235 @item @r{@emph{String Modes:}}
8238 @emph{Character String Mode} which is displayed by
8240 @code{CHARS(<string length>)}
8242 followed by the keyword @code{VARYING} if the String Mode is a varying
8245 @emph{Bit String Mode} which is displayed by
8252 @item @r{@emph{Array Mode:}}
8253 The Array Mode is displayed by the keyword @code{ARRAY(<range>)}
8254 followed by the element mode (which may in turn be an array mode).
8256 (@value{GDBP}) ptype x
8259 SET (karli = 10, susi = 20, fritzi = 100)
8262 @item @r{@emph{Structure Mode}}
8263 The Structure mode is displayed by the keyword @code{STRUCT(<field
8264 list>)}. The @code{<field list>} consists of names and modes of fields
8265 of the structure. Variant structures have the keyword @code{CASE <field>
8266 OF <variant fields> ESAC} in their field list. Since the current version
8267 of the GNU Chill compiler doesn't implement tag processing (no runtime
8268 checks of variant fields, and therefore no debugging info), the output
8269 always displays all variant fields.
8271 (@value{GDBP}) ptype str
8286 @subsubsection Locations and their accesses
8288 A location in Chill is an object which can contain values.
8290 A value of a location is generally accessed by the (declared) name of
8291 the location. The output conforms to the specification of values in
8292 Chill programs. How values are specified
8293 is the topic of the next section, @ref{Values and their Operations}.
8295 The pseudo-location @code{RESULT} (or @code{result}) can be used to
8296 display or change the result of a currently-active procedure:
8303 This does the same as the Chill action @code{RESULT EXPR} (which
8304 is not available in @value{GDBN}).
8306 Values of reference mode locations are printed by @code{PTR(<hex
8307 value>)} in case of a free reference mode, and by @code{(REF <reference
8308 mode>) (<hex-value>)} in case of a bound reference. @code{<hex value>}
8309 represents the address where the reference points to. To access the
8310 value of the location referenced by the pointer, use the dereference
8313 Values of procedure mode locations are displayed by
8316 (<argument modes> ) <return mode> @} <address> <name of procedure
8319 @code{<argument modes>} is a list of modes according to the parameter
8320 specification of the procedure and @code{<address>} shows the address of
8324 Locations of instance modes are displayed just like a structure with two
8325 fields specifying the @emph{process type} and the @emph{copy number} of
8326 the investigated instance location@footnote{This comes from the current
8327 implementation of instances. They are implemented as a structure (no
8328 na). The output should be something like @code{[<name of the process>;
8329 <instance number>]}.}. The field names are @code{__proc_type} and
8332 Locations of synchronization modes are displayed like a structure with
8333 the field name @code{__event_data} in case of a event mode location, and
8334 like a structure with the field @code{__buffer_data} in case of a buffer
8335 mode location (refer to previous paragraph).
8337 Structure Mode locations are printed by @code{[.<field name>: <value>,
8338 ...]}. The @code{<field name>} corresponds to the structure mode
8339 definition and the layout of @code{<value>} varies depending of the mode
8340 of the field. If the investigated structure mode location is of variant
8341 structure mode, the variant parts of the structure are enclosed in curled
8342 braces (@samp{@{@}}). Fields enclosed by @samp{@{,@}} are residing
8343 on the same memory location and represent the current values of the
8344 memory location in their specific modes. Since no tag processing is done
8345 all variants are displayed. A variant field is printed by
8346 @code{(<variant name>) = .<field name>: <value>}. (who implements the
8349 (@value{GDBP}) print str1 $4 = [.as: 0, .bs: karli, .<TAG>: { (karli) =
8350 [.cs: []], (susi) = [.ds: susi]}]
8354 Substructures of string mode-, array mode- or structure mode-values
8355 (e.g. array slices, fields of structure locations) are accessed using
8356 certain operations which are described in the next section, @ref{Values
8357 and their Operations}.
8359 A location value may be interpreted as having a different mode using the
8360 location conversion. This mode conversion is written as @code{<mode
8361 name>(<location>)}. The user has to consider that the sizes of the modes
8362 have to be equal otherwise an error occurs. Furthermore, no range
8363 checking of the location against the destination mode is performed, and
8364 therefore the result can be quite confusing.
8367 (@value{GDBP}) print int (s(3 up 4)) XXX TO be filled in !! XXX
8370 @node Values and their Operations
8371 @subsubsection Values and their Operations
8373 Values are used to alter locations, to investigate complex structures in
8374 more detail or to filter relevant information out of a large amount of
8375 data. There are several (mode dependent) operations defined which enable
8376 such investigations. These operations are not only applicable to
8377 constant values but also to locations, which can become quite useful
8378 when debugging complex structures. During parsing the command line
8379 (e.g. evaluating an expression) @value{GDBN} treats location names as
8380 the values behind these locations.
8382 This section describes how values have to be specified and which
8383 operations are legal to be used with such values.
8386 @item Literal Values
8387 Literal values are specified in the same manner as in @sc{gnu} Chill programs.
8388 For detailed specification refer to the @sc{gnu} Chill implementation Manual
8390 @c FIXME: if the Chill Manual is a Texinfo documents, the above should
8391 @c be converted to a @ref.
8396 @emph{Integer Literals} are specified in the same manner as in Chill
8397 programs (refer to the Chill Standard z200/88 chpt 5.2.4.2)
8399 @emph{Boolean Literals} are defined by @code{TRUE} and @code{FALSE}.
8401 @emph{Character Literals} are defined by @code{'<character>'}. (e.g.
8404 @emph{Set Literals} are defined by a name which was specified in a set
8405 mode. The value delivered by a Set Literal is the set value. This is
8406 comparable to an enumeration in C/C@t{++} language.
8408 @emph{Emptiness Literal} is predefined by @code{NULL}. The value of the
8409 emptiness literal delivers either the empty reference value, the empty
8410 procedure value or the empty instance value.
8413 @emph{Character String Literals} are defined by a sequence of characters
8414 enclosed in single- or double quotes. If a single- or double quote has
8415 to be part of the string literal it has to be stuffed (specified twice).
8417 @emph{Bitstring Literals} are specified in the same manner as in Chill
8418 programs (refer z200/88 chpt 5.2.4.8).
8420 @emph{Floating point literals} are specified in the same manner as in
8421 (gnu-)Chill programs (refer @sc{gnu} Chill implementation Manual chapter 1.5).
8426 A tuple is specified by @code{<mode name>[<tuple>]}, where @code{<mode
8427 name>} can be omitted if the mode of the tuple is unambiguous. This
8428 unambiguity is derived from the context of a evaluated expression.
8429 @code{<tuple>} can be one of the following:
8432 @item @emph{Powerset Tuple}
8433 @item @emph{Array Tuple}
8434 @item @emph{Structure Tuple}
8435 Powerset tuples, array tuples and structure tuples are specified in the
8436 same manner as in Chill programs refer to z200/88 chpt 5.2.5.
8439 @item String Element Value
8440 A string element value is specified by
8442 @code{<string value>(<index>)}
8444 where @code{<index>} is a integer expression. It delivers a character
8445 value which is equivalent to the character indexed by @code{<index>} in
8448 @item String Slice Value
8449 A string slice value is specified by @code{<string value>(<slice
8450 spec>)}, where @code{<slice spec>} can be either a range of integer
8451 expressions or specified by @code{<start expr> up <size>}.
8452 @code{<size>} denotes the number of elements which the slice contains.
8453 The delivered value is a string value, which is part of the specified
8456 @item Array Element Values
8457 An array element value is specified by @code{<array value>(<expr>)} and
8458 delivers a array element value of the mode of the specified array.
8460 @item Array Slice Values
8461 An array slice is specified by @code{<array value>(<slice spec>)}, where
8462 @code{<slice spec>} can be either a range specified by expressions or by
8463 @code{<start expr> up <size>}. @code{<size>} denotes the number of
8464 arrayelements the slice contains. The delivered value is an array value
8465 which is part of the specified array.
8467 @item Structure Field Values
8468 A structure field value is derived by @code{<structure value>.<field
8469 name>}, where @code{<field name>} indicates the name of a field specified
8470 in the mode definition of the structure. The mode of the delivered value
8471 corresponds to this mode definition in the structure definition.
8473 @item Procedure Call Value
8474 The procedure call value is derived from the return value of the
8475 procedure@footnote{If a procedure call is used for instance in an
8476 expression, then this procedure is called with all its side
8477 effects. This can lead to confusing results if used carelessly.}.
8479 Values of duration mode locations are represented by @code{ULONG} literals.
8481 Values of time mode locations appear as
8483 @code{TIME(<secs>:<nsecs>)}
8488 This is not implemented yet:
8489 @item Built-in Value
8491 The following built in functions are provided:
8503 @item @code{UPPER()}
8504 @item @code{LOWER()}
8505 @item @code{LENGTH()}
8509 @item @code{ARCSIN()}
8510 @item @code{ARCCOS()}
8511 @item @code{ARCTAN()}
8518 For a detailed description refer to the GNU Chill implementation manual
8522 @item Zero-adic Operator Value
8523 The zero-adic operator value is derived from the instance value for the
8524 current active process.
8526 @item Expression Values
8527 The value delivered by an expression is the result of the evaluation of
8528 the specified expression. If there are error conditions (mode
8529 incompatibility, etc.) the evaluation of expressions is aborted with a
8530 corresponding error message. Expressions may be parenthesised which
8531 causes the evaluation of this expression before any other expression
8532 which uses the result of the parenthesised expression. The following
8533 operators are supported by @value{GDBN}:
8536 @item @code{OR, ORIF, XOR}
8537 @itemx @code{AND, ANDIF}
8539 Logical operators defined over operands of boolean mode.
8542 Equality and inequality operators defined over all modes.
8546 Relational operators defined over predefined modes.
8549 @itemx @code{*, /, MOD, REM}
8550 Arithmetic operators defined over predefined modes.
8553 Change sign operator.
8556 String concatenation operator.
8559 String repetition operator.
8562 Referenced location operator which can be used either to take the
8563 address of a location (@code{->loc}), or to dereference a reference
8564 location (@code{loc->}).
8566 @item @code{OR, XOR}
8569 Powerset and bitstring operators.
8573 Powerset inclusion operators.
8576 Membership operator.
8580 @node Chill type and range checks
8581 @subsubsection Chill type and range checks
8583 @value{GDBN} considers two Chill variables mode equivalent if the sizes
8584 of the two modes are equal. This rule applies recursively to more
8585 complex datatypes which means that complex modes are treated
8586 equivalent if all element modes (which also can be complex modes like
8587 structures, arrays, etc.) have the same size.
8589 Range checking is done on all mathematical operations, assignment, array
8590 index bounds and all built in procedures.
8592 Strong type checks are forced using the @value{GDBN} command @code{set
8593 check strong}. This enforces strong type and range checks on all
8594 operations where Chill constructs are used (expressions, built in
8595 functions, etc.) in respect to the semantics as defined in the z.200
8596 language specification.
8598 All checks can be disabled by the @value{GDBN} command @code{set check
8602 @c Deviations from the Chill Standard Z200/88
8603 see last paragraph ?
8606 @node Chill defaults
8607 @subsubsection Chill defaults
8609 If type and range checking are set automatically by @value{GDBN}, they
8610 both default to @code{on} whenever the working language changes to
8611 Chill. This happens regardless of whether you or @value{GDBN}
8612 selected the working language.
8614 If you allow @value{GDBN} to set the language automatically, then entering
8615 code compiled from a file whose name ends with @file{.ch} sets the
8616 working language to Chill. @xref{Automatically, ,Having @value{GDBN} set
8617 the language automatically}, for further details.
8620 @chapter Examining the Symbol Table
8622 The commands described in this chapter allow you to inquire about the
8623 symbols (names of variables, functions and types) defined in your
8624 program. This information is inherent in the text of your program and
8625 does not change as your program executes. @value{GDBN} finds it in your
8626 program's symbol table, in the file indicated when you started @value{GDBN}
8627 (@pxref{File Options, ,Choosing files}), or by one of the
8628 file-management commands (@pxref{Files, ,Commands to specify files}).
8630 @cindex symbol names
8631 @cindex names of symbols
8632 @cindex quoting names
8633 Occasionally, you may need to refer to symbols that contain unusual
8634 characters, which @value{GDBN} ordinarily treats as word delimiters. The
8635 most frequent case is in referring to static variables in other
8636 source files (@pxref{Variables,,Program variables}). File names
8637 are recorded in object files as debugging symbols, but @value{GDBN} would
8638 ordinarily parse a typical file name, like @file{foo.c}, as the three words
8639 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
8640 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
8647 looks up the value of @code{x} in the scope of the file @file{foo.c}.
8650 @kindex info address
8651 @cindex address of a symbol
8652 @item info address @var{symbol}
8653 Describe where the data for @var{symbol} is stored. For a register
8654 variable, this says which register it is kept in. For a non-register
8655 local variable, this prints the stack-frame offset at which the variable
8658 Note the contrast with @samp{print &@var{symbol}}, which does not work
8659 at all for a register variable, and for a stack local variable prints
8660 the exact address of the current instantiation of the variable.
8663 @cindex symbol from address
8664 @item info symbol @var{addr}
8665 Print the name of a symbol which is stored at the address @var{addr}.
8666 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
8667 nearest symbol and an offset from it:
8670 (@value{GDBP}) info symbol 0x54320
8671 _initialize_vx + 396 in section .text
8675 This is the opposite of the @code{info address} command. You can use
8676 it to find out the name of a variable or a function given its address.
8679 @item whatis @var{expr}
8680 Print the data type of expression @var{expr}. @var{expr} is not
8681 actually evaluated, and any side-effecting operations (such as
8682 assignments or function calls) inside it do not take place.
8683 @xref{Expressions, ,Expressions}.
8686 Print the data type of @code{$}, the last value in the value history.
8689 @item ptype @var{typename}
8690 Print a description of data type @var{typename}. @var{typename} may be
8691 the name of a type, or for C code it may have the form @samp{class
8692 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
8693 @var{union-tag}} or @samp{enum @var{enum-tag}}.
8695 @item ptype @var{expr}
8697 Print a description of the type of expression @var{expr}. @code{ptype}
8698 differs from @code{whatis} by printing a detailed description, instead
8699 of just the name of the type.
8701 For example, for this variable declaration:
8704 struct complex @{double real; double imag;@} v;
8708 the two commands give this output:
8712 (@value{GDBP}) whatis v
8713 type = struct complex
8714 (@value{GDBP}) ptype v
8715 type = struct complex @{
8723 As with @code{whatis}, using @code{ptype} without an argument refers to
8724 the type of @code{$}, the last value in the value history.
8727 @item info types @var{regexp}
8729 Print a brief description of all types whose names match @var{regexp}
8730 (or all types in your program, if you supply no argument). Each
8731 complete typename is matched as though it were a complete line; thus,
8732 @samp{i type value} gives information on all types in your program whose
8733 names include the string @code{value}, but @samp{i type ^value$} gives
8734 information only on types whose complete name is @code{value}.
8736 This command differs from @code{ptype} in two ways: first, like
8737 @code{whatis}, it does not print a detailed description; second, it
8738 lists all source files where a type is defined.
8741 @cindex local variables
8742 @item info scope @var{addr}
8743 List all the variables local to a particular scope. This command
8744 accepts a location---a function name, a source line, or an address
8745 preceded by a @samp{*}, and prints all the variables local to the
8746 scope defined by that location. For example:
8749 (@value{GDBP}) @b{info scope command_line_handler}
8750 Scope for command_line_handler:
8751 Symbol rl is an argument at stack/frame offset 8, length 4.
8752 Symbol linebuffer is in static storage at address 0x150a18, length 4.
8753 Symbol linelength is in static storage at address 0x150a1c, length 4.
8754 Symbol p is a local variable in register $esi, length 4.
8755 Symbol p1 is a local variable in register $ebx, length 4.
8756 Symbol nline is a local variable in register $edx, length 4.
8757 Symbol repeat is a local variable at frame offset -8, length 4.
8761 This command is especially useful for determining what data to collect
8762 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
8767 Show the name of the current source file---that is, the source file for
8768 the function containing the current point of execution---and the language
8771 @kindex info sources
8773 Print the names of all source files in your program for which there is
8774 debugging information, organized into two lists: files whose symbols
8775 have already been read, and files whose symbols will be read when needed.
8777 @kindex info functions
8778 @item info functions
8779 Print the names and data types of all defined functions.
8781 @item info functions @var{regexp}
8782 Print the names and data types of all defined functions
8783 whose names contain a match for regular expression @var{regexp}.
8784 Thus, @samp{info fun step} finds all functions whose names
8785 include @code{step}; @samp{info fun ^step} finds those whose names
8786 start with @code{step}. If a function name contains characters
8787 that conflict with the regular expression language (eg.
8788 @samp{operator*()}), they may be quoted with a backslash.
8790 @kindex info variables
8791 @item info variables
8792 Print the names and data types of all variables that are declared
8793 outside of functions (i.e., excluding local variables).
8795 @item info variables @var{regexp}
8796 Print the names and data types of all variables (except for local
8797 variables) whose names contain a match for regular expression
8801 This was never implemented.
8802 @kindex info methods
8804 @itemx info methods @var{regexp}
8805 The @code{info methods} command permits the user to examine all defined
8806 methods within C@t{++} program, or (with the @var{regexp} argument) a
8807 specific set of methods found in the various C@t{++} classes. Many
8808 C@t{++} classes provide a large number of methods. Thus, the output
8809 from the @code{ptype} command can be overwhelming and hard to use. The
8810 @code{info-methods} command filters the methods, printing only those
8811 which match the regular-expression @var{regexp}.
8814 @cindex reloading symbols
8815 Some systems allow individual object files that make up your program to
8816 be replaced without stopping and restarting your program. For example,
8817 in VxWorks you can simply recompile a defective object file and keep on
8818 running. If you are running on one of these systems, you can allow
8819 @value{GDBN} to reload the symbols for automatically relinked modules:
8822 @kindex set symbol-reloading
8823 @item set symbol-reloading on
8824 Replace symbol definitions for the corresponding source file when an
8825 object file with a particular name is seen again.
8827 @item set symbol-reloading off
8828 Do not replace symbol definitions when encountering object files of the
8829 same name more than once. This is the default state; if you are not
8830 running on a system that permits automatic relinking of modules, you
8831 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
8832 may discard symbols when linking large programs, that may contain
8833 several modules (from different directories or libraries) with the same
8836 @kindex show symbol-reloading
8837 @item show symbol-reloading
8838 Show the current @code{on} or @code{off} setting.
8841 @kindex set opaque-type-resolution
8842 @item set opaque-type-resolution on
8843 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
8844 declared as a pointer to a @code{struct}, @code{class}, or
8845 @code{union}---for example, @code{struct MyType *}---that is used in one
8846 source file although the full declaration of @code{struct MyType} is in
8847 another source file. The default is on.
8849 A change in the setting of this subcommand will not take effect until
8850 the next time symbols for a file are loaded.
8852 @item set opaque-type-resolution off
8853 Tell @value{GDBN} not to resolve opaque types. In this case, the type
8854 is printed as follows:
8856 @{<no data fields>@}
8859 @kindex show opaque-type-resolution
8860 @item show opaque-type-resolution
8861 Show whether opaque types are resolved or not.
8863 @kindex maint print symbols
8865 @kindex maint print psymbols
8866 @cindex partial symbol dump
8867 @item maint print symbols @var{filename}
8868 @itemx maint print psymbols @var{filename}
8869 @itemx maint print msymbols @var{filename}
8870 Write a dump of debugging symbol data into the file @var{filename}.
8871 These commands are used to debug the @value{GDBN} symbol-reading code. Only
8872 symbols with debugging data are included. If you use @samp{maint print
8873 symbols}, @value{GDBN} includes all the symbols for which it has already
8874 collected full details: that is, @var{filename} reflects symbols for
8875 only those files whose symbols @value{GDBN} has read. You can use the
8876 command @code{info sources} to find out which files these are. If you
8877 use @samp{maint print psymbols} instead, the dump shows information about
8878 symbols that @value{GDBN} only knows partially---that is, symbols defined in
8879 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
8880 @samp{maint print msymbols} dumps just the minimal symbol information
8881 required for each object file from which @value{GDBN} has read some symbols.
8882 @xref{Files, ,Commands to specify files}, for a discussion of how
8883 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
8887 @chapter Altering Execution
8889 Once you think you have found an error in your program, you might want to
8890 find out for certain whether correcting the apparent error would lead to
8891 correct results in the rest of the run. You can find the answer by
8892 experiment, using the @value{GDBN} features for altering execution of the
8895 For example, you can store new values into variables or memory
8896 locations, give your program a signal, restart it at a different
8897 address, or even return prematurely from a function.
8900 * Assignment:: Assignment to variables
8901 * Jumping:: Continuing at a different address
8902 * Signaling:: Giving your program a signal
8903 * Returning:: Returning from a function
8904 * Calling:: Calling your program's functions
8905 * Patching:: Patching your program
8909 @section Assignment to variables
8912 @cindex setting variables
8913 To alter the value of a variable, evaluate an assignment expression.
8914 @xref{Expressions, ,Expressions}. For example,
8921 stores the value 4 into the variable @code{x}, and then prints the
8922 value of the assignment expression (which is 4).
8923 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
8924 information on operators in supported languages.
8926 @kindex set variable
8927 @cindex variables, setting
8928 If you are not interested in seeing the value of the assignment, use the
8929 @code{set} command instead of the @code{print} command. @code{set} is
8930 really the same as @code{print} except that the expression's value is
8931 not printed and is not put in the value history (@pxref{Value History,
8932 ,Value history}). The expression is evaluated only for its effects.
8934 If the beginning of the argument string of the @code{set} command
8935 appears identical to a @code{set} subcommand, use the @code{set
8936 variable} command instead of just @code{set}. This command is identical
8937 to @code{set} except for its lack of subcommands. For example, if your
8938 program has a variable @code{width}, you get an error if you try to set
8939 a new value with just @samp{set width=13}, because @value{GDBN} has the
8940 command @code{set width}:
8943 (@value{GDBP}) whatis width
8945 (@value{GDBP}) p width
8947 (@value{GDBP}) set width=47
8948 Invalid syntax in expression.
8952 The invalid expression, of course, is @samp{=47}. In
8953 order to actually set the program's variable @code{width}, use
8956 (@value{GDBP}) set var width=47
8959 Because the @code{set} command has many subcommands that can conflict
8960 with the names of program variables, it is a good idea to use the
8961 @code{set variable} command instead of just @code{set}. For example, if
8962 your program has a variable @code{g}, you run into problems if you try
8963 to set a new value with just @samp{set g=4}, because @value{GDBN} has
8964 the command @code{set gnutarget}, abbreviated @code{set g}:
8968 (@value{GDBP}) whatis g
8972 (@value{GDBP}) set g=4
8976 The program being debugged has been started already.
8977 Start it from the beginning? (y or n) y
8978 Starting program: /home/smith/cc_progs/a.out
8979 "/home/smith/cc_progs/a.out": can't open to read symbols:
8981 (@value{GDBP}) show g
8982 The current BFD target is "=4".
8987 The program variable @code{g} did not change, and you silently set the
8988 @code{gnutarget} to an invalid value. In order to set the variable
8992 (@value{GDBP}) set var g=4
8995 @value{GDBN} allows more implicit conversions in assignments than C; you can
8996 freely store an integer value into a pointer variable or vice versa,
8997 and you can convert any structure to any other structure that is the
8998 same length or shorter.
8999 @comment FIXME: how do structs align/pad in these conversions?
9000 @comment /doc@cygnus.com 18dec1990
9002 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
9003 construct to generate a value of specified type at a specified address
9004 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
9005 to memory location @code{0x83040} as an integer (which implies a certain size
9006 and representation in memory), and
9009 set @{int@}0x83040 = 4
9013 stores the value 4 into that memory location.
9016 @section Continuing at a different address
9018 Ordinarily, when you continue your program, you do so at the place where
9019 it stopped, with the @code{continue} command. You can instead continue at
9020 an address of your own choosing, with the following commands:
9024 @item jump @var{linespec}
9025 Resume execution at line @var{linespec}. Execution stops again
9026 immediately if there is a breakpoint there. @xref{List, ,Printing
9027 source lines}, for a description of the different forms of
9028 @var{linespec}. It is common practice to use the @code{tbreak} command
9029 in conjunction with @code{jump}. @xref{Set Breaks, ,Setting
9032 The @code{jump} command does not change the current stack frame, or
9033 the stack pointer, or the contents of any memory location or any
9034 register other than the program counter. If line @var{linespec} is in
9035 a different function from the one currently executing, the results may
9036 be bizarre if the two functions expect different patterns of arguments or
9037 of local variables. For this reason, the @code{jump} command requests
9038 confirmation if the specified line is not in the function currently
9039 executing. However, even bizarre results are predictable if you are
9040 well acquainted with the machine-language code of your program.
9042 @item jump *@var{address}
9043 Resume execution at the instruction at address @var{address}.
9046 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
9047 On many systems, you can get much the same effect as the @code{jump}
9048 command by storing a new value into the register @code{$pc}. The
9049 difference is that this does not start your program running; it only
9050 changes the address of where it @emph{will} run when you continue. For
9058 makes the next @code{continue} command or stepping command execute at
9059 address @code{0x485}, rather than at the address where your program stopped.
9060 @xref{Continuing and Stepping, ,Continuing and stepping}.
9062 The most common occasion to use the @code{jump} command is to back
9063 up---perhaps with more breakpoints set---over a portion of a program
9064 that has already executed, in order to examine its execution in more
9069 @section Giving your program a signal
9073 @item signal @var{signal}
9074 Resume execution where your program stopped, but immediately give it the
9075 signal @var{signal}. @var{signal} can be the name or the number of a
9076 signal. For example, on many systems @code{signal 2} and @code{signal
9077 SIGINT} are both ways of sending an interrupt signal.
9079 Alternatively, if @var{signal} is zero, continue execution without
9080 giving a signal. This is useful when your program stopped on account of
9081 a signal and would ordinary see the signal when resumed with the
9082 @code{continue} command; @samp{signal 0} causes it to resume without a
9085 @code{signal} does not repeat when you press @key{RET} a second time
9086 after executing the command.
9090 Invoking the @code{signal} command is not the same as invoking the
9091 @code{kill} utility from the shell. Sending a signal with @code{kill}
9092 causes @value{GDBN} to decide what to do with the signal depending on
9093 the signal handling tables (@pxref{Signals}). The @code{signal} command
9094 passes the signal directly to your program.
9098 @section Returning from a function
9101 @cindex returning from a function
9104 @itemx return @var{expression}
9105 You can cancel execution of a function call with the @code{return}
9106 command. If you give an
9107 @var{expression} argument, its value is used as the function's return
9111 When you use @code{return}, @value{GDBN} discards the selected stack frame
9112 (and all frames within it). You can think of this as making the
9113 discarded frame return prematurely. If you wish to specify a value to
9114 be returned, give that value as the argument to @code{return}.
9116 This pops the selected stack frame (@pxref{Selection, ,Selecting a
9117 frame}), and any other frames inside of it, leaving its caller as the
9118 innermost remaining frame. That frame becomes selected. The
9119 specified value is stored in the registers used for returning values
9122 The @code{return} command does not resume execution; it leaves the
9123 program stopped in the state that would exist if the function had just
9124 returned. In contrast, the @code{finish} command (@pxref{Continuing
9125 and Stepping, ,Continuing and stepping}) resumes execution until the
9126 selected stack frame returns naturally.
9129 @section Calling program functions
9131 @cindex calling functions
9134 @item call @var{expr}
9135 Evaluate the expression @var{expr} without displaying @code{void}
9139 You can use this variant of the @code{print} command if you want to
9140 execute a function from your program, but without cluttering the output
9141 with @code{void} returned values. If the result is not void, it
9142 is printed and saved in the value history.
9144 For the A29K, a user-controlled variable @code{call_scratch_address},
9145 specifies the location of a scratch area to be used when @value{GDBN}
9146 calls a function in the target. This is necessary because the usual
9147 method of putting the scratch area on the stack does not work in systems
9148 that have separate instruction and data spaces.
9151 @section Patching programs
9153 @cindex patching binaries
9154 @cindex writing into executables
9155 @cindex writing into corefiles
9157 By default, @value{GDBN} opens the file containing your program's
9158 executable code (or the corefile) read-only. This prevents accidental
9159 alterations to machine code; but it also prevents you from intentionally
9160 patching your program's binary.
9162 If you'd like to be able to patch the binary, you can specify that
9163 explicitly with the @code{set write} command. For example, you might
9164 want to turn on internal debugging flags, or even to make emergency
9170 @itemx set write off
9171 If you specify @samp{set write on}, @value{GDBN} opens executable and
9172 core files for both reading and writing; if you specify @samp{set write
9173 off} (the default), @value{GDBN} opens them read-only.
9175 If you have already loaded a file, you must load it again (using the
9176 @code{exec-file} or @code{core-file} command) after changing @code{set
9177 write}, for your new setting to take effect.
9181 Display whether executable files and core files are opened for writing
9186 @chapter @value{GDBN} Files
9188 @value{GDBN} needs to know the file name of the program to be debugged,
9189 both in order to read its symbol table and in order to start your
9190 program. To debug a core dump of a previous run, you must also tell
9191 @value{GDBN} the name of the core dump file.
9194 * Files:: Commands to specify files
9195 * Symbol Errors:: Errors reading symbol files
9199 @section Commands to specify files
9201 @cindex symbol table
9202 @cindex core dump file
9204 You may want to specify executable and core dump file names. The usual
9205 way to do this is at start-up time, using the arguments to
9206 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
9207 Out of @value{GDBN}}).
9209 Occasionally it is necessary to change to a different file during a
9210 @value{GDBN} session. Or you may run @value{GDBN} and forget to specify
9211 a file you want to use. In these situations the @value{GDBN} commands
9212 to specify new files are useful.
9215 @cindex executable file
9217 @item file @var{filename}
9218 Use @var{filename} as the program to be debugged. It is read for its
9219 symbols and for the contents of pure memory. It is also the program
9220 executed when you use the @code{run} command. If you do not specify a
9221 directory and the file is not found in the @value{GDBN} working directory,
9222 @value{GDBN} uses the environment variable @code{PATH} as a list of
9223 directories to search, just as the shell does when looking for a program
9224 to run. You can change the value of this variable, for both @value{GDBN}
9225 and your program, using the @code{path} command.
9227 On systems with memory-mapped files, an auxiliary file named
9228 @file{@var{filename}.syms} may hold symbol table information for
9229 @var{filename}. If so, @value{GDBN} maps in the symbol table from
9230 @file{@var{filename}.syms}, starting up more quickly. See the
9231 descriptions of the file options @samp{-mapped} and @samp{-readnow}
9232 (available on the command line, and with the commands @code{file},
9233 @code{symbol-file}, or @code{add-symbol-file}, described below),
9234 for more information.
9237 @code{file} with no argument makes @value{GDBN} discard any information it
9238 has on both executable file and the symbol table.
9241 @item exec-file @r{[} @var{filename} @r{]}
9242 Specify that the program to be run (but not the symbol table) is found
9243 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
9244 if necessary to locate your program. Omitting @var{filename} means to
9245 discard information on the executable file.
9248 @item symbol-file @r{[} @var{filename} @r{]}
9249 Read symbol table information from file @var{filename}. @code{PATH} is
9250 searched when necessary. Use the @code{file} command to get both symbol
9251 table and program to run from the same file.
9253 @code{symbol-file} with no argument clears out @value{GDBN} information on your
9254 program's symbol table.
9256 The @code{symbol-file} command causes @value{GDBN} to forget the contents
9257 of its convenience variables, the value history, and all breakpoints and
9258 auto-display expressions. This is because they may contain pointers to
9259 the internal data recording symbols and data types, which are part of
9260 the old symbol table data being discarded inside @value{GDBN}.
9262 @code{symbol-file} does not repeat if you press @key{RET} again after
9265 When @value{GDBN} is configured for a particular environment, it
9266 understands debugging information in whatever format is the standard
9267 generated for that environment; you may use either a @sc{gnu} compiler, or
9268 other compilers that adhere to the local conventions.
9269 Best results are usually obtained from @sc{gnu} compilers; for example,
9270 using @code{@value{GCC}} you can generate debugging information for
9273 For most kinds of object files, with the exception of old SVR3 systems
9274 using COFF, the @code{symbol-file} command does not normally read the
9275 symbol table in full right away. Instead, it scans the symbol table
9276 quickly to find which source files and which symbols are present. The
9277 details are read later, one source file at a time, as they are needed.
9279 The purpose of this two-stage reading strategy is to make @value{GDBN}
9280 start up faster. For the most part, it is invisible except for
9281 occasional pauses while the symbol table details for a particular source
9282 file are being read. (The @code{set verbose} command can turn these
9283 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
9284 warnings and messages}.)
9286 We have not implemented the two-stage strategy for COFF yet. When the
9287 symbol table is stored in COFF format, @code{symbol-file} reads the
9288 symbol table data in full right away. Note that ``stabs-in-COFF''
9289 still does the two-stage strategy, since the debug info is actually
9293 @cindex reading symbols immediately
9294 @cindex symbols, reading immediately
9296 @cindex memory-mapped symbol file
9297 @cindex saving symbol table
9298 @item symbol-file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
9299 @itemx file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
9300 You can override the @value{GDBN} two-stage strategy for reading symbol
9301 tables by using the @samp{-readnow} option with any of the commands that
9302 load symbol table information, if you want to be sure @value{GDBN} has the
9303 entire symbol table available.
9305 If memory-mapped files are available on your system through the
9306 @code{mmap} system call, you can use another option, @samp{-mapped}, to
9307 cause @value{GDBN} to write the symbols for your program into a reusable
9308 file. Future @value{GDBN} debugging sessions map in symbol information
9309 from this auxiliary symbol file (if the program has not changed), rather
9310 than spending time reading the symbol table from the executable
9311 program. Using the @samp{-mapped} option has the same effect as
9312 starting @value{GDBN} with the @samp{-mapped} command-line option.
9314 You can use both options together, to make sure the auxiliary symbol
9315 file has all the symbol information for your program.
9317 The auxiliary symbol file for a program called @var{myprog} is called
9318 @samp{@var{myprog}.syms}. Once this file exists (so long as it is newer
9319 than the corresponding executable), @value{GDBN} always attempts to use
9320 it when you debug @var{myprog}; no special options or commands are
9323 The @file{.syms} file is specific to the host machine where you run
9324 @value{GDBN}. It holds an exact image of the internal @value{GDBN}
9325 symbol table. It cannot be shared across multiple host platforms.
9327 @c FIXME: for now no mention of directories, since this seems to be in
9328 @c flux. 13mar1992 status is that in theory GDB would look either in
9329 @c current dir or in same dir as myprog; but issues like competing
9330 @c GDB's, or clutter in system dirs, mean that in practice right now
9331 @c only current dir is used. FFish says maybe a special GDB hierarchy
9332 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
9337 @item core-file @r{[} @var{filename} @r{]}
9338 Specify the whereabouts of a core dump file to be used as the ``contents
9339 of memory''. Traditionally, core files contain only some parts of the
9340 address space of the process that generated them; @value{GDBN} can access the
9341 executable file itself for other parts.
9343 @code{core-file} with no argument specifies that no core file is
9346 Note that the core file is ignored when your program is actually running
9347 under @value{GDBN}. So, if you have been running your program and you
9348 wish to debug a core file instead, you must kill the subprocess in which
9349 the program is running. To do this, use the @code{kill} command
9350 (@pxref{Kill Process, ,Killing the child process}).
9352 @kindex add-symbol-file
9353 @cindex dynamic linking
9354 @item add-symbol-file @var{filename} @var{address}
9355 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
9356 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
9357 The @code{add-symbol-file} command reads additional symbol table
9358 information from the file @var{filename}. You would use this command
9359 when @var{filename} has been dynamically loaded (by some other means)
9360 into the program that is running. @var{address} should be the memory
9361 address at which the file has been loaded; @value{GDBN} cannot figure
9362 this out for itself. You can additionally specify an arbitrary number
9363 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
9364 section name and base address for that section. You can specify any
9365 @var{address} as an expression.
9367 The symbol table of the file @var{filename} is added to the symbol table
9368 originally read with the @code{symbol-file} command. You can use the
9369 @code{add-symbol-file} command any number of times; the new symbol data
9370 thus read keeps adding to the old. To discard all old symbol data
9371 instead, use the @code{symbol-file} command without any arguments.
9373 @cindex relocatable object files, reading symbols from
9374 @cindex object files, relocatable, reading symbols from
9375 @cindex reading symbols from relocatable object files
9376 @cindex symbols, reading from relocatable object files
9377 @cindex @file{.o} files, reading symbols from
9378 Although @var{filename} is typically a shared library file, an
9379 executable file, or some other object file which has been fully
9380 relocated for loading into a process, you can also load symbolic
9381 information from relocatable @file{.o} files, as long as:
9385 the file's symbolic information refers only to linker symbols defined in
9386 that file, not to symbols defined by other object files,
9388 every section the file's symbolic information refers to has actually
9389 been loaded into the inferior, as it appears in the file, and
9391 you can determine the address at which every section was loaded, and
9392 provide these to the @code{add-symbol-file} command.
9396 Some embedded operating systems, like Sun Chorus and VxWorks, can load
9397 relocatable files into an already running program; such systems
9398 typically make the requirements above easy to meet. However, it's
9399 important to recognize that many native systems use complex link
9400 procedures (@code{.linkonce} section factoring and C++ constructor table
9401 assembly, for example) that make the requirements difficult to meet. In
9402 general, one cannot assume that using @code{add-symbol-file} to read a
9403 relocatable object file's symbolic information will have the same effect
9404 as linking the relocatable object file into the program in the normal
9407 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
9409 You can use the @samp{-mapped} and @samp{-readnow} options just as with
9410 the @code{symbol-file} command, to change how @value{GDBN} manages the symbol
9411 table information for @var{filename}.
9413 @kindex add-shared-symbol-file
9414 @item add-shared-symbol-file
9415 The @code{add-shared-symbol-file} command can be used only under Harris' CXUX
9416 operating system for the Motorola 88k. @value{GDBN} automatically looks for
9417 shared libraries, however if @value{GDBN} does not find yours, you can run
9418 @code{add-shared-symbol-file}. It takes no arguments.
9422 The @code{section} command changes the base address of section SECTION of
9423 the exec file to ADDR. This can be used if the exec file does not contain
9424 section addresses, (such as in the a.out format), or when the addresses
9425 specified in the file itself are wrong. Each section must be changed
9426 separately. The @code{info files} command, described below, lists all
9427 the sections and their addresses.
9433 @code{info files} and @code{info target} are synonymous; both print the
9434 current target (@pxref{Targets, ,Specifying a Debugging Target}),
9435 including the names of the executable and core dump files currently in
9436 use by @value{GDBN}, and the files from which symbols were loaded. The
9437 command @code{help target} lists all possible targets rather than
9440 @kindex maint info sections
9441 @item maint info sections
9442 Another command that can give you extra information about program sections
9443 is @code{maint info sections}. In addition to the section information
9444 displayed by @code{info files}, this command displays the flags and file
9445 offset of each section in the executable and core dump files. In addition,
9446 @code{maint info sections} provides the following command options (which
9447 may be arbitrarily combined):
9451 Display sections for all loaded object files, including shared libraries.
9452 @item @var{sections}
9453 Display info only for named @var{sections}.
9454 @item @var{section-flags}
9455 Display info only for sections for which @var{section-flags} are true.
9456 The section flags that @value{GDBN} currently knows about are:
9459 Section will have space allocated in the process when loaded.
9460 Set for all sections except those containing debug information.
9462 Section will be loaded from the file into the child process memory.
9463 Set for pre-initialized code and data, clear for @code{.bss} sections.
9465 Section needs to be relocated before loading.
9467 Section cannot be modified by the child process.
9469 Section contains executable code only.
9471 Section contains data only (no executable code).
9473 Section will reside in ROM.
9475 Section contains data for constructor/destructor lists.
9477 Section is not empty.
9479 An instruction to the linker to not output the section.
9480 @item COFF_SHARED_LIBRARY
9481 A notification to the linker that the section contains
9482 COFF shared library information.
9484 Section contains common symbols.
9489 All file-specifying commands allow both absolute and relative file names
9490 as arguments. @value{GDBN} always converts the file name to an absolute file
9491 name and remembers it that way.
9493 @cindex shared libraries
9494 @value{GDBN} supports HP-UX, SunOS, SVr4, Irix 5, and IBM RS/6000 shared
9497 @value{GDBN} automatically loads symbol definitions from shared libraries
9498 when you use the @code{run} command, or when you examine a core file.
9499 (Before you issue the @code{run} command, @value{GDBN} does not understand
9500 references to a function in a shared library, however---unless you are
9501 debugging a core file).
9503 On HP-UX, if the program loads a library explicitly, @value{GDBN}
9504 automatically loads the symbols at the time of the @code{shl_load} call.
9506 @c FIXME: some @value{GDBN} release may permit some refs to undef
9507 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
9508 @c FIXME...lib; check this from time to time when updating manual
9510 There are times, however, when you may wish to not automatically load
9511 symbol definitions from shared libraries, such as when they are
9512 particularly large or there are many of them.
9514 To control the automatic loading of shared library symbols, use the
9518 @kindex set auto-solib-add
9519 @item set auto-solib-add @var{mode}
9520 If @var{mode} is @code{on}, symbols from all shared object libraries
9521 will be loaded automatically when the inferior begins execution, you
9522 attach to an independently started inferior, or when the dynamic linker
9523 informs @value{GDBN} that a new library has been loaded. If @var{mode}
9524 is @code{off}, symbols must be loaded manually, using the
9525 @code{sharedlibrary} command. The default value is @code{on}.
9527 @kindex show auto-solib-add
9528 @item show auto-solib-add
9529 Display the current autoloading mode.
9532 To explicitly load shared library symbols, use the @code{sharedlibrary}
9536 @kindex info sharedlibrary
9539 @itemx info sharedlibrary
9540 Print the names of the shared libraries which are currently loaded.
9542 @kindex sharedlibrary
9544 @item sharedlibrary @var{regex}
9545 @itemx share @var{regex}
9546 Load shared object library symbols for files matching a
9547 Unix regular expression.
9548 As with files loaded automatically, it only loads shared libraries
9549 required by your program for a core file or after typing @code{run}. If
9550 @var{regex} is omitted all shared libraries required by your program are
9554 On some systems, such as HP-UX systems, @value{GDBN} supports
9555 autoloading shared library symbols until a limiting threshold size is
9556 reached. This provides the benefit of allowing autoloading to remain on
9557 by default, but avoids autoloading excessively large shared libraries,
9558 up to a threshold that is initially set, but which you can modify if you
9561 Beyond that threshold, symbols from shared libraries must be explicitly
9562 loaded. To load these symbols, use the command @code{sharedlibrary
9563 @var{filename}}. The base address of the shared library is determined
9564 automatically by @value{GDBN} and need not be specified.
9566 To display or set the threshold, use the commands:
9569 @kindex set auto-solib-limit
9570 @item set auto-solib-limit @var{threshold}
9571 Set the autoloading size threshold, in an integral number of megabytes.
9572 If @var{threshold} is nonzero and shared library autoloading is enabled,
9573 symbols from all shared object libraries will be loaded until the total
9574 size of the loaded shared library symbols exceeds this threshold.
9575 Otherwise, symbols must be loaded manually, using the
9576 @code{sharedlibrary} command. The default threshold is 100 (i.e. 100
9579 @kindex show auto-solib-limit
9580 @item show auto-solib-limit
9581 Display the current autoloading size threshold, in megabytes.
9585 @section Errors reading symbol files
9587 While reading a symbol file, @value{GDBN} occasionally encounters problems,
9588 such as symbol types it does not recognize, or known bugs in compiler
9589 output. By default, @value{GDBN} does not notify you of such problems, since
9590 they are relatively common and primarily of interest to people
9591 debugging compilers. If you are interested in seeing information
9592 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
9593 only one message about each such type of problem, no matter how many
9594 times the problem occurs; or you can ask @value{GDBN} to print more messages,
9595 to see how many times the problems occur, with the @code{set
9596 complaints} command (@pxref{Messages/Warnings, ,Optional warnings and
9599 The messages currently printed, and their meanings, include:
9602 @item inner block not inside outer block in @var{symbol}
9604 The symbol information shows where symbol scopes begin and end
9605 (such as at the start of a function or a block of statements). This
9606 error indicates that an inner scope block is not fully contained
9607 in its outer scope blocks.
9609 @value{GDBN} circumvents the problem by treating the inner block as if it had
9610 the same scope as the outer block. In the error message, @var{symbol}
9611 may be shown as ``@code{(don't know)}'' if the outer block is not a
9614 @item block at @var{address} out of order
9616 The symbol information for symbol scope blocks should occur in
9617 order of increasing addresses. This error indicates that it does not
9620 @value{GDBN} does not circumvent this problem, and has trouble
9621 locating symbols in the source file whose symbols it is reading. (You
9622 can often determine what source file is affected by specifying
9623 @code{set verbose on}. @xref{Messages/Warnings, ,Optional warnings and
9626 @item bad block start address patched
9628 The symbol information for a symbol scope block has a start address
9629 smaller than the address of the preceding source line. This is known
9630 to occur in the SunOS 4.1.1 (and earlier) C compiler.
9632 @value{GDBN} circumvents the problem by treating the symbol scope block as
9633 starting on the previous source line.
9635 @item bad string table offset in symbol @var{n}
9638 Symbol number @var{n} contains a pointer into the string table which is
9639 larger than the size of the string table.
9641 @value{GDBN} circumvents the problem by considering the symbol to have the
9642 name @code{foo}, which may cause other problems if many symbols end up
9645 @item unknown symbol type @code{0x@var{nn}}
9647 The symbol information contains new data types that @value{GDBN} does
9648 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
9649 uncomprehended information, in hexadecimal.
9651 @value{GDBN} circumvents the error by ignoring this symbol information.
9652 This usually allows you to debug your program, though certain symbols
9653 are not accessible. If you encounter such a problem and feel like
9654 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
9655 on @code{complain}, then go up to the function @code{read_dbx_symtab}
9656 and examine @code{*bufp} to see the symbol.
9658 @item stub type has NULL name
9660 @value{GDBN} could not find the full definition for a struct or class.
9662 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
9663 The symbol information for a C@t{++} member function is missing some
9664 information that recent versions of the compiler should have output for
9667 @item info mismatch between compiler and debugger
9669 @value{GDBN} could not parse a type specification output by the compiler.
9674 @chapter Specifying a Debugging Target
9676 @cindex debugging target
9679 A @dfn{target} is the execution environment occupied by your program.
9681 Often, @value{GDBN} runs in the same host environment as your program;
9682 in that case, the debugging target is specified as a side effect when
9683 you use the @code{file} or @code{core} commands. When you need more
9684 flexibility---for example, running @value{GDBN} on a physically separate
9685 host, or controlling a standalone system over a serial port or a
9686 realtime system over a TCP/IP connection---you can use the @code{target}
9687 command to specify one of the target types configured for @value{GDBN}
9688 (@pxref{Target Commands, ,Commands for managing targets}).
9691 * Active Targets:: Active targets
9692 * Target Commands:: Commands for managing targets
9693 * Byte Order:: Choosing target byte order
9694 * Remote:: Remote debugging
9695 * KOD:: Kernel Object Display
9699 @node Active Targets
9700 @section Active targets
9702 @cindex stacking targets
9703 @cindex active targets
9704 @cindex multiple targets
9706 There are three classes of targets: processes, core files, and
9707 executable files. @value{GDBN} can work concurrently on up to three
9708 active targets, one in each class. This allows you to (for example)
9709 start a process and inspect its activity without abandoning your work on
9712 For example, if you execute @samp{gdb a.out}, then the executable file
9713 @code{a.out} is the only active target. If you designate a core file as
9714 well---presumably from a prior run that crashed and coredumped---then
9715 @value{GDBN} has two active targets and uses them in tandem, looking
9716 first in the corefile target, then in the executable file, to satisfy
9717 requests for memory addresses. (Typically, these two classes of target
9718 are complementary, since core files contain only a program's
9719 read-write memory---variables and so on---plus machine status, while
9720 executable files contain only the program text and initialized data.)
9722 When you type @code{run}, your executable file becomes an active process
9723 target as well. When a process target is active, all @value{GDBN}
9724 commands requesting memory addresses refer to that target; addresses in
9725 an active core file or executable file target are obscured while the
9726 process target is active.
9728 Use the @code{core-file} and @code{exec-file} commands to select a new
9729 core file or executable target (@pxref{Files, ,Commands to specify
9730 files}). To specify as a target a process that is already running, use
9731 the @code{attach} command (@pxref{Attach, ,Debugging an already-running
9734 @node Target Commands
9735 @section Commands for managing targets
9738 @item target @var{type} @var{parameters}
9739 Connects the @value{GDBN} host environment to a target machine or
9740 process. A target is typically a protocol for talking to debugging
9741 facilities. You use the argument @var{type} to specify the type or
9742 protocol of the target machine.
9744 Further @var{parameters} are interpreted by the target protocol, but
9745 typically include things like device names or host names to connect
9746 with, process numbers, and baud rates.
9748 The @code{target} command does not repeat if you press @key{RET} again
9749 after executing the command.
9753 Displays the names of all targets available. To display targets
9754 currently selected, use either @code{info target} or @code{info files}
9755 (@pxref{Files, ,Commands to specify files}).
9757 @item help target @var{name}
9758 Describe a particular target, including any parameters necessary to
9761 @kindex set gnutarget
9762 @item set gnutarget @var{args}
9763 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
9764 knows whether it is reading an @dfn{executable},
9765 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
9766 with the @code{set gnutarget} command. Unlike most @code{target} commands,
9767 with @code{gnutarget} the @code{target} refers to a program, not a machine.
9770 @emph{Warning:} To specify a file format with @code{set gnutarget},
9771 you must know the actual BFD name.
9775 @xref{Files, , Commands to specify files}.
9777 @kindex show gnutarget
9778 @item show gnutarget
9779 Use the @code{show gnutarget} command to display what file format
9780 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
9781 @value{GDBN} will determine the file format for each file automatically,
9782 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
9785 Here are some common targets (available, or not, depending on the GDB
9790 @item target exec @var{program}
9791 An executable file. @samp{target exec @var{program}} is the same as
9792 @samp{exec-file @var{program}}.
9795 @item target core @var{filename}
9796 A core dump file. @samp{target core @var{filename}} is the same as
9797 @samp{core-file @var{filename}}.
9799 @kindex target remote
9800 @item target remote @var{dev}
9801 Remote serial target in GDB-specific protocol. The argument @var{dev}
9802 specifies what serial device to use for the connection (e.g.
9803 @file{/dev/ttya}). @xref{Remote, ,Remote debugging}. @code{target remote}
9804 supports the @code{load} command. This is only useful if you have
9805 some other way of getting the stub to the target system, and you can put
9806 it somewhere in memory where it won't get clobbered by the download.
9810 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
9818 works; however, you cannot assume that a specific memory map, device
9819 drivers, or even basic I/O is available, although some simulators do
9820 provide these. For info about any processor-specific simulator details,
9821 see the appropriate section in @ref{Embedded Processors, ,Embedded
9826 Some configurations may include these targets as well:
9831 @item target nrom @var{dev}
9832 NetROM ROM emulator. This target only supports downloading.
9836 Different targets are available on different configurations of @value{GDBN};
9837 your configuration may have more or fewer targets.
9839 Many remote targets require you to download the executable's code
9840 once you've successfully established a connection.
9844 @kindex load @var{filename}
9845 @item load @var{filename}
9846 Depending on what remote debugging facilities are configured into
9847 @value{GDBN}, the @code{load} command may be available. Where it exists, it
9848 is meant to make @var{filename} (an executable) available for debugging
9849 on the remote system---by downloading, or dynamic linking, for example.
9850 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
9851 the @code{add-symbol-file} command.
9853 If your @value{GDBN} does not have a @code{load} command, attempting to
9854 execute it gets the error message ``@code{You can't do that when your
9855 target is @dots{}}''
9857 The file is loaded at whatever address is specified in the executable.
9858 For some object file formats, you can specify the load address when you
9859 link the program; for other formats, like a.out, the object file format
9860 specifies a fixed address.
9861 @c FIXME! This would be a good place for an xref to the GNU linker doc.
9863 @code{load} does not repeat if you press @key{RET} again after using it.
9867 @section Choosing target byte order
9869 @cindex choosing target byte order
9870 @cindex target byte order
9872 Some types of processors, such as the MIPS, PowerPC, and Hitachi SH,
9873 offer the ability to run either big-endian or little-endian byte
9874 orders. Usually the executable or symbol will include a bit to
9875 designate the endian-ness, and you will not need to worry about
9876 which to use. However, you may still find it useful to adjust
9877 @value{GDBN}'s idea of processor endian-ness manually.
9880 @kindex set endian big
9881 @item set endian big
9882 Instruct @value{GDBN} to assume the target is big-endian.
9884 @kindex set endian little
9885 @item set endian little
9886 Instruct @value{GDBN} to assume the target is little-endian.
9888 @kindex set endian auto
9889 @item set endian auto
9890 Instruct @value{GDBN} to use the byte order associated with the
9894 Display @value{GDBN}'s current idea of the target byte order.
9898 Note that these commands merely adjust interpretation of symbolic
9899 data on the host, and that they have absolutely no effect on the
9903 @section Remote debugging
9904 @cindex remote debugging
9906 If you are trying to debug a program running on a machine that cannot run
9907 @value{GDBN} in the usual way, it is often useful to use remote debugging.
9908 For example, you might use remote debugging on an operating system kernel,
9909 or on a small system which does not have a general purpose operating system
9910 powerful enough to run a full-featured debugger.
9912 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
9913 to make this work with particular debugging targets. In addition,
9914 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
9915 but not specific to any particular target system) which you can use if you
9916 write the remote stubs---the code that runs on the remote system to
9917 communicate with @value{GDBN}.
9919 Other remote targets may be available in your
9920 configuration of @value{GDBN}; use @code{help target} to list them.
9923 * Remote Serial:: @value{GDBN} remote serial protocol
9927 @subsection The @value{GDBN} remote serial protocol
9929 @cindex remote serial debugging, overview
9930 To debug a program running on another machine (the debugging
9931 @dfn{target} machine), you must first arrange for all the usual
9932 prerequisites for the program to run by itself. For example, for a C
9937 A startup routine to set up the C runtime environment; these usually
9938 have a name like @file{crt0}. The startup routine may be supplied by
9939 your hardware supplier, or you may have to write your own.
9942 A C subroutine library to support your program's
9943 subroutine calls, notably managing input and output.
9946 A way of getting your program to the other machine---for example, a
9947 download program. These are often supplied by the hardware
9948 manufacturer, but you may have to write your own from hardware
9952 The next step is to arrange for your program to use a serial port to
9953 communicate with the machine where @value{GDBN} is running (the @dfn{host}
9954 machine). In general terms, the scheme looks like this:
9958 @value{GDBN} already understands how to use this protocol; when everything
9959 else is set up, you can simply use the @samp{target remote} command
9960 (@pxref{Targets,,Specifying a Debugging Target}).
9962 @item On the target,
9963 you must link with your program a few special-purpose subroutines that
9964 implement the @value{GDBN} remote serial protocol. The file containing these
9965 subroutines is called a @dfn{debugging stub}.
9967 On certain remote targets, you can use an auxiliary program
9968 @code{gdbserver} instead of linking a stub into your program.
9969 @xref{Server,,Using the @code{gdbserver} program}, for details.
9972 The debugging stub is specific to the architecture of the remote
9973 machine; for example, use @file{sparc-stub.c} to debug programs on
9976 @cindex remote serial stub list
9977 These working remote stubs are distributed with @value{GDBN}:
9982 @cindex @file{i386-stub.c}
9985 For Intel 386 and compatible architectures.
9988 @cindex @file{m68k-stub.c}
9989 @cindex Motorola 680x0
9991 For Motorola 680x0 architectures.
9994 @cindex @file{sh-stub.c}
9997 For Hitachi SH architectures.
10000 @cindex @file{sparc-stub.c}
10002 For @sc{sparc} architectures.
10004 @item sparcl-stub.c
10005 @cindex @file{sparcl-stub.c}
10008 For Fujitsu @sc{sparclite} architectures.
10012 The @file{README} file in the @value{GDBN} distribution may list other
10013 recently added stubs.
10016 * Stub Contents:: What the stub can do for you
10017 * Bootstrapping:: What you must do for the stub
10018 * Debug Session:: Putting it all together
10019 * Protocol:: Definition of the communication protocol
10020 * Server:: Using the `gdbserver' program
10021 * NetWare:: Using the `gdbserve.nlm' program
10024 @node Stub Contents
10025 @subsubsection What the stub can do for you
10027 @cindex remote serial stub
10028 The debugging stub for your architecture supplies these three
10032 @item set_debug_traps
10033 @kindex set_debug_traps
10034 @cindex remote serial stub, initialization
10035 This routine arranges for @code{handle_exception} to run when your
10036 program stops. You must call this subroutine explicitly near the
10037 beginning of your program.
10039 @item handle_exception
10040 @kindex handle_exception
10041 @cindex remote serial stub, main routine
10042 This is the central workhorse, but your program never calls it
10043 explicitly---the setup code arranges for @code{handle_exception} to
10044 run when a trap is triggered.
10046 @code{handle_exception} takes control when your program stops during
10047 execution (for example, on a breakpoint), and mediates communications
10048 with @value{GDBN} on the host machine. This is where the communications
10049 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
10050 representative on the target machine. It begins by sending summary
10051 information on the state of your program, then continues to execute,
10052 retrieving and transmitting any information @value{GDBN} needs, until you
10053 execute a @value{GDBN} command that makes your program resume; at that point,
10054 @code{handle_exception} returns control to your own code on the target
10058 @cindex @code{breakpoint} subroutine, remote
10059 Use this auxiliary subroutine to make your program contain a
10060 breakpoint. Depending on the particular situation, this may be the only
10061 way for @value{GDBN} to get control. For instance, if your target
10062 machine has some sort of interrupt button, you won't need to call this;
10063 pressing the interrupt button transfers control to
10064 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
10065 simply receiving characters on the serial port may also trigger a trap;
10066 again, in that situation, you don't need to call @code{breakpoint} from
10067 your own program---simply running @samp{target remote} from the host
10068 @value{GDBN} session gets control.
10070 Call @code{breakpoint} if none of these is true, or if you simply want
10071 to make certain your program stops at a predetermined point for the
10072 start of your debugging session.
10075 @node Bootstrapping
10076 @subsubsection What you must do for the stub
10078 @cindex remote stub, support routines
10079 The debugging stubs that come with @value{GDBN} are set up for a particular
10080 chip architecture, but they have no information about the rest of your
10081 debugging target machine.
10083 First of all you need to tell the stub how to communicate with the
10087 @item int getDebugChar()
10088 @kindex getDebugChar
10089 Write this subroutine to read a single character from the serial port.
10090 It may be identical to @code{getchar} for your target system; a
10091 different name is used to allow you to distinguish the two if you wish.
10093 @item void putDebugChar(int)
10094 @kindex putDebugChar
10095 Write this subroutine to write a single character to the serial port.
10096 It may be identical to @code{putchar} for your target system; a
10097 different name is used to allow you to distinguish the two if you wish.
10100 @cindex control C, and remote debugging
10101 @cindex interrupting remote targets
10102 If you want @value{GDBN} to be able to stop your program while it is
10103 running, you need to use an interrupt-driven serial driver, and arrange
10104 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
10105 character). That is the character which @value{GDBN} uses to tell the
10106 remote system to stop.
10108 Getting the debugging target to return the proper status to @value{GDBN}
10109 probably requires changes to the standard stub; one quick and dirty way
10110 is to just execute a breakpoint instruction (the ``dirty'' part is that
10111 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
10113 Other routines you need to supply are:
10116 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
10117 @kindex exceptionHandler
10118 Write this function to install @var{exception_address} in the exception
10119 handling tables. You need to do this because the stub does not have any
10120 way of knowing what the exception handling tables on your target system
10121 are like (for example, the processor's table might be in @sc{rom},
10122 containing entries which point to a table in @sc{ram}).
10123 @var{exception_number} is the exception number which should be changed;
10124 its meaning is architecture-dependent (for example, different numbers
10125 might represent divide by zero, misaligned access, etc). When this
10126 exception occurs, control should be transferred directly to
10127 @var{exception_address}, and the processor state (stack, registers,
10128 and so on) should be just as it is when a processor exception occurs. So if
10129 you want to use a jump instruction to reach @var{exception_address}, it
10130 should be a simple jump, not a jump to subroutine.
10132 For the 386, @var{exception_address} should be installed as an interrupt
10133 gate so that interrupts are masked while the handler runs. The gate
10134 should be at privilege level 0 (the most privileged level). The
10135 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
10136 help from @code{exceptionHandler}.
10138 @item void flush_i_cache()
10139 @kindex flush_i_cache
10140 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
10141 instruction cache, if any, on your target machine. If there is no
10142 instruction cache, this subroutine may be a no-op.
10144 On target machines that have instruction caches, @value{GDBN} requires this
10145 function to make certain that the state of your program is stable.
10149 You must also make sure this library routine is available:
10152 @item void *memset(void *, int, int)
10154 This is the standard library function @code{memset} that sets an area of
10155 memory to a known value. If you have one of the free versions of
10156 @code{libc.a}, @code{memset} can be found there; otherwise, you must
10157 either obtain it from your hardware manufacturer, or write your own.
10160 If you do not use the GNU C compiler, you may need other standard
10161 library subroutines as well; this varies from one stub to another,
10162 but in general the stubs are likely to use any of the common library
10163 subroutines which @code{@value{GCC}} generates as inline code.
10166 @node Debug Session
10167 @subsubsection Putting it all together
10169 @cindex remote serial debugging summary
10170 In summary, when your program is ready to debug, you must follow these
10175 Make sure you have defined the supporting low-level routines
10176 (@pxref{Bootstrapping,,What you must do for the stub}):
10178 @code{getDebugChar}, @code{putDebugChar},
10179 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
10183 Insert these lines near the top of your program:
10191 For the 680x0 stub only, you need to provide a variable called
10192 @code{exceptionHook}. Normally you just use:
10195 void (*exceptionHook)() = 0;
10199 but if before calling @code{set_debug_traps}, you set it to point to a
10200 function in your program, that function is called when
10201 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
10202 error). The function indicated by @code{exceptionHook} is called with
10203 one parameter: an @code{int} which is the exception number.
10206 Compile and link together: your program, the @value{GDBN} debugging stub for
10207 your target architecture, and the supporting subroutines.
10210 Make sure you have a serial connection between your target machine and
10211 the @value{GDBN} host, and identify the serial port on the host.
10214 @c The "remote" target now provides a `load' command, so we should
10215 @c document that. FIXME.
10216 Download your program to your target machine (or get it there by
10217 whatever means the manufacturer provides), and start it.
10220 To start remote debugging, run @value{GDBN} on the host machine, and specify
10221 as an executable file the program that is running in the remote machine.
10222 This tells @value{GDBN} how to find your program's symbols and the contents
10226 @cindex serial line, @code{target remote}
10227 Establish communication using the @code{target remote} command.
10228 Its argument specifies how to communicate with the target
10229 machine---either via a devicename attached to a direct serial line, or a
10230 TCP port (usually to a terminal server which in turn has a serial line
10231 to the target). For example, to use a serial line connected to the
10232 device named @file{/dev/ttyb}:
10235 target remote /dev/ttyb
10238 @cindex TCP port, @code{target remote}
10239 To use a TCP connection, use an argument of the form
10240 @code{@var{host}:port}. For example, to connect to port 2828 on a
10241 terminal server named @code{manyfarms}:
10244 target remote manyfarms:2828
10247 If your remote target is actually running on the same machine as
10248 your debugger session (e.g.@: a simulator of your target running on
10249 the same host), you can omit the hostname. For example, to connect
10250 to port 1234 on your local machine:
10253 target remote :1234
10257 Note that the colon is still required here.
10260 Now you can use all the usual commands to examine and change data and to
10261 step and continue the remote program.
10263 To resume the remote program and stop debugging it, use the @code{detach}
10266 @cindex interrupting remote programs
10267 @cindex remote programs, interrupting
10268 Whenever @value{GDBN} is waiting for the remote program, if you type the
10269 interrupt character (often @key{C-C}), @value{GDBN} attempts to stop the
10270 program. This may or may not succeed, depending in part on the hardware
10271 and the serial drivers the remote system uses. If you type the
10272 interrupt character once again, @value{GDBN} displays this prompt:
10275 Interrupted while waiting for the program.
10276 Give up (and stop debugging it)? (y or n)
10279 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
10280 (If you decide you want to try again later, you can use @samp{target
10281 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
10282 goes back to waiting.
10285 @subsubsection Communication protocol
10287 @cindex debugging stub, example
10288 @cindex remote stub, example
10289 @cindex stub example, remote debugging
10290 The stub files provided with @value{GDBN} implement the target side of the
10291 communication protocol, and the @value{GDBN} side is implemented in the
10292 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
10293 these subroutines to communicate, and ignore the details. (If you're
10294 implementing your own stub file, you can still ignore the details: start
10295 with one of the existing stub files. @file{sparc-stub.c} is the best
10296 organized, and therefore the easiest to read.)
10298 However, there may be occasions when you need to know something about
10299 the protocol---for example, if there is only one serial port to your
10300 target machine, you might want your program to do something special if
10301 it recognizes a packet meant for @value{GDBN}.
10303 In the examples below, @samp{<-} and @samp{->} are used to indicate
10304 transmitted and received data respectfully.
10306 @cindex protocol, @value{GDBN} remote serial
10307 @cindex serial protocol, @value{GDBN} remote
10308 @cindex remote serial protocol
10309 All @value{GDBN} commands and responses (other than acknowledgments) are
10310 sent as a @var{packet}. A @var{packet} is introduced with the character
10311 @samp{$}, the actual @var{packet-data}, and the terminating character
10312 @samp{#} followed by a two-digit @var{checksum}:
10315 @code{$}@var{packet-data}@code{#}@var{checksum}
10319 @cindex checksum, for @value{GDBN} remote
10321 The two-digit @var{checksum} is computed as the modulo 256 sum of all
10322 characters between the leading @samp{$} and the trailing @samp{#} (an
10323 eight bit unsigned checksum).
10325 Implementors should note that prior to @value{GDBN} 5.0 the protocol
10326 specification also included an optional two-digit @var{sequence-id}:
10329 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
10332 @cindex sequence-id, for @value{GDBN} remote
10334 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
10335 has never output @var{sequence-id}s. Stubs that handle packets added
10336 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
10338 @cindex acknowledgment, for @value{GDBN} remote
10339 When either the host or the target machine receives a packet, the first
10340 response expected is an acknowledgment: either @samp{+} (to indicate
10341 the package was received correctly) or @samp{-} (to request
10345 <- @code{$}@var{packet-data}@code{#}@var{checksum}
10350 The host (@value{GDBN}) sends @var{command}s, and the target (the
10351 debugging stub incorporated in your program) sends a @var{response}. In
10352 the case of step and continue @var{command}s, the response is only sent
10353 when the operation has completed (the target has again stopped).
10355 @var{packet-data} consists of a sequence of characters with the
10356 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
10359 Fields within the packet should be separated using @samp{,} @samp{;} or
10360 @samp{:}. Except where otherwise noted all numbers are represented in
10361 HEX with leading zeros suppressed.
10363 Implementors should note that prior to @value{GDBN} 5.0, the character
10364 @samp{:} could not appear as the third character in a packet (as it
10365 would potentially conflict with the @var{sequence-id}).
10367 Response @var{data} can be run-length encoded to save space. A @samp{*}
10368 means that the next character is an @sc{ascii} encoding giving a repeat count
10369 which stands for that many repetitions of the character preceding the
10370 @samp{*}. The encoding is @code{n+29}, yielding a printable character
10371 where @code{n >=3} (which is where rle starts to win). The printable
10372 characters @samp{$}, @samp{#}, @samp{+} and @samp{-} or with a numeric
10373 value greater than 126 should not be used.
10375 Some remote systems have used a different run-length encoding mechanism
10376 loosely refered to as the cisco encoding. Following the @samp{*}
10377 character are two hex digits that indicate the size of the packet.
10384 means the same as "0000".
10386 The error response returned for some packets includes a two character
10387 error number. That number is not well defined.
10389 For any @var{command} not supported by the stub, an empty response
10390 (@samp{$#00}) should be returned. That way it is possible to extend the
10391 protocol. A newer @value{GDBN} can tell if a packet is supported based
10394 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
10395 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
10398 Below is a complete list of all currently defined @var{command}s and
10399 their corresponding response @var{data}:
10401 @multitable @columnfractions .30 .30 .40
10406 @item extended mode
10409 Enable extended mode. In extended mode, the remote server is made
10410 persistent. The @samp{R} packet is used to restart the program being
10413 @tab reply @samp{OK}
10415 The remote target both supports and has enabled extended mode.
10420 Indicate the reason the target halted. The reply is the same as for step
10429 @tab Reserved for future use
10431 @item set program arguments @strong{(reserved)}
10432 @tab @code{A}@var{arglen}@code{,}@var{argnum}@code{,}@var{arg}@code{,...}
10437 Initialized @samp{argv[]} array passed into program. @var{arglen}
10438 specifies the number of bytes in the hex encoded byte stream @var{arg}.
10439 See @file{gdbserver} for more details.
10441 @tab reply @code{OK}
10443 @tab reply @code{E}@var{NN}
10445 @item set baud @strong{(deprecated)}
10446 @tab @code{b}@var{baud}
10448 Change the serial line speed to @var{baud}. JTC: @emph{When does the
10449 transport layer state change? When it's received, or after the ACK is
10450 transmitted. In either case, there are problems if the command or the
10451 acknowledgment packet is dropped.} Stan: @emph{If people really wanted
10452 to add something like this, and get it working for the first time, they
10453 ought to modify ser-unix.c to send some kind of out-of-band message to a
10454 specially-setup stub and have the switch happen "in between" packets, so
10455 that from remote protocol's point of view, nothing actually
10458 @item set breakpoint @strong{(deprecated)}
10459 @tab @code{B}@var{addr},@var{mode}
10461 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
10462 breakpoint at @var{addr}. @emph{This has been replaced by the @samp{Z} and
10466 @tab @code{c}@var{addr}
10468 @var{addr} is address to resume. If @var{addr} is omitted, resume at
10474 @item continue with signal
10475 @tab @code{C}@var{sig}@code{;}@var{addr}
10477 Continue with signal @var{sig} (hex signal number). If
10478 @code{;}@var{addr} is omitted, resume at same address.
10483 @item toggle debug @strong{(deprecated)}
10491 Detach @value{GDBN} from the remote system. Sent to the remote target before
10492 @value{GDBN} disconnects.
10494 @tab reply @emph{no response}
10496 @value{GDBN} does not check for any response after sending this packet.
10500 @tab Reserved for future use
10504 @tab Reserved for future use
10508 @tab Reserved for future use
10512 @tab Reserved for future use
10514 @item read registers
10516 @tab Read general registers.
10518 @tab reply @var{XX...}
10520 Each byte of register data is described by two hex digits. The bytes
10521 with the register are transmitted in target byte order. The size of
10522 each register and their position within the @samp{g} @var{packet} are
10523 determined by the @value{GDBN} internal macros @var{REGISTER_RAW_SIZE} and
10524 @var{REGISTER_NAME} macros. The specification of several standard
10525 @code{g} packets is specified below.
10527 @tab @code{E}@var{NN}
10531 @tab @code{G}@var{XX...}
10533 See @samp{g} for a description of the @var{XX...} data.
10535 @tab reply @code{OK}
10538 @tab reply @code{E}@var{NN}
10543 @tab Reserved for future use
10546 @tab @code{H}@var{c}@var{t...}
10548 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
10549 @samp{G}, et.al.). @var{c} = @samp{c} for thread used in step and
10550 continue; @var{t...} can be -1 for all threads. @var{c} = @samp{g} for
10551 thread used in other operations. If zero, pick a thread, any thread.
10553 @tab reply @code{OK}
10556 @tab reply @code{E}@var{NN}
10560 @c 'H': How restrictive (or permissive) is the thread model. If a
10561 @c thread is selected and stopped, are other threads allowed
10562 @c to continue to execute? As I mentioned above, I think the
10563 @c semantics of each command when a thread is selected must be
10564 @c described. For example:
10566 @c 'g': If the stub supports threads and a specific thread is
10567 @c selected, returns the register block from that thread;
10568 @c otherwise returns current registers.
10570 @c 'G' If the stub supports threads and a specific thread is
10571 @c selected, sets the registers of the register block of
10572 @c that thread; otherwise sets current registers.
10574 @item cycle step @strong{(draft)}
10575 @tab @code{i}@var{addr}@code{,}@var{nnn}
10577 Step the remote target by a single clock cycle. If @code{,}@var{nnn} is
10578 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
10579 step starting at that address.
10581 @item signal then cycle step @strong{(reserved)}
10584 See @samp{i} and @samp{S} for likely syntax and semantics.
10588 @tab Reserved for future use
10592 @tab Reserved for future use
10597 FIXME: @emph{There is no description of how operate when a specific
10598 thread context has been selected (ie. does 'k' kill only that thread?)}.
10602 @tab Reserved for future use
10606 @tab Reserved for future use
10609 @tab @code{m}@var{addr}@code{,}@var{length}
10611 Read @var{length} bytes of memory starting at address @var{addr}.
10612 Neither @value{GDBN} nor the stub assume that sized memory transfers are assumed
10613 using word alligned accesses. FIXME: @emph{A word aligned memory
10614 transfer mechanism is needed.}
10616 @tab reply @var{XX...}
10618 @var{XX...} is mem contents. Can be fewer bytes than requested if able
10619 to read only part of the data. Neither @value{GDBN} nor the stub assume that
10620 sized memory transfers are assumed using word alligned accesses. FIXME:
10621 @emph{A word aligned memory transfer mechanism is needed.}
10623 @tab reply @code{E}@var{NN}
10624 @tab @var{NN} is errno
10627 @tab @code{M}@var{addr},@var{length}@code{:}@var{XX...}
10629 Write @var{length} bytes of memory starting at address @var{addr}.
10630 @var{XX...} is the data.
10632 @tab reply @code{OK}
10635 @tab reply @code{E}@var{NN}
10637 for an error (this includes the case where only part of the data was
10642 @tab Reserved for future use
10646 @tab Reserved for future use
10650 @tab Reserved for future use
10654 @tab Reserved for future use
10656 @item read reg @strong{(reserved)}
10657 @tab @code{p}@var{n...}
10659 See write register.
10661 @tab return @var{r....}
10662 @tab The hex encoded value of the register in target byte order.
10665 @tab @code{P}@var{n...}@code{=}@var{r...}
10667 Write register @var{n...} with value @var{r...}, which contains two hex
10668 digits for each byte in the register (target byte order).
10670 @tab reply @code{OK}
10673 @tab reply @code{E}@var{NN}
10676 @item general query
10677 @tab @code{q}@var{query}
10679 Request info about @var{query}. In general @value{GDBN} queries
10680 have a leading upper case letter. Custom vendor queries should use a
10681 company prefix (in lower case) ex: @samp{qfsf.var}. @var{query} may
10682 optionally be followed by a @samp{,} or @samp{;} separated list. Stubs
10683 must ensure that they match the full @var{query} name.
10685 @tab reply @code{XX...}
10686 @tab Hex encoded data from query. The reply can not be empty.
10688 @tab reply @code{E}@var{NN}
10692 @tab Indicating an unrecognized @var{query}.
10695 @tab @code{Q}@var{var}@code{=}@var{val}
10697 Set value of @var{var} to @var{val}. See @samp{q} for a discussing of
10698 naming conventions.
10700 @item reset @strong{(deprecated)}
10703 Reset the entire system.
10705 @item remote restart
10706 @tab @code{R}@var{XX}
10708 Restart the program being debugged. @var{XX}, while needed, is ignored.
10709 This packet is only available in extended mode.
10714 The @samp{R} packet has no reply.
10717 @tab @code{s}@var{addr}
10719 @var{addr} is address to resume. If @var{addr} is omitted, resume at
10725 @item step with signal
10726 @tab @code{S}@var{sig}@code{;}@var{addr}
10728 Like @samp{C} but step not continue.
10734 @tab @code{t}@var{addr}@code{:}@var{PP}@code{,}@var{MM}
10736 Search backwards starting at address @var{addr} for a match with pattern
10737 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4
10738 bytes. @var{addr} must be at least 3 digits.
10741 @tab @code{T}@var{XX}
10742 @tab Find out if the thread XX is alive.
10744 @tab reply @code{OK}
10745 @tab thread is still alive
10747 @tab reply @code{E}@var{NN}
10748 @tab thread is dead
10752 @tab Reserved for future use
10756 @tab Reserved for future use
10760 @tab Reserved for future use
10764 @tab Reserved for future use
10768 @tab Reserved for future use
10772 @tab Reserved for future use
10776 @tab Reserved for future use
10778 @item write mem (binary)
10779 @tab @code{X}@var{addr}@code{,}@var{length}@var{:}@var{XX...}
10781 @var{addr} is address, @var{length} is number of bytes, @var{XX...} is
10782 binary data. The characters @code{$}, @code{#}, and @code{0x7d} are
10783 escaped using @code{0x7d}.
10785 @tab reply @code{OK}
10788 @tab reply @code{E}@var{NN}
10793 @tab Reserved for future use
10797 @tab Reserved for future use
10799 @item remove break or watchpoint @strong{(draft)}
10800 @tab @code{z}@var{t}@code{,}@var{addr}@code{,}@var{length}
10804 @item insert break or watchpoint @strong{(draft)}
10805 @tab @code{Z}@var{t}@code{,}@var{addr}@code{,}@var{length}
10807 @var{t} is type: @samp{0} - software breakpoint, @samp{1} - hardware
10808 breakpoint, @samp{2} - write watchpoint, @samp{3} - read watchpoint,
10809 @samp{4} - access watchpoint; @var{addr} is address; @var{length} is in
10810 bytes. For a software breakpoint, @var{length} specifies the size of
10811 the instruction to be patched. For hardware breakpoints and watchpoints
10812 @var{length} specifies the memory region to be monitored. To avoid
10813 potential problems with duplicate packets, the operations should be
10814 implemented in an idempotent way.
10816 @tab reply @code{E}@var{NN}
10819 @tab reply @code{OK}
10823 @tab If not supported.
10827 @tab Reserved for future use
10831 The @samp{C}, @samp{c}, @samp{S}, @samp{s} and @samp{?} packets can
10832 receive any of the below as a reply. In the case of the @samp{C},
10833 @samp{c}, @samp{S} and @samp{s} packets, that reply is only returned
10834 when the target halts. In the below the exact meaning of @samp{signal
10835 number} is poorly defined. In general one of the UNIX signal numbering
10836 conventions is used.
10838 @multitable @columnfractions .4 .6
10840 @item @code{S}@var{AA}
10841 @tab @var{AA} is the signal number
10843 @item @code{T}@var{AA}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}
10845 @var{AA} = two hex digit signal number; @var{n...} = register number
10846 (hex), @var{r...} = target byte ordered register contents, size defined
10847 by @code{REGISTER_RAW_SIZE}; @var{n...} = @samp{thread}, @var{r...} =
10848 thread process ID, this is a hex integer; @var{n...} = other string not
10849 starting with valid hex digit. @value{GDBN} should ignore this
10850 @var{n...}, @var{r...} pair and go on to the next. This way we can
10851 extend the protocol.
10853 @item @code{W}@var{AA}
10855 The process exited, and @var{AA} is the exit status. This is only
10856 applicable for certains sorts of targets.
10858 @item @code{X}@var{AA}
10860 The process terminated with signal @var{AA}.
10862 @item @code{N}@var{AA}@code{;}@var{t...}@code{;}@var{d...}@code{;}@var{b...} @strong{(obsolete)}
10864 @var{AA} = signal number; @var{t...} = address of symbol "_start";
10865 @var{d...} = base of data section; @var{b...} = base of bss section.
10866 @emph{Note: only used by Cisco Systems targets. The difference between
10867 this reply and the "qOffsets" query is that the 'N' packet may arrive
10868 spontaneously whereas the 'qOffsets' is a query initiated by the host
10871 @item @code{O}@var{XX...}
10873 @var{XX...} is hex encoding of @sc{ascii} data. This can happen at any time
10874 while the program is running and the debugger should continue to wait
10879 The following set and query packets have already been defined.
10881 @multitable @columnfractions .2 .2 .6
10883 @item current thread
10884 @tab @code{q}@code{C}
10885 @tab Return the current thread id.
10887 @tab reply @code{QC}@var{pid}
10889 Where @var{pid} is a HEX encoded 16 bit process id.
10892 @tab Any other reply implies the old pid.
10894 @item all thread ids
10895 @tab @code{q}@code{fThreadInfo}
10897 @tab @code{q}@code{sThreadInfo}
10899 Obtain a list of active thread ids from the target (OS). Since there
10900 may be too many active threads to fit into one reply packet, this query
10901 works iteratively: it may require more than one query/reply sequence to
10902 obtain the entire list of threads. The first query of the sequence will
10903 be the @code{qf}@code{ThreadInfo} query; subsequent queries in the
10904 sequence will be the @code{qs}@code{ThreadInfo} query.
10907 @tab NOTE: replaces the @code{qL} query (see below).
10909 @tab reply @code{m}@var{<id>}
10910 @tab A single thread id
10912 @tab reply @code{m}@var{<id>},@var{<id>...}
10913 @tab a comma-separated list of thread ids
10915 @tab reply @code{l}
10916 @tab (lower case 'el') denotes end of list.
10920 In response to each query, the target will reply with a list of one
10921 or more thread ids, in big-endian hex, separated by commas. GDB will
10922 respond to each reply with a request for more thread ids (using the
10923 @code{qs} form of the query), until the target responds with @code{l}
10924 (lower-case el, for @code{'last'}).
10926 @item extra thread info
10927 @tab @code{q}@code{ThreadExtraInfo}@code{,}@var{id}
10932 Where @var{<id>} is a thread-id in big-endian hex.
10933 Obtain a printable string description of a thread's attributes from
10934 the target OS. This string may contain anything that the target OS
10935 thinks is interesting for @value{GDBN} to tell the user about the thread.
10936 The string is displayed in @value{GDBN}'s @samp{info threads} display.
10937 Some examples of possible thread extra info strings are "Runnable", or
10938 "Blocked on Mutex".
10940 @tab reply @var{XX...}
10942 Where @var{XX...} is a hex encoding of @sc{ascii} data, comprising the
10943 printable string containing the extra information about the thread's
10946 @item query @var{LIST} or @var{threadLIST} @strong{(deprecated)}
10947 @tab @code{q}@code{L}@var{startflag}@var{threadcount}@var{nextthread}
10952 Obtain thread information from RTOS. Where: @var{startflag} (one hex
10953 digit) is one to indicate the first query and zero to indicate a
10954 subsequent query; @var{threadcount} (two hex digits) is the maximum
10955 number of threads the response packet can contain; and @var{nextthread}
10956 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
10957 returned in the response as @var{argthread}.
10960 @tab NOTE: this query is replaced by the @code{q}@code{fThreadInfo}
10963 @tab reply @code{q}@code{M}@var{count}@var{done}@var{argthread}@var{thread...}
10968 Where: @var{count} (two hex digits) is the number of threads being
10969 returned; @var{done} (one hex digit) is zero to indicate more threads
10970 and one indicates no further threads; @var{argthreadid} (eight hex
10971 digits) is @var{nextthread} from the request packet; @var{thread...} is
10972 a sequence of thread IDs from the target. @var{threadid} (eight hex
10973 digits). See @code{remote.c:parse_threadlist_response()}.
10975 @item compute CRC of memory block
10976 @tab @code{q}@code{CRC:}@var{addr}@code{,}@var{length}
10979 @tab reply @code{E}@var{NN}
10980 @tab An error (such as memory fault)
10982 @tab reply @code{C}@var{CRC32}
10983 @tab A 32 bit cyclic redundancy check of the specified memory region.
10985 @item query sect offs
10986 @tab @code{q}@code{Offsets}
10988 Get section offsets that the target used when re-locating the downloaded
10989 image. @emph{Note: while a @code{Bss} offset is included in the
10990 response, @value{GDBN} ignores this and instead applies the @code{Data}
10991 offset to the @code{Bss} section.}
10993 @tab reply @code{Text=}@var{xxx}@code{;Data=}@var{yyy}@code{;Bss=}@var{zzz}
10995 @item thread info request
10996 @tab @code{q}@code{P}@var{mode}@var{threadid}
11001 Returns information on @var{threadid}. Where: @var{mode} is a hex
11002 encoded 32 bit mode; @var{threadid} is a hex encoded 64 bit thread ID.
11006 See @code{remote.c:remote_unpack_thread_info_response()}.
11008 @item remote command
11009 @tab @code{q}@code{Rcmd,}@var{COMMAND}
11014 @var{COMMAND} (hex encoded) is passed to the local interpreter for
11015 execution. Invalid commands should be reported using the output string.
11016 Before the final result packet, the target may also respond with a
11017 number of intermediate @code{O}@var{OUTPUT} console output
11018 packets. @emph{Implementors should note that providing access to a
11019 stubs's interpreter may have security implications}.
11021 @tab reply @code{OK}
11023 A command response with no output.
11025 @tab reply @var{OUTPUT}
11027 A command response with the hex encoded output string @var{OUTPUT}.
11029 @tab reply @code{E}@var{NN}
11031 Indicate a badly formed request.
11036 When @samp{q}@samp{Rcmd} is not recognized.
11038 @item symbol lookup
11039 @tab @code{qSymbol::}
11041 Notify the target that @value{GDBN} is prepared to serve symbol lookup
11042 requests. Accept requests from the target for the values of symbols.
11047 @tab reply @code{OK}
11049 The target does not need to look up any (more) symbols.
11051 @tab reply @code{qSymbol:}@var{sym_name}
11053 The target requests the value of symbol @var{sym_name} (hex encoded).
11054 @value{GDBN} may provide the value by using the
11055 @code{qSymbol:}@var{sym_value}:@var{sym_name}
11056 message, described below.
11059 @tab @code{qSymbol:}@var{sym_value}:@var{sym_name}
11061 Set the value of SYM_NAME to SYM_VALUE.
11065 @var{sym_name} (hex encoded) is the name of a symbol whose value
11066 the target has previously requested.
11070 @var{sym_value} (hex) is the value for symbol @var{sym_name}.
11071 If @value{GDBN} cannot supply a value for @var{sym_name}, then this
11072 field will be empty.
11074 @tab reply @code{OK}
11076 The target does not need to look up any (more) symbols.
11078 @tab reply @code{qSymbol:}@var{sym_name}
11080 The target requests the value of a new symbol @var{sym_name} (hex encoded).
11081 @value{GDBN} will continue to supply the values of symbols (if available),
11082 until the target ceases to request them.
11086 The following @samp{g}/@samp{G} packets have previously been defined.
11087 In the below, some thirty-two bit registers are transferred as sixty-four
11088 bits. Those registers should be zero/sign extended (which?) to fill the
11089 space allocated. Register bytes are transfered in target byte order.
11090 The two nibbles within a register byte are transfered most-significant -
11093 @multitable @columnfractions .5 .5
11097 All registers are transfered as thirty-two bit quantities in the order:
11098 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
11099 registers; fsr; fir; fp.
11103 All registers are transfered as sixty-four bit quantities (including
11104 thirty-two bit registers such as @code{sr}). The ordering is the same
11109 Example sequence of a target being re-started. Notice how the restart
11110 does not get any direct output:
11115 @emph{target restarts}
11118 -> @code{T001:1234123412341234}
11122 Example sequence of a target being stepped by a single instruction:
11130 -> @code{T001:1234123412341234}
11139 @subsubsection Using the @code{gdbserver} program
11142 @cindex remote connection without stubs
11143 @code{gdbserver} is a control program for Unix-like systems, which
11144 allows you to connect your program with a remote @value{GDBN} via
11145 @code{target remote}---but without linking in the usual debugging stub.
11147 @code{gdbserver} is not a complete replacement for the debugging stubs,
11148 because it requires essentially the same operating-system facilities
11149 that @value{GDBN} itself does. In fact, a system that can run
11150 @code{gdbserver} to connect to a remote @value{GDBN} could also run
11151 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
11152 because it is a much smaller program than @value{GDBN} itself. It is
11153 also easier to port than all of @value{GDBN}, so you may be able to get
11154 started more quickly on a new system by using @code{gdbserver}.
11155 Finally, if you develop code for real-time systems, you may find that
11156 the tradeoffs involved in real-time operation make it more convenient to
11157 do as much development work as possible on another system, for example
11158 by cross-compiling. You can use @code{gdbserver} to make a similar
11159 choice for debugging.
11161 @value{GDBN} and @code{gdbserver} communicate via either a serial line
11162 or a TCP connection, using the standard @value{GDBN} remote serial
11166 @item On the target machine,
11167 you need to have a copy of the program you want to debug.
11168 @code{gdbserver} does not need your program's symbol table, so you can
11169 strip the program if necessary to save space. @value{GDBN} on the host
11170 system does all the symbol handling.
11172 To use the server, you must tell it how to communicate with @value{GDBN};
11173 the name of your program; and the arguments for your program. The
11177 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
11180 @var{comm} is either a device name (to use a serial line) or a TCP
11181 hostname and portnumber. For example, to debug Emacs with the argument
11182 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
11186 target> gdbserver /dev/com1 emacs foo.txt
11189 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
11192 To use a TCP connection instead of a serial line:
11195 target> gdbserver host:2345 emacs foo.txt
11198 The only difference from the previous example is the first argument,
11199 specifying that you are communicating with the host @value{GDBN} via
11200 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
11201 expect a TCP connection from machine @samp{host} to local TCP port 2345.
11202 (Currently, the @samp{host} part is ignored.) You can choose any number
11203 you want for the port number as long as it does not conflict with any
11204 TCP ports already in use on the target system (for example, @code{23} is
11205 reserved for @code{telnet}).@footnote{If you choose a port number that
11206 conflicts with another service, @code{gdbserver} prints an error message
11207 and exits.} You must use the same port number with the host @value{GDBN}
11208 @code{target remote} command.
11210 @item On the @value{GDBN} host machine,
11211 you need an unstripped copy of your program, since @value{GDBN} needs
11212 symbols and debugging information. Start up @value{GDBN} as usual,
11213 using the name of the local copy of your program as the first argument.
11214 (You may also need the @w{@samp{--baud}} option if the serial line is
11215 running at anything other than 9600@dmn{bps}.) After that, use @code{target
11216 remote} to establish communications with @code{gdbserver}. Its argument
11217 is either a device name (usually a serial device, like
11218 @file{/dev/ttyb}), or a TCP port descriptor in the form
11219 @code{@var{host}:@var{PORT}}. For example:
11222 (@value{GDBP}) target remote /dev/ttyb
11226 communicates with the server via serial line @file{/dev/ttyb}, and
11229 (@value{GDBP}) target remote the-target:2345
11233 communicates via a TCP connection to port 2345 on host @w{@file{the-target}}.
11234 For TCP connections, you must start up @code{gdbserver} prior to using
11235 the @code{target remote} command. Otherwise you may get an error whose
11236 text depends on the host system, but which usually looks something like
11237 @samp{Connection refused}.
11241 @subsubsection Using the @code{gdbserve.nlm} program
11243 @kindex gdbserve.nlm
11244 @code{gdbserve.nlm} is a control program for NetWare systems, which
11245 allows you to connect your program with a remote @value{GDBN} via
11246 @code{target remote}.
11248 @value{GDBN} and @code{gdbserve.nlm} communicate via a serial line,
11249 using the standard @value{GDBN} remote serial protocol.
11252 @item On the target machine,
11253 you need to have a copy of the program you want to debug.
11254 @code{gdbserve.nlm} does not need your program's symbol table, so you
11255 can strip the program if necessary to save space. @value{GDBN} on the
11256 host system does all the symbol handling.
11258 To use the server, you must tell it how to communicate with
11259 @value{GDBN}; the name of your program; and the arguments for your
11260 program. The syntax is:
11263 load gdbserve [ BOARD=@var{board} ] [ PORT=@var{port} ]
11264 [ BAUD=@var{baud} ] @var{program} [ @var{args} @dots{} ]
11267 @var{board} and @var{port} specify the serial line; @var{baud} specifies
11268 the baud rate used by the connection. @var{port} and @var{node} default
11269 to 0, @var{baud} defaults to 9600@dmn{bps}.
11271 For example, to debug Emacs with the argument @samp{foo.txt}and
11272 communicate with @value{GDBN} over serial port number 2 or board 1
11273 using a 19200@dmn{bps} connection:
11276 load gdbserve BOARD=1 PORT=2 BAUD=19200 emacs foo.txt
11279 @item On the @value{GDBN} host machine,
11280 you need an unstripped copy of your program, since @value{GDBN} needs
11281 symbols and debugging information. Start up @value{GDBN} as usual,
11282 using the name of the local copy of your program as the first argument.
11283 (You may also need the @w{@samp{--baud}} option if the serial line is
11284 running at anything other than 9600@dmn{bps}. After that, use @code{target
11285 remote} to establish communications with @code{gdbserve.nlm}. Its
11286 argument is a device name (usually a serial device, like
11287 @file{/dev/ttyb}). For example:
11290 (@value{GDBP}) target remote /dev/ttyb
11294 communications with the server via serial line @file{/dev/ttyb}.
11298 @section Kernel Object Display
11300 @cindex kernel object display
11301 @cindex kernel object
11304 Some targets support kernel object display. Using this facility,
11305 @value{GDBN} communicates specially with the underlying operating system
11306 and can display information about operating system-level objects such as
11307 mutexes and other synchronization objects. Exactly which objects can be
11308 displayed is determined on a per-OS basis.
11310 Use the @code{set os} command to set the operating system. This tells
11311 @value{GDBN} which kernel object display module to initialize:
11314 (@value{GDBP}) set os cisco
11317 If @code{set os} succeeds, @value{GDBN} will display some information
11318 about the operating system, and will create a new @code{info} command
11319 which can be used to query the target. The @code{info} command is named
11320 after the operating system:
11323 (@value{GDBP}) info cisco
11324 List of Cisco Kernel Objects
11326 any Any and all objects
11329 Further subcommands can be used to query about particular objects known
11332 There is currently no way to determine whether a given operating system
11333 is supported other than to try it.
11336 @node Configurations
11337 @chapter Configuration-Specific Information
11339 While nearly all @value{GDBN} commands are available for all native and
11340 cross versions of the debugger, there are some exceptions. This chapter
11341 describes things that are only available in certain configurations.
11343 There are three major categories of configurations: native
11344 configurations, where the host and target are the same, embedded
11345 operating system configurations, which are usually the same for several
11346 different processor architectures, and bare embedded processors, which
11347 are quite different from each other.
11352 * Embedded Processors::
11359 This section describes details specific to particular native
11364 * SVR4 Process Information:: SVR4 process information
11365 * DJGPP Native:: Features specific to the DJGPP port
11371 On HP-UX systems, if you refer to a function or variable name that
11372 begins with a dollar sign, @value{GDBN} searches for a user or system
11373 name first, before it searches for a convenience variable.
11375 @node SVR4 Process Information
11376 @subsection SVR4 process information
11379 @cindex process image
11381 Many versions of SVR4 provide a facility called @samp{/proc} that can be
11382 used to examine the image of a running process using file-system
11383 subroutines. If @value{GDBN} is configured for an operating system with
11384 this facility, the command @code{info proc} is available to report on
11385 several kinds of information about the process running your program.
11386 @code{info proc} works only on SVR4 systems that include the
11387 @code{procfs} code. This includes OSF/1 (Digital Unix), Solaris, Irix,
11388 and Unixware, but not HP-UX or Linux, for example.
11393 Summarize available information about the process.
11395 @kindex info proc mappings
11396 @item info proc mappings
11397 Report on the address ranges accessible in the program, with information
11398 on whether your program may read, write, or execute each range.
11400 @comment These sub-options of 'info proc' were not included when
11401 @comment procfs.c was re-written. Keep their descriptions around
11402 @comment against the day when someone finds the time to put them back in.
11403 @kindex info proc times
11404 @item info proc times
11405 Starting time, user CPU time, and system CPU time for your program and
11408 @kindex info proc id
11410 Report on the process IDs related to your program: its own process ID,
11411 the ID of its parent, the process group ID, and the session ID.
11413 @kindex info proc status
11414 @item info proc status
11415 General information on the state of the process. If the process is
11416 stopped, this report includes the reason for stopping, and any signal
11419 @item info proc all
11420 Show all the above information about the process.
11425 @subsection Features for Debugging @sc{djgpp} Programs
11426 @cindex @sc{djgpp} debugging
11427 @cindex native @sc{djgpp} debugging
11428 @cindex MS-DOS-specific commands
11430 @sc{djgpp} is the port of @sc{gnu} development tools to MS-DOS and
11431 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
11432 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
11433 top of real-mode DOS systems and their emulations.
11435 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
11436 defines a few commands specific to the @sc{djgpp} port. This
11437 subsection describes those commands.
11442 This is a prefix of @sc{djgpp}-specific commands which print
11443 information about the target system and important OS structures.
11446 @cindex MS-DOS system info
11447 @cindex free memory information (MS-DOS)
11448 @item info dos sysinfo
11449 This command displays assorted information about the underlying
11450 platform: the CPU type and features, the OS version and flavor, the
11451 DPMI version, and the available conventional and DPMI memory.
11456 @cindex segment descriptor tables
11457 @cindex descriptor tables display
11459 @itemx info dos ldt
11460 @itemx info dos idt
11461 These 3 commands display entries from, respectively, Global, Local,
11462 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
11463 tables are data structures which store a descriptor for each segment
11464 that is currently in use. The segment's selector is an index into a
11465 descriptor table; the table entry for that index holds the
11466 descriptor's base address and limit, and its attributes and access
11469 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
11470 segment (used for both data and the stack), and a DOS segment (which
11471 allows access to DOS/BIOS data structures and absolute addresses in
11472 conventional memory). However, the DPMI host will usually define
11473 additional segments in order to support the DPMI environment.
11475 @cindex garbled pointers
11476 These commands allow to display entries from the descriptor tables.
11477 Without an argument, all entries from the specified table are
11478 displayed. An argument, which should be an integer expression, means
11479 display a single entry whose index is given by the argument. For
11480 example, here's a convenient way to display information about the
11481 debugged program's data segment:
11484 (@value{GDBP}) info dos ldt $ds
11485 0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)
11489 This comes in handy when you want to see whether a pointer is outside
11490 the data segment's limit (i.e.@: @dfn{garbled}).
11492 @cindex page tables display (MS-DOS)
11494 @itemx info dos pte
11495 These two commands display entries from, respectively, the Page
11496 Directory and the Page Tables. Page Directories and Page Tables are
11497 data structures which control how virtual memory addresses are mapped
11498 into physical addresses. A Page Table includes an entry for every
11499 page of memory that is mapped into the program's address space; there
11500 may be several Page Tables, each one holding up to 4096 entries. A
11501 Page Directory has up to 4096 entries, one each for every Page Table
11502 that is currently in use.
11504 Without an argument, @kbd{info dos pde} displays the entire Page
11505 Directory, and @kbd{info dos pte} displays all the entries in all of
11506 the Page Tables. An argument, an integer expression, given to the
11507 @kbd{info dos pde} command means display only that entry from the Page
11508 Directory table. An argument given to the @kbd{info dos pte} command
11509 means display entries from a single Page Table, the one pointed to by
11510 the specified entry in the Page Directory.
11512 These commands are useful when your program uses @dfn{DMA} (Direct
11513 Memory Access), which needs physical addresses to program the DMA
11516 These commands are supported only with some DPMI servers.
11518 @cindex physical address from linear address
11519 @item info dos address-pte
11520 This command displays the Page Table entry for a specified linear
11521 address. The argument linear address should already have the
11522 appropriate segment's base address added to it, because this command
11523 accepts addresses which may belong to @emph{any} segment. For
11524 example, here's how to display the Page Table entry for the page where
11525 the variable @code{i} is stored:
11528 (@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i
11529 Page Table entry for address 0x11a00d30:
11530 Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30
11534 This says that @code{i} is stored at offset @code{0xd30} from the page
11535 whose physical base address is @code{0x02698000}, and prints all the
11536 attributes of that page.
11538 Note that you must cast the addresses of variables to a @code{char *},
11539 since otherwise the value of @code{__djgpp_base_address}, the base
11540 address of all variables and functions in a @sc{djgpp} program, will
11541 be added using the rules of C pointer arithmetics: if @code{i} is
11542 declared an @code{int}, @value{GDBN} will add 4 times the value of
11543 @code{__djgpp_base_address} to the address of @code{i}.
11545 Here's another example, it displays the Page Table entry for the
11549 (@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)
11550 Page Table entry for address 0x29110:
11551 Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110
11555 (The @code{+ 3} offset is because the transfer buffer's address is the
11556 3rd member of the @code{_go32_info_block} structure.) The output of
11557 this command clearly shows that addresses in conventional memory are
11558 mapped 1:1, i.e.@: the physical and linear addresses are identical.
11560 This command is supported only with some DPMI servers.
11564 @section Embedded Operating Systems
11566 This section describes configurations involving the debugging of
11567 embedded operating systems that are available for several different
11571 * VxWorks:: Using @value{GDBN} with VxWorks
11574 @value{GDBN} includes the ability to debug programs running on
11575 various real-time operating systems.
11578 @subsection Using @value{GDBN} with VxWorks
11584 @kindex target vxworks
11585 @item target vxworks @var{machinename}
11586 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
11587 is the target system's machine name or IP address.
11591 On VxWorks, @code{load} links @var{filename} dynamically on the
11592 current target system as well as adding its symbols in @value{GDBN}.
11594 @value{GDBN} enables developers to spawn and debug tasks running on networked
11595 VxWorks targets from a Unix host. Already-running tasks spawned from
11596 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
11597 both the Unix host and on the VxWorks target. The program
11598 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
11599 installed with the name @code{vxgdb}, to distinguish it from a
11600 @value{GDBN} for debugging programs on the host itself.)
11603 @item VxWorks-timeout @var{args}
11604 @kindex vxworks-timeout
11605 All VxWorks-based targets now support the option @code{vxworks-timeout}.
11606 This option is set by the user, and @var{args} represents the number of
11607 seconds @value{GDBN} waits for responses to rpc's. You might use this if
11608 your VxWorks target is a slow software simulator or is on the far side
11609 of a thin network line.
11612 The following information on connecting to VxWorks was current when
11613 this manual was produced; newer releases of VxWorks may use revised
11616 @kindex INCLUDE_RDB
11617 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
11618 to include the remote debugging interface routines in the VxWorks
11619 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
11620 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
11621 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
11622 source debugging task @code{tRdbTask} when VxWorks is booted. For more
11623 information on configuring and remaking VxWorks, see the manufacturer's
11625 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
11627 Once you have included @file{rdb.a} in your VxWorks system image and set
11628 your Unix execution search path to find @value{GDBN}, you are ready to
11629 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
11630 @code{vxgdb}, depending on your installation).
11632 @value{GDBN} comes up showing the prompt:
11639 * VxWorks Connection:: Connecting to VxWorks
11640 * VxWorks Download:: VxWorks download
11641 * VxWorks Attach:: Running tasks
11644 @node VxWorks Connection
11645 @subsubsection Connecting to VxWorks
11647 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
11648 network. To connect to a target whose host name is ``@code{tt}'', type:
11651 (vxgdb) target vxworks tt
11655 @value{GDBN} displays messages like these:
11658 Attaching remote machine across net...
11663 @value{GDBN} then attempts to read the symbol tables of any object modules
11664 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
11665 these files by searching the directories listed in the command search
11666 path (@pxref{Environment, ,Your program's environment}); if it fails
11667 to find an object file, it displays a message such as:
11670 prog.o: No such file or directory.
11673 When this happens, add the appropriate directory to the search path with
11674 the @value{GDBN} command @code{path}, and execute the @code{target}
11677 @node VxWorks Download
11678 @subsubsection VxWorks download
11680 @cindex download to VxWorks
11681 If you have connected to the VxWorks target and you want to debug an
11682 object that has not yet been loaded, you can use the @value{GDBN}
11683 @code{load} command to download a file from Unix to VxWorks
11684 incrementally. The object file given as an argument to the @code{load}
11685 command is actually opened twice: first by the VxWorks target in order
11686 to download the code, then by @value{GDBN} in order to read the symbol
11687 table. This can lead to problems if the current working directories on
11688 the two systems differ. If both systems have NFS mounted the same
11689 filesystems, you can avoid these problems by using absolute paths.
11690 Otherwise, it is simplest to set the working directory on both systems
11691 to the directory in which the object file resides, and then to reference
11692 the file by its name, without any path. For instance, a program
11693 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
11694 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
11695 program, type this on VxWorks:
11698 -> cd "@var{vxpath}/vw/demo/rdb"
11702 Then, in @value{GDBN}, type:
11705 (vxgdb) cd @var{hostpath}/vw/demo/rdb
11706 (vxgdb) load prog.o
11709 @value{GDBN} displays a response similar to this:
11712 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
11715 You can also use the @code{load} command to reload an object module
11716 after editing and recompiling the corresponding source file. Note that
11717 this makes @value{GDBN} delete all currently-defined breakpoints,
11718 auto-displays, and convenience variables, and to clear the value
11719 history. (This is necessary in order to preserve the integrity of
11720 debugger's data structures that reference the target system's symbol
11723 @node VxWorks Attach
11724 @subsubsection Running tasks
11726 @cindex running VxWorks tasks
11727 You can also attach to an existing task using the @code{attach} command as
11731 (vxgdb) attach @var{task}
11735 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
11736 or suspended when you attach to it. Running tasks are suspended at
11737 the time of attachment.
11739 @node Embedded Processors
11740 @section Embedded Processors
11742 This section goes into details specific to particular embedded
11746 * A29K Embedded:: AMD A29K Embedded
11748 * H8/300:: Hitachi H8/300
11749 * H8/500:: Hitachi H8/500
11750 * i960:: Intel i960
11751 * M32R/D:: Mitsubishi M32R/D
11752 * M68K:: Motorola M68K
11753 * M88K:: Motorola M88K
11754 * MIPS Embedded:: MIPS Embedded
11755 * PA:: HP PA Embedded
11758 * Sparclet:: Tsqware Sparclet
11759 * Sparclite:: Fujitsu Sparclite
11760 * ST2000:: Tandem ST2000
11761 * Z8000:: Zilog Z8000
11764 @node A29K Embedded
11765 @subsection AMD A29K Embedded
11770 * Comms (EB29K):: Communications setup
11771 * gdb-EB29K:: EB29K cross-debugging
11772 * Remote Log:: Remote log
11777 @kindex target adapt
11778 @item target adapt @var{dev}
11779 Adapt monitor for A29K.
11781 @kindex target amd-eb
11782 @item target amd-eb @var{dev} @var{speed} @var{PROG}
11784 Remote PC-resident AMD EB29K board, attached over serial lines.
11785 @var{dev} is the serial device, as for @code{target remote};
11786 @var{speed} allows you to specify the linespeed; and @var{PROG} is the
11787 name of the program to be debugged, as it appears to DOS on the PC.
11788 @xref{A29K EB29K, ,EBMON protocol for AMD29K}.
11793 @subsubsection A29K UDI
11796 @cindex AMD29K via UDI
11798 @value{GDBN} supports AMD's UDI (``Universal Debugger Interface'')
11799 protocol for debugging the a29k processor family. To use this
11800 configuration with AMD targets running the MiniMON monitor, you need the
11801 program @code{MONTIP}, available from AMD at no charge. You can also
11802 use @value{GDBN} with the UDI-conformant a29k simulator program
11803 @code{ISSTIP}, also available from AMD.
11806 @item target udi @var{keyword}
11808 Select the UDI interface to a remote a29k board or simulator, where
11809 @var{keyword} is an entry in the AMD configuration file @file{udi_soc}.
11810 This file contains keyword entries which specify parameters used to
11811 connect to a29k targets. If the @file{udi_soc} file is not in your
11812 working directory, you must set the environment variable @samp{UDICONF}
11817 @subsubsection EBMON protocol for AMD29K
11819 @cindex EB29K board
11820 @cindex running 29K programs
11822 AMD distributes a 29K development board meant to fit in a PC, together
11823 with a DOS-hosted monitor program called @code{EBMON}. As a shorthand
11824 term, this development system is called the ``EB29K''. To use
11825 @value{GDBN} from a Unix system to run programs on the EB29K board, you
11826 must first connect a serial cable between the PC (which hosts the EB29K
11827 board) and a serial port on the Unix system. In the following, we
11828 assume you've hooked the cable between the PC's @file{COM1} port and
11829 @file{/dev/ttya} on the Unix system.
11831 @node Comms (EB29K)
11832 @subsubsection Communications setup
11834 The next step is to set up the PC's port, by doing something like this
11838 C:\> MODE com1:9600,n,8,1,none
11842 This example---run on an MS DOS 4.0 system---sets the PC port to 9600
11843 bps, no parity, eight data bits, one stop bit, and no ``retry'' action;
11844 you must match the communications parameters when establishing the Unix
11845 end of the connection as well.
11846 @c FIXME: Who knows what this "no retry action" crud from the DOS manual may
11847 @c mean? It's optional; leave it out? ---doc@cygnus.com, 25feb91
11849 @c It's optional, but it's unwise to omit it: who knows what is the
11850 @c default value set when the DOS machines boots? "No retry" means that
11851 @c the DOS serial device driver won't retry the operation if it fails;
11852 @c I understand that this is needed because the GDB serial protocol
11853 @c handles any errors and retransmissions itself. ---Eli Zaretskii, 3sep99
11855 To give control of the PC to the Unix side of the serial line, type
11856 the following at the DOS console:
11863 (Later, if you wish to return control to the DOS console, you can use
11864 the command @code{CTTY con}---but you must send it over the device that
11865 had control, in our example over the @file{COM1} serial line.)
11867 From the Unix host, use a communications program such as @code{tip} or
11868 @code{cu} to communicate with the PC; for example,
11871 cu -s 9600 -l /dev/ttya
11875 The @code{cu} options shown specify, respectively, the linespeed and the
11876 serial port to use. If you use @code{tip} instead, your command line
11877 may look something like the following:
11880 tip -9600 /dev/ttya
11884 Your system may require a different name where we show
11885 @file{/dev/ttya} as the argument to @code{tip}. The communications
11886 parameters, including which port to use, are associated with the
11887 @code{tip} argument in the ``remote'' descriptions file---normally the
11888 system table @file{/etc/remote}.
11889 @c FIXME: What if anything needs doing to match the "n,8,1,none" part of
11890 @c the DOS side's comms setup? cu can support -o (odd
11891 @c parity), -e (even parity)---apparently no settings for no parity or
11892 @c for character size. Taken from stty maybe...? John points out tip
11893 @c can set these as internal variables, eg ~s parity=none; man stty
11894 @c suggests that it *might* work to stty these options with stdin or
11895 @c stdout redirected... ---doc@cygnus.com, 25feb91
11897 @c There's nothing to be done for the "none" part of the DOS MODE
11898 @c command. The rest of the parameters should be matched by the
11899 @c baudrate, bits, and parity used by the Unix side. ---Eli Zaretskii, 3Sep99
11902 Using the @code{tip} or @code{cu} connection, change the DOS working
11903 directory to the directory containing a copy of your 29K program, then
11904 start the PC program @code{EBMON} (an EB29K control program supplied
11905 with your board by AMD). You should see an initial display from
11906 @code{EBMON} similar to the one that follows, ending with the
11907 @code{EBMON} prompt @samp{#}---
11912 G:\> CD \usr\joe\work29k
11914 G:\USR\JOE\WORK29K> EBMON
11915 Am29000 PC Coprocessor Board Monitor, version 3.0-18
11916 Copyright 1990 Advanced Micro Devices, Inc.
11917 Written by Gibbons and Associates, Inc.
11919 Enter '?' or 'H' for help
11921 PC Coprocessor Type = EB29K
11923 Memory Base = 0xd0000
11925 Data Memory Size = 2048KB
11926 Available I-RAM Range = 0x8000 to 0x1fffff
11927 Available D-RAM Range = 0x80002000 to 0x801fffff
11930 Register Stack Size = 0x800
11931 Memory Stack Size = 0x1800
11934 Am29027 Available = No
11935 Byte Write Available = Yes
11940 Then exit the @code{cu} or @code{tip} program (done in the example by
11941 typing @code{~.} at the @code{EBMON} prompt). @code{EBMON} keeps
11942 running, ready for @value{GDBN} to take over.
11944 For this example, we've assumed what is probably the most convenient
11945 way to make sure the same 29K program is on both the PC and the Unix
11946 system: a PC/NFS connection that establishes ``drive @file{G:}'' on the
11947 PC as a file system on the Unix host. If you do not have PC/NFS or
11948 something similar connecting the two systems, you must arrange some
11949 other way---perhaps floppy-disk transfer---of getting the 29K program
11950 from the Unix system to the PC; @value{GDBN} does @emph{not} download it over the
11954 @subsubsection EB29K cross-debugging
11956 Finally, @code{cd} to the directory containing an image of your 29K
11957 program on the Unix system, and start @value{GDBN}---specifying as argument the
11958 name of your 29K program:
11961 cd /usr/joe/work29k
11966 Now you can use the @code{target} command:
11969 target amd-eb /dev/ttya 9600 MYFOO
11970 @c FIXME: test above 'target amd-eb' as spelled, with caps! caps are meant to
11971 @c emphasize that this is the name as seen by DOS (since I think DOS is
11972 @c single-minded about case of letters). ---doc@cygnus.com, 25feb91
11976 In this example, we've assumed your program is in a file called
11977 @file{myfoo}. Note that the filename given as the last argument to
11978 @code{target amd-eb} should be the name of the program as it appears to DOS.
11979 In our example this is simply @code{MYFOO}, but in general it can include
11980 a DOS path, and depending on your transfer mechanism may not resemble
11981 the name on the Unix side.
11983 At this point, you can set any breakpoints you wish; when you are ready
11984 to see your program run on the 29K board, use the @value{GDBN} command
11987 To stop debugging the remote program, use the @value{GDBN} @code{detach}
11990 To return control of the PC to its console, use @code{tip} or @code{cu}
11991 once again, after your @value{GDBN} session has concluded, to attach to
11992 @code{EBMON}. You can then type the command @code{q} to shut down
11993 @code{EBMON}, returning control to the DOS command-line interpreter.
11994 Type @kbd{CTTY con} to return command input to the main DOS console,
11995 and type @kbd{~.} to leave @code{tip} or @code{cu}.
11998 @subsubsection Remote log
11999 @cindex @file{eb.log}, a log file for EB29K
12000 @cindex log file for EB29K
12002 The @code{target amd-eb} command creates a file @file{eb.log} in the
12003 current working directory, to help debug problems with the connection.
12004 @file{eb.log} records all the output from @code{EBMON}, including echoes
12005 of the commands sent to it. Running @samp{tail -f} on this file in
12006 another window often helps to understand trouble with @code{EBMON}, or
12007 unexpected events on the PC side of the connection.
12015 @item target rdi @var{dev}
12016 ARM Angel monitor, via RDI library interface to ADP protocol. You may
12017 use this target to communicate with both boards running the Angel
12018 monitor, or with the EmbeddedICE JTAG debug device.
12021 @item target rdp @var{dev}
12027 @subsection Hitachi H8/300
12031 @kindex target hms@r{, with H8/300}
12032 @item target hms @var{dev}
12033 A Hitachi SH, H8/300, or H8/500 board, attached via serial line to your host.
12034 Use special commands @code{device} and @code{speed} to control the serial
12035 line and the communications speed used.
12037 @kindex target e7000@r{, with H8/300}
12038 @item target e7000 @var{dev}
12039 E7000 emulator for Hitachi H8 and SH.
12041 @kindex target sh3@r{, with H8/300}
12042 @kindex target sh3e@r{, with H8/300}
12043 @item target sh3 @var{dev}
12044 @itemx target sh3e @var{dev}
12045 Hitachi SH-3 and SH-3E target systems.
12049 @cindex download to H8/300 or H8/500
12050 @cindex H8/300 or H8/500 download
12051 @cindex download to Hitachi SH
12052 @cindex Hitachi SH download
12053 When you select remote debugging to a Hitachi SH, H8/300, or H8/500
12054 board, the @code{load} command downloads your program to the Hitachi
12055 board and also opens it as the current executable target for
12056 @value{GDBN} on your host (like the @code{file} command).
12058 @value{GDBN} needs to know these things to talk to your
12059 Hitachi SH, H8/300, or H8/500:
12063 that you want to use @samp{target hms}, the remote debugging interface
12064 for Hitachi microprocessors, or @samp{target e7000}, the in-circuit
12065 emulator for the Hitachi SH and the Hitachi 300H. (@samp{target hms} is
12066 the default when @value{GDBN} is configured specifically for the Hitachi SH,
12067 H8/300, or H8/500.)
12070 what serial device connects your host to your Hitachi board (the first
12071 serial device available on your host is the default).
12074 what speed to use over the serial device.
12078 * Hitachi Boards:: Connecting to Hitachi boards.
12079 * Hitachi ICE:: Using the E7000 In-Circuit Emulator.
12080 * Hitachi Special:: Special @value{GDBN} commands for Hitachi micros.
12083 @node Hitachi Boards
12084 @subsubsection Connecting to Hitachi boards
12086 @c only for Unix hosts
12088 @cindex serial device, Hitachi micros
12089 Use the special @code{@value{GDBN}} command @samp{device @var{port}} if you
12090 need to explicitly set the serial device. The default @var{port} is the
12091 first available port on your host. This is only necessary on Unix
12092 hosts, where it is typically something like @file{/dev/ttya}.
12095 @cindex serial line speed, Hitachi micros
12096 @code{@value{GDBN}} has another special command to set the communications
12097 speed: @samp{speed @var{bps}}. This command also is only used from Unix
12098 hosts; on DOS hosts, set the line speed as usual from outside @value{GDBN} with
12099 the DOS @code{mode} command (for instance,
12100 @w{@kbd{mode com2:9600,n,8,1,p}} for a 9600@dmn{bps} connection).
12102 The @samp{device} and @samp{speed} commands are available only when you
12103 use a Unix host to debug your Hitachi microprocessor programs. If you
12105 @value{GDBN} depends on an auxiliary terminate-and-stay-resident program
12106 called @code{asynctsr} to communicate with the development board
12107 through a PC serial port. You must also use the DOS @code{mode} command
12108 to set up the serial port on the DOS side.
12110 The following sample session illustrates the steps needed to start a
12111 program under @value{GDBN} control on an H8/300. The example uses a
12112 sample H8/300 program called @file{t.x}. The procedure is the same for
12113 the Hitachi SH and the H8/500.
12115 First hook up your development board. In this example, we use a
12116 board attached to serial port @code{COM2}; if you use a different serial
12117 port, substitute its name in the argument of the @code{mode} command.
12118 When you call @code{asynctsr}, the auxiliary comms program used by the
12119 debugger, you give it just the numeric part of the serial port's name;
12120 for example, @samp{asyncstr 2} below runs @code{asyncstr} on
12124 C:\H8300\TEST> asynctsr 2
12125 C:\H8300\TEST> mode com2:9600,n,8,1,p
12127 Resident portion of MODE loaded
12129 COM2: 9600, n, 8, 1, p
12134 @emph{Warning:} We have noticed a bug in PC-NFS that conflicts with
12135 @code{asynctsr}. If you also run PC-NFS on your DOS host, you may need to
12136 disable it, or even boot without it, to use @code{asynctsr} to control
12137 your development board.
12140 @kindex target hms@r{, and serial protocol}
12141 Now that serial communications are set up, and the development board is
12142 connected, you can start up @value{GDBN}. Call @code{@value{GDBP}} with
12143 the name of your program as the argument. @code{@value{GDBN}} prompts
12144 you, as usual, with the prompt @samp{(@value{GDBP})}. Use two special
12145 commands to begin your debugging session: @samp{target hms} to specify
12146 cross-debugging to the Hitachi board, and the @code{load} command to
12147 download your program to the board. @code{load} displays the names of
12148 the program's sections, and a @samp{*} for each 2K of data downloaded.
12149 (If you want to refresh @value{GDBN} data on symbols or on the
12150 executable file without downloading, use the @value{GDBN} commands
12151 @code{file} or @code{symbol-file}. These commands, and @code{load}
12152 itself, are described in @ref{Files,,Commands to specify files}.)
12155 (eg-C:\H8300\TEST) @value{GDBP} t.x
12156 @value{GDBN} is free software and you are welcome to distribute copies
12157 of it under certain conditions; type "show copying" to see
12159 There is absolutely no warranty for @value{GDBN}; type "show warranty"
12161 @value{GDBN} @value{GDBVN}, Copyright 1992 Free Software Foundation, Inc...
12162 (@value{GDBP}) target hms
12163 Connected to remote H8/300 HMS system.
12164 (@value{GDBP}) load t.x
12165 .text : 0x8000 .. 0xabde ***********
12166 .data : 0xabde .. 0xad30 *
12167 .stack : 0xf000 .. 0xf014 *
12170 At this point, you're ready to run or debug your program. From here on,
12171 you can use all the usual @value{GDBN} commands. The @code{break} command
12172 sets breakpoints; the @code{run} command starts your program;
12173 @code{print} or @code{x} display data; the @code{continue} command
12174 resumes execution after stopping at a breakpoint. You can use the
12175 @code{help} command at any time to find out more about @value{GDBN} commands.
12177 Remember, however, that @emph{operating system} facilities aren't
12178 available on your development board; for example, if your program hangs,
12179 you can't send an interrupt---but you can press the @sc{reset} switch!
12181 Use the @sc{reset} button on the development board
12184 to interrupt your program (don't use @kbd{ctl-C} on the DOS host---it has
12185 no way to pass an interrupt signal to the development board); and
12188 to return to the @value{GDBN} command prompt after your program finishes
12189 normally. The communications protocol provides no other way for @value{GDBN}
12190 to detect program completion.
12193 In either case, @value{GDBN} sees the effect of a @sc{reset} on the
12194 development board as a ``normal exit'' of your program.
12197 @subsubsection Using the E7000 in-circuit emulator
12199 @kindex target e7000@r{, with Hitachi ICE}
12200 You can use the E7000 in-circuit emulator to develop code for either the
12201 Hitachi SH or the H8/300H. Use one of these forms of the @samp{target
12202 e7000} command to connect @value{GDBN} to your E7000:
12205 @item target e7000 @var{port} @var{speed}
12206 Use this form if your E7000 is connected to a serial port. The
12207 @var{port} argument identifies what serial port to use (for example,
12208 @samp{com2}). The third argument is the line speed in bits per second
12209 (for example, @samp{9600}).
12211 @item target e7000 @var{hostname}
12212 If your E7000 is installed as a host on a TCP/IP network, you can just
12213 specify its hostname; @value{GDBN} uses @code{telnet} to connect.
12216 @node Hitachi Special
12217 @subsubsection Special @value{GDBN} commands for Hitachi micros
12219 Some @value{GDBN} commands are available only for the H8/300:
12223 @kindex set machine
12224 @kindex show machine
12225 @item set machine h8300
12226 @itemx set machine h8300h
12227 Condition @value{GDBN} for one of the two variants of the H8/300
12228 architecture with @samp{set machine}. You can use @samp{show machine}
12229 to check which variant is currently in effect.
12238 @kindex set memory @var{mod}
12239 @cindex memory models, H8/500
12240 @item set memory @var{mod}
12242 Specify which H8/500 memory model (@var{mod}) you are using with
12243 @samp{set memory}; check which memory model is in effect with @samp{show
12244 memory}. The accepted values for @var{mod} are @code{small},
12245 @code{big}, @code{medium}, and @code{compact}.
12250 @subsection Intel i960
12254 @kindex target mon960
12255 @item target mon960 @var{dev}
12256 MON960 monitor for Intel i960.
12258 @kindex target nindy
12259 @item target nindy @var{devicename}
12260 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
12261 the name of the serial device to use for the connection, e.g.
12268 @dfn{Nindy} is a ROM Monitor program for Intel 960 target systems. When
12269 @value{GDBN} is configured to control a remote Intel 960 using Nindy, you can
12270 tell @value{GDBN} how to connect to the 960 in several ways:
12274 Through command line options specifying serial port, version of the
12275 Nindy protocol, and communications speed;
12278 By responding to a prompt on startup;
12281 By using the @code{target} command at any point during your @value{GDBN}
12282 session. @xref{Target Commands, ,Commands for managing targets}.
12286 @cindex download to Nindy-960
12287 With the Nindy interface to an Intel 960 board, @code{load}
12288 downloads @var{filename} to the 960 as well as adding its symbols in
12292 * Nindy Startup:: Startup with Nindy
12293 * Nindy Options:: Options for Nindy
12294 * Nindy Reset:: Nindy reset command
12297 @node Nindy Startup
12298 @subsubsection Startup with Nindy
12300 If you simply start @code{@value{GDBP}} without using any command-line
12301 options, you are prompted for what serial port to use, @emph{before} you
12302 reach the ordinary @value{GDBN} prompt:
12305 Attach /dev/ttyNN -- specify NN, or "quit" to quit:
12309 Respond to the prompt with whatever suffix (after @samp{/dev/tty})
12310 identifies the serial port you want to use. You can, if you choose,
12311 simply start up with no Nindy connection by responding to the prompt
12312 with an empty line. If you do this and later wish to attach to Nindy,
12313 use @code{target} (@pxref{Target Commands, ,Commands for managing targets}).
12315 @node Nindy Options
12316 @subsubsection Options for Nindy
12318 These are the startup options for beginning your @value{GDBN} session with a
12319 Nindy-960 board attached:
12322 @item -r @var{port}
12323 Specify the serial port name of a serial interface to be used to connect
12324 to the target system. This option is only available when @value{GDBN} is
12325 configured for the Intel 960 target architecture. You may specify
12326 @var{port} as any of: a full pathname (e.g. @samp{-r /dev/ttya}), a
12327 device name in @file{/dev} (e.g. @samp{-r ttya}), or simply the unique
12328 suffix for a specific @code{tty} (e.g. @samp{-r a}).
12331 (An uppercase letter ``O'', not a zero.) Specify that @value{GDBN} should use
12332 the ``old'' Nindy monitor protocol to connect to the target system.
12333 This option is only available when @value{GDBN} is configured for the Intel 960
12334 target architecture.
12337 @emph{Warning:} if you specify @samp{-O}, but are actually trying to
12338 connect to a target system that expects the newer protocol, the connection
12339 fails, appearing to be a speed mismatch. @value{GDBN} repeatedly
12340 attempts to reconnect at several different line speeds. You can abort
12341 this process with an interrupt.
12345 Specify that @value{GDBN} should first send a @code{BREAK} signal to the target
12346 system, in an attempt to reset it, before connecting to a Nindy target.
12349 @emph{Warning:} Many target systems do not have the hardware that this
12350 requires; it only works with a few boards.
12354 The standard @samp{-b} option controls the line speed used on the serial
12359 @subsubsection Nindy reset command
12364 For a Nindy target, this command sends a ``break'' to the remote target
12365 system; this is only useful if the target has been equipped with a
12366 circuit to perform a hard reset (or some other interesting action) when
12367 a break is detected.
12372 @subsection Mitsubishi M32R/D
12376 @kindex target m32r
12377 @item target m32r @var{dev}
12378 Mitsubishi M32R/D ROM monitor.
12385 The Motorola m68k configuration includes ColdFire support, and
12386 target command for the following ROM monitors.
12390 @kindex target abug
12391 @item target abug @var{dev}
12392 ABug ROM monitor for M68K.
12394 @kindex target cpu32bug
12395 @item target cpu32bug @var{dev}
12396 CPU32BUG monitor, running on a CPU32 (M68K) board.
12398 @kindex target dbug
12399 @item target dbug @var{dev}
12400 dBUG ROM monitor for Motorola ColdFire.
12403 @item target est @var{dev}
12404 EST-300 ICE monitor, running on a CPU32 (M68K) board.
12406 @kindex target rom68k
12407 @item target rom68k @var{dev}
12408 ROM 68K monitor, running on an M68K IDP board.
12412 If @value{GDBN} is configured with @code{m68*-ericsson-*}, it will
12413 instead have only a single special target command:
12417 @kindex target es1800
12418 @item target es1800 @var{dev}
12419 ES-1800 emulator for M68K.
12427 @kindex target rombug
12428 @item target rombug @var{dev}
12429 ROMBUG ROM monitor for OS/9000.
12439 @item target bug @var{dev}
12440 BUG monitor, running on a MVME187 (m88k) board.
12444 @node MIPS Embedded
12445 @subsection MIPS Embedded
12447 @cindex MIPS boards
12448 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
12449 MIPS board attached to a serial line. This is available when
12450 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
12453 Use these @value{GDBN} commands to specify the connection to your target board:
12456 @item target mips @var{port}
12457 @kindex target mips @var{port}
12458 To run a program on the board, start up @code{@value{GDBP}} with the
12459 name of your program as the argument. To connect to the board, use the
12460 command @samp{target mips @var{port}}, where @var{port} is the name of
12461 the serial port connected to the board. If the program has not already
12462 been downloaded to the board, you may use the @code{load} command to
12463 download it. You can then use all the usual @value{GDBN} commands.
12465 For example, this sequence connects to the target board through a serial
12466 port, and loads and runs a program called @var{prog} through the
12470 host$ @value{GDBP} @var{prog}
12471 @value{GDBN} is free software and @dots{}
12472 (@value{GDBP}) target mips /dev/ttyb
12473 (@value{GDBP}) load @var{prog}
12477 @item target mips @var{hostname}:@var{portnumber}
12478 On some @value{GDBN} host configurations, you can specify a TCP
12479 connection (for instance, to a serial line managed by a terminal
12480 concentrator) instead of a serial port, using the syntax
12481 @samp{@var{hostname}:@var{portnumber}}.
12483 @item target pmon @var{port}
12484 @kindex target pmon @var{port}
12487 @item target ddb @var{port}
12488 @kindex target ddb @var{port}
12489 NEC's DDB variant of PMON for Vr4300.
12491 @item target lsi @var{port}
12492 @kindex target lsi @var{port}
12493 LSI variant of PMON.
12495 @kindex target r3900
12496 @item target r3900 @var{dev}
12497 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
12499 @kindex target array
12500 @item target array @var{dev}
12501 Array Tech LSI33K RAID controller board.
12507 @value{GDBN} also supports these special commands for MIPS targets:
12510 @item set processor @var{args}
12511 @itemx show processor
12512 @kindex set processor @var{args}
12513 @kindex show processor
12514 Use the @code{set processor} command to set the type of MIPS
12515 processor when you want to access processor-type-specific registers.
12516 For example, @code{set processor @var{r3041}} tells @value{GDBN}
12517 to use the CPU registers appropriate for the 3041 chip.
12518 Use the @code{show processor} command to see what MIPS processor @value{GDBN}
12519 is using. Use the @code{info reg} command to see what registers
12520 @value{GDBN} is using.
12522 @item set mipsfpu double
12523 @itemx set mipsfpu single
12524 @itemx set mipsfpu none
12525 @itemx show mipsfpu
12526 @kindex set mipsfpu
12527 @kindex show mipsfpu
12528 @cindex MIPS remote floating point
12529 @cindex floating point, MIPS remote
12530 If your target board does not support the MIPS floating point
12531 coprocessor, you should use the command @samp{set mipsfpu none} (if you
12532 need this, you may wish to put the command in your @value{GDBN} init
12533 file). This tells @value{GDBN} how to find the return value of
12534 functions which return floating point values. It also allows
12535 @value{GDBN} to avoid saving the floating point registers when calling
12536 functions on the board. If you are using a floating point coprocessor
12537 with only single precision floating point support, as on the @sc{r4650}
12538 processor, use the command @samp{set mipsfpu single}. The default
12539 double precision floating point coprocessor may be selected using
12540 @samp{set mipsfpu double}.
12542 In previous versions the only choices were double precision or no
12543 floating point, so @samp{set mipsfpu on} will select double precision
12544 and @samp{set mipsfpu off} will select no floating point.
12546 As usual, you can inquire about the @code{mipsfpu} variable with
12547 @samp{show mipsfpu}.
12549 @item set remotedebug @var{n}
12550 @itemx show remotedebug
12551 @kindex set remotedebug@r{, MIPS protocol}
12552 @kindex show remotedebug@r{, MIPS protocol}
12553 @cindex @code{remotedebug}, MIPS protocol
12554 @cindex MIPS @code{remotedebug} protocol
12555 @c FIXME! For this to be useful, you must know something about the MIPS
12556 @c FIXME...protocol. Where is it described?
12557 You can see some debugging information about communications with the board
12558 by setting the @code{remotedebug} variable. If you set it to @code{1} using
12559 @samp{set remotedebug 1}, every packet is displayed. If you set it
12560 to @code{2}, every character is displayed. You can check the current value
12561 at any time with the command @samp{show remotedebug}.
12563 @item set timeout @var{seconds}
12564 @itemx set retransmit-timeout @var{seconds}
12565 @itemx show timeout
12566 @itemx show retransmit-timeout
12567 @cindex @code{timeout}, MIPS protocol
12568 @cindex @code{retransmit-timeout}, MIPS protocol
12569 @kindex set timeout
12570 @kindex show timeout
12571 @kindex set retransmit-timeout
12572 @kindex show retransmit-timeout
12573 You can control the timeout used while waiting for a packet, in the MIPS
12574 remote protocol, with the @code{set timeout @var{seconds}} command. The
12575 default is 5 seconds. Similarly, you can control the timeout used while
12576 waiting for an acknowledgement of a packet with the @code{set
12577 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
12578 You can inspect both values with @code{show timeout} and @code{show
12579 retransmit-timeout}. (These commands are @emph{only} available when
12580 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
12582 The timeout set by @code{set timeout} does not apply when @value{GDBN}
12583 is waiting for your program to stop. In that case, @value{GDBN} waits
12584 forever because it has no way of knowing how long the program is going
12585 to run before stopping.
12589 @subsection PowerPC
12593 @kindex target dink32
12594 @item target dink32 @var{dev}
12595 DINK32 ROM monitor.
12597 @kindex target ppcbug
12598 @item target ppcbug @var{dev}
12599 @kindex target ppcbug1
12600 @item target ppcbug1 @var{dev}
12601 PPCBUG ROM monitor for PowerPC.
12604 @item target sds @var{dev}
12605 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
12610 @subsection HP PA Embedded
12614 @kindex target op50n
12615 @item target op50n @var{dev}
12616 OP50N monitor, running on an OKI HPPA board.
12618 @kindex target w89k
12619 @item target w89k @var{dev}
12620 W89K monitor, running on a Winbond HPPA board.
12625 @subsection Hitachi SH
12629 @kindex target hms@r{, with Hitachi SH}
12630 @item target hms @var{dev}
12631 A Hitachi SH board attached via serial line to your host. Use special
12632 commands @code{device} and @code{speed} to control the serial line and
12633 the communications speed used.
12635 @kindex target e7000@r{, with Hitachi SH}
12636 @item target e7000 @var{dev}
12637 E7000 emulator for Hitachi SH.
12639 @kindex target sh3@r{, with SH}
12640 @kindex target sh3e@r{, with SH}
12641 @item target sh3 @var{dev}
12642 @item target sh3e @var{dev}
12643 Hitachi SH-3 and SH-3E target systems.
12648 @subsection Tsqware Sparclet
12652 @value{GDBN} enables developers to debug tasks running on
12653 Sparclet targets from a Unix host.
12654 @value{GDBN} uses code that runs on
12655 both the Unix host and on the Sparclet target. The program
12656 @code{@value{GDBP}} is installed and executed on the Unix host.
12659 @item remotetimeout @var{args}
12660 @kindex remotetimeout
12661 @value{GDBN} supports the option @code{remotetimeout}.
12662 This option is set by the user, and @var{args} represents the number of
12663 seconds @value{GDBN} waits for responses.
12666 @cindex compiling, on Sparclet
12667 When compiling for debugging, include the options @samp{-g} to get debug
12668 information and @samp{-Ttext} to relocate the program to where you wish to
12669 load it on the target. You may also want to add the options @samp{-n} or
12670 @samp{-N} in order to reduce the size of the sections. Example:
12673 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
12676 You can use @code{objdump} to verify that the addresses are what you intended:
12679 sparclet-aout-objdump --headers --syms prog
12682 @cindex running, on Sparclet
12684 your Unix execution search path to find @value{GDBN}, you are ready to
12685 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
12686 (or @code{sparclet-aout-gdb}, depending on your installation).
12688 @value{GDBN} comes up showing the prompt:
12695 * Sparclet File:: Setting the file to debug
12696 * Sparclet Connection:: Connecting to Sparclet
12697 * Sparclet Download:: Sparclet download
12698 * Sparclet Execution:: Running and debugging
12701 @node Sparclet File
12702 @subsubsection Setting file to debug
12704 The @value{GDBN} command @code{file} lets you choose with program to debug.
12707 (gdbslet) file prog
12711 @value{GDBN} then attempts to read the symbol table of @file{prog}.
12712 @value{GDBN} locates
12713 the file by searching the directories listed in the command search
12715 If the file was compiled with debug information (option "-g"), source
12716 files will be searched as well.
12717 @value{GDBN} locates
12718 the source files by searching the directories listed in the directory search
12719 path (@pxref{Environment, ,Your program's environment}).
12721 to find a file, it displays a message such as:
12724 prog: No such file or directory.
12727 When this happens, add the appropriate directories to the search paths with
12728 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
12729 @code{target} command again.
12731 @node Sparclet Connection
12732 @subsubsection Connecting to Sparclet
12734 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
12735 To connect to a target on serial port ``@code{ttya}'', type:
12738 (gdbslet) target sparclet /dev/ttya
12739 Remote target sparclet connected to /dev/ttya
12740 main () at ../prog.c:3
12744 @value{GDBN} displays messages like these:
12750 @node Sparclet Download
12751 @subsubsection Sparclet download
12753 @cindex download to Sparclet
12754 Once connected to the Sparclet target,
12755 you can use the @value{GDBN}
12756 @code{load} command to download the file from the host to the target.
12757 The file name and load offset should be given as arguments to the @code{load}
12759 Since the file format is aout, the program must be loaded to the starting
12760 address. You can use @code{objdump} to find out what this value is. The load
12761 offset is an offset which is added to the VMA (virtual memory address)
12762 of each of the file's sections.
12763 For instance, if the program
12764 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
12765 and bss at 0x12010170, in @value{GDBN}, type:
12768 (gdbslet) load prog 0x12010000
12769 Loading section .text, size 0xdb0 vma 0x12010000
12772 If the code is loaded at a different address then what the program was linked
12773 to, you may need to use the @code{section} and @code{add-symbol-file} commands
12774 to tell @value{GDBN} where to map the symbol table.
12776 @node Sparclet Execution
12777 @subsubsection Running and debugging
12779 @cindex running and debugging Sparclet programs
12780 You can now begin debugging the task using @value{GDBN}'s execution control
12781 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
12782 manual for the list of commands.
12786 Breakpoint 1 at 0x12010000: file prog.c, line 3.
12788 Starting program: prog
12789 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
12790 3 char *symarg = 0;
12792 4 char *execarg = "hello!";
12797 @subsection Fujitsu Sparclite
12801 @kindex target sparclite
12802 @item target sparclite @var{dev}
12803 Fujitsu sparclite boards, used only for the purpose of loading.
12804 You must use an additional command to debug the program.
12805 For example: target remote @var{dev} using @value{GDBN} standard
12811 @subsection Tandem ST2000
12813 @value{GDBN} may be used with a Tandem ST2000 phone switch, running Tandem's
12816 To connect your ST2000 to the host system, see the manufacturer's
12817 manual. Once the ST2000 is physically attached, you can run:
12820 target st2000 @var{dev} @var{speed}
12824 to establish it as your debugging environment. @var{dev} is normally
12825 the name of a serial device, such as @file{/dev/ttya}, connected to the
12826 ST2000 via a serial line. You can instead specify @var{dev} as a TCP
12827 connection (for example, to a serial line attached via a terminal
12828 concentrator) using the syntax @code{@var{hostname}:@var{portnumber}}.
12830 The @code{load} and @code{attach} commands are @emph{not} defined for
12831 this target; you must load your program into the ST2000 as you normally
12832 would for standalone operation. @value{GDBN} reads debugging information
12833 (such as symbols) from a separate, debugging version of the program
12834 available on your host computer.
12835 @c FIXME!! This is terribly vague; what little content is here is
12836 @c basically hearsay.
12838 @cindex ST2000 auxiliary commands
12839 These auxiliary @value{GDBN} commands are available to help you with the ST2000
12843 @item st2000 @var{command}
12844 @kindex st2000 @var{cmd}
12845 @cindex STDBUG commands (ST2000)
12846 @cindex commands to STDBUG (ST2000)
12847 Send a @var{command} to the STDBUG monitor. See the manufacturer's
12848 manual for available commands.
12851 @cindex connect (to STDBUG)
12852 Connect the controlling terminal to the STDBUG command monitor. When
12853 you are done interacting with STDBUG, typing either of two character
12854 sequences gets you back to the @value{GDBN} command prompt:
12855 @kbd{@key{RET}~.} (Return, followed by tilde and period) or
12856 @kbd{@key{RET}~@key{C-d}} (Return, followed by tilde and control-D).
12860 @subsection Zilog Z8000
12863 @cindex simulator, Z8000
12864 @cindex Zilog Z8000 simulator
12866 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
12869 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
12870 unsegmented variant of the Z8000 architecture) or the Z8001 (the
12871 segmented variant). The simulator recognizes which architecture is
12872 appropriate by inspecting the object code.
12875 @item target sim @var{args}
12877 @kindex target sim@r{, with Z8000}
12878 Debug programs on a simulated CPU. If the simulator supports setup
12879 options, specify them via @var{args}.
12883 After specifying this target, you can debug programs for the simulated
12884 CPU in the same style as programs for your host computer; use the
12885 @code{file} command to load a new program image, the @code{run} command
12886 to run your program, and so on.
12888 As well as making available all the usual machine registers
12889 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
12890 additional items of information as specially named registers:
12895 Counts clock-ticks in the simulator.
12898 Counts instructions run in the simulator.
12901 Execution time in 60ths of a second.
12905 You can refer to these values in @value{GDBN} expressions with the usual
12906 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
12907 conditional breakpoint that suspends only after at least 5000
12908 simulated clock ticks.
12910 @node Architectures
12911 @section Architectures
12913 This section describes characteristics of architectures that affect
12914 all uses of @value{GDBN} with the architecture, both native and cross.
12927 @kindex set rstack_high_address
12928 @cindex AMD 29K register stack
12929 @cindex register stack, AMD29K
12930 @item set rstack_high_address @var{address}
12931 On AMD 29000 family processors, registers are saved in a separate
12932 @dfn{register stack}. There is no way for @value{GDBN} to determine the
12933 extent of this stack. Normally, @value{GDBN} just assumes that the
12934 stack is ``large enough''. This may result in @value{GDBN} referencing
12935 memory locations that do not exist. If necessary, you can get around
12936 this problem by specifying the ending address of the register stack with
12937 the @code{set rstack_high_address} command. The argument should be an
12938 address, which you probably want to precede with @samp{0x} to specify in
12941 @kindex show rstack_high_address
12942 @item show rstack_high_address
12943 Display the current limit of the register stack, on AMD 29000 family
12951 See the following section.
12956 @cindex stack on Alpha
12957 @cindex stack on MIPS
12958 @cindex Alpha stack
12960 Alpha- and MIPS-based computers use an unusual stack frame, which
12961 sometimes requires @value{GDBN} to search backward in the object code to
12962 find the beginning of a function.
12964 @cindex response time, MIPS debugging
12965 To improve response time (especially for embedded applications, where
12966 @value{GDBN} may be restricted to a slow serial line for this search)
12967 you may want to limit the size of this search, using one of these
12971 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
12972 @item set heuristic-fence-post @var{limit}
12973 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
12974 search for the beginning of a function. A value of @var{0} (the
12975 default) means there is no limit. However, except for @var{0}, the
12976 larger the limit the more bytes @code{heuristic-fence-post} must search
12977 and therefore the longer it takes to run.
12979 @item show heuristic-fence-post
12980 Display the current limit.
12984 These commands are available @emph{only} when @value{GDBN} is configured
12985 for debugging programs on Alpha or MIPS processors.
12988 @node Controlling GDB
12989 @chapter Controlling @value{GDBN}
12991 You can alter the way @value{GDBN} interacts with you by using the
12992 @code{set} command. For commands controlling how @value{GDBN} displays
12993 data, see @ref{Print Settings, ,Print settings}. Other settings are
12998 * Editing:: Command editing
12999 * History:: Command history
13000 * Screen Size:: Screen size
13001 * Numbers:: Numbers
13002 * Messages/Warnings:: Optional warnings and messages
13003 * Debugging Output:: Optional messages about internal happenings
13011 @value{GDBN} indicates its readiness to read a command by printing a string
13012 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
13013 can change the prompt string with the @code{set prompt} command. For
13014 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
13015 the prompt in one of the @value{GDBN} sessions so that you can always tell
13016 which one you are talking to.
13018 @emph{Note:} @code{set prompt} does not add a space for you after the
13019 prompt you set. This allows you to set a prompt which ends in a space
13020 or a prompt that does not.
13024 @item set prompt @var{newprompt}
13025 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
13027 @kindex show prompt
13029 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
13033 @section Command editing
13035 @cindex command line editing
13037 @value{GDBN} reads its input commands via the @dfn{readline} interface. This
13038 @sc{gnu} library provides consistent behavior for programs which provide a
13039 command line interface to the user. Advantages are @sc{gnu} Emacs-style
13040 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
13041 substitution, and a storage and recall of command history across
13042 debugging sessions.
13044 You may control the behavior of command line editing in @value{GDBN} with the
13045 command @code{set}.
13048 @kindex set editing
13051 @itemx set editing on
13052 Enable command line editing (enabled by default).
13054 @item set editing off
13055 Disable command line editing.
13057 @kindex show editing
13059 Show whether command line editing is enabled.
13063 @section Command history
13065 @value{GDBN} can keep track of the commands you type during your
13066 debugging sessions, so that you can be certain of precisely what
13067 happened. Use these commands to manage the @value{GDBN} command
13071 @cindex history substitution
13072 @cindex history file
13073 @kindex set history filename
13074 @kindex GDBHISTFILE
13075 @item set history filename @var{fname}
13076 Set the name of the @value{GDBN} command history file to @var{fname}.
13077 This is the file where @value{GDBN} reads an initial command history
13078 list, and where it writes the command history from this session when it
13079 exits. You can access this list through history expansion or through
13080 the history command editing characters listed below. This file defaults
13081 to the value of the environment variable @code{GDBHISTFILE}, or to
13082 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
13085 @cindex history save
13086 @kindex set history save
13087 @item set history save
13088 @itemx set history save on
13089 Record command history in a file, whose name may be specified with the
13090 @code{set history filename} command. By default, this option is disabled.
13092 @item set history save off
13093 Stop recording command history in a file.
13095 @cindex history size
13096 @kindex set history size
13097 @item set history size @var{size}
13098 Set the number of commands which @value{GDBN} keeps in its history list.
13099 This defaults to the value of the environment variable
13100 @code{HISTSIZE}, or to 256 if this variable is not set.
13103 @cindex history expansion
13104 History expansion assigns special meaning to the character @kbd{!}.
13105 @ifset have-readline-appendices
13106 @xref{Event Designators}.
13109 Since @kbd{!} is also the logical not operator in C, history expansion
13110 is off by default. If you decide to enable history expansion with the
13111 @code{set history expansion on} command, you may sometimes need to
13112 follow @kbd{!} (when it is used as logical not, in an expression) with
13113 a space or a tab to prevent it from being expanded. The readline
13114 history facilities do not attempt substitution on the strings
13115 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
13117 The commands to control history expansion are:
13120 @kindex set history expansion
13121 @item set history expansion on
13122 @itemx set history expansion
13123 Enable history expansion. History expansion is off by default.
13125 @item set history expansion off
13126 Disable history expansion.
13128 The readline code comes with more complete documentation of
13129 editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs
13130 or @code{vi} may wish to read it.
13131 @ifset have-readline-appendices
13132 @xref{Command Line Editing}.
13136 @kindex show history
13138 @itemx show history filename
13139 @itemx show history save
13140 @itemx show history size
13141 @itemx show history expansion
13142 These commands display the state of the @value{GDBN} history parameters.
13143 @code{show history} by itself displays all four states.
13149 @item show commands
13150 Display the last ten commands in the command history.
13152 @item show commands @var{n}
13153 Print ten commands centered on command number @var{n}.
13155 @item show commands +
13156 Print ten commands just after the commands last printed.
13160 @section Screen size
13161 @cindex size of screen
13162 @cindex pauses in output
13164 Certain commands to @value{GDBN} may produce large amounts of
13165 information output to the screen. To help you read all of it,
13166 @value{GDBN} pauses and asks you for input at the end of each page of
13167 output. Type @key{RET} when you want to continue the output, or @kbd{q}
13168 to discard the remaining output. Also, the screen width setting
13169 determines when to wrap lines of output. Depending on what is being
13170 printed, @value{GDBN} tries to break the line at a readable place,
13171 rather than simply letting it overflow onto the following line.
13173 Normally @value{GDBN} knows the size of the screen from the terminal
13174 driver software. For example, on Unix @value{GDBN} uses the termcap data base
13175 together with the value of the @code{TERM} environment variable and the
13176 @code{stty rows} and @code{stty cols} settings. If this is not correct,
13177 you can override it with the @code{set height} and @code{set
13184 @kindex show height
13185 @item set height @var{lpp}
13187 @itemx set width @var{cpl}
13189 These @code{set} commands specify a screen height of @var{lpp} lines and
13190 a screen width of @var{cpl} characters. The associated @code{show}
13191 commands display the current settings.
13193 If you specify a height of zero lines, @value{GDBN} does not pause during
13194 output no matter how long the output is. This is useful if output is to a
13195 file or to an editor buffer.
13197 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
13198 from wrapping its output.
13203 @cindex number representation
13204 @cindex entering numbers
13206 You can always enter numbers in octal, decimal, or hexadecimal in
13207 @value{GDBN} by the usual conventions: octal numbers begin with
13208 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
13209 begin with @samp{0x}. Numbers that begin with none of these are, by
13210 default, entered in base 10; likewise, the default display for
13211 numbers---when no particular format is specified---is base 10. You can
13212 change the default base for both input and output with the @code{set
13216 @kindex set input-radix
13217 @item set input-radix @var{base}
13218 Set the default base for numeric input. Supported choices
13219 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
13220 specified either unambiguously or using the current default radix; for
13230 sets the base to decimal. On the other hand, @samp{set radix 10}
13231 leaves the radix unchanged no matter what it was.
13233 @kindex set output-radix
13234 @item set output-radix @var{base}
13235 Set the default base for numeric display. Supported choices
13236 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
13237 specified either unambiguously or using the current default radix.
13239 @kindex show input-radix
13240 @item show input-radix
13241 Display the current default base for numeric input.
13243 @kindex show output-radix
13244 @item show output-radix
13245 Display the current default base for numeric display.
13248 @node Messages/Warnings
13249 @section Optional warnings and messages
13251 By default, @value{GDBN} is silent about its inner workings. If you are
13252 running on a slow machine, you may want to use the @code{set verbose}
13253 command. This makes @value{GDBN} tell you when it does a lengthy
13254 internal operation, so you will not think it has crashed.
13256 Currently, the messages controlled by @code{set verbose} are those
13257 which announce that the symbol table for a source file is being read;
13258 see @code{symbol-file} in @ref{Files, ,Commands to specify files}.
13261 @kindex set verbose
13262 @item set verbose on
13263 Enables @value{GDBN} output of certain informational messages.
13265 @item set verbose off
13266 Disables @value{GDBN} output of certain informational messages.
13268 @kindex show verbose
13270 Displays whether @code{set verbose} is on or off.
13273 By default, if @value{GDBN} encounters bugs in the symbol table of an
13274 object file, it is silent; but if you are debugging a compiler, you may
13275 find this information useful (@pxref{Symbol Errors, ,Errors reading
13280 @kindex set complaints
13281 @item set complaints @var{limit}
13282 Permits @value{GDBN} to output @var{limit} complaints about each type of
13283 unusual symbols before becoming silent about the problem. Set
13284 @var{limit} to zero to suppress all complaints; set it to a large number
13285 to prevent complaints from being suppressed.
13287 @kindex show complaints
13288 @item show complaints
13289 Displays how many symbol complaints @value{GDBN} is permitted to produce.
13293 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
13294 lot of stupid questions to confirm certain commands. For example, if
13295 you try to run a program which is already running:
13299 The program being debugged has been started already.
13300 Start it from the beginning? (y or n)
13303 If you are willing to unflinchingly face the consequences of your own
13304 commands, you can disable this ``feature'':
13308 @kindex set confirm
13310 @cindex confirmation
13311 @cindex stupid questions
13312 @item set confirm off
13313 Disables confirmation requests.
13315 @item set confirm on
13316 Enables confirmation requests (the default).
13318 @kindex show confirm
13320 Displays state of confirmation requests.
13324 @node Debugging Output
13325 @section Optional messages about internal happenings
13327 @kindex set debug arch
13328 @item set debug arch
13329 Turns on or off display of gdbarch debugging info. The default is off
13330 @kindex show debug arch
13331 @item show debug arch
13332 Displays the current state of displaying gdbarch debugging info.
13333 @kindex set debug event
13334 @item set debug event
13335 Turns on or off display of @value{GDBN} event debugging info. The
13337 @kindex show debug event
13338 @item show debug event
13339 Displays the current state of displaying @value{GDBN} event debugging
13341 @kindex set debug expression
13342 @item set debug expression
13343 Turns on or off display of @value{GDBN} expression debugging info. The
13345 @kindex show debug expression
13346 @item show debug expression
13347 Displays the current state of displaying @value{GDBN} expression
13349 @kindex set debug overload
13350 @item set debug overload
13351 Turns on or off display of @value{GDBN} C@t{++} overload debugging
13352 info. This includes info such as ranking of functions, etc. The default
13354 @kindex show debug overload
13355 @item show debug overload
13356 Displays the current state of displaying @value{GDBN} C@t{++} overload
13358 @kindex set debug remote
13359 @cindex packets, reporting on stdout
13360 @cindex serial connections, debugging
13361 @item set debug remote
13362 Turns on or off display of reports on all packets sent back and forth across
13363 the serial line to the remote machine. The info is printed on the
13364 @value{GDBN} standard output stream. The default is off.
13365 @kindex show debug remote
13366 @item show debug remote
13367 Displays the state of display of remote packets.
13368 @kindex set debug serial
13369 @item set debug serial
13370 Turns on or off display of @value{GDBN} serial debugging info. The
13372 @kindex show debug serial
13373 @item show debug serial
13374 Displays the current state of displaying @value{GDBN} serial debugging
13376 @kindex set debug target
13377 @item set debug target
13378 Turns on or off display of @value{GDBN} target debugging info. This info
13379 includes what is going on at the target level of GDB, as it happens. The
13381 @kindex show debug target
13382 @item show debug target
13383 Displays the current state of displaying @value{GDBN} target debugging
13385 @kindex set debug varobj
13386 @item set debug varobj
13387 Turns on or off display of @value{GDBN} variable object debugging
13388 info. The default is off.
13389 @kindex show debug varobj
13390 @item show debug varobj
13391 Displays the current state of displaying @value{GDBN} variable object
13396 @chapter Canned Sequences of Commands
13398 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
13399 command lists}), @value{GDBN} provides two ways to store sequences of
13400 commands for execution as a unit: user-defined commands and command
13404 * Define:: User-defined commands
13405 * Hooks:: User-defined command hooks
13406 * Command Files:: Command files
13407 * Output:: Commands for controlled output
13411 @section User-defined commands
13413 @cindex user-defined command
13414 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
13415 which you assign a new name as a command. This is done with the
13416 @code{define} command. User commands may accept up to 10 arguments
13417 separated by whitespace. Arguments are accessed within the user command
13418 via @var{$arg0@dots{}$arg9}. A trivial example:
13422 print $arg0 + $arg1 + $arg2
13426 To execute the command use:
13433 This defines the command @code{adder}, which prints the sum of
13434 its three arguments. Note the arguments are text substitutions, so they may
13435 reference variables, use complex expressions, or even perform inferior
13441 @item define @var{commandname}
13442 Define a command named @var{commandname}. If there is already a command
13443 by that name, you are asked to confirm that you want to redefine it.
13445 The definition of the command is made up of other @value{GDBN} command lines,
13446 which are given following the @code{define} command. The end of these
13447 commands is marked by a line containing @code{end}.
13452 Takes a single argument, which is an expression to evaluate.
13453 It is followed by a series of commands that are executed
13454 only if the expression is true (nonzero).
13455 There can then optionally be a line @code{else}, followed
13456 by a series of commands that are only executed if the expression
13457 was false. The end of the list is marked by a line containing @code{end}.
13461 The syntax is similar to @code{if}: the command takes a single argument,
13462 which is an expression to evaluate, and must be followed by the commands to
13463 execute, one per line, terminated by an @code{end}.
13464 The commands are executed repeatedly as long as the expression
13468 @item document @var{commandname}
13469 Document the user-defined command @var{commandname}, so that it can be
13470 accessed by @code{help}. The command @var{commandname} must already be
13471 defined. This command reads lines of documentation just as @code{define}
13472 reads the lines of the command definition, ending with @code{end}.
13473 After the @code{document} command is finished, @code{help} on command
13474 @var{commandname} displays the documentation you have written.
13476 You may use the @code{document} command again to change the
13477 documentation of a command. Redefining the command with @code{define}
13478 does not change the documentation.
13480 @kindex help user-defined
13481 @item help user-defined
13482 List all user-defined commands, with the first line of the documentation
13487 @itemx show user @var{commandname}
13488 Display the @value{GDBN} commands used to define @var{commandname} (but
13489 not its documentation). If no @var{commandname} is given, display the
13490 definitions for all user-defined commands.
13494 When user-defined commands are executed, the
13495 commands of the definition are not printed. An error in any command
13496 stops execution of the user-defined command.
13498 If used interactively, commands that would ask for confirmation proceed
13499 without asking when used inside a user-defined command. Many @value{GDBN}
13500 commands that normally print messages to say what they are doing omit the
13501 messages when used in a user-defined command.
13504 @section User-defined command hooks
13505 @cindex command hooks
13506 @cindex hooks, for commands
13507 @cindex hooks, pre-command
13511 You may define @dfn{hooks}, which are a special kind of user-defined
13512 command. Whenever you run the command @samp{foo}, if the user-defined
13513 command @samp{hook-foo} exists, it is executed (with no arguments)
13514 before that command.
13516 @cindex hooks, post-command
13519 A hook may also be defined which is run after the command you executed.
13520 Whenever you run the command @samp{foo}, if the user-defined command
13521 @samp{hookpost-foo} exists, it is executed (with no arguments) after
13522 that command. Post-execution hooks may exist simultaneously with
13523 pre-execution hooks, for the same command.
13525 It is valid for a hook to call the command which it hooks. If this
13526 occurs, the hook is not re-executed, thereby avoiding infinte recursion.
13528 @c It would be nice if hookpost could be passed a parameter indicating
13529 @c if the command it hooks executed properly or not. FIXME!
13531 @kindex stop@r{, a pseudo-command}
13532 In addition, a pseudo-command, @samp{stop} exists. Defining
13533 (@samp{hook-stop}) makes the associated commands execute every time
13534 execution stops in your program: before breakpoint commands are run,
13535 displays are printed, or the stack frame is printed.
13537 For example, to ignore @code{SIGALRM} signals while
13538 single-stepping, but treat them normally during normal execution,
13543 handle SIGALRM nopass
13547 handle SIGALRM pass
13550 define hook-continue
13551 handle SIGLARM pass
13555 As a further example, to hook at the begining and end of the @code{echo}
13556 command, and to add extra text to the beginning and end of the message,
13564 define hookpost-echo
13568 (@value{GDBP}) echo Hello World
13569 <<<---Hello World--->>>
13574 You can define a hook for any single-word command in @value{GDBN}, but
13575 not for command aliases; you should define a hook for the basic command
13576 name, e.g. @code{backtrace} rather than @code{bt}.
13577 @c FIXME! So how does Joe User discover whether a command is an alias
13579 If an error occurs during the execution of your hook, execution of
13580 @value{GDBN} commands stops and @value{GDBN} issues a prompt
13581 (before the command that you actually typed had a chance to run).
13583 If you try to define a hook which does not match any known command, you
13584 get a warning from the @code{define} command.
13586 @node Command Files
13587 @section Command files
13589 @cindex command files
13590 A command file for @value{GDBN} is a file of lines that are @value{GDBN}
13591 commands. Comments (lines starting with @kbd{#}) may also be included.
13592 An empty line in a command file does nothing; it does not mean to repeat
13593 the last command, as it would from the terminal.
13596 @cindex @file{.gdbinit}
13597 @cindex @file{gdb.ini}
13598 When you start @value{GDBN}, it automatically executes commands from its
13599 @dfn{init files}, normally called @file{.gdbinit}@footnote{The DJGPP
13600 port of @value{GDBN} uses the name @file{gdb.ini} instead, due to the
13601 limitations of file names imposed by DOS filesystems.}.
13602 During startup, @value{GDBN} does the following:
13606 Reads the init file (if any) in your home directory@footnote{On
13607 DOS/Windows systems, the home directory is the one pointed to by the
13608 @code{HOME} environment variable.}.
13611 Processes command line options and operands.
13614 Reads the init file (if any) in the current working directory.
13617 Reads command files specified by the @samp{-x} option.
13620 The init file in your home directory can set options (such as @samp{set
13621 complaints}) that affect subsequent processing of command line options
13622 and operands. Init files are not executed if you use the @samp{-nx}
13623 option (@pxref{Mode Options, ,Choosing modes}).
13625 @cindex init file name
13626 On some configurations of @value{GDBN}, the init file is known by a
13627 different name (these are typically environments where a specialized
13628 form of @value{GDBN} may need to coexist with other forms, hence a
13629 different name for the specialized version's init file). These are the
13630 environments with special init file names:
13632 @cindex @file{.vxgdbinit}
13635 VxWorks (Wind River Systems real-time OS): @file{.vxgdbinit}
13637 @cindex @file{.os68gdbinit}
13639 OS68K (Enea Data Systems real-time OS): @file{.os68gdbinit}
13641 @cindex @file{.esgdbinit}
13643 ES-1800 (Ericsson Telecom AB M68000 emulator): @file{.esgdbinit}
13646 You can also request the execution of a command file with the
13647 @code{source} command:
13651 @item source @var{filename}
13652 Execute the command file @var{filename}.
13655 The lines in a command file are executed sequentially. They are not
13656 printed as they are executed. An error in any command terminates execution
13657 of the command file.
13659 Commands that would ask for confirmation if used interactively proceed
13660 without asking when used in a command file. Many @value{GDBN} commands that
13661 normally print messages to say what they are doing omit the messages
13662 when called from command files.
13664 @value{GDBN} also accepts command input from standard input. In this
13665 mode, normal output goes to standard output and error output goes to
13666 standard error. Errors in a command file supplied on standard input do
13667 not terminate execution of the command file --- execution continues with
13671 gdb < cmds > log 2>&1
13674 (The syntax above will vary depending on the shell used.) This example
13675 will execute commands from the file @file{cmds}. All output and errors
13676 would be directed to @file{log}.
13679 @section Commands for controlled output
13681 During the execution of a command file or a user-defined command, normal
13682 @value{GDBN} output is suppressed; the only output that appears is what is
13683 explicitly printed by the commands in the definition. This section
13684 describes three commands useful for generating exactly the output you
13689 @item echo @var{text}
13690 @c I do not consider backslash-space a standard C escape sequence
13691 @c because it is not in ANSI.
13692 Print @var{text}. Nonprinting characters can be included in
13693 @var{text} using C escape sequences, such as @samp{\n} to print a
13694 newline. @strong{No newline is printed unless you specify one.}
13695 In addition to the standard C escape sequences, a backslash followed
13696 by a space stands for a space. This is useful for displaying a
13697 string with spaces at the beginning or the end, since leading and
13698 trailing spaces are otherwise trimmed from all arguments.
13699 To print @samp{@w{ }and foo =@w{ }}, use the command
13700 @samp{echo \@w{ }and foo = \@w{ }}.
13702 A backslash at the end of @var{text} can be used, as in C, to continue
13703 the command onto subsequent lines. For example,
13706 echo This is some text\n\
13707 which is continued\n\
13708 onto several lines.\n
13711 produces the same output as
13714 echo This is some text\n
13715 echo which is continued\n
13716 echo onto several lines.\n
13720 @item output @var{expression}
13721 Print the value of @var{expression} and nothing but that value: no
13722 newlines, no @samp{$@var{nn} = }. The value is not entered in the
13723 value history either. @xref{Expressions, ,Expressions}, for more information
13726 @item output/@var{fmt} @var{expression}
13727 Print the value of @var{expression} in format @var{fmt}. You can use
13728 the same formats as for @code{print}. @xref{Output Formats,,Output
13729 formats}, for more information.
13732 @item printf @var{string}, @var{expressions}@dots{}
13733 Print the values of the @var{expressions} under the control of
13734 @var{string}. The @var{expressions} are separated by commas and may be
13735 either numbers or pointers. Their values are printed as specified by
13736 @var{string}, exactly as if your program were to execute the C
13738 @c FIXME: the above implies that at least all ANSI C formats are
13739 @c supported, but it isn't true: %E and %G don't work (or so it seems).
13740 @c Either this is a bug, or the manual should document what formats are
13744 printf (@var{string}, @var{expressions}@dots{});
13747 For example, you can print two values in hex like this:
13750 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
13753 The only backslash-escape sequences that you can use in the format
13754 string are the simple ones that consist of backslash followed by a
13759 @chapter @value{GDBN} Text User Interface
13763 * TUI Overview:: TUI overview
13764 * TUI Keys:: TUI key bindings
13765 * TUI Commands:: TUI specific commands
13766 * TUI Configuration:: TUI configuration variables
13769 The @value{GDBN} Text User Interface, TUI in short,
13770 is a terminal interface which uses the @code{curses} library
13771 to show the source file, the assembly output, the program registers
13772 and @value{GDBN} commands in separate text windows.
13773 The TUI is available only when @value{GDBN} is configured
13774 with the @code{--enable-tui} configure option (@pxref{Configure Options}).
13777 @section TUI overview
13779 The TUI has two display modes that can be switched while
13784 A curses (or TUI) mode in which it displays several text
13785 windows on the terminal.
13788 A standard mode which corresponds to the @value{GDBN} configured without
13792 In the TUI mode, @value{GDBN} can display several text window
13797 This window is the @value{GDBN} command window with the @value{GDBN}
13798 prompt and the @value{GDBN} outputs. The @value{GDBN} input is still
13799 managed using readline but through the TUI. The @emph{command}
13800 window is always visible.
13803 The source window shows the source file of the program. The current
13804 line as well as active breakpoints are displayed in this window.
13805 The current program position is shown with the @samp{>} marker and
13806 active breakpoints are shown with @samp{*} markers.
13809 The assembly window shows the disassembly output of the program.
13812 This window shows the processor registers. It detects when
13813 a register is changed and when this is the case, registers that have
13814 changed are highlighted.
13818 The source, assembly and register windows are attached to the thread
13819 and the frame position. They are updated when the current thread
13820 changes, when the frame changes or when the program counter changes.
13821 These three windows are arranged by the TUI according to several
13822 layouts. The layout defines which of these three windows are visible.
13823 The following layouts are available:
13833 source and assembly
13836 source and registers
13839 assembly and registers
13844 @section TUI Key Bindings
13845 @cindex TUI key bindings
13847 The TUI installs several key bindings in the readline keymaps
13848 (@pxref{Command Line Editing}).
13849 They allow to leave or enter in the TUI mode or they operate
13850 directly on the TUI layout and windows. The following key bindings
13851 are installed for both TUI mode and the @value{GDBN} standard mode.
13860 Enter or leave the TUI mode. When the TUI mode is left,
13861 the curses window management is left and @value{GDBN} operates using
13862 its standard mode writing on the terminal directly. When the TUI
13863 mode is entered, the control is given back to the curses windows.
13864 The screen is then refreshed.
13868 Use a TUI layout with only one window. The layout will
13869 either be @samp{source} or @samp{assembly}. When the TUI mode
13870 is not active, it will switch to the TUI mode.
13872 Think of this key binding as the Emacs @kbd{C-x 1} binding.
13876 Use a TUI layout with at least two windows. When the current
13877 layout shows already two windows, a next layout with two windows is used.
13878 When a new layout is chosen, one window will always be common to the
13879 previous layout and the new one.
13881 Think of it as the Emacs @kbd{C-x 2} binding.
13885 The following key bindings are handled only by the TUI mode:
13890 Scroll the active window one page up.
13894 Scroll the active window one page down.
13898 Scroll the active window one line up.
13902 Scroll the active window one line down.
13906 Scroll the active window one column left.
13910 Scroll the active window one column right.
13914 Refresh the screen.
13918 In the TUI mode, the arrow keys are used by the active window
13919 for scrolling. This means they are not available for readline. It is
13920 necessary to use other readline key bindings such as @key{C-p}, @key{C-n},
13921 @key{C-b} and @key{C-f}.
13924 @section TUI specific commands
13925 @cindex TUI commands
13927 The TUI has specific commands to control the text windows.
13928 These commands are always available, that is they do not depend on
13929 the current terminal mode in which @value{GDBN} runs. When @value{GDBN}
13930 is in the standard mode, using these commands will automatically switch
13935 @kindex layout next
13936 Display the next layout.
13939 @kindex layout prev
13940 Display the previous layout.
13944 Display the source window only.
13948 Display the assembly window only.
13951 @kindex layout split
13952 Display the source and assembly window.
13955 @kindex layout regs
13956 Display the register window together with the source or assembly window.
13958 @item focus next | prev | src | asm | regs | split
13960 Set the focus to the named window.
13961 This command allows to change the active window so that scrolling keys
13962 can be affected to another window.
13966 Refresh the screen. This is similar to using @key{C-L} key.
13970 Update the source window and the current execution point.
13972 @item winheight @var{name} +@var{count}
13973 @itemx winheight @var{name} -@var{count}
13975 Change the height of the window @var{name} by @var{count}
13976 lines. Positive counts increase the height, while negative counts
13981 @node TUI Configuration
13982 @section TUI configuration variables
13983 @cindex TUI configuration variables
13985 The TUI has several configuration variables that control the
13986 appearance of windows on the terminal.
13989 @item set tui border-kind @var{kind}
13990 @kindex set tui border-kind
13991 Select the border appearance for the source, assembly and register windows.
13992 The possible values are the following:
13995 Use a space character to draw the border.
13998 Use ascii characters + - and | to draw the border.
14001 Use the Alternate Character Set to draw the border. The border is
14002 drawn using character line graphics if the terminal supports them.
14006 @item set tui active-border-mode @var{mode}
14007 @kindex set tui active-border-mode
14008 Select the attributes to display the border of the active window.
14009 The possible values are @code{normal}, @code{standout}, @code{reverse},
14010 @code{half}, @code{half-standout}, @code{bold} and @code{bold-standout}.
14012 @item set tui border-mode @var{mode}
14013 @kindex set tui border-mode
14014 Select the attributes to display the border of other windows.
14015 The @var{mode} can be one of the following:
14018 Use normal attributes to display the border.
14024 Use reverse video mode.
14027 Use half bright mode.
14029 @item half-standout
14030 Use half bright and standout mode.
14033 Use extra bright or bold mode.
14035 @item bold-standout
14036 Use extra bright or bold and standout mode.
14043 @chapter Using @value{GDBN} under @sc{gnu} Emacs
14046 @cindex @sc{gnu} Emacs
14047 A special interface allows you to use @sc{gnu} Emacs to view (and
14048 edit) the source files for the program you are debugging with
14051 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
14052 executable file you want to debug as an argument. This command starts
14053 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
14054 created Emacs buffer.
14055 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
14057 Using @value{GDBN} under Emacs is just like using @value{GDBN} normally except for two
14062 All ``terminal'' input and output goes through the Emacs buffer.
14065 This applies both to @value{GDBN} commands and their output, and to the input
14066 and output done by the program you are debugging.
14068 This is useful because it means that you can copy the text of previous
14069 commands and input them again; you can even use parts of the output
14072 All the facilities of Emacs' Shell mode are available for interacting
14073 with your program. In particular, you can send signals the usual
14074 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
14079 @value{GDBN} displays source code through Emacs.
14082 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
14083 source file for that frame and puts an arrow (@samp{=>}) at the
14084 left margin of the current line. Emacs uses a separate buffer for
14085 source display, and splits the screen to show both your @value{GDBN} session
14088 Explicit @value{GDBN} @code{list} or search commands still produce output as
14089 usual, but you probably have no reason to use them from Emacs.
14092 @emph{Warning:} If the directory where your program resides is not your
14093 current directory, it can be easy to confuse Emacs about the location of
14094 the source files, in which case the auxiliary display buffer does not
14095 appear to show your source. @value{GDBN} can find programs by searching your
14096 environment's @code{PATH} variable, so the @value{GDBN} input and output
14097 session proceeds normally; but Emacs does not get enough information
14098 back from @value{GDBN} to locate the source files in this situation. To
14099 avoid this problem, either start @value{GDBN} mode from the directory where
14100 your program resides, or specify an absolute file name when prompted for the
14101 @kbd{M-x gdb} argument.
14103 A similar confusion can result if you use the @value{GDBN} @code{file} command to
14104 switch to debugging a program in some other location, from an existing
14105 @value{GDBN} buffer in Emacs.
14108 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
14109 you need to call @value{GDBN} by a different name (for example, if you keep
14110 several configurations around, with different names) you can set the
14111 Emacs variable @code{gdb-command-name}; for example,
14114 (setq gdb-command-name "mygdb")
14118 (preceded by @kbd{M-:} or @kbd{ESC :}, or typed in the @code{*scratch*} buffer, or
14119 in your @file{.emacs} file) makes Emacs call the program named
14120 ``@code{mygdb}'' instead.
14122 In the @value{GDBN} I/O buffer, you can use these special Emacs commands in
14123 addition to the standard Shell mode commands:
14127 Describe the features of Emacs' @value{GDBN} Mode.
14130 Execute to another source line, like the @value{GDBN} @code{step} command; also
14131 update the display window to show the current file and location.
14134 Execute to next source line in this function, skipping all function
14135 calls, like the @value{GDBN} @code{next} command. Then update the display window
14136 to show the current file and location.
14139 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
14140 display window accordingly.
14142 @item M-x gdb-nexti
14143 Execute to next instruction, using the @value{GDBN} @code{nexti} command; update
14144 display window accordingly.
14147 Execute until exit from the selected stack frame, like the @value{GDBN}
14148 @code{finish} command.
14151 Continue execution of your program, like the @value{GDBN} @code{continue}
14154 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
14157 Go up the number of frames indicated by the numeric argument
14158 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
14159 like the @value{GDBN} @code{up} command.
14161 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}.
14164 Go down the number of frames indicated by the numeric argument, like the
14165 @value{GDBN} @code{down} command.
14167 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}.
14170 Read the number where the cursor is positioned, and insert it at the end
14171 of the @value{GDBN} I/O buffer. For example, if you wish to disassemble code
14172 around an address that was displayed earlier, type @kbd{disassemble};
14173 then move the cursor to the address display, and pick up the
14174 argument for @code{disassemble} by typing @kbd{C-x &}.
14176 You can customize this further by defining elements of the list
14177 @code{gdb-print-command}; once it is defined, you can format or
14178 otherwise process numbers picked up by @kbd{C-x &} before they are
14179 inserted. A numeric argument to @kbd{C-x &} indicates that you
14180 wish special formatting, and also acts as an index to pick an element of the
14181 list. If the list element is a string, the number to be inserted is
14182 formatted using the Emacs function @code{format}; otherwise the number
14183 is passed as an argument to the corresponding list element.
14186 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
14187 tells @value{GDBN} to set a breakpoint on the source line point is on.
14189 If you accidentally delete the source-display buffer, an easy way to get
14190 it back is to type the command @code{f} in the @value{GDBN} buffer, to
14191 request a frame display; when you run under Emacs, this recreates
14192 the source buffer if necessary to show you the context of the current
14195 The source files displayed in Emacs are in ordinary Emacs buffers
14196 which are visiting the source files in the usual way. You can edit
14197 the files with these buffers if you wish; but keep in mind that @value{GDBN}
14198 communicates with Emacs in terms of line numbers. If you add or
14199 delete lines from the text, the line numbers that @value{GDBN} knows cease
14200 to correspond properly with the code.
14202 @c The following dropped because Epoch is nonstandard. Reactivate
14203 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
14205 @kindex Emacs Epoch environment
14209 Version 18 of @sc{gnu} Emacs has a built-in window system
14210 called the @code{epoch}
14211 environment. Users of this environment can use a new command,
14212 @code{inspect} which performs identically to @code{print} except that
14213 each value is printed in its own window.
14216 @include annotate.texi
14217 @include gdbmi.texinfo
14220 @chapter Reporting Bugs in @value{GDBN}
14221 @cindex bugs in @value{GDBN}
14222 @cindex reporting bugs in @value{GDBN}
14224 Your bug reports play an essential role in making @value{GDBN} reliable.
14226 Reporting a bug may help you by bringing a solution to your problem, or it
14227 may not. But in any case the principal function of a bug report is to help
14228 the entire community by making the next version of @value{GDBN} work better. Bug
14229 reports are your contribution to the maintenance of @value{GDBN}.
14231 In order for a bug report to serve its purpose, you must include the
14232 information that enables us to fix the bug.
14235 * Bug Criteria:: Have you found a bug?
14236 * Bug Reporting:: How to report bugs
14240 @section Have you found a bug?
14241 @cindex bug criteria
14243 If you are not sure whether you have found a bug, here are some guidelines:
14246 @cindex fatal signal
14247 @cindex debugger crash
14248 @cindex crash of debugger
14250 If the debugger gets a fatal signal, for any input whatever, that is a
14251 @value{GDBN} bug. Reliable debuggers never crash.
14253 @cindex error on valid input
14255 If @value{GDBN} produces an error message for valid input, that is a
14256 bug. (Note that if you're cross debugging, the problem may also be
14257 somewhere in the connection to the target.)
14259 @cindex invalid input
14261 If @value{GDBN} does not produce an error message for invalid input,
14262 that is a bug. However, you should note that your idea of
14263 ``invalid input'' might be our idea of ``an extension'' or ``support
14264 for traditional practice''.
14267 If you are an experienced user of debugging tools, your suggestions
14268 for improvement of @value{GDBN} are welcome in any case.
14271 @node Bug Reporting
14272 @section How to report bugs
14273 @cindex bug reports
14274 @cindex @value{GDBN} bugs, reporting
14276 A number of companies and individuals offer support for @sc{gnu} products.
14277 If you obtained @value{GDBN} from a support organization, we recommend you
14278 contact that organization first.
14280 You can find contact information for many support companies and
14281 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
14283 @c should add a web page ref...
14285 In any event, we also recommend that you send bug reports for
14286 @value{GDBN} to this addresses:
14292 @strong{Do not send bug reports to @samp{info-gdb}, or to
14293 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
14294 not want to receive bug reports. Those that do have arranged to receive
14297 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
14298 serves as a repeater. The mailing list and the newsgroup carry exactly
14299 the same messages. Often people think of posting bug reports to the
14300 newsgroup instead of mailing them. This appears to work, but it has one
14301 problem which can be crucial: a newsgroup posting often lacks a mail
14302 path back to the sender. Thus, if we need to ask for more information,
14303 we may be unable to reach you. For this reason, it is better to send
14304 bug reports to the mailing list.
14306 As a last resort, send bug reports on paper to:
14309 @sc{gnu} Debugger Bugs
14310 Free Software Foundation Inc.
14311 59 Temple Place - Suite 330
14312 Boston, MA 02111-1307
14316 The fundamental principle of reporting bugs usefully is this:
14317 @strong{report all the facts}. If you are not sure whether to state a
14318 fact or leave it out, state it!
14320 Often people omit facts because they think they know what causes the
14321 problem and assume that some details do not matter. Thus, you might
14322 assume that the name of the variable you use in an example does not matter.
14323 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
14324 stray memory reference which happens to fetch from the location where that
14325 name is stored in memory; perhaps, if the name were different, the contents
14326 of that location would fool the debugger into doing the right thing despite
14327 the bug. Play it safe and give a specific, complete example. That is the
14328 easiest thing for you to do, and the most helpful.
14330 Keep in mind that the purpose of a bug report is to enable us to fix the
14331 bug. It may be that the bug has been reported previously, but neither
14332 you nor we can know that unless your bug report is complete and
14335 Sometimes people give a few sketchy facts and ask, ``Does this ring a
14336 bell?'' Those bug reports are useless, and we urge everyone to
14337 @emph{refuse to respond to them} except to chide the sender to report
14340 To enable us to fix the bug, you should include all these things:
14344 The version of @value{GDBN}. @value{GDBN} announces it if you start
14345 with no arguments; you can also print it at any time using @code{show
14348 Without this, we will not know whether there is any point in looking for
14349 the bug in the current version of @value{GDBN}.
14352 The type of machine you are using, and the operating system name and
14356 What compiler (and its version) was used to compile @value{GDBN}---e.g.
14357 ``@value{GCC}--2.8.1''.
14360 What compiler (and its version) was used to compile the program you are
14361 debugging---e.g. ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
14362 C Compiler''. For GCC, you can say @code{gcc --version} to get this
14363 information; for other compilers, see the documentation for those
14367 The command arguments you gave the compiler to compile your example and
14368 observe the bug. For example, did you use @samp{-O}? To guarantee
14369 you will not omit something important, list them all. A copy of the
14370 Makefile (or the output from make) is sufficient.
14372 If we were to try to guess the arguments, we would probably guess wrong
14373 and then we might not encounter the bug.
14376 A complete input script, and all necessary source files, that will
14380 A description of what behavior you observe that you believe is
14381 incorrect. For example, ``It gets a fatal signal.''
14383 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
14384 will certainly notice it. But if the bug is incorrect output, we might
14385 not notice unless it is glaringly wrong. You might as well not give us
14386 a chance to make a mistake.
14388 Even if the problem you experience is a fatal signal, you should still
14389 say so explicitly. Suppose something strange is going on, such as, your
14390 copy of @value{GDBN} is out of synch, or you have encountered a bug in
14391 the C library on your system. (This has happened!) Your copy might
14392 crash and ours would not. If you told us to expect a crash, then when
14393 ours fails to crash, we would know that the bug was not happening for
14394 us. If you had not told us to expect a crash, then we would not be able
14395 to draw any conclusion from our observations.
14398 If you wish to suggest changes to the @value{GDBN} source, send us context
14399 diffs. If you even discuss something in the @value{GDBN} source, refer to
14400 it by context, not by line number.
14402 The line numbers in our development sources will not match those in your
14403 sources. Your line numbers would convey no useful information to us.
14407 Here are some things that are not necessary:
14411 A description of the envelope of the bug.
14413 Often people who encounter a bug spend a lot of time investigating
14414 which changes to the input file will make the bug go away and which
14415 changes will not affect it.
14417 This is often time consuming and not very useful, because the way we
14418 will find the bug is by running a single example under the debugger
14419 with breakpoints, not by pure deduction from a series of examples.
14420 We recommend that you save your time for something else.
14422 Of course, if you can find a simpler example to report @emph{instead}
14423 of the original one, that is a convenience for us. Errors in the
14424 output will be easier to spot, running under the debugger will take
14425 less time, and so on.
14427 However, simplification is not vital; if you do not want to do this,
14428 report the bug anyway and send us the entire test case you used.
14431 A patch for the bug.
14433 A patch for the bug does help us if it is a good one. But do not omit
14434 the necessary information, such as the test case, on the assumption that
14435 a patch is all we need. We might see problems with your patch and decide
14436 to fix the problem another way, or we might not understand it at all.
14438 Sometimes with a program as complicated as @value{GDBN} it is very hard to
14439 construct an example that will make the program follow a certain path
14440 through the code. If you do not send us the example, we will not be able
14441 to construct one, so we will not be able to verify that the bug is fixed.
14443 And if we cannot understand what bug you are trying to fix, or why your
14444 patch should be an improvement, we will not install it. A test case will
14445 help us to understand.
14448 A guess about what the bug is or what it depends on.
14450 Such guesses are usually wrong. Even we cannot guess right about such
14451 things without first using the debugger to find the facts.
14454 @c The readline documentation is distributed with the readline code
14455 @c and consists of the two following files:
14457 @c inc-hist.texinfo
14458 @c Use -I with makeinfo to point to the appropriate directory,
14459 @c environment var TEXINPUTS with TeX.
14460 @include rluser.texinfo
14461 @include inc-hist.texinfo
14464 @node Formatting Documentation
14465 @appendix Formatting Documentation
14467 @cindex @value{GDBN} reference card
14468 @cindex reference card
14469 The @value{GDBN} 4 release includes an already-formatted reference card, ready
14470 for printing with PostScript or Ghostscript, in the @file{gdb}
14471 subdirectory of the main source directory@footnote{In
14472 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
14473 release.}. If you can use PostScript or Ghostscript with your printer,
14474 you can print the reference card immediately with @file{refcard.ps}.
14476 The release also includes the source for the reference card. You
14477 can format it, using @TeX{}, by typing:
14483 The @value{GDBN} reference card is designed to print in @dfn{landscape}
14484 mode on US ``letter'' size paper;
14485 that is, on a sheet 11 inches wide by 8.5 inches
14486 high. You will need to specify this form of printing as an option to
14487 your @sc{dvi} output program.
14489 @cindex documentation
14491 All the documentation for @value{GDBN} comes as part of the machine-readable
14492 distribution. The documentation is written in Texinfo format, which is
14493 a documentation system that uses a single source file to produce both
14494 on-line information and a printed manual. You can use one of the Info
14495 formatting commands to create the on-line version of the documentation
14496 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
14498 @value{GDBN} includes an already formatted copy of the on-line Info
14499 version of this manual in the @file{gdb} subdirectory. The main Info
14500 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
14501 subordinate files matching @samp{gdb.info*} in the same directory. If
14502 necessary, you can print out these files, or read them with any editor;
14503 but they are easier to read using the @code{info} subsystem in @sc{gnu}
14504 Emacs or the standalone @code{info} program, available as part of the
14505 @sc{gnu} Texinfo distribution.
14507 If you want to format these Info files yourself, you need one of the
14508 Info formatting programs, such as @code{texinfo-format-buffer} or
14511 If you have @code{makeinfo} installed, and are in the top level
14512 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
14513 version @value{GDBVN}), you can make the Info file by typing:
14520 If you want to typeset and print copies of this manual, you need @TeX{},
14521 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
14522 Texinfo definitions file.
14524 @TeX{} is a typesetting program; it does not print files directly, but
14525 produces output files called @sc{dvi} files. To print a typeset
14526 document, you need a program to print @sc{dvi} files. If your system
14527 has @TeX{} installed, chances are it has such a program. The precise
14528 command to use depends on your system; @kbd{lpr -d} is common; another
14529 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
14530 require a file name without any extension or a @samp{.dvi} extension.
14532 @TeX{} also requires a macro definitions file called
14533 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
14534 written in Texinfo format. On its own, @TeX{} cannot either read or
14535 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
14536 and is located in the @file{gdb-@var{version-number}/texinfo}
14539 If you have @TeX{} and a @sc{dvi} printer program installed, you can
14540 typeset and print this manual. First switch to the the @file{gdb}
14541 subdirectory of the main source directory (for example, to
14542 @file{gdb-@value{GDBVN}/gdb}) and type:
14548 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
14550 @node Installing GDB
14551 @appendix Installing @value{GDBN}
14552 @cindex configuring @value{GDBN}
14553 @cindex installation
14555 @value{GDBN} comes with a @code{configure} script that automates the process
14556 of preparing @value{GDBN} for installation; you can then use @code{make} to
14557 build the @code{gdb} program.
14559 @c irrelevant in info file; it's as current as the code it lives with.
14560 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
14561 look at the @file{README} file in the sources; we may have improved the
14562 installation procedures since publishing this manual.}
14565 The @value{GDBN} distribution includes all the source code you need for
14566 @value{GDBN} in a single directory, whose name is usually composed by
14567 appending the version number to @samp{gdb}.
14569 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
14570 @file{gdb-@value{GDBVN}} directory. That directory contains:
14573 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
14574 script for configuring @value{GDBN} and all its supporting libraries
14576 @item gdb-@value{GDBVN}/gdb
14577 the source specific to @value{GDBN} itself
14579 @item gdb-@value{GDBVN}/bfd
14580 source for the Binary File Descriptor library
14582 @item gdb-@value{GDBVN}/include
14583 @sc{gnu} include files
14585 @item gdb-@value{GDBVN}/libiberty
14586 source for the @samp{-liberty} free software library
14588 @item gdb-@value{GDBVN}/opcodes
14589 source for the library of opcode tables and disassemblers
14591 @item gdb-@value{GDBVN}/readline
14592 source for the @sc{gnu} command-line interface
14594 @item gdb-@value{GDBVN}/glob
14595 source for the @sc{gnu} filename pattern-matching subroutine
14597 @item gdb-@value{GDBVN}/mmalloc
14598 source for the @sc{gnu} memory-mapped malloc package
14601 The simplest way to configure and build @value{GDBN} is to run @code{configure}
14602 from the @file{gdb-@var{version-number}} source directory, which in
14603 this example is the @file{gdb-@value{GDBVN}} directory.
14605 First switch to the @file{gdb-@var{version-number}} source directory
14606 if you are not already in it; then run @code{configure}. Pass the
14607 identifier for the platform on which @value{GDBN} will run as an
14613 cd gdb-@value{GDBVN}
14614 ./configure @var{host}
14619 where @var{host} is an identifier such as @samp{sun4} or
14620 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
14621 (You can often leave off @var{host}; @code{configure} tries to guess the
14622 correct value by examining your system.)
14624 Running @samp{configure @var{host}} and then running @code{make} builds the
14625 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
14626 libraries, then @code{gdb} itself. The configured source files, and the
14627 binaries, are left in the corresponding source directories.
14630 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
14631 system does not recognize this automatically when you run a different
14632 shell, you may need to run @code{sh} on it explicitly:
14635 sh configure @var{host}
14638 If you run @code{configure} from a directory that contains source
14639 directories for multiple libraries or programs, such as the
14640 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN}, @code{configure}
14641 creates configuration files for every directory level underneath (unless
14642 you tell it not to, with the @samp{--norecursion} option).
14644 You can run the @code{configure} script from any of the
14645 subordinate directories in the @value{GDBN} distribution if you only want to
14646 configure that subdirectory, but be sure to specify a path to it.
14648 For example, with version @value{GDBVN}, type the following to configure only
14649 the @code{bfd} subdirectory:
14653 cd gdb-@value{GDBVN}/bfd
14654 ../configure @var{host}
14658 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
14659 However, you should make sure that the shell on your path (named by
14660 the @samp{SHELL} environment variable) is publicly readable. Remember
14661 that @value{GDBN} uses the shell to start your program---some systems refuse to
14662 let @value{GDBN} debug child processes whose programs are not readable.
14665 * Separate Objdir:: Compiling @value{GDBN} in another directory
14666 * Config Names:: Specifying names for hosts and targets
14667 * Configure Options:: Summary of options for configure
14670 @node Separate Objdir
14671 @section Compiling @value{GDBN} in another directory
14673 If you want to run @value{GDBN} versions for several host or target machines,
14674 you need a different @code{gdb} compiled for each combination of
14675 host and target. @code{configure} is designed to make this easy by
14676 allowing you to generate each configuration in a separate subdirectory,
14677 rather than in the source directory. If your @code{make} program
14678 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
14679 @code{make} in each of these directories builds the @code{gdb}
14680 program specified there.
14682 To build @code{gdb} in a separate directory, run @code{configure}
14683 with the @samp{--srcdir} option to specify where to find the source.
14684 (You also need to specify a path to find @code{configure}
14685 itself from your working directory. If the path to @code{configure}
14686 would be the same as the argument to @samp{--srcdir}, you can leave out
14687 the @samp{--srcdir} option; it is assumed.)
14689 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
14690 separate directory for a Sun 4 like this:
14694 cd gdb-@value{GDBVN}
14697 ../gdb-@value{GDBVN}/configure sun4
14702 When @code{configure} builds a configuration using a remote source
14703 directory, it creates a tree for the binaries with the same structure
14704 (and using the same names) as the tree under the source directory. In
14705 the example, you'd find the Sun 4 library @file{libiberty.a} in the
14706 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
14707 @file{gdb-sun4/gdb}.
14709 One popular reason to build several @value{GDBN} configurations in separate
14710 directories is to configure @value{GDBN} for cross-compiling (where
14711 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
14712 programs that run on another machine---the @dfn{target}).
14713 You specify a cross-debugging target by
14714 giving the @samp{--target=@var{target}} option to @code{configure}.
14716 When you run @code{make} to build a program or library, you must run
14717 it in a configured directory---whatever directory you were in when you
14718 called @code{configure} (or one of its subdirectories).
14720 The @code{Makefile} that @code{configure} generates in each source
14721 directory also runs recursively. If you type @code{make} in a source
14722 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
14723 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
14724 will build all the required libraries, and then build GDB.
14726 When you have multiple hosts or targets configured in separate
14727 directories, you can run @code{make} on them in parallel (for example,
14728 if they are NFS-mounted on each of the hosts); they will not interfere
14732 @section Specifying names for hosts and targets
14734 The specifications used for hosts and targets in the @code{configure}
14735 script are based on a three-part naming scheme, but some short predefined
14736 aliases are also supported. The full naming scheme encodes three pieces
14737 of information in the following pattern:
14740 @var{architecture}-@var{vendor}-@var{os}
14743 For example, you can use the alias @code{sun4} as a @var{host} argument,
14744 or as the value for @var{target} in a @code{--target=@var{target}}
14745 option. The equivalent full name is @samp{sparc-sun-sunos4}.
14747 The @code{configure} script accompanying @value{GDBN} does not provide
14748 any query facility to list all supported host and target names or
14749 aliases. @code{configure} calls the Bourne shell script
14750 @code{config.sub} to map abbreviations to full names; you can read the
14751 script, if you wish, or you can use it to test your guesses on
14752 abbreviations---for example:
14755 % sh config.sub i386-linux
14757 % sh config.sub alpha-linux
14758 alpha-unknown-linux-gnu
14759 % sh config.sub hp9k700
14761 % sh config.sub sun4
14762 sparc-sun-sunos4.1.1
14763 % sh config.sub sun3
14764 m68k-sun-sunos4.1.1
14765 % sh config.sub i986v
14766 Invalid configuration `i986v': machine `i986v' not recognized
14770 @code{config.sub} is also distributed in the @value{GDBN} source
14771 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
14773 @node Configure Options
14774 @section @code{configure} options
14776 Here is a summary of the @code{configure} options and arguments that
14777 are most often useful for building @value{GDBN}. @code{configure} also has
14778 several other options not listed here. @inforef{What Configure
14779 Does,,configure.info}, for a full explanation of @code{configure}.
14782 configure @r{[}--help@r{]}
14783 @r{[}--prefix=@var{dir}@r{]}
14784 @r{[}--exec-prefix=@var{dir}@r{]}
14785 @r{[}--srcdir=@var{dirname}@r{]}
14786 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
14787 @r{[}--target=@var{target}@r{]}
14792 You may introduce options with a single @samp{-} rather than
14793 @samp{--} if you prefer; but you may abbreviate option names if you use
14798 Display a quick summary of how to invoke @code{configure}.
14800 @item --prefix=@var{dir}
14801 Configure the source to install programs and files under directory
14804 @item --exec-prefix=@var{dir}
14805 Configure the source to install programs under directory
14808 @c avoid splitting the warning from the explanation:
14810 @item --srcdir=@var{dirname}
14811 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
14812 @code{make} that implements the @code{VPATH} feature.}@*
14813 Use this option to make configurations in directories separate from the
14814 @value{GDBN} source directories. Among other things, you can use this to
14815 build (or maintain) several configurations simultaneously, in separate
14816 directories. @code{configure} writes configuration specific files in
14817 the current directory, but arranges for them to use the source in the
14818 directory @var{dirname}. @code{configure} creates directories under
14819 the working directory in parallel to the source directories below
14822 @item --norecursion
14823 Configure only the directory level where @code{configure} is executed; do not
14824 propagate configuration to subdirectories.
14826 @item --target=@var{target}
14827 Configure @value{GDBN} for cross-debugging programs running on the specified
14828 @var{target}. Without this option, @value{GDBN} is configured to debug
14829 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
14831 There is no convenient way to generate a list of all available targets.
14833 @item @var{host} @dots{}
14834 Configure @value{GDBN} to run on the specified @var{host}.
14836 There is no convenient way to generate a list of all available hosts.
14839 There are many other options available as well, but they are generally
14840 needed for special purposes only.
14848 % I think something like @colophon should be in texinfo. In the
14850 \long\def\colophon{\hbox to0pt{}\vfill
14851 \centerline{The body of this manual is set in}
14852 \centerline{\fontname\tenrm,}
14853 \centerline{with headings in {\bf\fontname\tenbf}}
14854 \centerline{and examples in {\tt\fontname\tentt}.}
14855 \centerline{{\it\fontname\tenit\/},}
14856 \centerline{{\bf\fontname\tenbf}, and}
14857 \centerline{{\sl\fontname\tensl\/}}
14858 \centerline{are used for emphasis.}\vfill}
14860 % Blame: doc@cygnus.com, 1991.
14863 @c TeX can handle the contents at the start but makeinfo 3.12 can not