1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1998,
3 @c 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006
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!
31 @c This is updated by GNU Press.
34 @c !!set GDB edit command default editor
37 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
39 @c This is a dir.info fragment to support semi-automated addition of
40 @c manuals to an info tree.
41 @dircategory Software development
43 * Gdb: (gdb). The GNU debugger.
47 This file documents the @sc{gnu} debugger @value{GDBN}.
50 This is the @value{EDITION} Edition, of @cite{Debugging with
51 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
52 Version @value{GDBVN}.
54 Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1998,@*
55 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006@*
56 Free Software Foundation, Inc.
58 Permission is granted to copy, distribute and/or modify this document
59 under the terms of the GNU Free Documentation License, Version 1.1 or
60 any later version published by the Free Software Foundation; with the
61 Invariant Sections being ``Free Software'' and ``Free Software Needs
62 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
63 and with the Back-Cover Texts as in (a) below.
65 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
66 this GNU Manual. Buying copies from GNU Press supports the FSF in
67 developing GNU and promoting software freedom.''
71 @title Debugging with @value{GDBN}
72 @subtitle The @sc{gnu} Source-Level Debugger
74 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
75 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
79 \hfill (Send bugs and comments on @value{GDBN} to bug-gdb\@gnu.org.)\par
80 \hfill {\it Debugging with @value{GDBN}}\par
81 \hfill \TeX{}info \texinfoversion\par
85 @vskip 0pt plus 1filll
86 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995,
87 1996, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2006
88 Free Software Foundation, Inc.
90 Published by the Free Software Foundation @*
91 51 Franklin Street, Fifth Floor,
92 Boston, MA 02110-1301, USA@*
95 Permission is granted to copy, distribute and/or modify this document
96 under the terms of the GNU Free Documentation License, Version 1.1 or
97 any later version published by the Free Software Foundation; with the
98 Invariant Sections being ``Free Software'' and ``Free Software Needs
99 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
100 and with the Back-Cover Texts as in (a) below.
102 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
103 this GNU Manual. Buying copies from GNU Press supports the FSF in
104 developing GNU and promoting software freedom.''
106 This edition of the GDB manual is dedicated to the memory of Fred
107 Fish. Fred was a long-standing contributor to GDB and to Free
108 software in general. We will miss him.
113 @node Top, Summary, (dir), (dir)
115 @top Debugging with @value{GDBN}
117 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
119 This is the @value{EDITION} Edition, for @value{GDBN} Version
122 Copyright (C) 1988-2006 Free Software Foundation, Inc.
124 This edition of the GDB manual is dedicated to the memory of Fred
125 Fish. Fred was a long-standing contributor to GDB and to Free
126 software in general. We will miss him.
129 * Summary:: Summary of @value{GDBN}
130 * Sample Session:: A sample @value{GDBN} session
132 * Invocation:: Getting in and out of @value{GDBN}
133 * Commands:: @value{GDBN} commands
134 * Running:: Running programs under @value{GDBN}
135 * Stopping:: Stopping and continuing
136 * Stack:: Examining the stack
137 * Source:: Examining source files
138 * Data:: Examining data
139 * Macros:: Preprocessor Macros
140 * Tracepoints:: Debugging remote targets non-intrusively
141 * Overlays:: Debugging programs that use overlays
143 * Languages:: Using @value{GDBN} with different languages
145 * Symbols:: Examining the symbol table
146 * Altering:: Altering execution
147 * GDB Files:: @value{GDBN} files
148 * Targets:: Specifying a debugging target
149 * Remote Debugging:: Debugging remote programs
150 * Configurations:: Configuration-specific information
151 * Controlling GDB:: Controlling @value{GDBN}
152 * Sequences:: Canned sequences of commands
153 * Interpreters:: Command Interpreters
154 * TUI:: @value{GDBN} Text User Interface
155 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
156 * GDB/MI:: @value{GDBN}'s Machine Interface.
157 * Annotations:: @value{GDBN}'s annotation interface.
159 * GDB Bugs:: Reporting bugs in @value{GDBN}
161 * Command Line Editing:: Command Line Editing
162 * Using History Interactively:: Using History Interactively
163 * Formatting Documentation:: How to format and print @value{GDBN} documentation
164 * Installing GDB:: Installing GDB
165 * Maintenance Commands:: Maintenance Commands
166 * Remote Protocol:: GDB Remote Serial Protocol
167 * Agent Expressions:: The GDB Agent Expression Mechanism
168 * Target Descriptions:: How targets can describe themselves to
170 * Copying:: GNU General Public License says
171 how you can copy and share GDB
172 * GNU Free Documentation License:: The license for this documentation
181 @unnumbered Summary of @value{GDBN}
183 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
184 going on ``inside'' another program while it executes---or what another
185 program was doing at the moment it crashed.
187 @value{GDBN} can do four main kinds of things (plus other things in support of
188 these) to help you catch bugs in the act:
192 Start your program, specifying anything that might affect its behavior.
195 Make your program stop on specified conditions.
198 Examine what has happened, when your program has stopped.
201 Change things in your program, so you can experiment with correcting the
202 effects of one bug and go on to learn about another.
205 You can use @value{GDBN} to debug programs written in C and C@t{++}.
206 For more information, see @ref{Supported Languages,,Supported Languages}.
207 For more information, see @ref{C,,C and C++}.
210 Support for Modula-2 is partial. For information on Modula-2, see
211 @ref{Modula-2,,Modula-2}.
214 Debugging Pascal programs which use sets, subranges, file variables, or
215 nested functions does not currently work. @value{GDBN} does not support
216 entering expressions, printing values, or similar features using Pascal
220 @value{GDBN} can be used to debug programs written in Fortran, although
221 it may be necessary to refer to some variables with a trailing
224 @value{GDBN} can be used to debug programs written in Objective-C,
225 using either the Apple/NeXT or the GNU Objective-C runtime.
228 * Free Software:: Freely redistributable software
229 * Contributors:: Contributors to GDB
233 @unnumberedsec Free Software
235 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
236 General Public License
237 (GPL). The GPL gives you the freedom to copy or adapt a licensed
238 program---but every person getting a copy also gets with it the
239 freedom to modify that copy (which means that they must get access to
240 the source code), and the freedom to distribute further copies.
241 Typical software companies use copyrights to limit your freedoms; the
242 Free Software Foundation uses the GPL to preserve these freedoms.
244 Fundamentally, the General Public License is a license which says that
245 you have these freedoms and that you cannot take these freedoms away
248 @unnumberedsec Free Software Needs Free Documentation
250 The biggest deficiency in the free software community today is not in
251 the software---it is the lack of good free documentation that we can
252 include with the free software. Many of our most important
253 programs do not come with free reference manuals and free introductory
254 texts. Documentation is an essential part of any software package;
255 when an important free software package does not come with a free
256 manual and a free tutorial, that is a major gap. We have many such
259 Consider Perl, for instance. The tutorial manuals that people
260 normally use are non-free. How did this come about? Because the
261 authors of those manuals published them with restrictive terms---no
262 copying, no modification, source files not available---which exclude
263 them from the free software world.
265 That wasn't the first time this sort of thing happened, and it was far
266 from the last. Many times we have heard a GNU user eagerly describe a
267 manual that he is writing, his intended contribution to the community,
268 only to learn that he had ruined everything by signing a publication
269 contract to make it non-free.
271 Free documentation, like free software, is a matter of freedom, not
272 price. The problem with the non-free manual is not that publishers
273 charge a price for printed copies---that in itself is fine. (The Free
274 Software Foundation sells printed copies of manuals, too.) The
275 problem is the restrictions on the use of the manual. Free manuals
276 are available in source code form, and give you permission to copy and
277 modify. Non-free manuals do not allow this.
279 The criteria of freedom for a free manual are roughly the same as for
280 free software. Redistribution (including the normal kinds of
281 commercial redistribution) must be permitted, so that the manual can
282 accompany every copy of the program, both on-line and on paper.
284 Permission for modification of the technical content is crucial too.
285 When people modify the software, adding or changing features, if they
286 are conscientious they will change the manual too---so they can
287 provide accurate and clear documentation for the modified program. A
288 manual that leaves you no choice but to write a new manual to document
289 a changed version of the program is not really available to our
292 Some kinds of limits on the way modification is handled are
293 acceptable. For example, requirements to preserve the original
294 author's copyright notice, the distribution terms, or the list of
295 authors, are ok. It is also no problem to require modified versions
296 to include notice that they were modified. Even entire sections that
297 may not be deleted or changed are acceptable, as long as they deal
298 with nontechnical topics (like this one). These kinds of restrictions
299 are acceptable because they don't obstruct the community's normal use
302 However, it must be possible to modify all the @emph{technical}
303 content of the manual, and then distribute the result in all the usual
304 media, through all the usual channels. Otherwise, the restrictions
305 obstruct the use of the manual, it is not free, and we need another
306 manual to replace it.
308 Please spread the word about this issue. Our community continues to
309 lose manuals to proprietary publishing. If we spread the word that
310 free software needs free reference manuals and free tutorials, perhaps
311 the next person who wants to contribute by writing documentation will
312 realize, before it is too late, that only free manuals contribute to
313 the free software community.
315 If you are writing documentation, please insist on publishing it under
316 the GNU Free Documentation License or another free documentation
317 license. Remember that this decision requires your approval---you
318 don't have to let the publisher decide. Some commercial publishers
319 will use a free license if you insist, but they will not propose the
320 option; it is up to you to raise the issue and say firmly that this is
321 what you want. If the publisher you are dealing with refuses, please
322 try other publishers. If you're not sure whether a proposed license
323 is free, write to @email{licensing@@gnu.org}.
325 You can encourage commercial publishers to sell more free, copylefted
326 manuals and tutorials by buying them, and particularly by buying
327 copies from the publishers that paid for their writing or for major
328 improvements. Meanwhile, try to avoid buying non-free documentation
329 at all. Check the distribution terms of a manual before you buy it,
330 and insist that whoever seeks your business must respect your freedom.
331 Check the history of the book, and try to reward the publishers that
332 have paid or pay the authors to work on it.
334 The Free Software Foundation maintains a list of free documentation
335 published by other publishers, at
336 @url{http://www.fsf.org/doc/other-free-books.html}.
339 @unnumberedsec Contributors to @value{GDBN}
341 Richard Stallman was the original author of @value{GDBN}, and of many
342 other @sc{gnu} programs. Many others have contributed to its
343 development. This section attempts to credit major contributors. One
344 of the virtues of free software is that everyone is free to contribute
345 to it; with regret, we cannot actually acknowledge everyone here. The
346 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
347 blow-by-blow account.
349 Changes much prior to version 2.0 are lost in the mists of time.
352 @emph{Plea:} Additions to this section are particularly welcome. If you
353 or your friends (or enemies, to be evenhanded) have been unfairly
354 omitted from this list, we would like to add your names!
357 So that they may not regard their many labors as thankless, we
358 particularly thank those who shepherded @value{GDBN} through major
360 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
361 Jim Blandy (release 4.18);
362 Jason Molenda (release 4.17);
363 Stan Shebs (release 4.14);
364 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
365 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
366 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
367 Jim Kingdon (releases 3.5, 3.4, and 3.3);
368 and Randy Smith (releases 3.2, 3.1, and 3.0).
370 Richard Stallman, assisted at various times by Peter TerMaat, Chris
371 Hanson, and Richard Mlynarik, handled releases through 2.8.
373 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
374 in @value{GDBN}, with significant additional contributions from Per
375 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
376 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
377 much general update work leading to release 3.0).
379 @value{GDBN} uses the BFD subroutine library to examine multiple
380 object-file formats; BFD was a joint project of David V.
381 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
383 David Johnson wrote the original COFF support; Pace Willison did
384 the original support for encapsulated COFF.
386 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
388 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
389 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
391 Jean-Daniel Fekete contributed Sun 386i support.
392 Chris Hanson improved the HP9000 support.
393 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
394 David Johnson contributed Encore Umax support.
395 Jyrki Kuoppala contributed Altos 3068 support.
396 Jeff Law contributed HP PA and SOM support.
397 Keith Packard contributed NS32K support.
398 Doug Rabson contributed Acorn Risc Machine support.
399 Bob Rusk contributed Harris Nighthawk CX-UX support.
400 Chris Smith contributed Convex support (and Fortran debugging).
401 Jonathan Stone contributed Pyramid support.
402 Michael Tiemann contributed SPARC support.
403 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
404 Pace Willison contributed Intel 386 support.
405 Jay Vosburgh contributed Symmetry support.
406 Marko Mlinar contributed OpenRISC 1000 support.
408 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
410 Rich Schaefer and Peter Schauer helped with support of SunOS shared
413 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
414 about several machine instruction sets.
416 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
417 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
418 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
419 and RDI targets, respectively.
421 Brian Fox is the author of the readline libraries providing
422 command-line editing and command history.
424 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
425 Modula-2 support, and contributed the Languages chapter of this manual.
427 Fred Fish wrote most of the support for Unix System Vr4.
428 He also enhanced the command-completion support to cover C@t{++} overloaded
431 Hitachi America (now Renesas America), Ltd. sponsored the support for
432 H8/300, H8/500, and Super-H processors.
434 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
436 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
439 Toshiba sponsored the support for the TX39 Mips processor.
441 Matsushita sponsored the support for the MN10200 and MN10300 processors.
443 Fujitsu sponsored the support for SPARClite and FR30 processors.
445 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
448 Michael Snyder added support for tracepoints.
450 Stu Grossman wrote gdbserver.
452 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
453 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
455 The following people at the Hewlett-Packard Company contributed
456 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
457 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
458 compiler, and the Text User Interface (nee Terminal User Interface):
459 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
460 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
461 provided HP-specific information in this manual.
463 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
464 Robert Hoehne made significant contributions to the DJGPP port.
466 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
467 development since 1991. Cygnus engineers who have worked on @value{GDBN}
468 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
469 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
470 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
471 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
472 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
473 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
474 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
475 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
476 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
477 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
478 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
479 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
480 Zuhn have made contributions both large and small.
482 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
483 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
485 Jim Blandy added support for preprocessor macros, while working for Red
488 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
489 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
490 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
491 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
492 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
493 with the migration of old architectures to this new framework.
495 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
496 unwinder framework, this consisting of a fresh new design featuring
497 frame IDs, independent frame sniffers, and the sentinel frame. Mark
498 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
499 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
500 trad unwinders. The architecture-specific changes, each involving a
501 complete rewrite of the architecture's frame code, were carried out by
502 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
503 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
504 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
505 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
508 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
509 Tensilica, Inc.@: contributed support for Xtensa processors. Others
510 who have worked on the Xtensa port of @value{GDBN} in the past include
511 Steve Tjiang, John Newlin, and Scott Foehner.
514 @chapter A Sample @value{GDBN} Session
516 You can use this manual at your leisure to read all about @value{GDBN}.
517 However, a handful of commands are enough to get started using the
518 debugger. This chapter illustrates those commands.
521 In this sample session, we emphasize user input like this: @b{input},
522 to make it easier to pick out from the surrounding output.
525 @c FIXME: this example may not be appropriate for some configs, where
526 @c FIXME...primary interest is in remote use.
528 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
529 processor) exhibits the following bug: sometimes, when we change its
530 quote strings from the default, the commands used to capture one macro
531 definition within another stop working. In the following short @code{m4}
532 session, we define a macro @code{foo} which expands to @code{0000}; we
533 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
534 same thing. However, when we change the open quote string to
535 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
536 procedure fails to define a new synonym @code{baz}:
545 @b{define(bar,defn(`foo'))}
549 @b{changequote(<QUOTE>,<UNQUOTE>)}
551 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
554 m4: End of input: 0: fatal error: EOF in string
558 Let us use @value{GDBN} to try to see what is going on.
561 $ @b{@value{GDBP} m4}
562 @c FIXME: this falsifies the exact text played out, to permit smallbook
563 @c FIXME... format to come out better.
564 @value{GDBN} is free software and you are welcome to distribute copies
565 of it under certain conditions; type "show copying" to see
567 There is absolutely no warranty for @value{GDBN}; type "show warranty"
570 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
575 @value{GDBN} reads only enough symbol data to know where to find the
576 rest when needed; as a result, the first prompt comes up very quickly.
577 We now tell @value{GDBN} to use a narrower display width than usual, so
578 that examples fit in this manual.
581 (@value{GDBP}) @b{set width 70}
585 We need to see how the @code{m4} built-in @code{changequote} works.
586 Having looked at the source, we know the relevant subroutine is
587 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
588 @code{break} command.
591 (@value{GDBP}) @b{break m4_changequote}
592 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
596 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
597 control; as long as control does not reach the @code{m4_changequote}
598 subroutine, the program runs as usual:
601 (@value{GDBP}) @b{run}
602 Starting program: /work/Editorial/gdb/gnu/m4/m4
610 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
611 suspends execution of @code{m4}, displaying information about the
612 context where it stops.
615 @b{changequote(<QUOTE>,<UNQUOTE>)}
617 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
619 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
623 Now we use the command @code{n} (@code{next}) to advance execution to
624 the next line of the current function.
628 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
633 @code{set_quotes} looks like a promising subroutine. We can go into it
634 by using the command @code{s} (@code{step}) instead of @code{next}.
635 @code{step} goes to the next line to be executed in @emph{any}
636 subroutine, so it steps into @code{set_quotes}.
640 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
642 530 if (lquote != def_lquote)
646 The display that shows the subroutine where @code{m4} is now
647 suspended (and its arguments) is called a stack frame display. It
648 shows a summary of the stack. We can use the @code{backtrace}
649 command (which can also be spelled @code{bt}), to see where we are
650 in the stack as a whole: the @code{backtrace} command displays a
651 stack frame for each active subroutine.
654 (@value{GDBP}) @b{bt}
655 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
657 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
659 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
660 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
662 #4 0x79dc in expand_input () at macro.c:40
663 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
667 We step through a few more lines to see what happens. The first two
668 times, we can use @samp{s}; the next two times we use @code{n} to avoid
669 falling into the @code{xstrdup} subroutine.
673 0x3b5c 532 if (rquote != def_rquote)
675 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
676 def_lquote : xstrdup(lq);
678 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
681 538 len_lquote = strlen(rquote);
685 The last line displayed looks a little odd; we can examine the variables
686 @code{lquote} and @code{rquote} to see if they are in fact the new left
687 and right quotes we specified. We use the command @code{p}
688 (@code{print}) to see their values.
691 (@value{GDBP}) @b{p lquote}
692 $1 = 0x35d40 "<QUOTE>"
693 (@value{GDBP}) @b{p rquote}
694 $2 = 0x35d50 "<UNQUOTE>"
698 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
699 To look at some context, we can display ten lines of source
700 surrounding the current line with the @code{l} (@code{list}) command.
706 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
708 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
711 538 len_lquote = strlen(rquote);
712 539 len_rquote = strlen(lquote);
719 Let us step past the two lines that set @code{len_lquote} and
720 @code{len_rquote}, and then examine the values of those variables.
724 539 len_rquote = strlen(lquote);
727 (@value{GDBP}) @b{p len_lquote}
729 (@value{GDBP}) @b{p len_rquote}
734 That certainly looks wrong, assuming @code{len_lquote} and
735 @code{len_rquote} are meant to be the lengths of @code{lquote} and
736 @code{rquote} respectively. We can set them to better values using
737 the @code{p} command, since it can print the value of
738 any expression---and that expression can include subroutine calls and
742 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
744 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
749 Is that enough to fix the problem of using the new quotes with the
750 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
751 executing with the @code{c} (@code{continue}) command, and then try the
752 example that caused trouble initially:
758 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
765 Success! The new quotes now work just as well as the default ones. The
766 problem seems to have been just the two typos defining the wrong
767 lengths. We allow @code{m4} exit by giving it an EOF as input:
771 Program exited normally.
775 The message @samp{Program exited normally.} is from @value{GDBN}; it
776 indicates @code{m4} has finished executing. We can end our @value{GDBN}
777 session with the @value{GDBN} @code{quit} command.
780 (@value{GDBP}) @b{quit}
784 @chapter Getting In and Out of @value{GDBN}
786 This chapter discusses how to start @value{GDBN}, and how to get out of it.
790 type @samp{@value{GDBP}} to start @value{GDBN}.
792 type @kbd{quit} or @kbd{Ctrl-d} to exit.
796 * Invoking GDB:: How to start @value{GDBN}
797 * Quitting GDB:: How to quit @value{GDBN}
798 * Shell Commands:: How to use shell commands inside @value{GDBN}
799 * Logging Output:: How to log @value{GDBN}'s output to a file
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 @value{GDBP} --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
886 * Startup:: What @value{GDBN} does during startup
890 @subsection Choosing Files
892 When @value{GDBN} starts, it reads any arguments other than options as
893 specifying an executable file and core file (or process ID). This is
894 the same as if the arguments were specified by the @samp{-se} and
895 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
896 first argument that does not have an associated option flag as
897 equivalent to the @samp{-se} option followed by that argument; and the
898 second argument that does not have an associated option flag, if any, as
899 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
900 If the second argument begins with a decimal digit, @value{GDBN} will
901 first attempt to attach to it as a process, and if that fails, attempt
902 to open it as a corefile. If you have a corefile whose name begins with
903 a digit, you can prevent @value{GDBN} from treating it as a pid by
904 prefixing it with @file{./}, e.g.@: @file{./12345}.
906 If @value{GDBN} has not been configured to included core file support,
907 such as for most embedded targets, then it will complain about a second
908 argument and ignore it.
910 Many options have both long and short forms; both are shown in the
911 following list. @value{GDBN} also recognizes the long forms if you truncate
912 them, so long as enough of the option is present to be unambiguous.
913 (If you prefer, you can flag option arguments with @samp{--} rather
914 than @samp{-}, though we illustrate the more usual convention.)
916 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
917 @c way, both those who look for -foo and --foo in the index, will find
921 @item -symbols @var{file}
923 @cindex @code{--symbols}
925 Read symbol table from file @var{file}.
927 @item -exec @var{file}
929 @cindex @code{--exec}
931 Use file @var{file} as the executable file to execute when appropriate,
932 and for examining pure data in conjunction with a core dump.
936 Read symbol table from file @var{file} and use it as the executable
939 @item -core @var{file}
941 @cindex @code{--core}
943 Use file @var{file} as a core dump to examine.
945 @item -pid @var{number}
946 @itemx -p @var{number}
949 Connect to process ID @var{number}, as with the @code{attach} command.
951 @item -command @var{file}
953 @cindex @code{--command}
955 Execute @value{GDBN} commands from file @var{file}. @xref{Command
956 Files,, Command files}.
958 @item -eval-command @var{command}
959 @itemx -ex @var{command}
960 @cindex @code{--eval-command}
962 Execute a single @value{GDBN} command.
964 This option may be used multiple times to call multiple commands. It may
965 also be interleaved with @samp{-command} as required.
968 @value{GDBP} -ex 'target sim' -ex 'load' \
969 -x setbreakpoints -ex 'run' a.out
972 @item -directory @var{directory}
973 @itemx -d @var{directory}
974 @cindex @code{--directory}
976 Add @var{directory} to the path to search for source and script files.
980 @cindex @code{--readnow}
982 Read each symbol file's entire symbol table immediately, rather than
983 the default, which is to read it incrementally as it is needed.
984 This makes startup slower, but makes future operations faster.
989 @subsection Choosing Modes
991 You can run @value{GDBN} in various alternative modes---for example, in
992 batch mode or quiet mode.
999 Do not execute commands found in any initialization files. Normally,
1000 @value{GDBN} executes the commands in these files after all the command
1001 options and arguments have been processed. @xref{Command Files,,Command
1007 @cindex @code{--quiet}
1008 @cindex @code{--silent}
1010 ``Quiet''. Do not print the introductory and copyright messages. These
1011 messages are also suppressed in batch mode.
1014 @cindex @code{--batch}
1015 Run in batch mode. Exit with status @code{0} after processing all the
1016 command files specified with @samp{-x} (and all commands from
1017 initialization files, if not inhibited with @samp{-n}). Exit with
1018 nonzero status if an error occurs in executing the @value{GDBN} commands
1019 in the command files.
1021 Batch mode may be useful for running @value{GDBN} as a filter, for
1022 example to download and run a program on another computer; in order to
1023 make this more useful, the message
1026 Program exited normally.
1030 (which is ordinarily issued whenever a program running under
1031 @value{GDBN} control terminates) is not issued when running in batch
1035 @cindex @code{--batch-silent}
1036 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1037 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1038 unaffected). This is much quieter than @samp{-silent} and would be useless
1039 for an interactive session.
1041 This is particularly useful when using targets that give @samp{Loading section}
1042 messages, for example.
1044 Note that targets that give their output via @value{GDBN}, as opposed to
1045 writing directly to @code{stdout}, will also be made silent.
1047 @item -return-child-result
1048 @cindex @code{--return-child-result}
1049 The return code from @value{GDBN} will be the return code from the child
1050 process (the process being debugged), with the following exceptions:
1054 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1055 internal error. In this case the exit code is the same as it would have been
1056 without @samp{-return-child-result}.
1058 The user quits with an explicit value. E.g., @samp{quit 1}.
1060 The child process never runs, or is not allowed to terminate, in which case
1061 the exit code will be -1.
1064 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1065 when @value{GDBN} is being used as a remote program loader or simulator
1070 @cindex @code{--nowindows}
1072 ``No windows''. If @value{GDBN} comes with a graphical user interface
1073 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1074 interface. If no GUI is available, this option has no effect.
1078 @cindex @code{--windows}
1080 If @value{GDBN} includes a GUI, then this option requires it to be
1083 @item -cd @var{directory}
1085 Run @value{GDBN} using @var{directory} as its working directory,
1086 instead of the current directory.
1090 @cindex @code{--fullname}
1092 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1093 subprocess. It tells @value{GDBN} to output the full file name and line
1094 number in a standard, recognizable fashion each time a stack frame is
1095 displayed (which includes each time your program stops). This
1096 recognizable format looks like two @samp{\032} characters, followed by
1097 the file name, line number and character position separated by colons,
1098 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1099 @samp{\032} characters as a signal to display the source code for the
1103 @cindex @code{--epoch}
1104 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1105 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1106 routines so as to allow Epoch to display values of expressions in a
1109 @item -annotate @var{level}
1110 @cindex @code{--annotate}
1111 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1112 effect is identical to using @samp{set annotate @var{level}}
1113 (@pxref{Annotations}). The annotation @var{level} controls how much
1114 information @value{GDBN} prints together with its prompt, values of
1115 expressions, source lines, and other types of output. Level 0 is the
1116 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1117 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1118 that control @value{GDBN}, and level 2 has been deprecated.
1120 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1124 @cindex @code{--args}
1125 Change interpretation of command line so that arguments following the
1126 executable file are passed as command line arguments to the inferior.
1127 This option stops option processing.
1129 @item -baud @var{bps}
1131 @cindex @code{--baud}
1133 Set the line speed (baud rate or bits per second) of any serial
1134 interface used by @value{GDBN} for remote debugging.
1136 @item -l @var{timeout}
1138 Set the timeout (in seconds) of any communication used by @value{GDBN}
1139 for remote debugging.
1141 @item -tty @var{device}
1142 @itemx -t @var{device}
1143 @cindex @code{--tty}
1145 Run using @var{device} for your program's standard input and output.
1146 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1148 @c resolve the situation of these eventually
1150 @cindex @code{--tui}
1151 Activate the @dfn{Text User Interface} when starting. The Text User
1152 Interface manages several text windows on the terminal, showing
1153 source, assembly, registers and @value{GDBN} command outputs
1154 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1155 Text User Interface can be enabled by invoking the program
1156 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1157 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1160 @c @cindex @code{--xdb}
1161 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1162 @c For information, see the file @file{xdb_trans.html}, which is usually
1163 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1166 @item -interpreter @var{interp}
1167 @cindex @code{--interpreter}
1168 Use the interpreter @var{interp} for interface with the controlling
1169 program or device. This option is meant to be set by programs which
1170 communicate with @value{GDBN} using it as a back end.
1171 @xref{Interpreters, , Command Interpreters}.
1173 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1174 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1175 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1176 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1177 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1178 @sc{gdb/mi} interfaces are no longer supported.
1181 @cindex @code{--write}
1182 Open the executable and core files for both reading and writing. This
1183 is equivalent to the @samp{set write on} command inside @value{GDBN}
1187 @cindex @code{--statistics}
1188 This option causes @value{GDBN} to print statistics about time and
1189 memory usage after it completes each command and returns to the prompt.
1192 @cindex @code{--version}
1193 This option causes @value{GDBN} to print its version number and
1194 no-warranty blurb, and exit.
1199 @subsection What @value{GDBN} Does During Startup
1200 @cindex @value{GDBN} startup
1202 Here's the description of what @value{GDBN} does during session startup:
1206 Sets up the command interpreter as specified by the command line
1207 (@pxref{Mode Options, interpreter}).
1211 Reads the @dfn{init file} (if any) in your home directory@footnote{On
1212 DOS/Windows systems, the home directory is the one pointed to by the
1213 @code{HOME} environment variable.} and executes all the commands in
1217 Processes command line options and operands.
1220 Reads and executes the commands from init file (if any) in the current
1221 working directory. This is only done if the current directory is
1222 different from your home directory. Thus, you can have more than one
1223 init file, one generic in your home directory, and another, specific
1224 to the program you are debugging, in the directory where you invoke
1228 Reads command files specified by the @samp{-x} option. @xref{Command
1229 Files}, for more details about @value{GDBN} command files.
1232 Reads the command history recorded in the @dfn{history file}.
1233 @xref{Command History}, for more details about the command history and the
1234 files where @value{GDBN} records it.
1237 Init files use the same syntax as @dfn{command files} (@pxref{Command
1238 Files}) and are processed by @value{GDBN} in the same way. The init
1239 file in your home directory can set options (such as @samp{set
1240 complaints}) that affect subsequent processing of command line options
1241 and operands. Init files are not executed if you use the @samp{-nx}
1242 option (@pxref{Mode Options, ,Choosing Modes}).
1244 @cindex init file name
1245 @cindex @file{.gdbinit}
1246 @cindex @file{gdb.ini}
1247 The @value{GDBN} init files are normally called @file{.gdbinit}.
1248 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1249 the limitations of file names imposed by DOS filesystems. The Windows
1250 ports of @value{GDBN} use the standard name, but if they find a
1251 @file{gdb.ini} file, they warn you about that and suggest to rename
1252 the file to the standard name.
1256 @section Quitting @value{GDBN}
1257 @cindex exiting @value{GDBN}
1258 @cindex leaving @value{GDBN}
1261 @kindex quit @r{[}@var{expression}@r{]}
1262 @kindex q @r{(@code{quit})}
1263 @item quit @r{[}@var{expression}@r{]}
1265 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1266 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1267 do not supply @var{expression}, @value{GDBN} will terminate normally;
1268 otherwise it will terminate using the result of @var{expression} as the
1273 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1274 terminates the action of any @value{GDBN} command that is in progress and
1275 returns to @value{GDBN} command level. It is safe to type the interrupt
1276 character at any time because @value{GDBN} does not allow it to take effect
1277 until a time when it is safe.
1279 If you have been using @value{GDBN} to control an attached process or
1280 device, you can release it with the @code{detach} command
1281 (@pxref{Attach, ,Debugging an Already-running Process}).
1283 @node Shell Commands
1284 @section Shell Commands
1286 If you need to execute occasional shell commands during your
1287 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1288 just use the @code{shell} command.
1292 @cindex shell escape
1293 @item shell @var{command string}
1294 Invoke a standard shell to execute @var{command string}.
1295 If it exists, the environment variable @code{SHELL} determines which
1296 shell to run. Otherwise @value{GDBN} uses the default shell
1297 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1300 The utility @code{make} is often needed in development environments.
1301 You do not have to use the @code{shell} command for this purpose in
1306 @cindex calling make
1307 @item make @var{make-args}
1308 Execute the @code{make} program with the specified
1309 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1312 @node Logging Output
1313 @section Logging Output
1314 @cindex logging @value{GDBN} output
1315 @cindex save @value{GDBN} output to a file
1317 You may want to save the output of @value{GDBN} commands to a file.
1318 There are several commands to control @value{GDBN}'s logging.
1322 @item set logging on
1324 @item set logging off
1326 @cindex logging file name
1327 @item set logging file @var{file}
1328 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1329 @item set logging overwrite [on|off]
1330 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1331 you want @code{set logging on} to overwrite the logfile instead.
1332 @item set logging redirect [on|off]
1333 By default, @value{GDBN} output will go to both the terminal and the logfile.
1334 Set @code{redirect} if you want output to go only to the log file.
1335 @kindex show logging
1337 Show the current values of the logging settings.
1341 @chapter @value{GDBN} Commands
1343 You can abbreviate a @value{GDBN} command to the first few letters of the command
1344 name, if that abbreviation is unambiguous; and you can repeat certain
1345 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1346 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1347 show you the alternatives available, if there is more than one possibility).
1350 * Command Syntax:: How to give commands to @value{GDBN}
1351 * Completion:: Command completion
1352 * Help:: How to ask @value{GDBN} for help
1355 @node Command Syntax
1356 @section Command Syntax
1358 A @value{GDBN} command is a single line of input. There is no limit on
1359 how long it can be. It starts with a command name, which is followed by
1360 arguments whose meaning depends on the command name. For example, the
1361 command @code{step} accepts an argument which is the number of times to
1362 step, as in @samp{step 5}. You can also use the @code{step} command
1363 with no arguments. Some commands do not allow any arguments.
1365 @cindex abbreviation
1366 @value{GDBN} command names may always be truncated if that abbreviation is
1367 unambiguous. Other possible command abbreviations are listed in the
1368 documentation for individual commands. In some cases, even ambiguous
1369 abbreviations are allowed; for example, @code{s} is specially defined as
1370 equivalent to @code{step} even though there are other commands whose
1371 names start with @code{s}. You can test abbreviations by using them as
1372 arguments to the @code{help} command.
1374 @cindex repeating commands
1375 @kindex RET @r{(repeat last command)}
1376 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1377 repeat the previous command. Certain commands (for example, @code{run})
1378 will not repeat this way; these are commands whose unintentional
1379 repetition might cause trouble and which you are unlikely to want to
1380 repeat. User-defined commands can disable this feature; see
1381 @ref{Define, dont-repeat}.
1383 The @code{list} and @code{x} commands, when you repeat them with
1384 @key{RET}, construct new arguments rather than repeating
1385 exactly as typed. This permits easy scanning of source or memory.
1387 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1388 output, in a way similar to the common utility @code{more}
1389 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1390 @key{RET} too many in this situation, @value{GDBN} disables command
1391 repetition after any command that generates this sort of display.
1393 @kindex # @r{(a comment)}
1395 Any text from a @kbd{#} to the end of the line is a comment; it does
1396 nothing. This is useful mainly in command files (@pxref{Command
1397 Files,,Command Files}).
1399 @cindex repeating command sequences
1400 @kindex Ctrl-o @r{(operate-and-get-next)}
1401 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1402 commands. This command accepts the current line, like @key{RET}, and
1403 then fetches the next line relative to the current line from the history
1407 @section Command Completion
1410 @cindex word completion
1411 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1412 only one possibility; it can also show you what the valid possibilities
1413 are for the next word in a command, at any time. This works for @value{GDBN}
1414 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1416 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1417 of a word. If there is only one possibility, @value{GDBN} fills in the
1418 word, and waits for you to finish the command (or press @key{RET} to
1419 enter it). For example, if you type
1421 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1422 @c complete accuracy in these examples; space introduced for clarity.
1423 @c If texinfo enhancements make it unnecessary, it would be nice to
1424 @c replace " @key" by "@key" in the following...
1426 (@value{GDBP}) info bre @key{TAB}
1430 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1431 the only @code{info} subcommand beginning with @samp{bre}:
1434 (@value{GDBP}) info breakpoints
1438 You can either press @key{RET} at this point, to run the @code{info
1439 breakpoints} command, or backspace and enter something else, if
1440 @samp{breakpoints} does not look like the command you expected. (If you
1441 were sure you wanted @code{info breakpoints} in the first place, you
1442 might as well just type @key{RET} immediately after @samp{info bre},
1443 to exploit command abbreviations rather than command completion).
1445 If there is more than one possibility for the next word when you press
1446 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1447 characters and try again, or just press @key{TAB} a second time;
1448 @value{GDBN} displays all the possible completions for that word. For
1449 example, you might want to set a breakpoint on a subroutine whose name
1450 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1451 just sounds the bell. Typing @key{TAB} again displays all the
1452 function names in your program that begin with those characters, for
1456 (@value{GDBP}) b make_ @key{TAB}
1457 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1458 make_a_section_from_file make_environ
1459 make_abs_section make_function_type
1460 make_blockvector make_pointer_type
1461 make_cleanup make_reference_type
1462 make_command make_symbol_completion_list
1463 (@value{GDBP}) b make_
1467 After displaying the available possibilities, @value{GDBN} copies your
1468 partial input (@samp{b make_} in the example) so you can finish the
1471 If you just want to see the list of alternatives in the first place, you
1472 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1473 means @kbd{@key{META} ?}. You can type this either by holding down a
1474 key designated as the @key{META} shift on your keyboard (if there is
1475 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1477 @cindex quotes in commands
1478 @cindex completion of quoted strings
1479 Sometimes the string you need, while logically a ``word'', may contain
1480 parentheses or other characters that @value{GDBN} normally excludes from
1481 its notion of a word. To permit word completion to work in this
1482 situation, you may enclose words in @code{'} (single quote marks) in
1483 @value{GDBN} commands.
1485 The most likely situation where you might need this is in typing the
1486 name of a C@t{++} function. This is because C@t{++} allows function
1487 overloading (multiple definitions of the same function, distinguished
1488 by argument type). For example, when you want to set a breakpoint you
1489 may need to distinguish whether you mean the version of @code{name}
1490 that takes an @code{int} parameter, @code{name(int)}, or the version
1491 that takes a @code{float} parameter, @code{name(float)}. To use the
1492 word-completion facilities in this situation, type a single quote
1493 @code{'} at the beginning of the function name. This alerts
1494 @value{GDBN} that it may need to consider more information than usual
1495 when you press @key{TAB} or @kbd{M-?} to request word completion:
1498 (@value{GDBP}) b 'bubble( @kbd{M-?}
1499 bubble(double,double) bubble(int,int)
1500 (@value{GDBP}) b 'bubble(
1503 In some cases, @value{GDBN} can tell that completing a name requires using
1504 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1505 completing as much as it can) if you do not type the quote in the first
1509 (@value{GDBP}) b bub @key{TAB}
1510 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1511 (@value{GDBP}) b 'bubble(
1515 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1516 you have not yet started typing the argument list when you ask for
1517 completion on an overloaded symbol.
1519 For more information about overloaded functions, see @ref{C Plus Plus
1520 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1521 overload-resolution off} to disable overload resolution;
1522 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1526 @section Getting Help
1527 @cindex online documentation
1530 You can always ask @value{GDBN} itself for information on its commands,
1531 using the command @code{help}.
1534 @kindex h @r{(@code{help})}
1537 You can use @code{help} (abbreviated @code{h}) with no arguments to
1538 display a short list of named classes of commands:
1542 List of classes of commands:
1544 aliases -- Aliases of other commands
1545 breakpoints -- Making program stop at certain points
1546 data -- Examining data
1547 files -- Specifying and examining files
1548 internals -- Maintenance commands
1549 obscure -- Obscure features
1550 running -- Running the program
1551 stack -- Examining the stack
1552 status -- Status inquiries
1553 support -- Support facilities
1554 tracepoints -- Tracing of program execution without
1555 stopping the program
1556 user-defined -- User-defined commands
1558 Type "help" followed by a class name for a list of
1559 commands in that class.
1560 Type "help" followed by command name for full
1562 Command name abbreviations are allowed if unambiguous.
1565 @c the above line break eliminates huge line overfull...
1567 @item help @var{class}
1568 Using one of the general help classes as an argument, you can get a
1569 list of the individual commands in that class. For example, here is the
1570 help display for the class @code{status}:
1573 (@value{GDBP}) help status
1578 @c Line break in "show" line falsifies real output, but needed
1579 @c to fit in smallbook page size.
1580 info -- Generic command for showing things
1581 about the program being debugged
1582 show -- Generic command for showing things
1585 Type "help" followed by command name for full
1587 Command name abbreviations are allowed if unambiguous.
1591 @item help @var{command}
1592 With a command name as @code{help} argument, @value{GDBN} displays a
1593 short paragraph on how to use that command.
1596 @item apropos @var{args}
1597 The @code{apropos} command searches through all of the @value{GDBN}
1598 commands, and their documentation, for the regular expression specified in
1599 @var{args}. It prints out all matches found. For example:
1610 set symbol-reloading -- Set dynamic symbol table reloading
1611 multiple times in one run
1612 show symbol-reloading -- Show dynamic symbol table reloading
1613 multiple times in one run
1618 @item complete @var{args}
1619 The @code{complete @var{args}} command lists all the possible completions
1620 for the beginning of a command. Use @var{args} to specify the beginning of the
1621 command you want completed. For example:
1627 @noindent results in:
1638 @noindent This is intended for use by @sc{gnu} Emacs.
1641 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1642 and @code{show} to inquire about the state of your program, or the state
1643 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1644 manual introduces each of them in the appropriate context. The listings
1645 under @code{info} and under @code{show} in the Index point to
1646 all the sub-commands. @xref{Index}.
1651 @kindex i @r{(@code{info})}
1653 This command (abbreviated @code{i}) is for describing the state of your
1654 program. For example, you can show the arguments passed to a function
1655 with @code{info args}, list the registers currently in use with @code{info
1656 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1657 You can get a complete list of the @code{info} sub-commands with
1658 @w{@code{help info}}.
1662 You can assign the result of an expression to an environment variable with
1663 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1664 @code{set prompt $}.
1668 In contrast to @code{info}, @code{show} is for describing the state of
1669 @value{GDBN} itself.
1670 You can change most of the things you can @code{show}, by using the
1671 related command @code{set}; for example, you can control what number
1672 system is used for displays with @code{set radix}, or simply inquire
1673 which is currently in use with @code{show radix}.
1676 To display all the settable parameters and their current
1677 values, you can use @code{show} with no arguments; you may also use
1678 @code{info set}. Both commands produce the same display.
1679 @c FIXME: "info set" violates the rule that "info" is for state of
1680 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1681 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1685 Here are three miscellaneous @code{show} subcommands, all of which are
1686 exceptional in lacking corresponding @code{set} commands:
1689 @kindex show version
1690 @cindex @value{GDBN} version number
1692 Show what version of @value{GDBN} is running. You should include this
1693 information in @value{GDBN} bug-reports. If multiple versions of
1694 @value{GDBN} are in use at your site, you may need to determine which
1695 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1696 commands are introduced, and old ones may wither away. Also, many
1697 system vendors ship variant versions of @value{GDBN}, and there are
1698 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1699 The version number is the same as the one announced when you start
1702 @kindex show copying
1703 @kindex info copying
1704 @cindex display @value{GDBN} copyright
1707 Display information about permission for copying @value{GDBN}.
1709 @kindex show warranty
1710 @kindex info warranty
1712 @itemx info warranty
1713 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1714 if your version of @value{GDBN} comes with one.
1719 @chapter Running Programs Under @value{GDBN}
1721 When you run a program under @value{GDBN}, you must first generate
1722 debugging information when you compile it.
1724 You may start @value{GDBN} with its arguments, if any, in an environment
1725 of your choice. If you are doing native debugging, you may redirect
1726 your program's input and output, debug an already running process, or
1727 kill a child process.
1730 * Compilation:: Compiling for debugging
1731 * Starting:: Starting your program
1732 * Arguments:: Your program's arguments
1733 * Environment:: Your program's environment
1735 * Working Directory:: Your program's working directory
1736 * Input/Output:: Your program's input and output
1737 * Attach:: Debugging an already-running process
1738 * Kill Process:: Killing the child process
1740 * Threads:: Debugging programs with multiple threads
1741 * Processes:: Debugging programs with multiple processes
1742 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1746 @section Compiling for Debugging
1748 In order to debug a program effectively, you need to generate
1749 debugging information when you compile it. This debugging information
1750 is stored in the object file; it describes the data type of each
1751 variable or function and the correspondence between source line numbers
1752 and addresses in the executable code.
1754 To request debugging information, specify the @samp{-g} option when you run
1757 Programs that are to be shipped to your customers are compiled with
1758 optimizations, using the @samp{-O} compiler option. However, many
1759 compilers are unable to handle the @samp{-g} and @samp{-O} options
1760 together. Using those compilers, you cannot generate optimized
1761 executables containing debugging information.
1763 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1764 without @samp{-O}, making it possible to debug optimized code. We
1765 recommend that you @emph{always} use @samp{-g} whenever you compile a
1766 program. You may think your program is correct, but there is no sense
1767 in pushing your luck.
1769 @cindex optimized code, debugging
1770 @cindex debugging optimized code
1771 When you debug a program compiled with @samp{-g -O}, remember that the
1772 optimizer is rearranging your code; the debugger shows you what is
1773 really there. Do not be too surprised when the execution path does not
1774 exactly match your source file! An extreme example: if you define a
1775 variable, but never use it, @value{GDBN} never sees that
1776 variable---because the compiler optimizes it out of existence.
1778 Some things do not work as well with @samp{-g -O} as with just
1779 @samp{-g}, particularly on machines with instruction scheduling. If in
1780 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1781 please report it to us as a bug (including a test case!).
1782 @xref{Variables}, for more information about debugging optimized code.
1784 Older versions of the @sc{gnu} C compiler permitted a variant option
1785 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1786 format; if your @sc{gnu} C compiler has this option, do not use it.
1788 @value{GDBN} knows about preprocessor macros and can show you their
1789 expansion (@pxref{Macros}). Most compilers do not include information
1790 about preprocessor macros in the debugging information if you specify
1791 the @option{-g} flag alone, because this information is rather large.
1792 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1793 provides macro information if you specify the options
1794 @option{-gdwarf-2} and @option{-g3}; the former option requests
1795 debugging information in the Dwarf 2 format, and the latter requests
1796 ``extra information''. In the future, we hope to find more compact
1797 ways to represent macro information, so that it can be included with
1802 @section Starting your Program
1808 @kindex r @r{(@code{run})}
1811 Use the @code{run} command to start your program under @value{GDBN}.
1812 You must first specify the program name (except on VxWorks) with an
1813 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1814 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1815 (@pxref{Files, ,Commands to Specify Files}).
1819 If you are running your program in an execution environment that
1820 supports processes, @code{run} creates an inferior process and makes
1821 that process run your program. In some environments without processes,
1822 @code{run} jumps to the start of your program. Other targets,
1823 like @samp{remote}, are always running. If you get an error
1824 message like this one:
1827 The "remote" target does not support "run".
1828 Try "help target" or "continue".
1832 then use @code{continue} to run your program. You may need @code{load}
1833 first (@pxref{load}).
1835 The execution of a program is affected by certain information it
1836 receives from its superior. @value{GDBN} provides ways to specify this
1837 information, which you must do @emph{before} starting your program. (You
1838 can change it after starting your program, but such changes only affect
1839 your program the next time you start it.) This information may be
1840 divided into four categories:
1843 @item The @emph{arguments.}
1844 Specify the arguments to give your program as the arguments of the
1845 @code{run} command. If a shell is available on your target, the shell
1846 is used to pass the arguments, so that you may use normal conventions
1847 (such as wildcard expansion or variable substitution) in describing
1849 In Unix systems, you can control which shell is used with the
1850 @code{SHELL} environment variable.
1851 @xref{Arguments, ,Your Program's Arguments}.
1853 @item The @emph{environment.}
1854 Your program normally inherits its environment from @value{GDBN}, but you can
1855 use the @value{GDBN} commands @code{set environment} and @code{unset
1856 environment} to change parts of the environment that affect
1857 your program. @xref{Environment, ,Your Program's Environment}.
1859 @item The @emph{working directory.}
1860 Your program inherits its working directory from @value{GDBN}. You can set
1861 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1862 @xref{Working Directory, ,Your Program's Working Directory}.
1864 @item The @emph{standard input and output.}
1865 Your program normally uses the same device for standard input and
1866 standard output as @value{GDBN} is using. You can redirect input and output
1867 in the @code{run} command line, or you can use the @code{tty} command to
1868 set a different device for your program.
1869 @xref{Input/Output, ,Your Program's Input and Output}.
1872 @emph{Warning:} While input and output redirection work, you cannot use
1873 pipes to pass the output of the program you are debugging to another
1874 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1878 When you issue the @code{run} command, your program begins to execute
1879 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1880 of how to arrange for your program to stop. Once your program has
1881 stopped, you may call functions in your program, using the @code{print}
1882 or @code{call} commands. @xref{Data, ,Examining Data}.
1884 If the modification time of your symbol file has changed since the last
1885 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1886 table, and reads it again. When it does this, @value{GDBN} tries to retain
1887 your current breakpoints.
1892 @cindex run to main procedure
1893 The name of the main procedure can vary from language to language.
1894 With C or C@t{++}, the main procedure name is always @code{main}, but
1895 other languages such as Ada do not require a specific name for their
1896 main procedure. The debugger provides a convenient way to start the
1897 execution of the program and to stop at the beginning of the main
1898 procedure, depending on the language used.
1900 The @samp{start} command does the equivalent of setting a temporary
1901 breakpoint at the beginning of the main procedure and then invoking
1902 the @samp{run} command.
1904 @cindex elaboration phase
1905 Some programs contain an @dfn{elaboration} phase where some startup code is
1906 executed before the main procedure is called. This depends on the
1907 languages used to write your program. In C@t{++}, for instance,
1908 constructors for static and global objects are executed before
1909 @code{main} is called. It is therefore possible that the debugger stops
1910 before reaching the main procedure. However, the temporary breakpoint
1911 will remain to halt execution.
1913 Specify the arguments to give to your program as arguments to the
1914 @samp{start} command. These arguments will be given verbatim to the
1915 underlying @samp{run} command. Note that the same arguments will be
1916 reused if no argument is provided during subsequent calls to
1917 @samp{start} or @samp{run}.
1919 It is sometimes necessary to debug the program during elaboration. In
1920 these cases, using the @code{start} command would stop the execution of
1921 your program too late, as the program would have already completed the
1922 elaboration phase. Under these circumstances, insert breakpoints in your
1923 elaboration code before running your program.
1925 @kindex set exec-wrapper
1926 @item set exec-wrapper @var{wrapper}
1927 @itemx show exec-wrapper
1928 @itemx unset exec-wrapper
1929 When @samp{exec-wrapper} is set, the specified wrapper is used to
1930 launch programs for debugging. @value{GDBN} starts your program
1931 with a shell command of the form @kbd{exec @var{wrapper}
1932 @var{program}}. Quoting is added to @var{program} and its
1933 arguments, but not to @var{wrapper}, so you should add quotes if
1934 appropriate for your shell. The wrapper runs until it executes
1935 your program, and then @value{GDBN} takes control.
1937 You can use any program that eventually calls @code{execve} with
1938 its arguments as a wrapper. Several standard Unix utilities do
1939 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
1940 with @code{exec "$@@"} will also work.
1942 For example, you can use @code{env} to pass an environment variable to
1943 the debugged program, without setting the variable in your shell's
1947 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
1951 This command is available when debugging locally on most targets, excluding
1952 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
1957 @section Your Program's Arguments
1959 @cindex arguments (to your program)
1960 The arguments to your program can be specified by the arguments of the
1962 They are passed to a shell, which expands wildcard characters and
1963 performs redirection of I/O, and thence to your program. Your
1964 @code{SHELL} environment variable (if it exists) specifies what shell
1965 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
1966 the default shell (@file{/bin/sh} on Unix).
1968 On non-Unix systems, the program is usually invoked directly by
1969 @value{GDBN}, which emulates I/O redirection via the appropriate system
1970 calls, and the wildcard characters are expanded by the startup code of
1971 the program, not by the shell.
1973 @code{run} with no arguments uses the same arguments used by the previous
1974 @code{run}, or those set by the @code{set args} command.
1979 Specify the arguments to be used the next time your program is run. If
1980 @code{set args} has no arguments, @code{run} executes your program
1981 with no arguments. Once you have run your program with arguments,
1982 using @code{set args} before the next @code{run} is the only way to run
1983 it again without arguments.
1987 Show the arguments to give your program when it is started.
1991 @section Your Program's Environment
1993 @cindex environment (of your program)
1994 The @dfn{environment} consists of a set of environment variables and
1995 their values. Environment variables conventionally record such things as
1996 your user name, your home directory, your terminal type, and your search
1997 path for programs to run. Usually you set up environment variables with
1998 the shell and they are inherited by all the other programs you run. When
1999 debugging, it can be useful to try running your program with a modified
2000 environment without having to start @value{GDBN} over again.
2004 @item path @var{directory}
2005 Add @var{directory} to the front of the @code{PATH} environment variable
2006 (the search path for executables) that will be passed to your program.
2007 The value of @code{PATH} used by @value{GDBN} does not change.
2008 You may specify several directory names, separated by whitespace or by a
2009 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2010 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2011 is moved to the front, so it is searched sooner.
2013 You can use the string @samp{$cwd} to refer to whatever is the current
2014 working directory at the time @value{GDBN} searches the path. If you
2015 use @samp{.} instead, it refers to the directory where you executed the
2016 @code{path} command. @value{GDBN} replaces @samp{.} in the
2017 @var{directory} argument (with the current path) before adding
2018 @var{directory} to the search path.
2019 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2020 @c document that, since repeating it would be a no-op.
2024 Display the list of search paths for executables (the @code{PATH}
2025 environment variable).
2027 @kindex show environment
2028 @item show environment @r{[}@var{varname}@r{]}
2029 Print the value of environment variable @var{varname} to be given to
2030 your program when it starts. If you do not supply @var{varname},
2031 print the names and values of all environment variables to be given to
2032 your program. You can abbreviate @code{environment} as @code{env}.
2034 @kindex set environment
2035 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2036 Set environment variable @var{varname} to @var{value}. The value
2037 changes for your program only, not for @value{GDBN} itself. @var{value} may
2038 be any string; the values of environment variables are just strings, and
2039 any interpretation is supplied by your program itself. The @var{value}
2040 parameter is optional; if it is eliminated, the variable is set to a
2042 @c "any string" here does not include leading, trailing
2043 @c blanks. Gnu asks: does anyone care?
2045 For example, this command:
2052 tells the debugged program, when subsequently run, that its user is named
2053 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2054 are not actually required.)
2056 @kindex unset environment
2057 @item unset environment @var{varname}
2058 Remove variable @var{varname} from the environment to be passed to your
2059 program. This is different from @samp{set env @var{varname} =};
2060 @code{unset environment} removes the variable from the environment,
2061 rather than assigning it an empty value.
2064 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2066 by your @code{SHELL} environment variable if it exists (or
2067 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2068 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2069 @file{.bashrc} for BASH---any variables you set in that file affect
2070 your program. You may wish to move setting of environment variables to
2071 files that are only run when you sign on, such as @file{.login} or
2074 @node Working Directory
2075 @section Your Program's Working Directory
2077 @cindex working directory (of your program)
2078 Each time you start your program with @code{run}, it inherits its
2079 working directory from the current working directory of @value{GDBN}.
2080 The @value{GDBN} working directory is initially whatever it inherited
2081 from its parent process (typically the shell), but you can specify a new
2082 working directory in @value{GDBN} with the @code{cd} command.
2084 The @value{GDBN} working directory also serves as a default for the commands
2085 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2090 @cindex change working directory
2091 @item cd @var{directory}
2092 Set the @value{GDBN} working directory to @var{directory}.
2096 Print the @value{GDBN} working directory.
2099 It is generally impossible to find the current working directory of
2100 the process being debugged (since a program can change its directory
2101 during its run). If you work on a system where @value{GDBN} is
2102 configured with the @file{/proc} support, you can use the @code{info
2103 proc} command (@pxref{SVR4 Process Information}) to find out the
2104 current working directory of the debuggee.
2107 @section Your Program's Input and Output
2112 By default, the program you run under @value{GDBN} does input and output to
2113 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2114 to its own terminal modes to interact with you, but it records the terminal
2115 modes your program was using and switches back to them when you continue
2116 running your program.
2119 @kindex info terminal
2121 Displays information recorded by @value{GDBN} about the terminal modes your
2125 You can redirect your program's input and/or output using shell
2126 redirection with the @code{run} command. For example,
2133 starts your program, diverting its output to the file @file{outfile}.
2136 @cindex controlling terminal
2137 Another way to specify where your program should do input and output is
2138 with the @code{tty} command. This command accepts a file name as
2139 argument, and causes this file to be the default for future @code{run}
2140 commands. It also resets the controlling terminal for the child
2141 process, for future @code{run} commands. For example,
2148 directs that processes started with subsequent @code{run} commands
2149 default to do input and output on the terminal @file{/dev/ttyb} and have
2150 that as their controlling terminal.
2152 An explicit redirection in @code{run} overrides the @code{tty} command's
2153 effect on the input/output device, but not its effect on the controlling
2156 When you use the @code{tty} command or redirect input in the @code{run}
2157 command, only the input @emph{for your program} is affected. The input
2158 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2159 for @code{set inferior-tty}.
2161 @cindex inferior tty
2162 @cindex set inferior controlling terminal
2163 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2164 display the name of the terminal that will be used for future runs of your
2168 @item set inferior-tty /dev/ttyb
2169 @kindex set inferior-tty
2170 Set the tty for the program being debugged to /dev/ttyb.
2172 @item show inferior-tty
2173 @kindex show inferior-tty
2174 Show the current tty for the program being debugged.
2178 @section Debugging an Already-running Process
2183 @item attach @var{process-id}
2184 This command attaches to a running process---one that was started
2185 outside @value{GDBN}. (@code{info files} shows your active
2186 targets.) The command takes as argument a process ID. The usual way to
2187 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2188 or with the @samp{jobs -l} shell command.
2190 @code{attach} does not repeat if you press @key{RET} a second time after
2191 executing the command.
2194 To use @code{attach}, your program must be running in an environment
2195 which supports processes; for example, @code{attach} does not work for
2196 programs on bare-board targets that lack an operating system. You must
2197 also have permission to send the process a signal.
2199 When you use @code{attach}, the debugger finds the program running in
2200 the process first by looking in the current working directory, then (if
2201 the program is not found) by using the source file search path
2202 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2203 the @code{file} command to load the program. @xref{Files, ,Commands to
2206 The first thing @value{GDBN} does after arranging to debug the specified
2207 process is to stop it. You can examine and modify an attached process
2208 with all the @value{GDBN} commands that are ordinarily available when
2209 you start processes with @code{run}. You can insert breakpoints; you
2210 can step and continue; you can modify storage. If you would rather the
2211 process continue running, you may use the @code{continue} command after
2212 attaching @value{GDBN} to the process.
2217 When you have finished debugging the attached process, you can use the
2218 @code{detach} command to release it from @value{GDBN} control. Detaching
2219 the process continues its execution. After the @code{detach} command,
2220 that process and @value{GDBN} become completely independent once more, and you
2221 are ready to @code{attach} another process or start one with @code{run}.
2222 @code{detach} does not repeat if you press @key{RET} again after
2223 executing the command.
2226 If you exit @value{GDBN} while you have an attached process, you detach
2227 that process. If you use the @code{run} command, you kill that process.
2228 By default, @value{GDBN} asks for confirmation if you try to do either of these
2229 things; you can control whether or not you need to confirm by using the
2230 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2234 @section Killing the Child Process
2239 Kill the child process in which your program is running under @value{GDBN}.
2242 This command is useful if you wish to debug a core dump instead of a
2243 running process. @value{GDBN} ignores any core dump file while your program
2246 On some operating systems, a program cannot be executed outside @value{GDBN}
2247 while you have breakpoints set on it inside @value{GDBN}. You can use the
2248 @code{kill} command in this situation to permit running your program
2249 outside the debugger.
2251 The @code{kill} command is also useful if you wish to recompile and
2252 relink your program, since on many systems it is impossible to modify an
2253 executable file while it is running in a process. In this case, when you
2254 next type @code{run}, @value{GDBN} notices that the file has changed, and
2255 reads the symbol table again (while trying to preserve your current
2256 breakpoint settings).
2259 @section Debugging Programs with Multiple Threads
2261 @cindex threads of execution
2262 @cindex multiple threads
2263 @cindex switching threads
2264 In some operating systems, such as HP-UX and Solaris, a single program
2265 may have more than one @dfn{thread} of execution. The precise semantics
2266 of threads differ from one operating system to another, but in general
2267 the threads of a single program are akin to multiple processes---except
2268 that they share one address space (that is, they can all examine and
2269 modify the same variables). On the other hand, each thread has its own
2270 registers and execution stack, and perhaps private memory.
2272 @value{GDBN} provides these facilities for debugging multi-thread
2276 @item automatic notification of new threads
2277 @item @samp{thread @var{threadno}}, a command to switch among threads
2278 @item @samp{info threads}, a command to inquire about existing threads
2279 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2280 a command to apply a command to a list of threads
2281 @item thread-specific breakpoints
2282 @item @samp{set print thread-events}, which controls printing of
2283 messages on thread start and exit.
2287 @emph{Warning:} These facilities are not yet available on every
2288 @value{GDBN} configuration where the operating system supports threads.
2289 If your @value{GDBN} does not support threads, these commands have no
2290 effect. For example, a system without thread support shows no output
2291 from @samp{info threads}, and always rejects the @code{thread} command,
2295 (@value{GDBP}) info threads
2296 (@value{GDBP}) thread 1
2297 Thread ID 1 not known. Use the "info threads" command to
2298 see the IDs of currently known threads.
2300 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2301 @c doesn't support threads"?
2304 @cindex focus of debugging
2305 @cindex current thread
2306 The @value{GDBN} thread debugging facility allows you to observe all
2307 threads while your program runs---but whenever @value{GDBN} takes
2308 control, one thread in particular is always the focus of debugging.
2309 This thread is called the @dfn{current thread}. Debugging commands show
2310 program information from the perspective of the current thread.
2312 @cindex @code{New} @var{systag} message
2313 @cindex thread identifier (system)
2314 @c FIXME-implementors!! It would be more helpful if the [New...] message
2315 @c included GDB's numeric thread handle, so you could just go to that
2316 @c thread without first checking `info threads'.
2317 Whenever @value{GDBN} detects a new thread in your program, it displays
2318 the target system's identification for the thread with a message in the
2319 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2320 whose form varies depending on the particular system. For example, on
2321 @sc{gnu}/Linux, you might see
2324 [New Thread 46912507313328 (LWP 25582)]
2328 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2329 the @var{systag} is simply something like @samp{process 368}, with no
2332 @c FIXME!! (1) Does the [New...] message appear even for the very first
2333 @c thread of a program, or does it only appear for the
2334 @c second---i.e.@: when it becomes obvious we have a multithread
2336 @c (2) *Is* there necessarily a first thread always? Or do some
2337 @c multithread systems permit starting a program with multiple
2338 @c threads ab initio?
2340 @cindex thread number
2341 @cindex thread identifier (GDB)
2342 For debugging purposes, @value{GDBN} associates its own thread
2343 number---always a single integer---with each thread in your program.
2346 @kindex info threads
2348 Display a summary of all threads currently in your
2349 program. @value{GDBN} displays for each thread (in this order):
2353 the thread number assigned by @value{GDBN}
2356 the target system's thread identifier (@var{systag})
2359 the current stack frame summary for that thread
2363 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2364 indicates the current thread.
2368 @c end table here to get a little more width for example
2371 (@value{GDBP}) info threads
2372 3 process 35 thread 27 0x34e5 in sigpause ()
2373 2 process 35 thread 23 0x34e5 in sigpause ()
2374 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2380 @cindex debugging multithreaded programs (on HP-UX)
2381 @cindex thread identifier (GDB), on HP-UX
2382 For debugging purposes, @value{GDBN} associates its own thread
2383 number---a small integer assigned in thread-creation order---with each
2384 thread in your program.
2386 @cindex @code{New} @var{systag} message, on HP-UX
2387 @cindex thread identifier (system), on HP-UX
2388 @c FIXME-implementors!! It would be more helpful if the [New...] message
2389 @c included GDB's numeric thread handle, so you could just go to that
2390 @c thread without first checking `info threads'.
2391 Whenever @value{GDBN} detects a new thread in your program, it displays
2392 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2393 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2394 whose form varies depending on the particular system. For example, on
2398 [New thread 2 (system thread 26594)]
2402 when @value{GDBN} notices a new thread.
2405 @kindex info threads (HP-UX)
2407 Display a summary of all threads currently in your
2408 program. @value{GDBN} displays for each thread (in this order):
2411 @item the thread number assigned by @value{GDBN}
2413 @item the target system's thread identifier (@var{systag})
2415 @item the current stack frame summary for that thread
2419 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2420 indicates the current thread.
2424 @c end table here to get a little more width for example
2427 (@value{GDBP}) info threads
2428 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2430 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2431 from /usr/lib/libc.2
2432 1 system thread 27905 0x7b003498 in _brk () \@*
2433 from /usr/lib/libc.2
2436 On Solaris, you can display more information about user threads with a
2437 Solaris-specific command:
2440 @item maint info sol-threads
2441 @kindex maint info sol-threads
2442 @cindex thread info (Solaris)
2443 Display info on Solaris user threads.
2447 @kindex thread @var{threadno}
2448 @item thread @var{threadno}
2449 Make thread number @var{threadno} the current thread. The command
2450 argument @var{threadno} is the internal @value{GDBN} thread number, as
2451 shown in the first field of the @samp{info threads} display.
2452 @value{GDBN} responds by displaying the system identifier of the thread
2453 you selected, and its current stack frame summary:
2456 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2457 (@value{GDBP}) thread 2
2458 [Switching to process 35 thread 23]
2459 0x34e5 in sigpause ()
2463 As with the @samp{[New @dots{}]} message, the form of the text after
2464 @samp{Switching to} depends on your system's conventions for identifying
2467 @kindex thread apply
2468 @cindex apply command to several threads
2469 @item thread apply [@var{threadno}] [@var{all}] @var{command}
2470 The @code{thread apply} command allows you to apply the named
2471 @var{command} to one or more threads. Specify the numbers of the
2472 threads that you want affected with the command argument
2473 @var{threadno}. It can be a single thread number, one of the numbers
2474 shown in the first field of the @samp{info threads} display; or it
2475 could be a range of thread numbers, as in @code{2-4}. To apply a
2476 command to all threads, type @kbd{thread apply all @var{command}}.
2478 @kindex set print thread-events
2479 @cindex print messages on thread start and exit
2480 @item set print thread-events
2481 @itemx set print thread-events on
2482 @itemx set print thread-events off
2483 The @code{set print thread-events} command allows you to enable or
2484 disable printing of messages when @value{GDBN} notices that new threads have
2485 started or that threads have exited. By default, these messages will
2486 be printed if detection of these events is supported by the target.
2487 Note that these messages cannot be disabled on all targets.
2489 @kindex show print thread-events
2490 @item show print thread-events
2491 Show whether messages will be printed when @value{GDBN} detects that threads
2492 have started and exited.
2495 @cindex automatic thread selection
2496 @cindex switching threads automatically
2497 @cindex threads, automatic switching
2498 Whenever @value{GDBN} stops your program, due to a breakpoint or a
2499 signal, it automatically selects the thread where that breakpoint or
2500 signal happened. @value{GDBN} alerts you to the context switch with a
2501 message of the form @samp{[Switching to @var{systag}]} to identify the
2504 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2505 more information about how @value{GDBN} behaves when you stop and start
2506 programs with multiple threads.
2508 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2509 watchpoints in programs with multiple threads.
2512 @section Debugging Programs with Multiple Processes
2514 @cindex fork, debugging programs which call
2515 @cindex multiple processes
2516 @cindex processes, multiple
2517 On most systems, @value{GDBN} has no special support for debugging
2518 programs which create additional processes using the @code{fork}
2519 function. When a program forks, @value{GDBN} will continue to debug the
2520 parent process and the child process will run unimpeded. If you have
2521 set a breakpoint in any code which the child then executes, the child
2522 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2523 will cause it to terminate.
2525 However, if you want to debug the child process there is a workaround
2526 which isn't too painful. Put a call to @code{sleep} in the code which
2527 the child process executes after the fork. It may be useful to sleep
2528 only if a certain environment variable is set, or a certain file exists,
2529 so that the delay need not occur when you don't want to run @value{GDBN}
2530 on the child. While the child is sleeping, use the @code{ps} program to
2531 get its process ID. Then tell @value{GDBN} (a new invocation of
2532 @value{GDBN} if you are also debugging the parent process) to attach to
2533 the child process (@pxref{Attach}). From that point on you can debug
2534 the child process just like any other process which you attached to.
2536 On some systems, @value{GDBN} provides support for debugging programs that
2537 create additional processes using the @code{fork} or @code{vfork} functions.
2538 Currently, the only platforms with this feature are HP-UX (11.x and later
2539 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2541 By default, when a program forks, @value{GDBN} will continue to debug
2542 the parent process and the child process will run unimpeded.
2544 If you want to follow the child process instead of the parent process,
2545 use the command @w{@code{set follow-fork-mode}}.
2548 @kindex set follow-fork-mode
2549 @item set follow-fork-mode @var{mode}
2550 Set the debugger response to a program call of @code{fork} or
2551 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2552 process. The @var{mode} argument can be:
2556 The original process is debugged after a fork. The child process runs
2557 unimpeded. This is the default.
2560 The new process is debugged after a fork. The parent process runs
2565 @kindex show follow-fork-mode
2566 @item show follow-fork-mode
2567 Display the current debugger response to a @code{fork} or @code{vfork} call.
2570 @cindex debugging multiple processes
2571 On Linux, if you want to debug both the parent and child processes, use the
2572 command @w{@code{set detach-on-fork}}.
2575 @kindex set detach-on-fork
2576 @item set detach-on-fork @var{mode}
2577 Tells gdb whether to detach one of the processes after a fork, or
2578 retain debugger control over them both.
2582 The child process (or parent process, depending on the value of
2583 @code{follow-fork-mode}) will be detached and allowed to run
2584 independently. This is the default.
2587 Both processes will be held under the control of @value{GDBN}.
2588 One process (child or parent, depending on the value of
2589 @code{follow-fork-mode}) is debugged as usual, while the other
2594 @kindex show detach-on-fork
2595 @item show detach-on-fork
2596 Show whether detach-on-fork mode is on/off.
2599 If you choose to set @samp{detach-on-fork} mode off, then
2600 @value{GDBN} will retain control of all forked processes (including
2601 nested forks). You can list the forked processes under the control of
2602 @value{GDBN} by using the @w{@code{info forks}} command, and switch
2603 from one fork to another by using the @w{@code{fork}} command.
2608 Print a list of all forked processes under the control of @value{GDBN}.
2609 The listing will include a fork id, a process id, and the current
2610 position (program counter) of the process.
2612 @kindex fork @var{fork-id}
2613 @item fork @var{fork-id}
2614 Make fork number @var{fork-id} the current process. The argument
2615 @var{fork-id} is the internal fork number assigned by @value{GDBN},
2616 as shown in the first field of the @samp{info forks} display.
2618 @kindex process @var{process-id}
2619 @item process @var{process-id}
2620 Make process number @var{process-id} the current process. The
2621 argument @var{process-id} must be one that is listed in the output of
2626 To quit debugging one of the forked processes, you can either detach
2627 from it by using the @w{@code{detach fork}} command (allowing it to
2628 run independently), or delete (and kill) it using the
2629 @w{@code{delete fork}} command.
2632 @kindex detach fork @var{fork-id}
2633 @item detach fork @var{fork-id}
2634 Detach from the process identified by @value{GDBN} fork number
2635 @var{fork-id}, and remove it from the fork list. The process will be
2636 allowed to run independently.
2638 @kindex delete fork @var{fork-id}
2639 @item delete fork @var{fork-id}
2640 Kill the process identified by @value{GDBN} fork number @var{fork-id},
2641 and remove it from the fork list.
2645 If you ask to debug a child process and a @code{vfork} is followed by an
2646 @code{exec}, @value{GDBN} executes the new target up to the first
2647 breakpoint in the new target. If you have a breakpoint set on
2648 @code{main} in your original program, the breakpoint will also be set on
2649 the child process's @code{main}.
2651 When a child process is spawned by @code{vfork}, you cannot debug the
2652 child or parent until an @code{exec} call completes.
2654 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2655 call executes, the new target restarts. To restart the parent process,
2656 use the @code{file} command with the parent executable name as its
2659 You can use the @code{catch} command to make @value{GDBN} stop whenever
2660 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2661 Catchpoints, ,Setting Catchpoints}.
2663 @node Checkpoint/Restart
2664 @section Setting a @emph{Bookmark} to Return to Later
2669 @cindex snapshot of a process
2670 @cindex rewind program state
2672 On certain operating systems@footnote{Currently, only
2673 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
2674 program's state, called a @dfn{checkpoint}, and come back to it
2677 Returning to a checkpoint effectively undoes everything that has
2678 happened in the program since the @code{checkpoint} was saved. This
2679 includes changes in memory, registers, and even (within some limits)
2680 system state. Effectively, it is like going back in time to the
2681 moment when the checkpoint was saved.
2683 Thus, if you're stepping thru a program and you think you're
2684 getting close to the point where things go wrong, you can save
2685 a checkpoint. Then, if you accidentally go too far and miss
2686 the critical statement, instead of having to restart your program
2687 from the beginning, you can just go back to the checkpoint and
2688 start again from there.
2690 This can be especially useful if it takes a lot of time or
2691 steps to reach the point where you think the bug occurs.
2693 To use the @code{checkpoint}/@code{restart} method of debugging:
2698 Save a snapshot of the debugged program's current execution state.
2699 The @code{checkpoint} command takes no arguments, but each checkpoint
2700 is assigned a small integer id, similar to a breakpoint id.
2702 @kindex info checkpoints
2703 @item info checkpoints
2704 List the checkpoints that have been saved in the current debugging
2705 session. For each checkpoint, the following information will be
2712 @item Source line, or label
2715 @kindex restart @var{checkpoint-id}
2716 @item restart @var{checkpoint-id}
2717 Restore the program state that was saved as checkpoint number
2718 @var{checkpoint-id}. All program variables, registers, stack frames
2719 etc.@: will be returned to the values that they had when the checkpoint
2720 was saved. In essence, gdb will ``wind back the clock'' to the point
2721 in time when the checkpoint was saved.
2723 Note that breakpoints, @value{GDBN} variables, command history etc.
2724 are not affected by restoring a checkpoint. In general, a checkpoint
2725 only restores things that reside in the program being debugged, not in
2728 @kindex delete checkpoint @var{checkpoint-id}
2729 @item delete checkpoint @var{checkpoint-id}
2730 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
2734 Returning to a previously saved checkpoint will restore the user state
2735 of the program being debugged, plus a significant subset of the system
2736 (OS) state, including file pointers. It won't ``un-write'' data from
2737 a file, but it will rewind the file pointer to the previous location,
2738 so that the previously written data can be overwritten. For files
2739 opened in read mode, the pointer will also be restored so that the
2740 previously read data can be read again.
2742 Of course, characters that have been sent to a printer (or other
2743 external device) cannot be ``snatched back'', and characters received
2744 from eg.@: a serial device can be removed from internal program buffers,
2745 but they cannot be ``pushed back'' into the serial pipeline, ready to
2746 be received again. Similarly, the actual contents of files that have
2747 been changed cannot be restored (at this time).
2749 However, within those constraints, you actually can ``rewind'' your
2750 program to a previously saved point in time, and begin debugging it
2751 again --- and you can change the course of events so as to debug a
2752 different execution path this time.
2754 @cindex checkpoints and process id
2755 Finally, there is one bit of internal program state that will be
2756 different when you return to a checkpoint --- the program's process
2757 id. Each checkpoint will have a unique process id (or @var{pid}),
2758 and each will be different from the program's original @var{pid}.
2759 If your program has saved a local copy of its process id, this could
2760 potentially pose a problem.
2762 @subsection A Non-obvious Benefit of Using Checkpoints
2764 On some systems such as @sc{gnu}/Linux, address space randomization
2765 is performed on new processes for security reasons. This makes it
2766 difficult or impossible to set a breakpoint, or watchpoint, on an
2767 absolute address if you have to restart the program, since the
2768 absolute location of a symbol will change from one execution to the
2771 A checkpoint, however, is an @emph{identical} copy of a process.
2772 Therefore if you create a checkpoint at (eg.@:) the start of main,
2773 and simply return to that checkpoint instead of restarting the
2774 process, you can avoid the effects of address randomization and
2775 your symbols will all stay in the same place.
2778 @chapter Stopping and Continuing
2780 The principal purposes of using a debugger are so that you can stop your
2781 program before it terminates; or so that, if your program runs into
2782 trouble, you can investigate and find out why.
2784 Inside @value{GDBN}, your program may stop for any of several reasons,
2785 such as a signal, a breakpoint, or reaching a new line after a
2786 @value{GDBN} command such as @code{step}. You may then examine and
2787 change variables, set new breakpoints or remove old ones, and then
2788 continue execution. Usually, the messages shown by @value{GDBN} provide
2789 ample explanation of the status of your program---but you can also
2790 explicitly request this information at any time.
2793 @kindex info program
2795 Display information about the status of your program: whether it is
2796 running or not, what process it is, and why it stopped.
2800 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2801 * Continuing and Stepping:: Resuming execution
2803 * Thread Stops:: Stopping and starting multi-thread programs
2807 @section Breakpoints, Watchpoints, and Catchpoints
2810 A @dfn{breakpoint} makes your program stop whenever a certain point in
2811 the program is reached. For each breakpoint, you can add conditions to
2812 control in finer detail whether your program stops. You can set
2813 breakpoints with the @code{break} command and its variants (@pxref{Set
2814 Breaks, ,Setting Breakpoints}), to specify the place where your program
2815 should stop by line number, function name or exact address in the
2818 On some systems, you can set breakpoints in shared libraries before
2819 the executable is run. There is a minor limitation on HP-UX systems:
2820 you must wait until the executable is run in order to set breakpoints
2821 in shared library routines that are not called directly by the program
2822 (for example, routines that are arguments in a @code{pthread_create}
2826 @cindex data breakpoints
2827 @cindex memory tracing
2828 @cindex breakpoint on memory address
2829 @cindex breakpoint on variable modification
2830 A @dfn{watchpoint} is a special breakpoint that stops your program
2831 when the value of an expression changes. The expression may be a value
2832 of a variable, or it could involve values of one or more variables
2833 combined by operators, such as @samp{a + b}. This is sometimes called
2834 @dfn{data breakpoints}. You must use a different command to set
2835 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
2836 from that, you can manage a watchpoint like any other breakpoint: you
2837 enable, disable, and delete both breakpoints and watchpoints using the
2840 You can arrange to have values from your program displayed automatically
2841 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2845 @cindex breakpoint on events
2846 A @dfn{catchpoint} is another special breakpoint that stops your program
2847 when a certain kind of event occurs, such as the throwing of a C@t{++}
2848 exception or the loading of a library. As with watchpoints, you use a
2849 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
2850 Catchpoints}), but aside from that, you can manage a catchpoint like any
2851 other breakpoint. (To stop when your program receives a signal, use the
2852 @code{handle} command; see @ref{Signals, ,Signals}.)
2854 @cindex breakpoint numbers
2855 @cindex numbers for breakpoints
2856 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
2857 catchpoint when you create it; these numbers are successive integers
2858 starting with one. In many of the commands for controlling various
2859 features of breakpoints you use the breakpoint number to say which
2860 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
2861 @dfn{disabled}; if disabled, it has no effect on your program until you
2864 @cindex breakpoint ranges
2865 @cindex ranges of breakpoints
2866 Some @value{GDBN} commands accept a range of breakpoints on which to
2867 operate. A breakpoint range is either a single breakpoint number, like
2868 @samp{5}, or two such numbers, in increasing order, separated by a
2869 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
2870 all breakpoints in that range are operated on.
2873 * Set Breaks:: Setting breakpoints
2874 * Set Watchpoints:: Setting watchpoints
2875 * Set Catchpoints:: Setting catchpoints
2876 * Delete Breaks:: Deleting breakpoints
2877 * Disabling:: Disabling breakpoints
2878 * Conditions:: Break conditions
2879 * Break Commands:: Breakpoint command lists
2880 * Error in Breakpoints:: ``Cannot insert breakpoints''
2881 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
2885 @subsection Setting Breakpoints
2887 @c FIXME LMB what does GDB do if no code on line of breakpt?
2888 @c consider in particular declaration with/without initialization.
2890 @c FIXME 2 is there stuff on this already? break at fun start, already init?
2893 @kindex b @r{(@code{break})}
2894 @vindex $bpnum@r{, convenience variable}
2895 @cindex latest breakpoint
2896 Breakpoints are set with the @code{break} command (abbreviated
2897 @code{b}). The debugger convenience variable @samp{$bpnum} records the
2898 number of the breakpoint you've set most recently; see @ref{Convenience
2899 Vars,, Convenience Variables}, for a discussion of what you can do with
2900 convenience variables.
2903 @item break @var{location}
2904 Set a breakpoint at the given @var{location}, which can specify a
2905 function name, a line number, or an address of an instruction.
2906 (@xref{Specify Location}, for a list of all the possible ways to
2907 specify a @var{location}.) The breakpoint will stop your program just
2908 before it executes any of the code in the specified @var{location}.
2910 When using source languages that permit overloading of symbols, such as
2911 C@t{++}, a function name may refer to more than one possible place to break.
2912 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
2916 When called without any arguments, @code{break} sets a breakpoint at
2917 the next instruction to be executed in the selected stack frame
2918 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
2919 innermost, this makes your program stop as soon as control
2920 returns to that frame. This is similar to the effect of a
2921 @code{finish} command in the frame inside the selected frame---except
2922 that @code{finish} does not leave an active breakpoint. If you use
2923 @code{break} without an argument in the innermost frame, @value{GDBN} stops
2924 the next time it reaches the current location; this may be useful
2927 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
2928 least one instruction has been executed. If it did not do this, you
2929 would be unable to proceed past a breakpoint without first disabling the
2930 breakpoint. This rule applies whether or not the breakpoint already
2931 existed when your program stopped.
2933 @item break @dots{} if @var{cond}
2934 Set a breakpoint with condition @var{cond}; evaluate the expression
2935 @var{cond} each time the breakpoint is reached, and stop only if the
2936 value is nonzero---that is, if @var{cond} evaluates as true.
2937 @samp{@dots{}} stands for one of the possible arguments described
2938 above (or no argument) specifying where to break. @xref{Conditions,
2939 ,Break Conditions}, for more information on breakpoint conditions.
2942 @item tbreak @var{args}
2943 Set a breakpoint enabled only for one stop. @var{args} are the
2944 same as for the @code{break} command, and the breakpoint is set in the same
2945 way, but the breakpoint is automatically deleted after the first time your
2946 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
2949 @cindex hardware breakpoints
2950 @item hbreak @var{args}
2951 Set a hardware-assisted breakpoint. @var{args} are the same as for the
2952 @code{break} command and the breakpoint is set in the same way, but the
2953 breakpoint requires hardware support and some target hardware may not
2954 have this support. The main purpose of this is EPROM/ROM code
2955 debugging, so you can set a breakpoint at an instruction without
2956 changing the instruction. This can be used with the new trap-generation
2957 provided by SPARClite DSU and most x86-based targets. These targets
2958 will generate traps when a program accesses some data or instruction
2959 address that is assigned to the debug registers. However the hardware
2960 breakpoint registers can take a limited number of breakpoints. For
2961 example, on the DSU, only two data breakpoints can be set at a time, and
2962 @value{GDBN} will reject this command if more than two are used. Delete
2963 or disable unused hardware breakpoints before setting new ones
2964 (@pxref{Disabling, ,Disabling Breakpoints}).
2965 @xref{Conditions, ,Break Conditions}.
2966 For remote targets, you can restrict the number of hardware
2967 breakpoints @value{GDBN} will use, see @ref{set remote
2968 hardware-breakpoint-limit}.
2971 @item thbreak @var{args}
2972 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
2973 are the same as for the @code{hbreak} command and the breakpoint is set in
2974 the same way. However, like the @code{tbreak} command,
2975 the breakpoint is automatically deleted after the
2976 first time your program stops there. Also, like the @code{hbreak}
2977 command, the breakpoint requires hardware support and some target hardware
2978 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
2979 See also @ref{Conditions, ,Break Conditions}.
2982 @cindex regular expression
2983 @cindex breakpoints in functions matching a regexp
2984 @cindex set breakpoints in many functions
2985 @item rbreak @var{regex}
2986 Set breakpoints on all functions matching the regular expression
2987 @var{regex}. This command sets an unconditional breakpoint on all
2988 matches, printing a list of all breakpoints it set. Once these
2989 breakpoints are set, they are treated just like the breakpoints set with
2990 the @code{break} command. You can delete them, disable them, or make
2991 them conditional the same way as any other breakpoint.
2993 The syntax of the regular expression is the standard one used with tools
2994 like @file{grep}. Note that this is different from the syntax used by
2995 shells, so for instance @code{foo*} matches all functions that include
2996 an @code{fo} followed by zero or more @code{o}s. There is an implicit
2997 @code{.*} leading and trailing the regular expression you supply, so to
2998 match only functions that begin with @code{foo}, use @code{^foo}.
3000 @cindex non-member C@t{++} functions, set breakpoint in
3001 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3002 breakpoints on overloaded functions that are not members of any special
3005 @cindex set breakpoints on all functions
3006 The @code{rbreak} command can be used to set breakpoints in
3007 @strong{all} the functions in a program, like this:
3010 (@value{GDBP}) rbreak .
3013 @kindex info breakpoints
3014 @cindex @code{$_} and @code{info breakpoints}
3015 @item info breakpoints @r{[}@var{n}@r{]}
3016 @itemx info break @r{[}@var{n}@r{]}
3017 @itemx info watchpoints @r{[}@var{n}@r{]}
3018 Print a table of all breakpoints, watchpoints, and catchpoints set and
3019 not deleted. Optional argument @var{n} means print information only
3020 about the specified breakpoint (or watchpoint or catchpoint). For
3021 each breakpoint, following columns are printed:
3024 @item Breakpoint Numbers
3026 Breakpoint, watchpoint, or catchpoint.
3028 Whether the breakpoint is marked to be disabled or deleted when hit.
3029 @item Enabled or Disabled
3030 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3031 that are not enabled.
3033 Where the breakpoint is in your program, as a memory address. For a
3034 pending breakpoint whose address is not yet known, this field will
3035 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3036 library that has the symbol or line referred by breakpoint is loaded.
3037 See below for details. A breakpoint with several locations will
3038 have @samp{<MULTIPLE>} in this field---see below for details.
3040 Where the breakpoint is in the source for your program, as a file and
3041 line number. For a pending breakpoint, the original string passed to
3042 the breakpoint command will be listed as it cannot be resolved until
3043 the appropriate shared library is loaded in the future.
3047 If a breakpoint is conditional, @code{info break} shows the condition on
3048 the line following the affected breakpoint; breakpoint commands, if any,
3049 are listed after that. A pending breakpoint is allowed to have a condition
3050 specified for it. The condition is not parsed for validity until a shared
3051 library is loaded that allows the pending breakpoint to resolve to a
3055 @code{info break} with a breakpoint
3056 number @var{n} as argument lists only that breakpoint. The
3057 convenience variable @code{$_} and the default examining-address for
3058 the @code{x} command are set to the address of the last breakpoint
3059 listed (@pxref{Memory, ,Examining Memory}).
3062 @code{info break} displays a count of the number of times the breakpoint
3063 has been hit. This is especially useful in conjunction with the
3064 @code{ignore} command. You can ignore a large number of breakpoint
3065 hits, look at the breakpoint info to see how many times the breakpoint
3066 was hit, and then run again, ignoring one less than that number. This
3067 will get you quickly to the last hit of that breakpoint.
3070 @value{GDBN} allows you to set any number of breakpoints at the same place in
3071 your program. There is nothing silly or meaningless about this. When
3072 the breakpoints are conditional, this is even useful
3073 (@pxref{Conditions, ,Break Conditions}).
3075 @cindex multiple locations, breakpoints
3076 @cindex breakpoints, multiple locations
3077 It is possible that a breakpoint corresponds to several locations
3078 in your program. Examples of this situation are:
3082 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3083 instances of the function body, used in different cases.
3086 For a C@t{++} template function, a given line in the function can
3087 correspond to any number of instantiations.
3090 For an inlined function, a given source line can correspond to
3091 several places where that function is inlined.
3094 In all those cases, @value{GDBN} will insert a breakpoint at all
3095 the relevant locations@footnote{
3096 As of this writing, multiple-location breakpoints work only if there's
3097 line number information for all the locations. This means that they
3098 will generally not work in system libraries, unless you have debug
3099 info with line numbers for them.}.
3101 A breakpoint with multiple locations is displayed in the breakpoint
3102 table using several rows---one header row, followed by one row for
3103 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3104 address column. The rows for individual locations contain the actual
3105 addresses for locations, and show the functions to which those
3106 locations belong. The number column for a location is of the form
3107 @var{breakpoint-number}.@var{location-number}.
3112 Num Type Disp Enb Address What
3113 1 breakpoint keep y <MULTIPLE>
3115 breakpoint already hit 1 time
3116 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3117 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3120 Each location can be individually enabled or disabled by passing
3121 @var{breakpoint-number}.@var{location-number} as argument to the
3122 @code{enable} and @code{disable} commands. Note that you cannot
3123 delete the individual locations from the list, you can only delete the
3124 entire list of locations that belong to their parent breakpoint (with
3125 the @kbd{delete @var{num}} command, where @var{num} is the number of
3126 the parent breakpoint, 1 in the above example). Disabling or enabling
3127 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3128 that belong to that breakpoint.
3130 @cindex pending breakpoints
3131 It's quite common to have a breakpoint inside a shared library.
3132 Shared libraries can be loaded and unloaded explicitly,
3133 and possibly repeatedly, as the program is executed. To support
3134 this use case, @value{GDBN} updates breakpoint locations whenever
3135 any shared library is loaded or unloaded. Typically, you would
3136 set a breakpoint in a shared library at the beginning of your
3137 debugging session, when the library is not loaded, and when the
3138 symbols from the library are not available. When you try to set
3139 breakpoint, @value{GDBN} will ask you if you want to set
3140 a so called @dfn{pending breakpoint}---breakpoint whose address
3141 is not yet resolved.
3143 After the program is run, whenever a new shared library is loaded,
3144 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3145 shared library contains the symbol or line referred to by some
3146 pending breakpoint, that breakpoint is resolved and becomes an
3147 ordinary breakpoint. When a library is unloaded, all breakpoints
3148 that refer to its symbols or source lines become pending again.
3150 This logic works for breakpoints with multiple locations, too. For
3151 example, if you have a breakpoint in a C@t{++} template function, and
3152 a newly loaded shared library has an instantiation of that template,
3153 a new location is added to the list of locations for the breakpoint.
3155 Except for having unresolved address, pending breakpoints do not
3156 differ from regular breakpoints. You can set conditions or commands,
3157 enable and disable them and perform other breakpoint operations.
3159 @value{GDBN} provides some additional commands for controlling what
3160 happens when the @samp{break} command cannot resolve breakpoint
3161 address specification to an address:
3163 @kindex set breakpoint pending
3164 @kindex show breakpoint pending
3166 @item set breakpoint pending auto
3167 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3168 location, it queries you whether a pending breakpoint should be created.
3170 @item set breakpoint pending on
3171 This indicates that an unrecognized breakpoint location should automatically
3172 result in a pending breakpoint being created.
3174 @item set breakpoint pending off
3175 This indicates that pending breakpoints are not to be created. Any
3176 unrecognized breakpoint location results in an error. This setting does
3177 not affect any pending breakpoints previously created.
3179 @item show breakpoint pending
3180 Show the current behavior setting for creating pending breakpoints.
3183 The settings above only affect the @code{break} command and its
3184 variants. Once breakpoint is set, it will be automatically updated
3185 as shared libraries are loaded and unloaded.
3187 @cindex automatic hardware breakpoints
3188 For some targets, @value{GDBN} can automatically decide if hardware or
3189 software breakpoints should be used, depending on whether the
3190 breakpoint address is read-only or read-write. This applies to
3191 breakpoints set with the @code{break} command as well as to internal
3192 breakpoints set by commands like @code{next} and @code{finish}. For
3193 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3196 You can control this automatic behaviour with the following commands::
3198 @kindex set breakpoint auto-hw
3199 @kindex show breakpoint auto-hw
3201 @item set breakpoint auto-hw on
3202 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3203 will try to use the target memory map to decide if software or hardware
3204 breakpoint must be used.
3206 @item set breakpoint auto-hw off
3207 This indicates @value{GDBN} should not automatically select breakpoint
3208 type. If the target provides a memory map, @value{GDBN} will warn when
3209 trying to set software breakpoint at a read-only address.
3213 @cindex negative breakpoint numbers
3214 @cindex internal @value{GDBN} breakpoints
3215 @value{GDBN} itself sometimes sets breakpoints in your program for
3216 special purposes, such as proper handling of @code{longjmp} (in C
3217 programs). These internal breakpoints are assigned negative numbers,
3218 starting with @code{-1}; @samp{info breakpoints} does not display them.
3219 You can see these breakpoints with the @value{GDBN} maintenance command
3220 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3223 @node Set Watchpoints
3224 @subsection Setting Watchpoints
3226 @cindex setting watchpoints
3227 You can use a watchpoint to stop execution whenever the value of an
3228 expression changes, without having to predict a particular place where
3229 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3230 The expression may be as simple as the value of a single variable, or
3231 as complex as many variables combined by operators. Examples include:
3235 A reference to the value of a single variable.
3238 An address cast to an appropriate data type. For example,
3239 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3240 address (assuming an @code{int} occupies 4 bytes).
3243 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3244 expression can use any operators valid in the program's native
3245 language (@pxref{Languages}).
3248 You can set a watchpoint on an expression even if the expression can
3249 not be evaluated yet. For instance, you can set a watchpoint on
3250 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3251 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3252 the expression produces a valid value. If the expression becomes
3253 valid in some other way than changing a variable (e.g.@: if the memory
3254 pointed to by @samp{*global_ptr} becomes readable as the result of a
3255 @code{malloc} call), @value{GDBN} may not stop until the next time
3256 the expression changes.
3258 @cindex software watchpoints
3259 @cindex hardware watchpoints
3260 Depending on your system, watchpoints may be implemented in software or
3261 hardware. @value{GDBN} does software watchpointing by single-stepping your
3262 program and testing the variable's value each time, which is hundreds of
3263 times slower than normal execution. (But this may still be worth it, to
3264 catch errors where you have no clue what part of your program is the
3267 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3268 x86-based targets, @value{GDBN} includes support for hardware
3269 watchpoints, which do not slow down the running of your program.
3273 @item watch @var{expr} @r{[}thread @var{threadnum}@r{]}
3274 Set a watchpoint for an expression. @value{GDBN} will break when the
3275 expression @var{expr} is written into by the program and its value
3276 changes. The simplest (and the most popular) use of this command is
3277 to watch the value of a single variable:
3280 (@value{GDBP}) watch foo
3283 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3284 clause, @value{GDBN} breaks only when the thread identified by
3285 @var{threadnum} changes the value of @var{expr}. If any other threads
3286 change the value of @var{expr}, @value{GDBN} will not break. Note
3287 that watchpoints restricted to a single thread in this way only work
3288 with Hardware Watchpoints.
3291 @item rwatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3292 Set a watchpoint that will break when the value of @var{expr} is read
3296 @item awatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3297 Set a watchpoint that will break when @var{expr} is either read from
3298 or written into by the program.
3300 @kindex info watchpoints @r{[}@var{n}@r{]}
3301 @item info watchpoints
3302 This command prints a list of watchpoints, breakpoints, and catchpoints;
3303 it is the same as @code{info break} (@pxref{Set Breaks}).
3306 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3307 watchpoints execute very quickly, and the debugger reports a change in
3308 value at the exact instruction where the change occurs. If @value{GDBN}
3309 cannot set a hardware watchpoint, it sets a software watchpoint, which
3310 executes more slowly and reports the change in value at the next
3311 @emph{statement}, not the instruction, after the change occurs.
3313 @cindex use only software watchpoints
3314 You can force @value{GDBN} to use only software watchpoints with the
3315 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3316 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3317 the underlying system supports them. (Note that hardware-assisted
3318 watchpoints that were set @emph{before} setting
3319 @code{can-use-hw-watchpoints} to zero will still use the hardware
3320 mechanism of watching expression values.)
3323 @item set can-use-hw-watchpoints
3324 @kindex set can-use-hw-watchpoints
3325 Set whether or not to use hardware watchpoints.
3327 @item show can-use-hw-watchpoints
3328 @kindex show can-use-hw-watchpoints
3329 Show the current mode of using hardware watchpoints.
3332 For remote targets, you can restrict the number of hardware
3333 watchpoints @value{GDBN} will use, see @ref{set remote
3334 hardware-breakpoint-limit}.
3336 When you issue the @code{watch} command, @value{GDBN} reports
3339 Hardware watchpoint @var{num}: @var{expr}
3343 if it was able to set a hardware watchpoint.
3345 Currently, the @code{awatch} and @code{rwatch} commands can only set
3346 hardware watchpoints, because accesses to data that don't change the
3347 value of the watched expression cannot be detected without examining
3348 every instruction as it is being executed, and @value{GDBN} does not do
3349 that currently. If @value{GDBN} finds that it is unable to set a
3350 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3351 will print a message like this:
3354 Expression cannot be implemented with read/access watchpoint.
3357 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3358 data type of the watched expression is wider than what a hardware
3359 watchpoint on the target machine can handle. For example, some systems
3360 can only watch regions that are up to 4 bytes wide; on such systems you
3361 cannot set hardware watchpoints for an expression that yields a
3362 double-precision floating-point number (which is typically 8 bytes
3363 wide). As a work-around, it might be possible to break the large region
3364 into a series of smaller ones and watch them with separate watchpoints.
3366 If you set too many hardware watchpoints, @value{GDBN} might be unable
3367 to insert all of them when you resume the execution of your program.
3368 Since the precise number of active watchpoints is unknown until such
3369 time as the program is about to be resumed, @value{GDBN} might not be
3370 able to warn you about this when you set the watchpoints, and the
3371 warning will be printed only when the program is resumed:
3374 Hardware watchpoint @var{num}: Could not insert watchpoint
3378 If this happens, delete or disable some of the watchpoints.
3380 Watching complex expressions that reference many variables can also
3381 exhaust the resources available for hardware-assisted watchpoints.
3382 That's because @value{GDBN} needs to watch every variable in the
3383 expression with separately allocated resources.
3385 If you call a function interactively using @code{print} or @code{call},
3386 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3387 kind of breakpoint or the call completes.
3389 @value{GDBN} automatically deletes watchpoints that watch local
3390 (automatic) variables, or expressions that involve such variables, when
3391 they go out of scope, that is, when the execution leaves the block in
3392 which these variables were defined. In particular, when the program
3393 being debugged terminates, @emph{all} local variables go out of scope,
3394 and so only watchpoints that watch global variables remain set. If you
3395 rerun the program, you will need to set all such watchpoints again. One
3396 way of doing that would be to set a code breakpoint at the entry to the
3397 @code{main} function and when it breaks, set all the watchpoints.
3399 @cindex watchpoints and threads
3400 @cindex threads and watchpoints
3401 In multi-threaded programs, watchpoints will detect changes to the
3402 watched expression from every thread.
3405 @emph{Warning:} In multi-threaded programs, software watchpoints
3406 have only limited usefulness. If @value{GDBN} creates a software
3407 watchpoint, it can only watch the value of an expression @emph{in a
3408 single thread}. If you are confident that the expression can only
3409 change due to the current thread's activity (and if you are also
3410 confident that no other thread can become current), then you can use
3411 software watchpoints as usual. However, @value{GDBN} may not notice
3412 when a non-current thread's activity changes the expression. (Hardware
3413 watchpoints, in contrast, watch an expression in all threads.)
3416 @xref{set remote hardware-watchpoint-limit}.
3418 @node Set Catchpoints
3419 @subsection Setting Catchpoints
3420 @cindex catchpoints, setting
3421 @cindex exception handlers
3422 @cindex event handling
3424 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3425 kinds of program events, such as C@t{++} exceptions or the loading of a
3426 shared library. Use the @code{catch} command to set a catchpoint.
3430 @item catch @var{event}
3431 Stop when @var{event} occurs. @var{event} can be any of the following:
3434 @cindex stop on C@t{++} exceptions
3435 The throwing of a C@t{++} exception.
3438 The catching of a C@t{++} exception.
3441 @cindex Ada exception catching
3442 @cindex catch Ada exceptions
3443 An Ada exception being raised. If an exception name is specified
3444 at the end of the command (eg @code{catch exception Program_Error}),
3445 the debugger will stop only when this specific exception is raised.
3446 Otherwise, the debugger stops execution when any Ada exception is raised.
3448 @item exception unhandled
3449 An exception that was raised but is not handled by the program.
3452 A failed Ada assertion.
3455 @cindex break on fork/exec
3456 A call to @code{exec}. This is currently only available for HP-UX
3460 A call to @code{fork}. This is currently only available for HP-UX
3464 A call to @code{vfork}. This is currently only available for HP-UX
3468 @itemx load @var{libname}
3469 @cindex break on load/unload of shared library
3470 The dynamic loading of any shared library, or the loading of the library
3471 @var{libname}. This is currently only available for HP-UX.
3474 @itemx unload @var{libname}
3475 The unloading of any dynamically loaded shared library, or the unloading
3476 of the library @var{libname}. This is currently only available for HP-UX.
3479 @item tcatch @var{event}
3480 Set a catchpoint that is enabled only for one stop. The catchpoint is
3481 automatically deleted after the first time the event is caught.
3485 Use the @code{info break} command to list the current catchpoints.
3487 There are currently some limitations to C@t{++} exception handling
3488 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
3492 If you call a function interactively, @value{GDBN} normally returns
3493 control to you when the function has finished executing. If the call
3494 raises an exception, however, the call may bypass the mechanism that
3495 returns control to you and cause your program either to abort or to
3496 simply continue running until it hits a breakpoint, catches a signal
3497 that @value{GDBN} is listening for, or exits. This is the case even if
3498 you set a catchpoint for the exception; catchpoints on exceptions are
3499 disabled within interactive calls.
3502 You cannot raise an exception interactively.
3505 You cannot install an exception handler interactively.
3508 @cindex raise exceptions
3509 Sometimes @code{catch} is not the best way to debug exception handling:
3510 if you need to know exactly where an exception is raised, it is better to
3511 stop @emph{before} the exception handler is called, since that way you
3512 can see the stack before any unwinding takes place. If you set a
3513 breakpoint in an exception handler instead, it may not be easy to find
3514 out where the exception was raised.
3516 To stop just before an exception handler is called, you need some
3517 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
3518 raised by calling a library function named @code{__raise_exception}
3519 which has the following ANSI C interface:
3522 /* @var{addr} is where the exception identifier is stored.
3523 @var{id} is the exception identifier. */
3524 void __raise_exception (void **addr, void *id);
3528 To make the debugger catch all exceptions before any stack
3529 unwinding takes place, set a breakpoint on @code{__raise_exception}
3530 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
3532 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
3533 that depends on the value of @var{id}, you can stop your program when
3534 a specific exception is raised. You can use multiple conditional
3535 breakpoints to stop your program when any of a number of exceptions are
3540 @subsection Deleting Breakpoints
3542 @cindex clearing breakpoints, watchpoints, catchpoints
3543 @cindex deleting breakpoints, watchpoints, catchpoints
3544 It is often necessary to eliminate a breakpoint, watchpoint, or
3545 catchpoint once it has done its job and you no longer want your program
3546 to stop there. This is called @dfn{deleting} the breakpoint. A
3547 breakpoint that has been deleted no longer exists; it is forgotten.
3549 With the @code{clear} command you can delete breakpoints according to
3550 where they are in your program. With the @code{delete} command you can
3551 delete individual breakpoints, watchpoints, or catchpoints by specifying
3552 their breakpoint numbers.
3554 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
3555 automatically ignores breakpoints on the first instruction to be executed
3556 when you continue execution without changing the execution address.
3561 Delete any breakpoints at the next instruction to be executed in the
3562 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
3563 the innermost frame is selected, this is a good way to delete a
3564 breakpoint where your program just stopped.
3566 @item clear @var{location}
3567 Delete any breakpoints set at the specified @var{location}.
3568 @xref{Specify Location}, for the various forms of @var{location}; the
3569 most useful ones are listed below:
3572 @item clear @var{function}
3573 @itemx clear @var{filename}:@var{function}
3574 Delete any breakpoints set at entry to the named @var{function}.
3576 @item clear @var{linenum}
3577 @itemx clear @var{filename}:@var{linenum}
3578 Delete any breakpoints set at or within the code of the specified
3579 @var{linenum} of the specified @var{filename}.
3582 @cindex delete breakpoints
3584 @kindex d @r{(@code{delete})}
3585 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3586 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
3587 ranges specified as arguments. If no argument is specified, delete all
3588 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
3589 confirm off}). You can abbreviate this command as @code{d}.
3593 @subsection Disabling Breakpoints
3595 @cindex enable/disable a breakpoint
3596 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
3597 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
3598 it had been deleted, but remembers the information on the breakpoint so
3599 that you can @dfn{enable} it again later.
3601 You disable and enable breakpoints, watchpoints, and catchpoints with
3602 the @code{enable} and @code{disable} commands, optionally specifying one
3603 or more breakpoint numbers as arguments. Use @code{info break} or
3604 @code{info watch} to print a list of breakpoints, watchpoints, and
3605 catchpoints if you do not know which numbers to use.
3607 Disabling and enabling a breakpoint that has multiple locations
3608 affects all of its locations.
3610 A breakpoint, watchpoint, or catchpoint can have any of four different
3611 states of enablement:
3615 Enabled. The breakpoint stops your program. A breakpoint set
3616 with the @code{break} command starts out in this state.
3618 Disabled. The breakpoint has no effect on your program.
3620 Enabled once. The breakpoint stops your program, but then becomes
3623 Enabled for deletion. The breakpoint stops your program, but
3624 immediately after it does so it is deleted permanently. A breakpoint
3625 set with the @code{tbreak} command starts out in this state.
3628 You can use the following commands to enable or disable breakpoints,
3629 watchpoints, and catchpoints:
3633 @kindex dis @r{(@code{disable})}
3634 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3635 Disable the specified breakpoints---or all breakpoints, if none are
3636 listed. A disabled breakpoint has no effect but is not forgotten. All
3637 options such as ignore-counts, conditions and commands are remembered in
3638 case the breakpoint is enabled again later. You may abbreviate
3639 @code{disable} as @code{dis}.
3642 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3643 Enable the specified breakpoints (or all defined breakpoints). They
3644 become effective once again in stopping your program.
3646 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
3647 Enable the specified breakpoints temporarily. @value{GDBN} disables any
3648 of these breakpoints immediately after stopping your program.
3650 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
3651 Enable the specified breakpoints to work once, then die. @value{GDBN}
3652 deletes any of these breakpoints as soon as your program stops there.
3653 Breakpoints set by the @code{tbreak} command start out in this state.
3656 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
3657 @c confusing: tbreak is also initially enabled.
3658 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
3659 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
3660 subsequently, they become disabled or enabled only when you use one of
3661 the commands above. (The command @code{until} can set and delete a
3662 breakpoint of its own, but it does not change the state of your other
3663 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
3667 @subsection Break Conditions
3668 @cindex conditional breakpoints
3669 @cindex breakpoint conditions
3671 @c FIXME what is scope of break condition expr? Context where wanted?
3672 @c in particular for a watchpoint?
3673 The simplest sort of breakpoint breaks every time your program reaches a
3674 specified place. You can also specify a @dfn{condition} for a
3675 breakpoint. A condition is just a Boolean expression in your
3676 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
3677 a condition evaluates the expression each time your program reaches it,
3678 and your program stops only if the condition is @emph{true}.
3680 This is the converse of using assertions for program validation; in that
3681 situation, you want to stop when the assertion is violated---that is,
3682 when the condition is false. In C, if you want to test an assertion expressed
3683 by the condition @var{assert}, you should set the condition
3684 @samp{! @var{assert}} on the appropriate breakpoint.
3686 Conditions are also accepted for watchpoints; you may not need them,
3687 since a watchpoint is inspecting the value of an expression anyhow---but
3688 it might be simpler, say, to just set a watchpoint on a variable name,
3689 and specify a condition that tests whether the new value is an interesting
3692 Break conditions can have side effects, and may even call functions in
3693 your program. This can be useful, for example, to activate functions
3694 that log program progress, or to use your own print functions to
3695 format special data structures. The effects are completely predictable
3696 unless there is another enabled breakpoint at the same address. (In
3697 that case, @value{GDBN} might see the other breakpoint first and stop your
3698 program without checking the condition of this one.) Note that
3699 breakpoint commands are usually more convenient and flexible than break
3701 purpose of performing side effects when a breakpoint is reached
3702 (@pxref{Break Commands, ,Breakpoint Command Lists}).
3704 Break conditions can be specified when a breakpoint is set, by using
3705 @samp{if} in the arguments to the @code{break} command. @xref{Set
3706 Breaks, ,Setting Breakpoints}. They can also be changed at any time
3707 with the @code{condition} command.
3709 You can also use the @code{if} keyword with the @code{watch} command.
3710 The @code{catch} command does not recognize the @code{if} keyword;
3711 @code{condition} is the only way to impose a further condition on a
3716 @item condition @var{bnum} @var{expression}
3717 Specify @var{expression} as the break condition for breakpoint,
3718 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
3719 breakpoint @var{bnum} stops your program only if the value of
3720 @var{expression} is true (nonzero, in C). When you use
3721 @code{condition}, @value{GDBN} checks @var{expression} immediately for
3722 syntactic correctness, and to determine whether symbols in it have
3723 referents in the context of your breakpoint. If @var{expression} uses
3724 symbols not referenced in the context of the breakpoint, @value{GDBN}
3725 prints an error message:
3728 No symbol "foo" in current context.
3733 not actually evaluate @var{expression} at the time the @code{condition}
3734 command (or a command that sets a breakpoint with a condition, like
3735 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
3737 @item condition @var{bnum}
3738 Remove the condition from breakpoint number @var{bnum}. It becomes
3739 an ordinary unconditional breakpoint.
3742 @cindex ignore count (of breakpoint)
3743 A special case of a breakpoint condition is to stop only when the
3744 breakpoint has been reached a certain number of times. This is so
3745 useful that there is a special way to do it, using the @dfn{ignore
3746 count} of the breakpoint. Every breakpoint has an ignore count, which
3747 is an integer. Most of the time, the ignore count is zero, and
3748 therefore has no effect. But if your program reaches a breakpoint whose
3749 ignore count is positive, then instead of stopping, it just decrements
3750 the ignore count by one and continues. As a result, if the ignore count
3751 value is @var{n}, the breakpoint does not stop the next @var{n} times
3752 your program reaches it.
3756 @item ignore @var{bnum} @var{count}
3757 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
3758 The next @var{count} times the breakpoint is reached, your program's
3759 execution does not stop; other than to decrement the ignore count, @value{GDBN}
3762 To make the breakpoint stop the next time it is reached, specify
3765 When you use @code{continue} to resume execution of your program from a
3766 breakpoint, you can specify an ignore count directly as an argument to
3767 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
3768 Stepping,,Continuing and Stepping}.
3770 If a breakpoint has a positive ignore count and a condition, the
3771 condition is not checked. Once the ignore count reaches zero,
3772 @value{GDBN} resumes checking the condition.
3774 You could achieve the effect of the ignore count with a condition such
3775 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
3776 is decremented each time. @xref{Convenience Vars, ,Convenience
3780 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
3783 @node Break Commands
3784 @subsection Breakpoint Command Lists
3786 @cindex breakpoint commands
3787 You can give any breakpoint (or watchpoint or catchpoint) a series of
3788 commands to execute when your program stops due to that breakpoint. For
3789 example, you might want to print the values of certain expressions, or
3790 enable other breakpoints.
3794 @kindex end@r{ (breakpoint commands)}
3795 @item commands @r{[}@var{bnum}@r{]}
3796 @itemx @dots{} @var{command-list} @dots{}
3798 Specify a list of commands for breakpoint number @var{bnum}. The commands
3799 themselves appear on the following lines. Type a line containing just
3800 @code{end} to terminate the commands.
3802 To remove all commands from a breakpoint, type @code{commands} and
3803 follow it immediately with @code{end}; that is, give no commands.
3805 With no @var{bnum} argument, @code{commands} refers to the last
3806 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
3807 recently encountered).
3810 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
3811 disabled within a @var{command-list}.
3813 You can use breakpoint commands to start your program up again. Simply
3814 use the @code{continue} command, or @code{step}, or any other command
3815 that resumes execution.
3817 Any other commands in the command list, after a command that resumes
3818 execution, are ignored. This is because any time you resume execution
3819 (even with a simple @code{next} or @code{step}), you may encounter
3820 another breakpoint---which could have its own command list, leading to
3821 ambiguities about which list to execute.
3824 If the first command you specify in a command list is @code{silent}, the
3825 usual message about stopping at a breakpoint is not printed. This may
3826 be desirable for breakpoints that are to print a specific message and
3827 then continue. If none of the remaining commands print anything, you
3828 see no sign that the breakpoint was reached. @code{silent} is
3829 meaningful only at the beginning of a breakpoint command list.
3831 The commands @code{echo}, @code{output}, and @code{printf} allow you to
3832 print precisely controlled output, and are often useful in silent
3833 breakpoints. @xref{Output, ,Commands for Controlled Output}.
3835 For example, here is how you could use breakpoint commands to print the
3836 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
3842 printf "x is %d\n",x
3847 One application for breakpoint commands is to compensate for one bug so
3848 you can test for another. Put a breakpoint just after the erroneous line
3849 of code, give it a condition to detect the case in which something
3850 erroneous has been done, and give it commands to assign correct values
3851 to any variables that need them. End with the @code{continue} command
3852 so that your program does not stop, and start with the @code{silent}
3853 command so that no output is produced. Here is an example:
3864 @c @ifclear BARETARGET
3865 @node Error in Breakpoints
3866 @subsection ``Cannot insert breakpoints''
3868 @c FIXME!! 14/6/95 Is there a real example of this? Let's use it.
3870 Under some operating systems, breakpoints cannot be used in a program if
3871 any other process is running that program. In this situation,
3872 attempting to run or continue a program with a breakpoint causes
3873 @value{GDBN} to print an error message:
3876 Cannot insert breakpoints.
3877 The same program may be running in another process.
3880 When this happens, you have three ways to proceed:
3884 Remove or disable the breakpoints, then continue.
3887 Suspend @value{GDBN}, and copy the file containing your program to a new
3888 name. Resume @value{GDBN} and use the @code{exec-file} command to specify
3889 that @value{GDBN} should run your program under that name.
3890 Then start your program again.
3893 Relink your program so that the text segment is nonsharable, using the
3894 linker option @samp{-N}. The operating system limitation may not apply
3895 to nonsharable executables.
3899 A similar message can be printed if you request too many active
3900 hardware-assisted breakpoints and watchpoints:
3902 @c FIXME: the precise wording of this message may change; the relevant
3903 @c source change is not committed yet (Sep 3, 1999).
3905 Stopped; cannot insert breakpoints.
3906 You may have requested too many hardware breakpoints and watchpoints.
3910 This message is printed when you attempt to resume the program, since
3911 only then @value{GDBN} knows exactly how many hardware breakpoints and
3912 watchpoints it needs to insert.
3914 When this message is printed, you need to disable or remove some of the
3915 hardware-assisted breakpoints and watchpoints, and then continue.
3917 @node Breakpoint-related Warnings
3918 @subsection ``Breakpoint address adjusted...''
3919 @cindex breakpoint address adjusted
3921 Some processor architectures place constraints on the addresses at
3922 which breakpoints may be placed. For architectures thus constrained,
3923 @value{GDBN} will attempt to adjust the breakpoint's address to comply
3924 with the constraints dictated by the architecture.
3926 One example of such an architecture is the Fujitsu FR-V. The FR-V is
3927 a VLIW architecture in which a number of RISC-like instructions may be
3928 bundled together for parallel execution. The FR-V architecture
3929 constrains the location of a breakpoint instruction within such a
3930 bundle to the instruction with the lowest address. @value{GDBN}
3931 honors this constraint by adjusting a breakpoint's address to the
3932 first in the bundle.
3934 It is not uncommon for optimized code to have bundles which contain
3935 instructions from different source statements, thus it may happen that
3936 a breakpoint's address will be adjusted from one source statement to
3937 another. Since this adjustment may significantly alter @value{GDBN}'s
3938 breakpoint related behavior from what the user expects, a warning is
3939 printed when the breakpoint is first set and also when the breakpoint
3942 A warning like the one below is printed when setting a breakpoint
3943 that's been subject to address adjustment:
3946 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
3949 Such warnings are printed both for user settable and @value{GDBN}'s
3950 internal breakpoints. If you see one of these warnings, you should
3951 verify that a breakpoint set at the adjusted address will have the
3952 desired affect. If not, the breakpoint in question may be removed and
3953 other breakpoints may be set which will have the desired behavior.
3954 E.g., it may be sufficient to place the breakpoint at a later
3955 instruction. A conditional breakpoint may also be useful in some
3956 cases to prevent the breakpoint from triggering too often.
3958 @value{GDBN} will also issue a warning when stopping at one of these
3959 adjusted breakpoints:
3962 warning: Breakpoint 1 address previously adjusted from 0x00010414
3966 When this warning is encountered, it may be too late to take remedial
3967 action except in cases where the breakpoint is hit earlier or more
3968 frequently than expected.
3970 @node Continuing and Stepping
3971 @section Continuing and Stepping
3975 @cindex resuming execution
3976 @dfn{Continuing} means resuming program execution until your program
3977 completes normally. In contrast, @dfn{stepping} means executing just
3978 one more ``step'' of your program, where ``step'' may mean either one
3979 line of source code, or one machine instruction (depending on what
3980 particular command you use). Either when continuing or when stepping,
3981 your program may stop even sooner, due to a breakpoint or a signal. (If
3982 it stops due to a signal, you may want to use @code{handle}, or use
3983 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
3987 @kindex c @r{(@code{continue})}
3988 @kindex fg @r{(resume foreground execution)}
3989 @item continue @r{[}@var{ignore-count}@r{]}
3990 @itemx c @r{[}@var{ignore-count}@r{]}
3991 @itemx fg @r{[}@var{ignore-count}@r{]}
3992 Resume program execution, at the address where your program last stopped;
3993 any breakpoints set at that address are bypassed. The optional argument
3994 @var{ignore-count} allows you to specify a further number of times to
3995 ignore a breakpoint at this location; its effect is like that of
3996 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
3998 The argument @var{ignore-count} is meaningful only when your program
3999 stopped due to a breakpoint. At other times, the argument to
4000 @code{continue} is ignored.
4002 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4003 debugged program is deemed to be the foreground program) are provided
4004 purely for convenience, and have exactly the same behavior as
4008 To resume execution at a different place, you can use @code{return}
4009 (@pxref{Returning, ,Returning from a Function}) to go back to the
4010 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4011 Different Address}) to go to an arbitrary location in your program.
4013 A typical technique for using stepping is to set a breakpoint
4014 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4015 beginning of the function or the section of your program where a problem
4016 is believed to lie, run your program until it stops at that breakpoint,
4017 and then step through the suspect area, examining the variables that are
4018 interesting, until you see the problem happen.
4022 @kindex s @r{(@code{step})}
4024 Continue running your program until control reaches a different source
4025 line, then stop it and return control to @value{GDBN}. This command is
4026 abbreviated @code{s}.
4029 @c "without debugging information" is imprecise; actually "without line
4030 @c numbers in the debugging information". (gcc -g1 has debugging info but
4031 @c not line numbers). But it seems complex to try to make that
4032 @c distinction here.
4033 @emph{Warning:} If you use the @code{step} command while control is
4034 within a function that was compiled without debugging information,
4035 execution proceeds until control reaches a function that does have
4036 debugging information. Likewise, it will not step into a function which
4037 is compiled without debugging information. To step through functions
4038 without debugging information, use the @code{stepi} command, described
4042 The @code{step} command only stops at the first instruction of a source
4043 line. This prevents the multiple stops that could otherwise occur in
4044 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4045 to stop if a function that has debugging information is called within
4046 the line. In other words, @code{step} @emph{steps inside} any functions
4047 called within the line.
4049 Also, the @code{step} command only enters a function if there is line
4050 number information for the function. Otherwise it acts like the
4051 @code{next} command. This avoids problems when using @code{cc -gl}
4052 on MIPS machines. Previously, @code{step} entered subroutines if there
4053 was any debugging information about the routine.
4055 @item step @var{count}
4056 Continue running as in @code{step}, but do so @var{count} times. If a
4057 breakpoint is reached, or a signal not related to stepping occurs before
4058 @var{count} steps, stepping stops right away.
4061 @kindex n @r{(@code{next})}
4062 @item next @r{[}@var{count}@r{]}
4063 Continue to the next source line in the current (innermost) stack frame.
4064 This is similar to @code{step}, but function calls that appear within
4065 the line of code are executed without stopping. Execution stops when
4066 control reaches a different line of code at the original stack level
4067 that was executing when you gave the @code{next} command. This command
4068 is abbreviated @code{n}.
4070 An argument @var{count} is a repeat count, as for @code{step}.
4073 @c FIX ME!! Do we delete this, or is there a way it fits in with
4074 @c the following paragraph? --- Vctoria
4076 @c @code{next} within a function that lacks debugging information acts like
4077 @c @code{step}, but any function calls appearing within the code of the
4078 @c function are executed without stopping.
4080 The @code{next} command only stops at the first instruction of a
4081 source line. This prevents multiple stops that could otherwise occur in
4082 @code{switch} statements, @code{for} loops, etc.
4084 @kindex set step-mode
4086 @cindex functions without line info, and stepping
4087 @cindex stepping into functions with no line info
4088 @itemx set step-mode on
4089 The @code{set step-mode on} command causes the @code{step} command to
4090 stop at the first instruction of a function which contains no debug line
4091 information rather than stepping over it.
4093 This is useful in cases where you may be interested in inspecting the
4094 machine instructions of a function which has no symbolic info and do not
4095 want @value{GDBN} to automatically skip over this function.
4097 @item set step-mode off
4098 Causes the @code{step} command to step over any functions which contains no
4099 debug information. This is the default.
4101 @item show step-mode
4102 Show whether @value{GDBN} will stop in or step over functions without
4103 source line debug information.
4107 Continue running until just after function in the selected stack frame
4108 returns. Print the returned value (if any).
4110 Contrast this with the @code{return} command (@pxref{Returning,
4111 ,Returning from a Function}).
4114 @kindex u @r{(@code{until})}
4115 @cindex run until specified location
4118 Continue running until a source line past the current line, in the
4119 current stack frame, is reached. This command is used to avoid single
4120 stepping through a loop more than once. It is like the @code{next}
4121 command, except that when @code{until} encounters a jump, it
4122 automatically continues execution until the program counter is greater
4123 than the address of the jump.
4125 This means that when you reach the end of a loop after single stepping
4126 though it, @code{until} makes your program continue execution until it
4127 exits the loop. In contrast, a @code{next} command at the end of a loop
4128 simply steps back to the beginning of the loop, which forces you to step
4129 through the next iteration.
4131 @code{until} always stops your program if it attempts to exit the current
4134 @code{until} may produce somewhat counterintuitive results if the order
4135 of machine code does not match the order of the source lines. For
4136 example, in the following excerpt from a debugging session, the @code{f}
4137 (@code{frame}) command shows that execution is stopped at line
4138 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4142 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4144 (@value{GDBP}) until
4145 195 for ( ; argc > 0; NEXTARG) @{
4148 This happened because, for execution efficiency, the compiler had
4149 generated code for the loop closure test at the end, rather than the
4150 start, of the loop---even though the test in a C @code{for}-loop is
4151 written before the body of the loop. The @code{until} command appeared
4152 to step back to the beginning of the loop when it advanced to this
4153 expression; however, it has not really gone to an earlier
4154 statement---not in terms of the actual machine code.
4156 @code{until} with no argument works by means of single
4157 instruction stepping, and hence is slower than @code{until} with an
4160 @item until @var{location}
4161 @itemx u @var{location}
4162 Continue running your program until either the specified location is
4163 reached, or the current stack frame returns. @var{location} is any of
4164 the forms described in @ref{Specify Location}.
4165 This form of the command uses temporary breakpoints, and
4166 hence is quicker than @code{until} without an argument. The specified
4167 location is actually reached only if it is in the current frame. This
4168 implies that @code{until} can be used to skip over recursive function
4169 invocations. For instance in the code below, if the current location is
4170 line @code{96}, issuing @code{until 99} will execute the program up to
4171 line @code{99} in the same invocation of factorial, i.e., after the inner
4172 invocations have returned.
4175 94 int factorial (int value)
4177 96 if (value > 1) @{
4178 97 value *= factorial (value - 1);
4185 @kindex advance @var{location}
4186 @itemx advance @var{location}
4187 Continue running the program up to the given @var{location}. An argument is
4188 required, which should be of one of the forms described in
4189 @ref{Specify Location}.
4190 Execution will also stop upon exit from the current stack
4191 frame. This command is similar to @code{until}, but @code{advance} will
4192 not skip over recursive function calls, and the target location doesn't
4193 have to be in the same frame as the current one.
4197 @kindex si @r{(@code{stepi})}
4199 @itemx stepi @var{arg}
4201 Execute one machine instruction, then stop and return to the debugger.
4203 It is often useful to do @samp{display/i $pc} when stepping by machine
4204 instructions. This makes @value{GDBN} automatically display the next
4205 instruction to be executed, each time your program stops. @xref{Auto
4206 Display,, Automatic Display}.
4208 An argument is a repeat count, as in @code{step}.
4212 @kindex ni @r{(@code{nexti})}
4214 @itemx nexti @var{arg}
4216 Execute one machine instruction, but if it is a function call,
4217 proceed until the function returns.
4219 An argument is a repeat count, as in @code{next}.
4226 A signal is an asynchronous event that can happen in a program. The
4227 operating system defines the possible kinds of signals, and gives each
4228 kind a name and a number. For example, in Unix @code{SIGINT} is the
4229 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4230 @code{SIGSEGV} is the signal a program gets from referencing a place in
4231 memory far away from all the areas in use; @code{SIGALRM} occurs when
4232 the alarm clock timer goes off (which happens only if your program has
4233 requested an alarm).
4235 @cindex fatal signals
4236 Some signals, including @code{SIGALRM}, are a normal part of the
4237 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4238 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4239 program has not specified in advance some other way to handle the signal.
4240 @code{SIGINT} does not indicate an error in your program, but it is normally
4241 fatal so it can carry out the purpose of the interrupt: to kill the program.
4243 @value{GDBN} has the ability to detect any occurrence of a signal in your
4244 program. You can tell @value{GDBN} in advance what to do for each kind of
4247 @cindex handling signals
4248 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4249 @code{SIGALRM} be silently passed to your program
4250 (so as not to interfere with their role in the program's functioning)
4251 but to stop your program immediately whenever an error signal happens.
4252 You can change these settings with the @code{handle} command.
4255 @kindex info signals
4259 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4260 handle each one. You can use this to see the signal numbers of all
4261 the defined types of signals.
4263 @item info signals @var{sig}
4264 Similar, but print information only about the specified signal number.
4266 @code{info handle} is an alias for @code{info signals}.
4269 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4270 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4271 can be the number of a signal or its name (with or without the
4272 @samp{SIG} at the beginning); a list of signal numbers of the form
4273 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4274 known signals. Optional arguments @var{keywords}, described below,
4275 say what change to make.
4279 The keywords allowed by the @code{handle} command can be abbreviated.
4280 Their full names are:
4284 @value{GDBN} should not stop your program when this signal happens. It may
4285 still print a message telling you that the signal has come in.
4288 @value{GDBN} should stop your program when this signal happens. This implies
4289 the @code{print} keyword as well.
4292 @value{GDBN} should print a message when this signal happens.
4295 @value{GDBN} should not mention the occurrence of the signal at all. This
4296 implies the @code{nostop} keyword as well.
4300 @value{GDBN} should allow your program to see this signal; your program
4301 can handle the signal, or else it may terminate if the signal is fatal
4302 and not handled. @code{pass} and @code{noignore} are synonyms.
4306 @value{GDBN} should not allow your program to see this signal.
4307 @code{nopass} and @code{ignore} are synonyms.
4311 When a signal stops your program, the signal is not visible to the
4313 continue. Your program sees the signal then, if @code{pass} is in
4314 effect for the signal in question @emph{at that time}. In other words,
4315 after @value{GDBN} reports a signal, you can use the @code{handle}
4316 command with @code{pass} or @code{nopass} to control whether your
4317 program sees that signal when you continue.
4319 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4320 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4321 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4324 You can also use the @code{signal} command to prevent your program from
4325 seeing a signal, or cause it to see a signal it normally would not see,
4326 or to give it any signal at any time. For example, if your program stopped
4327 due to some sort of memory reference error, you might store correct
4328 values into the erroneous variables and continue, hoping to see more
4329 execution; but your program would probably terminate immediately as
4330 a result of the fatal signal once it saw the signal. To prevent this,
4331 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4335 @section Stopping and Starting Multi-thread Programs
4337 When your program has multiple threads (@pxref{Threads,, Debugging
4338 Programs with Multiple Threads}), you can choose whether to set
4339 breakpoints on all threads, or on a particular thread.
4342 @cindex breakpoints and threads
4343 @cindex thread breakpoints
4344 @kindex break @dots{} thread @var{threadno}
4345 @item break @var{linespec} thread @var{threadno}
4346 @itemx break @var{linespec} thread @var{threadno} if @dots{}
4347 @var{linespec} specifies source lines; there are several ways of
4348 writing them (@pxref{Specify Location}), but the effect is always to
4349 specify some source line.
4351 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
4352 to specify that you only want @value{GDBN} to stop the program when a
4353 particular thread reaches this breakpoint. @var{threadno} is one of the
4354 numeric thread identifiers assigned by @value{GDBN}, shown in the first
4355 column of the @samp{info threads} display.
4357 If you do not specify @samp{thread @var{threadno}} when you set a
4358 breakpoint, the breakpoint applies to @emph{all} threads of your
4361 You can use the @code{thread} qualifier on conditional breakpoints as
4362 well; in this case, place @samp{thread @var{threadno}} before the
4363 breakpoint condition, like this:
4366 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
4371 @cindex stopped threads
4372 @cindex threads, stopped
4373 Whenever your program stops under @value{GDBN} for any reason,
4374 @emph{all} threads of execution stop, not just the current thread. This
4375 allows you to examine the overall state of the program, including
4376 switching between threads, without worrying that things may change
4379 @cindex thread breakpoints and system calls
4380 @cindex system calls and thread breakpoints
4381 @cindex premature return from system calls
4382 There is an unfortunate side effect. If one thread stops for a
4383 breakpoint, or for some other reason, and another thread is blocked in a
4384 system call, then the system call may return prematurely. This is a
4385 consequence of the interaction between multiple threads and the signals
4386 that @value{GDBN} uses to implement breakpoints and other events that
4389 To handle this problem, your program should check the return value of
4390 each system call and react appropriately. This is good programming
4393 For example, do not write code like this:
4399 The call to @code{sleep} will return early if a different thread stops
4400 at a breakpoint or for some other reason.
4402 Instead, write this:
4407 unslept = sleep (unslept);
4410 A system call is allowed to return early, so the system is still
4411 conforming to its specification. But @value{GDBN} does cause your
4412 multi-threaded program to behave differently than it would without
4415 Also, @value{GDBN} uses internal breakpoints in the thread library to
4416 monitor certain events such as thread creation and thread destruction.
4417 When such an event happens, a system call in another thread may return
4418 prematurely, even though your program does not appear to stop.
4420 @cindex continuing threads
4421 @cindex threads, continuing
4422 Conversely, whenever you restart the program, @emph{all} threads start
4423 executing. @emph{This is true even when single-stepping} with commands
4424 like @code{step} or @code{next}.
4426 In particular, @value{GDBN} cannot single-step all threads in lockstep.
4427 Since thread scheduling is up to your debugging target's operating
4428 system (not controlled by @value{GDBN}), other threads may
4429 execute more than one statement while the current thread completes a
4430 single step. Moreover, in general other threads stop in the middle of a
4431 statement, rather than at a clean statement boundary, when the program
4434 You might even find your program stopped in another thread after
4435 continuing or even single-stepping. This happens whenever some other
4436 thread runs into a breakpoint, a signal, or an exception before the
4437 first thread completes whatever you requested.
4439 On some OSes, you can lock the OS scheduler and thus allow only a single
4443 @item set scheduler-locking @var{mode}
4444 @cindex scheduler locking mode
4445 @cindex lock scheduler
4446 Set the scheduler locking mode. If it is @code{off}, then there is no
4447 locking and any thread may run at any time. If @code{on}, then only the
4448 current thread may run when the inferior is resumed. The @code{step}
4449 mode optimizes for single-stepping. It stops other threads from
4450 ``seizing the prompt'' by preempting the current thread while you are
4451 stepping. Other threads will only rarely (or never) get a chance to run
4452 when you step. They are more likely to run when you @samp{next} over a
4453 function call, and they are completely free to run when you use commands
4454 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
4455 thread hits a breakpoint during its timeslice, they will never steal the
4456 @value{GDBN} prompt away from the thread that you are debugging.
4458 @item show scheduler-locking
4459 Display the current scheduler locking mode.
4464 @chapter Examining the Stack
4466 When your program has stopped, the first thing you need to know is where it
4467 stopped and how it got there.
4470 Each time your program performs a function call, information about the call
4472 That information includes the location of the call in your program,
4473 the arguments of the call,
4474 and the local variables of the function being called.
4475 The information is saved in a block of data called a @dfn{stack frame}.
4476 The stack frames are allocated in a region of memory called the @dfn{call
4479 When your program stops, the @value{GDBN} commands for examining the
4480 stack allow you to see all of this information.
4482 @cindex selected frame
4483 One of the stack frames is @dfn{selected} by @value{GDBN} and many
4484 @value{GDBN} commands refer implicitly to the selected frame. In
4485 particular, whenever you ask @value{GDBN} for the value of a variable in
4486 your program, the value is found in the selected frame. There are
4487 special @value{GDBN} commands to select whichever frame you are
4488 interested in. @xref{Selection, ,Selecting a Frame}.
4490 When your program stops, @value{GDBN} automatically selects the
4491 currently executing frame and describes it briefly, similar to the
4492 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
4495 * Frames:: Stack frames
4496 * Backtrace:: Backtraces
4497 * Selection:: Selecting a frame
4498 * Frame Info:: Information on a frame
4503 @section Stack Frames
4505 @cindex frame, definition
4507 The call stack is divided up into contiguous pieces called @dfn{stack
4508 frames}, or @dfn{frames} for short; each frame is the data associated
4509 with one call to one function. The frame contains the arguments given
4510 to the function, the function's local variables, and the address at
4511 which the function is executing.
4513 @cindex initial frame
4514 @cindex outermost frame
4515 @cindex innermost frame
4516 When your program is started, the stack has only one frame, that of the
4517 function @code{main}. This is called the @dfn{initial} frame or the
4518 @dfn{outermost} frame. Each time a function is called, a new frame is
4519 made. Each time a function returns, the frame for that function invocation
4520 is eliminated. If a function is recursive, there can be many frames for
4521 the same function. The frame for the function in which execution is
4522 actually occurring is called the @dfn{innermost} frame. This is the most
4523 recently created of all the stack frames that still exist.
4525 @cindex frame pointer
4526 Inside your program, stack frames are identified by their addresses. A
4527 stack frame consists of many bytes, each of which has its own address; each
4528 kind of computer has a convention for choosing one byte whose
4529 address serves as the address of the frame. Usually this address is kept
4530 in a register called the @dfn{frame pointer register}
4531 (@pxref{Registers, $fp}) while execution is going on in that frame.
4533 @cindex frame number
4534 @value{GDBN} assigns numbers to all existing stack frames, starting with
4535 zero for the innermost frame, one for the frame that called it,
4536 and so on upward. These numbers do not really exist in your program;
4537 they are assigned by @value{GDBN} to give you a way of designating stack
4538 frames in @value{GDBN} commands.
4540 @c The -fomit-frame-pointer below perennially causes hbox overflow
4541 @c underflow problems.
4542 @cindex frameless execution
4543 Some compilers provide a way to compile functions so that they operate
4544 without stack frames. (For example, the @value{NGCC} option
4546 @samp{-fomit-frame-pointer}
4548 generates functions without a frame.)
4549 This is occasionally done with heavily used library functions to save
4550 the frame setup time. @value{GDBN} has limited facilities for dealing
4551 with these function invocations. If the innermost function invocation
4552 has no stack frame, @value{GDBN} nevertheless regards it as though
4553 it had a separate frame, which is numbered zero as usual, allowing
4554 correct tracing of the function call chain. However, @value{GDBN} has
4555 no provision for frameless functions elsewhere in the stack.
4558 @kindex frame@r{, command}
4559 @cindex current stack frame
4560 @item frame @var{args}
4561 The @code{frame} command allows you to move from one stack frame to another,
4562 and to print the stack frame you select. @var{args} may be either the
4563 address of the frame or the stack frame number. Without an argument,
4564 @code{frame} prints the current stack frame.
4566 @kindex select-frame
4567 @cindex selecting frame silently
4569 The @code{select-frame} command allows you to move from one stack frame
4570 to another without printing the frame. This is the silent version of
4578 @cindex call stack traces
4579 A backtrace is a summary of how your program got where it is. It shows one
4580 line per frame, for many frames, starting with the currently executing
4581 frame (frame zero), followed by its caller (frame one), and on up the
4586 @kindex bt @r{(@code{backtrace})}
4589 Print a backtrace of the entire stack: one line per frame for all
4590 frames in the stack.
4592 You can stop the backtrace at any time by typing the system interrupt
4593 character, normally @kbd{Ctrl-c}.
4595 @item backtrace @var{n}
4597 Similar, but print only the innermost @var{n} frames.
4599 @item backtrace -@var{n}
4601 Similar, but print only the outermost @var{n} frames.
4603 @item backtrace full
4605 @itemx bt full @var{n}
4606 @itemx bt full -@var{n}
4607 Print the values of the local variables also. @var{n} specifies the
4608 number of frames to print, as described above.
4613 The names @code{where} and @code{info stack} (abbreviated @code{info s})
4614 are additional aliases for @code{backtrace}.
4616 @cindex multiple threads, backtrace
4617 In a multi-threaded program, @value{GDBN} by default shows the
4618 backtrace only for the current thread. To display the backtrace for
4619 several or all of the threads, use the command @code{thread apply}
4620 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
4621 apply all backtrace}, @value{GDBN} will display the backtrace for all
4622 the threads; this is handy when you debug a core dump of a
4623 multi-threaded program.
4625 Each line in the backtrace shows the frame number and the function name.
4626 The program counter value is also shown---unless you use @code{set
4627 print address off}. The backtrace also shows the source file name and
4628 line number, as well as the arguments to the function. The program
4629 counter value is omitted if it is at the beginning of the code for that
4632 Here is an example of a backtrace. It was made with the command
4633 @samp{bt 3}, so it shows the innermost three frames.
4637 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
4639 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
4640 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
4642 (More stack frames follow...)
4647 The display for frame zero does not begin with a program counter
4648 value, indicating that your program has stopped at the beginning of the
4649 code for line @code{993} of @code{builtin.c}.
4651 @cindex value optimized out, in backtrace
4652 @cindex function call arguments, optimized out
4653 If your program was compiled with optimizations, some compilers will
4654 optimize away arguments passed to functions if those arguments are
4655 never used after the call. Such optimizations generate code that
4656 passes arguments through registers, but doesn't store those arguments
4657 in the stack frame. @value{GDBN} has no way of displaying such
4658 arguments in stack frames other than the innermost one. Here's what
4659 such a backtrace might look like:
4663 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
4665 #1 0x6e38 in expand_macro (sym=<value optimized out>) at macro.c:242
4666 #2 0x6840 in expand_token (obs=0x0, t=<value optimized out>, td=0xf7fffb08)
4668 (More stack frames follow...)
4673 The values of arguments that were not saved in their stack frames are
4674 shown as @samp{<value optimized out>}.
4676 If you need to display the values of such optimized-out arguments,
4677 either deduce that from other variables whose values depend on the one
4678 you are interested in, or recompile without optimizations.
4680 @cindex backtrace beyond @code{main} function
4681 @cindex program entry point
4682 @cindex startup code, and backtrace
4683 Most programs have a standard user entry point---a place where system
4684 libraries and startup code transition into user code. For C this is
4685 @code{main}@footnote{
4686 Note that embedded programs (the so-called ``free-standing''
4687 environment) are not required to have a @code{main} function as the
4688 entry point. They could even have multiple entry points.}.
4689 When @value{GDBN} finds the entry function in a backtrace
4690 it will terminate the backtrace, to avoid tracing into highly
4691 system-specific (and generally uninteresting) code.
4693 If you need to examine the startup code, or limit the number of levels
4694 in a backtrace, you can change this behavior:
4697 @item set backtrace past-main
4698 @itemx set backtrace past-main on
4699 @kindex set backtrace
4700 Backtraces will continue past the user entry point.
4702 @item set backtrace past-main off
4703 Backtraces will stop when they encounter the user entry point. This is the
4706 @item show backtrace past-main
4707 @kindex show backtrace
4708 Display the current user entry point backtrace policy.
4710 @item set backtrace past-entry
4711 @itemx set backtrace past-entry on
4712 Backtraces will continue past the internal entry point of an application.
4713 This entry point is encoded by the linker when the application is built,
4714 and is likely before the user entry point @code{main} (or equivalent) is called.
4716 @item set backtrace past-entry off
4717 Backtraces will stop when they encounter the internal entry point of an
4718 application. This is the default.
4720 @item show backtrace past-entry
4721 Display the current internal entry point backtrace policy.
4723 @item set backtrace limit @var{n}
4724 @itemx set backtrace limit 0
4725 @cindex backtrace limit
4726 Limit the backtrace to @var{n} levels. A value of zero means
4729 @item show backtrace limit
4730 Display the current limit on backtrace levels.
4734 @section Selecting a Frame
4736 Most commands for examining the stack and other data in your program work on
4737 whichever stack frame is selected at the moment. Here are the commands for
4738 selecting a stack frame; all of them finish by printing a brief description
4739 of the stack frame just selected.
4742 @kindex frame@r{, selecting}
4743 @kindex f @r{(@code{frame})}
4746 Select frame number @var{n}. Recall that frame zero is the innermost
4747 (currently executing) frame, frame one is the frame that called the
4748 innermost one, and so on. The highest-numbered frame is the one for
4751 @item frame @var{addr}
4753 Select the frame at address @var{addr}. This is useful mainly if the
4754 chaining of stack frames has been damaged by a bug, making it
4755 impossible for @value{GDBN} to assign numbers properly to all frames. In
4756 addition, this can be useful when your program has multiple stacks and
4757 switches between them.
4759 On the SPARC architecture, @code{frame} needs two addresses to
4760 select an arbitrary frame: a frame pointer and a stack pointer.
4762 On the MIPS and Alpha architecture, it needs two addresses: a stack
4763 pointer and a program counter.
4765 On the 29k architecture, it needs three addresses: a register stack
4766 pointer, a program counter, and a memory stack pointer.
4770 Move @var{n} frames up the stack. For positive numbers @var{n}, this
4771 advances toward the outermost frame, to higher frame numbers, to frames
4772 that have existed longer. @var{n} defaults to one.
4775 @kindex do @r{(@code{down})}
4777 Move @var{n} frames down the stack. For positive numbers @var{n}, this
4778 advances toward the innermost frame, to lower frame numbers, to frames
4779 that were created more recently. @var{n} defaults to one. You may
4780 abbreviate @code{down} as @code{do}.
4783 All of these commands end by printing two lines of output describing the
4784 frame. The first line shows the frame number, the function name, the
4785 arguments, and the source file and line number of execution in that
4786 frame. The second line shows the text of that source line.
4794 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
4796 10 read_input_file (argv[i]);
4800 After such a printout, the @code{list} command with no arguments
4801 prints ten lines centered on the point of execution in the frame.
4802 You can also edit the program at the point of execution with your favorite
4803 editing program by typing @code{edit}.
4804 @xref{List, ,Printing Source Lines},
4808 @kindex down-silently
4810 @item up-silently @var{n}
4811 @itemx down-silently @var{n}
4812 These two commands are variants of @code{up} and @code{down},
4813 respectively; they differ in that they do their work silently, without
4814 causing display of the new frame. They are intended primarily for use
4815 in @value{GDBN} command scripts, where the output might be unnecessary and
4820 @section Information About a Frame
4822 There are several other commands to print information about the selected
4828 When used without any argument, this command does not change which
4829 frame is selected, but prints a brief description of the currently
4830 selected stack frame. It can be abbreviated @code{f}. With an
4831 argument, this command is used to select a stack frame.
4832 @xref{Selection, ,Selecting a Frame}.
4835 @kindex info f @r{(@code{info frame})}
4838 This command prints a verbose description of the selected stack frame,
4843 the address of the frame
4845 the address of the next frame down (called by this frame)
4847 the address of the next frame up (caller of this frame)
4849 the language in which the source code corresponding to this frame is written
4851 the address of the frame's arguments
4853 the address of the frame's local variables
4855 the program counter saved in it (the address of execution in the caller frame)
4857 which registers were saved in the frame
4860 @noindent The verbose description is useful when
4861 something has gone wrong that has made the stack format fail to fit
4862 the usual conventions.
4864 @item info frame @var{addr}
4865 @itemx info f @var{addr}
4866 Print a verbose description of the frame at address @var{addr}, without
4867 selecting that frame. The selected frame remains unchanged by this
4868 command. This requires the same kind of address (more than one for some
4869 architectures) that you specify in the @code{frame} command.
4870 @xref{Selection, ,Selecting a Frame}.
4874 Print the arguments of the selected frame, each on a separate line.
4878 Print the local variables of the selected frame, each on a separate
4879 line. These are all variables (declared either static or automatic)
4880 accessible at the point of execution of the selected frame.
4883 @cindex catch exceptions, list active handlers
4884 @cindex exception handlers, how to list
4886 Print a list of all the exception handlers that are active in the
4887 current stack frame at the current point of execution. To see other
4888 exception handlers, visit the associated frame (using the @code{up},
4889 @code{down}, or @code{frame} commands); then type @code{info catch}.
4890 @xref{Set Catchpoints, , Setting Catchpoints}.
4896 @chapter Examining Source Files
4898 @value{GDBN} can print parts of your program's source, since the debugging
4899 information recorded in the program tells @value{GDBN} what source files were
4900 used to build it. When your program stops, @value{GDBN} spontaneously prints
4901 the line where it stopped. Likewise, when you select a stack frame
4902 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
4903 execution in that frame has stopped. You can print other portions of
4904 source files by explicit command.
4906 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
4907 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
4908 @value{GDBN} under @sc{gnu} Emacs}.
4911 * List:: Printing source lines
4912 * Specify Location:: How to specify code locations
4913 * Edit:: Editing source files
4914 * Search:: Searching source files
4915 * Source Path:: Specifying source directories
4916 * Machine Code:: Source and machine code
4920 @section Printing Source Lines
4923 @kindex l @r{(@code{list})}
4924 To print lines from a source file, use the @code{list} command
4925 (abbreviated @code{l}). By default, ten lines are printed.
4926 There are several ways to specify what part of the file you want to
4927 print; see @ref{Specify Location}, for the full list.
4929 Here are the forms of the @code{list} command most commonly used:
4932 @item list @var{linenum}
4933 Print lines centered around line number @var{linenum} in the
4934 current source file.
4936 @item list @var{function}
4937 Print lines centered around the beginning of function
4941 Print more lines. If the last lines printed were printed with a
4942 @code{list} command, this prints lines following the last lines
4943 printed; however, if the last line printed was a solitary line printed
4944 as part of displaying a stack frame (@pxref{Stack, ,Examining the
4945 Stack}), this prints lines centered around that line.
4948 Print lines just before the lines last printed.
4951 @cindex @code{list}, how many lines to display
4952 By default, @value{GDBN} prints ten source lines with any of these forms of
4953 the @code{list} command. You can change this using @code{set listsize}:
4956 @kindex set listsize
4957 @item set listsize @var{count}
4958 Make the @code{list} command display @var{count} source lines (unless
4959 the @code{list} argument explicitly specifies some other number).
4961 @kindex show listsize
4963 Display the number of lines that @code{list} prints.
4966 Repeating a @code{list} command with @key{RET} discards the argument,
4967 so it is equivalent to typing just @code{list}. This is more useful
4968 than listing the same lines again. An exception is made for an
4969 argument of @samp{-}; that argument is preserved in repetition so that
4970 each repetition moves up in the source file.
4972 In general, the @code{list} command expects you to supply zero, one or two
4973 @dfn{linespecs}. Linespecs specify source lines; there are several ways
4974 of writing them (@pxref{Specify Location}), but the effect is always
4975 to specify some source line.
4977 Here is a complete description of the possible arguments for @code{list}:
4980 @item list @var{linespec}
4981 Print lines centered around the line specified by @var{linespec}.
4983 @item list @var{first},@var{last}
4984 Print lines from @var{first} to @var{last}. Both arguments are
4985 linespecs. When a @code{list} command has two linespecs, and the
4986 source file of the second linespec is omitted, this refers to
4987 the same source file as the first linespec.
4989 @item list ,@var{last}
4990 Print lines ending with @var{last}.
4992 @item list @var{first},
4993 Print lines starting with @var{first}.
4996 Print lines just after the lines last printed.
4999 Print lines just before the lines last printed.
5002 As described in the preceding table.
5005 @node Specify Location
5006 @section Specifying a Location
5007 @cindex specifying location
5010 Several @value{GDBN} commands accept arguments that specify a location
5011 of your program's code. Since @value{GDBN} is a source-level
5012 debugger, a location usually specifies some line in the source code;
5013 for that reason, locations are also known as @dfn{linespecs}.
5015 Here are all the different ways of specifying a code location that
5016 @value{GDBN} understands:
5020 Specifies the line number @var{linenum} of the current source file.
5023 @itemx +@var{offset}
5024 Specifies the line @var{offset} lines before or after the @dfn{current
5025 line}. For the @code{list} command, the current line is the last one
5026 printed; for the breakpoint commands, this is the line at which
5027 execution stopped in the currently selected @dfn{stack frame}
5028 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
5029 used as the second of the two linespecs in a @code{list} command,
5030 this specifies the line @var{offset} lines up or down from the first
5033 @item @var{filename}:@var{linenum}
5034 Specifies the line @var{linenum} in the source file @var{filename}.
5036 @item @var{function}
5037 Specifies the line that begins the body of the function @var{function}.
5038 For example, in C, this is the line with the open brace.
5040 @item @var{filename}:@var{function}
5041 Specifies the line that begins the body of the function @var{function}
5042 in the file @var{filename}. You only need the file name with a
5043 function name to avoid ambiguity when there are identically named
5044 functions in different source files.
5046 @item *@var{address}
5047 Specifies the program address @var{address}. For line-oriented
5048 commands, such as @code{list} and @code{edit}, this specifies a source
5049 line that contains @var{address}. For @code{break} and other
5050 breakpoint oriented commands, this can be used to set breakpoints in
5051 parts of your program which do not have debugging information or
5054 Here @var{address} may be any expression valid in the current working
5055 language (@pxref{Languages, working language}) that specifies a code
5056 address. In addition, as a convenience, @value{GDBN} extends the
5057 semantics of expressions used in locations to cover the situations
5058 that frequently happen during debugging. Here are the various forms
5062 @item @var{expression}
5063 Any expression valid in the current working language.
5065 @item @var{funcaddr}
5066 An address of a function or procedure derived from its name. In C,
5067 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
5068 simply the function's name @var{function} (and actually a special case
5069 of a valid expression). In Pascal and Modula-2, this is
5070 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
5071 (although the Pascal form also works).
5073 This form specifies the address of the function's first instruction,
5074 before the stack frame and arguments have been set up.
5076 @item '@var{filename}'::@var{funcaddr}
5077 Like @var{funcaddr} above, but also specifies the name of the source
5078 file explicitly. This is useful if the name of the function does not
5079 specify the function unambiguously, e.g., if there are several
5080 functions with identical names in different source files.
5087 @section Editing Source Files
5088 @cindex editing source files
5091 @kindex e @r{(@code{edit})}
5092 To edit the lines in a source file, use the @code{edit} command.
5093 The editing program of your choice
5094 is invoked with the current line set to
5095 the active line in the program.
5096 Alternatively, there are several ways to specify what part of the file you
5097 want to print if you want to see other parts of the program:
5100 @item edit @var{location}
5101 Edit the source file specified by @code{location}. Editing starts at
5102 that @var{location}, e.g., at the specified source line of the
5103 specified file. @xref{Specify Location}, for all the possible forms
5104 of the @var{location} argument; here are the forms of the @code{edit}
5105 command most commonly used:
5108 @item edit @var{number}
5109 Edit the current source file with @var{number} as the active line number.
5111 @item edit @var{function}
5112 Edit the file containing @var{function} at the beginning of its definition.
5117 @subsection Choosing your Editor
5118 You can customize @value{GDBN} to use any editor you want
5120 The only restriction is that your editor (say @code{ex}), recognizes the
5121 following command-line syntax:
5123 ex +@var{number} file
5125 The optional numeric value +@var{number} specifies the number of the line in
5126 the file where to start editing.}.
5127 By default, it is @file{@value{EDITOR}}, but you can change this
5128 by setting the environment variable @code{EDITOR} before using
5129 @value{GDBN}. For example, to configure @value{GDBN} to use the
5130 @code{vi} editor, you could use these commands with the @code{sh} shell:
5136 or in the @code{csh} shell,
5138 setenv EDITOR /usr/bin/vi
5143 @section Searching Source Files
5144 @cindex searching source files
5146 There are two commands for searching through the current source file for a
5151 @kindex forward-search
5152 @item forward-search @var{regexp}
5153 @itemx search @var{regexp}
5154 The command @samp{forward-search @var{regexp}} checks each line,
5155 starting with the one following the last line listed, for a match for
5156 @var{regexp}. It lists the line that is found. You can use the
5157 synonym @samp{search @var{regexp}} or abbreviate the command name as
5160 @kindex reverse-search
5161 @item reverse-search @var{regexp}
5162 The command @samp{reverse-search @var{regexp}} checks each line, starting
5163 with the one before the last line listed and going backward, for a match
5164 for @var{regexp}. It lists the line that is found. You can abbreviate
5165 this command as @code{rev}.
5169 @section Specifying Source Directories
5172 @cindex directories for source files
5173 Executable programs sometimes do not record the directories of the source
5174 files from which they were compiled, just the names. Even when they do,
5175 the directories could be moved between the compilation and your debugging
5176 session. @value{GDBN} has a list of directories to search for source files;
5177 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
5178 it tries all the directories in the list, in the order they are present
5179 in the list, until it finds a file with the desired name.
5181 For example, suppose an executable references the file
5182 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
5183 @file{/mnt/cross}. The file is first looked up literally; if this
5184 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
5185 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
5186 message is printed. @value{GDBN} does not look up the parts of the
5187 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
5188 Likewise, the subdirectories of the source path are not searched: if
5189 the source path is @file{/mnt/cross}, and the binary refers to
5190 @file{foo.c}, @value{GDBN} would not find it under
5191 @file{/mnt/cross/usr/src/foo-1.0/lib}.
5193 Plain file names, relative file names with leading directories, file
5194 names containing dots, etc.@: are all treated as described above; for
5195 instance, if the source path is @file{/mnt/cross}, and the source file
5196 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
5197 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
5198 that---@file{/mnt/cross/foo.c}.
5200 Note that the executable search path is @emph{not} used to locate the
5203 Whenever you reset or rearrange the source path, @value{GDBN} clears out
5204 any information it has cached about where source files are found and where
5205 each line is in the file.
5209 When you start @value{GDBN}, its source path includes only @samp{cdir}
5210 and @samp{cwd}, in that order.
5211 To add other directories, use the @code{directory} command.
5213 The search path is used to find both program source files and @value{GDBN}
5214 script files (read using the @samp{-command} option and @samp{source} command).
5216 In addition to the source path, @value{GDBN} provides a set of commands
5217 that manage a list of source path substitution rules. A @dfn{substitution
5218 rule} specifies how to rewrite source directories stored in the program's
5219 debug information in case the sources were moved to a different
5220 directory between compilation and debugging. A rule is made of
5221 two strings, the first specifying what needs to be rewritten in
5222 the path, and the second specifying how it should be rewritten.
5223 In @ref{set substitute-path}, we name these two parts @var{from} and
5224 @var{to} respectively. @value{GDBN} does a simple string replacement
5225 of @var{from} with @var{to} at the start of the directory part of the
5226 source file name, and uses that result instead of the original file
5227 name to look up the sources.
5229 Using the previous example, suppose the @file{foo-1.0} tree has been
5230 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
5231 @value{GDBN} to replace @file{/usr/src} in all source path names with
5232 @file{/mnt/cross}. The first lookup will then be
5233 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
5234 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
5235 substitution rule, use the @code{set substitute-path} command
5236 (@pxref{set substitute-path}).
5238 To avoid unexpected substitution results, a rule is applied only if the
5239 @var{from} part of the directory name ends at a directory separator.
5240 For instance, a rule substituting @file{/usr/source} into
5241 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
5242 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
5243 is applied only at the beginning of the directory name, this rule will
5244 not be applied to @file{/root/usr/source/baz.c} either.
5246 In many cases, you can achieve the same result using the @code{directory}
5247 command. However, @code{set substitute-path} can be more efficient in
5248 the case where the sources are organized in a complex tree with multiple
5249 subdirectories. With the @code{directory} command, you need to add each
5250 subdirectory of your project. If you moved the entire tree while
5251 preserving its internal organization, then @code{set substitute-path}
5252 allows you to direct the debugger to all the sources with one single
5255 @code{set substitute-path} is also more than just a shortcut command.
5256 The source path is only used if the file at the original location no
5257 longer exists. On the other hand, @code{set substitute-path} modifies
5258 the debugger behavior to look at the rewritten location instead. So, if
5259 for any reason a source file that is not relevant to your executable is
5260 located at the original location, a substitution rule is the only
5261 method available to point @value{GDBN} at the new location.
5264 @item directory @var{dirname} @dots{}
5265 @item dir @var{dirname} @dots{}
5266 Add directory @var{dirname} to the front of the source path. Several
5267 directory names may be given to this command, separated by @samp{:}
5268 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
5269 part of absolute file names) or
5270 whitespace. You may specify a directory that is already in the source
5271 path; this moves it forward, so @value{GDBN} searches it sooner.
5275 @vindex $cdir@r{, convenience variable}
5276 @vindex $cwd@r{, convenience variable}
5277 @cindex compilation directory
5278 @cindex current directory
5279 @cindex working directory
5280 @cindex directory, current
5281 @cindex directory, compilation
5282 You can use the string @samp{$cdir} to refer to the compilation
5283 directory (if one is recorded), and @samp{$cwd} to refer to the current
5284 working directory. @samp{$cwd} is not the same as @samp{.}---the former
5285 tracks the current working directory as it changes during your @value{GDBN}
5286 session, while the latter is immediately expanded to the current
5287 directory at the time you add an entry to the source path.
5290 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
5292 @c RET-repeat for @code{directory} is explicitly disabled, but since
5293 @c repeating it would be a no-op we do not say that. (thanks to RMS)
5295 @item show directories
5296 @kindex show directories
5297 Print the source path: show which directories it contains.
5299 @anchor{set substitute-path}
5300 @item set substitute-path @var{from} @var{to}
5301 @kindex set substitute-path
5302 Define a source path substitution rule, and add it at the end of the
5303 current list of existing substitution rules. If a rule with the same
5304 @var{from} was already defined, then the old rule is also deleted.
5306 For example, if the file @file{/foo/bar/baz.c} was moved to
5307 @file{/mnt/cross/baz.c}, then the command
5310 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
5314 will tell @value{GDBN} to replace @samp{/usr/src} with
5315 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
5316 @file{baz.c} even though it was moved.
5318 In the case when more than one substitution rule have been defined,
5319 the rules are evaluated one by one in the order where they have been
5320 defined. The first one matching, if any, is selected to perform
5323 For instance, if we had entered the following commands:
5326 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
5327 (@value{GDBP}) set substitute-path /usr/src /mnt/src
5331 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
5332 @file{/mnt/include/defs.h} by using the first rule. However, it would
5333 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
5334 @file{/mnt/src/lib/foo.c}.
5337 @item unset substitute-path [path]
5338 @kindex unset substitute-path
5339 If a path is specified, search the current list of substitution rules
5340 for a rule that would rewrite that path. Delete that rule if found.
5341 A warning is emitted by the debugger if no rule could be found.
5343 If no path is specified, then all substitution rules are deleted.
5345 @item show substitute-path [path]
5346 @kindex show substitute-path
5347 If a path is specified, then print the source path substitution rule
5348 which would rewrite that path, if any.
5350 If no path is specified, then print all existing source path substitution
5355 If your source path is cluttered with directories that are no longer of
5356 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
5357 versions of source. You can correct the situation as follows:
5361 Use @code{directory} with no argument to reset the source path to its default value.
5364 Use @code{directory} with suitable arguments to reinstall the
5365 directories you want in the source path. You can add all the
5366 directories in one command.
5370 @section Source and Machine Code
5371 @cindex source line and its code address
5373 You can use the command @code{info line} to map source lines to program
5374 addresses (and vice versa), and the command @code{disassemble} to display
5375 a range of addresses as machine instructions. When run under @sc{gnu} Emacs
5376 mode, the @code{info line} command causes the arrow to point to the
5377 line specified. Also, @code{info line} prints addresses in symbolic form as
5382 @item info line @var{linespec}
5383 Print the starting and ending addresses of the compiled code for
5384 source line @var{linespec}. You can specify source lines in any of
5385 the ways documented in @ref{Specify Location}.
5388 For example, we can use @code{info line} to discover the location of
5389 the object code for the first line of function
5390 @code{m4_changequote}:
5392 @c FIXME: I think this example should also show the addresses in
5393 @c symbolic form, as they usually would be displayed.
5395 (@value{GDBP}) info line m4_changequote
5396 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
5400 @cindex code address and its source line
5401 We can also inquire (using @code{*@var{addr}} as the form for
5402 @var{linespec}) what source line covers a particular address:
5404 (@value{GDBP}) info line *0x63ff
5405 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
5408 @cindex @code{$_} and @code{info line}
5409 @cindex @code{x} command, default address
5410 @kindex x@r{(examine), and} info line
5411 After @code{info line}, the default address for the @code{x} command
5412 is changed to the starting address of the line, so that @samp{x/i} is
5413 sufficient to begin examining the machine code (@pxref{Memory,
5414 ,Examining Memory}). Also, this address is saved as the value of the
5415 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
5420 @cindex assembly instructions
5421 @cindex instructions, assembly
5422 @cindex machine instructions
5423 @cindex listing machine instructions
5425 This specialized command dumps a range of memory as machine
5426 instructions. The default memory range is the function surrounding the
5427 program counter of the selected frame. A single argument to this
5428 command is a program counter value; @value{GDBN} dumps the function
5429 surrounding this value. Two arguments specify a range of addresses
5430 (first inclusive, second exclusive) to dump.
5433 The following example shows the disassembly of a range of addresses of
5434 HP PA-RISC 2.0 code:
5437 (@value{GDBP}) disas 0x32c4 0x32e4
5438 Dump of assembler code from 0x32c4 to 0x32e4:
5439 0x32c4 <main+204>: addil 0,dp
5440 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
5441 0x32cc <main+212>: ldil 0x3000,r31
5442 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
5443 0x32d4 <main+220>: ldo 0(r31),rp
5444 0x32d8 <main+224>: addil -0x800,dp
5445 0x32dc <main+228>: ldo 0x588(r1),r26
5446 0x32e0 <main+232>: ldil 0x3000,r31
5447 End of assembler dump.
5450 Some architectures have more than one commonly-used set of instruction
5451 mnemonics or other syntax.
5453 For programs that were dynamically linked and use shared libraries,
5454 instructions that call functions or branch to locations in the shared
5455 libraries might show a seemingly bogus location---it's actually a
5456 location of the relocation table. On some architectures, @value{GDBN}
5457 might be able to resolve these to actual function names.
5460 @kindex set disassembly-flavor
5461 @cindex Intel disassembly flavor
5462 @cindex AT&T disassembly flavor
5463 @item set disassembly-flavor @var{instruction-set}
5464 Select the instruction set to use when disassembling the
5465 program via the @code{disassemble} or @code{x/i} commands.
5467 Currently this command is only defined for the Intel x86 family. You
5468 can set @var{instruction-set} to either @code{intel} or @code{att}.
5469 The default is @code{att}, the AT&T flavor used by default by Unix
5470 assemblers for x86-based targets.
5472 @kindex show disassembly-flavor
5473 @item show disassembly-flavor
5474 Show the current setting of the disassembly flavor.
5479 @chapter Examining Data
5481 @cindex printing data
5482 @cindex examining data
5485 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
5486 @c document because it is nonstandard... Under Epoch it displays in a
5487 @c different window or something like that.
5488 The usual way to examine data in your program is with the @code{print}
5489 command (abbreviated @code{p}), or its synonym @code{inspect}. It
5490 evaluates and prints the value of an expression of the language your
5491 program is written in (@pxref{Languages, ,Using @value{GDBN} with
5492 Different Languages}).
5495 @item print @var{expr}
5496 @itemx print /@var{f} @var{expr}
5497 @var{expr} is an expression (in the source language). By default the
5498 value of @var{expr} is printed in a format appropriate to its data type;
5499 you can choose a different format by specifying @samp{/@var{f}}, where
5500 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
5504 @itemx print /@var{f}
5505 @cindex reprint the last value
5506 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
5507 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
5508 conveniently inspect the same value in an alternative format.
5511 A more low-level way of examining data is with the @code{x} command.
5512 It examines data in memory at a specified address and prints it in a
5513 specified format. @xref{Memory, ,Examining Memory}.
5515 If you are interested in information about types, or about how the
5516 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
5517 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
5521 * Expressions:: Expressions
5522 * Ambiguous Expressions:: Ambiguous Expressions
5523 * Variables:: Program variables
5524 * Arrays:: Artificial arrays
5525 * Output Formats:: Output formats
5526 * Memory:: Examining memory
5527 * Auto Display:: Automatic display
5528 * Print Settings:: Print settings
5529 * Value History:: Value history
5530 * Convenience Vars:: Convenience variables
5531 * Registers:: Registers
5532 * Floating Point Hardware:: Floating point hardware
5533 * Vector Unit:: Vector Unit
5534 * OS Information:: Auxiliary data provided by operating system
5535 * Memory Region Attributes:: Memory region attributes
5536 * Dump/Restore Files:: Copy between memory and a file
5537 * Core File Generation:: Cause a program dump its core
5538 * Character Sets:: Debugging programs that use a different
5539 character set than GDB does
5540 * Caching Remote Data:: Data caching for remote targets
5544 @section Expressions
5547 @code{print} and many other @value{GDBN} commands accept an expression and
5548 compute its value. Any kind of constant, variable or operator defined
5549 by the programming language you are using is valid in an expression in
5550 @value{GDBN}. This includes conditional expressions, function calls,
5551 casts, and string constants. It also includes preprocessor macros, if
5552 you compiled your program to include this information; see
5555 @cindex arrays in expressions
5556 @value{GDBN} supports array constants in expressions input by
5557 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
5558 you can use the command @code{print @{1, 2, 3@}} to create an array
5559 of three integers. If you pass an array to a function or assign it
5560 to a program variable, @value{GDBN} copies the array to memory that
5561 is @code{malloc}ed in the target program.
5563 Because C is so widespread, most of the expressions shown in examples in
5564 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
5565 Languages}, for information on how to use expressions in other
5568 In this section, we discuss operators that you can use in @value{GDBN}
5569 expressions regardless of your programming language.
5571 @cindex casts, in expressions
5572 Casts are supported in all languages, not just in C, because it is so
5573 useful to cast a number into a pointer in order to examine a structure
5574 at that address in memory.
5575 @c FIXME: casts supported---Mod2 true?
5577 @value{GDBN} supports these operators, in addition to those common
5578 to programming languages:
5582 @samp{@@} is a binary operator for treating parts of memory as arrays.
5583 @xref{Arrays, ,Artificial Arrays}, for more information.
5586 @samp{::} allows you to specify a variable in terms of the file or
5587 function where it is defined. @xref{Variables, ,Program Variables}.
5589 @cindex @{@var{type}@}
5590 @cindex type casting memory
5591 @cindex memory, viewing as typed object
5592 @cindex casts, to view memory
5593 @item @{@var{type}@} @var{addr}
5594 Refers to an object of type @var{type} stored at address @var{addr} in
5595 memory. @var{addr} may be any expression whose value is an integer or
5596 pointer (but parentheses are required around binary operators, just as in
5597 a cast). This construct is allowed regardless of what kind of data is
5598 normally supposed to reside at @var{addr}.
5601 @node Ambiguous Expressions
5602 @section Ambiguous Expressions
5603 @cindex ambiguous expressions
5605 Expressions can sometimes contain some ambiguous elements. For instance,
5606 some programming languages (notably Ada, C@t{++} and Objective-C) permit
5607 a single function name to be defined several times, for application in
5608 different contexts. This is called @dfn{overloading}. Another example
5609 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
5610 templates and is typically instantiated several times, resulting in
5611 the same function name being defined in different contexts.
5613 In some cases and depending on the language, it is possible to adjust
5614 the expression to remove the ambiguity. For instance in C@t{++}, you
5615 can specify the signature of the function you want to break on, as in
5616 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
5617 qualified name of your function often makes the expression unambiguous
5620 When an ambiguity that needs to be resolved is detected, the debugger
5621 has the capability to display a menu of numbered choices for each
5622 possibility, and then waits for the selection with the prompt @samp{>}.
5623 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
5624 aborts the current command. If the command in which the expression was
5625 used allows more than one choice to be selected, the next option in the
5626 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
5629 For example, the following session excerpt shows an attempt to set a
5630 breakpoint at the overloaded symbol @code{String::after}.
5631 We choose three particular definitions of that function name:
5633 @c FIXME! This is likely to change to show arg type lists, at least
5636 (@value{GDBP}) b String::after
5639 [2] file:String.cc; line number:867
5640 [3] file:String.cc; line number:860
5641 [4] file:String.cc; line number:875
5642 [5] file:String.cc; line number:853
5643 [6] file:String.cc; line number:846
5644 [7] file:String.cc; line number:735
5646 Breakpoint 1 at 0xb26c: file String.cc, line 867.
5647 Breakpoint 2 at 0xb344: file String.cc, line 875.
5648 Breakpoint 3 at 0xafcc: file String.cc, line 846.
5649 Multiple breakpoints were set.
5650 Use the "delete" command to delete unwanted
5657 @kindex set multiple-symbols
5658 @item set multiple-symbols @var{mode}
5659 @cindex multiple-symbols menu
5661 This option allows you to adjust the debugger behavior when an expression
5664 By default, @var{mode} is set to @code{all}. If the command with which
5665 the expression is used allows more than one choice, then @value{GDBN}
5666 automatically selects all possible choices. For instance, inserting
5667 a breakpoint on a function using an ambiguous name results in a breakpoint
5668 inserted on each possible match. However, if a unique choice must be made,
5669 then @value{GDBN} uses the menu to help you disambiguate the expression.
5670 For instance, printing the address of an overloaded function will result
5671 in the use of the menu.
5673 When @var{mode} is set to @code{ask}, the debugger always uses the menu
5674 when an ambiguity is detected.
5676 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
5677 an error due to the ambiguity and the command is aborted.
5679 @kindex show multiple-symbols
5680 @item show multiple-symbols
5681 Show the current value of the @code{multiple-symbols} setting.
5685 @section Program Variables
5687 The most common kind of expression to use is the name of a variable
5690 Variables in expressions are understood in the selected stack frame
5691 (@pxref{Selection, ,Selecting a Frame}); they must be either:
5695 global (or file-static)
5702 visible according to the scope rules of the
5703 programming language from the point of execution in that frame
5706 @noindent This means that in the function
5721 you can examine and use the variable @code{a} whenever your program is
5722 executing within the function @code{foo}, but you can only use or
5723 examine the variable @code{b} while your program is executing inside
5724 the block where @code{b} is declared.
5726 @cindex variable name conflict
5727 There is an exception: you can refer to a variable or function whose
5728 scope is a single source file even if the current execution point is not
5729 in this file. But it is possible to have more than one such variable or
5730 function with the same name (in different source files). If that
5731 happens, referring to that name has unpredictable effects. If you wish,
5732 you can specify a static variable in a particular function or file,
5733 using the colon-colon (@code{::}) notation:
5735 @cindex colon-colon, context for variables/functions
5737 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
5738 @cindex @code{::}, context for variables/functions
5741 @var{file}::@var{variable}
5742 @var{function}::@var{variable}
5746 Here @var{file} or @var{function} is the name of the context for the
5747 static @var{variable}. In the case of file names, you can use quotes to
5748 make sure @value{GDBN} parses the file name as a single word---for example,
5749 to print a global value of @code{x} defined in @file{f2.c}:
5752 (@value{GDBP}) p 'f2.c'::x
5755 @cindex C@t{++} scope resolution
5756 This use of @samp{::} is very rarely in conflict with the very similar
5757 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
5758 scope resolution operator in @value{GDBN} expressions.
5759 @c FIXME: Um, so what happens in one of those rare cases where it's in
5762 @cindex wrong values
5763 @cindex variable values, wrong
5764 @cindex function entry/exit, wrong values of variables
5765 @cindex optimized code, wrong values of variables
5767 @emph{Warning:} Occasionally, a local variable may appear to have the
5768 wrong value at certain points in a function---just after entry to a new
5769 scope, and just before exit.
5771 You may see this problem when you are stepping by machine instructions.
5772 This is because, on most machines, it takes more than one instruction to
5773 set up a stack frame (including local variable definitions); if you are
5774 stepping by machine instructions, variables may appear to have the wrong
5775 values until the stack frame is completely built. On exit, it usually
5776 also takes more than one machine instruction to destroy a stack frame;
5777 after you begin stepping through that group of instructions, local
5778 variable definitions may be gone.
5780 This may also happen when the compiler does significant optimizations.
5781 To be sure of always seeing accurate values, turn off all optimization
5784 @cindex ``No symbol "foo" in current context''
5785 Another possible effect of compiler optimizations is to optimize
5786 unused variables out of existence, or assign variables to registers (as
5787 opposed to memory addresses). Depending on the support for such cases
5788 offered by the debug info format used by the compiler, @value{GDBN}
5789 might not be able to display values for such local variables. If that
5790 happens, @value{GDBN} will print a message like this:
5793 No symbol "foo" in current context.
5796 To solve such problems, either recompile without optimizations, or use a
5797 different debug info format, if the compiler supports several such
5798 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
5799 usually supports the @option{-gstabs+} option. @option{-gstabs+}
5800 produces debug info in a format that is superior to formats such as
5801 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
5802 an effective form for debug info. @xref{Debugging Options,,Options
5803 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
5804 Compiler Collection (GCC)}.
5805 @xref{C, ,C and C@t{++}}, for more information about debug info formats
5806 that are best suited to C@t{++} programs.
5808 If you ask to print an object whose contents are unknown to
5809 @value{GDBN}, e.g., because its data type is not completely specified
5810 by the debug information, @value{GDBN} will say @samp{<incomplete
5811 type>}. @xref{Symbols, incomplete type}, for more about this.
5813 Strings are identified as arrays of @code{char} values without specified
5814 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
5815 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
5816 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
5817 defines literal string type @code{"char"} as @code{char} without a sign.
5822 signed char var1[] = "A";
5825 You get during debugging
5830 $2 = @{65 'A', 0 '\0'@}
5834 @section Artificial Arrays
5836 @cindex artificial array
5838 @kindex @@@r{, referencing memory as an array}
5839 It is often useful to print out several successive objects of the
5840 same type in memory; a section of an array, or an array of
5841 dynamically determined size for which only a pointer exists in the
5844 You can do this by referring to a contiguous span of memory as an
5845 @dfn{artificial array}, using the binary operator @samp{@@}. The left
5846 operand of @samp{@@} should be the first element of the desired array
5847 and be an individual object. The right operand should be the desired length
5848 of the array. The result is an array value whose elements are all of
5849 the type of the left argument. The first element is actually the left
5850 argument; the second element comes from bytes of memory immediately
5851 following those that hold the first element, and so on. Here is an
5852 example. If a program says
5855 int *array = (int *) malloc (len * sizeof (int));
5859 you can print the contents of @code{array} with
5865 The left operand of @samp{@@} must reside in memory. Array values made
5866 with @samp{@@} in this way behave just like other arrays in terms of
5867 subscripting, and are coerced to pointers when used in expressions.
5868 Artificial arrays most often appear in expressions via the value history
5869 (@pxref{Value History, ,Value History}), after printing one out.
5871 Another way to create an artificial array is to use a cast.
5872 This re-interprets a value as if it were an array.
5873 The value need not be in memory:
5875 (@value{GDBP}) p/x (short[2])0x12345678
5876 $1 = @{0x1234, 0x5678@}
5879 As a convenience, if you leave the array length out (as in
5880 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
5881 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
5883 (@value{GDBP}) p/x (short[])0x12345678
5884 $2 = @{0x1234, 0x5678@}
5887 Sometimes the artificial array mechanism is not quite enough; in
5888 moderately complex data structures, the elements of interest may not
5889 actually be adjacent---for example, if you are interested in the values
5890 of pointers in an array. One useful work-around in this situation is
5891 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
5892 Variables}) as a counter in an expression that prints the first
5893 interesting value, and then repeat that expression via @key{RET}. For
5894 instance, suppose you have an array @code{dtab} of pointers to
5895 structures, and you are interested in the values of a field @code{fv}
5896 in each structure. Here is an example of what you might type:
5906 @node Output Formats
5907 @section Output Formats
5909 @cindex formatted output
5910 @cindex output formats
5911 By default, @value{GDBN} prints a value according to its data type. Sometimes
5912 this is not what you want. For example, you might want to print a number
5913 in hex, or a pointer in decimal. Or you might want to view data in memory
5914 at a certain address as a character string or as an instruction. To do
5915 these things, specify an @dfn{output format} when you print a value.
5917 The simplest use of output formats is to say how to print a value
5918 already computed. This is done by starting the arguments of the
5919 @code{print} command with a slash and a format letter. The format
5920 letters supported are:
5924 Regard the bits of the value as an integer, and print the integer in
5928 Print as integer in signed decimal.
5931 Print as integer in unsigned decimal.
5934 Print as integer in octal.
5937 Print as integer in binary. The letter @samp{t} stands for ``two''.
5938 @footnote{@samp{b} cannot be used because these format letters are also
5939 used with the @code{x} command, where @samp{b} stands for ``byte'';
5940 see @ref{Memory,,Examining Memory}.}
5943 @cindex unknown address, locating
5944 @cindex locate address
5945 Print as an address, both absolute in hexadecimal and as an offset from
5946 the nearest preceding symbol. You can use this format used to discover
5947 where (in what function) an unknown address is located:
5950 (@value{GDBP}) p/a 0x54320
5951 $3 = 0x54320 <_initialize_vx+396>
5955 The command @code{info symbol 0x54320} yields similar results.
5956 @xref{Symbols, info symbol}.
5959 Regard as an integer and print it as a character constant. This
5960 prints both the numerical value and its character representation. The
5961 character representation is replaced with the octal escape @samp{\nnn}
5962 for characters outside the 7-bit @sc{ascii} range.
5964 Without this format, @value{GDBN} displays @code{char},
5965 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
5966 constants. Single-byte members of vectors are displayed as integer
5970 Regard the bits of the value as a floating point number and print
5971 using typical floating point syntax.
5974 @cindex printing strings
5975 @cindex printing byte arrays
5976 Regard as a string, if possible. With this format, pointers to single-byte
5977 data are displayed as null-terminated strings and arrays of single-byte data
5978 are displayed as fixed-length strings. Other values are displayed in their
5981 Without this format, @value{GDBN} displays pointers to and arrays of
5982 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
5983 strings. Single-byte members of a vector are displayed as an integer
5987 For example, to print the program counter in hex (@pxref{Registers}), type
5994 Note that no space is required before the slash; this is because command
5995 names in @value{GDBN} cannot contain a slash.
5997 To reprint the last value in the value history with a different format,
5998 you can use the @code{print} command with just a format and no
5999 expression. For example, @samp{p/x} reprints the last value in hex.
6002 @section Examining Memory
6004 You can use the command @code{x} (for ``examine'') to examine memory in
6005 any of several formats, independently of your program's data types.
6007 @cindex examining memory
6009 @kindex x @r{(examine memory)}
6010 @item x/@var{nfu} @var{addr}
6013 Use the @code{x} command to examine memory.
6016 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
6017 much memory to display and how to format it; @var{addr} is an
6018 expression giving the address where you want to start displaying memory.
6019 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
6020 Several commands set convenient defaults for @var{addr}.
6023 @item @var{n}, the repeat count
6024 The repeat count is a decimal integer; the default is 1. It specifies
6025 how much memory (counting by units @var{u}) to display.
6026 @c This really is **decimal**; unaffected by 'set radix' as of GDB
6029 @item @var{f}, the display format
6030 The display format is one of the formats used by @code{print}
6031 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
6032 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
6033 The default is @samp{x} (hexadecimal) initially. The default changes
6034 each time you use either @code{x} or @code{print}.
6036 @item @var{u}, the unit size
6037 The unit size is any of
6043 Halfwords (two bytes).
6045 Words (four bytes). This is the initial default.
6047 Giant words (eight bytes).
6050 Each time you specify a unit size with @code{x}, that size becomes the
6051 default unit the next time you use @code{x}. (For the @samp{s} and
6052 @samp{i} formats, the unit size is ignored and is normally not written.)
6054 @item @var{addr}, starting display address
6055 @var{addr} is the address where you want @value{GDBN} to begin displaying
6056 memory. The expression need not have a pointer value (though it may);
6057 it is always interpreted as an integer address of a byte of memory.
6058 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
6059 @var{addr} is usually just after the last address examined---but several
6060 other commands also set the default address: @code{info breakpoints} (to
6061 the address of the last breakpoint listed), @code{info line} (to the
6062 starting address of a line), and @code{print} (if you use it to display
6063 a value from memory).
6066 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
6067 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
6068 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
6069 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
6070 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
6072 Since the letters indicating unit sizes are all distinct from the
6073 letters specifying output formats, you do not have to remember whether
6074 unit size or format comes first; either order works. The output
6075 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
6076 (However, the count @var{n} must come first; @samp{wx4} does not work.)
6078 Even though the unit size @var{u} is ignored for the formats @samp{s}
6079 and @samp{i}, you might still want to use a count @var{n}; for example,
6080 @samp{3i} specifies that you want to see three machine instructions,
6081 including any operands. For convenience, especially when used with
6082 the @code{display} command, the @samp{i} format also prints branch delay
6083 slot instructions, if any, beyond the count specified, which immediately
6084 follow the last instruction that is within the count. The command
6085 @code{disassemble} gives an alternative way of inspecting machine
6086 instructions; see @ref{Machine Code,,Source and Machine Code}.
6088 All the defaults for the arguments to @code{x} are designed to make it
6089 easy to continue scanning memory with minimal specifications each time
6090 you use @code{x}. For example, after you have inspected three machine
6091 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
6092 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
6093 the repeat count @var{n} is used again; the other arguments default as
6094 for successive uses of @code{x}.
6096 @cindex @code{$_}, @code{$__}, and value history
6097 The addresses and contents printed by the @code{x} command are not saved
6098 in the value history because there is often too much of them and they
6099 would get in the way. Instead, @value{GDBN} makes these values available for
6100 subsequent use in expressions as values of the convenience variables
6101 @code{$_} and @code{$__}. After an @code{x} command, the last address
6102 examined is available for use in expressions in the convenience variable
6103 @code{$_}. The contents of that address, as examined, are available in
6104 the convenience variable @code{$__}.
6106 If the @code{x} command has a repeat count, the address and contents saved
6107 are from the last memory unit printed; this is not the same as the last
6108 address printed if several units were printed on the last line of output.
6110 @cindex remote memory comparison
6111 @cindex verify remote memory image
6112 When you are debugging a program running on a remote target machine
6113 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
6114 remote machine's memory against the executable file you downloaded to
6115 the target. The @code{compare-sections} command is provided for such
6119 @kindex compare-sections
6120 @item compare-sections @r{[}@var{section-name}@r{]}
6121 Compare the data of a loadable section @var{section-name} in the
6122 executable file of the program being debugged with the same section in
6123 the remote machine's memory, and report any mismatches. With no
6124 arguments, compares all loadable sections. This command's
6125 availability depends on the target's support for the @code{"qCRC"}
6130 @section Automatic Display
6131 @cindex automatic display
6132 @cindex display of expressions
6134 If you find that you want to print the value of an expression frequently
6135 (to see how it changes), you might want to add it to the @dfn{automatic
6136 display list} so that @value{GDBN} prints its value each time your program stops.
6137 Each expression added to the list is given a number to identify it;
6138 to remove an expression from the list, you specify that number.
6139 The automatic display looks like this:
6143 3: bar[5] = (struct hack *) 0x3804
6147 This display shows item numbers, expressions and their current values. As with
6148 displays you request manually using @code{x} or @code{print}, you can
6149 specify the output format you prefer; in fact, @code{display} decides
6150 whether to use @code{print} or @code{x} depending your format
6151 specification---it uses @code{x} if you specify either the @samp{i}
6152 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
6156 @item display @var{expr}
6157 Add the expression @var{expr} to the list of expressions to display
6158 each time your program stops. @xref{Expressions, ,Expressions}.
6160 @code{display} does not repeat if you press @key{RET} again after using it.
6162 @item display/@var{fmt} @var{expr}
6163 For @var{fmt} specifying only a display format and not a size or
6164 count, add the expression @var{expr} to the auto-display list but
6165 arrange to display it each time in the specified format @var{fmt}.
6166 @xref{Output Formats,,Output Formats}.
6168 @item display/@var{fmt} @var{addr}
6169 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
6170 number of units, add the expression @var{addr} as a memory address to
6171 be examined each time your program stops. Examining means in effect
6172 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
6175 For example, @samp{display/i $pc} can be helpful, to see the machine
6176 instruction about to be executed each time execution stops (@samp{$pc}
6177 is a common name for the program counter; @pxref{Registers, ,Registers}).
6180 @kindex delete display
6182 @item undisplay @var{dnums}@dots{}
6183 @itemx delete display @var{dnums}@dots{}
6184 Remove item numbers @var{dnums} from the list of expressions to display.
6186 @code{undisplay} does not repeat if you press @key{RET} after using it.
6187 (Otherwise you would just get the error @samp{No display number @dots{}}.)
6189 @kindex disable display
6190 @item disable display @var{dnums}@dots{}
6191 Disable the display of item numbers @var{dnums}. A disabled display
6192 item is not printed automatically, but is not forgotten. It may be
6193 enabled again later.
6195 @kindex enable display
6196 @item enable display @var{dnums}@dots{}
6197 Enable display of item numbers @var{dnums}. It becomes effective once
6198 again in auto display of its expression, until you specify otherwise.
6201 Display the current values of the expressions on the list, just as is
6202 done when your program stops.
6204 @kindex info display
6206 Print the list of expressions previously set up to display
6207 automatically, each one with its item number, but without showing the
6208 values. This includes disabled expressions, which are marked as such.
6209 It also includes expressions which would not be displayed right now
6210 because they refer to automatic variables not currently available.
6213 @cindex display disabled out of scope
6214 If a display expression refers to local variables, then it does not make
6215 sense outside the lexical context for which it was set up. Such an
6216 expression is disabled when execution enters a context where one of its
6217 variables is not defined. For example, if you give the command
6218 @code{display last_char} while inside a function with an argument
6219 @code{last_char}, @value{GDBN} displays this argument while your program
6220 continues to stop inside that function. When it stops elsewhere---where
6221 there is no variable @code{last_char}---the display is disabled
6222 automatically. The next time your program stops where @code{last_char}
6223 is meaningful, you can enable the display expression once again.
6225 @node Print Settings
6226 @section Print Settings
6228 @cindex format options
6229 @cindex print settings
6230 @value{GDBN} provides the following ways to control how arrays, structures,
6231 and symbols are printed.
6234 These settings are useful for debugging programs in any language:
6238 @item set print address
6239 @itemx set print address on
6240 @cindex print/don't print memory addresses
6241 @value{GDBN} prints memory addresses showing the location of stack
6242 traces, structure values, pointer values, breakpoints, and so forth,
6243 even when it also displays the contents of those addresses. The default
6244 is @code{on}. For example, this is what a stack frame display looks like with
6245 @code{set print address on}:
6250 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
6252 530 if (lquote != def_lquote)
6256 @item set print address off
6257 Do not print addresses when displaying their contents. For example,
6258 this is the same stack frame displayed with @code{set print address off}:
6262 (@value{GDBP}) set print addr off
6264 #0 set_quotes (lq="<<", rq=">>") at input.c:530
6265 530 if (lquote != def_lquote)
6269 You can use @samp{set print address off} to eliminate all machine
6270 dependent displays from the @value{GDBN} interface. For example, with
6271 @code{print address off}, you should get the same text for backtraces on
6272 all machines---whether or not they involve pointer arguments.
6275 @item show print address
6276 Show whether or not addresses are to be printed.
6279 When @value{GDBN} prints a symbolic address, it normally prints the
6280 closest earlier symbol plus an offset. If that symbol does not uniquely
6281 identify the address (for example, it is a name whose scope is a single
6282 source file), you may need to clarify. One way to do this is with
6283 @code{info line}, for example @samp{info line *0x4537}. Alternately,
6284 you can set @value{GDBN} to print the source file and line number when
6285 it prints a symbolic address:
6288 @item set print symbol-filename on
6289 @cindex source file and line of a symbol
6290 @cindex symbol, source file and line
6291 Tell @value{GDBN} to print the source file name and line number of a
6292 symbol in the symbolic form of an address.
6294 @item set print symbol-filename off
6295 Do not print source file name and line number of a symbol. This is the
6298 @item show print symbol-filename
6299 Show whether or not @value{GDBN} will print the source file name and
6300 line number of a symbol in the symbolic form of an address.
6303 Another situation where it is helpful to show symbol filenames and line
6304 numbers is when disassembling code; @value{GDBN} shows you the line
6305 number and source file that corresponds to each instruction.
6307 Also, you may wish to see the symbolic form only if the address being
6308 printed is reasonably close to the closest earlier symbol:
6311 @item set print max-symbolic-offset @var{max-offset}
6312 @cindex maximum value for offset of closest symbol
6313 Tell @value{GDBN} to only display the symbolic form of an address if the
6314 offset between the closest earlier symbol and the address is less than
6315 @var{max-offset}. The default is 0, which tells @value{GDBN}
6316 to always print the symbolic form of an address if any symbol precedes it.
6318 @item show print max-symbolic-offset
6319 Ask how large the maximum offset is that @value{GDBN} prints in a
6323 @cindex wild pointer, interpreting
6324 @cindex pointer, finding referent
6325 If you have a pointer and you are not sure where it points, try
6326 @samp{set print symbol-filename on}. Then you can determine the name
6327 and source file location of the variable where it points, using
6328 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
6329 For example, here @value{GDBN} shows that a variable @code{ptt} points
6330 at another variable @code{t}, defined in @file{hi2.c}:
6333 (@value{GDBP}) set print symbol-filename on
6334 (@value{GDBP}) p/a ptt
6335 $4 = 0xe008 <t in hi2.c>
6339 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
6340 does not show the symbol name and filename of the referent, even with
6341 the appropriate @code{set print} options turned on.
6344 Other settings control how different kinds of objects are printed:
6347 @item set print array
6348 @itemx set print array on
6349 @cindex pretty print arrays
6350 Pretty print arrays. This format is more convenient to read,
6351 but uses more space. The default is off.
6353 @item set print array off
6354 Return to compressed format for arrays.
6356 @item show print array
6357 Show whether compressed or pretty format is selected for displaying
6360 @cindex print array indexes
6361 @item set print array-indexes
6362 @itemx set print array-indexes on
6363 Print the index of each element when displaying arrays. May be more
6364 convenient to locate a given element in the array or quickly find the
6365 index of a given element in that printed array. The default is off.
6367 @item set print array-indexes off
6368 Stop printing element indexes when displaying arrays.
6370 @item show print array-indexes
6371 Show whether the index of each element is printed when displaying
6374 @item set print elements @var{number-of-elements}
6375 @cindex number of array elements to print
6376 @cindex limit on number of printed array elements
6377 Set a limit on how many elements of an array @value{GDBN} will print.
6378 If @value{GDBN} is printing a large array, it stops printing after it has
6379 printed the number of elements set by the @code{set print elements} command.
6380 This limit also applies to the display of strings.
6381 When @value{GDBN} starts, this limit is set to 200.
6382 Setting @var{number-of-elements} to zero means that the printing is unlimited.
6384 @item show print elements
6385 Display the number of elements of a large array that @value{GDBN} will print.
6386 If the number is 0, then the printing is unlimited.
6388 @item set print frame-arguments @var{value}
6389 @cindex printing frame argument values
6390 @cindex print all frame argument values
6391 @cindex print frame argument values for scalars only
6392 @cindex do not print frame argument values
6393 This command allows to control how the values of arguments are printed
6394 when the debugger prints a frame (@pxref{Frames}). The possible
6399 The values of all arguments are printed. This is the default.
6402 Print the value of an argument only if it is a scalar. The value of more
6403 complex arguments such as arrays, structures, unions, etc, is replaced
6404 by @code{@dots{}}. Here is an example where only scalar arguments are shown:
6407 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
6412 None of the argument values are printed. Instead, the value of each argument
6413 is replaced by @code{@dots{}}. In this case, the example above now becomes:
6416 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
6421 By default, all argument values are always printed. But this command
6422 can be useful in several cases. For instance, it can be used to reduce
6423 the amount of information printed in each frame, making the backtrace
6424 more readable. Also, this command can be used to improve performance
6425 when displaying Ada frames, because the computation of large arguments
6426 can sometimes be CPU-intensive, especiallly in large applications.
6427 Setting @code{print frame-arguments} to @code{scalars} or @code{none}
6428 avoids this computation, thus speeding up the display of each Ada frame.
6430 @item show print frame-arguments
6431 Show how the value of arguments should be displayed when printing a frame.
6433 @item set print repeats
6434 @cindex repeated array elements
6435 Set the threshold for suppressing display of repeated array
6436 elements. When the number of consecutive identical elements of an
6437 array exceeds the threshold, @value{GDBN} prints the string
6438 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
6439 identical repetitions, instead of displaying the identical elements
6440 themselves. Setting the threshold to zero will cause all elements to
6441 be individually printed. The default threshold is 10.
6443 @item show print repeats
6444 Display the current threshold for printing repeated identical
6447 @item set print null-stop
6448 @cindex @sc{null} elements in arrays
6449 Cause @value{GDBN} to stop printing the characters of an array when the first
6450 @sc{null} is encountered. This is useful when large arrays actually
6451 contain only short strings.
6454 @item show print null-stop
6455 Show whether @value{GDBN} stops printing an array on the first
6456 @sc{null} character.
6458 @item set print pretty on
6459 @cindex print structures in indented form
6460 @cindex indentation in structure display
6461 Cause @value{GDBN} to print structures in an indented format with one member
6462 per line, like this:
6477 @item set print pretty off
6478 Cause @value{GDBN} to print structures in a compact format, like this:
6482 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
6483 meat = 0x54 "Pork"@}
6488 This is the default format.
6490 @item show print pretty
6491 Show which format @value{GDBN} is using to print structures.
6493 @item set print sevenbit-strings on
6494 @cindex eight-bit characters in strings
6495 @cindex octal escapes in strings
6496 Print using only seven-bit characters; if this option is set,
6497 @value{GDBN} displays any eight-bit characters (in strings or
6498 character values) using the notation @code{\}@var{nnn}. This setting is
6499 best if you are working in English (@sc{ascii}) and you use the
6500 high-order bit of characters as a marker or ``meta'' bit.
6502 @item set print sevenbit-strings off
6503 Print full eight-bit characters. This allows the use of more
6504 international character sets, and is the default.
6506 @item show print sevenbit-strings
6507 Show whether or not @value{GDBN} is printing only seven-bit characters.
6509 @item set print union on
6510 @cindex unions in structures, printing
6511 Tell @value{GDBN} to print unions which are contained in structures
6512 and other unions. This is the default setting.
6514 @item set print union off
6515 Tell @value{GDBN} not to print unions which are contained in
6516 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
6519 @item show print union
6520 Ask @value{GDBN} whether or not it will print unions which are contained in
6521 structures and other unions.
6523 For example, given the declarations
6526 typedef enum @{Tree, Bug@} Species;
6527 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
6528 typedef enum @{Caterpillar, Cocoon, Butterfly@}
6539 struct thing foo = @{Tree, @{Acorn@}@};
6543 with @code{set print union on} in effect @samp{p foo} would print
6546 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
6550 and with @code{set print union off} in effect it would print
6553 $1 = @{it = Tree, form = @{...@}@}
6557 @code{set print union} affects programs written in C-like languages
6563 These settings are of interest when debugging C@t{++} programs:
6566 @cindex demangling C@t{++} names
6567 @item set print demangle
6568 @itemx set print demangle on
6569 Print C@t{++} names in their source form rather than in the encoded
6570 (``mangled'') form passed to the assembler and linker for type-safe
6571 linkage. The default is on.
6573 @item show print demangle
6574 Show whether C@t{++} names are printed in mangled or demangled form.
6576 @item set print asm-demangle
6577 @itemx set print asm-demangle on
6578 Print C@t{++} names in their source form rather than their mangled form, even
6579 in assembler code printouts such as instruction disassemblies.
6582 @item show print asm-demangle
6583 Show whether C@t{++} names in assembly listings are printed in mangled
6586 @cindex C@t{++} symbol decoding style
6587 @cindex symbol decoding style, C@t{++}
6588 @kindex set demangle-style
6589 @item set demangle-style @var{style}
6590 Choose among several encoding schemes used by different compilers to
6591 represent C@t{++} names. The choices for @var{style} are currently:
6595 Allow @value{GDBN} to choose a decoding style by inspecting your program.
6598 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
6599 This is the default.
6602 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
6605 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
6608 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
6609 @strong{Warning:} this setting alone is not sufficient to allow
6610 debugging @code{cfront}-generated executables. @value{GDBN} would
6611 require further enhancement to permit that.
6614 If you omit @var{style}, you will see a list of possible formats.
6616 @item show demangle-style
6617 Display the encoding style currently in use for decoding C@t{++} symbols.
6619 @item set print object
6620 @itemx set print object on
6621 @cindex derived type of an object, printing
6622 @cindex display derived types
6623 When displaying a pointer to an object, identify the @emph{actual}
6624 (derived) type of the object rather than the @emph{declared} type, using
6625 the virtual function table.
6627 @item set print object off
6628 Display only the declared type of objects, without reference to the
6629 virtual function table. This is the default setting.
6631 @item show print object
6632 Show whether actual, or declared, object types are displayed.
6634 @item set print static-members
6635 @itemx set print static-members on
6636 @cindex static members of C@t{++} objects
6637 Print static members when displaying a C@t{++} object. The default is on.
6639 @item set print static-members off
6640 Do not print static members when displaying a C@t{++} object.
6642 @item show print static-members
6643 Show whether C@t{++} static members are printed or not.
6645 @item set print pascal_static-members
6646 @itemx set print pascal_static-members on
6647 @cindex static members of Pascal objects
6648 @cindex Pascal objects, static members display
6649 Print static members when displaying a Pascal object. The default is on.
6651 @item set print pascal_static-members off
6652 Do not print static members when displaying a Pascal object.
6654 @item show print pascal_static-members
6655 Show whether Pascal static members are printed or not.
6657 @c These don't work with HP ANSI C++ yet.
6658 @item set print vtbl
6659 @itemx set print vtbl on
6660 @cindex pretty print C@t{++} virtual function tables
6661 @cindex virtual functions (C@t{++}) display
6662 @cindex VTBL display
6663 Pretty print C@t{++} virtual function tables. The default is off.
6664 (The @code{vtbl} commands do not work on programs compiled with the HP
6665 ANSI C@t{++} compiler (@code{aCC}).)
6667 @item set print vtbl off
6668 Do not pretty print C@t{++} virtual function tables.
6670 @item show print vtbl
6671 Show whether C@t{++} virtual function tables are pretty printed, or not.
6675 @section Value History
6677 @cindex value history
6678 @cindex history of values printed by @value{GDBN}
6679 Values printed by the @code{print} command are saved in the @value{GDBN}
6680 @dfn{value history}. This allows you to refer to them in other expressions.
6681 Values are kept until the symbol table is re-read or discarded
6682 (for example with the @code{file} or @code{symbol-file} commands).
6683 When the symbol table changes, the value history is discarded,
6684 since the values may contain pointers back to the types defined in the
6689 @cindex history number
6690 The values printed are given @dfn{history numbers} by which you can
6691 refer to them. These are successive integers starting with one.
6692 @code{print} shows you the history number assigned to a value by
6693 printing @samp{$@var{num} = } before the value; here @var{num} is the
6696 To refer to any previous value, use @samp{$} followed by the value's
6697 history number. The way @code{print} labels its output is designed to
6698 remind you of this. Just @code{$} refers to the most recent value in
6699 the history, and @code{$$} refers to the value before that.
6700 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
6701 is the value just prior to @code{$$}, @code{$$1} is equivalent to
6702 @code{$$}, and @code{$$0} is equivalent to @code{$}.
6704 For example, suppose you have just printed a pointer to a structure and
6705 want to see the contents of the structure. It suffices to type
6711 If you have a chain of structures where the component @code{next} points
6712 to the next one, you can print the contents of the next one with this:
6719 You can print successive links in the chain by repeating this
6720 command---which you can do by just typing @key{RET}.
6722 Note that the history records values, not expressions. If the value of
6723 @code{x} is 4 and you type these commands:
6731 then the value recorded in the value history by the @code{print} command
6732 remains 4 even though the value of @code{x} has changed.
6737 Print the last ten values in the value history, with their item numbers.
6738 This is like @samp{p@ $$9} repeated ten times, except that @code{show
6739 values} does not change the history.
6741 @item show values @var{n}
6742 Print ten history values centered on history item number @var{n}.
6745 Print ten history values just after the values last printed. If no more
6746 values are available, @code{show values +} produces no display.
6749 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
6750 same effect as @samp{show values +}.
6752 @node Convenience Vars
6753 @section Convenience Variables
6755 @cindex convenience variables
6756 @cindex user-defined variables
6757 @value{GDBN} provides @dfn{convenience variables} that you can use within
6758 @value{GDBN} to hold on to a value and refer to it later. These variables
6759 exist entirely within @value{GDBN}; they are not part of your program, and
6760 setting a convenience variable has no direct effect on further execution
6761 of your program. That is why you can use them freely.
6763 Convenience variables are prefixed with @samp{$}. Any name preceded by
6764 @samp{$} can be used for a convenience variable, unless it is one of
6765 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
6766 (Value history references, in contrast, are @emph{numbers} preceded
6767 by @samp{$}. @xref{Value History, ,Value History}.)
6769 You can save a value in a convenience variable with an assignment
6770 expression, just as you would set a variable in your program.
6774 set $foo = *object_ptr
6778 would save in @code{$foo} the value contained in the object pointed to by
6781 Using a convenience variable for the first time creates it, but its
6782 value is @code{void} until you assign a new value. You can alter the
6783 value with another assignment at any time.
6785 Convenience variables have no fixed types. You can assign a convenience
6786 variable any type of value, including structures and arrays, even if
6787 that variable already has a value of a different type. The convenience
6788 variable, when used as an expression, has the type of its current value.
6791 @kindex show convenience
6792 @cindex show all user variables
6793 @item show convenience
6794 Print a list of convenience variables used so far, and their values.
6795 Abbreviated @code{show conv}.
6797 @kindex init-if-undefined
6798 @cindex convenience variables, initializing
6799 @item init-if-undefined $@var{variable} = @var{expression}
6800 Set a convenience variable if it has not already been set. This is useful
6801 for user-defined commands that keep some state. It is similar, in concept,
6802 to using local static variables with initializers in C (except that
6803 convenience variables are global). It can also be used to allow users to
6804 override default values used in a command script.
6806 If the variable is already defined then the expression is not evaluated so
6807 any side-effects do not occur.
6810 One of the ways to use a convenience variable is as a counter to be
6811 incremented or a pointer to be advanced. For example, to print
6812 a field from successive elements of an array of structures:
6816 print bar[$i++]->contents
6820 Repeat that command by typing @key{RET}.
6822 Some convenience variables are created automatically by @value{GDBN} and given
6823 values likely to be useful.
6826 @vindex $_@r{, convenience variable}
6828 The variable @code{$_} is automatically set by the @code{x} command to
6829 the last address examined (@pxref{Memory, ,Examining Memory}). Other
6830 commands which provide a default address for @code{x} to examine also
6831 set @code{$_} to that address; these commands include @code{info line}
6832 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
6833 except when set by the @code{x} command, in which case it is a pointer
6834 to the type of @code{$__}.
6836 @vindex $__@r{, convenience variable}
6838 The variable @code{$__} is automatically set by the @code{x} command
6839 to the value found in the last address examined. Its type is chosen
6840 to match the format in which the data was printed.
6843 @vindex $_exitcode@r{, convenience variable}
6844 The variable @code{$_exitcode} is automatically set to the exit code when
6845 the program being debugged terminates.
6848 On HP-UX systems, if you refer to a function or variable name that
6849 begins with a dollar sign, @value{GDBN} searches for a user or system
6850 name first, before it searches for a convenience variable.
6856 You can refer to machine register contents, in expressions, as variables
6857 with names starting with @samp{$}. The names of registers are different
6858 for each machine; use @code{info registers} to see the names used on
6862 @kindex info registers
6863 @item info registers
6864 Print the names and values of all registers except floating-point
6865 and vector registers (in the selected stack frame).
6867 @kindex info all-registers
6868 @cindex floating point registers
6869 @item info all-registers
6870 Print the names and values of all registers, including floating-point
6871 and vector registers (in the selected stack frame).
6873 @item info registers @var{regname} @dots{}
6874 Print the @dfn{relativized} value of each specified register @var{regname}.
6875 As discussed in detail below, register values are normally relative to
6876 the selected stack frame. @var{regname} may be any register name valid on
6877 the machine you are using, with or without the initial @samp{$}.
6880 @cindex stack pointer register
6881 @cindex program counter register
6882 @cindex process status register
6883 @cindex frame pointer register
6884 @cindex standard registers
6885 @value{GDBN} has four ``standard'' register names that are available (in
6886 expressions) on most machines---whenever they do not conflict with an
6887 architecture's canonical mnemonics for registers. The register names
6888 @code{$pc} and @code{$sp} are used for the program counter register and
6889 the stack pointer. @code{$fp} is used for a register that contains a
6890 pointer to the current stack frame, and @code{$ps} is used for a
6891 register that contains the processor status. For example,
6892 you could print the program counter in hex with
6899 or print the instruction to be executed next with
6906 or add four to the stack pointer@footnote{This is a way of removing
6907 one word from the stack, on machines where stacks grow downward in
6908 memory (most machines, nowadays). This assumes that the innermost
6909 stack frame is selected; setting @code{$sp} is not allowed when other
6910 stack frames are selected. To pop entire frames off the stack,
6911 regardless of machine architecture, use @code{return};
6912 see @ref{Returning, ,Returning from a Function}.} with
6918 Whenever possible, these four standard register names are available on
6919 your machine even though the machine has different canonical mnemonics,
6920 so long as there is no conflict. The @code{info registers} command
6921 shows the canonical names. For example, on the SPARC, @code{info
6922 registers} displays the processor status register as @code{$psr} but you
6923 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
6924 is an alias for the @sc{eflags} register.
6926 @value{GDBN} always considers the contents of an ordinary register as an
6927 integer when the register is examined in this way. Some machines have
6928 special registers which can hold nothing but floating point; these
6929 registers are considered to have floating point values. There is no way
6930 to refer to the contents of an ordinary register as floating point value
6931 (although you can @emph{print} it as a floating point value with
6932 @samp{print/f $@var{regname}}).
6934 Some registers have distinct ``raw'' and ``virtual'' data formats. This
6935 means that the data format in which the register contents are saved by
6936 the operating system is not the same one that your program normally
6937 sees. For example, the registers of the 68881 floating point
6938 coprocessor are always saved in ``extended'' (raw) format, but all C
6939 programs expect to work with ``double'' (virtual) format. In such
6940 cases, @value{GDBN} normally works with the virtual format only (the format
6941 that makes sense for your program), but the @code{info registers} command
6942 prints the data in both formats.
6944 @cindex SSE registers (x86)
6945 @cindex MMX registers (x86)
6946 Some machines have special registers whose contents can be interpreted
6947 in several different ways. For example, modern x86-based machines
6948 have SSE and MMX registers that can hold several values packed
6949 together in several different formats. @value{GDBN} refers to such
6950 registers in @code{struct} notation:
6953 (@value{GDBP}) print $xmm1
6955 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
6956 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
6957 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
6958 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
6959 v4_int32 = @{0, 20657912, 11, 13@},
6960 v2_int64 = @{88725056443645952, 55834574859@},
6961 uint128 = 0x0000000d0000000b013b36f800000000
6966 To set values of such registers, you need to tell @value{GDBN} which
6967 view of the register you wish to change, as if you were assigning
6968 value to a @code{struct} member:
6971 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
6974 Normally, register values are relative to the selected stack frame
6975 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
6976 value that the register would contain if all stack frames farther in
6977 were exited and their saved registers restored. In order to see the
6978 true contents of hardware registers, you must select the innermost
6979 frame (with @samp{frame 0}).
6981 However, @value{GDBN} must deduce where registers are saved, from the machine
6982 code generated by your compiler. If some registers are not saved, or if
6983 @value{GDBN} is unable to locate the saved registers, the selected stack
6984 frame makes no difference.
6986 @node Floating Point Hardware
6987 @section Floating Point Hardware
6988 @cindex floating point
6990 Depending on the configuration, @value{GDBN} may be able to give
6991 you more information about the status of the floating point hardware.
6996 Display hardware-dependent information about the floating
6997 point unit. The exact contents and layout vary depending on the
6998 floating point chip. Currently, @samp{info float} is supported on
6999 the ARM and x86 machines.
7003 @section Vector Unit
7006 Depending on the configuration, @value{GDBN} may be able to give you
7007 more information about the status of the vector unit.
7012 Display information about the vector unit. The exact contents and
7013 layout vary depending on the hardware.
7016 @node OS Information
7017 @section Operating System Auxiliary Information
7018 @cindex OS information
7020 @value{GDBN} provides interfaces to useful OS facilities that can help
7021 you debug your program.
7023 @cindex @code{ptrace} system call
7024 @cindex @code{struct user} contents
7025 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
7026 machines), it interfaces with the inferior via the @code{ptrace}
7027 system call. The operating system creates a special sata structure,
7028 called @code{struct user}, for this interface. You can use the
7029 command @code{info udot} to display the contents of this data
7035 Display the contents of the @code{struct user} maintained by the OS
7036 kernel for the program being debugged. @value{GDBN} displays the
7037 contents of @code{struct user} as a list of hex numbers, similar to
7038 the @code{examine} command.
7041 @cindex auxiliary vector
7042 @cindex vector, auxiliary
7043 Some operating systems supply an @dfn{auxiliary vector} to programs at
7044 startup. This is akin to the arguments and environment that you
7045 specify for a program, but contains a system-dependent variety of
7046 binary values that tell system libraries important details about the
7047 hardware, operating system, and process. Each value's purpose is
7048 identified by an integer tag; the meanings are well-known but system-specific.
7049 Depending on the configuration and operating system facilities,
7050 @value{GDBN} may be able to show you this information. For remote
7051 targets, this functionality may further depend on the remote stub's
7052 support of the @samp{qXfer:auxv:read} packet, see
7053 @ref{qXfer auxiliary vector read}.
7058 Display the auxiliary vector of the inferior, which can be either a
7059 live process or a core dump file. @value{GDBN} prints each tag value
7060 numerically, and also shows names and text descriptions for recognized
7061 tags. Some values in the vector are numbers, some bit masks, and some
7062 pointers to strings or other data. @value{GDBN} displays each value in the
7063 most appropriate form for a recognized tag, and in hexadecimal for
7064 an unrecognized tag.
7068 @node Memory Region Attributes
7069 @section Memory Region Attributes
7070 @cindex memory region attributes
7072 @dfn{Memory region attributes} allow you to describe special handling
7073 required by regions of your target's memory. @value{GDBN} uses
7074 attributes to determine whether to allow certain types of memory
7075 accesses; whether to use specific width accesses; and whether to cache
7076 target memory. By default the description of memory regions is
7077 fetched from the target (if the current target supports this), but the
7078 user can override the fetched regions.
7080 Defined memory regions can be individually enabled and disabled. When a
7081 memory region is disabled, @value{GDBN} uses the default attributes when
7082 accessing memory in that region. Similarly, if no memory regions have
7083 been defined, @value{GDBN} uses the default attributes when accessing
7086 When a memory region is defined, it is given a number to identify it;
7087 to enable, disable, or remove a memory region, you specify that number.
7091 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
7092 Define a memory region bounded by @var{lower} and @var{upper} with
7093 attributes @var{attributes}@dots{}, and add it to the list of regions
7094 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
7095 case: it is treated as the target's maximum memory address.
7096 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
7099 Discard any user changes to the memory regions and use target-supplied
7100 regions, if available, or no regions if the target does not support.
7103 @item delete mem @var{nums}@dots{}
7104 Remove memory regions @var{nums}@dots{} from the list of regions
7105 monitored by @value{GDBN}.
7108 @item disable mem @var{nums}@dots{}
7109 Disable monitoring of memory regions @var{nums}@dots{}.
7110 A disabled memory region is not forgotten.
7111 It may be enabled again later.
7114 @item enable mem @var{nums}@dots{}
7115 Enable monitoring of memory regions @var{nums}@dots{}.
7119 Print a table of all defined memory regions, with the following columns
7123 @item Memory Region Number
7124 @item Enabled or Disabled.
7125 Enabled memory regions are marked with @samp{y}.
7126 Disabled memory regions are marked with @samp{n}.
7129 The address defining the inclusive lower bound of the memory region.
7132 The address defining the exclusive upper bound of the memory region.
7135 The list of attributes set for this memory region.
7140 @subsection Attributes
7142 @subsubsection Memory Access Mode
7143 The access mode attributes set whether @value{GDBN} may make read or
7144 write accesses to a memory region.
7146 While these attributes prevent @value{GDBN} from performing invalid
7147 memory accesses, they do nothing to prevent the target system, I/O DMA,
7148 etc.@: from accessing memory.
7152 Memory is read only.
7154 Memory is write only.
7156 Memory is read/write. This is the default.
7159 @subsubsection Memory Access Size
7160 The access size attribute tells @value{GDBN} to use specific sized
7161 accesses in the memory region. Often memory mapped device registers
7162 require specific sized accesses. If no access size attribute is
7163 specified, @value{GDBN} may use accesses of any size.
7167 Use 8 bit memory accesses.
7169 Use 16 bit memory accesses.
7171 Use 32 bit memory accesses.
7173 Use 64 bit memory accesses.
7176 @c @subsubsection Hardware/Software Breakpoints
7177 @c The hardware/software breakpoint attributes set whether @value{GDBN}
7178 @c will use hardware or software breakpoints for the internal breakpoints
7179 @c used by the step, next, finish, until, etc. commands.
7183 @c Always use hardware breakpoints
7184 @c @item swbreak (default)
7187 @subsubsection Data Cache
7188 The data cache attributes set whether @value{GDBN} will cache target
7189 memory. While this generally improves performance by reducing debug
7190 protocol overhead, it can lead to incorrect results because @value{GDBN}
7191 does not know about volatile variables or memory mapped device
7196 Enable @value{GDBN} to cache target memory.
7198 Disable @value{GDBN} from caching target memory. This is the default.
7201 @subsection Memory Access Checking
7202 @value{GDBN} can be instructed to refuse accesses to memory that is
7203 not explicitly described. This can be useful if accessing such
7204 regions has undesired effects for a specific target, or to provide
7205 better error checking. The following commands control this behaviour.
7208 @kindex set mem inaccessible-by-default
7209 @item set mem inaccessible-by-default [on|off]
7210 If @code{on} is specified, make @value{GDBN} treat memory not
7211 explicitly described by the memory ranges as non-existent and refuse accesses
7212 to such memory. The checks are only performed if there's at least one
7213 memory range defined. If @code{off} is specified, make @value{GDBN}
7214 treat the memory not explicitly described by the memory ranges as RAM.
7215 The default value is @code{on}.
7216 @kindex show mem inaccessible-by-default
7217 @item show mem inaccessible-by-default
7218 Show the current handling of accesses to unknown memory.
7222 @c @subsubsection Memory Write Verification
7223 @c The memory write verification attributes set whether @value{GDBN}
7224 @c will re-reads data after each write to verify the write was successful.
7228 @c @item noverify (default)
7231 @node Dump/Restore Files
7232 @section Copy Between Memory and a File
7233 @cindex dump/restore files
7234 @cindex append data to a file
7235 @cindex dump data to a file
7236 @cindex restore data from a file
7238 You can use the commands @code{dump}, @code{append}, and
7239 @code{restore} to copy data between target memory and a file. The
7240 @code{dump} and @code{append} commands write data to a file, and the
7241 @code{restore} command reads data from a file back into the inferior's
7242 memory. Files may be in binary, Motorola S-record, Intel hex, or
7243 Tektronix Hex format; however, @value{GDBN} can only append to binary
7249 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
7250 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
7251 Dump the contents of memory from @var{start_addr} to @var{end_addr},
7252 or the value of @var{expr}, to @var{filename} in the given format.
7254 The @var{format} parameter may be any one of:
7261 Motorola S-record format.
7263 Tektronix Hex format.
7266 @value{GDBN} uses the same definitions of these formats as the
7267 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
7268 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
7272 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
7273 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
7274 Append the contents of memory from @var{start_addr} to @var{end_addr},
7275 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
7276 (@value{GDBN} can only append data to files in raw binary form.)
7279 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
7280 Restore the contents of file @var{filename} into memory. The
7281 @code{restore} command can automatically recognize any known @sc{bfd}
7282 file format, except for raw binary. To restore a raw binary file you
7283 must specify the optional keyword @code{binary} after the filename.
7285 If @var{bias} is non-zero, its value will be added to the addresses
7286 contained in the file. Binary files always start at address zero, so
7287 they will be restored at address @var{bias}. Other bfd files have
7288 a built-in location; they will be restored at offset @var{bias}
7291 If @var{start} and/or @var{end} are non-zero, then only data between
7292 file offset @var{start} and file offset @var{end} will be restored.
7293 These offsets are relative to the addresses in the file, before
7294 the @var{bias} argument is applied.
7298 @node Core File Generation
7299 @section How to Produce a Core File from Your Program
7300 @cindex dump core from inferior
7302 A @dfn{core file} or @dfn{core dump} is a file that records the memory
7303 image of a running process and its process status (register values
7304 etc.). Its primary use is post-mortem debugging of a program that
7305 crashed while it ran outside a debugger. A program that crashes
7306 automatically produces a core file, unless this feature is disabled by
7307 the user. @xref{Files}, for information on invoking @value{GDBN} in
7308 the post-mortem debugging mode.
7310 Occasionally, you may wish to produce a core file of the program you
7311 are debugging in order to preserve a snapshot of its state.
7312 @value{GDBN} has a special command for that.
7316 @kindex generate-core-file
7317 @item generate-core-file [@var{file}]
7318 @itemx gcore [@var{file}]
7319 Produce a core dump of the inferior process. The optional argument
7320 @var{file} specifies the file name where to put the core dump. If not
7321 specified, the file name defaults to @file{core.@var{pid}}, where
7322 @var{pid} is the inferior process ID.
7324 Note that this command is implemented only for some systems (as of
7325 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
7328 @node Character Sets
7329 @section Character Sets
7330 @cindex character sets
7332 @cindex translating between character sets
7333 @cindex host character set
7334 @cindex target character set
7336 If the program you are debugging uses a different character set to
7337 represent characters and strings than the one @value{GDBN} uses itself,
7338 @value{GDBN} can automatically translate between the character sets for
7339 you. The character set @value{GDBN} uses we call the @dfn{host
7340 character set}; the one the inferior program uses we call the
7341 @dfn{target character set}.
7343 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
7344 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
7345 remote protocol (@pxref{Remote Debugging}) to debug a program
7346 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
7347 then the host character set is Latin-1, and the target character set is
7348 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
7349 target-charset EBCDIC-US}, then @value{GDBN} translates between
7350 @sc{ebcdic} and Latin 1 as you print character or string values, or use
7351 character and string literals in expressions.
7353 @value{GDBN} has no way to automatically recognize which character set
7354 the inferior program uses; you must tell it, using the @code{set
7355 target-charset} command, described below.
7357 Here are the commands for controlling @value{GDBN}'s character set
7361 @item set target-charset @var{charset}
7362 @kindex set target-charset
7363 Set the current target character set to @var{charset}. We list the
7364 character set names @value{GDBN} recognizes below, but if you type
7365 @code{set target-charset} followed by @key{TAB}@key{TAB}, @value{GDBN} will
7366 list the target character sets it supports.
7370 @item set host-charset @var{charset}
7371 @kindex set host-charset
7372 Set the current host character set to @var{charset}.
7374 By default, @value{GDBN} uses a host character set appropriate to the
7375 system it is running on; you can override that default using the
7376 @code{set host-charset} command.
7378 @value{GDBN} can only use certain character sets as its host character
7379 set. We list the character set names @value{GDBN} recognizes below, and
7380 indicate which can be host character sets, but if you type
7381 @code{set target-charset} followed by @key{TAB}@key{TAB}, @value{GDBN} will
7382 list the host character sets it supports.
7384 @item set charset @var{charset}
7386 Set the current host and target character sets to @var{charset}. As
7387 above, if you type @code{set charset} followed by @key{TAB}@key{TAB},
7388 @value{GDBN} will list the name of the character sets that can be used
7389 for both host and target.
7393 @kindex show charset
7394 Show the names of the current host and target charsets.
7396 @itemx show host-charset
7397 @kindex show host-charset
7398 Show the name of the current host charset.
7400 @itemx show target-charset
7401 @kindex show target-charset
7402 Show the name of the current target charset.
7406 @value{GDBN} currently includes support for the following character
7412 @cindex ASCII character set
7413 Seven-bit U.S. @sc{ascii}. @value{GDBN} can use this as its host
7417 @cindex ISO 8859-1 character set
7418 @cindex ISO Latin 1 character set
7419 The ISO Latin 1 character set. This extends @sc{ascii} with accented
7420 characters needed for French, German, and Spanish. @value{GDBN} can use
7421 this as its host character set.
7425 @cindex EBCDIC character set
7426 @cindex IBM1047 character set
7427 Variants of the @sc{ebcdic} character set, used on some of IBM's
7428 mainframe operating systems. (@sc{gnu}/Linux on the S/390 uses U.S. @sc{ascii}.)
7429 @value{GDBN} cannot use these as its host character set.
7433 Note that these are all single-byte character sets. More work inside
7434 @value{GDBN} is needed to support multi-byte or variable-width character
7435 encodings, like the UTF-8 and UCS-2 encodings of Unicode.
7437 Here is an example of @value{GDBN}'s character set support in action.
7438 Assume that the following source code has been placed in the file
7439 @file{charset-test.c}:
7445 = @{72, 101, 108, 108, 111, 44, 32, 119,
7446 111, 114, 108, 100, 33, 10, 0@};
7447 char ibm1047_hello[]
7448 = @{200, 133, 147, 147, 150, 107, 64, 166,
7449 150, 153, 147, 132, 90, 37, 0@};
7453 printf ("Hello, world!\n");
7457 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
7458 containing the string @samp{Hello, world!} followed by a newline,
7459 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
7461 We compile the program, and invoke the debugger on it:
7464 $ gcc -g charset-test.c -o charset-test
7465 $ gdb -nw charset-test
7466 GNU gdb 2001-12-19-cvs
7467 Copyright 2001 Free Software Foundation, Inc.
7472 We can use the @code{show charset} command to see what character sets
7473 @value{GDBN} is currently using to interpret and display characters and
7477 (@value{GDBP}) show charset
7478 The current host and target character set is `ISO-8859-1'.
7482 For the sake of printing this manual, let's use @sc{ascii} as our
7483 initial character set:
7485 (@value{GDBP}) set charset ASCII
7486 (@value{GDBP}) show charset
7487 The current host and target character set is `ASCII'.
7491 Let's assume that @sc{ascii} is indeed the correct character set for our
7492 host system --- in other words, let's assume that if @value{GDBN} prints
7493 characters using the @sc{ascii} character set, our terminal will display
7494 them properly. Since our current target character set is also
7495 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
7498 (@value{GDBP}) print ascii_hello
7499 $1 = 0x401698 "Hello, world!\n"
7500 (@value{GDBP}) print ascii_hello[0]
7505 @value{GDBN} uses the target character set for character and string
7506 literals you use in expressions:
7509 (@value{GDBP}) print '+'
7514 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
7517 @value{GDBN} relies on the user to tell it which character set the
7518 target program uses. If we print @code{ibm1047_hello} while our target
7519 character set is still @sc{ascii}, we get jibberish:
7522 (@value{GDBP}) print ibm1047_hello
7523 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
7524 (@value{GDBP}) print ibm1047_hello[0]
7529 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
7530 @value{GDBN} tells us the character sets it supports:
7533 (@value{GDBP}) set target-charset
7534 ASCII EBCDIC-US IBM1047 ISO-8859-1
7535 (@value{GDBP}) set target-charset
7538 We can select @sc{ibm1047} as our target character set, and examine the
7539 program's strings again. Now the @sc{ascii} string is wrong, but
7540 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
7541 target character set, @sc{ibm1047}, to the host character set,
7542 @sc{ascii}, and they display correctly:
7545 (@value{GDBP}) set target-charset IBM1047
7546 (@value{GDBP}) show charset
7547 The current host character set is `ASCII'.
7548 The current target character set is `IBM1047'.
7549 (@value{GDBP}) print ascii_hello
7550 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
7551 (@value{GDBP}) print ascii_hello[0]
7553 (@value{GDBP}) print ibm1047_hello
7554 $8 = 0x4016a8 "Hello, world!\n"
7555 (@value{GDBP}) print ibm1047_hello[0]
7560 As above, @value{GDBN} uses the target character set for character and
7561 string literals you use in expressions:
7564 (@value{GDBP}) print '+'
7569 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
7572 @node Caching Remote Data
7573 @section Caching Data of Remote Targets
7574 @cindex caching data of remote targets
7576 @value{GDBN} can cache data exchanged between the debugger and a
7577 remote target (@pxref{Remote Debugging}). Such caching generally improves
7578 performance, because it reduces the overhead of the remote protocol by
7579 bundling memory reads and writes into large chunks. Unfortunately,
7580 @value{GDBN} does not currently know anything about volatile
7581 registers, and thus data caching will produce incorrect results when
7582 volatile registers are in use.
7585 @kindex set remotecache
7586 @item set remotecache on
7587 @itemx set remotecache off
7588 Set caching state for remote targets. When @code{ON}, use data
7589 caching. By default, this option is @code{OFF}.
7591 @kindex show remotecache
7592 @item show remotecache
7593 Show the current state of data caching for remote targets.
7597 Print the information about the data cache performance. The
7598 information displayed includes: the dcache width and depth; and for
7599 each cache line, how many times it was referenced, and its data and
7600 state (dirty, bad, ok, etc.). This command is useful for debugging
7601 the data cache operation.
7606 @chapter C Preprocessor Macros
7608 Some languages, such as C and C@t{++}, provide a way to define and invoke
7609 ``preprocessor macros'' which expand into strings of tokens.
7610 @value{GDBN} can evaluate expressions containing macro invocations, show
7611 the result of macro expansion, and show a macro's definition, including
7612 where it was defined.
7614 You may need to compile your program specially to provide @value{GDBN}
7615 with information about preprocessor macros. Most compilers do not
7616 include macros in their debugging information, even when you compile
7617 with the @option{-g} flag. @xref{Compilation}.
7619 A program may define a macro at one point, remove that definition later,
7620 and then provide a different definition after that. Thus, at different
7621 points in the program, a macro may have different definitions, or have
7622 no definition at all. If there is a current stack frame, @value{GDBN}
7623 uses the macros in scope at that frame's source code line. Otherwise,
7624 @value{GDBN} uses the macros in scope at the current listing location;
7627 At the moment, @value{GDBN} does not support the @code{##}
7628 token-splicing operator, the @code{#} stringification operator, or
7629 variable-arity macros.
7631 Whenever @value{GDBN} evaluates an expression, it always expands any
7632 macro invocations present in the expression. @value{GDBN} also provides
7633 the following commands for working with macros explicitly.
7637 @kindex macro expand
7638 @cindex macro expansion, showing the results of preprocessor
7639 @cindex preprocessor macro expansion, showing the results of
7640 @cindex expanding preprocessor macros
7641 @item macro expand @var{expression}
7642 @itemx macro exp @var{expression}
7643 Show the results of expanding all preprocessor macro invocations in
7644 @var{expression}. Since @value{GDBN} simply expands macros, but does
7645 not parse the result, @var{expression} need not be a valid expression;
7646 it can be any string of tokens.
7649 @item macro expand-once @var{expression}
7650 @itemx macro exp1 @var{expression}
7651 @cindex expand macro once
7652 @i{(This command is not yet implemented.)} Show the results of
7653 expanding those preprocessor macro invocations that appear explicitly in
7654 @var{expression}. Macro invocations appearing in that expansion are
7655 left unchanged. This command allows you to see the effect of a
7656 particular macro more clearly, without being confused by further
7657 expansions. Since @value{GDBN} simply expands macros, but does not
7658 parse the result, @var{expression} need not be a valid expression; it
7659 can be any string of tokens.
7662 @cindex macro definition, showing
7663 @cindex definition, showing a macro's
7664 @item info macro @var{macro}
7665 Show the definition of the macro named @var{macro}, and describe the
7666 source location where that definition was established.
7668 @kindex macro define
7669 @cindex user-defined macros
7670 @cindex defining macros interactively
7671 @cindex macros, user-defined
7672 @item macro define @var{macro} @var{replacement-list}
7673 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
7674 @i{(This command is not yet implemented.)} Introduce a definition for a
7675 preprocessor macro named @var{macro}, invocations of which are replaced
7676 by the tokens given in @var{replacement-list}. The first form of this
7677 command defines an ``object-like'' macro, which takes no arguments; the
7678 second form defines a ``function-like'' macro, which takes the arguments
7679 given in @var{arglist}.
7681 A definition introduced by this command is in scope in every expression
7682 evaluated in @value{GDBN}, until it is removed with the @command{macro
7683 undef} command, described below. The definition overrides all
7684 definitions for @var{macro} present in the program being debugged, as
7685 well as any previous user-supplied definition.
7688 @item macro undef @var{macro}
7689 @i{(This command is not yet implemented.)} Remove any user-supplied
7690 definition for the macro named @var{macro}. This command only affects
7691 definitions provided with the @command{macro define} command, described
7692 above; it cannot remove definitions present in the program being
7697 @i{(This command is not yet implemented.)} List all the macros
7698 defined using the @code{macro define} command.
7701 @cindex macros, example of debugging with
7702 Here is a transcript showing the above commands in action. First, we
7703 show our source files:
7711 #define ADD(x) (M + x)
7716 printf ("Hello, world!\n");
7718 printf ("We're so creative.\n");
7720 printf ("Goodbye, world!\n");
7727 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
7728 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
7729 compiler includes information about preprocessor macros in the debugging
7733 $ gcc -gdwarf-2 -g3 sample.c -o sample
7737 Now, we start @value{GDBN} on our sample program:
7741 GNU gdb 2002-05-06-cvs
7742 Copyright 2002 Free Software Foundation, Inc.
7743 GDB is free software, @dots{}
7747 We can expand macros and examine their definitions, even when the
7748 program is not running. @value{GDBN} uses the current listing position
7749 to decide which macro definitions are in scope:
7752 (@value{GDBP}) list main
7755 5 #define ADD(x) (M + x)
7760 10 printf ("Hello, world!\n");
7762 12 printf ("We're so creative.\n");
7763 (@value{GDBP}) info macro ADD
7764 Defined at /home/jimb/gdb/macros/play/sample.c:5
7765 #define ADD(x) (M + x)
7766 (@value{GDBP}) info macro Q
7767 Defined at /home/jimb/gdb/macros/play/sample.h:1
7768 included at /home/jimb/gdb/macros/play/sample.c:2
7770 (@value{GDBP}) macro expand ADD(1)
7771 expands to: (42 + 1)
7772 (@value{GDBP}) macro expand-once ADD(1)
7773 expands to: once (M + 1)
7777 In the example above, note that @command{macro expand-once} expands only
7778 the macro invocation explicit in the original text --- the invocation of
7779 @code{ADD} --- but does not expand the invocation of the macro @code{M},
7780 which was introduced by @code{ADD}.
7782 Once the program is running, @value{GDBN} uses the macro definitions in
7783 force at the source line of the current stack frame:
7786 (@value{GDBP}) break main
7787 Breakpoint 1 at 0x8048370: file sample.c, line 10.
7789 Starting program: /home/jimb/gdb/macros/play/sample
7791 Breakpoint 1, main () at sample.c:10
7792 10 printf ("Hello, world!\n");
7796 At line 10, the definition of the macro @code{N} at line 9 is in force:
7799 (@value{GDBP}) info macro N
7800 Defined at /home/jimb/gdb/macros/play/sample.c:9
7802 (@value{GDBP}) macro expand N Q M
7804 (@value{GDBP}) print N Q M
7809 As we step over directives that remove @code{N}'s definition, and then
7810 give it a new definition, @value{GDBN} finds the definition (or lack
7811 thereof) in force at each point:
7816 12 printf ("We're so creative.\n");
7817 (@value{GDBP}) info macro N
7818 The symbol `N' has no definition as a C/C++ preprocessor macro
7819 at /home/jimb/gdb/macros/play/sample.c:12
7822 14 printf ("Goodbye, world!\n");
7823 (@value{GDBP}) info macro N
7824 Defined at /home/jimb/gdb/macros/play/sample.c:13
7826 (@value{GDBP}) macro expand N Q M
7827 expands to: 1729 < 42
7828 (@value{GDBP}) print N Q M
7835 @chapter Tracepoints
7836 @c This chapter is based on the documentation written by Michael
7837 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
7840 In some applications, it is not feasible for the debugger to interrupt
7841 the program's execution long enough for the developer to learn
7842 anything helpful about its behavior. If the program's correctness
7843 depends on its real-time behavior, delays introduced by a debugger
7844 might cause the program to change its behavior drastically, or perhaps
7845 fail, even when the code itself is correct. It is useful to be able
7846 to observe the program's behavior without interrupting it.
7848 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
7849 specify locations in the program, called @dfn{tracepoints}, and
7850 arbitrary expressions to evaluate when those tracepoints are reached.
7851 Later, using the @code{tfind} command, you can examine the values
7852 those expressions had when the program hit the tracepoints. The
7853 expressions may also denote objects in memory---structures or arrays,
7854 for example---whose values @value{GDBN} should record; while visiting
7855 a particular tracepoint, you may inspect those objects as if they were
7856 in memory at that moment. However, because @value{GDBN} records these
7857 values without interacting with you, it can do so quickly and
7858 unobtrusively, hopefully not disturbing the program's behavior.
7860 The tracepoint facility is currently available only for remote
7861 targets. @xref{Targets}. In addition, your remote target must know
7862 how to collect trace data. This functionality is implemented in the
7863 remote stub; however, none of the stubs distributed with @value{GDBN}
7864 support tracepoints as of this writing. The format of the remote
7865 packets used to implement tracepoints are described in @ref{Tracepoint
7868 This chapter describes the tracepoint commands and features.
7872 * Analyze Collected Data::
7873 * Tracepoint Variables::
7876 @node Set Tracepoints
7877 @section Commands to Set Tracepoints
7879 Before running such a @dfn{trace experiment}, an arbitrary number of
7880 tracepoints can be set. Like a breakpoint (@pxref{Set Breaks}), a
7881 tracepoint has a number assigned to it by @value{GDBN}. Like with
7882 breakpoints, tracepoint numbers are successive integers starting from
7883 one. Many of the commands associated with tracepoints take the
7884 tracepoint number as their argument, to identify which tracepoint to
7887 For each tracepoint, you can specify, in advance, some arbitrary set
7888 of data that you want the target to collect in the trace buffer when
7889 it hits that tracepoint. The collected data can include registers,
7890 local variables, or global data. Later, you can use @value{GDBN}
7891 commands to examine the values these data had at the time the
7894 This section describes commands to set tracepoints and associated
7895 conditions and actions.
7898 * Create and Delete Tracepoints::
7899 * Enable and Disable Tracepoints::
7900 * Tracepoint Passcounts::
7901 * Tracepoint Actions::
7902 * Listing Tracepoints::
7903 * Starting and Stopping Trace Experiments::
7906 @node Create and Delete Tracepoints
7907 @subsection Create and Delete Tracepoints
7910 @cindex set tracepoint
7913 The @code{trace} command is very similar to the @code{break} command.
7914 Its argument can be a source line, a function name, or an address in
7915 the target program. @xref{Set Breaks}. The @code{trace} command
7916 defines a tracepoint, which is a point in the target program where the
7917 debugger will briefly stop, collect some data, and then allow the
7918 program to continue. Setting a tracepoint or changing its commands
7919 doesn't take effect until the next @code{tstart} command; thus, you
7920 cannot change the tracepoint attributes once a trace experiment is
7923 Here are some examples of using the @code{trace} command:
7926 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
7928 (@value{GDBP}) @b{trace +2} // 2 lines forward
7930 (@value{GDBP}) @b{trace my_function} // first source line of function
7932 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
7934 (@value{GDBP}) @b{trace *0x2117c4} // an address
7938 You can abbreviate @code{trace} as @code{tr}.
7941 @cindex last tracepoint number
7942 @cindex recent tracepoint number
7943 @cindex tracepoint number
7944 The convenience variable @code{$tpnum} records the tracepoint number
7945 of the most recently set tracepoint.
7947 @kindex delete tracepoint
7948 @cindex tracepoint deletion
7949 @item delete tracepoint @r{[}@var{num}@r{]}
7950 Permanently delete one or more tracepoints. With no argument, the
7951 default is to delete all tracepoints.
7956 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
7958 (@value{GDBP}) @b{delete trace} // remove all tracepoints
7962 You can abbreviate this command as @code{del tr}.
7965 @node Enable and Disable Tracepoints
7966 @subsection Enable and Disable Tracepoints
7969 @kindex disable tracepoint
7970 @item disable tracepoint @r{[}@var{num}@r{]}
7971 Disable tracepoint @var{num}, or all tracepoints if no argument
7972 @var{num} is given. A disabled tracepoint will have no effect during
7973 the next trace experiment, but it is not forgotten. You can re-enable
7974 a disabled tracepoint using the @code{enable tracepoint} command.
7976 @kindex enable tracepoint
7977 @item enable tracepoint @r{[}@var{num}@r{]}
7978 Enable tracepoint @var{num}, or all tracepoints. The enabled
7979 tracepoints will become effective the next time a trace experiment is
7983 @node Tracepoint Passcounts
7984 @subsection Tracepoint Passcounts
7988 @cindex tracepoint pass count
7989 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
7990 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
7991 automatically stop a trace experiment. If a tracepoint's passcount is
7992 @var{n}, then the trace experiment will be automatically stopped on
7993 the @var{n}'th time that tracepoint is hit. If the tracepoint number
7994 @var{num} is not specified, the @code{passcount} command sets the
7995 passcount of the most recently defined tracepoint. If no passcount is
7996 given, the trace experiment will run until stopped explicitly by the
8002 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
8003 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
8005 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
8006 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
8007 (@value{GDBP}) @b{trace foo}
8008 (@value{GDBP}) @b{pass 3}
8009 (@value{GDBP}) @b{trace bar}
8010 (@value{GDBP}) @b{pass 2}
8011 (@value{GDBP}) @b{trace baz}
8012 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
8013 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
8014 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
8015 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
8019 @node Tracepoint Actions
8020 @subsection Tracepoint Action Lists
8024 @cindex tracepoint actions
8025 @item actions @r{[}@var{num}@r{]}
8026 This command will prompt for a list of actions to be taken when the
8027 tracepoint is hit. If the tracepoint number @var{num} is not
8028 specified, this command sets the actions for the one that was most
8029 recently defined (so that you can define a tracepoint and then say
8030 @code{actions} without bothering about its number). You specify the
8031 actions themselves on the following lines, one action at a time, and
8032 terminate the actions list with a line containing just @code{end}. So
8033 far, the only defined actions are @code{collect} and
8034 @code{while-stepping}.
8036 @cindex remove actions from a tracepoint
8037 To remove all actions from a tracepoint, type @samp{actions @var{num}}
8038 and follow it immediately with @samp{end}.
8041 (@value{GDBP}) @b{collect @var{data}} // collect some data
8043 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
8045 (@value{GDBP}) @b{end} // signals the end of actions.
8048 In the following example, the action list begins with @code{collect}
8049 commands indicating the things to be collected when the tracepoint is
8050 hit. Then, in order to single-step and collect additional data
8051 following the tracepoint, a @code{while-stepping} command is used,
8052 followed by the list of things to be collected while stepping. The
8053 @code{while-stepping} command is terminated by its own separate
8054 @code{end} command. Lastly, the action list is terminated by an
8058 (@value{GDBP}) @b{trace foo}
8059 (@value{GDBP}) @b{actions}
8060 Enter actions for tracepoint 1, one per line:
8069 @kindex collect @r{(tracepoints)}
8070 @item collect @var{expr1}, @var{expr2}, @dots{}
8071 Collect values of the given expressions when the tracepoint is hit.
8072 This command accepts a comma-separated list of any valid expressions.
8073 In addition to global, static, or local variables, the following
8074 special arguments are supported:
8078 collect all registers
8081 collect all function arguments
8084 collect all local variables.
8087 You can give several consecutive @code{collect} commands, each one
8088 with a single argument, or one @code{collect} command with several
8089 arguments separated by commas: the effect is the same.
8091 The command @code{info scope} (@pxref{Symbols, info scope}) is
8092 particularly useful for figuring out what data to collect.
8094 @kindex while-stepping @r{(tracepoints)}
8095 @item while-stepping @var{n}
8096 Perform @var{n} single-step traces after the tracepoint, collecting
8097 new data at each step. The @code{while-stepping} command is
8098 followed by the list of what to collect while stepping (followed by
8099 its own @code{end} command):
8103 > collect $regs, myglobal
8109 You may abbreviate @code{while-stepping} as @code{ws} or
8113 @node Listing Tracepoints
8114 @subsection Listing Tracepoints
8117 @kindex info tracepoints
8119 @cindex information about tracepoints
8120 @item info tracepoints @r{[}@var{num}@r{]}
8121 Display information about the tracepoint @var{num}. If you don't specify
8122 a tracepoint number, displays information about all the tracepoints
8123 defined so far. For each tracepoint, the following information is
8130 whether it is enabled or disabled
8134 its passcount as given by the @code{passcount @var{n}} command
8136 its step count as given by the @code{while-stepping @var{n}} command
8138 where in the source files is the tracepoint set
8140 its action list as given by the @code{actions} command
8144 (@value{GDBP}) @b{info trace}
8145 Num Enb Address PassC StepC What
8146 1 y 0x002117c4 0 0 <gdb_asm>
8147 2 y 0x0020dc64 0 0 in g_test at g_test.c:1375
8148 3 y 0x0020b1f4 0 0 in get_data at ../foo.c:41
8153 This command can be abbreviated @code{info tp}.
8156 @node Starting and Stopping Trace Experiments
8157 @subsection Starting and Stopping Trace Experiments
8161 @cindex start a new trace experiment
8162 @cindex collected data discarded
8164 This command takes no arguments. It starts the trace experiment, and
8165 begins collecting data. This has the side effect of discarding all
8166 the data collected in the trace buffer during the previous trace
8170 @cindex stop a running trace experiment
8172 This command takes no arguments. It ends the trace experiment, and
8173 stops collecting data.
8175 @strong{Note}: a trace experiment and data collection may stop
8176 automatically if any tracepoint's passcount is reached
8177 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
8180 @cindex status of trace data collection
8181 @cindex trace experiment, status of
8183 This command displays the status of the current trace data
8187 Here is an example of the commands we described so far:
8190 (@value{GDBP}) @b{trace gdb_c_test}
8191 (@value{GDBP}) @b{actions}
8192 Enter actions for tracepoint #1, one per line.
8193 > collect $regs,$locals,$args
8198 (@value{GDBP}) @b{tstart}
8199 [time passes @dots{}]
8200 (@value{GDBP}) @b{tstop}
8204 @node Analyze Collected Data
8205 @section Using the Collected Data
8207 After the tracepoint experiment ends, you use @value{GDBN} commands
8208 for examining the trace data. The basic idea is that each tracepoint
8209 collects a trace @dfn{snapshot} every time it is hit and another
8210 snapshot every time it single-steps. All these snapshots are
8211 consecutively numbered from zero and go into a buffer, and you can
8212 examine them later. The way you examine them is to @dfn{focus} on a
8213 specific trace snapshot. When the remote stub is focused on a trace
8214 snapshot, it will respond to all @value{GDBN} requests for memory and
8215 registers by reading from the buffer which belongs to that snapshot,
8216 rather than from @emph{real} memory or registers of the program being
8217 debugged. This means that @strong{all} @value{GDBN} commands
8218 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
8219 behave as if we were currently debugging the program state as it was
8220 when the tracepoint occurred. Any requests for data that are not in
8221 the buffer will fail.
8224 * tfind:: How to select a trace snapshot
8225 * tdump:: How to display all data for a snapshot
8226 * save-tracepoints:: How to save tracepoints for a future run
8230 @subsection @code{tfind @var{n}}
8233 @cindex select trace snapshot
8234 @cindex find trace snapshot
8235 The basic command for selecting a trace snapshot from the buffer is
8236 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
8237 counting from zero. If no argument @var{n} is given, the next
8238 snapshot is selected.
8240 Here are the various forms of using the @code{tfind} command.
8244 Find the first snapshot in the buffer. This is a synonym for
8245 @code{tfind 0} (since 0 is the number of the first snapshot).
8248 Stop debugging trace snapshots, resume @emph{live} debugging.
8251 Same as @samp{tfind none}.
8254 No argument means find the next trace snapshot.
8257 Find the previous trace snapshot before the current one. This permits
8258 retracing earlier steps.
8260 @item tfind tracepoint @var{num}
8261 Find the next snapshot associated with tracepoint @var{num}. Search
8262 proceeds forward from the last examined trace snapshot. If no
8263 argument @var{num} is given, it means find the next snapshot collected
8264 for the same tracepoint as the current snapshot.
8266 @item tfind pc @var{addr}
8267 Find the next snapshot associated with the value @var{addr} of the
8268 program counter. Search proceeds forward from the last examined trace
8269 snapshot. If no argument @var{addr} is given, it means find the next
8270 snapshot with the same value of PC as the current snapshot.
8272 @item tfind outside @var{addr1}, @var{addr2}
8273 Find the next snapshot whose PC is outside the given range of
8276 @item tfind range @var{addr1}, @var{addr2}
8277 Find the next snapshot whose PC is between @var{addr1} and
8278 @var{addr2}. @c FIXME: Is the range inclusive or exclusive?
8280 @item tfind line @r{[}@var{file}:@r{]}@var{n}
8281 Find the next snapshot associated with the source line @var{n}. If
8282 the optional argument @var{file} is given, refer to line @var{n} in
8283 that source file. Search proceeds forward from the last examined
8284 trace snapshot. If no argument @var{n} is given, it means find the
8285 next line other than the one currently being examined; thus saying
8286 @code{tfind line} repeatedly can appear to have the same effect as
8287 stepping from line to line in a @emph{live} debugging session.
8290 The default arguments for the @code{tfind} commands are specifically
8291 designed to make it easy to scan through the trace buffer. For
8292 instance, @code{tfind} with no argument selects the next trace
8293 snapshot, and @code{tfind -} with no argument selects the previous
8294 trace snapshot. So, by giving one @code{tfind} command, and then
8295 simply hitting @key{RET} repeatedly you can examine all the trace
8296 snapshots in order. Or, by saying @code{tfind -} and then hitting
8297 @key{RET} repeatedly you can examine the snapshots in reverse order.
8298 The @code{tfind line} command with no argument selects the snapshot
8299 for the next source line executed. The @code{tfind pc} command with
8300 no argument selects the next snapshot with the same program counter
8301 (PC) as the current frame. The @code{tfind tracepoint} command with
8302 no argument selects the next trace snapshot collected by the same
8303 tracepoint as the current one.
8305 In addition to letting you scan through the trace buffer manually,
8306 these commands make it easy to construct @value{GDBN} scripts that
8307 scan through the trace buffer and print out whatever collected data
8308 you are interested in. Thus, if we want to examine the PC, FP, and SP
8309 registers from each trace frame in the buffer, we can say this:
8312 (@value{GDBP}) @b{tfind start}
8313 (@value{GDBP}) @b{while ($trace_frame != -1)}
8314 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
8315 $trace_frame, $pc, $sp, $fp
8319 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
8320 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
8321 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
8322 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
8323 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
8324 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
8325 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
8326 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
8327 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
8328 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
8329 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
8332 Or, if we want to examine the variable @code{X} at each source line in
8336 (@value{GDBP}) @b{tfind start}
8337 (@value{GDBP}) @b{while ($trace_frame != -1)}
8338 > printf "Frame %d, X == %d\n", $trace_frame, X
8348 @subsection @code{tdump}
8350 @cindex dump all data collected at tracepoint
8351 @cindex tracepoint data, display
8353 This command takes no arguments. It prints all the data collected at
8354 the current trace snapshot.
8357 (@value{GDBP}) @b{trace 444}
8358 (@value{GDBP}) @b{actions}
8359 Enter actions for tracepoint #2, one per line:
8360 > collect $regs, $locals, $args, gdb_long_test
8363 (@value{GDBP}) @b{tstart}
8365 (@value{GDBP}) @b{tfind line 444}
8366 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
8368 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
8370 (@value{GDBP}) @b{tdump}
8371 Data collected at tracepoint 2, trace frame 1:
8372 d0 0xc4aa0085 -995491707
8376 d4 0x71aea3d 119204413
8381 a1 0x3000668 50333288
8384 a4 0x3000698 50333336
8386 fp 0x30bf3c 0x30bf3c
8387 sp 0x30bf34 0x30bf34
8389 pc 0x20b2c8 0x20b2c8
8393 p = 0x20e5b4 "gdb-test"
8400 gdb_long_test = 17 '\021'
8405 @node save-tracepoints
8406 @subsection @code{save-tracepoints @var{filename}}
8407 @kindex save-tracepoints
8408 @cindex save tracepoints for future sessions
8410 This command saves all current tracepoint definitions together with
8411 their actions and passcounts, into a file @file{@var{filename}}
8412 suitable for use in a later debugging session. To read the saved
8413 tracepoint definitions, use the @code{source} command (@pxref{Command
8416 @node Tracepoint Variables
8417 @section Convenience Variables for Tracepoints
8418 @cindex tracepoint variables
8419 @cindex convenience variables for tracepoints
8422 @vindex $trace_frame
8423 @item (int) $trace_frame
8424 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
8425 snapshot is selected.
8428 @item (int) $tracepoint
8429 The tracepoint for the current trace snapshot.
8432 @item (int) $trace_line
8433 The line number for the current trace snapshot.
8436 @item (char []) $trace_file
8437 The source file for the current trace snapshot.
8440 @item (char []) $trace_func
8441 The name of the function containing @code{$tracepoint}.
8444 Note: @code{$trace_file} is not suitable for use in @code{printf},
8445 use @code{output} instead.
8447 Here's a simple example of using these convenience variables for
8448 stepping through all the trace snapshots and printing some of their
8452 (@value{GDBP}) @b{tfind start}
8454 (@value{GDBP}) @b{while $trace_frame != -1}
8455 > output $trace_file
8456 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
8462 @chapter Debugging Programs That Use Overlays
8465 If your program is too large to fit completely in your target system's
8466 memory, you can sometimes use @dfn{overlays} to work around this
8467 problem. @value{GDBN} provides some support for debugging programs that
8471 * How Overlays Work:: A general explanation of overlays.
8472 * Overlay Commands:: Managing overlays in @value{GDBN}.
8473 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
8474 mapped by asking the inferior.
8475 * Overlay Sample Program:: A sample program using overlays.
8478 @node How Overlays Work
8479 @section How Overlays Work
8480 @cindex mapped overlays
8481 @cindex unmapped overlays
8482 @cindex load address, overlay's
8483 @cindex mapped address
8484 @cindex overlay area
8486 Suppose you have a computer whose instruction address space is only 64
8487 kilobytes long, but which has much more memory which can be accessed by
8488 other means: special instructions, segment registers, or memory
8489 management hardware, for example. Suppose further that you want to
8490 adapt a program which is larger than 64 kilobytes to run on this system.
8492 One solution is to identify modules of your program which are relatively
8493 independent, and need not call each other directly; call these modules
8494 @dfn{overlays}. Separate the overlays from the main program, and place
8495 their machine code in the larger memory. Place your main program in
8496 instruction memory, but leave at least enough space there to hold the
8497 largest overlay as well.
8499 Now, to call a function located in an overlay, you must first copy that
8500 overlay's machine code from the large memory into the space set aside
8501 for it in the instruction memory, and then jump to its entry point
8504 @c NB: In the below the mapped area's size is greater or equal to the
8505 @c size of all overlays. This is intentional to remind the developer
8506 @c that overlays don't necessarily need to be the same size.
8510 Data Instruction Larger
8511 Address Space Address Space Address Space
8512 +-----------+ +-----------+ +-----------+
8514 +-----------+ +-----------+ +-----------+<-- overlay 1
8515 | program | | main | .----| overlay 1 | load address
8516 | variables | | program | | +-----------+
8517 | and heap | | | | | |
8518 +-----------+ | | | +-----------+<-- overlay 2
8519 | | +-----------+ | | | load address
8520 +-----------+ | | | .-| overlay 2 |
8522 mapped --->+-----------+ | | +-----------+
8524 | overlay | <-' | | |
8525 | area | <---' +-----------+<-- overlay 3
8526 | | <---. | | load address
8527 +-----------+ `--| overlay 3 |
8534 @anchor{A code overlay}A code overlay
8538 The diagram (@pxref{A code overlay}) shows a system with separate data
8539 and instruction address spaces. To map an overlay, the program copies
8540 its code from the larger address space to the instruction address space.
8541 Since the overlays shown here all use the same mapped address, only one
8542 may be mapped at a time. For a system with a single address space for
8543 data and instructions, the diagram would be similar, except that the
8544 program variables and heap would share an address space with the main
8545 program and the overlay area.
8547 An overlay loaded into instruction memory and ready for use is called a
8548 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
8549 instruction memory. An overlay not present (or only partially present)
8550 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
8551 is its address in the larger memory. The mapped address is also called
8552 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
8553 called the @dfn{load memory address}, or @dfn{LMA}.
8555 Unfortunately, overlays are not a completely transparent way to adapt a
8556 program to limited instruction memory. They introduce a new set of
8557 global constraints you must keep in mind as you design your program:
8562 Before calling or returning to a function in an overlay, your program
8563 must make sure that overlay is actually mapped. Otherwise, the call or
8564 return will transfer control to the right address, but in the wrong
8565 overlay, and your program will probably crash.
8568 If the process of mapping an overlay is expensive on your system, you
8569 will need to choose your overlays carefully to minimize their effect on
8570 your program's performance.
8573 The executable file you load onto your system must contain each
8574 overlay's instructions, appearing at the overlay's load address, not its
8575 mapped address. However, each overlay's instructions must be relocated
8576 and its symbols defined as if the overlay were at its mapped address.
8577 You can use GNU linker scripts to specify different load and relocation
8578 addresses for pieces of your program; see @ref{Overlay Description,,,
8579 ld.info, Using ld: the GNU linker}.
8582 The procedure for loading executable files onto your system must be able
8583 to load their contents into the larger address space as well as the
8584 instruction and data spaces.
8588 The overlay system described above is rather simple, and could be
8589 improved in many ways:
8594 If your system has suitable bank switch registers or memory management
8595 hardware, you could use those facilities to make an overlay's load area
8596 contents simply appear at their mapped address in instruction space.
8597 This would probably be faster than copying the overlay to its mapped
8598 area in the usual way.
8601 If your overlays are small enough, you could set aside more than one
8602 overlay area, and have more than one overlay mapped at a time.
8605 You can use overlays to manage data, as well as instructions. In
8606 general, data overlays are even less transparent to your design than
8607 code overlays: whereas code overlays only require care when you call or
8608 return to functions, data overlays require care every time you access
8609 the data. Also, if you change the contents of a data overlay, you
8610 must copy its contents back out to its load address before you can copy a
8611 different data overlay into the same mapped area.
8616 @node Overlay Commands
8617 @section Overlay Commands
8619 To use @value{GDBN}'s overlay support, each overlay in your program must
8620 correspond to a separate section of the executable file. The section's
8621 virtual memory address and load memory address must be the overlay's
8622 mapped and load addresses. Identifying overlays with sections allows
8623 @value{GDBN} to determine the appropriate address of a function or
8624 variable, depending on whether the overlay is mapped or not.
8626 @value{GDBN}'s overlay commands all start with the word @code{overlay};
8627 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
8632 Disable @value{GDBN}'s overlay support. When overlay support is
8633 disabled, @value{GDBN} assumes that all functions and variables are
8634 always present at their mapped addresses. By default, @value{GDBN}'s
8635 overlay support is disabled.
8637 @item overlay manual
8638 @cindex manual overlay debugging
8639 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
8640 relies on you to tell it which overlays are mapped, and which are not,
8641 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
8642 commands described below.
8644 @item overlay map-overlay @var{overlay}
8645 @itemx overlay map @var{overlay}
8646 @cindex map an overlay
8647 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
8648 be the name of the object file section containing the overlay. When an
8649 overlay is mapped, @value{GDBN} assumes it can find the overlay's
8650 functions and variables at their mapped addresses. @value{GDBN} assumes
8651 that any other overlays whose mapped ranges overlap that of
8652 @var{overlay} are now unmapped.
8654 @item overlay unmap-overlay @var{overlay}
8655 @itemx overlay unmap @var{overlay}
8656 @cindex unmap an overlay
8657 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
8658 must be the name of the object file section containing the overlay.
8659 When an overlay is unmapped, @value{GDBN} assumes it can find the
8660 overlay's functions and variables at their load addresses.
8663 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
8664 consults a data structure the overlay manager maintains in the inferior
8665 to see which overlays are mapped. For details, see @ref{Automatic
8668 @item overlay load-target
8670 @cindex reloading the overlay table
8671 Re-read the overlay table from the inferior. Normally, @value{GDBN}
8672 re-reads the table @value{GDBN} automatically each time the inferior
8673 stops, so this command should only be necessary if you have changed the
8674 overlay mapping yourself using @value{GDBN}. This command is only
8675 useful when using automatic overlay debugging.
8677 @item overlay list-overlays
8679 @cindex listing mapped overlays
8680 Display a list of the overlays currently mapped, along with their mapped
8681 addresses, load addresses, and sizes.
8685 Normally, when @value{GDBN} prints a code address, it includes the name
8686 of the function the address falls in:
8689 (@value{GDBP}) print main
8690 $3 = @{int ()@} 0x11a0 <main>
8693 When overlay debugging is enabled, @value{GDBN} recognizes code in
8694 unmapped overlays, and prints the names of unmapped functions with
8695 asterisks around them. For example, if @code{foo} is a function in an
8696 unmapped overlay, @value{GDBN} prints it this way:
8699 (@value{GDBP}) overlay list
8700 No sections are mapped.
8701 (@value{GDBP}) print foo
8702 $5 = @{int (int)@} 0x100000 <*foo*>
8705 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
8709 (@value{GDBP}) overlay list
8710 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
8711 mapped at 0x1016 - 0x104a
8712 (@value{GDBP}) print foo
8713 $6 = @{int (int)@} 0x1016 <foo>
8716 When overlay debugging is enabled, @value{GDBN} can find the correct
8717 address for functions and variables in an overlay, whether or not the
8718 overlay is mapped. This allows most @value{GDBN} commands, like
8719 @code{break} and @code{disassemble}, to work normally, even on unmapped
8720 code. However, @value{GDBN}'s breakpoint support has some limitations:
8724 @cindex breakpoints in overlays
8725 @cindex overlays, setting breakpoints in
8726 You can set breakpoints in functions in unmapped overlays, as long as
8727 @value{GDBN} can write to the overlay at its load address.
8729 @value{GDBN} can not set hardware or simulator-based breakpoints in
8730 unmapped overlays. However, if you set a breakpoint at the end of your
8731 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
8732 you are using manual overlay management), @value{GDBN} will re-set its
8733 breakpoints properly.
8737 @node Automatic Overlay Debugging
8738 @section Automatic Overlay Debugging
8739 @cindex automatic overlay debugging
8741 @value{GDBN} can automatically track which overlays are mapped and which
8742 are not, given some simple co-operation from the overlay manager in the
8743 inferior. If you enable automatic overlay debugging with the
8744 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
8745 looks in the inferior's memory for certain variables describing the
8746 current state of the overlays.
8748 Here are the variables your overlay manager must define to support
8749 @value{GDBN}'s automatic overlay debugging:
8753 @item @code{_ovly_table}:
8754 This variable must be an array of the following structures:
8759 /* The overlay's mapped address. */
8762 /* The size of the overlay, in bytes. */
8765 /* The overlay's load address. */
8768 /* Non-zero if the overlay is currently mapped;
8770 unsigned long mapped;
8774 @item @code{_novlys}:
8775 This variable must be a four-byte signed integer, holding the total
8776 number of elements in @code{_ovly_table}.
8780 To decide whether a particular overlay is mapped or not, @value{GDBN}
8781 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
8782 @code{lma} members equal the VMA and LMA of the overlay's section in the
8783 executable file. When @value{GDBN} finds a matching entry, it consults
8784 the entry's @code{mapped} member to determine whether the overlay is
8787 In addition, your overlay manager may define a function called
8788 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
8789 will silently set a breakpoint there. If the overlay manager then
8790 calls this function whenever it has changed the overlay table, this
8791 will enable @value{GDBN} to accurately keep track of which overlays
8792 are in program memory, and update any breakpoints that may be set
8793 in overlays. This will allow breakpoints to work even if the
8794 overlays are kept in ROM or other non-writable memory while they
8795 are not being executed.
8797 @node Overlay Sample Program
8798 @section Overlay Sample Program
8799 @cindex overlay example program
8801 When linking a program which uses overlays, you must place the overlays
8802 at their load addresses, while relocating them to run at their mapped
8803 addresses. To do this, you must write a linker script (@pxref{Overlay
8804 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
8805 since linker scripts are specific to a particular host system, target
8806 architecture, and target memory layout, this manual cannot provide
8807 portable sample code demonstrating @value{GDBN}'s overlay support.
8809 However, the @value{GDBN} source distribution does contain an overlaid
8810 program, with linker scripts for a few systems, as part of its test
8811 suite. The program consists of the following files from
8812 @file{gdb/testsuite/gdb.base}:
8816 The main program file.
8818 A simple overlay manager, used by @file{overlays.c}.
8823 Overlay modules, loaded and used by @file{overlays.c}.
8826 Linker scripts for linking the test program on the @code{d10v-elf}
8827 and @code{m32r-elf} targets.
8830 You can build the test program using the @code{d10v-elf} GCC
8831 cross-compiler like this:
8834 $ d10v-elf-gcc -g -c overlays.c
8835 $ d10v-elf-gcc -g -c ovlymgr.c
8836 $ d10v-elf-gcc -g -c foo.c
8837 $ d10v-elf-gcc -g -c bar.c
8838 $ d10v-elf-gcc -g -c baz.c
8839 $ d10v-elf-gcc -g -c grbx.c
8840 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
8841 baz.o grbx.o -Wl,-Td10v.ld -o overlays
8844 The build process is identical for any other architecture, except that
8845 you must substitute the appropriate compiler and linker script for the
8846 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
8850 @chapter Using @value{GDBN} with Different Languages
8853 Although programming languages generally have common aspects, they are
8854 rarely expressed in the same manner. For instance, in ANSI C,
8855 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
8856 Modula-2, it is accomplished by @code{p^}. Values can also be
8857 represented (and displayed) differently. Hex numbers in C appear as
8858 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
8860 @cindex working language
8861 Language-specific information is built into @value{GDBN} for some languages,
8862 allowing you to express operations like the above in your program's
8863 native language, and allowing @value{GDBN} to output values in a manner
8864 consistent with the syntax of your program's native language. The
8865 language you use to build expressions is called the @dfn{working
8869 * Setting:: Switching between source languages
8870 * Show:: Displaying the language
8871 * Checks:: Type and range checks
8872 * Supported Languages:: Supported languages
8873 * Unsupported Languages:: Unsupported languages
8877 @section Switching Between Source Languages
8879 There are two ways to control the working language---either have @value{GDBN}
8880 set it automatically, or select it manually yourself. You can use the
8881 @code{set language} command for either purpose. On startup, @value{GDBN}
8882 defaults to setting the language automatically. The working language is
8883 used to determine how expressions you type are interpreted, how values
8886 In addition to the working language, every source file that
8887 @value{GDBN} knows about has its own working language. For some object
8888 file formats, the compiler might indicate which language a particular
8889 source file is in. However, most of the time @value{GDBN} infers the
8890 language from the name of the file. The language of a source file
8891 controls whether C@t{++} names are demangled---this way @code{backtrace} can
8892 show each frame appropriately for its own language. There is no way to
8893 set the language of a source file from within @value{GDBN}, but you can
8894 set the language associated with a filename extension. @xref{Show, ,
8895 Displaying the Language}.
8897 This is most commonly a problem when you use a program, such
8898 as @code{cfront} or @code{f2c}, that generates C but is written in
8899 another language. In that case, make the
8900 program use @code{#line} directives in its C output; that way
8901 @value{GDBN} will know the correct language of the source code of the original
8902 program, and will display that source code, not the generated C code.
8905 * Filenames:: Filename extensions and languages.
8906 * Manually:: Setting the working language manually
8907 * Automatically:: Having @value{GDBN} infer the source language
8911 @subsection List of Filename Extensions and Languages
8913 If a source file name ends in one of the following extensions, then
8914 @value{GDBN} infers that its language is the one indicated.
8935 Objective-C source file
8942 Modula-2 source file
8946 Assembler source file. This actually behaves almost like C, but
8947 @value{GDBN} does not skip over function prologues when stepping.
8950 In addition, you may set the language associated with a filename
8951 extension. @xref{Show, , Displaying the Language}.
8954 @subsection Setting the Working Language
8956 If you allow @value{GDBN} to set the language automatically,
8957 expressions are interpreted the same way in your debugging session and
8960 @kindex set language
8961 If you wish, you may set the language manually. To do this, issue the
8962 command @samp{set language @var{lang}}, where @var{lang} is the name of
8964 @code{c} or @code{modula-2}.
8965 For a list of the supported languages, type @samp{set language}.
8967 Setting the language manually prevents @value{GDBN} from updating the working
8968 language automatically. This can lead to confusion if you try
8969 to debug a program when the working language is not the same as the
8970 source language, when an expression is acceptable to both
8971 languages---but means different things. For instance, if the current
8972 source file were written in C, and @value{GDBN} was parsing Modula-2, a
8980 might not have the effect you intended. In C, this means to add
8981 @code{b} and @code{c} and place the result in @code{a}. The result
8982 printed would be the value of @code{a}. In Modula-2, this means to compare
8983 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
8986 @subsection Having @value{GDBN} Infer the Source Language
8988 To have @value{GDBN} set the working language automatically, use
8989 @samp{set language local} or @samp{set language auto}. @value{GDBN}
8990 then infers the working language. That is, when your program stops in a
8991 frame (usually by encountering a breakpoint), @value{GDBN} sets the
8992 working language to the language recorded for the function in that
8993 frame. If the language for a frame is unknown (that is, if the function
8994 or block corresponding to the frame was defined in a source file that
8995 does not have a recognized extension), the current working language is
8996 not changed, and @value{GDBN} issues a warning.
8998 This may not seem necessary for most programs, which are written
8999 entirely in one source language. However, program modules and libraries
9000 written in one source language can be used by a main program written in
9001 a different source language. Using @samp{set language auto} in this
9002 case frees you from having to set the working language manually.
9005 @section Displaying the Language
9007 The following commands help you find out which language is the
9008 working language, and also what language source files were written in.
9012 @kindex show language
9013 Display the current working language. This is the
9014 language you can use with commands such as @code{print} to
9015 build and compute expressions that may involve variables in your program.
9018 @kindex info frame@r{, show the source language}
9019 Display the source language for this frame. This language becomes the
9020 working language if you use an identifier from this frame.
9021 @xref{Frame Info, ,Information about a Frame}, to identify the other
9022 information listed here.
9025 @kindex info source@r{, show the source language}
9026 Display the source language of this source file.
9027 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
9028 information listed here.
9031 In unusual circumstances, you may have source files with extensions
9032 not in the standard list. You can then set the extension associated
9033 with a language explicitly:
9036 @item set extension-language @var{ext} @var{language}
9037 @kindex set extension-language
9038 Tell @value{GDBN} that source files with extension @var{ext} are to be
9039 assumed as written in the source language @var{language}.
9041 @item info extensions
9042 @kindex info extensions
9043 List all the filename extensions and the associated languages.
9047 @section Type and Range Checking
9050 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
9051 checking are included, but they do not yet have any effect. This
9052 section documents the intended facilities.
9054 @c FIXME remove warning when type/range code added
9056 Some languages are designed to guard you against making seemingly common
9057 errors through a series of compile- and run-time checks. These include
9058 checking the type of arguments to functions and operators, and making
9059 sure mathematical overflows are caught at run time. Checks such as
9060 these help to ensure a program's correctness once it has been compiled
9061 by eliminating type mismatches, and providing active checks for range
9062 errors when your program is running.
9064 @value{GDBN} can check for conditions like the above if you wish.
9065 Although @value{GDBN} does not check the statements in your program,
9066 it can check expressions entered directly into @value{GDBN} for
9067 evaluation via the @code{print} command, for example. As with the
9068 working language, @value{GDBN} can also decide whether or not to check
9069 automatically based on your program's source language.
9070 @xref{Supported Languages, ,Supported Languages}, for the default
9071 settings of supported languages.
9074 * Type Checking:: An overview of type checking
9075 * Range Checking:: An overview of range checking
9078 @cindex type checking
9079 @cindex checks, type
9081 @subsection An Overview of Type Checking
9083 Some languages, such as Modula-2, are strongly typed, meaning that the
9084 arguments to operators and functions have to be of the correct type,
9085 otherwise an error occurs. These checks prevent type mismatch
9086 errors from ever causing any run-time problems. For example,
9094 The second example fails because the @code{CARDINAL} 1 is not
9095 type-compatible with the @code{REAL} 2.3.
9097 For the expressions you use in @value{GDBN} commands, you can tell the
9098 @value{GDBN} type checker to skip checking;
9099 to treat any mismatches as errors and abandon the expression;
9100 or to only issue warnings when type mismatches occur,
9101 but evaluate the expression anyway. When you choose the last of
9102 these, @value{GDBN} evaluates expressions like the second example above, but
9103 also issues a warning.
9105 Even if you turn type checking off, there may be other reasons
9106 related to type that prevent @value{GDBN} from evaluating an expression.
9107 For instance, @value{GDBN} does not know how to add an @code{int} and
9108 a @code{struct foo}. These particular type errors have nothing to do
9109 with the language in use, and usually arise from expressions, such as
9110 the one described above, which make little sense to evaluate anyway.
9112 Each language defines to what degree it is strict about type. For
9113 instance, both Modula-2 and C require the arguments to arithmetical
9114 operators to be numbers. In C, enumerated types and pointers can be
9115 represented as numbers, so that they are valid arguments to mathematical
9116 operators. @xref{Supported Languages, ,Supported Languages}, for further
9117 details on specific languages.
9119 @value{GDBN} provides some additional commands for controlling the type checker:
9121 @kindex set check type
9122 @kindex show check type
9124 @item set check type auto
9125 Set type checking on or off based on the current working language.
9126 @xref{Supported Languages, ,Supported Languages}, for the default settings for
9129 @item set check type on
9130 @itemx set check type off
9131 Set type checking on or off, overriding the default setting for the
9132 current working language. Issue a warning if the setting does not
9133 match the language default. If any type mismatches occur in
9134 evaluating an expression while type checking is on, @value{GDBN} prints a
9135 message and aborts evaluation of the expression.
9137 @item set check type warn
9138 Cause the type checker to issue warnings, but to always attempt to
9139 evaluate the expression. Evaluating the expression may still
9140 be impossible for other reasons. For example, @value{GDBN} cannot add
9141 numbers and structures.
9144 Show the current setting of the type checker, and whether or not @value{GDBN}
9145 is setting it automatically.
9148 @cindex range checking
9149 @cindex checks, range
9150 @node Range Checking
9151 @subsection An Overview of Range Checking
9153 In some languages (such as Modula-2), it is an error to exceed the
9154 bounds of a type; this is enforced with run-time checks. Such range
9155 checking is meant to ensure program correctness by making sure
9156 computations do not overflow, or indices on an array element access do
9157 not exceed the bounds of the array.
9159 For expressions you use in @value{GDBN} commands, you can tell
9160 @value{GDBN} to treat range errors in one of three ways: ignore them,
9161 always treat them as errors and abandon the expression, or issue
9162 warnings but evaluate the expression anyway.
9164 A range error can result from numerical overflow, from exceeding an
9165 array index bound, or when you type a constant that is not a member
9166 of any type. Some languages, however, do not treat overflows as an
9167 error. In many implementations of C, mathematical overflow causes the
9168 result to ``wrap around'' to lower values---for example, if @var{m} is
9169 the largest integer value, and @var{s} is the smallest, then
9172 @var{m} + 1 @result{} @var{s}
9175 This, too, is specific to individual languages, and in some cases
9176 specific to individual compilers or machines. @xref{Supported Languages, ,
9177 Supported Languages}, for further details on specific languages.
9179 @value{GDBN} provides some additional commands for controlling the range checker:
9181 @kindex set check range
9182 @kindex show check range
9184 @item set check range auto
9185 Set range checking on or off based on the current working language.
9186 @xref{Supported Languages, ,Supported Languages}, for the default settings for
9189 @item set check range on
9190 @itemx set check range off
9191 Set range checking on or off, overriding the default setting for the
9192 current working language. A warning is issued if the setting does not
9193 match the language default. If a range error occurs and range checking is on,
9194 then a message is printed and evaluation of the expression is aborted.
9196 @item set check range warn
9197 Output messages when the @value{GDBN} range checker detects a range error,
9198 but attempt to evaluate the expression anyway. Evaluating the
9199 expression may still be impossible for other reasons, such as accessing
9200 memory that the process does not own (a typical example from many Unix
9204 Show the current setting of the range checker, and whether or not it is
9205 being set automatically by @value{GDBN}.
9208 @node Supported Languages
9209 @section Supported Languages
9211 @value{GDBN} supports C, C@t{++}, Objective-C, Fortran, Java, Pascal,
9212 assembly, Modula-2, and Ada.
9213 @c This is false ...
9214 Some @value{GDBN} features may be used in expressions regardless of the
9215 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
9216 and the @samp{@{type@}addr} construct (@pxref{Expressions,
9217 ,Expressions}) can be used with the constructs of any supported
9220 The following sections detail to what degree each source language is
9221 supported by @value{GDBN}. These sections are not meant to be language
9222 tutorials or references, but serve only as a reference guide to what the
9223 @value{GDBN} expression parser accepts, and what input and output
9224 formats should look like for different languages. There are many good
9225 books written on each of these languages; please look to these for a
9226 language reference or tutorial.
9230 * Objective-C:: Objective-C
9233 * Modula-2:: Modula-2
9238 @subsection C and C@t{++}
9240 @cindex C and C@t{++}
9241 @cindex expressions in C or C@t{++}
9243 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
9244 to both languages. Whenever this is the case, we discuss those languages
9248 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
9249 @cindex @sc{gnu} C@t{++}
9250 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
9251 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
9252 effectively, you must compile your C@t{++} programs with a supported
9253 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
9254 compiler (@code{aCC}).
9256 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
9257 format; if it doesn't work on your system, try the stabs+ debugging
9258 format. You can select those formats explicitly with the @code{g++}
9259 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
9260 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
9261 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
9264 * C Operators:: C and C@t{++} operators
9265 * C Constants:: C and C@t{++} constants
9266 * C Plus Plus Expressions:: C@t{++} expressions
9267 * C Defaults:: Default settings for C and C@t{++}
9268 * C Checks:: C and C@t{++} type and range checks
9269 * Debugging C:: @value{GDBN} and C
9270 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
9271 * Decimal Floating Point:: Numbers in Decimal Floating Point format
9275 @subsubsection C and C@t{++} Operators
9277 @cindex C and C@t{++} operators
9279 Operators must be defined on values of specific types. For instance,
9280 @code{+} is defined on numbers, but not on structures. Operators are
9281 often defined on groups of types.
9283 For the purposes of C and C@t{++}, the following definitions hold:
9288 @emph{Integral types} include @code{int} with any of its storage-class
9289 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
9292 @emph{Floating-point types} include @code{float}, @code{double}, and
9293 @code{long double} (if supported by the target platform).
9296 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
9299 @emph{Scalar types} include all of the above.
9304 The following operators are supported. They are listed here
9305 in order of increasing precedence:
9309 The comma or sequencing operator. Expressions in a comma-separated list
9310 are evaluated from left to right, with the result of the entire
9311 expression being the last expression evaluated.
9314 Assignment. The value of an assignment expression is the value
9315 assigned. Defined on scalar types.
9318 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
9319 and translated to @w{@code{@var{a} = @var{a op b}}}.
9320 @w{@code{@var{op}=}} and @code{=} have the same precedence.
9321 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
9322 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
9325 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
9326 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
9330 Logical @sc{or}. Defined on integral types.
9333 Logical @sc{and}. Defined on integral types.
9336 Bitwise @sc{or}. Defined on integral types.
9339 Bitwise exclusive-@sc{or}. Defined on integral types.
9342 Bitwise @sc{and}. Defined on integral types.
9345 Equality and inequality. Defined on scalar types. The value of these
9346 expressions is 0 for false and non-zero for true.
9348 @item <@r{, }>@r{, }<=@r{, }>=
9349 Less than, greater than, less than or equal, greater than or equal.
9350 Defined on scalar types. The value of these expressions is 0 for false
9351 and non-zero for true.
9354 left shift, and right shift. Defined on integral types.
9357 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
9360 Addition and subtraction. Defined on integral types, floating-point types and
9363 @item *@r{, }/@r{, }%
9364 Multiplication, division, and modulus. Multiplication and division are
9365 defined on integral and floating-point types. Modulus is defined on
9369 Increment and decrement. When appearing before a variable, the
9370 operation is performed before the variable is used in an expression;
9371 when appearing after it, the variable's value is used before the
9372 operation takes place.
9375 Pointer dereferencing. Defined on pointer types. Same precedence as
9379 Address operator. Defined on variables. Same precedence as @code{++}.
9381 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
9382 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
9383 to examine the address
9384 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
9388 Negative. Defined on integral and floating-point types. Same
9389 precedence as @code{++}.
9392 Logical negation. Defined on integral types. Same precedence as
9396 Bitwise complement operator. Defined on integral types. Same precedence as
9401 Structure member, and pointer-to-structure member. For convenience,
9402 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
9403 pointer based on the stored type information.
9404 Defined on @code{struct} and @code{union} data.
9407 Dereferences of pointers to members.
9410 Array indexing. @code{@var{a}[@var{i}]} is defined as
9411 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
9414 Function parameter list. Same precedence as @code{->}.
9417 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
9418 and @code{class} types.
9421 Doubled colons also represent the @value{GDBN} scope operator
9422 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
9426 If an operator is redefined in the user code, @value{GDBN} usually
9427 attempts to invoke the redefined version instead of using the operator's
9431 @subsubsection C and C@t{++} Constants
9433 @cindex C and C@t{++} constants
9435 @value{GDBN} allows you to express the constants of C and C@t{++} in the
9440 Integer constants are a sequence of digits. Octal constants are
9441 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
9442 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
9443 @samp{l}, specifying that the constant should be treated as a
9447 Floating point constants are a sequence of digits, followed by a decimal
9448 point, followed by a sequence of digits, and optionally followed by an
9449 exponent. An exponent is of the form:
9450 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
9451 sequence of digits. The @samp{+} is optional for positive exponents.
9452 A floating-point constant may also end with a letter @samp{f} or
9453 @samp{F}, specifying that the constant should be treated as being of
9454 the @code{float} (as opposed to the default @code{double}) type; or with
9455 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
9459 Enumerated constants consist of enumerated identifiers, or their
9460 integral equivalents.
9463 Character constants are a single character surrounded by single quotes
9464 (@code{'}), or a number---the ordinal value of the corresponding character
9465 (usually its @sc{ascii} value). Within quotes, the single character may
9466 be represented by a letter or by @dfn{escape sequences}, which are of
9467 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
9468 of the character's ordinal value; or of the form @samp{\@var{x}}, where
9469 @samp{@var{x}} is a predefined special character---for example,
9470 @samp{\n} for newline.
9473 String constants are a sequence of character constants surrounded by
9474 double quotes (@code{"}). Any valid character constant (as described
9475 above) may appear. Double quotes within the string must be preceded by
9476 a backslash, so for instance @samp{"a\"b'c"} is a string of five
9480 Pointer constants are an integral value. You can also write pointers
9481 to constants using the C operator @samp{&}.
9484 Array constants are comma-separated lists surrounded by braces @samp{@{}
9485 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
9486 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
9487 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
9490 @node C Plus Plus Expressions
9491 @subsubsection C@t{++} Expressions
9493 @cindex expressions in C@t{++}
9494 @value{GDBN} expression handling can interpret most C@t{++} expressions.
9496 @cindex debugging C@t{++} programs
9497 @cindex C@t{++} compilers
9498 @cindex debug formats and C@t{++}
9499 @cindex @value{NGCC} and C@t{++}
9501 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
9502 proper compiler and the proper debug format. Currently, @value{GDBN}
9503 works best when debugging C@t{++} code that is compiled with
9504 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
9505 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
9506 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
9507 stabs+ as their default debug format, so you usually don't need to
9508 specify a debug format explicitly. Other compilers and/or debug formats
9509 are likely to work badly or not at all when using @value{GDBN} to debug
9515 @cindex member functions
9517 Member function calls are allowed; you can use expressions like
9520 count = aml->GetOriginal(x, y)
9523 @vindex this@r{, inside C@t{++} member functions}
9524 @cindex namespace in C@t{++}
9526 While a member function is active (in the selected stack frame), your
9527 expressions have the same namespace available as the member function;
9528 that is, @value{GDBN} allows implicit references to the class instance
9529 pointer @code{this} following the same rules as C@t{++}.
9531 @cindex call overloaded functions
9532 @cindex overloaded functions, calling
9533 @cindex type conversions in C@t{++}
9535 You can call overloaded functions; @value{GDBN} resolves the function
9536 call to the right definition, with some restrictions. @value{GDBN} does not
9537 perform overload resolution involving user-defined type conversions,
9538 calls to constructors, or instantiations of templates that do not exist
9539 in the program. It also cannot handle ellipsis argument lists or
9542 It does perform integral conversions and promotions, floating-point
9543 promotions, arithmetic conversions, pointer conversions, conversions of
9544 class objects to base classes, and standard conversions such as those of
9545 functions or arrays to pointers; it requires an exact match on the
9546 number of function arguments.
9548 Overload resolution is always performed, unless you have specified
9549 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
9550 ,@value{GDBN} Features for C@t{++}}.
9552 You must specify @code{set overload-resolution off} in order to use an
9553 explicit function signature to call an overloaded function, as in
9555 p 'foo(char,int)'('x', 13)
9558 The @value{GDBN} command-completion facility can simplify this;
9559 see @ref{Completion, ,Command Completion}.
9561 @cindex reference declarations
9563 @value{GDBN} understands variables declared as C@t{++} references; you can use
9564 them in expressions just as you do in C@t{++} source---they are automatically
9567 In the parameter list shown when @value{GDBN} displays a frame, the values of
9568 reference variables are not displayed (unlike other variables); this
9569 avoids clutter, since references are often used for large structures.
9570 The @emph{address} of a reference variable is always shown, unless
9571 you have specified @samp{set print address off}.
9574 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
9575 expressions can use it just as expressions in your program do. Since
9576 one scope may be defined in another, you can use @code{::} repeatedly if
9577 necessary, for example in an expression like
9578 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
9579 resolving name scope by reference to source files, in both C and C@t{++}
9580 debugging (@pxref{Variables, ,Program Variables}).
9583 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
9584 calling virtual functions correctly, printing out virtual bases of
9585 objects, calling functions in a base subobject, casting objects, and
9586 invoking user-defined operators.
9589 @subsubsection C and C@t{++} Defaults
9591 @cindex C and C@t{++} defaults
9593 If you allow @value{GDBN} to set type and range checking automatically, they
9594 both default to @code{off} whenever the working language changes to
9595 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
9596 selects the working language.
9598 If you allow @value{GDBN} to set the language automatically, it
9599 recognizes source files whose names end with @file{.c}, @file{.C}, or
9600 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
9601 these files, it sets the working language to C or C@t{++}.
9602 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
9603 for further details.
9605 @c Type checking is (a) primarily motivated by Modula-2, and (b)
9606 @c unimplemented. If (b) changes, it might make sense to let this node
9607 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
9610 @subsubsection C and C@t{++} Type and Range Checks
9612 @cindex C and C@t{++} checks
9614 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
9615 is not used. However, if you turn type checking on, @value{GDBN}
9616 considers two variables type equivalent if:
9620 The two variables are structured and have the same structure, union, or
9624 The two variables have the same type name, or types that have been
9625 declared equivalent through @code{typedef}.
9628 @c leaving this out because neither J Gilmore nor R Pesch understand it.
9631 The two @code{struct}, @code{union}, or @code{enum} variables are
9632 declared in the same declaration. (Note: this may not be true for all C
9637 Range checking, if turned on, is done on mathematical operations. Array
9638 indices are not checked, since they are often used to index a pointer
9639 that is not itself an array.
9642 @subsubsection @value{GDBN} and C
9644 The @code{set print union} and @code{show print union} commands apply to
9645 the @code{union} type. When set to @samp{on}, any @code{union} that is
9646 inside a @code{struct} or @code{class} is also printed. Otherwise, it
9647 appears as @samp{@{...@}}.
9649 The @code{@@} operator aids in the debugging of dynamic arrays, formed
9650 with pointers and a memory allocation function. @xref{Expressions,
9653 @node Debugging C Plus Plus
9654 @subsubsection @value{GDBN} Features for C@t{++}
9656 @cindex commands for C@t{++}
9658 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
9659 designed specifically for use with C@t{++}. Here is a summary:
9662 @cindex break in overloaded functions
9663 @item @r{breakpoint menus}
9664 When you want a breakpoint in a function whose name is overloaded,
9665 @value{GDBN} has the capability to display a menu of possible breakpoint
9666 locations to help you specify which function definition you want.
9667 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
9669 @cindex overloading in C@t{++}
9670 @item rbreak @var{regex}
9671 Setting breakpoints using regular expressions is helpful for setting
9672 breakpoints on overloaded functions that are not members of any special
9674 @xref{Set Breaks, ,Setting Breakpoints}.
9676 @cindex C@t{++} exception handling
9679 Debug C@t{++} exception handling using these commands. @xref{Set
9680 Catchpoints, , Setting Catchpoints}.
9683 @item ptype @var{typename}
9684 Print inheritance relationships as well as other information for type
9686 @xref{Symbols, ,Examining the Symbol Table}.
9688 @cindex C@t{++} symbol display
9689 @item set print demangle
9690 @itemx show print demangle
9691 @itemx set print asm-demangle
9692 @itemx show print asm-demangle
9693 Control whether C@t{++} symbols display in their source form, both when
9694 displaying code as C@t{++} source and when displaying disassemblies.
9695 @xref{Print Settings, ,Print Settings}.
9697 @item set print object
9698 @itemx show print object
9699 Choose whether to print derived (actual) or declared types of objects.
9700 @xref{Print Settings, ,Print Settings}.
9702 @item set print vtbl
9703 @itemx show print vtbl
9704 Control the format for printing virtual function tables.
9705 @xref{Print Settings, ,Print Settings}.
9706 (The @code{vtbl} commands do not work on programs compiled with the HP
9707 ANSI C@t{++} compiler (@code{aCC}).)
9709 @kindex set overload-resolution
9710 @cindex overloaded functions, overload resolution
9711 @item set overload-resolution on
9712 Enable overload resolution for C@t{++} expression evaluation. The default
9713 is on. For overloaded functions, @value{GDBN} evaluates the arguments
9714 and searches for a function whose signature matches the argument types,
9715 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
9716 Expressions, ,C@t{++} Expressions}, for details).
9717 If it cannot find a match, it emits a message.
9719 @item set overload-resolution off
9720 Disable overload resolution for C@t{++} expression evaluation. For
9721 overloaded functions that are not class member functions, @value{GDBN}
9722 chooses the first function of the specified name that it finds in the
9723 symbol table, whether or not its arguments are of the correct type. For
9724 overloaded functions that are class member functions, @value{GDBN}
9725 searches for a function whose signature @emph{exactly} matches the
9728 @kindex show overload-resolution
9729 @item show overload-resolution
9730 Show the current setting of overload resolution.
9732 @item @r{Overloaded symbol names}
9733 You can specify a particular definition of an overloaded symbol, using
9734 the same notation that is used to declare such symbols in C@t{++}: type
9735 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
9736 also use the @value{GDBN} command-line word completion facilities to list the
9737 available choices, or to finish the type list for you.
9738 @xref{Completion,, Command Completion}, for details on how to do this.
9741 @node Decimal Floating Point
9742 @subsubsection Decimal Floating Point format
9743 @cindex decimal floating point format
9745 @value{GDBN} can examine, set and perform computations with numbers in
9746 decimal floating point format, which in the C language correspond to the
9747 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
9748 specified by the extension to support decimal floating-point arithmetic.
9750 There are two encodings in use, depending on the architecture: BID (Binary
9751 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
9752 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
9755 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
9756 to manipulate decimal floating point numbers, it is not possible to convert
9757 (using a cast, for example) integers wider than 32-bit to decimal float.
9759 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
9760 point computations, error checking in decimal float operations ignores
9761 underflow, overflow and divide by zero exceptions.
9763 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
9764 to inspect @code{_Decimal128} values stored in floating point registers. See
9765 @ref{PowerPC,,PowerPC} for more details.
9768 @subsection Objective-C
9771 This section provides information about some commands and command
9772 options that are useful for debugging Objective-C code. See also
9773 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
9774 few more commands specific to Objective-C support.
9777 * Method Names in Commands::
9778 * The Print Command with Objective-C::
9781 @node Method Names in Commands
9782 @subsubsection Method Names in Commands
9784 The following commands have been extended to accept Objective-C method
9785 names as line specifications:
9787 @kindex clear@r{, and Objective-C}
9788 @kindex break@r{, and Objective-C}
9789 @kindex info line@r{, and Objective-C}
9790 @kindex jump@r{, and Objective-C}
9791 @kindex list@r{, and Objective-C}
9795 @item @code{info line}
9800 A fully qualified Objective-C method name is specified as
9803 -[@var{Class} @var{methodName}]
9806 where the minus sign is used to indicate an instance method and a
9807 plus sign (not shown) is used to indicate a class method. The class
9808 name @var{Class} and method name @var{methodName} are enclosed in
9809 brackets, similar to the way messages are specified in Objective-C
9810 source code. For example, to set a breakpoint at the @code{create}
9811 instance method of class @code{Fruit} in the program currently being
9815 break -[Fruit create]
9818 To list ten program lines around the @code{initialize} class method,
9822 list +[NSText initialize]
9825 In the current version of @value{GDBN}, the plus or minus sign is
9826 required. In future versions of @value{GDBN}, the plus or minus
9827 sign will be optional, but you can use it to narrow the search. It
9828 is also possible to specify just a method name:
9834 You must specify the complete method name, including any colons. If
9835 your program's source files contain more than one @code{create} method,
9836 you'll be presented with a numbered list of classes that implement that
9837 method. Indicate your choice by number, or type @samp{0} to exit if
9840 As another example, to clear a breakpoint established at the
9841 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
9844 clear -[NSWindow makeKeyAndOrderFront:]
9847 @node The Print Command with Objective-C
9848 @subsubsection The Print Command With Objective-C
9849 @cindex Objective-C, print objects
9850 @kindex print-object
9851 @kindex po @r{(@code{print-object})}
9853 The print command has also been extended to accept methods. For example:
9856 print -[@var{object} hash]
9859 @cindex print an Objective-C object description
9860 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
9862 will tell @value{GDBN} to send the @code{hash} message to @var{object}
9863 and print the result. Also, an additional command has been added,
9864 @code{print-object} or @code{po} for short, which is meant to print
9865 the description of an object. However, this command may only work
9866 with certain Objective-C libraries that have a particular hook
9867 function, @code{_NSPrintForDebugger}, defined.
9871 @cindex Fortran-specific support in @value{GDBN}
9873 @value{GDBN} can be used to debug programs written in Fortran, but it
9874 currently supports only the features of Fortran 77 language.
9876 @cindex trailing underscore, in Fortran symbols
9877 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
9878 among them) append an underscore to the names of variables and
9879 functions. When you debug programs compiled by those compilers, you
9880 will need to refer to variables and functions with a trailing
9884 * Fortran Operators:: Fortran operators and expressions
9885 * Fortran Defaults:: Default settings for Fortran
9886 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
9889 @node Fortran Operators
9890 @subsubsection Fortran Operators and Expressions
9892 @cindex Fortran operators and expressions
9894 Operators must be defined on values of specific types. For instance,
9895 @code{+} is defined on numbers, but not on characters or other non-
9896 arithmetic types. Operators are often defined on groups of types.
9900 The exponentiation operator. It raises the first operand to the power
9904 The range operator. Normally used in the form of array(low:high) to
9905 represent a section of array.
9908 The access component operator. Normally used to access elements in derived
9909 types. Also suitable for unions. As unions aren't part of regular Fortran,
9910 this can only happen when accessing a register that uses a gdbarch-defined
9914 @node Fortran Defaults
9915 @subsubsection Fortran Defaults
9917 @cindex Fortran Defaults
9919 Fortran symbols are usually case-insensitive, so @value{GDBN} by
9920 default uses case-insensitive matches for Fortran symbols. You can
9921 change that with the @samp{set case-insensitive} command, see
9922 @ref{Symbols}, for the details.
9924 @node Special Fortran Commands
9925 @subsubsection Special Fortran Commands
9927 @cindex Special Fortran commands
9929 @value{GDBN} has some commands to support Fortran-specific features,
9930 such as displaying common blocks.
9933 @cindex @code{COMMON} blocks, Fortran
9935 @item info common @r{[}@var{common-name}@r{]}
9936 This command prints the values contained in the Fortran @code{COMMON}
9937 block whose name is @var{common-name}. With no argument, the names of
9938 all @code{COMMON} blocks visible at the current program location are
9945 @cindex Pascal support in @value{GDBN}, limitations
9946 Debugging Pascal programs which use sets, subranges, file variables, or
9947 nested functions does not currently work. @value{GDBN} does not support
9948 entering expressions, printing values, or similar features using Pascal
9951 The Pascal-specific command @code{set print pascal_static-members}
9952 controls whether static members of Pascal objects are displayed.
9953 @xref{Print Settings, pascal_static-members}.
9956 @subsection Modula-2
9958 @cindex Modula-2, @value{GDBN} support
9960 The extensions made to @value{GDBN} to support Modula-2 only support
9961 output from the @sc{gnu} Modula-2 compiler (which is currently being
9962 developed). Other Modula-2 compilers are not currently supported, and
9963 attempting to debug executables produced by them is most likely
9964 to give an error as @value{GDBN} reads in the executable's symbol
9967 @cindex expressions in Modula-2
9969 * M2 Operators:: Built-in operators
9970 * Built-In Func/Proc:: Built-in functions and procedures
9971 * M2 Constants:: Modula-2 constants
9972 * M2 Types:: Modula-2 types
9973 * M2 Defaults:: Default settings for Modula-2
9974 * Deviations:: Deviations from standard Modula-2
9975 * M2 Checks:: Modula-2 type and range checks
9976 * M2 Scope:: The scope operators @code{::} and @code{.}
9977 * GDB/M2:: @value{GDBN} and Modula-2
9981 @subsubsection Operators
9982 @cindex Modula-2 operators
9984 Operators must be defined on values of specific types. For instance,
9985 @code{+} is defined on numbers, but not on structures. Operators are
9986 often defined on groups of types. For the purposes of Modula-2, the
9987 following definitions hold:
9992 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
9996 @emph{Character types} consist of @code{CHAR} and its subranges.
9999 @emph{Floating-point types} consist of @code{REAL}.
10002 @emph{Pointer types} consist of anything declared as @code{POINTER TO
10006 @emph{Scalar types} consist of all of the above.
10009 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
10012 @emph{Boolean types} consist of @code{BOOLEAN}.
10016 The following operators are supported, and appear in order of
10017 increasing precedence:
10021 Function argument or array index separator.
10024 Assignment. The value of @var{var} @code{:=} @var{value} is
10028 Less than, greater than on integral, floating-point, or enumerated
10032 Less than or equal to, greater than or equal to
10033 on integral, floating-point and enumerated types, or set inclusion on
10034 set types. Same precedence as @code{<}.
10036 @item =@r{, }<>@r{, }#
10037 Equality and two ways of expressing inequality, valid on scalar types.
10038 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
10039 available for inequality, since @code{#} conflicts with the script
10043 Set membership. Defined on set types and the types of their members.
10044 Same precedence as @code{<}.
10047 Boolean disjunction. Defined on boolean types.
10050 Boolean conjunction. Defined on boolean types.
10053 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
10056 Addition and subtraction on integral and floating-point types, or union
10057 and difference on set types.
10060 Multiplication on integral and floating-point types, or set intersection
10064 Division on floating-point types, or symmetric set difference on set
10065 types. Same precedence as @code{*}.
10068 Integer division and remainder. Defined on integral types. Same
10069 precedence as @code{*}.
10072 Negative. Defined on @code{INTEGER} and @code{REAL} data.
10075 Pointer dereferencing. Defined on pointer types.
10078 Boolean negation. Defined on boolean types. Same precedence as
10082 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
10083 precedence as @code{^}.
10086 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
10089 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
10093 @value{GDBN} and Modula-2 scope operators.
10097 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
10098 treats the use of the operator @code{IN}, or the use of operators
10099 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
10100 @code{<=}, and @code{>=} on sets as an error.
10104 @node Built-In Func/Proc
10105 @subsubsection Built-in Functions and Procedures
10106 @cindex Modula-2 built-ins
10108 Modula-2 also makes available several built-in procedures and functions.
10109 In describing these, the following metavariables are used:
10114 represents an @code{ARRAY} variable.
10117 represents a @code{CHAR} constant or variable.
10120 represents a variable or constant of integral type.
10123 represents an identifier that belongs to a set. Generally used in the
10124 same function with the metavariable @var{s}. The type of @var{s} should
10125 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
10128 represents a variable or constant of integral or floating-point type.
10131 represents a variable or constant of floating-point type.
10137 represents a variable.
10140 represents a variable or constant of one of many types. See the
10141 explanation of the function for details.
10144 All Modula-2 built-in procedures also return a result, described below.
10148 Returns the absolute value of @var{n}.
10151 If @var{c} is a lower case letter, it returns its upper case
10152 equivalent, otherwise it returns its argument.
10155 Returns the character whose ordinal value is @var{i}.
10158 Decrements the value in the variable @var{v} by one. Returns the new value.
10160 @item DEC(@var{v},@var{i})
10161 Decrements the value in the variable @var{v} by @var{i}. Returns the
10164 @item EXCL(@var{m},@var{s})
10165 Removes the element @var{m} from the set @var{s}. Returns the new
10168 @item FLOAT(@var{i})
10169 Returns the floating point equivalent of the integer @var{i}.
10171 @item HIGH(@var{a})
10172 Returns the index of the last member of @var{a}.
10175 Increments the value in the variable @var{v} by one. Returns the new value.
10177 @item INC(@var{v},@var{i})
10178 Increments the value in the variable @var{v} by @var{i}. Returns the
10181 @item INCL(@var{m},@var{s})
10182 Adds the element @var{m} to the set @var{s} if it is not already
10183 there. Returns the new set.
10186 Returns the maximum value of the type @var{t}.
10189 Returns the minimum value of the type @var{t}.
10192 Returns boolean TRUE if @var{i} is an odd number.
10195 Returns the ordinal value of its argument. For example, the ordinal
10196 value of a character is its @sc{ascii} value (on machines supporting the
10197 @sc{ascii} character set). @var{x} must be of an ordered type, which include
10198 integral, character and enumerated types.
10200 @item SIZE(@var{x})
10201 Returns the size of its argument. @var{x} can be a variable or a type.
10203 @item TRUNC(@var{r})
10204 Returns the integral part of @var{r}.
10206 @item TSIZE(@var{x})
10207 Returns the size of its argument. @var{x} can be a variable or a type.
10209 @item VAL(@var{t},@var{i})
10210 Returns the member of the type @var{t} whose ordinal value is @var{i}.
10214 @emph{Warning:} Sets and their operations are not yet supported, so
10215 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
10219 @cindex Modula-2 constants
10221 @subsubsection Constants
10223 @value{GDBN} allows you to express the constants of Modula-2 in the following
10229 Integer constants are simply a sequence of digits. When used in an
10230 expression, a constant is interpreted to be type-compatible with the
10231 rest of the expression. Hexadecimal integers are specified by a
10232 trailing @samp{H}, and octal integers by a trailing @samp{B}.
10235 Floating point constants appear as a sequence of digits, followed by a
10236 decimal point and another sequence of digits. An optional exponent can
10237 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
10238 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
10239 digits of the floating point constant must be valid decimal (base 10)
10243 Character constants consist of a single character enclosed by a pair of
10244 like quotes, either single (@code{'}) or double (@code{"}). They may
10245 also be expressed by their ordinal value (their @sc{ascii} value, usually)
10246 followed by a @samp{C}.
10249 String constants consist of a sequence of characters enclosed by a
10250 pair of like quotes, either single (@code{'}) or double (@code{"}).
10251 Escape sequences in the style of C are also allowed. @xref{C
10252 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
10256 Enumerated constants consist of an enumerated identifier.
10259 Boolean constants consist of the identifiers @code{TRUE} and
10263 Pointer constants consist of integral values only.
10266 Set constants are not yet supported.
10270 @subsubsection Modula-2 Types
10271 @cindex Modula-2 types
10273 Currently @value{GDBN} can print the following data types in Modula-2
10274 syntax: array types, record types, set types, pointer types, procedure
10275 types, enumerated types, subrange types and base types. You can also
10276 print the contents of variables declared using these type.
10277 This section gives a number of simple source code examples together with
10278 sample @value{GDBN} sessions.
10280 The first example contains the following section of code:
10289 and you can request @value{GDBN} to interrogate the type and value of
10290 @code{r} and @code{s}.
10293 (@value{GDBP}) print s
10295 (@value{GDBP}) ptype s
10297 (@value{GDBP}) print r
10299 (@value{GDBP}) ptype r
10304 Likewise if your source code declares @code{s} as:
10308 s: SET ['A'..'Z'] ;
10312 then you may query the type of @code{s} by:
10315 (@value{GDBP}) ptype s
10316 type = SET ['A'..'Z']
10320 Note that at present you cannot interactively manipulate set
10321 expressions using the debugger.
10323 The following example shows how you might declare an array in Modula-2
10324 and how you can interact with @value{GDBN} to print its type and contents:
10328 s: ARRAY [-10..10] OF CHAR ;
10332 (@value{GDBP}) ptype s
10333 ARRAY [-10..10] OF CHAR
10336 Note that the array handling is not yet complete and although the type
10337 is printed correctly, expression handling still assumes that all
10338 arrays have a lower bound of zero and not @code{-10} as in the example
10341 Here are some more type related Modula-2 examples:
10345 colour = (blue, red, yellow, green) ;
10346 t = [blue..yellow] ;
10354 The @value{GDBN} interaction shows how you can query the data type
10355 and value of a variable.
10358 (@value{GDBP}) print s
10360 (@value{GDBP}) ptype t
10361 type = [blue..yellow]
10365 In this example a Modula-2 array is declared and its contents
10366 displayed. Observe that the contents are written in the same way as
10367 their @code{C} counterparts.
10371 s: ARRAY [1..5] OF CARDINAL ;
10377 (@value{GDBP}) print s
10378 $1 = @{1, 0, 0, 0, 0@}
10379 (@value{GDBP}) ptype s
10380 type = ARRAY [1..5] OF CARDINAL
10383 The Modula-2 language interface to @value{GDBN} also understands
10384 pointer types as shown in this example:
10388 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
10395 and you can request that @value{GDBN} describes the type of @code{s}.
10398 (@value{GDBP}) ptype s
10399 type = POINTER TO ARRAY [1..5] OF CARDINAL
10402 @value{GDBN} handles compound types as we can see in this example.
10403 Here we combine array types, record types, pointer types and subrange
10414 myarray = ARRAY myrange OF CARDINAL ;
10415 myrange = [-2..2] ;
10417 s: POINTER TO ARRAY myrange OF foo ;
10421 and you can ask @value{GDBN} to describe the type of @code{s} as shown
10425 (@value{GDBP}) ptype s
10426 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
10429 f3 : ARRAY [-2..2] OF CARDINAL;
10434 @subsubsection Modula-2 Defaults
10435 @cindex Modula-2 defaults
10437 If type and range checking are set automatically by @value{GDBN}, they
10438 both default to @code{on} whenever the working language changes to
10439 Modula-2. This happens regardless of whether you or @value{GDBN}
10440 selected the working language.
10442 If you allow @value{GDBN} to set the language automatically, then entering
10443 code compiled from a file whose name ends with @file{.mod} sets the
10444 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
10445 Infer the Source Language}, for further details.
10448 @subsubsection Deviations from Standard Modula-2
10449 @cindex Modula-2, deviations from
10451 A few changes have been made to make Modula-2 programs easier to debug.
10452 This is done primarily via loosening its type strictness:
10456 Unlike in standard Modula-2, pointer constants can be formed by
10457 integers. This allows you to modify pointer variables during
10458 debugging. (In standard Modula-2, the actual address contained in a
10459 pointer variable is hidden from you; it can only be modified
10460 through direct assignment to another pointer variable or expression that
10461 returned a pointer.)
10464 C escape sequences can be used in strings and characters to represent
10465 non-printable characters. @value{GDBN} prints out strings with these
10466 escape sequences embedded. Single non-printable characters are
10467 printed using the @samp{CHR(@var{nnn})} format.
10470 The assignment operator (@code{:=}) returns the value of its right-hand
10474 All built-in procedures both modify @emph{and} return their argument.
10478 @subsubsection Modula-2 Type and Range Checks
10479 @cindex Modula-2 checks
10482 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
10485 @c FIXME remove warning when type/range checks added
10487 @value{GDBN} considers two Modula-2 variables type equivalent if:
10491 They are of types that have been declared equivalent via a @code{TYPE
10492 @var{t1} = @var{t2}} statement
10495 They have been declared on the same line. (Note: This is true of the
10496 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
10499 As long as type checking is enabled, any attempt to combine variables
10500 whose types are not equivalent is an error.
10502 Range checking is done on all mathematical operations, assignment, array
10503 index bounds, and all built-in functions and procedures.
10506 @subsubsection The Scope Operators @code{::} and @code{.}
10508 @cindex @code{.}, Modula-2 scope operator
10509 @cindex colon, doubled as scope operator
10511 @vindex colon-colon@r{, in Modula-2}
10512 @c Info cannot handle :: but TeX can.
10515 @vindex ::@r{, in Modula-2}
10518 There are a few subtle differences between the Modula-2 scope operator
10519 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
10524 @var{module} . @var{id}
10525 @var{scope} :: @var{id}
10529 where @var{scope} is the name of a module or a procedure,
10530 @var{module} the name of a module, and @var{id} is any declared
10531 identifier within your program, except another module.
10533 Using the @code{::} operator makes @value{GDBN} search the scope
10534 specified by @var{scope} for the identifier @var{id}. If it is not
10535 found in the specified scope, then @value{GDBN} searches all scopes
10536 enclosing the one specified by @var{scope}.
10538 Using the @code{.} operator makes @value{GDBN} search the current scope for
10539 the identifier specified by @var{id} that was imported from the
10540 definition module specified by @var{module}. With this operator, it is
10541 an error if the identifier @var{id} was not imported from definition
10542 module @var{module}, or if @var{id} is not an identifier in
10546 @subsubsection @value{GDBN} and Modula-2
10548 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
10549 Five subcommands of @code{set print} and @code{show print} apply
10550 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
10551 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
10552 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
10553 analogue in Modula-2.
10555 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
10556 with any language, is not useful with Modula-2. Its
10557 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
10558 created in Modula-2 as they can in C or C@t{++}. However, because an
10559 address can be specified by an integral constant, the construct
10560 @samp{@{@var{type}@}@var{adrexp}} is still useful.
10562 @cindex @code{#} in Modula-2
10563 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
10564 interpreted as the beginning of a comment. Use @code{<>} instead.
10570 The extensions made to @value{GDBN} for Ada only support
10571 output from the @sc{gnu} Ada (GNAT) compiler.
10572 Other Ada compilers are not currently supported, and
10573 attempting to debug executables produced by them is most likely
10577 @cindex expressions in Ada
10579 * Ada Mode Intro:: General remarks on the Ada syntax
10580 and semantics supported by Ada mode
10582 * Omissions from Ada:: Restrictions on the Ada expression syntax.
10583 * Additions to Ada:: Extensions of the Ada expression syntax.
10584 * Stopping Before Main Program:: Debugging the program during elaboration.
10585 * Ada Glitches:: Known peculiarities of Ada mode.
10588 @node Ada Mode Intro
10589 @subsubsection Introduction
10590 @cindex Ada mode, general
10592 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
10593 syntax, with some extensions.
10594 The philosophy behind the design of this subset is
10598 That @value{GDBN} should provide basic literals and access to operations for
10599 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
10600 leaving more sophisticated computations to subprograms written into the
10601 program (which therefore may be called from @value{GDBN}).
10604 That type safety and strict adherence to Ada language restrictions
10605 are not particularly important to the @value{GDBN} user.
10608 That brevity is important to the @value{GDBN} user.
10611 Thus, for brevity, the debugger acts as if there were
10612 implicit @code{with} and @code{use} clauses in effect for all user-written
10613 packages, making it unnecessary to fully qualify most names with
10614 their packages, regardless of context. Where this causes ambiguity,
10615 @value{GDBN} asks the user's intent.
10617 The debugger will start in Ada mode if it detects an Ada main program.
10618 As for other languages, it will enter Ada mode when stopped in a program that
10619 was translated from an Ada source file.
10621 While in Ada mode, you may use `@t{--}' for comments. This is useful
10622 mostly for documenting command files. The standard @value{GDBN} comment
10623 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
10624 middle (to allow based literals).
10626 The debugger supports limited overloading. Given a subprogram call in which
10627 the function symbol has multiple definitions, it will use the number of
10628 actual parameters and some information about their types to attempt to narrow
10629 the set of definitions. It also makes very limited use of context, preferring
10630 procedures to functions in the context of the @code{call} command, and
10631 functions to procedures elsewhere.
10633 @node Omissions from Ada
10634 @subsubsection Omissions from Ada
10635 @cindex Ada, omissions from
10637 Here are the notable omissions from the subset:
10641 Only a subset of the attributes are supported:
10645 @t{'First}, @t{'Last}, and @t{'Length}
10646 on array objects (not on types and subtypes).
10649 @t{'Min} and @t{'Max}.
10652 @t{'Pos} and @t{'Val}.
10658 @t{'Range} on array objects (not subtypes), but only as the right
10659 operand of the membership (@code{in}) operator.
10662 @t{'Access}, @t{'Unchecked_Access}, and
10663 @t{'Unrestricted_Access} (a GNAT extension).
10671 @code{Characters.Latin_1} are not available and
10672 concatenation is not implemented. Thus, escape characters in strings are
10673 not currently available.
10676 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
10677 equality of representations. They will generally work correctly
10678 for strings and arrays whose elements have integer or enumeration types.
10679 They may not work correctly for arrays whose element
10680 types have user-defined equality, for arrays of real values
10681 (in particular, IEEE-conformant floating point, because of negative
10682 zeroes and NaNs), and for arrays whose elements contain unused bits with
10683 indeterminate values.
10686 The other component-by-component array operations (@code{and}, @code{or},
10687 @code{xor}, @code{not}, and relational tests other than equality)
10688 are not implemented.
10691 @cindex array aggregates (Ada)
10692 @cindex record aggregates (Ada)
10693 @cindex aggregates (Ada)
10694 There is limited support for array and record aggregates. They are
10695 permitted only on the right sides of assignments, as in these examples:
10698 set An_Array := (1, 2, 3, 4, 5, 6)
10699 set An_Array := (1, others => 0)
10700 set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
10701 set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
10702 set A_Record := (1, "Peter", True);
10703 set A_Record := (Name => "Peter", Id => 1, Alive => True)
10707 discriminant's value by assigning an aggregate has an
10708 undefined effect if that discriminant is used within the record.
10709 However, you can first modify discriminants by directly assigning to
10710 them (which normally would not be allowed in Ada), and then performing an
10711 aggregate assignment. For example, given a variable @code{A_Rec}
10712 declared to have a type such as:
10715 type Rec (Len : Small_Integer := 0) is record
10717 Vals : IntArray (1 .. Len);
10721 you can assign a value with a different size of @code{Vals} with two
10726 set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
10729 As this example also illustrates, @value{GDBN} is very loose about the usual
10730 rules concerning aggregates. You may leave out some of the
10731 components of an array or record aggregate (such as the @code{Len}
10732 component in the assignment to @code{A_Rec} above); they will retain their
10733 original values upon assignment. You may freely use dynamic values as
10734 indices in component associations. You may even use overlapping or
10735 redundant component associations, although which component values are
10736 assigned in such cases is not defined.
10739 Calls to dispatching subprograms are not implemented.
10742 The overloading algorithm is much more limited (i.e., less selective)
10743 than that of real Ada. It makes only limited use of the context in
10744 which a subexpression appears to resolve its meaning, and it is much
10745 looser in its rules for allowing type matches. As a result, some
10746 function calls will be ambiguous, and the user will be asked to choose
10747 the proper resolution.
10750 The @code{new} operator is not implemented.
10753 Entry calls are not implemented.
10756 Aside from printing, arithmetic operations on the native VAX floating-point
10757 formats are not supported.
10760 It is not possible to slice a packed array.
10763 @node Additions to Ada
10764 @subsubsection Additions to Ada
10765 @cindex Ada, deviations from
10767 As it does for other languages, @value{GDBN} makes certain generic
10768 extensions to Ada (@pxref{Expressions}):
10772 If the expression @var{E} is a variable residing in memory (typically
10773 a local variable or array element) and @var{N} is a positive integer,
10774 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
10775 @var{N}-1 adjacent variables following it in memory as an array. In
10776 Ada, this operator is generally not necessary, since its prime use is
10777 in displaying parts of an array, and slicing will usually do this in
10778 Ada. However, there are occasional uses when debugging programs in
10779 which certain debugging information has been optimized away.
10782 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
10783 appears in function or file @var{B}.'' When @var{B} is a file name,
10784 you must typically surround it in single quotes.
10787 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
10788 @var{type} that appears at address @var{addr}.''
10791 A name starting with @samp{$} is a convenience variable
10792 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
10795 In addition, @value{GDBN} provides a few other shortcuts and outright
10796 additions specific to Ada:
10800 The assignment statement is allowed as an expression, returning
10801 its right-hand operand as its value. Thus, you may enter
10805 print A(tmp := y + 1)
10809 The semicolon is allowed as an ``operator,'' returning as its value
10810 the value of its right-hand operand.
10811 This allows, for example,
10812 complex conditional breaks:
10816 condition 1 (report(i); k += 1; A(k) > 100)
10820 Rather than use catenation and symbolic character names to introduce special
10821 characters into strings, one may instead use a special bracket notation,
10822 which is also used to print strings. A sequence of characters of the form
10823 @samp{["@var{XX}"]} within a string or character literal denotes the
10824 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
10825 sequence of characters @samp{["""]} also denotes a single quotation mark
10826 in strings. For example,
10828 "One line.["0a"]Next line.["0a"]"
10831 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
10835 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
10836 @t{'Max} is optional (and is ignored in any case). For example, it is valid
10844 When printing arrays, @value{GDBN} uses positional notation when the
10845 array has a lower bound of 1, and uses a modified named notation otherwise.
10846 For example, a one-dimensional array of three integers with a lower bound
10847 of 3 might print as
10854 That is, in contrast to valid Ada, only the first component has a @code{=>}
10858 You may abbreviate attributes in expressions with any unique,
10859 multi-character subsequence of
10860 their names (an exact match gets preference).
10861 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
10862 in place of @t{a'length}.
10865 @cindex quoting Ada internal identifiers
10866 Since Ada is case-insensitive, the debugger normally maps identifiers you type
10867 to lower case. The GNAT compiler uses upper-case characters for
10868 some of its internal identifiers, which are normally of no interest to users.
10869 For the rare occasions when you actually have to look at them,
10870 enclose them in angle brackets to avoid the lower-case mapping.
10873 @value{GDBP} print <JMPBUF_SAVE>[0]
10877 Printing an object of class-wide type or dereferencing an
10878 access-to-class-wide value will display all the components of the object's
10879 specific type (as indicated by its run-time tag). Likewise, component
10880 selection on such a value will operate on the specific type of the
10885 @node Stopping Before Main Program
10886 @subsubsection Stopping at the Very Beginning
10888 @cindex breakpointing Ada elaboration code
10889 It is sometimes necessary to debug the program during elaboration, and
10890 before reaching the main procedure.
10891 As defined in the Ada Reference
10892 Manual, the elaboration code is invoked from a procedure called
10893 @code{adainit}. To run your program up to the beginning of
10894 elaboration, simply use the following two commands:
10895 @code{tbreak adainit} and @code{run}.
10898 @subsubsection Known Peculiarities of Ada Mode
10899 @cindex Ada, problems
10901 Besides the omissions listed previously (@pxref{Omissions from Ada}),
10902 we know of several problems with and limitations of Ada mode in
10904 some of which will be fixed with planned future releases of the debugger
10905 and the GNU Ada compiler.
10909 Currently, the debugger
10910 has insufficient information to determine whether certain pointers represent
10911 pointers to objects or the objects themselves.
10912 Thus, the user may have to tack an extra @code{.all} after an expression
10913 to get it printed properly.
10916 Static constants that the compiler chooses not to materialize as objects in
10917 storage are invisible to the debugger.
10920 Named parameter associations in function argument lists are ignored (the
10921 argument lists are treated as positional).
10924 Many useful library packages are currently invisible to the debugger.
10927 Fixed-point arithmetic, conversions, input, and output is carried out using
10928 floating-point arithmetic, and may give results that only approximate those on
10932 The type of the @t{'Address} attribute may not be @code{System.Address}.
10935 The GNAT compiler never generates the prefix @code{Standard} for any of
10936 the standard symbols defined by the Ada language. @value{GDBN} knows about
10937 this: it will strip the prefix from names when you use it, and will never
10938 look for a name you have so qualified among local symbols, nor match against
10939 symbols in other packages or subprograms. If you have
10940 defined entities anywhere in your program other than parameters and
10941 local variables whose simple names match names in @code{Standard},
10942 GNAT's lack of qualification here can cause confusion. When this happens,
10943 you can usually resolve the confusion
10944 by qualifying the problematic names with package
10945 @code{Standard} explicitly.
10948 @node Unsupported Languages
10949 @section Unsupported Languages
10951 @cindex unsupported languages
10952 @cindex minimal language
10953 In addition to the other fully-supported programming languages,
10954 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
10955 It does not represent a real programming language, but provides a set
10956 of capabilities close to what the C or assembly languages provide.
10957 This should allow most simple operations to be performed while debugging
10958 an application that uses a language currently not supported by @value{GDBN}.
10960 If the language is set to @code{auto}, @value{GDBN} will automatically
10961 select this language if the current frame corresponds to an unsupported
10965 @chapter Examining the Symbol Table
10967 The commands described in this chapter allow you to inquire about the
10968 symbols (names of variables, functions and types) defined in your
10969 program. This information is inherent in the text of your program and
10970 does not change as your program executes. @value{GDBN} finds it in your
10971 program's symbol table, in the file indicated when you started @value{GDBN}
10972 (@pxref{File Options, ,Choosing Files}), or by one of the
10973 file-management commands (@pxref{Files, ,Commands to Specify Files}).
10975 @cindex symbol names
10976 @cindex names of symbols
10977 @cindex quoting names
10978 Occasionally, you may need to refer to symbols that contain unusual
10979 characters, which @value{GDBN} ordinarily treats as word delimiters. The
10980 most frequent case is in referring to static variables in other
10981 source files (@pxref{Variables,,Program Variables}). File names
10982 are recorded in object files as debugging symbols, but @value{GDBN} would
10983 ordinarily parse a typical file name, like @file{foo.c}, as the three words
10984 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
10985 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
10992 looks up the value of @code{x} in the scope of the file @file{foo.c}.
10995 @cindex case-insensitive symbol names
10996 @cindex case sensitivity in symbol names
10997 @kindex set case-sensitive
10998 @item set case-sensitive on
10999 @itemx set case-sensitive off
11000 @itemx set case-sensitive auto
11001 Normally, when @value{GDBN} looks up symbols, it matches their names
11002 with case sensitivity determined by the current source language.
11003 Occasionally, you may wish to control that. The command @code{set
11004 case-sensitive} lets you do that by specifying @code{on} for
11005 case-sensitive matches or @code{off} for case-insensitive ones. If
11006 you specify @code{auto}, case sensitivity is reset to the default
11007 suitable for the source language. The default is case-sensitive
11008 matches for all languages except for Fortran, for which the default is
11009 case-insensitive matches.
11011 @kindex show case-sensitive
11012 @item show case-sensitive
11013 This command shows the current setting of case sensitivity for symbols
11016 @kindex info address
11017 @cindex address of a symbol
11018 @item info address @var{symbol}
11019 Describe where the data for @var{symbol} is stored. For a register
11020 variable, this says which register it is kept in. For a non-register
11021 local variable, this prints the stack-frame offset at which the variable
11024 Note the contrast with @samp{print &@var{symbol}}, which does not work
11025 at all for a register variable, and for a stack local variable prints
11026 the exact address of the current instantiation of the variable.
11028 @kindex info symbol
11029 @cindex symbol from address
11030 @cindex closest symbol and offset for an address
11031 @item info symbol @var{addr}
11032 Print the name of a symbol which is stored at the address @var{addr}.
11033 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
11034 nearest symbol and an offset from it:
11037 (@value{GDBP}) info symbol 0x54320
11038 _initialize_vx + 396 in section .text
11042 This is the opposite of the @code{info address} command. You can use
11043 it to find out the name of a variable or a function given its address.
11046 @item whatis [@var{arg}]
11047 Print the data type of @var{arg}, which can be either an expression or
11048 a data type. With no argument, print the data type of @code{$}, the
11049 last value in the value history. If @var{arg} is an expression, it is
11050 not actually evaluated, and any side-effecting operations (such as
11051 assignments or function calls) inside it do not take place. If
11052 @var{arg} is a type name, it may be the name of a type or typedef, or
11053 for C code it may have the form @samp{class @var{class-name}},
11054 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
11055 @samp{enum @var{enum-tag}}.
11056 @xref{Expressions, ,Expressions}.
11059 @item ptype [@var{arg}]
11060 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
11061 detailed description of the type, instead of just the name of the type.
11062 @xref{Expressions, ,Expressions}.
11064 For example, for this variable declaration:
11067 struct complex @{double real; double imag;@} v;
11071 the two commands give this output:
11075 (@value{GDBP}) whatis v
11076 type = struct complex
11077 (@value{GDBP}) ptype v
11078 type = struct complex @{
11086 As with @code{whatis}, using @code{ptype} without an argument refers to
11087 the type of @code{$}, the last value in the value history.
11089 @cindex incomplete type
11090 Sometimes, programs use opaque data types or incomplete specifications
11091 of complex data structure. If the debug information included in the
11092 program does not allow @value{GDBN} to display a full declaration of
11093 the data type, it will say @samp{<incomplete type>}. For example,
11094 given these declarations:
11098 struct foo *fooptr;
11102 but no definition for @code{struct foo} itself, @value{GDBN} will say:
11105 (@value{GDBP}) ptype foo
11106 $1 = <incomplete type>
11110 ``Incomplete type'' is C terminology for data types that are not
11111 completely specified.
11114 @item info types @var{regexp}
11116 Print a brief description of all types whose names match the regular
11117 expression @var{regexp} (or all types in your program, if you supply
11118 no argument). Each complete typename is matched as though it were a
11119 complete line; thus, @samp{i type value} gives information on all
11120 types in your program whose names include the string @code{value}, but
11121 @samp{i type ^value$} gives information only on types whose complete
11122 name is @code{value}.
11124 This command differs from @code{ptype} in two ways: first, like
11125 @code{whatis}, it does not print a detailed description; second, it
11126 lists all source files where a type is defined.
11129 @cindex local variables
11130 @item info scope @var{location}
11131 List all the variables local to a particular scope. This command
11132 accepts a @var{location} argument---a function name, a source line, or
11133 an address preceded by a @samp{*}, and prints all the variables local
11134 to the scope defined by that location. (@xref{Specify Location}, for
11135 details about supported forms of @var{location}.) For example:
11138 (@value{GDBP}) @b{info scope command_line_handler}
11139 Scope for command_line_handler:
11140 Symbol rl is an argument at stack/frame offset 8, length 4.
11141 Symbol linebuffer is in static storage at address 0x150a18, length 4.
11142 Symbol linelength is in static storage at address 0x150a1c, length 4.
11143 Symbol p is a local variable in register $esi, length 4.
11144 Symbol p1 is a local variable in register $ebx, length 4.
11145 Symbol nline is a local variable in register $edx, length 4.
11146 Symbol repeat is a local variable at frame offset -8, length 4.
11150 This command is especially useful for determining what data to collect
11151 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
11154 @kindex info source
11156 Show information about the current source file---that is, the source file for
11157 the function containing the current point of execution:
11160 the name of the source file, and the directory containing it,
11162 the directory it was compiled in,
11164 its length, in lines,
11166 which programming language it is written in,
11168 whether the executable includes debugging information for that file, and
11169 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
11171 whether the debugging information includes information about
11172 preprocessor macros.
11176 @kindex info sources
11178 Print the names of all source files in your program for which there is
11179 debugging information, organized into two lists: files whose symbols
11180 have already been read, and files whose symbols will be read when needed.
11182 @kindex info functions
11183 @item info functions
11184 Print the names and data types of all defined functions.
11186 @item info functions @var{regexp}
11187 Print the names and data types of all defined functions
11188 whose names contain a match for regular expression @var{regexp}.
11189 Thus, @samp{info fun step} finds all functions whose names
11190 include @code{step}; @samp{info fun ^step} finds those whose names
11191 start with @code{step}. If a function name contains characters
11192 that conflict with the regular expression language (e.g.@:
11193 @samp{operator*()}), they may be quoted with a backslash.
11195 @kindex info variables
11196 @item info variables
11197 Print the names and data types of all variables that are declared
11198 outside of functions (i.e.@: excluding local variables).
11200 @item info variables @var{regexp}
11201 Print the names and data types of all variables (except for local
11202 variables) whose names contain a match for regular expression
11205 @kindex info classes
11206 @cindex Objective-C, classes and selectors
11208 @itemx info classes @var{regexp}
11209 Display all Objective-C classes in your program, or
11210 (with the @var{regexp} argument) all those matching a particular regular
11213 @kindex info selectors
11214 @item info selectors
11215 @itemx info selectors @var{regexp}
11216 Display all Objective-C selectors in your program, or
11217 (with the @var{regexp} argument) all those matching a particular regular
11221 This was never implemented.
11222 @kindex info methods
11224 @itemx info methods @var{regexp}
11225 The @code{info methods} command permits the user to examine all defined
11226 methods within C@t{++} program, or (with the @var{regexp} argument) a
11227 specific set of methods found in the various C@t{++} classes. Many
11228 C@t{++} classes provide a large number of methods. Thus, the output
11229 from the @code{ptype} command can be overwhelming and hard to use. The
11230 @code{info-methods} command filters the methods, printing only those
11231 which match the regular-expression @var{regexp}.
11234 @cindex reloading symbols
11235 Some systems allow individual object files that make up your program to
11236 be replaced without stopping and restarting your program. For example,
11237 in VxWorks you can simply recompile a defective object file and keep on
11238 running. If you are running on one of these systems, you can allow
11239 @value{GDBN} to reload the symbols for automatically relinked modules:
11242 @kindex set symbol-reloading
11243 @item set symbol-reloading on
11244 Replace symbol definitions for the corresponding source file when an
11245 object file with a particular name is seen again.
11247 @item set symbol-reloading off
11248 Do not replace symbol definitions when encountering object files of the
11249 same name more than once. This is the default state; if you are not
11250 running on a system that permits automatic relinking of modules, you
11251 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
11252 may discard symbols when linking large programs, that may contain
11253 several modules (from different directories or libraries) with the same
11256 @kindex show symbol-reloading
11257 @item show symbol-reloading
11258 Show the current @code{on} or @code{off} setting.
11261 @cindex opaque data types
11262 @kindex set opaque-type-resolution
11263 @item set opaque-type-resolution on
11264 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
11265 declared as a pointer to a @code{struct}, @code{class}, or
11266 @code{union}---for example, @code{struct MyType *}---that is used in one
11267 source file although the full declaration of @code{struct MyType} is in
11268 another source file. The default is on.
11270 A change in the setting of this subcommand will not take effect until
11271 the next time symbols for a file are loaded.
11273 @item set opaque-type-resolution off
11274 Tell @value{GDBN} not to resolve opaque types. In this case, the type
11275 is printed as follows:
11277 @{<no data fields>@}
11280 @kindex show opaque-type-resolution
11281 @item show opaque-type-resolution
11282 Show whether opaque types are resolved or not.
11284 @kindex maint print symbols
11285 @cindex symbol dump
11286 @kindex maint print psymbols
11287 @cindex partial symbol dump
11288 @item maint print symbols @var{filename}
11289 @itemx maint print psymbols @var{filename}
11290 @itemx maint print msymbols @var{filename}
11291 Write a dump of debugging symbol data into the file @var{filename}.
11292 These commands are used to debug the @value{GDBN} symbol-reading code. Only
11293 symbols with debugging data are included. If you use @samp{maint print
11294 symbols}, @value{GDBN} includes all the symbols for which it has already
11295 collected full details: that is, @var{filename} reflects symbols for
11296 only those files whose symbols @value{GDBN} has read. You can use the
11297 command @code{info sources} to find out which files these are. If you
11298 use @samp{maint print psymbols} instead, the dump shows information about
11299 symbols that @value{GDBN} only knows partially---that is, symbols defined in
11300 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
11301 @samp{maint print msymbols} dumps just the minimal symbol information
11302 required for each object file from which @value{GDBN} has read some symbols.
11303 @xref{Files, ,Commands to Specify Files}, for a discussion of how
11304 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
11306 @kindex maint info symtabs
11307 @kindex maint info psymtabs
11308 @cindex listing @value{GDBN}'s internal symbol tables
11309 @cindex symbol tables, listing @value{GDBN}'s internal
11310 @cindex full symbol tables, listing @value{GDBN}'s internal
11311 @cindex partial symbol tables, listing @value{GDBN}'s internal
11312 @item maint info symtabs @r{[} @var{regexp} @r{]}
11313 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
11315 List the @code{struct symtab} or @code{struct partial_symtab}
11316 structures whose names match @var{regexp}. If @var{regexp} is not
11317 given, list them all. The output includes expressions which you can
11318 copy into a @value{GDBN} debugging this one to examine a particular
11319 structure in more detail. For example:
11322 (@value{GDBP}) maint info psymtabs dwarf2read
11323 @{ objfile /home/gnu/build/gdb/gdb
11324 ((struct objfile *) 0x82e69d0)
11325 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
11326 ((struct partial_symtab *) 0x8474b10)
11329 text addresses 0x814d3c8 -- 0x8158074
11330 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
11331 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
11332 dependencies (none)
11335 (@value{GDBP}) maint info symtabs
11339 We see that there is one partial symbol table whose filename contains
11340 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
11341 and we see that @value{GDBN} has not read in any symtabs yet at all.
11342 If we set a breakpoint on a function, that will cause @value{GDBN} to
11343 read the symtab for the compilation unit containing that function:
11346 (@value{GDBP}) break dwarf2_psymtab_to_symtab
11347 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
11349 (@value{GDBP}) maint info symtabs
11350 @{ objfile /home/gnu/build/gdb/gdb
11351 ((struct objfile *) 0x82e69d0)
11352 @{ symtab /home/gnu/src/gdb/dwarf2read.c
11353 ((struct symtab *) 0x86c1f38)
11356 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
11357 linetable ((struct linetable *) 0x8370fa0)
11358 debugformat DWARF 2
11367 @chapter Altering Execution
11369 Once you think you have found an error in your program, you might want to
11370 find out for certain whether correcting the apparent error would lead to
11371 correct results in the rest of the run. You can find the answer by
11372 experiment, using the @value{GDBN} features for altering execution of the
11375 For example, you can store new values into variables or memory
11376 locations, give your program a signal, restart it at a different
11377 address, or even return prematurely from a function.
11380 * Assignment:: Assignment to variables
11381 * Jumping:: Continuing at a different address
11382 * Signaling:: Giving your program a signal
11383 * Returning:: Returning from a function
11384 * Calling:: Calling your program's functions
11385 * Patching:: Patching your program
11389 @section Assignment to Variables
11392 @cindex setting variables
11393 To alter the value of a variable, evaluate an assignment expression.
11394 @xref{Expressions, ,Expressions}. For example,
11401 stores the value 4 into the variable @code{x}, and then prints the
11402 value of the assignment expression (which is 4).
11403 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
11404 information on operators in supported languages.
11406 @kindex set variable
11407 @cindex variables, setting
11408 If you are not interested in seeing the value of the assignment, use the
11409 @code{set} command instead of the @code{print} command. @code{set} is
11410 really the same as @code{print} except that the expression's value is
11411 not printed and is not put in the value history (@pxref{Value History,
11412 ,Value History}). The expression is evaluated only for its effects.
11414 If the beginning of the argument string of the @code{set} command
11415 appears identical to a @code{set} subcommand, use the @code{set
11416 variable} command instead of just @code{set}. This command is identical
11417 to @code{set} except for its lack of subcommands. For example, if your
11418 program has a variable @code{width}, you get an error if you try to set
11419 a new value with just @samp{set width=13}, because @value{GDBN} has the
11420 command @code{set width}:
11423 (@value{GDBP}) whatis width
11425 (@value{GDBP}) p width
11427 (@value{GDBP}) set width=47
11428 Invalid syntax in expression.
11432 The invalid expression, of course, is @samp{=47}. In
11433 order to actually set the program's variable @code{width}, use
11436 (@value{GDBP}) set var width=47
11439 Because the @code{set} command has many subcommands that can conflict
11440 with the names of program variables, it is a good idea to use the
11441 @code{set variable} command instead of just @code{set}. For example, if
11442 your program has a variable @code{g}, you run into problems if you try
11443 to set a new value with just @samp{set g=4}, because @value{GDBN} has
11444 the command @code{set gnutarget}, abbreviated @code{set g}:
11448 (@value{GDBP}) whatis g
11452 (@value{GDBP}) set g=4
11456 The program being debugged has been started already.
11457 Start it from the beginning? (y or n) y
11458 Starting program: /home/smith/cc_progs/a.out
11459 "/home/smith/cc_progs/a.out": can't open to read symbols:
11460 Invalid bfd target.
11461 (@value{GDBP}) show g
11462 The current BFD target is "=4".
11467 The program variable @code{g} did not change, and you silently set the
11468 @code{gnutarget} to an invalid value. In order to set the variable
11472 (@value{GDBP}) set var g=4
11475 @value{GDBN} allows more implicit conversions in assignments than C; you can
11476 freely store an integer value into a pointer variable or vice versa,
11477 and you can convert any structure to any other structure that is the
11478 same length or shorter.
11479 @comment FIXME: how do structs align/pad in these conversions?
11480 @comment /doc@cygnus.com 18dec1990
11482 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
11483 construct to generate a value of specified type at a specified address
11484 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
11485 to memory location @code{0x83040} as an integer (which implies a certain size
11486 and representation in memory), and
11489 set @{int@}0x83040 = 4
11493 stores the value 4 into that memory location.
11496 @section Continuing at a Different Address
11498 Ordinarily, when you continue your program, you do so at the place where
11499 it stopped, with the @code{continue} command. You can instead continue at
11500 an address of your own choosing, with the following commands:
11504 @item jump @var{linespec}
11505 @itemx jump @var{location}
11506 Resume execution at line @var{linespec} or at address given by
11507 @var{location}. Execution stops again immediately if there is a
11508 breakpoint there. @xref{Specify Location}, for a description of the
11509 different forms of @var{linespec} and @var{location}. It is common
11510 practice to use the @code{tbreak} command in conjunction with
11511 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
11513 The @code{jump} command does not change the current stack frame, or
11514 the stack pointer, or the contents of any memory location or any
11515 register other than the program counter. If line @var{linespec} is in
11516 a different function from the one currently executing, the results may
11517 be bizarre if the two functions expect different patterns of arguments or
11518 of local variables. For this reason, the @code{jump} command requests
11519 confirmation if the specified line is not in the function currently
11520 executing. However, even bizarre results are predictable if you are
11521 well acquainted with the machine-language code of your program.
11524 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
11525 On many systems, you can get much the same effect as the @code{jump}
11526 command by storing a new value into the register @code{$pc}. The
11527 difference is that this does not start your program running; it only
11528 changes the address of where it @emph{will} run when you continue. For
11536 makes the next @code{continue} command or stepping command execute at
11537 address @code{0x485}, rather than at the address where your program stopped.
11538 @xref{Continuing and Stepping, ,Continuing and Stepping}.
11540 The most common occasion to use the @code{jump} command is to back
11541 up---perhaps with more breakpoints set---over a portion of a program
11542 that has already executed, in order to examine its execution in more
11547 @section Giving your Program a Signal
11548 @cindex deliver a signal to a program
11552 @item signal @var{signal}
11553 Resume execution where your program stopped, but immediately give it the
11554 signal @var{signal}. @var{signal} can be the name or the number of a
11555 signal. For example, on many systems @code{signal 2} and @code{signal
11556 SIGINT} are both ways of sending an interrupt signal.
11558 Alternatively, if @var{signal} is zero, continue execution without
11559 giving a signal. This is useful when your program stopped on account of
11560 a signal and would ordinary see the signal when resumed with the
11561 @code{continue} command; @samp{signal 0} causes it to resume without a
11564 @code{signal} does not repeat when you press @key{RET} a second time
11565 after executing the command.
11569 Invoking the @code{signal} command is not the same as invoking the
11570 @code{kill} utility from the shell. Sending a signal with @code{kill}
11571 causes @value{GDBN} to decide what to do with the signal depending on
11572 the signal handling tables (@pxref{Signals}). The @code{signal} command
11573 passes the signal directly to your program.
11577 @section Returning from a Function
11580 @cindex returning from a function
11583 @itemx return @var{expression}
11584 You can cancel execution of a function call with the @code{return}
11585 command. If you give an
11586 @var{expression} argument, its value is used as the function's return
11590 When you use @code{return}, @value{GDBN} discards the selected stack frame
11591 (and all frames within it). You can think of this as making the
11592 discarded frame return prematurely. If you wish to specify a value to
11593 be returned, give that value as the argument to @code{return}.
11595 This pops the selected stack frame (@pxref{Selection, ,Selecting a
11596 Frame}), and any other frames inside of it, leaving its caller as the
11597 innermost remaining frame. That frame becomes selected. The
11598 specified value is stored in the registers used for returning values
11601 The @code{return} command does not resume execution; it leaves the
11602 program stopped in the state that would exist if the function had just
11603 returned. In contrast, the @code{finish} command (@pxref{Continuing
11604 and Stepping, ,Continuing and Stepping}) resumes execution until the
11605 selected stack frame returns naturally.
11608 @section Calling Program Functions
11611 @cindex calling functions
11612 @cindex inferior functions, calling
11613 @item print @var{expr}
11614 Evaluate the expression @var{expr} and display the resulting value.
11615 @var{expr} may include calls to functions in the program being
11619 @item call @var{expr}
11620 Evaluate the expression @var{expr} without displaying @code{void}
11623 You can use this variant of the @code{print} command if you want to
11624 execute a function from your program that does not return anything
11625 (a.k.a.@: @dfn{a void function}), but without cluttering the output
11626 with @code{void} returned values that @value{GDBN} will otherwise
11627 print. If the result is not void, it is printed and saved in the
11631 It is possible for the function you call via the @code{print} or
11632 @code{call} command to generate a signal (e.g., if there's a bug in
11633 the function, or if you passed it incorrect arguments). What happens
11634 in that case is controlled by the @code{set unwindonsignal} command.
11637 @item set unwindonsignal
11638 @kindex set unwindonsignal
11639 @cindex unwind stack in called functions
11640 @cindex call dummy stack unwinding
11641 Set unwinding of the stack if a signal is received while in a function
11642 that @value{GDBN} called in the program being debugged. If set to on,
11643 @value{GDBN} unwinds the stack it created for the call and restores
11644 the context to what it was before the call. If set to off (the
11645 default), @value{GDBN} stops in the frame where the signal was
11648 @item show unwindonsignal
11649 @kindex show unwindonsignal
11650 Show the current setting of stack unwinding in the functions called by
11654 @cindex weak alias functions
11655 Sometimes, a function you wish to call is actually a @dfn{weak alias}
11656 for another function. In such case, @value{GDBN} might not pick up
11657 the type information, including the types of the function arguments,
11658 which causes @value{GDBN} to call the inferior function incorrectly.
11659 As a result, the called function will function erroneously and may
11660 even crash. A solution to that is to use the name of the aliased
11664 @section Patching Programs
11666 @cindex patching binaries
11667 @cindex writing into executables
11668 @cindex writing into corefiles
11670 By default, @value{GDBN} opens the file containing your program's
11671 executable code (or the corefile) read-only. This prevents accidental
11672 alterations to machine code; but it also prevents you from intentionally
11673 patching your program's binary.
11675 If you'd like to be able to patch the binary, you can specify that
11676 explicitly with the @code{set write} command. For example, you might
11677 want to turn on internal debugging flags, or even to make emergency
11683 @itemx set write off
11684 If you specify @samp{set write on}, @value{GDBN} opens executable and
11685 core files for both reading and writing; if you specify @samp{set write
11686 off} (the default), @value{GDBN} opens them read-only.
11688 If you have already loaded a file, you must load it again (using the
11689 @code{exec-file} or @code{core-file} command) after changing @code{set
11690 write}, for your new setting to take effect.
11694 Display whether executable files and core files are opened for writing
11695 as well as reading.
11699 @chapter @value{GDBN} Files
11701 @value{GDBN} needs to know the file name of the program to be debugged,
11702 both in order to read its symbol table and in order to start your
11703 program. To debug a core dump of a previous run, you must also tell
11704 @value{GDBN} the name of the core dump file.
11707 * Files:: Commands to specify files
11708 * Separate Debug Files:: Debugging information in separate files
11709 * Symbol Errors:: Errors reading symbol files
11713 @section Commands to Specify Files
11715 @cindex symbol table
11716 @cindex core dump file
11718 You may want to specify executable and core dump file names. The usual
11719 way to do this is at start-up time, using the arguments to
11720 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
11721 Out of @value{GDBN}}).
11723 Occasionally it is necessary to change to a different file during a
11724 @value{GDBN} session. Or you may run @value{GDBN} and forget to
11725 specify a file you want to use. Or you are debugging a remote target
11726 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
11727 Program}). In these situations the @value{GDBN} commands to specify
11728 new files are useful.
11731 @cindex executable file
11733 @item file @var{filename}
11734 Use @var{filename} as the program to be debugged. It is read for its
11735 symbols and for the contents of pure memory. It is also the program
11736 executed when you use the @code{run} command. If you do not specify a
11737 directory and the file is not found in the @value{GDBN} working directory,
11738 @value{GDBN} uses the environment variable @code{PATH} as a list of
11739 directories to search, just as the shell does when looking for a program
11740 to run. You can change the value of this variable, for both @value{GDBN}
11741 and your program, using the @code{path} command.
11743 @cindex unlinked object files
11744 @cindex patching object files
11745 You can load unlinked object @file{.o} files into @value{GDBN} using
11746 the @code{file} command. You will not be able to ``run'' an object
11747 file, but you can disassemble functions and inspect variables. Also,
11748 if the underlying BFD functionality supports it, you could use
11749 @kbd{gdb -write} to patch object files using this technique. Note
11750 that @value{GDBN} can neither interpret nor modify relocations in this
11751 case, so branches and some initialized variables will appear to go to
11752 the wrong place. But this feature is still handy from time to time.
11755 @code{file} with no argument makes @value{GDBN} discard any information it
11756 has on both executable file and the symbol table.
11759 @item exec-file @r{[} @var{filename} @r{]}
11760 Specify that the program to be run (but not the symbol table) is found
11761 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
11762 if necessary to locate your program. Omitting @var{filename} means to
11763 discard information on the executable file.
11765 @kindex symbol-file
11766 @item symbol-file @r{[} @var{filename} @r{]}
11767 Read symbol table information from file @var{filename}. @code{PATH} is
11768 searched when necessary. Use the @code{file} command to get both symbol
11769 table and program to run from the same file.
11771 @code{symbol-file} with no argument clears out @value{GDBN} information on your
11772 program's symbol table.
11774 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
11775 some breakpoints and auto-display expressions. This is because they may
11776 contain pointers to the internal data recording symbols and data types,
11777 which are part of the old symbol table data being discarded inside
11780 @code{symbol-file} does not repeat if you press @key{RET} again after
11783 When @value{GDBN} is configured for a particular environment, it
11784 understands debugging information in whatever format is the standard
11785 generated for that environment; you may use either a @sc{gnu} compiler, or
11786 other compilers that adhere to the local conventions.
11787 Best results are usually obtained from @sc{gnu} compilers; for example,
11788 using @code{@value{NGCC}} you can generate debugging information for
11791 For most kinds of object files, with the exception of old SVR3 systems
11792 using COFF, the @code{symbol-file} command does not normally read the
11793 symbol table in full right away. Instead, it scans the symbol table
11794 quickly to find which source files and which symbols are present. The
11795 details are read later, one source file at a time, as they are needed.
11797 The purpose of this two-stage reading strategy is to make @value{GDBN}
11798 start up faster. For the most part, it is invisible except for
11799 occasional pauses while the symbol table details for a particular source
11800 file are being read. (The @code{set verbose} command can turn these
11801 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
11802 Warnings and Messages}.)
11804 We have not implemented the two-stage strategy for COFF yet. When the
11805 symbol table is stored in COFF format, @code{symbol-file} reads the
11806 symbol table data in full right away. Note that ``stabs-in-COFF''
11807 still does the two-stage strategy, since the debug info is actually
11811 @cindex reading symbols immediately
11812 @cindex symbols, reading immediately
11813 @item symbol-file @var{filename} @r{[} -readnow @r{]}
11814 @itemx file @var{filename} @r{[} -readnow @r{]}
11815 You can override the @value{GDBN} two-stage strategy for reading symbol
11816 tables by using the @samp{-readnow} option with any of the commands that
11817 load symbol table information, if you want to be sure @value{GDBN} has the
11818 entire symbol table available.
11820 @c FIXME: for now no mention of directories, since this seems to be in
11821 @c flux. 13mar1992 status is that in theory GDB would look either in
11822 @c current dir or in same dir as myprog; but issues like competing
11823 @c GDB's, or clutter in system dirs, mean that in practice right now
11824 @c only current dir is used. FFish says maybe a special GDB hierarchy
11825 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
11829 @item core-file @r{[}@var{filename}@r{]}
11831 Specify the whereabouts of a core dump file to be used as the ``contents
11832 of memory''. Traditionally, core files contain only some parts of the
11833 address space of the process that generated them; @value{GDBN} can access the
11834 executable file itself for other parts.
11836 @code{core-file} with no argument specifies that no core file is
11839 Note that the core file is ignored when your program is actually running
11840 under @value{GDBN}. So, if you have been running your program and you
11841 wish to debug a core file instead, you must kill the subprocess in which
11842 the program is running. To do this, use the @code{kill} command
11843 (@pxref{Kill Process, ,Killing the Child Process}).
11845 @kindex add-symbol-file
11846 @cindex dynamic linking
11847 @item add-symbol-file @var{filename} @var{address}
11848 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
11849 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
11850 The @code{add-symbol-file} command reads additional symbol table
11851 information from the file @var{filename}. You would use this command
11852 when @var{filename} has been dynamically loaded (by some other means)
11853 into the program that is running. @var{address} should be the memory
11854 address at which the file has been loaded; @value{GDBN} cannot figure
11855 this out for itself. You can additionally specify an arbitrary number
11856 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
11857 section name and base address for that section. You can specify any
11858 @var{address} as an expression.
11860 The symbol table of the file @var{filename} is added to the symbol table
11861 originally read with the @code{symbol-file} command. You can use the
11862 @code{add-symbol-file} command any number of times; the new symbol data
11863 thus read keeps adding to the old. To discard all old symbol data
11864 instead, use the @code{symbol-file} command without any arguments.
11866 @cindex relocatable object files, reading symbols from
11867 @cindex object files, relocatable, reading symbols from
11868 @cindex reading symbols from relocatable object files
11869 @cindex symbols, reading from relocatable object files
11870 @cindex @file{.o} files, reading symbols from
11871 Although @var{filename} is typically a shared library file, an
11872 executable file, or some other object file which has been fully
11873 relocated for loading into a process, you can also load symbolic
11874 information from relocatable @file{.o} files, as long as:
11878 the file's symbolic information refers only to linker symbols defined in
11879 that file, not to symbols defined by other object files,
11881 every section the file's symbolic information refers to has actually
11882 been loaded into the inferior, as it appears in the file, and
11884 you can determine the address at which every section was loaded, and
11885 provide these to the @code{add-symbol-file} command.
11889 Some embedded operating systems, like Sun Chorus and VxWorks, can load
11890 relocatable files into an already running program; such systems
11891 typically make the requirements above easy to meet. However, it's
11892 important to recognize that many native systems use complex link
11893 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
11894 assembly, for example) that make the requirements difficult to meet. In
11895 general, one cannot assume that using @code{add-symbol-file} to read a
11896 relocatable object file's symbolic information will have the same effect
11897 as linking the relocatable object file into the program in the normal
11900 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
11902 @kindex add-symbol-file-from-memory
11903 @cindex @code{syscall DSO}
11904 @cindex load symbols from memory
11905 @item add-symbol-file-from-memory @var{address}
11906 Load symbols from the given @var{address} in a dynamically loaded
11907 object file whose image is mapped directly into the inferior's memory.
11908 For example, the Linux kernel maps a @code{syscall DSO} into each
11909 process's address space; this DSO provides kernel-specific code for
11910 some system calls. The argument can be any expression whose
11911 evaluation yields the address of the file's shared object file header.
11912 For this command to work, you must have used @code{symbol-file} or
11913 @code{exec-file} commands in advance.
11915 @kindex add-shared-symbol-files
11917 @item add-shared-symbol-files @var{library-file}
11918 @itemx assf @var{library-file}
11919 The @code{add-shared-symbol-files} command can currently be used only
11920 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
11921 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
11922 @value{GDBN} automatically looks for shared libraries, however if
11923 @value{GDBN} does not find yours, you can invoke
11924 @code{add-shared-symbol-files}. It takes one argument: the shared
11925 library's file name. @code{assf} is a shorthand alias for
11926 @code{add-shared-symbol-files}.
11929 @item section @var{section} @var{addr}
11930 The @code{section} command changes the base address of the named
11931 @var{section} of the exec file to @var{addr}. This can be used if the
11932 exec file does not contain section addresses, (such as in the
11933 @code{a.out} format), or when the addresses specified in the file
11934 itself are wrong. Each section must be changed separately. The
11935 @code{info files} command, described below, lists all the sections and
11939 @kindex info target
11942 @code{info files} and @code{info target} are synonymous; both print the
11943 current target (@pxref{Targets, ,Specifying a Debugging Target}),
11944 including the names of the executable and core dump files currently in
11945 use by @value{GDBN}, and the files from which symbols were loaded. The
11946 command @code{help target} lists all possible targets rather than
11949 @kindex maint info sections
11950 @item maint info sections
11951 Another command that can give you extra information about program sections
11952 is @code{maint info sections}. In addition to the section information
11953 displayed by @code{info files}, this command displays the flags and file
11954 offset of each section in the executable and core dump files. In addition,
11955 @code{maint info sections} provides the following command options (which
11956 may be arbitrarily combined):
11960 Display sections for all loaded object files, including shared libraries.
11961 @item @var{sections}
11962 Display info only for named @var{sections}.
11963 @item @var{section-flags}
11964 Display info only for sections for which @var{section-flags} are true.
11965 The section flags that @value{GDBN} currently knows about are:
11968 Section will have space allocated in the process when loaded.
11969 Set for all sections except those containing debug information.
11971 Section will be loaded from the file into the child process memory.
11972 Set for pre-initialized code and data, clear for @code{.bss} sections.
11974 Section needs to be relocated before loading.
11976 Section cannot be modified by the child process.
11978 Section contains executable code only.
11980 Section contains data only (no executable code).
11982 Section will reside in ROM.
11984 Section contains data for constructor/destructor lists.
11986 Section is not empty.
11988 An instruction to the linker to not output the section.
11989 @item COFF_SHARED_LIBRARY
11990 A notification to the linker that the section contains
11991 COFF shared library information.
11993 Section contains common symbols.
11996 @kindex set trust-readonly-sections
11997 @cindex read-only sections
11998 @item set trust-readonly-sections on
11999 Tell @value{GDBN} that readonly sections in your object file
12000 really are read-only (i.e.@: that their contents will not change).
12001 In that case, @value{GDBN} can fetch values from these sections
12002 out of the object file, rather than from the target program.
12003 For some targets (notably embedded ones), this can be a significant
12004 enhancement to debugging performance.
12006 The default is off.
12008 @item set trust-readonly-sections off
12009 Tell @value{GDBN} not to trust readonly sections. This means that
12010 the contents of the section might change while the program is running,
12011 and must therefore be fetched from the target when needed.
12013 @item show trust-readonly-sections
12014 Show the current setting of trusting readonly sections.
12017 All file-specifying commands allow both absolute and relative file names
12018 as arguments. @value{GDBN} always converts the file name to an absolute file
12019 name and remembers it that way.
12021 @cindex shared libraries
12022 @anchor{Shared Libraries}
12023 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
12024 and IBM RS/6000 AIX shared libraries.
12026 On MS-Windows @value{GDBN} must be linked with the Expat library to support
12027 shared libraries. @xref{Expat}.
12029 @value{GDBN} automatically loads symbol definitions from shared libraries
12030 when you use the @code{run} command, or when you examine a core file.
12031 (Before you issue the @code{run} command, @value{GDBN} does not understand
12032 references to a function in a shared library, however---unless you are
12033 debugging a core file).
12035 On HP-UX, if the program loads a library explicitly, @value{GDBN}
12036 automatically loads the symbols at the time of the @code{shl_load} call.
12038 @c FIXME: some @value{GDBN} release may permit some refs to undef
12039 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
12040 @c FIXME...lib; check this from time to time when updating manual
12042 There are times, however, when you may wish to not automatically load
12043 symbol definitions from shared libraries, such as when they are
12044 particularly large or there are many of them.
12046 To control the automatic loading of shared library symbols, use the
12050 @kindex set auto-solib-add
12051 @item set auto-solib-add @var{mode}
12052 If @var{mode} is @code{on}, symbols from all shared object libraries
12053 will be loaded automatically when the inferior begins execution, you
12054 attach to an independently started inferior, or when the dynamic linker
12055 informs @value{GDBN} that a new library has been loaded. If @var{mode}
12056 is @code{off}, symbols must be loaded manually, using the
12057 @code{sharedlibrary} command. The default value is @code{on}.
12059 @cindex memory used for symbol tables
12060 If your program uses lots of shared libraries with debug info that
12061 takes large amounts of memory, you can decrease the @value{GDBN}
12062 memory footprint by preventing it from automatically loading the
12063 symbols from shared libraries. To that end, type @kbd{set
12064 auto-solib-add off} before running the inferior, then load each
12065 library whose debug symbols you do need with @kbd{sharedlibrary
12066 @var{regexp}}, where @var{regexp} is a regular expression that matches
12067 the libraries whose symbols you want to be loaded.
12069 @kindex show auto-solib-add
12070 @item show auto-solib-add
12071 Display the current autoloading mode.
12074 @cindex load shared library
12075 To explicitly load shared library symbols, use the @code{sharedlibrary}
12079 @kindex info sharedlibrary
12082 @itemx info sharedlibrary
12083 Print the names of the shared libraries which are currently loaded.
12085 @kindex sharedlibrary
12087 @item sharedlibrary @var{regex}
12088 @itemx share @var{regex}
12089 Load shared object library symbols for files matching a
12090 Unix regular expression.
12091 As with files loaded automatically, it only loads shared libraries
12092 required by your program for a core file or after typing @code{run}. If
12093 @var{regex} is omitted all shared libraries required by your program are
12096 @item nosharedlibrary
12097 @kindex nosharedlibrary
12098 @cindex unload symbols from shared libraries
12099 Unload all shared object library symbols. This discards all symbols
12100 that have been loaded from all shared libraries. Symbols from shared
12101 libraries that were loaded by explicit user requests are not
12105 Sometimes you may wish that @value{GDBN} stops and gives you control
12106 when any of shared library events happen. Use the @code{set
12107 stop-on-solib-events} command for this:
12110 @item set stop-on-solib-events
12111 @kindex set stop-on-solib-events
12112 This command controls whether @value{GDBN} should give you control
12113 when the dynamic linker notifies it about some shared library event.
12114 The most common event of interest is loading or unloading of a new
12117 @item show stop-on-solib-events
12118 @kindex show stop-on-solib-events
12119 Show whether @value{GDBN} stops and gives you control when shared
12120 library events happen.
12123 Shared libraries are also supported in many cross or remote debugging
12124 configurations. A copy of the target's libraries need to be present on the
12125 host system; they need to be the same as the target libraries, although the
12126 copies on the target can be stripped as long as the copies on the host are
12129 @cindex where to look for shared libraries
12130 For remote debugging, you need to tell @value{GDBN} where the target
12131 libraries are, so that it can load the correct copies---otherwise, it
12132 may try to load the host's libraries. @value{GDBN} has two variables
12133 to specify the search directories for target libraries.
12136 @cindex prefix for shared library file names
12137 @cindex system root, alternate
12138 @kindex set solib-absolute-prefix
12139 @kindex set sysroot
12140 @item set sysroot @var{path}
12141 Use @var{path} as the system root for the program being debugged. Any
12142 absolute shared library paths will be prefixed with @var{path}; many
12143 runtime loaders store the absolute paths to the shared library in the
12144 target program's memory. If you use @code{set sysroot} to find shared
12145 libraries, they need to be laid out in the same way that they are on
12146 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
12149 The @code{set solib-absolute-prefix} command is an alias for @code{set
12152 @cindex default system root
12153 @cindex @samp{--with-sysroot}
12154 You can set the default system root by using the configure-time
12155 @samp{--with-sysroot} option. If the system root is inside
12156 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
12157 @samp{--exec-prefix}), then the default system root will be updated
12158 automatically if the installed @value{GDBN} is moved to a new
12161 @kindex show sysroot
12163 Display the current shared library prefix.
12165 @kindex set solib-search-path
12166 @item set solib-search-path @var{path}
12167 If this variable is set, @var{path} is a colon-separated list of
12168 directories to search for shared libraries. @samp{solib-search-path}
12169 is used after @samp{sysroot} fails to locate the library, or if the
12170 path to the library is relative instead of absolute. If you want to
12171 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
12172 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
12173 finding your host's libraries. @samp{sysroot} is preferred; setting
12174 it to a nonexistent directory may interfere with automatic loading
12175 of shared library symbols.
12177 @kindex show solib-search-path
12178 @item show solib-search-path
12179 Display the current shared library search path.
12183 @node Separate Debug Files
12184 @section Debugging Information in Separate Files
12185 @cindex separate debugging information files
12186 @cindex debugging information in separate files
12187 @cindex @file{.debug} subdirectories
12188 @cindex debugging information directory, global
12189 @cindex global debugging information directory
12190 @cindex build ID, and separate debugging files
12191 @cindex @file{.build-id} directory
12193 @value{GDBN} allows you to put a program's debugging information in a
12194 file separate from the executable itself, in a way that allows
12195 @value{GDBN} to find and load the debugging information automatically.
12196 Since debugging information can be very large---sometimes larger
12197 than the executable code itself---some systems distribute debugging
12198 information for their executables in separate files, which users can
12199 install only when they need to debug a problem.
12201 @value{GDBN} supports two ways of specifying the separate debug info
12206 The executable contains a @dfn{debug link} that specifies the name of
12207 the separate debug info file. The separate debug file's name is
12208 usually @file{@var{executable}.debug}, where @var{executable} is the
12209 name of the corresponding executable file without leading directories
12210 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
12211 debug link specifies a CRC32 checksum for the debug file, which
12212 @value{GDBN} uses to validate that the executable and the debug file
12213 came from the same build.
12216 The executable contains a @dfn{build ID}, a unique bit string that is
12217 also present in the corresponding debug info file. (This is supported
12218 only on some operating systems, notably those which use the ELF format
12219 for binary files and the @sc{gnu} Binutils.) For more details about
12220 this feature, see the description of the @option{--build-id}
12221 command-line option in @ref{Options, , Command Line Options, ld.info,
12222 The GNU Linker}. The debug info file's name is not specified
12223 explicitly by the build ID, but can be computed from the build ID, see
12227 Depending on the way the debug info file is specified, @value{GDBN}
12228 uses two different methods of looking for the debug file:
12232 For the ``debug link'' method, @value{GDBN} looks up the named file in
12233 the directory of the executable file, then in a subdirectory of that
12234 directory named @file{.debug}, and finally under the global debug
12235 directory, in a subdirectory whose name is identical to the leading
12236 directories of the executable's absolute file name.
12239 For the ``build ID'' method, @value{GDBN} looks in the
12240 @file{.build-id} subdirectory of the global debug directory for a file
12241 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
12242 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
12243 are the rest of the bit string. (Real build ID strings are 32 or more
12244 hex characters, not 10.)
12247 So, for example, suppose you ask @value{GDBN} to debug
12248 @file{/usr/bin/ls}, which has a debug link that specifies the
12249 file @file{ls.debug}, and a build ID whose value in hex is
12250 @code{abcdef1234}. If the global debug directory is
12251 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
12252 debug information files, in the indicated order:
12256 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
12258 @file{/usr/bin/ls.debug}
12260 @file{/usr/bin/.debug/ls.debug}
12262 @file{/usr/lib/debug/usr/bin/ls.debug}.
12265 You can set the global debugging info directory's name, and view the
12266 name @value{GDBN} is currently using.
12270 @kindex set debug-file-directory
12271 @item set debug-file-directory @var{directory}
12272 Set the directory which @value{GDBN} searches for separate debugging
12273 information files to @var{directory}.
12275 @kindex show debug-file-directory
12276 @item show debug-file-directory
12277 Show the directory @value{GDBN} searches for separate debugging
12282 @cindex @code{.gnu_debuglink} sections
12283 @cindex debug link sections
12284 A debug link is a special section of the executable file named
12285 @code{.gnu_debuglink}. The section must contain:
12289 A filename, with any leading directory components removed, followed by
12292 zero to three bytes of padding, as needed to reach the next four-byte
12293 boundary within the section, and
12295 a four-byte CRC checksum, stored in the same endianness used for the
12296 executable file itself. The checksum is computed on the debugging
12297 information file's full contents by the function given below, passing
12298 zero as the @var{crc} argument.
12301 Any executable file format can carry a debug link, as long as it can
12302 contain a section named @code{.gnu_debuglink} with the contents
12305 @cindex @code{.note.gnu.build-id} sections
12306 @cindex build ID sections
12307 The build ID is a special section in the executable file (and in other
12308 ELF binary files that @value{GDBN} may consider). This section is
12309 often named @code{.note.gnu.build-id}, but that name is not mandatory.
12310 It contains unique identification for the built files---the ID remains
12311 the same across multiple builds of the same build tree. The default
12312 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
12313 content for the build ID string. The same section with an identical
12314 value is present in the original built binary with symbols, in its
12315 stripped variant, and in the separate debugging information file.
12317 The debugging information file itself should be an ordinary
12318 executable, containing a full set of linker symbols, sections, and
12319 debugging information. The sections of the debugging information file
12320 should have the same names, addresses, and sizes as the original file,
12321 but they need not contain any data---much like a @code{.bss} section
12322 in an ordinary executable.
12324 The @sc{gnu} binary utilities (Binutils) package includes the
12325 @samp{objcopy} utility that can produce
12326 the separated executable / debugging information file pairs using the
12327 following commands:
12330 @kbd{objcopy --only-keep-debug foo foo.debug}
12335 These commands remove the debugging
12336 information from the executable file @file{foo} and place it in the file
12337 @file{foo.debug}. You can use the first, second or both methods to link the
12342 The debug link method needs the following additional command to also leave
12343 behind a debug link in @file{foo}:
12346 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
12349 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
12350 a version of the @code{strip} command such that the command @kbd{strip foo -f
12351 foo.debug} has the same functionality as the two @code{objcopy} commands and
12352 the @code{ln -s} command above, together.
12355 Build ID gets embedded into the main executable using @code{ld --build-id} or
12356 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
12357 compatibility fixes for debug files separation are present in @sc{gnu} binary
12358 utilities (Binutils) package since version 2.18.
12363 Since there are many different ways to compute CRC's for the debug
12364 link (different polynomials, reversals, byte ordering, etc.), the
12365 simplest way to describe the CRC used in @code{.gnu_debuglink}
12366 sections is to give the complete code for a function that computes it:
12368 @kindex gnu_debuglink_crc32
12371 gnu_debuglink_crc32 (unsigned long crc,
12372 unsigned char *buf, size_t len)
12374 static const unsigned long crc32_table[256] =
12376 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
12377 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
12378 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
12379 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
12380 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
12381 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
12382 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
12383 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
12384 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
12385 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
12386 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
12387 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
12388 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
12389 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
12390 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
12391 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
12392 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
12393 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
12394 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
12395 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
12396 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
12397 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
12398 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
12399 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
12400 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
12401 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
12402 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
12403 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
12404 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
12405 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
12406 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
12407 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
12408 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
12409 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
12410 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
12411 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
12412 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
12413 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
12414 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
12415 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
12416 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
12417 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
12418 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
12419 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
12420 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
12421 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
12422 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
12423 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
12424 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
12425 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
12426 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
12429 unsigned char *end;
12431 crc = ~crc & 0xffffffff;
12432 for (end = buf + len; buf < end; ++buf)
12433 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
12434 return ~crc & 0xffffffff;
12439 This computation does not apply to the ``build ID'' method.
12442 @node Symbol Errors
12443 @section Errors Reading Symbol Files
12445 While reading a symbol file, @value{GDBN} occasionally encounters problems,
12446 such as symbol types it does not recognize, or known bugs in compiler
12447 output. By default, @value{GDBN} does not notify you of such problems, since
12448 they are relatively common and primarily of interest to people
12449 debugging compilers. If you are interested in seeing information
12450 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
12451 only one message about each such type of problem, no matter how many
12452 times the problem occurs; or you can ask @value{GDBN} to print more messages,
12453 to see how many times the problems occur, with the @code{set
12454 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
12457 The messages currently printed, and their meanings, include:
12460 @item inner block not inside outer block in @var{symbol}
12462 The symbol information shows where symbol scopes begin and end
12463 (such as at the start of a function or a block of statements). This
12464 error indicates that an inner scope block is not fully contained
12465 in its outer scope blocks.
12467 @value{GDBN} circumvents the problem by treating the inner block as if it had
12468 the same scope as the outer block. In the error message, @var{symbol}
12469 may be shown as ``@code{(don't know)}'' if the outer block is not a
12472 @item block at @var{address} out of order
12474 The symbol information for symbol scope blocks should occur in
12475 order of increasing addresses. This error indicates that it does not
12478 @value{GDBN} does not circumvent this problem, and has trouble
12479 locating symbols in the source file whose symbols it is reading. (You
12480 can often determine what source file is affected by specifying
12481 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
12484 @item bad block start address patched
12486 The symbol information for a symbol scope block has a start address
12487 smaller than the address of the preceding source line. This is known
12488 to occur in the SunOS 4.1.1 (and earlier) C compiler.
12490 @value{GDBN} circumvents the problem by treating the symbol scope block as
12491 starting on the previous source line.
12493 @item bad string table offset in symbol @var{n}
12496 Symbol number @var{n} contains a pointer into the string table which is
12497 larger than the size of the string table.
12499 @value{GDBN} circumvents the problem by considering the symbol to have the
12500 name @code{foo}, which may cause other problems if many symbols end up
12503 @item unknown symbol type @code{0x@var{nn}}
12505 The symbol information contains new data types that @value{GDBN} does
12506 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
12507 uncomprehended information, in hexadecimal.
12509 @value{GDBN} circumvents the error by ignoring this symbol information.
12510 This usually allows you to debug your program, though certain symbols
12511 are not accessible. If you encounter such a problem and feel like
12512 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
12513 on @code{complain}, then go up to the function @code{read_dbx_symtab}
12514 and examine @code{*bufp} to see the symbol.
12516 @item stub type has NULL name
12518 @value{GDBN} could not find the full definition for a struct or class.
12520 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
12521 The symbol information for a C@t{++} member function is missing some
12522 information that recent versions of the compiler should have output for
12525 @item info mismatch between compiler and debugger
12527 @value{GDBN} could not parse a type specification output by the compiler.
12532 @chapter Specifying a Debugging Target
12534 @cindex debugging target
12535 A @dfn{target} is the execution environment occupied by your program.
12537 Often, @value{GDBN} runs in the same host environment as your program;
12538 in that case, the debugging target is specified as a side effect when
12539 you use the @code{file} or @code{core} commands. When you need more
12540 flexibility---for example, running @value{GDBN} on a physically separate
12541 host, or controlling a standalone system over a serial port or a
12542 realtime system over a TCP/IP connection---you can use the @code{target}
12543 command to specify one of the target types configured for @value{GDBN}
12544 (@pxref{Target Commands, ,Commands for Managing Targets}).
12546 @cindex target architecture
12547 It is possible to build @value{GDBN} for several different @dfn{target
12548 architectures}. When @value{GDBN} is built like that, you can choose
12549 one of the available architectures with the @kbd{set architecture}
12553 @kindex set architecture
12554 @kindex show architecture
12555 @item set architecture @var{arch}
12556 This command sets the current target architecture to @var{arch}. The
12557 value of @var{arch} can be @code{"auto"}, in addition to one of the
12558 supported architectures.
12560 @item show architecture
12561 Show the current target architecture.
12563 @item set processor
12565 @kindex set processor
12566 @kindex show processor
12567 These are alias commands for, respectively, @code{set architecture}
12568 and @code{show architecture}.
12572 * Active Targets:: Active targets
12573 * Target Commands:: Commands for managing targets
12574 * Byte Order:: Choosing target byte order
12577 @node Active Targets
12578 @section Active Targets
12580 @cindex stacking targets
12581 @cindex active targets
12582 @cindex multiple targets
12584 There are three classes of targets: processes, core files, and
12585 executable files. @value{GDBN} can work concurrently on up to three
12586 active targets, one in each class. This allows you to (for example)
12587 start a process and inspect its activity without abandoning your work on
12590 For example, if you execute @samp{gdb a.out}, then the executable file
12591 @code{a.out} is the only active target. If you designate a core file as
12592 well---presumably from a prior run that crashed and coredumped---then
12593 @value{GDBN} has two active targets and uses them in tandem, looking
12594 first in the corefile target, then in the executable file, to satisfy
12595 requests for memory addresses. (Typically, these two classes of target
12596 are complementary, since core files contain only a program's
12597 read-write memory---variables and so on---plus machine status, while
12598 executable files contain only the program text and initialized data.)
12600 When you type @code{run}, your executable file becomes an active process
12601 target as well. When a process target is active, all @value{GDBN}
12602 commands requesting memory addresses refer to that target; addresses in
12603 an active core file or executable file target are obscured while the
12604 process target is active.
12606 Use the @code{core-file} and @code{exec-file} commands to select a new
12607 core file or executable target (@pxref{Files, ,Commands to Specify
12608 Files}). To specify as a target a process that is already running, use
12609 the @code{attach} command (@pxref{Attach, ,Debugging an Already-running
12612 @node Target Commands
12613 @section Commands for Managing Targets
12616 @item target @var{type} @var{parameters}
12617 Connects the @value{GDBN} host environment to a target machine or
12618 process. A target is typically a protocol for talking to debugging
12619 facilities. You use the argument @var{type} to specify the type or
12620 protocol of the target machine.
12622 Further @var{parameters} are interpreted by the target protocol, but
12623 typically include things like device names or host names to connect
12624 with, process numbers, and baud rates.
12626 The @code{target} command does not repeat if you press @key{RET} again
12627 after executing the command.
12629 @kindex help target
12631 Displays the names of all targets available. To display targets
12632 currently selected, use either @code{info target} or @code{info files}
12633 (@pxref{Files, ,Commands to Specify Files}).
12635 @item help target @var{name}
12636 Describe a particular target, including any parameters necessary to
12639 @kindex set gnutarget
12640 @item set gnutarget @var{args}
12641 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
12642 knows whether it is reading an @dfn{executable},
12643 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
12644 with the @code{set gnutarget} command. Unlike most @code{target} commands,
12645 with @code{gnutarget} the @code{target} refers to a program, not a machine.
12648 @emph{Warning:} To specify a file format with @code{set gnutarget},
12649 you must know the actual BFD name.
12653 @xref{Files, , Commands to Specify Files}.
12655 @kindex show gnutarget
12656 @item show gnutarget
12657 Use the @code{show gnutarget} command to display what file format
12658 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
12659 @value{GDBN} will determine the file format for each file automatically,
12660 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
12663 @cindex common targets
12664 Here are some common targets (available, or not, depending on the GDB
12669 @item target exec @var{program}
12670 @cindex executable file target
12671 An executable file. @samp{target exec @var{program}} is the same as
12672 @samp{exec-file @var{program}}.
12674 @item target core @var{filename}
12675 @cindex core dump file target
12676 A core dump file. @samp{target core @var{filename}} is the same as
12677 @samp{core-file @var{filename}}.
12679 @item target remote @var{medium}
12680 @cindex remote target
12681 A remote system connected to @value{GDBN} via a serial line or network
12682 connection. This command tells @value{GDBN} to use its own remote
12683 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
12685 For example, if you have a board connected to @file{/dev/ttya} on the
12686 machine running @value{GDBN}, you could say:
12689 target remote /dev/ttya
12692 @code{target remote} supports the @code{load} command. This is only
12693 useful if you have some other way of getting the stub to the target
12694 system, and you can put it somewhere in memory where it won't get
12695 clobbered by the download.
12698 @cindex built-in simulator target
12699 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
12707 works; however, you cannot assume that a specific memory map, device
12708 drivers, or even basic I/O is available, although some simulators do
12709 provide these. For info about any processor-specific simulator details,
12710 see the appropriate section in @ref{Embedded Processors, ,Embedded
12715 Some configurations may include these targets as well:
12719 @item target nrom @var{dev}
12720 @cindex NetROM ROM emulator target
12721 NetROM ROM emulator. This target only supports downloading.
12725 Different targets are available on different configurations of @value{GDBN};
12726 your configuration may have more or fewer targets.
12728 Many remote targets require you to download the executable's code once
12729 you've successfully established a connection. You may wish to control
12730 various aspects of this process.
12735 @kindex set hash@r{, for remote monitors}
12736 @cindex hash mark while downloading
12737 This command controls whether a hash mark @samp{#} is displayed while
12738 downloading a file to the remote monitor. If on, a hash mark is
12739 displayed after each S-record is successfully downloaded to the
12743 @kindex show hash@r{, for remote monitors}
12744 Show the current status of displaying the hash mark.
12746 @item set debug monitor
12747 @kindex set debug monitor
12748 @cindex display remote monitor communications
12749 Enable or disable display of communications messages between
12750 @value{GDBN} and the remote monitor.
12752 @item show debug monitor
12753 @kindex show debug monitor
12754 Show the current status of displaying communications between
12755 @value{GDBN} and the remote monitor.
12760 @kindex load @var{filename}
12761 @item load @var{filename}
12763 Depending on what remote debugging facilities are configured into
12764 @value{GDBN}, the @code{load} command may be available. Where it exists, it
12765 is meant to make @var{filename} (an executable) available for debugging
12766 on the remote system---by downloading, or dynamic linking, for example.
12767 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
12768 the @code{add-symbol-file} command.
12770 If your @value{GDBN} does not have a @code{load} command, attempting to
12771 execute it gets the error message ``@code{You can't do that when your
12772 target is @dots{}}''
12774 The file is loaded at whatever address is specified in the executable.
12775 For some object file formats, you can specify the load address when you
12776 link the program; for other formats, like a.out, the object file format
12777 specifies a fixed address.
12778 @c FIXME! This would be a good place for an xref to the GNU linker doc.
12780 Depending on the remote side capabilities, @value{GDBN} may be able to
12781 load programs into flash memory.
12783 @code{load} does not repeat if you press @key{RET} again after using it.
12787 @section Choosing Target Byte Order
12789 @cindex choosing target byte order
12790 @cindex target byte order
12792 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
12793 offer the ability to run either big-endian or little-endian byte
12794 orders. Usually the executable or symbol will include a bit to
12795 designate the endian-ness, and you will not need to worry about
12796 which to use. However, you may still find it useful to adjust
12797 @value{GDBN}'s idea of processor endian-ness manually.
12801 @item set endian big
12802 Instruct @value{GDBN} to assume the target is big-endian.
12804 @item set endian little
12805 Instruct @value{GDBN} to assume the target is little-endian.
12807 @item set endian auto
12808 Instruct @value{GDBN} to use the byte order associated with the
12812 Display @value{GDBN}'s current idea of the target byte order.
12816 Note that these commands merely adjust interpretation of symbolic
12817 data on the host, and that they have absolutely no effect on the
12821 @node Remote Debugging
12822 @chapter Debugging Remote Programs
12823 @cindex remote debugging
12825 If you are trying to debug a program running on a machine that cannot run
12826 @value{GDBN} in the usual way, it is often useful to use remote debugging.
12827 For example, you might use remote debugging on an operating system kernel,
12828 or on a small system which does not have a general purpose operating system
12829 powerful enough to run a full-featured debugger.
12831 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
12832 to make this work with particular debugging targets. In addition,
12833 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
12834 but not specific to any particular target system) which you can use if you
12835 write the remote stubs---the code that runs on the remote system to
12836 communicate with @value{GDBN}.
12838 Other remote targets may be available in your
12839 configuration of @value{GDBN}; use @code{help target} to list them.
12842 * Connecting:: Connecting to a remote target
12843 * File Transfer:: Sending files to a remote system
12844 * Server:: Using the gdbserver program
12845 * Remote Configuration:: Remote configuration
12846 * Remote Stub:: Implementing a remote stub
12850 @section Connecting to a Remote Target
12852 On the @value{GDBN} host machine, you will need an unstripped copy of
12853 your program, since @value{GDBN} needs symbol and debugging information.
12854 Start up @value{GDBN} as usual, using the name of the local copy of your
12855 program as the first argument.
12857 @cindex @code{target remote}
12858 @value{GDBN} can communicate with the target over a serial line, or
12859 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
12860 each case, @value{GDBN} uses the same protocol for debugging your
12861 program; only the medium carrying the debugging packets varies. The
12862 @code{target remote} command establishes a connection to the target.
12863 Its arguments indicate which medium to use:
12867 @item target remote @var{serial-device}
12868 @cindex serial line, @code{target remote}
12869 Use @var{serial-device} to communicate with the target. For example,
12870 to use a serial line connected to the device named @file{/dev/ttyb}:
12873 target remote /dev/ttyb
12876 If you're using a serial line, you may want to give @value{GDBN} the
12877 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
12878 (@pxref{Remote Configuration, set remotebaud}) before the
12879 @code{target} command.
12881 @item target remote @code{@var{host}:@var{port}}
12882 @itemx target remote @code{tcp:@var{host}:@var{port}}
12883 @cindex @acronym{TCP} port, @code{target remote}
12884 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
12885 The @var{host} may be either a host name or a numeric @acronym{IP}
12886 address; @var{port} must be a decimal number. The @var{host} could be
12887 the target machine itself, if it is directly connected to the net, or
12888 it might be a terminal server which in turn has a serial line to the
12891 For example, to connect to port 2828 on a terminal server named
12895 target remote manyfarms:2828
12898 If your remote target is actually running on the same machine as your
12899 debugger session (e.g.@: a simulator for your target running on the
12900 same host), you can omit the hostname. For example, to connect to
12901 port 1234 on your local machine:
12904 target remote :1234
12908 Note that the colon is still required here.
12910 @item target remote @code{udp:@var{host}:@var{port}}
12911 @cindex @acronym{UDP} port, @code{target remote}
12912 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
12913 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
12916 target remote udp:manyfarms:2828
12919 When using a @acronym{UDP} connection for remote debugging, you should
12920 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
12921 can silently drop packets on busy or unreliable networks, which will
12922 cause havoc with your debugging session.
12924 @item target remote | @var{command}
12925 @cindex pipe, @code{target remote} to
12926 Run @var{command} in the background and communicate with it using a
12927 pipe. The @var{command} is a shell command, to be parsed and expanded
12928 by the system's command shell, @code{/bin/sh}; it should expect remote
12929 protocol packets on its standard input, and send replies on its
12930 standard output. You could use this to run a stand-alone simulator
12931 that speaks the remote debugging protocol, to make net connections
12932 using programs like @code{ssh}, or for other similar tricks.
12934 If @var{command} closes its standard output (perhaps by exiting),
12935 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
12936 program has already exited, this will have no effect.)
12940 Once the connection has been established, you can use all the usual
12941 commands to examine and change data. The remote program is already
12942 running; you can use @kbd{step} and @kbd{continue}, and you do not
12943 need to use @kbd{run}.
12945 @cindex interrupting remote programs
12946 @cindex remote programs, interrupting
12947 Whenever @value{GDBN} is waiting for the remote program, if you type the
12948 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
12949 program. This may or may not succeed, depending in part on the hardware
12950 and the serial drivers the remote system uses. If you type the
12951 interrupt character once again, @value{GDBN} displays this prompt:
12954 Interrupted while waiting for the program.
12955 Give up (and stop debugging it)? (y or n)
12958 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
12959 (If you decide you want to try again later, you can use @samp{target
12960 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
12961 goes back to waiting.
12964 @kindex detach (remote)
12966 When you have finished debugging the remote program, you can use the
12967 @code{detach} command to release it from @value{GDBN} control.
12968 Detaching from the target normally resumes its execution, but the results
12969 will depend on your particular remote stub. After the @code{detach}
12970 command, @value{GDBN} is free to connect to another target.
12974 The @code{disconnect} command behaves like @code{detach}, except that
12975 the target is generally not resumed. It will wait for @value{GDBN}
12976 (this instance or another one) to connect and continue debugging. After
12977 the @code{disconnect} command, @value{GDBN} is again free to connect to
12980 @cindex send command to remote monitor
12981 @cindex extend @value{GDBN} for remote targets
12982 @cindex add new commands for external monitor
12984 @item monitor @var{cmd}
12985 This command allows you to send arbitrary commands directly to the
12986 remote monitor. Since @value{GDBN} doesn't care about the commands it
12987 sends like this, this command is the way to extend @value{GDBN}---you
12988 can add new commands that only the external monitor will understand
12992 @node File Transfer
12993 @section Sending files to a remote system
12994 @cindex remote target, file transfer
12995 @cindex file transfer
12996 @cindex sending files to remote systems
12998 Some remote targets offer the ability to transfer files over the same
12999 connection used to communicate with @value{GDBN}. This is convenient
13000 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
13001 running @code{gdbserver} over a network interface. For other targets,
13002 e.g.@: embedded devices with only a single serial port, this may be
13003 the only way to upload or download files.
13005 Not all remote targets support these commands.
13009 @item remote put @var{hostfile} @var{targetfile}
13010 Copy file @var{hostfile} from the host system (the machine running
13011 @value{GDBN}) to @var{targetfile} on the target system.
13014 @item remote get @var{targetfile} @var{hostfile}
13015 Copy file @var{targetfile} from the target system to @var{hostfile}
13016 on the host system.
13018 @kindex remote delete
13019 @item remote delete @var{targetfile}
13020 Delete @var{targetfile} from the target system.
13025 @section Using the @code{gdbserver} Program
13028 @cindex remote connection without stubs
13029 @code{gdbserver} is a control program for Unix-like systems, which
13030 allows you to connect your program with a remote @value{GDBN} via
13031 @code{target remote}---but without linking in the usual debugging stub.
13033 @code{gdbserver} is not a complete replacement for the debugging stubs,
13034 because it requires essentially the same operating-system facilities
13035 that @value{GDBN} itself does. In fact, a system that can run
13036 @code{gdbserver} to connect to a remote @value{GDBN} could also run
13037 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
13038 because it is a much smaller program than @value{GDBN} itself. It is
13039 also easier to port than all of @value{GDBN}, so you may be able to get
13040 started more quickly on a new system by using @code{gdbserver}.
13041 Finally, if you develop code for real-time systems, you may find that
13042 the tradeoffs involved in real-time operation make it more convenient to
13043 do as much development work as possible on another system, for example
13044 by cross-compiling. You can use @code{gdbserver} to make a similar
13045 choice for debugging.
13047 @value{GDBN} and @code{gdbserver} communicate via either a serial line
13048 or a TCP connection, using the standard @value{GDBN} remote serial
13052 @emph{Warning:} @code{gdbserver} does not have any built-in security.
13053 Do not run @code{gdbserver} connected to any public network; a
13054 @value{GDBN} connection to @code{gdbserver} provides access to the
13055 target system with the same privileges as the user running
13059 @subsection Running @code{gdbserver}
13060 @cindex arguments, to @code{gdbserver}
13062 Run @code{gdbserver} on the target system. You need a copy of the
13063 program you want to debug, including any libraries it requires.
13064 @code{gdbserver} does not need your program's symbol table, so you can
13065 strip the program if necessary to save space. @value{GDBN} on the host
13066 system does all the symbol handling.
13068 To use the server, you must tell it how to communicate with @value{GDBN};
13069 the name of your program; and the arguments for your program. The usual
13073 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
13076 @var{comm} is either a device name (to use a serial line) or a TCP
13077 hostname and portnumber. For example, to debug Emacs with the argument
13078 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
13082 target> gdbserver /dev/com1 emacs foo.txt
13085 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
13088 To use a TCP connection instead of a serial line:
13091 target> gdbserver host:2345 emacs foo.txt
13094 The only difference from the previous example is the first argument,
13095 specifying that you are communicating with the host @value{GDBN} via
13096 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
13097 expect a TCP connection from machine @samp{host} to local TCP port 2345.
13098 (Currently, the @samp{host} part is ignored.) You can choose any number
13099 you want for the port number as long as it does not conflict with any
13100 TCP ports already in use on the target system (for example, @code{23} is
13101 reserved for @code{telnet}).@footnote{If you choose a port number that
13102 conflicts with another service, @code{gdbserver} prints an error message
13103 and exits.} You must use the same port number with the host @value{GDBN}
13104 @code{target remote} command.
13106 @subsubsection Attaching to a Running Program
13108 On some targets, @code{gdbserver} can also attach to running programs.
13109 This is accomplished via the @code{--attach} argument. The syntax is:
13112 target> gdbserver --attach @var{comm} @var{pid}
13115 @var{pid} is the process ID of a currently running process. It isn't necessary
13116 to point @code{gdbserver} at a binary for the running process.
13119 @cindex attach to a program by name
13120 You can debug processes by name instead of process ID if your target has the
13121 @code{pidof} utility:
13124 target> gdbserver --attach @var{comm} `pidof @var{program}`
13127 In case more than one copy of @var{program} is running, or @var{program}
13128 has multiple threads, most versions of @code{pidof} support the
13129 @code{-s} option to only return the first process ID.
13131 @subsubsection Multi-Process Mode for @code{gdbserver}
13132 @cindex gdbserver, multiple processes
13133 @cindex multiple processes with gdbserver
13135 When you connect to @code{gdbserver} using @code{target remote},
13136 @code{gdbserver} debugs the specified program only once. When the
13137 program exits, or you detach from it, @value{GDBN} closes the connection
13138 and @code{gdbserver} exits.
13140 If you connect using @kbd{target extended-remote}, @code{gdbserver}
13141 enters multi-process mode. When the debugged program exits, or you
13142 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
13143 though no program is running. The @code{run} and @code{attach}
13144 commands instruct @code{gdbserver} to run or attach to a new program.
13145 The @code{run} command uses @code{set remote exec-file} (@pxref{set
13146 remote exec-file}) to select the program to run. Command line
13147 arguments are supported, except for wildcard expansion and I/O
13148 redirection (@pxref{Arguments}).
13150 To start @code{gdbserver} without supplying an initial command to run
13151 or process ID to attach, use the @option{--multi} command line option.
13152 Then you can connect using @kbd{target extended-remote} and start
13153 the program you want to debug.
13155 @code{gdbserver} does not automatically exit in multi-process mode.
13156 You can terminate it by using @code{monitor exit}
13157 (@pxref{Monitor Commands for gdbserver}).
13159 @subsubsection Other Command-Line Arguments for @code{gdbserver}
13161 You can include @option{--debug} on the @code{gdbserver} command line.
13162 @code{gdbserver} will display extra status information about the debugging
13163 process. This option is intended for @code{gdbserver} development and
13164 for bug reports to the developers.
13166 The @option{--wrapper} option specifies a wrapper to launch programs
13167 for debugging. The option should be followed by the name of the
13168 wrapper, then any command-line arguments to pass to the wrapper, then
13169 @kbd{--} indicating the end of the wrapper arguments.
13171 @code{gdbserver} runs the specified wrapper program with a combined
13172 command line including the wrapper arguments, then the name of the
13173 program to debug, then any arguments to the program. The wrapper
13174 runs until it executes your program, and then @value{GDBN} gains control.
13176 You can use any program that eventually calls @code{execve} with
13177 its arguments as a wrapper. Several standard Unix utilities do
13178 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
13179 with @code{exec "$@@"} will also work.
13181 For example, you can use @code{env} to pass an environment variable to
13182 the debugged program, without setting the variable in @code{gdbserver}'s
13186 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
13189 @subsection Connecting to @code{gdbserver}
13191 Run @value{GDBN} on the host system.
13193 First make sure you have the necessary symbol files. Load symbols for
13194 your application using the @code{file} command before you connect. Use
13195 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
13196 was compiled with the correct sysroot using @code{--with-sysroot}).
13198 The symbol file and target libraries must exactly match the executable
13199 and libraries on the target, with one exception: the files on the host
13200 system should not be stripped, even if the files on the target system
13201 are. Mismatched or missing files will lead to confusing results
13202 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
13203 files may also prevent @code{gdbserver} from debugging multi-threaded
13206 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
13207 For TCP connections, you must start up @code{gdbserver} prior to using
13208 the @code{target remote} command. Otherwise you may get an error whose
13209 text depends on the host system, but which usually looks something like
13210 @samp{Connection refused}. Don't use the @code{load}
13211 command in @value{GDBN} when using @code{gdbserver}, since the program is
13212 already on the target.
13214 @subsection Monitor Commands for @code{gdbserver}
13215 @cindex monitor commands, for @code{gdbserver}
13216 @anchor{Monitor Commands for gdbserver}
13218 During a @value{GDBN} session using @code{gdbserver}, you can use the
13219 @code{monitor} command to send special requests to @code{gdbserver}.
13220 Here are the available commands.
13224 List the available monitor commands.
13226 @item monitor set debug 0
13227 @itemx monitor set debug 1
13228 Disable or enable general debugging messages.
13230 @item monitor set remote-debug 0
13231 @itemx monitor set remote-debug 1
13232 Disable or enable specific debugging messages associated with the remote
13233 protocol (@pxref{Remote Protocol}).
13236 Tell gdbserver to exit immediately. This command should be followed by
13237 @code{disconnect} to close the debugging session. @code{gdbserver} will
13238 detach from any attached processes and kill any processes it created.
13239 Use @code{monitor exit} to terminate @code{gdbserver} at the end
13240 of a multi-process mode debug session.
13244 @node Remote Configuration
13245 @section Remote Configuration
13248 @kindex show remote
13249 This section documents the configuration options available when
13250 debugging remote programs. For the options related to the File I/O
13251 extensions of the remote protocol, see @ref{system,
13252 system-call-allowed}.
13255 @item set remoteaddresssize @var{bits}
13256 @cindex address size for remote targets
13257 @cindex bits in remote address
13258 Set the maximum size of address in a memory packet to the specified
13259 number of bits. @value{GDBN} will mask off the address bits above
13260 that number, when it passes addresses to the remote target. The
13261 default value is the number of bits in the target's address.
13263 @item show remoteaddresssize
13264 Show the current value of remote address size in bits.
13266 @item set remotebaud @var{n}
13267 @cindex baud rate for remote targets
13268 Set the baud rate for the remote serial I/O to @var{n} baud. The
13269 value is used to set the speed of the serial port used for debugging
13272 @item show remotebaud
13273 Show the current speed of the remote connection.
13275 @item set remotebreak
13276 @cindex interrupt remote programs
13277 @cindex BREAK signal instead of Ctrl-C
13278 @anchor{set remotebreak}
13279 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
13280 when you type @kbd{Ctrl-c} to interrupt the program running
13281 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
13282 character instead. The default is off, since most remote systems
13283 expect to see @samp{Ctrl-C} as the interrupt signal.
13285 @item show remotebreak
13286 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
13287 interrupt the remote program.
13289 @item set remoteflow on
13290 @itemx set remoteflow off
13291 @kindex set remoteflow
13292 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
13293 on the serial port used to communicate to the remote target.
13295 @item show remoteflow
13296 @kindex show remoteflow
13297 Show the current setting of hardware flow control.
13299 @item set remotelogbase @var{base}
13300 Set the base (a.k.a.@: radix) of logging serial protocol
13301 communications to @var{base}. Supported values of @var{base} are:
13302 @code{ascii}, @code{octal}, and @code{hex}. The default is
13305 @item show remotelogbase
13306 Show the current setting of the radix for logging remote serial
13309 @item set remotelogfile @var{file}
13310 @cindex record serial communications on file
13311 Record remote serial communications on the named @var{file}. The
13312 default is not to record at all.
13314 @item show remotelogfile.
13315 Show the current setting of the file name on which to record the
13316 serial communications.
13318 @item set remotetimeout @var{num}
13319 @cindex timeout for serial communications
13320 @cindex remote timeout
13321 Set the timeout limit to wait for the remote target to respond to
13322 @var{num} seconds. The default is 2 seconds.
13324 @item show remotetimeout
13325 Show the current number of seconds to wait for the remote target
13328 @cindex limit hardware breakpoints and watchpoints
13329 @cindex remote target, limit break- and watchpoints
13330 @anchor{set remote hardware-watchpoint-limit}
13331 @anchor{set remote hardware-breakpoint-limit}
13332 @item set remote hardware-watchpoint-limit @var{limit}
13333 @itemx set remote hardware-breakpoint-limit @var{limit}
13334 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
13335 watchpoints. A limit of -1, the default, is treated as unlimited.
13337 @item set remote exec-file @var{filename}
13338 @itemx show remote exec-file
13339 @anchor{set remote exec-file}
13340 @cindex executable file, for remote target
13341 Select the file used for @code{run} with @code{target
13342 extended-remote}. This should be set to a filename valid on the
13343 target system. If it is not set, the target will use a default
13344 filename (e.g.@: the last program run).
13347 @cindex remote packets, enabling and disabling
13348 The @value{GDBN} remote protocol autodetects the packets supported by
13349 your debugging stub. If you need to override the autodetection, you
13350 can use these commands to enable or disable individual packets. Each
13351 packet can be set to @samp{on} (the remote target supports this
13352 packet), @samp{off} (the remote target does not support this packet),
13353 or @samp{auto} (detect remote target support for this packet). They
13354 all default to @samp{auto}. For more information about each packet,
13355 see @ref{Remote Protocol}.
13357 During normal use, you should not have to use any of these commands.
13358 If you do, that may be a bug in your remote debugging stub, or a bug
13359 in @value{GDBN}. You may want to report the problem to the
13360 @value{GDBN} developers.
13362 For each packet @var{name}, the command to enable or disable the
13363 packet is @code{set remote @var{name}-packet}. The available settings
13366 @multitable @columnfractions 0.28 0.32 0.25
13369 @tab Related Features
13371 @item @code{fetch-register}
13373 @tab @code{info registers}
13375 @item @code{set-register}
13379 @item @code{binary-download}
13381 @tab @code{load}, @code{set}
13383 @item @code{read-aux-vector}
13384 @tab @code{qXfer:auxv:read}
13385 @tab @code{info auxv}
13387 @item @code{symbol-lookup}
13388 @tab @code{qSymbol}
13389 @tab Detecting multiple threads
13391 @item @code{attach}
13392 @tab @code{vAttach}
13395 @item @code{verbose-resume}
13397 @tab Stepping or resuming multiple threads
13403 @item @code{software-breakpoint}
13407 @item @code{hardware-breakpoint}
13411 @item @code{write-watchpoint}
13415 @item @code{read-watchpoint}
13419 @item @code{access-watchpoint}
13423 @item @code{target-features}
13424 @tab @code{qXfer:features:read}
13425 @tab @code{set architecture}
13427 @item @code{library-info}
13428 @tab @code{qXfer:libraries:read}
13429 @tab @code{info sharedlibrary}
13431 @item @code{memory-map}
13432 @tab @code{qXfer:memory-map:read}
13433 @tab @code{info mem}
13435 @item @code{read-spu-object}
13436 @tab @code{qXfer:spu:read}
13437 @tab @code{info spu}
13439 @item @code{write-spu-object}
13440 @tab @code{qXfer:spu:write}
13441 @tab @code{info spu}
13443 @item @code{get-thread-local-@*storage-address}
13444 @tab @code{qGetTLSAddr}
13445 @tab Displaying @code{__thread} variables
13447 @item @code{supported-packets}
13448 @tab @code{qSupported}
13449 @tab Remote communications parameters
13451 @item @code{pass-signals}
13452 @tab @code{QPassSignals}
13453 @tab @code{handle @var{signal}}
13455 @item @code{hostio-close-packet}
13456 @tab @code{vFile:close}
13457 @tab @code{remote get}, @code{remote put}
13459 @item @code{hostio-open-packet}
13460 @tab @code{vFile:open}
13461 @tab @code{remote get}, @code{remote put}
13463 @item @code{hostio-pread-packet}
13464 @tab @code{vFile:pread}
13465 @tab @code{remote get}, @code{remote put}
13467 @item @code{hostio-pwrite-packet}
13468 @tab @code{vFile:pwrite}
13469 @tab @code{remote get}, @code{remote put}
13471 @item @code{hostio-unlink-packet}
13472 @tab @code{vFile:unlink}
13473 @tab @code{remote delete}
13477 @section Implementing a Remote Stub
13479 @cindex debugging stub, example
13480 @cindex remote stub, example
13481 @cindex stub example, remote debugging
13482 The stub files provided with @value{GDBN} implement the target side of the
13483 communication protocol, and the @value{GDBN} side is implemented in the
13484 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
13485 these subroutines to communicate, and ignore the details. (If you're
13486 implementing your own stub file, you can still ignore the details: start
13487 with one of the existing stub files. @file{sparc-stub.c} is the best
13488 organized, and therefore the easiest to read.)
13490 @cindex remote serial debugging, overview
13491 To debug a program running on another machine (the debugging
13492 @dfn{target} machine), you must first arrange for all the usual
13493 prerequisites for the program to run by itself. For example, for a C
13498 A startup routine to set up the C runtime environment; these usually
13499 have a name like @file{crt0}. The startup routine may be supplied by
13500 your hardware supplier, or you may have to write your own.
13503 A C subroutine library to support your program's
13504 subroutine calls, notably managing input and output.
13507 A way of getting your program to the other machine---for example, a
13508 download program. These are often supplied by the hardware
13509 manufacturer, but you may have to write your own from hardware
13513 The next step is to arrange for your program to use a serial port to
13514 communicate with the machine where @value{GDBN} is running (the @dfn{host}
13515 machine). In general terms, the scheme looks like this:
13519 @value{GDBN} already understands how to use this protocol; when everything
13520 else is set up, you can simply use the @samp{target remote} command
13521 (@pxref{Targets,,Specifying a Debugging Target}).
13523 @item On the target,
13524 you must link with your program a few special-purpose subroutines that
13525 implement the @value{GDBN} remote serial protocol. The file containing these
13526 subroutines is called a @dfn{debugging stub}.
13528 On certain remote targets, you can use an auxiliary program
13529 @code{gdbserver} instead of linking a stub into your program.
13530 @xref{Server,,Using the @code{gdbserver} Program}, for details.
13533 The debugging stub is specific to the architecture of the remote
13534 machine; for example, use @file{sparc-stub.c} to debug programs on
13537 @cindex remote serial stub list
13538 These working remote stubs are distributed with @value{GDBN}:
13543 @cindex @file{i386-stub.c}
13546 For Intel 386 and compatible architectures.
13549 @cindex @file{m68k-stub.c}
13550 @cindex Motorola 680x0
13552 For Motorola 680x0 architectures.
13555 @cindex @file{sh-stub.c}
13558 For Renesas SH architectures.
13561 @cindex @file{sparc-stub.c}
13563 For @sc{sparc} architectures.
13565 @item sparcl-stub.c
13566 @cindex @file{sparcl-stub.c}
13569 For Fujitsu @sc{sparclite} architectures.
13573 The @file{README} file in the @value{GDBN} distribution may list other
13574 recently added stubs.
13577 * Stub Contents:: What the stub can do for you
13578 * Bootstrapping:: What you must do for the stub
13579 * Debug Session:: Putting it all together
13582 @node Stub Contents
13583 @subsection What the Stub Can Do for You
13585 @cindex remote serial stub
13586 The debugging stub for your architecture supplies these three
13590 @item set_debug_traps
13591 @findex set_debug_traps
13592 @cindex remote serial stub, initialization
13593 This routine arranges for @code{handle_exception} to run when your
13594 program stops. You must call this subroutine explicitly near the
13595 beginning of your program.
13597 @item handle_exception
13598 @findex handle_exception
13599 @cindex remote serial stub, main routine
13600 This is the central workhorse, but your program never calls it
13601 explicitly---the setup code arranges for @code{handle_exception} to
13602 run when a trap is triggered.
13604 @code{handle_exception} takes control when your program stops during
13605 execution (for example, on a breakpoint), and mediates communications
13606 with @value{GDBN} on the host machine. This is where the communications
13607 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
13608 representative on the target machine. It begins by sending summary
13609 information on the state of your program, then continues to execute,
13610 retrieving and transmitting any information @value{GDBN} needs, until you
13611 execute a @value{GDBN} command that makes your program resume; at that point,
13612 @code{handle_exception} returns control to your own code on the target
13616 @cindex @code{breakpoint} subroutine, remote
13617 Use this auxiliary subroutine to make your program contain a
13618 breakpoint. Depending on the particular situation, this may be the only
13619 way for @value{GDBN} to get control. For instance, if your target
13620 machine has some sort of interrupt button, you won't need to call this;
13621 pressing the interrupt button transfers control to
13622 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
13623 simply receiving characters on the serial port may also trigger a trap;
13624 again, in that situation, you don't need to call @code{breakpoint} from
13625 your own program---simply running @samp{target remote} from the host
13626 @value{GDBN} session gets control.
13628 Call @code{breakpoint} if none of these is true, or if you simply want
13629 to make certain your program stops at a predetermined point for the
13630 start of your debugging session.
13633 @node Bootstrapping
13634 @subsection What You Must Do for the Stub
13636 @cindex remote stub, support routines
13637 The debugging stubs that come with @value{GDBN} are set up for a particular
13638 chip architecture, but they have no information about the rest of your
13639 debugging target machine.
13641 First of all you need to tell the stub how to communicate with the
13645 @item int getDebugChar()
13646 @findex getDebugChar
13647 Write this subroutine to read a single character from the serial port.
13648 It may be identical to @code{getchar} for your target system; a
13649 different name is used to allow you to distinguish the two if you wish.
13651 @item void putDebugChar(int)
13652 @findex putDebugChar
13653 Write this subroutine to write a single character to the serial port.
13654 It may be identical to @code{putchar} for your target system; a
13655 different name is used to allow you to distinguish the two if you wish.
13658 @cindex control C, and remote debugging
13659 @cindex interrupting remote targets
13660 If you want @value{GDBN} to be able to stop your program while it is
13661 running, you need to use an interrupt-driven serial driver, and arrange
13662 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
13663 character). That is the character which @value{GDBN} uses to tell the
13664 remote system to stop.
13666 Getting the debugging target to return the proper status to @value{GDBN}
13667 probably requires changes to the standard stub; one quick and dirty way
13668 is to just execute a breakpoint instruction (the ``dirty'' part is that
13669 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
13671 Other routines you need to supply are:
13674 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
13675 @findex exceptionHandler
13676 Write this function to install @var{exception_address} in the exception
13677 handling tables. You need to do this because the stub does not have any
13678 way of knowing what the exception handling tables on your target system
13679 are like (for example, the processor's table might be in @sc{rom},
13680 containing entries which point to a table in @sc{ram}).
13681 @var{exception_number} is the exception number which should be changed;
13682 its meaning is architecture-dependent (for example, different numbers
13683 might represent divide by zero, misaligned access, etc). When this
13684 exception occurs, control should be transferred directly to
13685 @var{exception_address}, and the processor state (stack, registers,
13686 and so on) should be just as it is when a processor exception occurs. So if
13687 you want to use a jump instruction to reach @var{exception_address}, it
13688 should be a simple jump, not a jump to subroutine.
13690 For the 386, @var{exception_address} should be installed as an interrupt
13691 gate so that interrupts are masked while the handler runs. The gate
13692 should be at privilege level 0 (the most privileged level). The
13693 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
13694 help from @code{exceptionHandler}.
13696 @item void flush_i_cache()
13697 @findex flush_i_cache
13698 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
13699 instruction cache, if any, on your target machine. If there is no
13700 instruction cache, this subroutine may be a no-op.
13702 On target machines that have instruction caches, @value{GDBN} requires this
13703 function to make certain that the state of your program is stable.
13707 You must also make sure this library routine is available:
13710 @item void *memset(void *, int, int)
13712 This is the standard library function @code{memset} that sets an area of
13713 memory to a known value. If you have one of the free versions of
13714 @code{libc.a}, @code{memset} can be found there; otherwise, you must
13715 either obtain it from your hardware manufacturer, or write your own.
13718 If you do not use the GNU C compiler, you may need other standard
13719 library subroutines as well; this varies from one stub to another,
13720 but in general the stubs are likely to use any of the common library
13721 subroutines which @code{@value{NGCC}} generates as inline code.
13724 @node Debug Session
13725 @subsection Putting it All Together
13727 @cindex remote serial debugging summary
13728 In summary, when your program is ready to debug, you must follow these
13733 Make sure you have defined the supporting low-level routines
13734 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
13736 @code{getDebugChar}, @code{putDebugChar},
13737 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
13741 Insert these lines near the top of your program:
13749 For the 680x0 stub only, you need to provide a variable called
13750 @code{exceptionHook}. Normally you just use:
13753 void (*exceptionHook)() = 0;
13757 but if before calling @code{set_debug_traps}, you set it to point to a
13758 function in your program, that function is called when
13759 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
13760 error). The function indicated by @code{exceptionHook} is called with
13761 one parameter: an @code{int} which is the exception number.
13764 Compile and link together: your program, the @value{GDBN} debugging stub for
13765 your target architecture, and the supporting subroutines.
13768 Make sure you have a serial connection between your target machine and
13769 the @value{GDBN} host, and identify the serial port on the host.
13772 @c The "remote" target now provides a `load' command, so we should
13773 @c document that. FIXME.
13774 Download your program to your target machine (or get it there by
13775 whatever means the manufacturer provides), and start it.
13778 Start @value{GDBN} on the host, and connect to the target
13779 (@pxref{Connecting,,Connecting to a Remote Target}).
13783 @node Configurations
13784 @chapter Configuration-Specific Information
13786 While nearly all @value{GDBN} commands are available for all native and
13787 cross versions of the debugger, there are some exceptions. This chapter
13788 describes things that are only available in certain configurations.
13790 There are three major categories of configurations: native
13791 configurations, where the host and target are the same, embedded
13792 operating system configurations, which are usually the same for several
13793 different processor architectures, and bare embedded processors, which
13794 are quite different from each other.
13799 * Embedded Processors::
13806 This section describes details specific to particular native
13811 * BSD libkvm Interface:: Debugging BSD kernel memory images
13812 * SVR4 Process Information:: SVR4 process information
13813 * DJGPP Native:: Features specific to the DJGPP port
13814 * Cygwin Native:: Features specific to the Cygwin port
13815 * Hurd Native:: Features specific to @sc{gnu} Hurd
13816 * Neutrino:: Features specific to QNX Neutrino
13822 On HP-UX systems, if you refer to a function or variable name that
13823 begins with a dollar sign, @value{GDBN} searches for a user or system
13824 name first, before it searches for a convenience variable.
13827 @node BSD libkvm Interface
13828 @subsection BSD libkvm Interface
13831 @cindex kernel memory image
13832 @cindex kernel crash dump
13834 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
13835 interface that provides a uniform interface for accessing kernel virtual
13836 memory images, including live systems and crash dumps. @value{GDBN}
13837 uses this interface to allow you to debug live kernels and kernel crash
13838 dumps on many native BSD configurations. This is implemented as a
13839 special @code{kvm} debugging target. For debugging a live system, load
13840 the currently running kernel into @value{GDBN} and connect to the
13844 (@value{GDBP}) @b{target kvm}
13847 For debugging crash dumps, provide the file name of the crash dump as an
13851 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
13854 Once connected to the @code{kvm} target, the following commands are
13860 Set current context from the @dfn{Process Control Block} (PCB) address.
13863 Set current context from proc address. This command isn't available on
13864 modern FreeBSD systems.
13867 @node SVR4 Process Information
13868 @subsection SVR4 Process Information
13870 @cindex examine process image
13871 @cindex process info via @file{/proc}
13873 Many versions of SVR4 and compatible systems provide a facility called
13874 @samp{/proc} that can be used to examine the image of a running
13875 process using file-system subroutines. If @value{GDBN} is configured
13876 for an operating system with this facility, the command @code{info
13877 proc} is available to report information about the process running
13878 your program, or about any process running on your system. @code{info
13879 proc} works only on SVR4 systems that include the @code{procfs} code.
13880 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
13881 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
13887 @itemx info proc @var{process-id}
13888 Summarize available information about any running process. If a
13889 process ID is specified by @var{process-id}, display information about
13890 that process; otherwise display information about the program being
13891 debugged. The summary includes the debugged process ID, the command
13892 line used to invoke it, its current working directory, and its
13893 executable file's absolute file name.
13895 On some systems, @var{process-id} can be of the form
13896 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
13897 within a process. If the optional @var{pid} part is missing, it means
13898 a thread from the process being debugged (the leading @samp{/} still
13899 needs to be present, or else @value{GDBN} will interpret the number as
13900 a process ID rather than a thread ID).
13902 @item info proc mappings
13903 @cindex memory address space mappings
13904 Report the memory address space ranges accessible in the program, with
13905 information on whether the process has read, write, or execute access
13906 rights to each range. On @sc{gnu}/Linux systems, each memory range
13907 includes the object file which is mapped to that range, instead of the
13908 memory access rights to that range.
13910 @item info proc stat
13911 @itemx info proc status
13912 @cindex process detailed status information
13913 These subcommands are specific to @sc{gnu}/Linux systems. They show
13914 the process-related information, including the user ID and group ID;
13915 how many threads are there in the process; its virtual memory usage;
13916 the signals that are pending, blocked, and ignored; its TTY; its
13917 consumption of system and user time; its stack size; its @samp{nice}
13918 value; etc. For more information, see the @samp{proc} man page
13919 (type @kbd{man 5 proc} from your shell prompt).
13921 @item info proc all
13922 Show all the information about the process described under all of the
13923 above @code{info proc} subcommands.
13926 @comment These sub-options of 'info proc' were not included when
13927 @comment procfs.c was re-written. Keep their descriptions around
13928 @comment against the day when someone finds the time to put them back in.
13929 @kindex info proc times
13930 @item info proc times
13931 Starting time, user CPU time, and system CPU time for your program and
13934 @kindex info proc id
13936 Report on the process IDs related to your program: its own process ID,
13937 the ID of its parent, the process group ID, and the session ID.
13940 @item set procfs-trace
13941 @kindex set procfs-trace
13942 @cindex @code{procfs} API calls
13943 This command enables and disables tracing of @code{procfs} API calls.
13945 @item show procfs-trace
13946 @kindex show procfs-trace
13947 Show the current state of @code{procfs} API call tracing.
13949 @item set procfs-file @var{file}
13950 @kindex set procfs-file
13951 Tell @value{GDBN} to write @code{procfs} API trace to the named
13952 @var{file}. @value{GDBN} appends the trace info to the previous
13953 contents of the file. The default is to display the trace on the
13956 @item show procfs-file
13957 @kindex show procfs-file
13958 Show the file to which @code{procfs} API trace is written.
13960 @item proc-trace-entry
13961 @itemx proc-trace-exit
13962 @itemx proc-untrace-entry
13963 @itemx proc-untrace-exit
13964 @kindex proc-trace-entry
13965 @kindex proc-trace-exit
13966 @kindex proc-untrace-entry
13967 @kindex proc-untrace-exit
13968 These commands enable and disable tracing of entries into and exits
13969 from the @code{syscall} interface.
13972 @kindex info pidlist
13973 @cindex process list, QNX Neutrino
13974 For QNX Neutrino only, this command displays the list of all the
13975 processes and all the threads within each process.
13978 @kindex info meminfo
13979 @cindex mapinfo list, QNX Neutrino
13980 For QNX Neutrino only, this command displays the list of all mapinfos.
13984 @subsection Features for Debugging @sc{djgpp} Programs
13985 @cindex @sc{djgpp} debugging
13986 @cindex native @sc{djgpp} debugging
13987 @cindex MS-DOS-specific commands
13990 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
13991 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
13992 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
13993 top of real-mode DOS systems and their emulations.
13995 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
13996 defines a few commands specific to the @sc{djgpp} port. This
13997 subsection describes those commands.
14002 This is a prefix of @sc{djgpp}-specific commands which print
14003 information about the target system and important OS structures.
14006 @cindex MS-DOS system info
14007 @cindex free memory information (MS-DOS)
14008 @item info dos sysinfo
14009 This command displays assorted information about the underlying
14010 platform: the CPU type and features, the OS version and flavor, the
14011 DPMI version, and the available conventional and DPMI memory.
14016 @cindex segment descriptor tables
14017 @cindex descriptor tables display
14019 @itemx info dos ldt
14020 @itemx info dos idt
14021 These 3 commands display entries from, respectively, Global, Local,
14022 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
14023 tables are data structures which store a descriptor for each segment
14024 that is currently in use. The segment's selector is an index into a
14025 descriptor table; the table entry for that index holds the
14026 descriptor's base address and limit, and its attributes and access
14029 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
14030 segment (used for both data and the stack), and a DOS segment (which
14031 allows access to DOS/BIOS data structures and absolute addresses in
14032 conventional memory). However, the DPMI host will usually define
14033 additional segments in order to support the DPMI environment.
14035 @cindex garbled pointers
14036 These commands allow to display entries from the descriptor tables.
14037 Without an argument, all entries from the specified table are
14038 displayed. An argument, which should be an integer expression, means
14039 display a single entry whose index is given by the argument. For
14040 example, here's a convenient way to display information about the
14041 debugged program's data segment:
14044 @exdent @code{(@value{GDBP}) info dos ldt $ds}
14045 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
14049 This comes in handy when you want to see whether a pointer is outside
14050 the data segment's limit (i.e.@: @dfn{garbled}).
14052 @cindex page tables display (MS-DOS)
14054 @itemx info dos pte
14055 These two commands display entries from, respectively, the Page
14056 Directory and the Page Tables. Page Directories and Page Tables are
14057 data structures which control how virtual memory addresses are mapped
14058 into physical addresses. A Page Table includes an entry for every
14059 page of memory that is mapped into the program's address space; there
14060 may be several Page Tables, each one holding up to 4096 entries. A
14061 Page Directory has up to 4096 entries, one each for every Page Table
14062 that is currently in use.
14064 Without an argument, @kbd{info dos pde} displays the entire Page
14065 Directory, and @kbd{info dos pte} displays all the entries in all of
14066 the Page Tables. An argument, an integer expression, given to the
14067 @kbd{info dos pde} command means display only that entry from the Page
14068 Directory table. An argument given to the @kbd{info dos pte} command
14069 means display entries from a single Page Table, the one pointed to by
14070 the specified entry in the Page Directory.
14072 @cindex direct memory access (DMA) on MS-DOS
14073 These commands are useful when your program uses @dfn{DMA} (Direct
14074 Memory Access), which needs physical addresses to program the DMA
14077 These commands are supported only with some DPMI servers.
14079 @cindex physical address from linear address
14080 @item info dos address-pte @var{addr}
14081 This command displays the Page Table entry for a specified linear
14082 address. The argument @var{addr} is a linear address which should
14083 already have the appropriate segment's base address added to it,
14084 because this command accepts addresses which may belong to @emph{any}
14085 segment. For example, here's how to display the Page Table entry for
14086 the page where a variable @code{i} is stored:
14089 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
14090 @exdent @code{Page Table entry for address 0x11a00d30:}
14091 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
14095 This says that @code{i} is stored at offset @code{0xd30} from the page
14096 whose physical base address is @code{0x02698000}, and shows all the
14097 attributes of that page.
14099 Note that you must cast the addresses of variables to a @code{char *},
14100 since otherwise the value of @code{__djgpp_base_address}, the base
14101 address of all variables and functions in a @sc{djgpp} program, will
14102 be added using the rules of C pointer arithmetics: if @code{i} is
14103 declared an @code{int}, @value{GDBN} will add 4 times the value of
14104 @code{__djgpp_base_address} to the address of @code{i}.
14106 Here's another example, it displays the Page Table entry for the
14110 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
14111 @exdent @code{Page Table entry for address 0x29110:}
14112 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
14116 (The @code{+ 3} offset is because the transfer buffer's address is the
14117 3rd member of the @code{_go32_info_block} structure.) The output
14118 clearly shows that this DPMI server maps the addresses in conventional
14119 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
14120 linear (@code{0x29110}) addresses are identical.
14122 This command is supported only with some DPMI servers.
14125 @cindex DOS serial data link, remote debugging
14126 In addition to native debugging, the DJGPP port supports remote
14127 debugging via a serial data link. The following commands are specific
14128 to remote serial debugging in the DJGPP port of @value{GDBN}.
14131 @kindex set com1base
14132 @kindex set com1irq
14133 @kindex set com2base
14134 @kindex set com2irq
14135 @kindex set com3base
14136 @kindex set com3irq
14137 @kindex set com4base
14138 @kindex set com4irq
14139 @item set com1base @var{addr}
14140 This command sets the base I/O port address of the @file{COM1} serial
14143 @item set com1irq @var{irq}
14144 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
14145 for the @file{COM1} serial port.
14147 There are similar commands @samp{set com2base}, @samp{set com3irq},
14148 etc.@: for setting the port address and the @code{IRQ} lines for the
14151 @kindex show com1base
14152 @kindex show com1irq
14153 @kindex show com2base
14154 @kindex show com2irq
14155 @kindex show com3base
14156 @kindex show com3irq
14157 @kindex show com4base
14158 @kindex show com4irq
14159 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
14160 display the current settings of the base address and the @code{IRQ}
14161 lines used by the COM ports.
14164 @kindex info serial
14165 @cindex DOS serial port status
14166 This command prints the status of the 4 DOS serial ports. For each
14167 port, it prints whether it's active or not, its I/O base address and
14168 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
14169 counts of various errors encountered so far.
14173 @node Cygwin Native
14174 @subsection Features for Debugging MS Windows PE Executables
14175 @cindex MS Windows debugging
14176 @cindex native Cygwin debugging
14177 @cindex Cygwin-specific commands
14179 @value{GDBN} supports native debugging of MS Windows programs, including
14180 DLLs with and without symbolic debugging information. There are various
14181 additional Cygwin-specific commands, described in this section.
14182 Working with DLLs that have no debugging symbols is described in
14183 @ref{Non-debug DLL Symbols}.
14188 This is a prefix of MS Windows-specific commands which print
14189 information about the target system and important OS structures.
14191 @item info w32 selector
14192 This command displays information returned by
14193 the Win32 API @code{GetThreadSelectorEntry} function.
14194 It takes an optional argument that is evaluated to
14195 a long value to give the information about this given selector.
14196 Without argument, this command displays information
14197 about the six segment registers.
14201 This is a Cygwin-specific alias of @code{info shared}.
14203 @kindex dll-symbols
14205 This command loads symbols from a dll similarly to
14206 add-sym command but without the need to specify a base address.
14208 @kindex set cygwin-exceptions
14209 @cindex debugging the Cygwin DLL
14210 @cindex Cygwin DLL, debugging
14211 @item set cygwin-exceptions @var{mode}
14212 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
14213 happen inside the Cygwin DLL. If @var{mode} is @code{off},
14214 @value{GDBN} will delay recognition of exceptions, and may ignore some
14215 exceptions which seem to be caused by internal Cygwin DLL
14216 ``bookkeeping''. This option is meant primarily for debugging the
14217 Cygwin DLL itself; the default value is @code{off} to avoid annoying
14218 @value{GDBN} users with false @code{SIGSEGV} signals.
14220 @kindex show cygwin-exceptions
14221 @item show cygwin-exceptions
14222 Displays whether @value{GDBN} will break on exceptions that happen
14223 inside the Cygwin DLL itself.
14225 @kindex set new-console
14226 @item set new-console @var{mode}
14227 If @var{mode} is @code{on} the debuggee will
14228 be started in a new console on next start.
14229 If @var{mode} is @code{off}i, the debuggee will
14230 be started in the same console as the debugger.
14232 @kindex show new-console
14233 @item show new-console
14234 Displays whether a new console is used
14235 when the debuggee is started.
14237 @kindex set new-group
14238 @item set new-group @var{mode}
14239 This boolean value controls whether the debuggee should
14240 start a new group or stay in the same group as the debugger.
14241 This affects the way the Windows OS handles
14244 @kindex show new-group
14245 @item show new-group
14246 Displays current value of new-group boolean.
14248 @kindex set debugevents
14249 @item set debugevents
14250 This boolean value adds debug output concerning kernel events related
14251 to the debuggee seen by the debugger. This includes events that
14252 signal thread and process creation and exit, DLL loading and
14253 unloading, console interrupts, and debugging messages produced by the
14254 Windows @code{OutputDebugString} API call.
14256 @kindex set debugexec
14257 @item set debugexec
14258 This boolean value adds debug output concerning execute events
14259 (such as resume thread) seen by the debugger.
14261 @kindex set debugexceptions
14262 @item set debugexceptions
14263 This boolean value adds debug output concerning exceptions in the
14264 debuggee seen by the debugger.
14266 @kindex set debugmemory
14267 @item set debugmemory
14268 This boolean value adds debug output concerning debuggee memory reads
14269 and writes by the debugger.
14273 This boolean values specifies whether the debuggee is called
14274 via a shell or directly (default value is on).
14278 Displays if the debuggee will be started with a shell.
14283 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
14286 @node Non-debug DLL Symbols
14287 @subsubsection Support for DLLs without Debugging Symbols
14288 @cindex DLLs with no debugging symbols
14289 @cindex Minimal symbols and DLLs
14291 Very often on windows, some of the DLLs that your program relies on do
14292 not include symbolic debugging information (for example,
14293 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
14294 symbols in a DLL, it relies on the minimal amount of symbolic
14295 information contained in the DLL's export table. This section
14296 describes working with such symbols, known internally to @value{GDBN} as
14297 ``minimal symbols''.
14299 Note that before the debugged program has started execution, no DLLs
14300 will have been loaded. The easiest way around this problem is simply to
14301 start the program --- either by setting a breakpoint or letting the
14302 program run once to completion. It is also possible to force
14303 @value{GDBN} to load a particular DLL before starting the executable ---
14304 see the shared library information in @ref{Files}, or the
14305 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
14306 explicitly loading symbols from a DLL with no debugging information will
14307 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
14308 which may adversely affect symbol lookup performance.
14310 @subsubsection DLL Name Prefixes
14312 In keeping with the naming conventions used by the Microsoft debugging
14313 tools, DLL export symbols are made available with a prefix based on the
14314 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
14315 also entered into the symbol table, so @code{CreateFileA} is often
14316 sufficient. In some cases there will be name clashes within a program
14317 (particularly if the executable itself includes full debugging symbols)
14318 necessitating the use of the fully qualified name when referring to the
14319 contents of the DLL. Use single-quotes around the name to avoid the
14320 exclamation mark (``!'') being interpreted as a language operator.
14322 Note that the internal name of the DLL may be all upper-case, even
14323 though the file name of the DLL is lower-case, or vice-versa. Since
14324 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
14325 some confusion. If in doubt, try the @code{info functions} and
14326 @code{info variables} commands or even @code{maint print msymbols}
14327 (@pxref{Symbols}). Here's an example:
14330 (@value{GDBP}) info function CreateFileA
14331 All functions matching regular expression "CreateFileA":
14333 Non-debugging symbols:
14334 0x77e885f4 CreateFileA
14335 0x77e885f4 KERNEL32!CreateFileA
14339 (@value{GDBP}) info function !
14340 All functions matching regular expression "!":
14342 Non-debugging symbols:
14343 0x6100114c cygwin1!__assert
14344 0x61004034 cygwin1!_dll_crt0@@0
14345 0x61004240 cygwin1!dll_crt0(per_process *)
14349 @subsubsection Working with Minimal Symbols
14351 Symbols extracted from a DLL's export table do not contain very much
14352 type information. All that @value{GDBN} can do is guess whether a symbol
14353 refers to a function or variable depending on the linker section that
14354 contains the symbol. Also note that the actual contents of the memory
14355 contained in a DLL are not available unless the program is running. This
14356 means that you cannot examine the contents of a variable or disassemble
14357 a function within a DLL without a running program.
14359 Variables are generally treated as pointers and dereferenced
14360 automatically. For this reason, it is often necessary to prefix a
14361 variable name with the address-of operator (``&'') and provide explicit
14362 type information in the command. Here's an example of the type of
14366 (@value{GDBP}) print 'cygwin1!__argv'
14371 (@value{GDBP}) x 'cygwin1!__argv'
14372 0x10021610: "\230y\""
14375 And two possible solutions:
14378 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
14379 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
14383 (@value{GDBP}) x/2x &'cygwin1!__argv'
14384 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
14385 (@value{GDBP}) x/x 0x10021608
14386 0x10021608: 0x0022fd98
14387 (@value{GDBP}) x/s 0x0022fd98
14388 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
14391 Setting a break point within a DLL is possible even before the program
14392 starts execution. However, under these circumstances, @value{GDBN} can't
14393 examine the initial instructions of the function in order to skip the
14394 function's frame set-up code. You can work around this by using ``*&''
14395 to set the breakpoint at a raw memory address:
14398 (@value{GDBP}) break *&'python22!PyOS_Readline'
14399 Breakpoint 1 at 0x1e04eff0
14402 The author of these extensions is not entirely convinced that setting a
14403 break point within a shared DLL like @file{kernel32.dll} is completely
14407 @subsection Commands Specific to @sc{gnu} Hurd Systems
14408 @cindex @sc{gnu} Hurd debugging
14410 This subsection describes @value{GDBN} commands specific to the
14411 @sc{gnu} Hurd native debugging.
14416 @kindex set signals@r{, Hurd command}
14417 @kindex set sigs@r{, Hurd command}
14418 This command toggles the state of inferior signal interception by
14419 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
14420 affected by this command. @code{sigs} is a shorthand alias for
14425 @kindex show signals@r{, Hurd command}
14426 @kindex show sigs@r{, Hurd command}
14427 Show the current state of intercepting inferior's signals.
14429 @item set signal-thread
14430 @itemx set sigthread
14431 @kindex set signal-thread
14432 @kindex set sigthread
14433 This command tells @value{GDBN} which thread is the @code{libc} signal
14434 thread. That thread is run when a signal is delivered to a running
14435 process. @code{set sigthread} is the shorthand alias of @code{set
14438 @item show signal-thread
14439 @itemx show sigthread
14440 @kindex show signal-thread
14441 @kindex show sigthread
14442 These two commands show which thread will run when the inferior is
14443 delivered a signal.
14446 @kindex set stopped@r{, Hurd command}
14447 This commands tells @value{GDBN} that the inferior process is stopped,
14448 as with the @code{SIGSTOP} signal. The stopped process can be
14449 continued by delivering a signal to it.
14452 @kindex show stopped@r{, Hurd command}
14453 This command shows whether @value{GDBN} thinks the debuggee is
14456 @item set exceptions
14457 @kindex set exceptions@r{, Hurd command}
14458 Use this command to turn off trapping of exceptions in the inferior.
14459 When exception trapping is off, neither breakpoints nor
14460 single-stepping will work. To restore the default, set exception
14463 @item show exceptions
14464 @kindex show exceptions@r{, Hurd command}
14465 Show the current state of trapping exceptions in the inferior.
14467 @item set task pause
14468 @kindex set task@r{, Hurd commands}
14469 @cindex task attributes (@sc{gnu} Hurd)
14470 @cindex pause current task (@sc{gnu} Hurd)
14471 This command toggles task suspension when @value{GDBN} has control.
14472 Setting it to on takes effect immediately, and the task is suspended
14473 whenever @value{GDBN} gets control. Setting it to off will take
14474 effect the next time the inferior is continued. If this option is set
14475 to off, you can use @code{set thread default pause on} or @code{set
14476 thread pause on} (see below) to pause individual threads.
14478 @item show task pause
14479 @kindex show task@r{, Hurd commands}
14480 Show the current state of task suspension.
14482 @item set task detach-suspend-count
14483 @cindex task suspend count
14484 @cindex detach from task, @sc{gnu} Hurd
14485 This command sets the suspend count the task will be left with when
14486 @value{GDBN} detaches from it.
14488 @item show task detach-suspend-count
14489 Show the suspend count the task will be left with when detaching.
14491 @item set task exception-port
14492 @itemx set task excp
14493 @cindex task exception port, @sc{gnu} Hurd
14494 This command sets the task exception port to which @value{GDBN} will
14495 forward exceptions. The argument should be the value of the @dfn{send
14496 rights} of the task. @code{set task excp} is a shorthand alias.
14498 @item set noninvasive
14499 @cindex noninvasive task options
14500 This command switches @value{GDBN} to a mode that is the least
14501 invasive as far as interfering with the inferior is concerned. This
14502 is the same as using @code{set task pause}, @code{set exceptions}, and
14503 @code{set signals} to values opposite to the defaults.
14505 @item info send-rights
14506 @itemx info receive-rights
14507 @itemx info port-rights
14508 @itemx info port-sets
14509 @itemx info dead-names
14512 @cindex send rights, @sc{gnu} Hurd
14513 @cindex receive rights, @sc{gnu} Hurd
14514 @cindex port rights, @sc{gnu} Hurd
14515 @cindex port sets, @sc{gnu} Hurd
14516 @cindex dead names, @sc{gnu} Hurd
14517 These commands display information about, respectively, send rights,
14518 receive rights, port rights, port sets, and dead names of a task.
14519 There are also shorthand aliases: @code{info ports} for @code{info
14520 port-rights} and @code{info psets} for @code{info port-sets}.
14522 @item set thread pause
14523 @kindex set thread@r{, Hurd command}
14524 @cindex thread properties, @sc{gnu} Hurd
14525 @cindex pause current thread (@sc{gnu} Hurd)
14526 This command toggles current thread suspension when @value{GDBN} has
14527 control. Setting it to on takes effect immediately, and the current
14528 thread is suspended whenever @value{GDBN} gets control. Setting it to
14529 off will take effect the next time the inferior is continued.
14530 Normally, this command has no effect, since when @value{GDBN} has
14531 control, the whole task is suspended. However, if you used @code{set
14532 task pause off} (see above), this command comes in handy to suspend
14533 only the current thread.
14535 @item show thread pause
14536 @kindex show thread@r{, Hurd command}
14537 This command shows the state of current thread suspension.
14539 @item set thread run
14540 This command sets whether the current thread is allowed to run.
14542 @item show thread run
14543 Show whether the current thread is allowed to run.
14545 @item set thread detach-suspend-count
14546 @cindex thread suspend count, @sc{gnu} Hurd
14547 @cindex detach from thread, @sc{gnu} Hurd
14548 This command sets the suspend count @value{GDBN} will leave on a
14549 thread when detaching. This number is relative to the suspend count
14550 found by @value{GDBN} when it notices the thread; use @code{set thread
14551 takeover-suspend-count} to force it to an absolute value.
14553 @item show thread detach-suspend-count
14554 Show the suspend count @value{GDBN} will leave on the thread when
14557 @item set thread exception-port
14558 @itemx set thread excp
14559 Set the thread exception port to which to forward exceptions. This
14560 overrides the port set by @code{set task exception-port} (see above).
14561 @code{set thread excp} is the shorthand alias.
14563 @item set thread takeover-suspend-count
14564 Normally, @value{GDBN}'s thread suspend counts are relative to the
14565 value @value{GDBN} finds when it notices each thread. This command
14566 changes the suspend counts to be absolute instead.
14568 @item set thread default
14569 @itemx show thread default
14570 @cindex thread default settings, @sc{gnu} Hurd
14571 Each of the above @code{set thread} commands has a @code{set thread
14572 default} counterpart (e.g., @code{set thread default pause}, @code{set
14573 thread default exception-port}, etc.). The @code{thread default}
14574 variety of commands sets the default thread properties for all
14575 threads; you can then change the properties of individual threads with
14576 the non-default commands.
14581 @subsection QNX Neutrino
14582 @cindex QNX Neutrino
14584 @value{GDBN} provides the following commands specific to the QNX
14588 @item set debug nto-debug
14589 @kindex set debug nto-debug
14590 When set to on, enables debugging messages specific to the QNX
14593 @item show debug nto-debug
14594 @kindex show debug nto-debug
14595 Show the current state of QNX Neutrino messages.
14600 @section Embedded Operating Systems
14602 This section describes configurations involving the debugging of
14603 embedded operating systems that are available for several different
14607 * VxWorks:: Using @value{GDBN} with VxWorks
14610 @value{GDBN} includes the ability to debug programs running on
14611 various real-time operating systems.
14614 @subsection Using @value{GDBN} with VxWorks
14620 @kindex target vxworks
14621 @item target vxworks @var{machinename}
14622 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
14623 is the target system's machine name or IP address.
14627 On VxWorks, @code{load} links @var{filename} dynamically on the
14628 current target system as well as adding its symbols in @value{GDBN}.
14630 @value{GDBN} enables developers to spawn and debug tasks running on networked
14631 VxWorks targets from a Unix host. Already-running tasks spawned from
14632 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
14633 both the Unix host and on the VxWorks target. The program
14634 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
14635 installed with the name @code{vxgdb}, to distinguish it from a
14636 @value{GDBN} for debugging programs on the host itself.)
14639 @item VxWorks-timeout @var{args}
14640 @kindex vxworks-timeout
14641 All VxWorks-based targets now support the option @code{vxworks-timeout}.
14642 This option is set by the user, and @var{args} represents the number of
14643 seconds @value{GDBN} waits for responses to rpc's. You might use this if
14644 your VxWorks target is a slow software simulator or is on the far side
14645 of a thin network line.
14648 The following information on connecting to VxWorks was current when
14649 this manual was produced; newer releases of VxWorks may use revised
14652 @findex INCLUDE_RDB
14653 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
14654 to include the remote debugging interface routines in the VxWorks
14655 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
14656 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
14657 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
14658 source debugging task @code{tRdbTask} when VxWorks is booted. For more
14659 information on configuring and remaking VxWorks, see the manufacturer's
14661 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
14663 Once you have included @file{rdb.a} in your VxWorks system image and set
14664 your Unix execution search path to find @value{GDBN}, you are ready to
14665 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
14666 @code{vxgdb}, depending on your installation).
14668 @value{GDBN} comes up showing the prompt:
14675 * VxWorks Connection:: Connecting to VxWorks
14676 * VxWorks Download:: VxWorks download
14677 * VxWorks Attach:: Running tasks
14680 @node VxWorks Connection
14681 @subsubsection Connecting to VxWorks
14683 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
14684 network. To connect to a target whose host name is ``@code{tt}'', type:
14687 (vxgdb) target vxworks tt
14691 @value{GDBN} displays messages like these:
14694 Attaching remote machine across net...
14699 @value{GDBN} then attempts to read the symbol tables of any object modules
14700 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
14701 these files by searching the directories listed in the command search
14702 path (@pxref{Environment, ,Your Program's Environment}); if it fails
14703 to find an object file, it displays a message such as:
14706 prog.o: No such file or directory.
14709 When this happens, add the appropriate directory to the search path with
14710 the @value{GDBN} command @code{path}, and execute the @code{target}
14713 @node VxWorks Download
14714 @subsubsection VxWorks Download
14716 @cindex download to VxWorks
14717 If you have connected to the VxWorks target and you want to debug an
14718 object that has not yet been loaded, you can use the @value{GDBN}
14719 @code{load} command to download a file from Unix to VxWorks
14720 incrementally. The object file given as an argument to the @code{load}
14721 command is actually opened twice: first by the VxWorks target in order
14722 to download the code, then by @value{GDBN} in order to read the symbol
14723 table. This can lead to problems if the current working directories on
14724 the two systems differ. If both systems have NFS mounted the same
14725 filesystems, you can avoid these problems by using absolute paths.
14726 Otherwise, it is simplest to set the working directory on both systems
14727 to the directory in which the object file resides, and then to reference
14728 the file by its name, without any path. For instance, a program
14729 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
14730 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
14731 program, type this on VxWorks:
14734 -> cd "@var{vxpath}/vw/demo/rdb"
14738 Then, in @value{GDBN}, type:
14741 (vxgdb) cd @var{hostpath}/vw/demo/rdb
14742 (vxgdb) load prog.o
14745 @value{GDBN} displays a response similar to this:
14748 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
14751 You can also use the @code{load} command to reload an object module
14752 after editing and recompiling the corresponding source file. Note that
14753 this makes @value{GDBN} delete all currently-defined breakpoints,
14754 auto-displays, and convenience variables, and to clear the value
14755 history. (This is necessary in order to preserve the integrity of
14756 debugger's data structures that reference the target system's symbol
14759 @node VxWorks Attach
14760 @subsubsection Running Tasks
14762 @cindex running VxWorks tasks
14763 You can also attach to an existing task using the @code{attach} command as
14767 (vxgdb) attach @var{task}
14771 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
14772 or suspended when you attach to it. Running tasks are suspended at
14773 the time of attachment.
14775 @node Embedded Processors
14776 @section Embedded Processors
14778 This section goes into details specific to particular embedded
14781 @cindex send command to simulator
14782 Whenever a specific embedded processor has a simulator, @value{GDBN}
14783 allows to send an arbitrary command to the simulator.
14786 @item sim @var{command}
14787 @kindex sim@r{, a command}
14788 Send an arbitrary @var{command} string to the simulator. Consult the
14789 documentation for the specific simulator in use for information about
14790 acceptable commands.
14796 * M32R/D:: Renesas M32R/D
14797 * M68K:: Motorola M68K
14798 * MIPS Embedded:: MIPS Embedded
14799 * OpenRISC 1000:: OpenRisc 1000
14800 * PA:: HP PA Embedded
14801 * PowerPC Embedded:: PowerPC Embedded
14802 * Sparclet:: Tsqware Sparclet
14803 * Sparclite:: Fujitsu Sparclite
14804 * Z8000:: Zilog Z8000
14807 * Super-H:: Renesas Super-H
14816 @item target rdi @var{dev}
14817 ARM Angel monitor, via RDI library interface to ADP protocol. You may
14818 use this target to communicate with both boards running the Angel
14819 monitor, or with the EmbeddedICE JTAG debug device.
14822 @item target rdp @var{dev}
14827 @value{GDBN} provides the following ARM-specific commands:
14830 @item set arm disassembler
14832 This commands selects from a list of disassembly styles. The
14833 @code{"std"} style is the standard style.
14835 @item show arm disassembler
14837 Show the current disassembly style.
14839 @item set arm apcs32
14840 @cindex ARM 32-bit mode
14841 This command toggles ARM operation mode between 32-bit and 26-bit.
14843 @item show arm apcs32
14844 Display the current usage of the ARM 32-bit mode.
14846 @item set arm fpu @var{fputype}
14847 This command sets the ARM floating-point unit (FPU) type. The
14848 argument @var{fputype} can be one of these:
14852 Determine the FPU type by querying the OS ABI.
14854 Software FPU, with mixed-endian doubles on little-endian ARM
14857 GCC-compiled FPA co-processor.
14859 Software FPU with pure-endian doubles.
14865 Show the current type of the FPU.
14868 This command forces @value{GDBN} to use the specified ABI.
14871 Show the currently used ABI.
14873 @item set debug arm
14874 Toggle whether to display ARM-specific debugging messages from the ARM
14875 target support subsystem.
14877 @item show debug arm
14878 Show whether ARM-specific debugging messages are enabled.
14881 The following commands are available when an ARM target is debugged
14882 using the RDI interface:
14885 @item rdilogfile @r{[}@var{file}@r{]}
14887 @cindex ADP (Angel Debugger Protocol) logging
14888 Set the filename for the ADP (Angel Debugger Protocol) packet log.
14889 With an argument, sets the log file to the specified @var{file}. With
14890 no argument, show the current log file name. The default log file is
14893 @item rdilogenable @r{[}@var{arg}@r{]}
14894 @kindex rdilogenable
14895 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
14896 enables logging, with an argument 0 or @code{"no"} disables it. With
14897 no arguments displays the current setting. When logging is enabled,
14898 ADP packets exchanged between @value{GDBN} and the RDI target device
14899 are logged to a file.
14901 @item set rdiromatzero
14902 @kindex set rdiromatzero
14903 @cindex ROM at zero address, RDI
14904 Tell @value{GDBN} whether the target has ROM at address 0. If on,
14905 vector catching is disabled, so that zero address can be used. If off
14906 (the default), vector catching is enabled. For this command to take
14907 effect, it needs to be invoked prior to the @code{target rdi} command.
14909 @item show rdiromatzero
14910 @kindex show rdiromatzero
14911 Show the current setting of ROM at zero address.
14913 @item set rdiheartbeat
14914 @kindex set rdiheartbeat
14915 @cindex RDI heartbeat
14916 Enable or disable RDI heartbeat packets. It is not recommended to
14917 turn on this option, since it confuses ARM and EPI JTAG interface, as
14918 well as the Angel monitor.
14920 @item show rdiheartbeat
14921 @kindex show rdiheartbeat
14922 Show the setting of RDI heartbeat packets.
14927 @subsection Renesas M32R/D and M32R/SDI
14930 @kindex target m32r
14931 @item target m32r @var{dev}
14932 Renesas M32R/D ROM monitor.
14934 @kindex target m32rsdi
14935 @item target m32rsdi @var{dev}
14936 Renesas M32R SDI server, connected via parallel port to the board.
14939 The following @value{GDBN} commands are specific to the M32R monitor:
14942 @item set download-path @var{path}
14943 @kindex set download-path
14944 @cindex find downloadable @sc{srec} files (M32R)
14945 Set the default path for finding downloadable @sc{srec} files.
14947 @item show download-path
14948 @kindex show download-path
14949 Show the default path for downloadable @sc{srec} files.
14951 @item set board-address @var{addr}
14952 @kindex set board-address
14953 @cindex M32-EVA target board address
14954 Set the IP address for the M32R-EVA target board.
14956 @item show board-address
14957 @kindex show board-address
14958 Show the current IP address of the target board.
14960 @item set server-address @var{addr}
14961 @kindex set server-address
14962 @cindex download server address (M32R)
14963 Set the IP address for the download server, which is the @value{GDBN}'s
14966 @item show server-address
14967 @kindex show server-address
14968 Display the IP address of the download server.
14970 @item upload @r{[}@var{file}@r{]}
14971 @kindex upload@r{, M32R}
14972 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
14973 upload capability. If no @var{file} argument is given, the current
14974 executable file is uploaded.
14976 @item tload @r{[}@var{file}@r{]}
14977 @kindex tload@r{, M32R}
14978 Test the @code{upload} command.
14981 The following commands are available for M32R/SDI:
14986 @cindex reset SDI connection, M32R
14987 This command resets the SDI connection.
14991 This command shows the SDI connection status.
14994 @kindex debug_chaos
14995 @cindex M32R/Chaos debugging
14996 Instructs the remote that M32R/Chaos debugging is to be used.
14998 @item use_debug_dma
14999 @kindex use_debug_dma
15000 Instructs the remote to use the DEBUG_DMA method of accessing memory.
15003 @kindex use_mon_code
15004 Instructs the remote to use the MON_CODE method of accessing memory.
15007 @kindex use_ib_break
15008 Instructs the remote to set breakpoints by IB break.
15010 @item use_dbt_break
15011 @kindex use_dbt_break
15012 Instructs the remote to set breakpoints by DBT.
15018 The Motorola m68k configuration includes ColdFire support, and a
15019 target command for the following ROM monitor.
15023 @kindex target dbug
15024 @item target dbug @var{dev}
15025 dBUG ROM monitor for Motorola ColdFire.
15029 @node MIPS Embedded
15030 @subsection MIPS Embedded
15032 @cindex MIPS boards
15033 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
15034 MIPS board attached to a serial line. This is available when
15035 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
15038 Use these @value{GDBN} commands to specify the connection to your target board:
15041 @item target mips @var{port}
15042 @kindex target mips @var{port}
15043 To run a program on the board, start up @code{@value{GDBP}} with the
15044 name of your program as the argument. To connect to the board, use the
15045 command @samp{target mips @var{port}}, where @var{port} is the name of
15046 the serial port connected to the board. If the program has not already
15047 been downloaded to the board, you may use the @code{load} command to
15048 download it. You can then use all the usual @value{GDBN} commands.
15050 For example, this sequence connects to the target board through a serial
15051 port, and loads and runs a program called @var{prog} through the
15055 host$ @value{GDBP} @var{prog}
15056 @value{GDBN} is free software and @dots{}
15057 (@value{GDBP}) target mips /dev/ttyb
15058 (@value{GDBP}) load @var{prog}
15062 @item target mips @var{hostname}:@var{portnumber}
15063 On some @value{GDBN} host configurations, you can specify a TCP
15064 connection (for instance, to a serial line managed by a terminal
15065 concentrator) instead of a serial port, using the syntax
15066 @samp{@var{hostname}:@var{portnumber}}.
15068 @item target pmon @var{port}
15069 @kindex target pmon @var{port}
15072 @item target ddb @var{port}
15073 @kindex target ddb @var{port}
15074 NEC's DDB variant of PMON for Vr4300.
15076 @item target lsi @var{port}
15077 @kindex target lsi @var{port}
15078 LSI variant of PMON.
15080 @kindex target r3900
15081 @item target r3900 @var{dev}
15082 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
15084 @kindex target array
15085 @item target array @var{dev}
15086 Array Tech LSI33K RAID controller board.
15092 @value{GDBN} also supports these special commands for MIPS targets:
15095 @item set mipsfpu double
15096 @itemx set mipsfpu single
15097 @itemx set mipsfpu none
15098 @itemx set mipsfpu auto
15099 @itemx show mipsfpu
15100 @kindex set mipsfpu
15101 @kindex show mipsfpu
15102 @cindex MIPS remote floating point
15103 @cindex floating point, MIPS remote
15104 If your target board does not support the MIPS floating point
15105 coprocessor, you should use the command @samp{set mipsfpu none} (if you
15106 need this, you may wish to put the command in your @value{GDBN} init
15107 file). This tells @value{GDBN} how to find the return value of
15108 functions which return floating point values. It also allows
15109 @value{GDBN} to avoid saving the floating point registers when calling
15110 functions on the board. If you are using a floating point coprocessor
15111 with only single precision floating point support, as on the @sc{r4650}
15112 processor, use the command @samp{set mipsfpu single}. The default
15113 double precision floating point coprocessor may be selected using
15114 @samp{set mipsfpu double}.
15116 In previous versions the only choices were double precision or no
15117 floating point, so @samp{set mipsfpu on} will select double precision
15118 and @samp{set mipsfpu off} will select no floating point.
15120 As usual, you can inquire about the @code{mipsfpu} variable with
15121 @samp{show mipsfpu}.
15123 @item set timeout @var{seconds}
15124 @itemx set retransmit-timeout @var{seconds}
15125 @itemx show timeout
15126 @itemx show retransmit-timeout
15127 @cindex @code{timeout}, MIPS protocol
15128 @cindex @code{retransmit-timeout}, MIPS protocol
15129 @kindex set timeout
15130 @kindex show timeout
15131 @kindex set retransmit-timeout
15132 @kindex show retransmit-timeout
15133 You can control the timeout used while waiting for a packet, in the MIPS
15134 remote protocol, with the @code{set timeout @var{seconds}} command. The
15135 default is 5 seconds. Similarly, you can control the timeout used while
15136 waiting for an acknowledgement of a packet with the @code{set
15137 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
15138 You can inspect both values with @code{show timeout} and @code{show
15139 retransmit-timeout}. (These commands are @emph{only} available when
15140 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
15142 The timeout set by @code{set timeout} does not apply when @value{GDBN}
15143 is waiting for your program to stop. In that case, @value{GDBN} waits
15144 forever because it has no way of knowing how long the program is going
15145 to run before stopping.
15147 @item set syn-garbage-limit @var{num}
15148 @kindex set syn-garbage-limit@r{, MIPS remote}
15149 @cindex synchronize with remote MIPS target
15150 Limit the maximum number of characters @value{GDBN} should ignore when
15151 it tries to synchronize with the remote target. The default is 10
15152 characters. Setting the limit to -1 means there's no limit.
15154 @item show syn-garbage-limit
15155 @kindex show syn-garbage-limit@r{, MIPS remote}
15156 Show the current limit on the number of characters to ignore when
15157 trying to synchronize with the remote system.
15159 @item set monitor-prompt @var{prompt}
15160 @kindex set monitor-prompt@r{, MIPS remote}
15161 @cindex remote monitor prompt
15162 Tell @value{GDBN} to expect the specified @var{prompt} string from the
15163 remote monitor. The default depends on the target:
15173 @item show monitor-prompt
15174 @kindex show monitor-prompt@r{, MIPS remote}
15175 Show the current strings @value{GDBN} expects as the prompt from the
15178 @item set monitor-warnings
15179 @kindex set monitor-warnings@r{, MIPS remote}
15180 Enable or disable monitor warnings about hardware breakpoints. This
15181 has effect only for the @code{lsi} target. When on, @value{GDBN} will
15182 display warning messages whose codes are returned by the @code{lsi}
15183 PMON monitor for breakpoint commands.
15185 @item show monitor-warnings
15186 @kindex show monitor-warnings@r{, MIPS remote}
15187 Show the current setting of printing monitor warnings.
15189 @item pmon @var{command}
15190 @kindex pmon@r{, MIPS remote}
15191 @cindex send PMON command
15192 This command allows sending an arbitrary @var{command} string to the
15193 monitor. The monitor must be in debug mode for this to work.
15196 @node OpenRISC 1000
15197 @subsection OpenRISC 1000
15198 @cindex OpenRISC 1000
15200 @cindex or1k boards
15201 See OR1k Architecture document (@uref{www.opencores.org}) for more information
15202 about platform and commands.
15206 @kindex target jtag
15207 @item target jtag jtag://@var{host}:@var{port}
15209 Connects to remote JTAG server.
15210 JTAG remote server can be either an or1ksim or JTAG server,
15211 connected via parallel port to the board.
15213 Example: @code{target jtag jtag://localhost:9999}
15216 @item or1ksim @var{command}
15217 If connected to @code{or1ksim} OpenRISC 1000 Architectural
15218 Simulator, proprietary commands can be executed.
15220 @kindex info or1k spr
15221 @item info or1k spr
15222 Displays spr groups.
15224 @item info or1k spr @var{group}
15225 @itemx info or1k spr @var{groupno}
15226 Displays register names in selected group.
15228 @item info or1k spr @var{group} @var{register}
15229 @itemx info or1k spr @var{register}
15230 @itemx info or1k spr @var{groupno} @var{registerno}
15231 @itemx info or1k spr @var{registerno}
15232 Shows information about specified spr register.
15235 @item spr @var{group} @var{register} @var{value}
15236 @itemx spr @var{register @var{value}}
15237 @itemx spr @var{groupno} @var{registerno @var{value}}
15238 @itemx spr @var{registerno @var{value}}
15239 Writes @var{value} to specified spr register.
15242 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
15243 It is very similar to @value{GDBN} trace, except it does not interfere with normal
15244 program execution and is thus much faster. Hardware breakpoints/watchpoint
15245 triggers can be set using:
15248 Load effective address/data
15250 Store effective address/data
15252 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
15257 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
15258 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
15260 @code{htrace} commands:
15261 @cindex OpenRISC 1000 htrace
15264 @item hwatch @var{conditional}
15265 Set hardware watchpoint on combination of Load/Store Effective Address(es)
15266 or Data. For example:
15268 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
15270 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
15274 Display information about current HW trace configuration.
15276 @item htrace trigger @var{conditional}
15277 Set starting criteria for HW trace.
15279 @item htrace qualifier @var{conditional}
15280 Set acquisition qualifier for HW trace.
15282 @item htrace stop @var{conditional}
15283 Set HW trace stopping criteria.
15285 @item htrace record [@var{data}]*
15286 Selects the data to be recorded, when qualifier is met and HW trace was
15289 @item htrace enable
15290 @itemx htrace disable
15291 Enables/disables the HW trace.
15293 @item htrace rewind [@var{filename}]
15294 Clears currently recorded trace data.
15296 If filename is specified, new trace file is made and any newly collected data
15297 will be written there.
15299 @item htrace print [@var{start} [@var{len}]]
15300 Prints trace buffer, using current record configuration.
15302 @item htrace mode continuous
15303 Set continuous trace mode.
15305 @item htrace mode suspend
15306 Set suspend trace mode.
15310 @node PowerPC Embedded
15311 @subsection PowerPC Embedded
15313 @value{GDBN} provides the following PowerPC-specific commands:
15316 @kindex set powerpc
15317 @item set powerpc soft-float
15318 @itemx show powerpc soft-float
15319 Force @value{GDBN} to use (or not use) a software floating point calling
15320 convention. By default, @value{GDBN} selects the calling convention based
15321 on the selected architecture and the provided executable file.
15323 @item set powerpc vector-abi
15324 @itemx show powerpc vector-abi
15325 Force @value{GDBN} to use the specified calling convention for vector
15326 arguments and return values. The valid options are @samp{auto};
15327 @samp{generic}, to avoid vector registers even if they are present;
15328 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
15329 registers. By default, @value{GDBN} selects the calling convention
15330 based on the selected architecture and the provided executable file.
15332 @kindex target dink32
15333 @item target dink32 @var{dev}
15334 DINK32 ROM monitor.
15336 @kindex target ppcbug
15337 @item target ppcbug @var{dev}
15338 @kindex target ppcbug1
15339 @item target ppcbug1 @var{dev}
15340 PPCBUG ROM monitor for PowerPC.
15343 @item target sds @var{dev}
15344 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
15347 @cindex SDS protocol
15348 The following commands specific to the SDS protocol are supported
15352 @item set sdstimeout @var{nsec}
15353 @kindex set sdstimeout
15354 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
15355 default is 2 seconds.
15357 @item show sdstimeout
15358 @kindex show sdstimeout
15359 Show the current value of the SDS timeout.
15361 @item sds @var{command}
15362 @kindex sds@r{, a command}
15363 Send the specified @var{command} string to the SDS monitor.
15368 @subsection HP PA Embedded
15372 @kindex target op50n
15373 @item target op50n @var{dev}
15374 OP50N monitor, running on an OKI HPPA board.
15376 @kindex target w89k
15377 @item target w89k @var{dev}
15378 W89K monitor, running on a Winbond HPPA board.
15383 @subsection Tsqware Sparclet
15387 @value{GDBN} enables developers to debug tasks running on
15388 Sparclet targets from a Unix host.
15389 @value{GDBN} uses code that runs on
15390 both the Unix host and on the Sparclet target. The program
15391 @code{@value{GDBP}} is installed and executed on the Unix host.
15394 @item remotetimeout @var{args}
15395 @kindex remotetimeout
15396 @value{GDBN} supports the option @code{remotetimeout}.
15397 This option is set by the user, and @var{args} represents the number of
15398 seconds @value{GDBN} waits for responses.
15401 @cindex compiling, on Sparclet
15402 When compiling for debugging, include the options @samp{-g} to get debug
15403 information and @samp{-Ttext} to relocate the program to where you wish to
15404 load it on the target. You may also want to add the options @samp{-n} or
15405 @samp{-N} in order to reduce the size of the sections. Example:
15408 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
15411 You can use @code{objdump} to verify that the addresses are what you intended:
15414 sparclet-aout-objdump --headers --syms prog
15417 @cindex running, on Sparclet
15419 your Unix execution search path to find @value{GDBN}, you are ready to
15420 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
15421 (or @code{sparclet-aout-gdb}, depending on your installation).
15423 @value{GDBN} comes up showing the prompt:
15430 * Sparclet File:: Setting the file to debug
15431 * Sparclet Connection:: Connecting to Sparclet
15432 * Sparclet Download:: Sparclet download
15433 * Sparclet Execution:: Running and debugging
15436 @node Sparclet File
15437 @subsubsection Setting File to Debug
15439 The @value{GDBN} command @code{file} lets you choose with program to debug.
15442 (gdbslet) file prog
15446 @value{GDBN} then attempts to read the symbol table of @file{prog}.
15447 @value{GDBN} locates
15448 the file by searching the directories listed in the command search
15450 If the file was compiled with debug information (option @samp{-g}), source
15451 files will be searched as well.
15452 @value{GDBN} locates
15453 the source files by searching the directories listed in the directory search
15454 path (@pxref{Environment, ,Your Program's Environment}).
15456 to find a file, it displays a message such as:
15459 prog: No such file or directory.
15462 When this happens, add the appropriate directories to the search paths with
15463 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
15464 @code{target} command again.
15466 @node Sparclet Connection
15467 @subsubsection Connecting to Sparclet
15469 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
15470 To connect to a target on serial port ``@code{ttya}'', type:
15473 (gdbslet) target sparclet /dev/ttya
15474 Remote target sparclet connected to /dev/ttya
15475 main () at ../prog.c:3
15479 @value{GDBN} displays messages like these:
15485 @node Sparclet Download
15486 @subsubsection Sparclet Download
15488 @cindex download to Sparclet
15489 Once connected to the Sparclet target,
15490 you can use the @value{GDBN}
15491 @code{load} command to download the file from the host to the target.
15492 The file name and load offset should be given as arguments to the @code{load}
15494 Since the file format is aout, the program must be loaded to the starting
15495 address. You can use @code{objdump} to find out what this value is. The load
15496 offset is an offset which is added to the VMA (virtual memory address)
15497 of each of the file's sections.
15498 For instance, if the program
15499 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
15500 and bss at 0x12010170, in @value{GDBN}, type:
15503 (gdbslet) load prog 0x12010000
15504 Loading section .text, size 0xdb0 vma 0x12010000
15507 If the code is loaded at a different address then what the program was linked
15508 to, you may need to use the @code{section} and @code{add-symbol-file} commands
15509 to tell @value{GDBN} where to map the symbol table.
15511 @node Sparclet Execution
15512 @subsubsection Running and Debugging
15514 @cindex running and debugging Sparclet programs
15515 You can now begin debugging the task using @value{GDBN}'s execution control
15516 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
15517 manual for the list of commands.
15521 Breakpoint 1 at 0x12010000: file prog.c, line 3.
15523 Starting program: prog
15524 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
15525 3 char *symarg = 0;
15527 4 char *execarg = "hello!";
15532 @subsection Fujitsu Sparclite
15536 @kindex target sparclite
15537 @item target sparclite @var{dev}
15538 Fujitsu sparclite boards, used only for the purpose of loading.
15539 You must use an additional command to debug the program.
15540 For example: target remote @var{dev} using @value{GDBN} standard
15546 @subsection Zilog Z8000
15549 @cindex simulator, Z8000
15550 @cindex Zilog Z8000 simulator
15552 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
15555 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
15556 unsegmented variant of the Z8000 architecture) or the Z8001 (the
15557 segmented variant). The simulator recognizes which architecture is
15558 appropriate by inspecting the object code.
15561 @item target sim @var{args}
15563 @kindex target sim@r{, with Z8000}
15564 Debug programs on a simulated CPU. If the simulator supports setup
15565 options, specify them via @var{args}.
15569 After specifying this target, you can debug programs for the simulated
15570 CPU in the same style as programs for your host computer; use the
15571 @code{file} command to load a new program image, the @code{run} command
15572 to run your program, and so on.
15574 As well as making available all the usual machine registers
15575 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
15576 additional items of information as specially named registers:
15581 Counts clock-ticks in the simulator.
15584 Counts instructions run in the simulator.
15587 Execution time in 60ths of a second.
15591 You can refer to these values in @value{GDBN} expressions with the usual
15592 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
15593 conditional breakpoint that suspends only after at least 5000
15594 simulated clock ticks.
15597 @subsection Atmel AVR
15600 When configured for debugging the Atmel AVR, @value{GDBN} supports the
15601 following AVR-specific commands:
15604 @item info io_registers
15605 @kindex info io_registers@r{, AVR}
15606 @cindex I/O registers (Atmel AVR)
15607 This command displays information about the AVR I/O registers. For
15608 each register, @value{GDBN} prints its number and value.
15615 When configured for debugging CRIS, @value{GDBN} provides the
15616 following CRIS-specific commands:
15619 @item set cris-version @var{ver}
15620 @cindex CRIS version
15621 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
15622 The CRIS version affects register names and sizes. This command is useful in
15623 case autodetection of the CRIS version fails.
15625 @item show cris-version
15626 Show the current CRIS version.
15628 @item set cris-dwarf2-cfi
15629 @cindex DWARF-2 CFI and CRIS
15630 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
15631 Change to @samp{off} when using @code{gcc-cris} whose version is below
15634 @item show cris-dwarf2-cfi
15635 Show the current state of using DWARF-2 CFI.
15637 @item set cris-mode @var{mode}
15639 Set the current CRIS mode to @var{mode}. It should only be changed when
15640 debugging in guru mode, in which case it should be set to
15641 @samp{guru} (the default is @samp{normal}).
15643 @item show cris-mode
15644 Show the current CRIS mode.
15648 @subsection Renesas Super-H
15651 For the Renesas Super-H processor, @value{GDBN} provides these
15656 @kindex regs@r{, Super-H}
15657 Show the values of all Super-H registers.
15659 @item set sh calling-convention @var{convention}
15660 @kindex set sh calling-convention
15661 Set the calling-convention used when calling functions from @value{GDBN}.
15662 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
15663 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
15664 convention. If the DWARF-2 information of the called function specifies
15665 that the function follows the Renesas calling convention, the function
15666 is called using the Renesas calling convention. If the calling convention
15667 is set to @samp{renesas}, the Renesas calling convention is always used,
15668 regardless of the DWARF-2 information. This can be used to override the
15669 default of @samp{gcc} if debug information is missing, or the compiler
15670 does not emit the DWARF-2 calling convention entry for a function.
15672 @item show sh calling-convention
15673 @kindex show sh calling-convention
15674 Show the current calling convention setting.
15679 @node Architectures
15680 @section Architectures
15682 This section describes characteristics of architectures that affect
15683 all uses of @value{GDBN} with the architecture, both native and cross.
15690 * HPPA:: HP PA architecture
15691 * SPU:: Cell Broadband Engine SPU architecture
15696 @subsection x86 Architecture-specific Issues
15699 @item set struct-convention @var{mode}
15700 @kindex set struct-convention
15701 @cindex struct return convention
15702 @cindex struct/union returned in registers
15703 Set the convention used by the inferior to return @code{struct}s and
15704 @code{union}s from functions to @var{mode}. Possible values of
15705 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
15706 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
15707 are returned on the stack, while @code{"reg"} means that a
15708 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
15709 be returned in a register.
15711 @item show struct-convention
15712 @kindex show struct-convention
15713 Show the current setting of the convention to return @code{struct}s
15722 @kindex set rstack_high_address
15723 @cindex AMD 29K register stack
15724 @cindex register stack, AMD29K
15725 @item set rstack_high_address @var{address}
15726 On AMD 29000 family processors, registers are saved in a separate
15727 @dfn{register stack}. There is no way for @value{GDBN} to determine the
15728 extent of this stack. Normally, @value{GDBN} just assumes that the
15729 stack is ``large enough''. This may result in @value{GDBN} referencing
15730 memory locations that do not exist. If necessary, you can get around
15731 this problem by specifying the ending address of the register stack with
15732 the @code{set rstack_high_address} command. The argument should be an
15733 address, which you probably want to precede with @samp{0x} to specify in
15736 @kindex show rstack_high_address
15737 @item show rstack_high_address
15738 Display the current limit of the register stack, on AMD 29000 family
15746 See the following section.
15751 @cindex stack on Alpha
15752 @cindex stack on MIPS
15753 @cindex Alpha stack
15755 Alpha- and MIPS-based computers use an unusual stack frame, which
15756 sometimes requires @value{GDBN} to search backward in the object code to
15757 find the beginning of a function.
15759 @cindex response time, MIPS debugging
15760 To improve response time (especially for embedded applications, where
15761 @value{GDBN} may be restricted to a slow serial line for this search)
15762 you may want to limit the size of this search, using one of these
15766 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
15767 @item set heuristic-fence-post @var{limit}
15768 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
15769 search for the beginning of a function. A value of @var{0} (the
15770 default) means there is no limit. However, except for @var{0}, the
15771 larger the limit the more bytes @code{heuristic-fence-post} must search
15772 and therefore the longer it takes to run. You should only need to use
15773 this command when debugging a stripped executable.
15775 @item show heuristic-fence-post
15776 Display the current limit.
15780 These commands are available @emph{only} when @value{GDBN} is configured
15781 for debugging programs on Alpha or MIPS processors.
15783 Several MIPS-specific commands are available when debugging MIPS
15787 @item set mips abi @var{arg}
15788 @kindex set mips abi
15789 @cindex set ABI for MIPS
15790 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
15791 values of @var{arg} are:
15795 The default ABI associated with the current binary (this is the
15806 @item show mips abi
15807 @kindex show mips abi
15808 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
15811 @itemx show mipsfpu
15812 @xref{MIPS Embedded, set mipsfpu}.
15814 @item set mips mask-address @var{arg}
15815 @kindex set mips mask-address
15816 @cindex MIPS addresses, masking
15817 This command determines whether the most-significant 32 bits of 64-bit
15818 MIPS addresses are masked off. The argument @var{arg} can be
15819 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
15820 setting, which lets @value{GDBN} determine the correct value.
15822 @item show mips mask-address
15823 @kindex show mips mask-address
15824 Show whether the upper 32 bits of MIPS addresses are masked off or
15827 @item set remote-mips64-transfers-32bit-regs
15828 @kindex set remote-mips64-transfers-32bit-regs
15829 This command controls compatibility with 64-bit MIPS targets that
15830 transfer data in 32-bit quantities. If you have an old MIPS 64 target
15831 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
15832 and 64 bits for other registers, set this option to @samp{on}.
15834 @item show remote-mips64-transfers-32bit-regs
15835 @kindex show remote-mips64-transfers-32bit-regs
15836 Show the current setting of compatibility with older MIPS 64 targets.
15838 @item set debug mips
15839 @kindex set debug mips
15840 This command turns on and off debugging messages for the MIPS-specific
15841 target code in @value{GDBN}.
15843 @item show debug mips
15844 @kindex show debug mips
15845 Show the current setting of MIPS debugging messages.
15851 @cindex HPPA support
15853 When @value{GDBN} is debugging the HP PA architecture, it provides the
15854 following special commands:
15857 @item set debug hppa
15858 @kindex set debug hppa
15859 This command determines whether HPPA architecture-specific debugging
15860 messages are to be displayed.
15862 @item show debug hppa
15863 Show whether HPPA debugging messages are displayed.
15865 @item maint print unwind @var{address}
15866 @kindex maint print unwind@r{, HPPA}
15867 This command displays the contents of the unwind table entry at the
15868 given @var{address}.
15874 @subsection Cell Broadband Engine SPU architecture
15875 @cindex Cell Broadband Engine
15878 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
15879 it provides the following special commands:
15882 @item info spu event
15884 Display SPU event facility status. Shows current event mask
15885 and pending event status.
15887 @item info spu signal
15888 Display SPU signal notification facility status. Shows pending
15889 signal-control word and signal notification mode of both signal
15890 notification channels.
15892 @item info spu mailbox
15893 Display SPU mailbox facility status. Shows all pending entries,
15894 in order of processing, in each of the SPU Write Outbound,
15895 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
15898 Display MFC DMA status. Shows all pending commands in the MFC
15899 DMA queue. For each entry, opcode, tag, class IDs, effective
15900 and local store addresses and transfer size are shown.
15902 @item info spu proxydma
15903 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
15904 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
15905 and local store addresses and transfer size are shown.
15910 @subsection PowerPC
15911 @cindex PowerPC architecture
15913 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
15914 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
15915 numbers stored in the floating point registers. These values must be stored
15916 in two consecutive registers, always starting at an even register like
15917 @code{f0} or @code{f2}.
15919 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
15920 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
15921 @code{f2} and @code{f3} for @code{$dl1} and so on.
15924 @node Controlling GDB
15925 @chapter Controlling @value{GDBN}
15927 You can alter the way @value{GDBN} interacts with you by using the
15928 @code{set} command. For commands controlling how @value{GDBN} displays
15929 data, see @ref{Print Settings, ,Print Settings}. Other settings are
15934 * Editing:: Command editing
15935 * Command History:: Command history
15936 * Screen Size:: Screen size
15937 * Numbers:: Numbers
15938 * ABI:: Configuring the current ABI
15939 * Messages/Warnings:: Optional warnings and messages
15940 * Debugging Output:: Optional messages about internal happenings
15948 @value{GDBN} indicates its readiness to read a command by printing a string
15949 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
15950 can change the prompt string with the @code{set prompt} command. For
15951 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
15952 the prompt in one of the @value{GDBN} sessions so that you can always tell
15953 which one you are talking to.
15955 @emph{Note:} @code{set prompt} does not add a space for you after the
15956 prompt you set. This allows you to set a prompt which ends in a space
15957 or a prompt that does not.
15961 @item set prompt @var{newprompt}
15962 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
15964 @kindex show prompt
15966 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
15970 @section Command Editing
15972 @cindex command line editing
15974 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
15975 @sc{gnu} library provides consistent behavior for programs which provide a
15976 command line interface to the user. Advantages are @sc{gnu} Emacs-style
15977 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
15978 substitution, and a storage and recall of command history across
15979 debugging sessions.
15981 You may control the behavior of command line editing in @value{GDBN} with the
15982 command @code{set}.
15985 @kindex set editing
15988 @itemx set editing on
15989 Enable command line editing (enabled by default).
15991 @item set editing off
15992 Disable command line editing.
15994 @kindex show editing
15996 Show whether command line editing is enabled.
15999 @xref{Command Line Editing}, for more details about the Readline
16000 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
16001 encouraged to read that chapter.
16003 @node Command History
16004 @section Command History
16005 @cindex command history
16007 @value{GDBN} can keep track of the commands you type during your
16008 debugging sessions, so that you can be certain of precisely what
16009 happened. Use these commands to manage the @value{GDBN} command
16012 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
16013 package, to provide the history facility. @xref{Using History
16014 Interactively}, for the detailed description of the History library.
16016 To issue a command to @value{GDBN} without affecting certain aspects of
16017 the state which is seen by users, prefix it with @samp{server }
16018 (@pxref{Server Prefix}). This
16019 means that this command will not affect the command history, nor will it
16020 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
16021 pressed on a line by itself.
16023 @cindex @code{server}, command prefix
16024 The server prefix does not affect the recording of values into the value
16025 history; to print a value without recording it into the value history,
16026 use the @code{output} command instead of the @code{print} command.
16028 Here is the description of @value{GDBN} commands related to command
16032 @cindex history substitution
16033 @cindex history file
16034 @kindex set history filename
16035 @cindex @env{GDBHISTFILE}, environment variable
16036 @item set history filename @var{fname}
16037 Set the name of the @value{GDBN} command history file to @var{fname}.
16038 This is the file where @value{GDBN} reads an initial command history
16039 list, and where it writes the command history from this session when it
16040 exits. You can access this list through history expansion or through
16041 the history command editing characters listed below. This file defaults
16042 to the value of the environment variable @code{GDBHISTFILE}, or to
16043 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
16046 @cindex save command history
16047 @kindex set history save
16048 @item set history save
16049 @itemx set history save on
16050 Record command history in a file, whose name may be specified with the
16051 @code{set history filename} command. By default, this option is disabled.
16053 @item set history save off
16054 Stop recording command history in a file.
16056 @cindex history size
16057 @kindex set history size
16058 @cindex @env{HISTSIZE}, environment variable
16059 @item set history size @var{size}
16060 Set the number of commands which @value{GDBN} keeps in its history list.
16061 This defaults to the value of the environment variable
16062 @code{HISTSIZE}, or to 256 if this variable is not set.
16065 History expansion assigns special meaning to the character @kbd{!}.
16066 @xref{Event Designators}, for more details.
16068 @cindex history expansion, turn on/off
16069 Since @kbd{!} is also the logical not operator in C, history expansion
16070 is off by default. If you decide to enable history expansion with the
16071 @code{set history expansion on} command, you may sometimes need to
16072 follow @kbd{!} (when it is used as logical not, in an expression) with
16073 a space or a tab to prevent it from being expanded. The readline
16074 history facilities do not attempt substitution on the strings
16075 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
16077 The commands to control history expansion are:
16080 @item set history expansion on
16081 @itemx set history expansion
16082 @kindex set history expansion
16083 Enable history expansion. History expansion is off by default.
16085 @item set history expansion off
16086 Disable history expansion.
16089 @kindex show history
16091 @itemx show history filename
16092 @itemx show history save
16093 @itemx show history size
16094 @itemx show history expansion
16095 These commands display the state of the @value{GDBN} history parameters.
16096 @code{show history} by itself displays all four states.
16101 @kindex show commands
16102 @cindex show last commands
16103 @cindex display command history
16104 @item show commands
16105 Display the last ten commands in the command history.
16107 @item show commands @var{n}
16108 Print ten commands centered on command number @var{n}.
16110 @item show commands +
16111 Print ten commands just after the commands last printed.
16115 @section Screen Size
16116 @cindex size of screen
16117 @cindex pauses in output
16119 Certain commands to @value{GDBN} may produce large amounts of
16120 information output to the screen. To help you read all of it,
16121 @value{GDBN} pauses and asks you for input at the end of each page of
16122 output. Type @key{RET} when you want to continue the output, or @kbd{q}
16123 to discard the remaining output. Also, the screen width setting
16124 determines when to wrap lines of output. Depending on what is being
16125 printed, @value{GDBN} tries to break the line at a readable place,
16126 rather than simply letting it overflow onto the following line.
16128 Normally @value{GDBN} knows the size of the screen from the terminal
16129 driver software. For example, on Unix @value{GDBN} uses the termcap data base
16130 together with the value of the @code{TERM} environment variable and the
16131 @code{stty rows} and @code{stty cols} settings. If this is not correct,
16132 you can override it with the @code{set height} and @code{set
16139 @kindex show height
16140 @item set height @var{lpp}
16142 @itemx set width @var{cpl}
16144 These @code{set} commands specify a screen height of @var{lpp} lines and
16145 a screen width of @var{cpl} characters. The associated @code{show}
16146 commands display the current settings.
16148 If you specify a height of zero lines, @value{GDBN} does not pause during
16149 output no matter how long the output is. This is useful if output is to a
16150 file or to an editor buffer.
16152 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
16153 from wrapping its output.
16155 @item set pagination on
16156 @itemx set pagination off
16157 @kindex set pagination
16158 Turn the output pagination on or off; the default is on. Turning
16159 pagination off is the alternative to @code{set height 0}.
16161 @item show pagination
16162 @kindex show pagination
16163 Show the current pagination mode.
16168 @cindex number representation
16169 @cindex entering numbers
16171 You can always enter numbers in octal, decimal, or hexadecimal in
16172 @value{GDBN} by the usual conventions: octal numbers begin with
16173 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
16174 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
16175 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
16176 10; likewise, the default display for numbers---when no particular
16177 format is specified---is base 10. You can change the default base for
16178 both input and output with the commands described below.
16181 @kindex set input-radix
16182 @item set input-radix @var{base}
16183 Set the default base for numeric input. Supported choices
16184 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
16185 specified either unambiguously or using the current input radix; for
16189 set input-radix 012
16190 set input-radix 10.
16191 set input-radix 0xa
16195 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
16196 leaves the input radix unchanged, no matter what it was, since
16197 @samp{10}, being without any leading or trailing signs of its base, is
16198 interpreted in the current radix. Thus, if the current radix is 16,
16199 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
16202 @kindex set output-radix
16203 @item set output-radix @var{base}
16204 Set the default base for numeric display. Supported choices
16205 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
16206 specified either unambiguously or using the current input radix.
16208 @kindex show input-radix
16209 @item show input-radix
16210 Display the current default base for numeric input.
16212 @kindex show output-radix
16213 @item show output-radix
16214 Display the current default base for numeric display.
16216 @item set radix @r{[}@var{base}@r{]}
16220 These commands set and show the default base for both input and output
16221 of numbers. @code{set radix} sets the radix of input and output to
16222 the same base; without an argument, it resets the radix back to its
16223 default value of 10.
16228 @section Configuring the Current ABI
16230 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
16231 application automatically. However, sometimes you need to override its
16232 conclusions. Use these commands to manage @value{GDBN}'s view of the
16239 One @value{GDBN} configuration can debug binaries for multiple operating
16240 system targets, either via remote debugging or native emulation.
16241 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
16242 but you can override its conclusion using the @code{set osabi} command.
16243 One example where this is useful is in debugging of binaries which use
16244 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
16245 not have the same identifying marks that the standard C library for your
16250 Show the OS ABI currently in use.
16253 With no argument, show the list of registered available OS ABI's.
16255 @item set osabi @var{abi}
16256 Set the current OS ABI to @var{abi}.
16259 @cindex float promotion
16261 Generally, the way that an argument of type @code{float} is passed to a
16262 function depends on whether the function is prototyped. For a prototyped
16263 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
16264 according to the architecture's convention for @code{float}. For unprototyped
16265 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
16266 @code{double} and then passed.
16268 Unfortunately, some forms of debug information do not reliably indicate whether
16269 a function is prototyped. If @value{GDBN} calls a function that is not marked
16270 as prototyped, it consults @kbd{set coerce-float-to-double}.
16273 @kindex set coerce-float-to-double
16274 @item set coerce-float-to-double
16275 @itemx set coerce-float-to-double on
16276 Arguments of type @code{float} will be promoted to @code{double} when passed
16277 to an unprototyped function. This is the default setting.
16279 @item set coerce-float-to-double off
16280 Arguments of type @code{float} will be passed directly to unprototyped
16283 @kindex show coerce-float-to-double
16284 @item show coerce-float-to-double
16285 Show the current setting of promoting @code{float} to @code{double}.
16289 @kindex show cp-abi
16290 @value{GDBN} needs to know the ABI used for your program's C@t{++}
16291 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
16292 used to build your application. @value{GDBN} only fully supports
16293 programs with a single C@t{++} ABI; if your program contains code using
16294 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
16295 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
16296 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
16297 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
16298 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
16299 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
16304 Show the C@t{++} ABI currently in use.
16307 With no argument, show the list of supported C@t{++} ABI's.
16309 @item set cp-abi @var{abi}
16310 @itemx set cp-abi auto
16311 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
16314 @node Messages/Warnings
16315 @section Optional Warnings and Messages
16317 @cindex verbose operation
16318 @cindex optional warnings
16319 By default, @value{GDBN} is silent about its inner workings. If you are
16320 running on a slow machine, you may want to use the @code{set verbose}
16321 command. This makes @value{GDBN} tell you when it does a lengthy
16322 internal operation, so you will not think it has crashed.
16324 Currently, the messages controlled by @code{set verbose} are those
16325 which announce that the symbol table for a source file is being read;
16326 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
16329 @kindex set verbose
16330 @item set verbose on
16331 Enables @value{GDBN} output of certain informational messages.
16333 @item set verbose off
16334 Disables @value{GDBN} output of certain informational messages.
16336 @kindex show verbose
16338 Displays whether @code{set verbose} is on or off.
16341 By default, if @value{GDBN} encounters bugs in the symbol table of an
16342 object file, it is silent; but if you are debugging a compiler, you may
16343 find this information useful (@pxref{Symbol Errors, ,Errors Reading
16348 @kindex set complaints
16349 @item set complaints @var{limit}
16350 Permits @value{GDBN} to output @var{limit} complaints about each type of
16351 unusual symbols before becoming silent about the problem. Set
16352 @var{limit} to zero to suppress all complaints; set it to a large number
16353 to prevent complaints from being suppressed.
16355 @kindex show complaints
16356 @item show complaints
16357 Displays how many symbol complaints @value{GDBN} is permitted to produce.
16361 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
16362 lot of stupid questions to confirm certain commands. For example, if
16363 you try to run a program which is already running:
16367 The program being debugged has been started already.
16368 Start it from the beginning? (y or n)
16371 If you are willing to unflinchingly face the consequences of your own
16372 commands, you can disable this ``feature'':
16376 @kindex set confirm
16378 @cindex confirmation
16379 @cindex stupid questions
16380 @item set confirm off
16381 Disables confirmation requests.
16383 @item set confirm on
16384 Enables confirmation requests (the default).
16386 @kindex show confirm
16388 Displays state of confirmation requests.
16392 @cindex command tracing
16393 If you need to debug user-defined commands or sourced files you may find it
16394 useful to enable @dfn{command tracing}. In this mode each command will be
16395 printed as it is executed, prefixed with one or more @samp{+} symbols, the
16396 quantity denoting the call depth of each command.
16399 @kindex set trace-commands
16400 @cindex command scripts, debugging
16401 @item set trace-commands on
16402 Enable command tracing.
16403 @item set trace-commands off
16404 Disable command tracing.
16405 @item show trace-commands
16406 Display the current state of command tracing.
16409 @node Debugging Output
16410 @section Optional Messages about Internal Happenings
16411 @cindex optional debugging messages
16413 @value{GDBN} has commands that enable optional debugging messages from
16414 various @value{GDBN} subsystems; normally these commands are of
16415 interest to @value{GDBN} maintainers, or when reporting a bug. This
16416 section documents those commands.
16419 @kindex set exec-done-display
16420 @item set exec-done-display
16421 Turns on or off the notification of asynchronous commands'
16422 completion. When on, @value{GDBN} will print a message when an
16423 asynchronous command finishes its execution. The default is off.
16424 @kindex show exec-done-display
16425 @item show exec-done-display
16426 Displays the current setting of asynchronous command completion
16429 @cindex gdbarch debugging info
16430 @cindex architecture debugging info
16431 @item set debug arch
16432 Turns on or off display of gdbarch debugging info. The default is off
16434 @item show debug arch
16435 Displays the current state of displaying gdbarch debugging info.
16436 @item set debug aix-thread
16437 @cindex AIX threads
16438 Display debugging messages about inner workings of the AIX thread
16440 @item show debug aix-thread
16441 Show the current state of AIX thread debugging info display.
16442 @item set debug event
16443 @cindex event debugging info
16444 Turns on or off display of @value{GDBN} event debugging info. The
16446 @item show debug event
16447 Displays the current state of displaying @value{GDBN} event debugging
16449 @item set debug expression
16450 @cindex expression debugging info
16451 Turns on or off display of debugging info about @value{GDBN}
16452 expression parsing. The default is off.
16453 @item show debug expression
16454 Displays the current state of displaying debugging info about
16455 @value{GDBN} expression parsing.
16456 @item set debug frame
16457 @cindex frame debugging info
16458 Turns on or off display of @value{GDBN} frame debugging info. The
16460 @item show debug frame
16461 Displays the current state of displaying @value{GDBN} frame debugging
16463 @item set debug infrun
16464 @cindex inferior debugging info
16465 Turns on or off display of @value{GDBN} debugging info for running the inferior.
16466 The default is off. @file{infrun.c} contains GDB's runtime state machine used
16467 for implementing operations such as single-stepping the inferior.
16468 @item show debug infrun
16469 Displays the current state of @value{GDBN} inferior debugging.
16470 @item set debug lin-lwp
16471 @cindex @sc{gnu}/Linux LWP debug messages
16472 @cindex Linux lightweight processes
16473 Turns on or off debugging messages from the Linux LWP debug support.
16474 @item show debug lin-lwp
16475 Show the current state of Linux LWP debugging messages.
16476 @item set debug lin-lwp-async
16477 @cindex @sc{gnu}/Linux LWP async debug messages
16478 @cindex Linux lightweight processes
16479 Turns on or off debugging messages from the Linux LWP async debug support.
16480 @item show debug lin-lwp-async
16481 Show the current state of Linux LWP async debugging messages.
16482 @item set debug observer
16483 @cindex observer debugging info
16484 Turns on or off display of @value{GDBN} observer debugging. This
16485 includes info such as the notification of observable events.
16486 @item show debug observer
16487 Displays the current state of observer debugging.
16488 @item set debug overload
16489 @cindex C@t{++} overload debugging info
16490 Turns on or off display of @value{GDBN} C@t{++} overload debugging
16491 info. This includes info such as ranking of functions, etc. The default
16493 @item show debug overload
16494 Displays the current state of displaying @value{GDBN} C@t{++} overload
16496 @cindex packets, reporting on stdout
16497 @cindex serial connections, debugging
16498 @cindex debug remote protocol
16499 @cindex remote protocol debugging
16500 @cindex display remote packets
16501 @item set debug remote
16502 Turns on or off display of reports on all packets sent back and forth across
16503 the serial line to the remote machine. The info is printed on the
16504 @value{GDBN} standard output stream. The default is off.
16505 @item show debug remote
16506 Displays the state of display of remote packets.
16507 @item set debug serial
16508 Turns on or off display of @value{GDBN} serial debugging info. The
16510 @item show debug serial
16511 Displays the current state of displaying @value{GDBN} serial debugging
16513 @item set debug solib-frv
16514 @cindex FR-V shared-library debugging
16515 Turns on or off debugging messages for FR-V shared-library code.
16516 @item show debug solib-frv
16517 Display the current state of FR-V shared-library code debugging
16519 @item set debug target
16520 @cindex target debugging info
16521 Turns on or off display of @value{GDBN} target debugging info. This info
16522 includes what is going on at the target level of GDB, as it happens. The
16523 default is 0. Set it to 1 to track events, and to 2 to also track the
16524 value of large memory transfers. Changes to this flag do not take effect
16525 until the next time you connect to a target or use the @code{run} command.
16526 @item show debug target
16527 Displays the current state of displaying @value{GDBN} target debugging
16529 @item set debug timestamp
16530 @cindex timestampping debugging info
16531 Turns on or off display of timestamps with @value{GDBN} debugging info.
16532 When enabled, seconds and microseconds are displayed before each debugging
16534 @item show debug timestamp
16535 Displays the current state of displaying timestamps with @value{GDBN}
16537 @item set debugvarobj
16538 @cindex variable object debugging info
16539 Turns on or off display of @value{GDBN} variable object debugging
16540 info. The default is off.
16541 @item show debugvarobj
16542 Displays the current state of displaying @value{GDBN} variable object
16544 @item set debug xml
16545 @cindex XML parser debugging
16546 Turns on or off debugging messages for built-in XML parsers.
16547 @item show debug xml
16548 Displays the current state of XML debugging messages.
16552 @chapter Canned Sequences of Commands
16554 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
16555 Command Lists}), @value{GDBN} provides two ways to store sequences of
16556 commands for execution as a unit: user-defined commands and command
16560 * Define:: How to define your own commands
16561 * Hooks:: Hooks for user-defined commands
16562 * Command Files:: How to write scripts of commands to be stored in a file
16563 * Output:: Commands for controlled output
16567 @section User-defined Commands
16569 @cindex user-defined command
16570 @cindex arguments, to user-defined commands
16571 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
16572 which you assign a new name as a command. This is done with the
16573 @code{define} command. User commands may accept up to 10 arguments
16574 separated by whitespace. Arguments are accessed within the user command
16575 via @code{$arg0@dots{}$arg9}. A trivial example:
16579 print $arg0 + $arg1 + $arg2
16584 To execute the command use:
16591 This defines the command @code{adder}, which prints the sum of
16592 its three arguments. Note the arguments are text substitutions, so they may
16593 reference variables, use complex expressions, or even perform inferior
16596 @cindex argument count in user-defined commands
16597 @cindex how many arguments (user-defined commands)
16598 In addition, @code{$argc} may be used to find out how many arguments have
16599 been passed. This expands to a number in the range 0@dots{}10.
16604 print $arg0 + $arg1
16607 print $arg0 + $arg1 + $arg2
16615 @item define @var{commandname}
16616 Define a command named @var{commandname}. If there is already a command
16617 by that name, you are asked to confirm that you want to redefine it.
16619 The definition of the command is made up of other @value{GDBN} command lines,
16620 which are given following the @code{define} command. The end of these
16621 commands is marked by a line containing @code{end}.
16624 @kindex end@r{ (user-defined commands)}
16625 @item document @var{commandname}
16626 Document the user-defined command @var{commandname}, so that it can be
16627 accessed by @code{help}. The command @var{commandname} must already be
16628 defined. This command reads lines of documentation just as @code{define}
16629 reads the lines of the command definition, ending with @code{end}.
16630 After the @code{document} command is finished, @code{help} on command
16631 @var{commandname} displays the documentation you have written.
16633 You may use the @code{document} command again to change the
16634 documentation of a command. Redefining the command with @code{define}
16635 does not change the documentation.
16637 @kindex dont-repeat
16638 @cindex don't repeat command
16640 Used inside a user-defined command, this tells @value{GDBN} that this
16641 command should not be repeated when the user hits @key{RET}
16642 (@pxref{Command Syntax, repeat last command}).
16644 @kindex help user-defined
16645 @item help user-defined
16646 List all user-defined commands, with the first line of the documentation
16651 @itemx show user @var{commandname}
16652 Display the @value{GDBN} commands used to define @var{commandname} (but
16653 not its documentation). If no @var{commandname} is given, display the
16654 definitions for all user-defined commands.
16656 @cindex infinite recursion in user-defined commands
16657 @kindex show max-user-call-depth
16658 @kindex set max-user-call-depth
16659 @item show max-user-call-depth
16660 @itemx set max-user-call-depth
16661 The value of @code{max-user-call-depth} controls how many recursion
16662 levels are allowed in user-defined commands before @value{GDBN} suspects an
16663 infinite recursion and aborts the command.
16666 In addition to the above commands, user-defined commands frequently
16667 use control flow commands, described in @ref{Command Files}.
16669 When user-defined commands are executed, the
16670 commands of the definition are not printed. An error in any command
16671 stops execution of the user-defined command.
16673 If used interactively, commands that would ask for confirmation proceed
16674 without asking when used inside a user-defined command. Many @value{GDBN}
16675 commands that normally print messages to say what they are doing omit the
16676 messages when used in a user-defined command.
16679 @section User-defined Command Hooks
16680 @cindex command hooks
16681 @cindex hooks, for commands
16682 @cindex hooks, pre-command
16685 You may define @dfn{hooks}, which are a special kind of user-defined
16686 command. Whenever you run the command @samp{foo}, if the user-defined
16687 command @samp{hook-foo} exists, it is executed (with no arguments)
16688 before that command.
16690 @cindex hooks, post-command
16692 A hook may also be defined which is run after the command you executed.
16693 Whenever you run the command @samp{foo}, if the user-defined command
16694 @samp{hookpost-foo} exists, it is executed (with no arguments) after
16695 that command. Post-execution hooks may exist simultaneously with
16696 pre-execution hooks, for the same command.
16698 It is valid for a hook to call the command which it hooks. If this
16699 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
16701 @c It would be nice if hookpost could be passed a parameter indicating
16702 @c if the command it hooks executed properly or not. FIXME!
16704 @kindex stop@r{, a pseudo-command}
16705 In addition, a pseudo-command, @samp{stop} exists. Defining
16706 (@samp{hook-stop}) makes the associated commands execute every time
16707 execution stops in your program: before breakpoint commands are run,
16708 displays are printed, or the stack frame is printed.
16710 For example, to ignore @code{SIGALRM} signals while
16711 single-stepping, but treat them normally during normal execution,
16716 handle SIGALRM nopass
16720 handle SIGALRM pass
16723 define hook-continue
16724 handle SIGALRM pass
16728 As a further example, to hook at the beginning and end of the @code{echo}
16729 command, and to add extra text to the beginning and end of the message,
16737 define hookpost-echo
16741 (@value{GDBP}) echo Hello World
16742 <<<---Hello World--->>>
16747 You can define a hook for any single-word command in @value{GDBN}, but
16748 not for command aliases; you should define a hook for the basic command
16749 name, e.g.@: @code{backtrace} rather than @code{bt}.
16750 @c FIXME! So how does Joe User discover whether a command is an alias
16752 If an error occurs during the execution of your hook, execution of
16753 @value{GDBN} commands stops and @value{GDBN} issues a prompt
16754 (before the command that you actually typed had a chance to run).
16756 If you try to define a hook which does not match any known command, you
16757 get a warning from the @code{define} command.
16759 @node Command Files
16760 @section Command Files
16762 @cindex command files
16763 @cindex scripting commands
16764 A command file for @value{GDBN} is a text file made of lines that are
16765 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
16766 also be included. An empty line in a command file does nothing; it
16767 does not mean to repeat the last command, as it would from the
16770 You can request the execution of a command file with the @code{source}
16775 @cindex execute commands from a file
16776 @item source [@code{-v}] @var{filename}
16777 Execute the command file @var{filename}.
16780 The lines in a command file are generally executed sequentially,
16781 unless the order of execution is changed by one of the
16782 @emph{flow-control commands} described below. The commands are not
16783 printed as they are executed. An error in any command terminates
16784 execution of the command file and control is returned to the console.
16786 @value{GDBN} searches for @var{filename} in the current directory and then
16787 on the search path (specified with the @samp{directory} command).
16789 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
16790 each command as it is executed. The option must be given before
16791 @var{filename}, and is interpreted as part of the filename anywhere else.
16793 Commands that would ask for confirmation if used interactively proceed
16794 without asking when used in a command file. Many @value{GDBN} commands that
16795 normally print messages to say what they are doing omit the messages
16796 when called from command files.
16798 @value{GDBN} also accepts command input from standard input. In this
16799 mode, normal output goes to standard output and error output goes to
16800 standard error. Errors in a command file supplied on standard input do
16801 not terminate execution of the command file---execution continues with
16805 gdb < cmds > log 2>&1
16808 (The syntax above will vary depending on the shell used.) This example
16809 will execute commands from the file @file{cmds}. All output and errors
16810 would be directed to @file{log}.
16812 Since commands stored on command files tend to be more general than
16813 commands typed interactively, they frequently need to deal with
16814 complicated situations, such as different or unexpected values of
16815 variables and symbols, changes in how the program being debugged is
16816 built, etc. @value{GDBN} provides a set of flow-control commands to
16817 deal with these complexities. Using these commands, you can write
16818 complex scripts that loop over data structures, execute commands
16819 conditionally, etc.
16826 This command allows to include in your script conditionally executed
16827 commands. The @code{if} command takes a single argument, which is an
16828 expression to evaluate. It is followed by a series of commands that
16829 are executed only if the expression is true (its value is nonzero).
16830 There can then optionally be an @code{else} line, followed by a series
16831 of commands that are only executed if the expression was false. The
16832 end of the list is marked by a line containing @code{end}.
16836 This command allows to write loops. Its syntax is similar to
16837 @code{if}: the command takes a single argument, which is an expression
16838 to evaluate, and must be followed by the commands to execute, one per
16839 line, terminated by an @code{end}. These commands are called the
16840 @dfn{body} of the loop. The commands in the body of @code{while} are
16841 executed repeatedly as long as the expression evaluates to true.
16845 This command exits the @code{while} loop in whose body it is included.
16846 Execution of the script continues after that @code{while}s @code{end}
16849 @kindex loop_continue
16850 @item loop_continue
16851 This command skips the execution of the rest of the body of commands
16852 in the @code{while} loop in whose body it is included. Execution
16853 branches to the beginning of the @code{while} loop, where it evaluates
16854 the controlling expression.
16856 @kindex end@r{ (if/else/while commands)}
16858 Terminate the block of commands that are the body of @code{if},
16859 @code{else}, or @code{while} flow-control commands.
16864 @section Commands for Controlled Output
16866 During the execution of a command file or a user-defined command, normal
16867 @value{GDBN} output is suppressed; the only output that appears is what is
16868 explicitly printed by the commands in the definition. This section
16869 describes three commands useful for generating exactly the output you
16874 @item echo @var{text}
16875 @c I do not consider backslash-space a standard C escape sequence
16876 @c because it is not in ANSI.
16877 Print @var{text}. Nonprinting characters can be included in
16878 @var{text} using C escape sequences, such as @samp{\n} to print a
16879 newline. @strong{No newline is printed unless you specify one.}
16880 In addition to the standard C escape sequences, a backslash followed
16881 by a space stands for a space. This is useful for displaying a
16882 string with spaces at the beginning or the end, since leading and
16883 trailing spaces are otherwise trimmed from all arguments.
16884 To print @samp{@w{ }and foo =@w{ }}, use the command
16885 @samp{echo \@w{ }and foo = \@w{ }}.
16887 A backslash at the end of @var{text} can be used, as in C, to continue
16888 the command onto subsequent lines. For example,
16891 echo This is some text\n\
16892 which is continued\n\
16893 onto several lines.\n
16896 produces the same output as
16899 echo This is some text\n
16900 echo which is continued\n
16901 echo onto several lines.\n
16905 @item output @var{expression}
16906 Print the value of @var{expression} and nothing but that value: no
16907 newlines, no @samp{$@var{nn} = }. The value is not entered in the
16908 value history either. @xref{Expressions, ,Expressions}, for more information
16911 @item output/@var{fmt} @var{expression}
16912 Print the value of @var{expression} in format @var{fmt}. You can use
16913 the same formats as for @code{print}. @xref{Output Formats,,Output
16914 Formats}, for more information.
16917 @item printf @var{template}, @var{expressions}@dots{}
16918 Print the values of one or more @var{expressions} under the control of
16919 the string @var{template}. To print several values, make
16920 @var{expressions} be a comma-separated list of individual expressions,
16921 which may be either numbers or pointers. Their values are printed as
16922 specified by @var{template}, exactly as a C program would do by
16923 executing the code below:
16926 printf (@var{template}, @var{expressions}@dots{});
16929 As in @code{C} @code{printf}, ordinary characters in @var{template}
16930 are printed verbatim, while @dfn{conversion specification} introduced
16931 by the @samp{%} character cause subsequent @var{expressions} to be
16932 evaluated, their values converted and formatted according to type and
16933 style information encoded in the conversion specifications, and then
16936 For example, you can print two values in hex like this:
16939 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
16942 @code{printf} supports all the standard @code{C} conversion
16943 specifications, including the flags and modifiers between the @samp{%}
16944 character and the conversion letter, with the following exceptions:
16948 The argument-ordering modifiers, such as @samp{2$}, are not supported.
16951 The modifier @samp{*} is not supported for specifying precision or
16955 The @samp{'} flag (for separation of digits into groups according to
16956 @code{LC_NUMERIC'}) is not supported.
16959 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
16963 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
16966 The conversion letters @samp{a} and @samp{A} are not supported.
16970 Note that the @samp{ll} type modifier is supported only if the
16971 underlying @code{C} implementation used to build @value{GDBN} supports
16972 the @code{long long int} type, and the @samp{L} type modifier is
16973 supported only if @code{long double} type is available.
16975 As in @code{C}, @code{printf} supports simple backslash-escape
16976 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
16977 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
16978 single character. Octal and hexadecimal escape sequences are not
16981 Additionally, @code{printf} supports conversion specifications for DFP
16982 (@dfn{Decimal Floating Point}) types using the following length modifiers
16983 together with a floating point specifier.
16988 @samp{H} for printing @code{Decimal32} types.
16991 @samp{D} for printing @code{Decimal64} types.
16994 @samp{DD} for printing @code{Decimal128} types.
16997 If the underlying @code{C} implementation used to build @value{GDBN} has
16998 support for the three length modifiers for DFP types, other modifiers
16999 such as width and precision will also be available for @value{GDBN} to use.
17001 In case there is no such @code{C} support, no additional modifiers will be
17002 available and the value will be printed in the standard way.
17004 Here's an example of printing DFP types using the above conversion letters:
17006 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
17012 @chapter Command Interpreters
17013 @cindex command interpreters
17015 @value{GDBN} supports multiple command interpreters, and some command
17016 infrastructure to allow users or user interface writers to switch
17017 between interpreters or run commands in other interpreters.
17019 @value{GDBN} currently supports two command interpreters, the console
17020 interpreter (sometimes called the command-line interpreter or @sc{cli})
17021 and the machine interface interpreter (or @sc{gdb/mi}). This manual
17022 describes both of these interfaces in great detail.
17024 By default, @value{GDBN} will start with the console interpreter.
17025 However, the user may choose to start @value{GDBN} with another
17026 interpreter by specifying the @option{-i} or @option{--interpreter}
17027 startup options. Defined interpreters include:
17031 @cindex console interpreter
17032 The traditional console or command-line interpreter. This is the most often
17033 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
17034 @value{GDBN} will use this interpreter.
17037 @cindex mi interpreter
17038 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
17039 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
17040 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
17044 @cindex mi2 interpreter
17045 The current @sc{gdb/mi} interface.
17048 @cindex mi1 interpreter
17049 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
17053 @cindex invoke another interpreter
17054 The interpreter being used by @value{GDBN} may not be dynamically
17055 switched at runtime. Although possible, this could lead to a very
17056 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
17057 enters the command "interpreter-set console" in a console view,
17058 @value{GDBN} would switch to using the console interpreter, rendering
17059 the IDE inoperable!
17061 @kindex interpreter-exec
17062 Although you may only choose a single interpreter at startup, you may execute
17063 commands in any interpreter from the current interpreter using the appropriate
17064 command. If you are running the console interpreter, simply use the
17065 @code{interpreter-exec} command:
17068 interpreter-exec mi "-data-list-register-names"
17071 @sc{gdb/mi} has a similar command, although it is only available in versions of
17072 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
17075 @chapter @value{GDBN} Text User Interface
17077 @cindex Text User Interface
17080 * TUI Overview:: TUI overview
17081 * TUI Keys:: TUI key bindings
17082 * TUI Single Key Mode:: TUI single key mode
17083 * TUI Commands:: TUI-specific commands
17084 * TUI Configuration:: TUI configuration variables
17087 The @value{GDBN} Text User Interface (TUI) is a terminal
17088 interface which uses the @code{curses} library to show the source
17089 file, the assembly output, the program registers and @value{GDBN}
17090 commands in separate text windows. The TUI mode is supported only
17091 on platforms where a suitable version of the @code{curses} library
17094 @pindex @value{GDBTUI}
17095 The TUI mode is enabled by default when you invoke @value{GDBN} as
17096 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
17097 You can also switch in and out of TUI mode while @value{GDBN} runs by
17098 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
17099 @xref{TUI Keys, ,TUI Key Bindings}.
17102 @section TUI Overview
17104 In TUI mode, @value{GDBN} can display several text windows:
17108 This window is the @value{GDBN} command window with the @value{GDBN}
17109 prompt and the @value{GDBN} output. The @value{GDBN} input is still
17110 managed using readline.
17113 The source window shows the source file of the program. The current
17114 line and active breakpoints are displayed in this window.
17117 The assembly window shows the disassembly output of the program.
17120 This window shows the processor registers. Registers are highlighted
17121 when their values change.
17124 The source and assembly windows show the current program position
17125 by highlighting the current line and marking it with a @samp{>} marker.
17126 Breakpoints are indicated with two markers. The first marker
17127 indicates the breakpoint type:
17131 Breakpoint which was hit at least once.
17134 Breakpoint which was never hit.
17137 Hardware breakpoint which was hit at least once.
17140 Hardware breakpoint which was never hit.
17143 The second marker indicates whether the breakpoint is enabled or not:
17147 Breakpoint is enabled.
17150 Breakpoint is disabled.
17153 The source, assembly and register windows are updated when the current
17154 thread changes, when the frame changes, or when the program counter
17157 These windows are not all visible at the same time. The command
17158 window is always visible. The others can be arranged in several
17169 source and assembly,
17172 source and registers, or
17175 assembly and registers.
17178 A status line above the command window shows the following information:
17182 Indicates the current @value{GDBN} target.
17183 (@pxref{Targets, ,Specifying a Debugging Target}).
17186 Gives the current process or thread number.
17187 When no process is being debugged, this field is set to @code{No process}.
17190 Gives the current function name for the selected frame.
17191 The name is demangled if demangling is turned on (@pxref{Print Settings}).
17192 When there is no symbol corresponding to the current program counter,
17193 the string @code{??} is displayed.
17196 Indicates the current line number for the selected frame.
17197 When the current line number is not known, the string @code{??} is displayed.
17200 Indicates the current program counter address.
17204 @section TUI Key Bindings
17205 @cindex TUI key bindings
17207 The TUI installs several key bindings in the readline keymaps
17208 (@pxref{Command Line Editing}). The following key bindings
17209 are installed for both TUI mode and the @value{GDBN} standard mode.
17218 Enter or leave the TUI mode. When leaving the TUI mode,
17219 the curses window management stops and @value{GDBN} operates using
17220 its standard mode, writing on the terminal directly. When reentering
17221 the TUI mode, control is given back to the curses windows.
17222 The screen is then refreshed.
17226 Use a TUI layout with only one window. The layout will
17227 either be @samp{source} or @samp{assembly}. When the TUI mode
17228 is not active, it will switch to the TUI mode.
17230 Think of this key binding as the Emacs @kbd{C-x 1} binding.
17234 Use a TUI layout with at least two windows. When the current
17235 layout already has two windows, the next layout with two windows is used.
17236 When a new layout is chosen, one window will always be common to the
17237 previous layout and the new one.
17239 Think of it as the Emacs @kbd{C-x 2} binding.
17243 Change the active window. The TUI associates several key bindings
17244 (like scrolling and arrow keys) with the active window. This command
17245 gives the focus to the next TUI window.
17247 Think of it as the Emacs @kbd{C-x o} binding.
17251 Switch in and out of the TUI SingleKey mode that binds single
17252 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
17255 The following key bindings only work in the TUI mode:
17260 Scroll the active window one page up.
17264 Scroll the active window one page down.
17268 Scroll the active window one line up.
17272 Scroll the active window one line down.
17276 Scroll the active window one column left.
17280 Scroll the active window one column right.
17284 Refresh the screen.
17287 Because the arrow keys scroll the active window in the TUI mode, they
17288 are not available for their normal use by readline unless the command
17289 window has the focus. When another window is active, you must use
17290 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
17291 and @kbd{C-f} to control the command window.
17293 @node TUI Single Key Mode
17294 @section TUI Single Key Mode
17295 @cindex TUI single key mode
17297 The TUI also provides a @dfn{SingleKey} mode, which binds several
17298 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
17299 switch into this mode, where the following key bindings are used:
17302 @kindex c @r{(SingleKey TUI key)}
17306 @kindex d @r{(SingleKey TUI key)}
17310 @kindex f @r{(SingleKey TUI key)}
17314 @kindex n @r{(SingleKey TUI key)}
17318 @kindex q @r{(SingleKey TUI key)}
17320 exit the SingleKey mode.
17322 @kindex r @r{(SingleKey TUI key)}
17326 @kindex s @r{(SingleKey TUI key)}
17330 @kindex u @r{(SingleKey TUI key)}
17334 @kindex v @r{(SingleKey TUI key)}
17338 @kindex w @r{(SingleKey TUI key)}
17343 Other keys temporarily switch to the @value{GDBN} command prompt.
17344 The key that was pressed is inserted in the editing buffer so that
17345 it is possible to type most @value{GDBN} commands without interaction
17346 with the TUI SingleKey mode. Once the command is entered the TUI
17347 SingleKey mode is restored. The only way to permanently leave
17348 this mode is by typing @kbd{q} or @kbd{C-x s}.
17352 @section TUI-specific Commands
17353 @cindex TUI commands
17355 The TUI has specific commands to control the text windows.
17356 These commands are always available, even when @value{GDBN} is not in
17357 the TUI mode. When @value{GDBN} is in the standard mode, most
17358 of these commands will automatically switch to the TUI mode.
17363 List and give the size of all displayed windows.
17367 Display the next layout.
17370 Display the previous layout.
17373 Display the source window only.
17376 Display the assembly window only.
17379 Display the source and assembly window.
17382 Display the register window together with the source or assembly window.
17386 Make the next window active for scrolling.
17389 Make the previous window active for scrolling.
17392 Make the source window active for scrolling.
17395 Make the assembly window active for scrolling.
17398 Make the register window active for scrolling.
17401 Make the command window active for scrolling.
17405 Refresh the screen. This is similar to typing @kbd{C-L}.
17407 @item tui reg float
17409 Show the floating point registers in the register window.
17411 @item tui reg general
17412 Show the general registers in the register window.
17415 Show the next register group. The list of register groups as well as
17416 their order is target specific. The predefined register groups are the
17417 following: @code{general}, @code{float}, @code{system}, @code{vector},
17418 @code{all}, @code{save}, @code{restore}.
17420 @item tui reg system
17421 Show the system registers in the register window.
17425 Update the source window and the current execution point.
17427 @item winheight @var{name} +@var{count}
17428 @itemx winheight @var{name} -@var{count}
17430 Change the height of the window @var{name} by @var{count}
17431 lines. Positive counts increase the height, while negative counts
17434 @item tabset @var{nchars}
17436 Set the width of tab stops to be @var{nchars} characters.
17439 @node TUI Configuration
17440 @section TUI Configuration Variables
17441 @cindex TUI configuration variables
17443 Several configuration variables control the appearance of TUI windows.
17446 @item set tui border-kind @var{kind}
17447 @kindex set tui border-kind
17448 Select the border appearance for the source, assembly and register windows.
17449 The possible values are the following:
17452 Use a space character to draw the border.
17455 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
17458 Use the Alternate Character Set to draw the border. The border is
17459 drawn using character line graphics if the terminal supports them.
17462 @item set tui border-mode @var{mode}
17463 @kindex set tui border-mode
17464 @itemx set tui active-border-mode @var{mode}
17465 @kindex set tui active-border-mode
17466 Select the display attributes for the borders of the inactive windows
17467 or the active window. The @var{mode} can be one of the following:
17470 Use normal attributes to display the border.
17476 Use reverse video mode.
17479 Use half bright mode.
17481 @item half-standout
17482 Use half bright and standout mode.
17485 Use extra bright or bold mode.
17487 @item bold-standout
17488 Use extra bright or bold and standout mode.
17493 @chapter Using @value{GDBN} under @sc{gnu} Emacs
17496 @cindex @sc{gnu} Emacs
17497 A special interface allows you to use @sc{gnu} Emacs to view (and
17498 edit) the source files for the program you are debugging with
17501 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
17502 executable file you want to debug as an argument. This command starts
17503 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
17504 created Emacs buffer.
17505 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
17507 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
17512 All ``terminal'' input and output goes through an Emacs buffer, called
17515 This applies both to @value{GDBN} commands and their output, and to the input
17516 and output done by the program you are debugging.
17518 This is useful because it means that you can copy the text of previous
17519 commands and input them again; you can even use parts of the output
17522 All the facilities of Emacs' Shell mode are available for interacting
17523 with your program. In particular, you can send signals the usual
17524 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
17528 @value{GDBN} displays source code through Emacs.
17530 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
17531 source file for that frame and puts an arrow (@samp{=>}) at the
17532 left margin of the current line. Emacs uses a separate buffer for
17533 source display, and splits the screen to show both your @value{GDBN} session
17536 Explicit @value{GDBN} @code{list} or search commands still produce output as
17537 usual, but you probably have no reason to use them from Emacs.
17540 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
17541 a graphical mode, enabled by default, which provides further buffers
17542 that can control the execution and describe the state of your program.
17543 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
17545 If you specify an absolute file name when prompted for the @kbd{M-x
17546 gdb} argument, then Emacs sets your current working directory to where
17547 your program resides. If you only specify the file name, then Emacs
17548 sets your current working directory to to the directory associated
17549 with the previous buffer. In this case, @value{GDBN} may find your
17550 program by searching your environment's @code{PATH} variable, but on
17551 some operating systems it might not find the source. So, although the
17552 @value{GDBN} input and output session proceeds normally, the auxiliary
17553 buffer does not display the current source and line of execution.
17555 The initial working directory of @value{GDBN} is printed on the top
17556 line of the GUD buffer and this serves as a default for the commands
17557 that specify files for @value{GDBN} to operate on. @xref{Files,
17558 ,Commands to Specify Files}.
17560 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
17561 need to call @value{GDBN} by a different name (for example, if you
17562 keep several configurations around, with different names) you can
17563 customize the Emacs variable @code{gud-gdb-command-name} to run the
17566 In the GUD buffer, you can use these special Emacs commands in
17567 addition to the standard Shell mode commands:
17571 Describe the features of Emacs' GUD Mode.
17574 Execute to another source line, like the @value{GDBN} @code{step} command; also
17575 update the display window to show the current file and location.
17578 Execute to next source line in this function, skipping all function
17579 calls, like the @value{GDBN} @code{next} command. Then update the display window
17580 to show the current file and location.
17583 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
17584 display window accordingly.
17587 Execute until exit from the selected stack frame, like the @value{GDBN}
17588 @code{finish} command.
17591 Continue execution of your program, like the @value{GDBN} @code{continue}
17595 Go up the number of frames indicated by the numeric argument
17596 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
17597 like the @value{GDBN} @code{up} command.
17600 Go down the number of frames indicated by the numeric argument, like the
17601 @value{GDBN} @code{down} command.
17604 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
17605 tells @value{GDBN} to set a breakpoint on the source line point is on.
17607 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
17608 separate frame which shows a backtrace when the GUD buffer is current.
17609 Move point to any frame in the stack and type @key{RET} to make it
17610 become the current frame and display the associated source in the
17611 source buffer. Alternatively, click @kbd{Mouse-2} to make the
17612 selected frame become the current one. In graphical mode, the
17613 speedbar displays watch expressions.
17615 If you accidentally delete the source-display buffer, an easy way to get
17616 it back is to type the command @code{f} in the @value{GDBN} buffer, to
17617 request a frame display; when you run under Emacs, this recreates
17618 the source buffer if necessary to show you the context of the current
17621 The source files displayed in Emacs are in ordinary Emacs buffers
17622 which are visiting the source files in the usual way. You can edit
17623 the files with these buffers if you wish; but keep in mind that @value{GDBN}
17624 communicates with Emacs in terms of line numbers. If you add or
17625 delete lines from the text, the line numbers that @value{GDBN} knows cease
17626 to correspond properly with the code.
17628 A more detailed description of Emacs' interaction with @value{GDBN} is
17629 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
17632 @c The following dropped because Epoch is nonstandard. Reactivate
17633 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
17635 @kindex Emacs Epoch environment
17639 Version 18 of @sc{gnu} Emacs has a built-in window system
17640 called the @code{epoch}
17641 environment. Users of this environment can use a new command,
17642 @code{inspect} which performs identically to @code{print} except that
17643 each value is printed in its own window.
17648 @chapter The @sc{gdb/mi} Interface
17650 @unnumberedsec Function and Purpose
17652 @cindex @sc{gdb/mi}, its purpose
17653 @sc{gdb/mi} is a line based machine oriented text interface to
17654 @value{GDBN} and is activated by specifying using the
17655 @option{--interpreter} command line option (@pxref{Mode Options}). It
17656 is specifically intended to support the development of systems which
17657 use the debugger as just one small component of a larger system.
17659 This chapter is a specification of the @sc{gdb/mi} interface. It is written
17660 in the form of a reference manual.
17662 Note that @sc{gdb/mi} is still under construction, so some of the
17663 features described below are incomplete and subject to change
17664 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
17666 @unnumberedsec Notation and Terminology
17668 @cindex notational conventions, for @sc{gdb/mi}
17669 This chapter uses the following notation:
17673 @code{|} separates two alternatives.
17676 @code{[ @var{something} ]} indicates that @var{something} is optional:
17677 it may or may not be given.
17680 @code{( @var{group} )*} means that @var{group} inside the parentheses
17681 may repeat zero or more times.
17684 @code{( @var{group} )+} means that @var{group} inside the parentheses
17685 may repeat one or more times.
17688 @code{"@var{string}"} means a literal @var{string}.
17692 @heading Dependencies
17696 * GDB/MI Command Syntax::
17697 * GDB/MI Compatibility with CLI::
17698 * GDB/MI Development and Front Ends::
17699 * GDB/MI Output Records::
17700 * GDB/MI Simple Examples::
17701 * GDB/MI Command Description Format::
17702 * GDB/MI Breakpoint Commands::
17703 * GDB/MI Program Context::
17704 * GDB/MI Thread Commands::
17705 * GDB/MI Program Execution::
17706 * GDB/MI Stack Manipulation::
17707 * GDB/MI Variable Objects::
17708 * GDB/MI Data Manipulation::
17709 * GDB/MI Tracepoint Commands::
17710 * GDB/MI Symbol Query::
17711 * GDB/MI File Commands::
17713 * GDB/MI Kod Commands::
17714 * GDB/MI Memory Overlay Commands::
17715 * GDB/MI Signal Handling Commands::
17717 * GDB/MI Target Manipulation::
17718 * GDB/MI File Transfer Commands::
17719 * GDB/MI Miscellaneous Commands::
17722 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
17723 @node GDB/MI Command Syntax
17724 @section @sc{gdb/mi} Command Syntax
17727 * GDB/MI Input Syntax::
17728 * GDB/MI Output Syntax::
17731 @node GDB/MI Input Syntax
17732 @subsection @sc{gdb/mi} Input Syntax
17734 @cindex input syntax for @sc{gdb/mi}
17735 @cindex @sc{gdb/mi}, input syntax
17737 @item @var{command} @expansion{}
17738 @code{@var{cli-command} | @var{mi-command}}
17740 @item @var{cli-command} @expansion{}
17741 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
17742 @var{cli-command} is any existing @value{GDBN} CLI command.
17744 @item @var{mi-command} @expansion{}
17745 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
17746 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
17748 @item @var{token} @expansion{}
17749 "any sequence of digits"
17751 @item @var{option} @expansion{}
17752 @code{"-" @var{parameter} [ " " @var{parameter} ]}
17754 @item @var{parameter} @expansion{}
17755 @code{@var{non-blank-sequence} | @var{c-string}}
17757 @item @var{operation} @expansion{}
17758 @emph{any of the operations described in this chapter}
17760 @item @var{non-blank-sequence} @expansion{}
17761 @emph{anything, provided it doesn't contain special characters such as
17762 "-", @var{nl}, """ and of course " "}
17764 @item @var{c-string} @expansion{}
17765 @code{""" @var{seven-bit-iso-c-string-content} """}
17767 @item @var{nl} @expansion{}
17776 The CLI commands are still handled by the @sc{mi} interpreter; their
17777 output is described below.
17780 The @code{@var{token}}, when present, is passed back when the command
17784 Some @sc{mi} commands accept optional arguments as part of the parameter
17785 list. Each option is identified by a leading @samp{-} (dash) and may be
17786 followed by an optional argument parameter. Options occur first in the
17787 parameter list and can be delimited from normal parameters using
17788 @samp{--} (this is useful when some parameters begin with a dash).
17795 We want easy access to the existing CLI syntax (for debugging).
17798 We want it to be easy to spot a @sc{mi} operation.
17801 @node GDB/MI Output Syntax
17802 @subsection @sc{gdb/mi} Output Syntax
17804 @cindex output syntax of @sc{gdb/mi}
17805 @cindex @sc{gdb/mi}, output syntax
17806 The output from @sc{gdb/mi} consists of zero or more out-of-band records
17807 followed, optionally, by a single result record. This result record
17808 is for the most recent command. The sequence of output records is
17809 terminated by @samp{(gdb)}.
17811 If an input command was prefixed with a @code{@var{token}} then the
17812 corresponding output for that command will also be prefixed by that same
17816 @item @var{output} @expansion{}
17817 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
17819 @item @var{result-record} @expansion{}
17820 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
17822 @item @var{out-of-band-record} @expansion{}
17823 @code{@var{async-record} | @var{stream-record}}
17825 @item @var{async-record} @expansion{}
17826 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
17828 @item @var{exec-async-output} @expansion{}
17829 @code{[ @var{token} ] "*" @var{async-output}}
17831 @item @var{status-async-output} @expansion{}
17832 @code{[ @var{token} ] "+" @var{async-output}}
17834 @item @var{notify-async-output} @expansion{}
17835 @code{[ @var{token} ] "=" @var{async-output}}
17837 @item @var{async-output} @expansion{}
17838 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
17840 @item @var{result-class} @expansion{}
17841 @code{"done" | "running" | "connected" | "error" | "exit"}
17843 @item @var{async-class} @expansion{}
17844 @code{"stopped" | @var{others}} (where @var{others} will be added
17845 depending on the needs---this is still in development).
17847 @item @var{result} @expansion{}
17848 @code{ @var{variable} "=" @var{value}}
17850 @item @var{variable} @expansion{}
17851 @code{ @var{string} }
17853 @item @var{value} @expansion{}
17854 @code{ @var{const} | @var{tuple} | @var{list} }
17856 @item @var{const} @expansion{}
17857 @code{@var{c-string}}
17859 @item @var{tuple} @expansion{}
17860 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
17862 @item @var{list} @expansion{}
17863 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
17864 @var{result} ( "," @var{result} )* "]" }
17866 @item @var{stream-record} @expansion{}
17867 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
17869 @item @var{console-stream-output} @expansion{}
17870 @code{"~" @var{c-string}}
17872 @item @var{target-stream-output} @expansion{}
17873 @code{"@@" @var{c-string}}
17875 @item @var{log-stream-output} @expansion{}
17876 @code{"&" @var{c-string}}
17878 @item @var{nl} @expansion{}
17881 @item @var{token} @expansion{}
17882 @emph{any sequence of digits}.
17890 All output sequences end in a single line containing a period.
17893 The @code{@var{token}} is from the corresponding request. If an execution
17894 command is interrupted by the @samp{-exec-interrupt} command, the
17895 @var{token} associated with the @samp{*stopped} message is the one of the
17896 original execution command, not the one of the interrupt command.
17899 @cindex status output in @sc{gdb/mi}
17900 @var{status-async-output} contains on-going status information about the
17901 progress of a slow operation. It can be discarded. All status output is
17902 prefixed by @samp{+}.
17905 @cindex async output in @sc{gdb/mi}
17906 @var{exec-async-output} contains asynchronous state change on the target
17907 (stopped, started, disappeared). All async output is prefixed by
17911 @cindex notify output in @sc{gdb/mi}
17912 @var{notify-async-output} contains supplementary information that the
17913 client should handle (e.g., a new breakpoint information). All notify
17914 output is prefixed by @samp{=}.
17917 @cindex console output in @sc{gdb/mi}
17918 @var{console-stream-output} is output that should be displayed as is in the
17919 console. It is the textual response to a CLI command. All the console
17920 output is prefixed by @samp{~}.
17923 @cindex target output in @sc{gdb/mi}
17924 @var{target-stream-output} is the output produced by the target program.
17925 All the target output is prefixed by @samp{@@}.
17928 @cindex log output in @sc{gdb/mi}
17929 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
17930 instance messages that should be displayed as part of an error log. All
17931 the log output is prefixed by @samp{&}.
17934 @cindex list output in @sc{gdb/mi}
17935 New @sc{gdb/mi} commands should only output @var{lists} containing
17941 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
17942 details about the various output records.
17944 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
17945 @node GDB/MI Compatibility with CLI
17946 @section @sc{gdb/mi} Compatibility with CLI
17948 @cindex compatibility, @sc{gdb/mi} and CLI
17949 @cindex @sc{gdb/mi}, compatibility with CLI
17951 For the developers convenience CLI commands can be entered directly,
17952 but there may be some unexpected behaviour. For example, commands
17953 that query the user will behave as if the user replied yes, breakpoint
17954 command lists are not executed and some CLI commands, such as
17955 @code{if}, @code{when} and @code{define}, prompt for further input with
17956 @samp{>}, which is not valid MI output.
17958 This feature may be removed at some stage in the future and it is
17959 recommended that front ends use the @code{-interpreter-exec} command
17960 (@pxref{-interpreter-exec}).
17962 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
17963 @node GDB/MI Development and Front Ends
17964 @section @sc{gdb/mi} Development and Front Ends
17965 @cindex @sc{gdb/mi} development
17967 The application which takes the MI output and presents the state of the
17968 program being debugged to the user is called a @dfn{front end}.
17970 Although @sc{gdb/mi} is still incomplete, it is currently being used
17971 by a variety of front ends to @value{GDBN}. This makes it difficult
17972 to introduce new functionality without breaking existing usage. This
17973 section tries to minimize the problems by describing how the protocol
17976 Some changes in MI need not break a carefully designed front end, and
17977 for these the MI version will remain unchanged. The following is a
17978 list of changes that may occur within one level, so front ends should
17979 parse MI output in a way that can handle them:
17983 New MI commands may be added.
17986 New fields may be added to the output of any MI command.
17989 The range of values for fields with specified values, e.g.,
17990 @code{in_scope} (@pxref{-var-update}) may be extended.
17992 @c The format of field's content e.g type prefix, may change so parse it
17993 @c at your own risk. Yes, in general?
17995 @c The order of fields may change? Shouldn't really matter but it might
17996 @c resolve inconsistencies.
17999 If the changes are likely to break front ends, the MI version level
18000 will be increased by one. This will allow the front end to parse the
18001 output according to the MI version. Apart from mi0, new versions of
18002 @value{GDBN} will not support old versions of MI and it will be the
18003 responsibility of the front end to work with the new one.
18005 @c Starting with mi3, add a new command -mi-version that prints the MI
18008 The best way to avoid unexpected changes in MI that might break your front
18009 end is to make your project known to @value{GDBN} developers and
18010 follow development on @email{gdb@@sourceware.org} and
18011 @email{gdb-patches@@sourceware.org}. There is also the mailing list
18012 @email{dmi-discuss@@lists.freestandards.org}, hosted by the Free Standards
18013 Group, which has the aim of creating a more general MI protocol
18014 called Debugger Machine Interface (DMI) that will become a standard
18015 for all debuggers, not just @value{GDBN}.
18016 @cindex mailing lists
18018 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18019 @node GDB/MI Output Records
18020 @section @sc{gdb/mi} Output Records
18023 * GDB/MI Result Records::
18024 * GDB/MI Stream Records::
18025 * GDB/MI Out-of-band Records::
18028 @node GDB/MI Result Records
18029 @subsection @sc{gdb/mi} Result Records
18031 @cindex result records in @sc{gdb/mi}
18032 @cindex @sc{gdb/mi}, result records
18033 In addition to a number of out-of-band notifications, the response to a
18034 @sc{gdb/mi} command includes one of the following result indications:
18038 @item "^done" [ "," @var{results} ]
18039 The synchronous operation was successful, @code{@var{results}} are the return
18044 @c Is this one correct? Should it be an out-of-band notification?
18045 The asynchronous operation was successfully started. The target is
18050 @value{GDBN} has connected to a remote target.
18052 @item "^error" "," @var{c-string}
18054 The operation failed. The @code{@var{c-string}} contains the corresponding
18059 @value{GDBN} has terminated.
18063 @node GDB/MI Stream Records
18064 @subsection @sc{gdb/mi} Stream Records
18066 @cindex @sc{gdb/mi}, stream records
18067 @cindex stream records in @sc{gdb/mi}
18068 @value{GDBN} internally maintains a number of output streams: the console, the
18069 target, and the log. The output intended for each of these streams is
18070 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
18072 Each stream record begins with a unique @dfn{prefix character} which
18073 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
18074 Syntax}). In addition to the prefix, each stream record contains a
18075 @code{@var{string-output}}. This is either raw text (with an implicit new
18076 line) or a quoted C string (which does not contain an implicit newline).
18079 @item "~" @var{string-output}
18080 The console output stream contains text that should be displayed in the
18081 CLI console window. It contains the textual responses to CLI commands.
18083 @item "@@" @var{string-output}
18084 The target output stream contains any textual output from the running
18085 target. This is only present when GDB's event loop is truly
18086 asynchronous, which is currently only the case for remote targets.
18088 @item "&" @var{string-output}
18089 The log stream contains debugging messages being produced by @value{GDBN}'s
18093 @node GDB/MI Out-of-band Records
18094 @subsection @sc{gdb/mi} Out-of-band Records
18096 @cindex out-of-band records in @sc{gdb/mi}
18097 @cindex @sc{gdb/mi}, out-of-band records
18098 @dfn{Out-of-band} records are used to notify the @sc{gdb/mi} client of
18099 additional changes that have occurred. Those changes can either be a
18100 consequence of @sc{gdb/mi} (e.g., a breakpoint modified) or a result of
18101 target activity (e.g., target stopped).
18103 The following is a preliminary list of possible out-of-band records.
18104 In particular, the @var{exec-async-output} records.
18107 @item *stopped,reason="@var{reason}"
18110 @var{reason} can be one of the following:
18113 @item breakpoint-hit
18114 A breakpoint was reached.
18115 @item watchpoint-trigger
18116 A watchpoint was triggered.
18117 @item read-watchpoint-trigger
18118 A read watchpoint was triggered.
18119 @item access-watchpoint-trigger
18120 An access watchpoint was triggered.
18121 @item function-finished
18122 An -exec-finish or similar CLI command was accomplished.
18123 @item location-reached
18124 An -exec-until or similar CLI command was accomplished.
18125 @item watchpoint-scope
18126 A watchpoint has gone out of scope.
18127 @item end-stepping-range
18128 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
18129 similar CLI command was accomplished.
18130 @item exited-signalled
18131 The inferior exited because of a signal.
18133 The inferior exited.
18134 @item exited-normally
18135 The inferior exited normally.
18136 @item signal-received
18137 A signal was received by the inferior.
18141 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18142 @node GDB/MI Simple Examples
18143 @section Simple Examples of @sc{gdb/mi} Interaction
18144 @cindex @sc{gdb/mi}, simple examples
18146 This subsection presents several simple examples of interaction using
18147 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
18148 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
18149 the output received from @sc{gdb/mi}.
18151 Note the line breaks shown in the examples are here only for
18152 readability, they don't appear in the real output.
18154 @subheading Setting a Breakpoint
18156 Setting a breakpoint generates synchronous output which contains detailed
18157 information of the breakpoint.
18160 -> -break-insert main
18161 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
18162 enabled="y",addr="0x08048564",func="main",file="myprog.c",
18163 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
18167 @subheading Program Execution
18169 Program execution generates asynchronous records and MI gives the
18170 reason that execution stopped.
18176 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
18177 frame=@{addr="0x08048564",func="main",
18178 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
18179 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
18184 <- *stopped,reason="exited-normally"
18188 @subheading Quitting @value{GDBN}
18190 Quitting @value{GDBN} just prints the result class @samp{^exit}.
18198 @subheading A Bad Command
18200 Here's what happens if you pass a non-existent command:
18204 <- ^error,msg="Undefined MI command: rubbish"
18209 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18210 @node GDB/MI Command Description Format
18211 @section @sc{gdb/mi} Command Description Format
18213 The remaining sections describe blocks of commands. Each block of
18214 commands is laid out in a fashion similar to this section.
18216 @subheading Motivation
18218 The motivation for this collection of commands.
18220 @subheading Introduction
18222 A brief introduction to this collection of commands as a whole.
18224 @subheading Commands
18226 For each command in the block, the following is described:
18228 @subsubheading Synopsis
18231 -command @var{args}@dots{}
18234 @subsubheading Result
18236 @subsubheading @value{GDBN} Command
18238 The corresponding @value{GDBN} CLI command(s), if any.
18240 @subsubheading Example
18242 Example(s) formatted for readability. Some of the described commands have
18243 not been implemented yet and these are labeled N.A.@: (not available).
18246 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18247 @node GDB/MI Breakpoint Commands
18248 @section @sc{gdb/mi} Breakpoint Commands
18250 @cindex breakpoint commands for @sc{gdb/mi}
18251 @cindex @sc{gdb/mi}, breakpoint commands
18252 This section documents @sc{gdb/mi} commands for manipulating
18255 @subheading The @code{-break-after} Command
18256 @findex -break-after
18258 @subsubheading Synopsis
18261 -break-after @var{number} @var{count}
18264 The breakpoint number @var{number} is not in effect until it has been
18265 hit @var{count} times. To see how this is reflected in the output of
18266 the @samp{-break-list} command, see the description of the
18267 @samp{-break-list} command below.
18269 @subsubheading @value{GDBN} Command
18271 The corresponding @value{GDBN} command is @samp{ignore}.
18273 @subsubheading Example
18278 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
18279 enabled="y",addr="0x000100d0",func="main",file="hello.c",
18280 fullname="/home/foo/hello.c",line="5",times="0"@}
18287 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
18288 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18289 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18290 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18291 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18292 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18293 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18294 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
18295 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
18296 line="5",times="0",ignore="3"@}]@}
18301 @subheading The @code{-break-catch} Command
18302 @findex -break-catch
18304 @subheading The @code{-break-commands} Command
18305 @findex -break-commands
18309 @subheading The @code{-break-condition} Command
18310 @findex -break-condition
18312 @subsubheading Synopsis
18315 -break-condition @var{number} @var{expr}
18318 Breakpoint @var{number} will stop the program only if the condition in
18319 @var{expr} is true. The condition becomes part of the
18320 @samp{-break-list} output (see the description of the @samp{-break-list}
18323 @subsubheading @value{GDBN} Command
18325 The corresponding @value{GDBN} command is @samp{condition}.
18327 @subsubheading Example
18331 -break-condition 1 1
18335 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
18336 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18337 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18338 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18339 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18340 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18341 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18342 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
18343 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
18344 line="5",cond="1",times="0",ignore="3"@}]@}
18348 @subheading The @code{-break-delete} Command
18349 @findex -break-delete
18351 @subsubheading Synopsis
18354 -break-delete ( @var{breakpoint} )+
18357 Delete the breakpoint(s) whose number(s) are specified in the argument
18358 list. This is obviously reflected in the breakpoint list.
18360 @subsubheading @value{GDBN} Command
18362 The corresponding @value{GDBN} command is @samp{delete}.
18364 @subsubheading Example
18372 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
18373 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18374 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18375 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18376 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18377 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18378 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18383 @subheading The @code{-break-disable} Command
18384 @findex -break-disable
18386 @subsubheading Synopsis
18389 -break-disable ( @var{breakpoint} )+
18392 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
18393 break list is now set to @samp{n} for the named @var{breakpoint}(s).
18395 @subsubheading @value{GDBN} Command
18397 The corresponding @value{GDBN} command is @samp{disable}.
18399 @subsubheading Example
18407 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
18408 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18409 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18410 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18411 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18412 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18413 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18414 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
18415 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
18416 line="5",times="0"@}]@}
18420 @subheading The @code{-break-enable} Command
18421 @findex -break-enable
18423 @subsubheading Synopsis
18426 -break-enable ( @var{breakpoint} )+
18429 Enable (previously disabled) @var{breakpoint}(s).
18431 @subsubheading @value{GDBN} Command
18433 The corresponding @value{GDBN} command is @samp{enable}.
18435 @subsubheading Example
18443 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
18444 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18445 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18446 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18447 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18448 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18449 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18450 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
18451 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
18452 line="5",times="0"@}]@}
18456 @subheading The @code{-break-info} Command
18457 @findex -break-info
18459 @subsubheading Synopsis
18462 -break-info @var{breakpoint}
18466 Get information about a single breakpoint.
18468 @subsubheading @value{GDBN} Command
18470 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
18472 @subsubheading Example
18475 @subheading The @code{-break-insert} Command
18476 @findex -break-insert
18478 @subsubheading Synopsis
18481 -break-insert [ -t ] [ -h ] [ -f ]
18482 [ -c @var{condition} ] [ -i @var{ignore-count} ]
18483 [ -p @var{thread} ] [ @var{location} ]
18487 If specified, @var{location}, can be one of:
18494 @item filename:linenum
18495 @item filename:function
18499 The possible optional parameters of this command are:
18503 Insert a temporary breakpoint.
18505 Insert a hardware breakpoint.
18506 @item -c @var{condition}
18507 Make the breakpoint conditional on @var{condition}.
18508 @item -i @var{ignore-count}
18509 Initialize the @var{ignore-count}.
18511 If @var{location} cannot be parsed (for example if it
18512 refers to unknown files or functions), create a pending
18513 breakpoint. Without this flag, @value{GDBN} will report
18514 an error, and won't create a breakpoint, if @var{location}
18518 @subsubheading Result
18520 The result is in the form:
18523 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
18524 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
18525 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
18526 times="@var{times}"@}
18530 where @var{number} is the @value{GDBN} number for this breakpoint,
18531 @var{funcname} is the name of the function where the breakpoint was
18532 inserted, @var{filename} is the name of the source file which contains
18533 this function, @var{lineno} is the source line number within that file
18534 and @var{times} the number of times that the breakpoint has been hit
18535 (always 0 for -break-insert but may be greater for -break-info or -break-list
18536 which use the same output).
18538 Note: this format is open to change.
18539 @c An out-of-band breakpoint instead of part of the result?
18541 @subsubheading @value{GDBN} Command
18543 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
18544 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
18546 @subsubheading Example
18551 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
18552 fullname="/home/foo/recursive2.c,line="4",times="0"@}
18554 -break-insert -t foo
18555 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
18556 fullname="/home/foo/recursive2.c,line="11",times="0"@}
18559 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
18560 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18561 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18562 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18563 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18564 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18565 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18566 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
18567 addr="0x0001072c", func="main",file="recursive2.c",
18568 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
18569 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
18570 addr="0x00010774",func="foo",file="recursive2.c",
18571 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
18573 -break-insert -r foo.*
18574 ~int foo(int, int);
18575 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
18576 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
18580 @subheading The @code{-break-list} Command
18581 @findex -break-list
18583 @subsubheading Synopsis
18589 Displays the list of inserted breakpoints, showing the following fields:
18593 number of the breakpoint
18595 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
18597 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
18600 is the breakpoint enabled or no: @samp{y} or @samp{n}
18602 memory location at which the breakpoint is set
18604 logical location of the breakpoint, expressed by function name, file
18607 number of times the breakpoint has been hit
18610 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
18611 @code{body} field is an empty list.
18613 @subsubheading @value{GDBN} Command
18615 The corresponding @value{GDBN} command is @samp{info break}.
18617 @subsubheading Example
18622 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
18623 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18624 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18625 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18626 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18627 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18628 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18629 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
18630 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
18631 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
18632 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
18633 line="13",times="0"@}]@}
18637 Here's an example of the result when there are no breakpoints:
18642 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
18643 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18644 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18645 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18646 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18647 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18648 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18653 @subheading The @code{-break-watch} Command
18654 @findex -break-watch
18656 @subsubheading Synopsis
18659 -break-watch [ -a | -r ]
18662 Create a watchpoint. With the @samp{-a} option it will create an
18663 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
18664 read from or on a write to the memory location. With the @samp{-r}
18665 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
18666 trigger only when the memory location is accessed for reading. Without
18667 either of the options, the watchpoint created is a regular watchpoint,
18668 i.e., it will trigger when the memory location is accessed for writing.
18669 @xref{Set Watchpoints, , Setting Watchpoints}.
18671 Note that @samp{-break-list} will report a single list of watchpoints and
18672 breakpoints inserted.
18674 @subsubheading @value{GDBN} Command
18676 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
18679 @subsubheading Example
18681 Setting a watchpoint on a variable in the @code{main} function:
18686 ^done,wpt=@{number="2",exp="x"@}
18691 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
18692 value=@{old="-268439212",new="55"@},
18693 frame=@{func="main",args=[],file="recursive2.c",
18694 fullname="/home/foo/bar/recursive2.c",line="5"@}
18698 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
18699 the program execution twice: first for the variable changing value, then
18700 for the watchpoint going out of scope.
18705 ^done,wpt=@{number="5",exp="C"@}
18710 *stopped,reason="watchpoint-trigger",
18711 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
18712 frame=@{func="callee4",args=[],
18713 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
18714 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
18719 *stopped,reason="watchpoint-scope",wpnum="5",
18720 frame=@{func="callee3",args=[@{name="strarg",
18721 value="0x11940 \"A string argument.\""@}],
18722 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
18723 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
18727 Listing breakpoints and watchpoints, at different points in the program
18728 execution. Note that once the watchpoint goes out of scope, it is
18734 ^done,wpt=@{number="2",exp="C"@}
18737 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
18738 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18739 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18740 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18741 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18742 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18743 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18744 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
18745 addr="0x00010734",func="callee4",
18746 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
18747 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
18748 bkpt=@{number="2",type="watchpoint",disp="keep",
18749 enabled="y",addr="",what="C",times="0"@}]@}
18754 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
18755 value=@{old="-276895068",new="3"@},
18756 frame=@{func="callee4",args=[],
18757 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
18758 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
18761 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
18762 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18763 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18764 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18765 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18766 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18767 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18768 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
18769 addr="0x00010734",func="callee4",
18770 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
18771 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
18772 bkpt=@{number="2",type="watchpoint",disp="keep",
18773 enabled="y",addr="",what="C",times="-5"@}]@}
18777 ^done,reason="watchpoint-scope",wpnum="2",
18778 frame=@{func="callee3",args=[@{name="strarg",
18779 value="0x11940 \"A string argument.\""@}],
18780 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
18781 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
18784 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
18785 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
18786 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
18787 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
18788 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
18789 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
18790 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
18791 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
18792 addr="0x00010734",func="callee4",
18793 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
18794 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
18799 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18800 @node GDB/MI Program Context
18801 @section @sc{gdb/mi} Program Context
18803 @subheading The @code{-exec-arguments} Command
18804 @findex -exec-arguments
18807 @subsubheading Synopsis
18810 -exec-arguments @var{args}
18813 Set the inferior program arguments, to be used in the next
18816 @subsubheading @value{GDBN} Command
18818 The corresponding @value{GDBN} command is @samp{set args}.
18820 @subsubheading Example
18823 Don't have one around.
18826 @subheading The @code{-exec-show-arguments} Command
18827 @findex -exec-show-arguments
18829 @subsubheading Synopsis
18832 -exec-show-arguments
18835 Print the arguments of the program.
18837 @subsubheading @value{GDBN} Command
18839 The corresponding @value{GDBN} command is @samp{show args}.
18841 @subsubheading Example
18845 @subheading The @code{-environment-cd} Command
18846 @findex -environment-cd
18848 @subsubheading Synopsis
18851 -environment-cd @var{pathdir}
18854 Set @value{GDBN}'s working directory.
18856 @subsubheading @value{GDBN} Command
18858 The corresponding @value{GDBN} command is @samp{cd}.
18860 @subsubheading Example
18864 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
18870 @subheading The @code{-environment-directory} Command
18871 @findex -environment-directory
18873 @subsubheading Synopsis
18876 -environment-directory [ -r ] [ @var{pathdir} ]+
18879 Add directories @var{pathdir} to beginning of search path for source files.
18880 If the @samp{-r} option is used, the search path is reset to the default
18881 search path. If directories @var{pathdir} are supplied in addition to the
18882 @samp{-r} option, the search path is first reset and then addition
18884 Multiple directories may be specified, separated by blanks. Specifying
18885 multiple directories in a single command
18886 results in the directories added to the beginning of the
18887 search path in the same order they were presented in the command.
18888 If blanks are needed as
18889 part of a directory name, double-quotes should be used around
18890 the name. In the command output, the path will show up separated
18891 by the system directory-separator character. The directory-separator
18892 character must not be used
18893 in any directory name.
18894 If no directories are specified, the current search path is displayed.
18896 @subsubheading @value{GDBN} Command
18898 The corresponding @value{GDBN} command is @samp{dir}.
18900 @subsubheading Example
18904 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
18905 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
18907 -environment-directory ""
18908 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
18910 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
18911 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
18913 -environment-directory -r
18914 ^done,source-path="$cdir:$cwd"
18919 @subheading The @code{-environment-path} Command
18920 @findex -environment-path
18922 @subsubheading Synopsis
18925 -environment-path [ -r ] [ @var{pathdir} ]+
18928 Add directories @var{pathdir} to beginning of search path for object files.
18929 If the @samp{-r} option is used, the search path is reset to the original
18930 search path that existed at gdb start-up. If directories @var{pathdir} are
18931 supplied in addition to the
18932 @samp{-r} option, the search path is first reset and then addition
18934 Multiple directories may be specified, separated by blanks. Specifying
18935 multiple directories in a single command
18936 results in the directories added to the beginning of the
18937 search path in the same order they were presented in the command.
18938 If blanks are needed as
18939 part of a directory name, double-quotes should be used around
18940 the name. In the command output, the path will show up separated
18941 by the system directory-separator character. The directory-separator
18942 character must not be used
18943 in any directory name.
18944 If no directories are specified, the current path is displayed.
18947 @subsubheading @value{GDBN} Command
18949 The corresponding @value{GDBN} command is @samp{path}.
18951 @subsubheading Example
18956 ^done,path="/usr/bin"
18958 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
18959 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
18961 -environment-path -r /usr/local/bin
18962 ^done,path="/usr/local/bin:/usr/bin"
18967 @subheading The @code{-environment-pwd} Command
18968 @findex -environment-pwd
18970 @subsubheading Synopsis
18976 Show the current working directory.
18978 @subsubheading @value{GDBN} Command
18980 The corresponding @value{GDBN} command is @samp{pwd}.
18982 @subsubheading Example
18987 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
18991 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18992 @node GDB/MI Thread Commands
18993 @section @sc{gdb/mi} Thread Commands
18996 @subheading The @code{-thread-info} Command
18997 @findex -thread-info
18999 @subsubheading Synopsis
19002 -thread-info [ @var{thread-id} ]
19005 Reports information about either a specific thread, if
19006 the @var{thread-id} parameter is present, or about all
19007 threads. When printing information about all threads,
19008 also reports the current thread.
19010 @subsubheading @value{GDBN} Command
19012 The @samp{info thread} command prints the same information
19015 @subsubheading Example
19020 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
19021 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},
19022 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
19023 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
19024 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@}@}],
19025 current-thread-id="1"
19029 @subheading The @code{-thread-list-ids} Command
19030 @findex -thread-list-ids
19032 @subsubheading Synopsis
19038 Produces a list of the currently known @value{GDBN} thread ids. At the
19039 end of the list it also prints the total number of such threads.
19041 @subsubheading @value{GDBN} Command
19043 Part of @samp{info threads} supplies the same information.
19045 @subsubheading Example
19047 No threads present, besides the main process:
19052 ^done,thread-ids=@{@},number-of-threads="0"
19062 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
19063 number-of-threads="3"
19068 @subheading The @code{-thread-select} Command
19069 @findex -thread-select
19071 @subsubheading Synopsis
19074 -thread-select @var{threadnum}
19077 Make @var{threadnum} the current thread. It prints the number of the new
19078 current thread, and the topmost frame for that thread.
19080 @subsubheading @value{GDBN} Command
19082 The corresponding @value{GDBN} command is @samp{thread}.
19084 @subsubheading Example
19091 *stopped,reason="end-stepping-range",thread-id="2",line="187",
19092 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
19096 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
19097 number-of-threads="3"
19100 ^done,new-thread-id="3",
19101 frame=@{level="0",func="vprintf",
19102 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
19103 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
19107 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19108 @node GDB/MI Program Execution
19109 @section @sc{gdb/mi} Program Execution
19111 These are the asynchronous commands which generate the out-of-band
19112 record @samp{*stopped}. Currently @value{GDBN} only really executes
19113 asynchronously with remote targets and this interaction is mimicked in
19116 @subheading The @code{-exec-continue} Command
19117 @findex -exec-continue
19119 @subsubheading Synopsis
19125 Resumes the execution of the inferior program until a breakpoint is
19126 encountered, or until the inferior exits.
19128 @subsubheading @value{GDBN} Command
19130 The corresponding @value{GDBN} corresponding is @samp{continue}.
19132 @subsubheading Example
19139 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
19140 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
19146 @subheading The @code{-exec-finish} Command
19147 @findex -exec-finish
19149 @subsubheading Synopsis
19155 Resumes the execution of the inferior program until the current
19156 function is exited. Displays the results returned by the function.
19158 @subsubheading @value{GDBN} Command
19160 The corresponding @value{GDBN} command is @samp{finish}.
19162 @subsubheading Example
19164 Function returning @code{void}.
19171 *stopped,reason="function-finished",frame=@{func="main",args=[],
19172 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
19176 Function returning other than @code{void}. The name of the internal
19177 @value{GDBN} variable storing the result is printed, together with the
19184 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
19185 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
19186 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19187 gdb-result-var="$1",return-value="0"
19192 @subheading The @code{-exec-interrupt} Command
19193 @findex -exec-interrupt
19195 @subsubheading Synopsis
19201 Interrupts the background execution of the target. Note how the token
19202 associated with the stop message is the one for the execution command
19203 that has been interrupted. The token for the interrupt itself only
19204 appears in the @samp{^done} output. If the user is trying to
19205 interrupt a non-running program, an error message will be printed.
19207 @subsubheading @value{GDBN} Command
19209 The corresponding @value{GDBN} command is @samp{interrupt}.
19211 @subsubheading Example
19222 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
19223 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
19224 fullname="/home/foo/bar/try.c",line="13"@}
19229 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
19234 @subheading The @code{-exec-next} Command
19237 @subsubheading Synopsis
19243 Resumes execution of the inferior program, stopping when the beginning
19244 of the next source line is reached.
19246 @subsubheading @value{GDBN} Command
19248 The corresponding @value{GDBN} command is @samp{next}.
19250 @subsubheading Example
19256 *stopped,reason="end-stepping-range",line="8",file="hello.c"
19261 @subheading The @code{-exec-next-instruction} Command
19262 @findex -exec-next-instruction
19264 @subsubheading Synopsis
19267 -exec-next-instruction
19270 Executes one machine instruction. If the instruction is a function
19271 call, continues until the function returns. If the program stops at an
19272 instruction in the middle of a source line, the address will be
19275 @subsubheading @value{GDBN} Command
19277 The corresponding @value{GDBN} command is @samp{nexti}.
19279 @subsubheading Example
19283 -exec-next-instruction
19287 *stopped,reason="end-stepping-range",
19288 addr="0x000100d4",line="5",file="hello.c"
19293 @subheading The @code{-exec-return} Command
19294 @findex -exec-return
19296 @subsubheading Synopsis
19302 Makes current function return immediately. Doesn't execute the inferior.
19303 Displays the new current frame.
19305 @subsubheading @value{GDBN} Command
19307 The corresponding @value{GDBN} command is @samp{return}.
19309 @subsubheading Example
19313 200-break-insert callee4
19314 200^done,bkpt=@{number="1",addr="0x00010734",
19315 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
19320 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
19321 frame=@{func="callee4",args=[],
19322 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
19323 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
19329 111^done,frame=@{level="0",func="callee3",
19330 args=[@{name="strarg",
19331 value="0x11940 \"A string argument.\""@}],
19332 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
19333 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
19338 @subheading The @code{-exec-run} Command
19341 @subsubheading Synopsis
19347 Starts execution of the inferior from the beginning. The inferior
19348 executes until either a breakpoint is encountered or the program
19349 exits. In the latter case the output will include an exit code, if
19350 the program has exited exceptionally.
19352 @subsubheading @value{GDBN} Command
19354 The corresponding @value{GDBN} command is @samp{run}.
19356 @subsubheading Examples
19361 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
19366 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
19367 frame=@{func="main",args=[],file="recursive2.c",
19368 fullname="/home/foo/bar/recursive2.c",line="4"@}
19373 Program exited normally:
19381 *stopped,reason="exited-normally"
19386 Program exited exceptionally:
19394 *stopped,reason="exited",exit-code="01"
19398 Another way the program can terminate is if it receives a signal such as
19399 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
19403 *stopped,reason="exited-signalled",signal-name="SIGINT",
19404 signal-meaning="Interrupt"
19408 @c @subheading -exec-signal
19411 @subheading The @code{-exec-step} Command
19414 @subsubheading Synopsis
19420 Resumes execution of the inferior program, stopping when the beginning
19421 of the next source line is reached, if the next source line is not a
19422 function call. If it is, stop at the first instruction of the called
19425 @subsubheading @value{GDBN} Command
19427 The corresponding @value{GDBN} command is @samp{step}.
19429 @subsubheading Example
19431 Stepping into a function:
19437 *stopped,reason="end-stepping-range",
19438 frame=@{func="foo",args=[@{name="a",value="10"@},
19439 @{name="b",value="0"@}],file="recursive2.c",
19440 fullname="/home/foo/bar/recursive2.c",line="11"@}
19450 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
19455 @subheading The @code{-exec-step-instruction} Command
19456 @findex -exec-step-instruction
19458 @subsubheading Synopsis
19461 -exec-step-instruction
19464 Resumes the inferior which executes one machine instruction. The
19465 output, once @value{GDBN} has stopped, will vary depending on whether
19466 we have stopped in the middle of a source line or not. In the former
19467 case, the address at which the program stopped will be printed as
19470 @subsubheading @value{GDBN} Command
19472 The corresponding @value{GDBN} command is @samp{stepi}.
19474 @subsubheading Example
19478 -exec-step-instruction
19482 *stopped,reason="end-stepping-range",
19483 frame=@{func="foo",args=[],file="try.c",
19484 fullname="/home/foo/bar/try.c",line="10"@}
19486 -exec-step-instruction
19490 *stopped,reason="end-stepping-range",
19491 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
19492 fullname="/home/foo/bar/try.c",line="10"@}
19497 @subheading The @code{-exec-until} Command
19498 @findex -exec-until
19500 @subsubheading Synopsis
19503 -exec-until [ @var{location} ]
19506 Executes the inferior until the @var{location} specified in the
19507 argument is reached. If there is no argument, the inferior executes
19508 until a source line greater than the current one is reached. The
19509 reason for stopping in this case will be @samp{location-reached}.
19511 @subsubheading @value{GDBN} Command
19513 The corresponding @value{GDBN} command is @samp{until}.
19515 @subsubheading Example
19519 -exec-until recursive2.c:6
19523 *stopped,reason="location-reached",frame=@{func="main",args=[],
19524 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
19529 @subheading -file-clear
19530 Is this going away????
19533 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19534 @node GDB/MI Stack Manipulation
19535 @section @sc{gdb/mi} Stack Manipulation Commands
19538 @subheading The @code{-stack-info-frame} Command
19539 @findex -stack-info-frame
19541 @subsubheading Synopsis
19547 Get info on the selected frame.
19549 @subsubheading @value{GDBN} Command
19551 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
19552 (without arguments).
19554 @subsubheading Example
19559 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
19560 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
19561 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
19565 @subheading The @code{-stack-info-depth} Command
19566 @findex -stack-info-depth
19568 @subsubheading Synopsis
19571 -stack-info-depth [ @var{max-depth} ]
19574 Return the depth of the stack. If the integer argument @var{max-depth}
19575 is specified, do not count beyond @var{max-depth} frames.
19577 @subsubheading @value{GDBN} Command
19579 There's no equivalent @value{GDBN} command.
19581 @subsubheading Example
19583 For a stack with frame levels 0 through 11:
19590 -stack-info-depth 4
19593 -stack-info-depth 12
19596 -stack-info-depth 11
19599 -stack-info-depth 13
19604 @subheading The @code{-stack-list-arguments} Command
19605 @findex -stack-list-arguments
19607 @subsubheading Synopsis
19610 -stack-list-arguments @var{show-values}
19611 [ @var{low-frame} @var{high-frame} ]
19614 Display a list of the arguments for the frames between @var{low-frame}
19615 and @var{high-frame} (inclusive). If @var{low-frame} and
19616 @var{high-frame} are not provided, list the arguments for the whole
19617 call stack. If the two arguments are equal, show the single frame
19618 at the corresponding level. It is an error if @var{low-frame} is
19619 larger than the actual number of frames. On the other hand,
19620 @var{high-frame} may be larger than the actual number of frames, in
19621 which case only existing frames will be returned.
19623 The @var{show-values} argument must have a value of 0 or 1. A value of
19624 0 means that only the names of the arguments are listed, a value of 1
19625 means that both names and values of the arguments are printed.
19627 @subsubheading @value{GDBN} Command
19629 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
19630 @samp{gdb_get_args} command which partially overlaps with the
19631 functionality of @samp{-stack-list-arguments}.
19633 @subsubheading Example
19640 frame=@{level="0",addr="0x00010734",func="callee4",
19641 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
19642 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
19643 frame=@{level="1",addr="0x0001076c",func="callee3",
19644 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
19645 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
19646 frame=@{level="2",addr="0x0001078c",func="callee2",
19647 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
19648 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
19649 frame=@{level="3",addr="0x000107b4",func="callee1",
19650 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
19651 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
19652 frame=@{level="4",addr="0x000107e0",func="main",
19653 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
19654 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
19656 -stack-list-arguments 0
19659 frame=@{level="0",args=[]@},
19660 frame=@{level="1",args=[name="strarg"]@},
19661 frame=@{level="2",args=[name="intarg",name="strarg"]@},
19662 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
19663 frame=@{level="4",args=[]@}]
19665 -stack-list-arguments 1
19668 frame=@{level="0",args=[]@},
19670 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
19671 frame=@{level="2",args=[
19672 @{name="intarg",value="2"@},
19673 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
19674 @{frame=@{level="3",args=[
19675 @{name="intarg",value="2"@},
19676 @{name="strarg",value="0x11940 \"A string argument.\""@},
19677 @{name="fltarg",value="3.5"@}]@},
19678 frame=@{level="4",args=[]@}]
19680 -stack-list-arguments 0 2 2
19681 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
19683 -stack-list-arguments 1 2 2
19684 ^done,stack-args=[frame=@{level="2",
19685 args=[@{name="intarg",value="2"@},
19686 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
19690 @c @subheading -stack-list-exception-handlers
19693 @subheading The @code{-stack-list-frames} Command
19694 @findex -stack-list-frames
19696 @subsubheading Synopsis
19699 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
19702 List the frames currently on the stack. For each frame it displays the
19707 The frame number, 0 being the topmost frame, i.e., the innermost function.
19709 The @code{$pc} value for that frame.
19713 File name of the source file where the function lives.
19715 Line number corresponding to the @code{$pc}.
19718 If invoked without arguments, this command prints a backtrace for the
19719 whole stack. If given two integer arguments, it shows the frames whose
19720 levels are between the two arguments (inclusive). If the two arguments
19721 are equal, it shows the single frame at the corresponding level. It is
19722 an error if @var{low-frame} is larger than the actual number of
19723 frames. On the other hand, @var{high-frame} may be larger than the
19724 actual number of frames, in which case only existing frames will be returned.
19726 @subsubheading @value{GDBN} Command
19728 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
19730 @subsubheading Example
19732 Full stack backtrace:
19738 [frame=@{level="0",addr="0x0001076c",func="foo",
19739 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
19740 frame=@{level="1",addr="0x000107a4",func="foo",
19741 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19742 frame=@{level="2",addr="0x000107a4",func="foo",
19743 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19744 frame=@{level="3",addr="0x000107a4",func="foo",
19745 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19746 frame=@{level="4",addr="0x000107a4",func="foo",
19747 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19748 frame=@{level="5",addr="0x000107a4",func="foo",
19749 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19750 frame=@{level="6",addr="0x000107a4",func="foo",
19751 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19752 frame=@{level="7",addr="0x000107a4",func="foo",
19753 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19754 frame=@{level="8",addr="0x000107a4",func="foo",
19755 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19756 frame=@{level="9",addr="0x000107a4",func="foo",
19757 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19758 frame=@{level="10",addr="0x000107a4",func="foo",
19759 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19760 frame=@{level="11",addr="0x00010738",func="main",
19761 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
19765 Show frames between @var{low_frame} and @var{high_frame}:
19769 -stack-list-frames 3 5
19771 [frame=@{level="3",addr="0x000107a4",func="foo",
19772 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19773 frame=@{level="4",addr="0x000107a4",func="foo",
19774 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
19775 frame=@{level="5",addr="0x000107a4",func="foo",
19776 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
19780 Show a single frame:
19784 -stack-list-frames 3 3
19786 [frame=@{level="3",addr="0x000107a4",func="foo",
19787 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
19792 @subheading The @code{-stack-list-locals} Command
19793 @findex -stack-list-locals
19795 @subsubheading Synopsis
19798 -stack-list-locals @var{print-values}
19801 Display the local variable names for the selected frame. If
19802 @var{print-values} is 0 or @code{--no-values}, print only the names of
19803 the variables; if it is 1 or @code{--all-values}, print also their
19804 values; and if it is 2 or @code{--simple-values}, print the name,
19805 type and value for simple data types and the name and type for arrays,
19806 structures and unions. In this last case, a frontend can immediately
19807 display the value of simple data types and create variable objects for
19808 other data types when the user wishes to explore their values in
19811 @subsubheading @value{GDBN} Command
19813 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
19815 @subsubheading Example
19819 -stack-list-locals 0
19820 ^done,locals=[name="A",name="B",name="C"]
19822 -stack-list-locals --all-values
19823 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
19824 @{name="C",value="@{1, 2, 3@}"@}]
19825 -stack-list-locals --simple-values
19826 ^done,locals=[@{name="A",type="int",value="1"@},
19827 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
19832 @subheading The @code{-stack-select-frame} Command
19833 @findex -stack-select-frame
19835 @subsubheading Synopsis
19838 -stack-select-frame @var{framenum}
19841 Change the selected frame. Select a different frame @var{framenum} on
19844 @subsubheading @value{GDBN} Command
19846 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
19847 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
19849 @subsubheading Example
19853 -stack-select-frame 2
19858 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19859 @node GDB/MI Variable Objects
19860 @section @sc{gdb/mi} Variable Objects
19864 @subheading Motivation for Variable Objects in @sc{gdb/mi}
19866 For the implementation of a variable debugger window (locals, watched
19867 expressions, etc.), we are proposing the adaptation of the existing code
19868 used by @code{Insight}.
19870 The two main reasons for that are:
19874 It has been proven in practice (it is already on its second generation).
19877 It will shorten development time (needless to say how important it is
19881 The original interface was designed to be used by Tcl code, so it was
19882 slightly changed so it could be used through @sc{gdb/mi}. This section
19883 describes the @sc{gdb/mi} operations that will be available and gives some
19884 hints about their use.
19886 @emph{Note}: In addition to the set of operations described here, we
19887 expect the @sc{gui} implementation of a variable window to require, at
19888 least, the following operations:
19891 @item @code{-gdb-show} @code{output-radix}
19892 @item @code{-stack-list-arguments}
19893 @item @code{-stack-list-locals}
19894 @item @code{-stack-select-frame}
19899 @subheading Introduction to Variable Objects
19901 @cindex variable objects in @sc{gdb/mi}
19903 Variable objects are "object-oriented" MI interface for examining and
19904 changing values of expressions. Unlike some other MI interfaces that
19905 work with expressions, variable objects are specifically designed for
19906 simple and efficient presentation in the frontend. A variable object
19907 is identified by string name. When a variable object is created, the
19908 frontend specifies the expression for that variable object. The
19909 expression can be a simple variable, or it can be an arbitrary complex
19910 expression, and can even involve CPU registers. After creating a
19911 variable object, the frontend can invoke other variable object
19912 operations---for example to obtain or change the value of a variable
19913 object, or to change display format.
19915 Variable objects have hierarchical tree structure. Any variable object
19916 that corresponds to a composite type, such as structure in C, has
19917 a number of child variable objects, for example corresponding to each
19918 element of a structure. A child variable object can itself have
19919 children, recursively. Recursion ends when we reach
19920 leaf variable objects, which always have built-in types. Child variable
19921 objects are created only by explicit request, so if a frontend
19922 is not interested in the children of a particular variable object, no
19923 child will be created.
19925 For a leaf variable object it is possible to obtain its value as a
19926 string, or set the value from a string. String value can be also
19927 obtained for a non-leaf variable object, but it's generally a string
19928 that only indicates the type of the object, and does not list its
19929 contents. Assignment to a non-leaf variable object is not allowed.
19931 A frontend does not need to read the values of all variable objects each time
19932 the program stops. Instead, MI provides an update command that lists all
19933 variable objects whose values has changed since the last update
19934 operation. This considerably reduces the amount of data that must
19935 be transferred to the frontend. As noted above, children variable
19936 objects are created on demand, and only leaf variable objects have a
19937 real value. As result, gdb will read target memory only for leaf
19938 variables that frontend has created.
19940 The automatic update is not always desirable. For example, a frontend
19941 might want to keep a value of some expression for future reference,
19942 and never update it. For another example, fetching memory is
19943 relatively slow for embedded targets, so a frontend might want
19944 to disable automatic update for the variables that are either not
19945 visible on the screen, or ``closed''. This is possible using so
19946 called ``frozen variable objects''. Such variable objects are never
19947 implicitly updated.
19949 The following is the complete set of @sc{gdb/mi} operations defined to
19950 access this functionality:
19952 @multitable @columnfractions .4 .6
19953 @item @strong{Operation}
19954 @tab @strong{Description}
19956 @item @code{-var-create}
19957 @tab create a variable object
19958 @item @code{-var-delete}
19959 @tab delete the variable object and/or its children
19960 @item @code{-var-set-format}
19961 @tab set the display format of this variable
19962 @item @code{-var-show-format}
19963 @tab show the display format of this variable
19964 @item @code{-var-info-num-children}
19965 @tab tells how many children this object has
19966 @item @code{-var-list-children}
19967 @tab return a list of the object's children
19968 @item @code{-var-info-type}
19969 @tab show the type of this variable object
19970 @item @code{-var-info-expression}
19971 @tab print parent-relative expression that this variable object represents
19972 @item @code{-var-info-path-expression}
19973 @tab print full expression that this variable object represents
19974 @item @code{-var-show-attributes}
19975 @tab is this variable editable? does it exist here?
19976 @item @code{-var-evaluate-expression}
19977 @tab get the value of this variable
19978 @item @code{-var-assign}
19979 @tab set the value of this variable
19980 @item @code{-var-update}
19981 @tab update the variable and its children
19982 @item @code{-var-set-frozen}
19983 @tab set frozeness attribute
19986 In the next subsection we describe each operation in detail and suggest
19987 how it can be used.
19989 @subheading Description And Use of Operations on Variable Objects
19991 @subheading The @code{-var-create} Command
19992 @findex -var-create
19994 @subsubheading Synopsis
19997 -var-create @{@var{name} | "-"@}
19998 @{@var{frame-addr} | "*"@} @var{expression}
20001 This operation creates a variable object, which allows the monitoring of
20002 a variable, the result of an expression, a memory cell or a CPU
20005 The @var{name} parameter is the string by which the object can be
20006 referenced. It must be unique. If @samp{-} is specified, the varobj
20007 system will generate a string ``varNNNNNN'' automatically. It will be
20008 unique provided that one does not specify @var{name} on that format.
20009 The command fails if a duplicate name is found.
20011 The frame under which the expression should be evaluated can be
20012 specified by @var{frame-addr}. A @samp{*} indicates that the current
20013 frame should be used.
20015 @var{expression} is any expression valid on the current language set (must not
20016 begin with a @samp{*}), or one of the following:
20020 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
20023 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
20026 @samp{$@var{regname}} --- a CPU register name
20029 @subsubheading Result
20031 This operation returns the name, number of children and the type of the
20032 object created. Type is returned as a string as the ones generated by
20033 the @value{GDBN} CLI:
20036 name="@var{name}",numchild="N",type="@var{type}"
20040 @subheading The @code{-var-delete} Command
20041 @findex -var-delete
20043 @subsubheading Synopsis
20046 -var-delete [ -c ] @var{name}
20049 Deletes a previously created variable object and all of its children.
20050 With the @samp{-c} option, just deletes the children.
20052 Returns an error if the object @var{name} is not found.
20055 @subheading The @code{-var-set-format} Command
20056 @findex -var-set-format
20058 @subsubheading Synopsis
20061 -var-set-format @var{name} @var{format-spec}
20064 Sets the output format for the value of the object @var{name} to be
20067 @anchor{-var-set-format}
20068 The syntax for the @var{format-spec} is as follows:
20071 @var{format-spec} @expansion{}
20072 @{binary | decimal | hexadecimal | octal | natural@}
20075 The natural format is the default format choosen automatically
20076 based on the variable type (like decimal for an @code{int}, hex
20077 for pointers, etc.).
20079 For a variable with children, the format is set only on the
20080 variable itself, and the children are not affected.
20082 @subheading The @code{-var-show-format} Command
20083 @findex -var-show-format
20085 @subsubheading Synopsis
20088 -var-show-format @var{name}
20091 Returns the format used to display the value of the object @var{name}.
20094 @var{format} @expansion{}
20099 @subheading The @code{-var-info-num-children} Command
20100 @findex -var-info-num-children
20102 @subsubheading Synopsis
20105 -var-info-num-children @var{name}
20108 Returns the number of children of a variable object @var{name}:
20115 @subheading The @code{-var-list-children} Command
20116 @findex -var-list-children
20118 @subsubheading Synopsis
20121 -var-list-children [@var{print-values}] @var{name}
20123 @anchor{-var-list-children}
20125 Return a list of the children of the specified variable object and
20126 create variable objects for them, if they do not already exist. With
20127 a single argument or if @var{print-values} has a value for of 0 or
20128 @code{--no-values}, print only the names of the variables; if
20129 @var{print-values} is 1 or @code{--all-values}, also print their
20130 values; and if it is 2 or @code{--simple-values} print the name and
20131 value for simple data types and just the name for arrays, structures
20134 @subsubheading Example
20138 -var-list-children n
20139 ^done,numchild=@var{n},children=[@{name=@var{name},
20140 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
20142 -var-list-children --all-values n
20143 ^done,numchild=@var{n},children=[@{name=@var{name},
20144 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
20148 @subheading The @code{-var-info-type} Command
20149 @findex -var-info-type
20151 @subsubheading Synopsis
20154 -var-info-type @var{name}
20157 Returns the type of the specified variable @var{name}. The type is
20158 returned as a string in the same format as it is output by the
20162 type=@var{typename}
20166 @subheading The @code{-var-info-expression} Command
20167 @findex -var-info-expression
20169 @subsubheading Synopsis
20172 -var-info-expression @var{name}
20175 Returns a string that is suitable for presenting this
20176 variable object in user interface. The string is generally
20177 not valid expression in the current language, and cannot be evaluated.
20179 For example, if @code{a} is an array, and variable object
20180 @code{A} was created for @code{a}, then we'll get this output:
20183 (gdb) -var-info-expression A.1
20184 ^done,lang="C",exp="1"
20188 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
20190 Note that the output of the @code{-var-list-children} command also
20191 includes those expressions, so the @code{-var-info-expression} command
20194 @subheading The @code{-var-info-path-expression} Command
20195 @findex -var-info-path-expression
20197 @subsubheading Synopsis
20200 -var-info-path-expression @var{name}
20203 Returns an expression that can be evaluated in the current
20204 context and will yield the same value that a variable object has.
20205 Compare this with the @code{-var-info-expression} command, which
20206 result can be used only for UI presentation. Typical use of
20207 the @code{-var-info-path-expression} command is creating a
20208 watchpoint from a variable object.
20210 For example, suppose @code{C} is a C@t{++} class, derived from class
20211 @code{Base}, and that the @code{Base} class has a member called
20212 @code{m_size}. Assume a variable @code{c} is has the type of
20213 @code{C} and a variable object @code{C} was created for variable
20214 @code{c}. Then, we'll get this output:
20216 (gdb) -var-info-path-expression C.Base.public.m_size
20217 ^done,path_expr=((Base)c).m_size)
20220 @subheading The @code{-var-show-attributes} Command
20221 @findex -var-show-attributes
20223 @subsubheading Synopsis
20226 -var-show-attributes @var{name}
20229 List attributes of the specified variable object @var{name}:
20232 status=@var{attr} [ ( ,@var{attr} )* ]
20236 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
20238 @subheading The @code{-var-evaluate-expression} Command
20239 @findex -var-evaluate-expression
20241 @subsubheading Synopsis
20244 -var-evaluate-expression [-f @var{format-spec}] @var{name}
20247 Evaluates the expression that is represented by the specified variable
20248 object and returns its value as a string. The format of the string
20249 can be specified with the @samp{-f} option. The possible values of
20250 this option are the same as for @code{-var-set-format}
20251 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
20252 the current display format will be used. The current display format
20253 can be changed using the @code{-var-set-format} command.
20259 Note that one must invoke @code{-var-list-children} for a variable
20260 before the value of a child variable can be evaluated.
20262 @subheading The @code{-var-assign} Command
20263 @findex -var-assign
20265 @subsubheading Synopsis
20268 -var-assign @var{name} @var{expression}
20271 Assigns the value of @var{expression} to the variable object specified
20272 by @var{name}. The object must be @samp{editable}. If the variable's
20273 value is altered by the assign, the variable will show up in any
20274 subsequent @code{-var-update} list.
20276 @subsubheading Example
20284 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
20288 @subheading The @code{-var-update} Command
20289 @findex -var-update
20291 @subsubheading Synopsis
20294 -var-update [@var{print-values}] @{@var{name} | "*"@}
20297 Reevaluate the expressions corresponding to the variable object
20298 @var{name} and all its direct and indirect children, and return the
20299 list of variable objects whose values have changed; @var{name} must
20300 be a root variable object. Here, ``changed'' means that the result of
20301 @code{-var-evaluate-expression} before and after the
20302 @code{-var-update} is different. If @samp{*} is used as the variable
20303 object names, all existing variable objects are updated, except
20304 for frozen ones (@pxref{-var-set-frozen}). The option
20305 @var{print-values} determines whether both names and values, or just
20306 names are printed. The possible values of this option are the same
20307 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
20308 recommended to use the @samp{--all-values} option, to reduce the
20309 number of MI commands needed on each program stop.
20312 @subsubheading Example
20319 -var-update --all-values var1
20320 ^done,changelist=[@{name="var1",value="3",in_scope="true",
20321 type_changed="false"@}]
20325 @anchor{-var-update}
20326 The field in_scope may take three values:
20330 The variable object's current value is valid.
20333 The variable object does not currently hold a valid value but it may
20334 hold one in the future if its associated expression comes back into
20338 The variable object no longer holds a valid value.
20339 This can occur when the executable file being debugged has changed,
20340 either through recompilation or by using the @value{GDBN} @code{file}
20341 command. The front end should normally choose to delete these variable
20345 In the future new values may be added to this list so the front should
20346 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
20348 @subheading The @code{-var-set-frozen} Command
20349 @findex -var-set-frozen
20350 @anchor{-var-set-frozen}
20352 @subsubheading Synopsis
20355 -var-set-frozen @var{name} @var{flag}
20358 Set the frozenness flag on the variable object @var{name}. The
20359 @var{flag} parameter should be either @samp{1} to make the variable
20360 frozen or @samp{0} to make it unfrozen. If a variable object is
20361 frozen, then neither itself, nor any of its children, are
20362 implicitly updated by @code{-var-update} of
20363 a parent variable or by @code{-var-update *}. Only
20364 @code{-var-update} of the variable itself will update its value and
20365 values of its children. After a variable object is unfrozen, it is
20366 implicitly updated by all subsequent @code{-var-update} operations.
20367 Unfreezing a variable does not update it, only subsequent
20368 @code{-var-update} does.
20370 @subsubheading Example
20374 -var-set-frozen V 1
20380 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20381 @node GDB/MI Data Manipulation
20382 @section @sc{gdb/mi} Data Manipulation
20384 @cindex data manipulation, in @sc{gdb/mi}
20385 @cindex @sc{gdb/mi}, data manipulation
20386 This section describes the @sc{gdb/mi} commands that manipulate data:
20387 examine memory and registers, evaluate expressions, etc.
20389 @c REMOVED FROM THE INTERFACE.
20390 @c @subheading -data-assign
20391 @c Change the value of a program variable. Plenty of side effects.
20392 @c @subsubheading GDB Command
20394 @c @subsubheading Example
20397 @subheading The @code{-data-disassemble} Command
20398 @findex -data-disassemble
20400 @subsubheading Synopsis
20404 [ -s @var{start-addr} -e @var{end-addr} ]
20405 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
20413 @item @var{start-addr}
20414 is the beginning address (or @code{$pc})
20415 @item @var{end-addr}
20417 @item @var{filename}
20418 is the name of the file to disassemble
20419 @item @var{linenum}
20420 is the line number to disassemble around
20422 is the number of disassembly lines to be produced. If it is -1,
20423 the whole function will be disassembled, in case no @var{end-addr} is
20424 specified. If @var{end-addr} is specified as a non-zero value, and
20425 @var{lines} is lower than the number of disassembly lines between
20426 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
20427 displayed; if @var{lines} is higher than the number of lines between
20428 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
20431 is either 0 (meaning only disassembly) or 1 (meaning mixed source and
20435 @subsubheading Result
20437 The output for each instruction is composed of four fields:
20446 Note that whatever included in the instruction field, is not manipulated
20447 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
20449 @subsubheading @value{GDBN} Command
20451 There's no direct mapping from this command to the CLI.
20453 @subsubheading Example
20455 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
20459 -data-disassemble -s $pc -e "$pc + 20" -- 0
20462 @{address="0x000107c0",func-name="main",offset="4",
20463 inst="mov 2, %o0"@},
20464 @{address="0x000107c4",func-name="main",offset="8",
20465 inst="sethi %hi(0x11800), %o2"@},
20466 @{address="0x000107c8",func-name="main",offset="12",
20467 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
20468 @{address="0x000107cc",func-name="main",offset="16",
20469 inst="sethi %hi(0x11800), %o2"@},
20470 @{address="0x000107d0",func-name="main",offset="20",
20471 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
20475 Disassemble the whole @code{main} function. Line 32 is part of
20479 -data-disassemble -f basics.c -l 32 -- 0
20481 @{address="0x000107bc",func-name="main",offset="0",
20482 inst="save %sp, -112, %sp"@},
20483 @{address="0x000107c0",func-name="main",offset="4",
20484 inst="mov 2, %o0"@},
20485 @{address="0x000107c4",func-name="main",offset="8",
20486 inst="sethi %hi(0x11800), %o2"@},
20488 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
20489 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
20493 Disassemble 3 instructions from the start of @code{main}:
20497 -data-disassemble -f basics.c -l 32 -n 3 -- 0
20499 @{address="0x000107bc",func-name="main",offset="0",
20500 inst="save %sp, -112, %sp"@},
20501 @{address="0x000107c0",func-name="main",offset="4",
20502 inst="mov 2, %o0"@},
20503 @{address="0x000107c4",func-name="main",offset="8",
20504 inst="sethi %hi(0x11800), %o2"@}]
20508 Disassemble 3 instructions from the start of @code{main} in mixed mode:
20512 -data-disassemble -f basics.c -l 32 -n 3 -- 1
20514 src_and_asm_line=@{line="31",
20515 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
20516 testsuite/gdb.mi/basics.c",line_asm_insn=[
20517 @{address="0x000107bc",func-name="main",offset="0",
20518 inst="save %sp, -112, %sp"@}]@},
20519 src_and_asm_line=@{line="32",
20520 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
20521 testsuite/gdb.mi/basics.c",line_asm_insn=[
20522 @{address="0x000107c0",func-name="main",offset="4",
20523 inst="mov 2, %o0"@},
20524 @{address="0x000107c4",func-name="main",offset="8",
20525 inst="sethi %hi(0x11800), %o2"@}]@}]
20530 @subheading The @code{-data-evaluate-expression} Command
20531 @findex -data-evaluate-expression
20533 @subsubheading Synopsis
20536 -data-evaluate-expression @var{expr}
20539 Evaluate @var{expr} as an expression. The expression could contain an
20540 inferior function call. The function call will execute synchronously.
20541 If the expression contains spaces, it must be enclosed in double quotes.
20543 @subsubheading @value{GDBN} Command
20545 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
20546 @samp{call}. In @code{gdbtk} only, there's a corresponding
20547 @samp{gdb_eval} command.
20549 @subsubheading Example
20551 In the following example, the numbers that precede the commands are the
20552 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
20553 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
20557 211-data-evaluate-expression A
20560 311-data-evaluate-expression &A
20561 311^done,value="0xefffeb7c"
20563 411-data-evaluate-expression A+3
20566 511-data-evaluate-expression "A + 3"
20572 @subheading The @code{-data-list-changed-registers} Command
20573 @findex -data-list-changed-registers
20575 @subsubheading Synopsis
20578 -data-list-changed-registers
20581 Display a list of the registers that have changed.
20583 @subsubheading @value{GDBN} Command
20585 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
20586 has the corresponding command @samp{gdb_changed_register_list}.
20588 @subsubheading Example
20590 On a PPC MBX board:
20598 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
20599 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
20602 -data-list-changed-registers
20603 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
20604 "10","11","13","14","15","16","17","18","19","20","21","22","23",
20605 "24","25","26","27","28","30","31","64","65","66","67","69"]
20610 @subheading The @code{-data-list-register-names} Command
20611 @findex -data-list-register-names
20613 @subsubheading Synopsis
20616 -data-list-register-names [ ( @var{regno} )+ ]
20619 Show a list of register names for the current target. If no arguments
20620 are given, it shows a list of the names of all the registers. If
20621 integer numbers are given as arguments, it will print a list of the
20622 names of the registers corresponding to the arguments. To ensure
20623 consistency between a register name and its number, the output list may
20624 include empty register names.
20626 @subsubheading @value{GDBN} Command
20628 @value{GDBN} does not have a command which corresponds to
20629 @samp{-data-list-register-names}. In @code{gdbtk} there is a
20630 corresponding command @samp{gdb_regnames}.
20632 @subsubheading Example
20634 For the PPC MBX board:
20637 -data-list-register-names
20638 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
20639 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
20640 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
20641 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
20642 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
20643 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
20644 "", "pc","ps","cr","lr","ctr","xer"]
20646 -data-list-register-names 1 2 3
20647 ^done,register-names=["r1","r2","r3"]
20651 @subheading The @code{-data-list-register-values} Command
20652 @findex -data-list-register-values
20654 @subsubheading Synopsis
20657 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
20660 Display the registers' contents. @var{fmt} is the format according to
20661 which the registers' contents are to be returned, followed by an optional
20662 list of numbers specifying the registers to display. A missing list of
20663 numbers indicates that the contents of all the registers must be returned.
20665 Allowed formats for @var{fmt} are:
20682 @subsubheading @value{GDBN} Command
20684 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
20685 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
20687 @subsubheading Example
20689 For a PPC MBX board (note: line breaks are for readability only, they
20690 don't appear in the actual output):
20694 -data-list-register-values r 64 65
20695 ^done,register-values=[@{number="64",value="0xfe00a300"@},
20696 @{number="65",value="0x00029002"@}]
20698 -data-list-register-values x
20699 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
20700 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
20701 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
20702 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
20703 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
20704 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
20705 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
20706 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
20707 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
20708 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
20709 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
20710 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
20711 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
20712 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
20713 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
20714 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
20715 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
20716 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
20717 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
20718 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
20719 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
20720 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
20721 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
20722 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
20723 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
20724 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
20725 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
20726 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
20727 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
20728 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
20729 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
20730 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
20731 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
20732 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
20733 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
20734 @{number="69",value="0x20002b03"@}]
20739 @subheading The @code{-data-read-memory} Command
20740 @findex -data-read-memory
20742 @subsubheading Synopsis
20745 -data-read-memory [ -o @var{byte-offset} ]
20746 @var{address} @var{word-format} @var{word-size}
20747 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
20754 @item @var{address}
20755 An expression specifying the address of the first memory word to be
20756 read. Complex expressions containing embedded white space should be
20757 quoted using the C convention.
20759 @item @var{word-format}
20760 The format to be used to print the memory words. The notation is the
20761 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
20764 @item @var{word-size}
20765 The size of each memory word in bytes.
20767 @item @var{nr-rows}
20768 The number of rows in the output table.
20770 @item @var{nr-cols}
20771 The number of columns in the output table.
20774 If present, indicates that each row should include an @sc{ascii} dump. The
20775 value of @var{aschar} is used as a padding character when a byte is not a
20776 member of the printable @sc{ascii} character set (printable @sc{ascii}
20777 characters are those whose code is between 32 and 126, inclusively).
20779 @item @var{byte-offset}
20780 An offset to add to the @var{address} before fetching memory.
20783 This command displays memory contents as a table of @var{nr-rows} by
20784 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
20785 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
20786 (returned as @samp{total-bytes}). Should less than the requested number
20787 of bytes be returned by the target, the missing words are identified
20788 using @samp{N/A}. The number of bytes read from the target is returned
20789 in @samp{nr-bytes} and the starting address used to read memory in
20792 The address of the next/previous row or page is available in
20793 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
20796 @subsubheading @value{GDBN} Command
20798 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
20799 @samp{gdb_get_mem} memory read command.
20801 @subsubheading Example
20803 Read six bytes of memory starting at @code{bytes+6} but then offset by
20804 @code{-6} bytes. Format as three rows of two columns. One byte per
20805 word. Display each word in hex.
20809 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
20810 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
20811 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
20812 prev-page="0x0000138a",memory=[
20813 @{addr="0x00001390",data=["0x00","0x01"]@},
20814 @{addr="0x00001392",data=["0x02","0x03"]@},
20815 @{addr="0x00001394",data=["0x04","0x05"]@}]
20819 Read two bytes of memory starting at address @code{shorts + 64} and
20820 display as a single word formatted in decimal.
20824 5-data-read-memory shorts+64 d 2 1 1
20825 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
20826 next-row="0x00001512",prev-row="0x0000150e",
20827 next-page="0x00001512",prev-page="0x0000150e",memory=[
20828 @{addr="0x00001510",data=["128"]@}]
20832 Read thirty two bytes of memory starting at @code{bytes+16} and format
20833 as eight rows of four columns. Include a string encoding with @samp{x}
20834 used as the non-printable character.
20838 4-data-read-memory bytes+16 x 1 8 4 x
20839 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
20840 next-row="0x000013c0",prev-row="0x0000139c",
20841 next-page="0x000013c0",prev-page="0x00001380",memory=[
20842 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
20843 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
20844 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
20845 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
20846 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
20847 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
20848 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
20849 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
20853 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20854 @node GDB/MI Tracepoint Commands
20855 @section @sc{gdb/mi} Tracepoint Commands
20857 The tracepoint commands are not yet implemented.
20859 @c @subheading -trace-actions
20861 @c @subheading -trace-delete
20863 @c @subheading -trace-disable
20865 @c @subheading -trace-dump
20867 @c @subheading -trace-enable
20869 @c @subheading -trace-exists
20871 @c @subheading -trace-find
20873 @c @subheading -trace-frame-number
20875 @c @subheading -trace-info
20877 @c @subheading -trace-insert
20879 @c @subheading -trace-list
20881 @c @subheading -trace-pass-count
20883 @c @subheading -trace-save
20885 @c @subheading -trace-start
20887 @c @subheading -trace-stop
20890 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20891 @node GDB/MI Symbol Query
20892 @section @sc{gdb/mi} Symbol Query Commands
20895 @subheading The @code{-symbol-info-address} Command
20896 @findex -symbol-info-address
20898 @subsubheading Synopsis
20901 -symbol-info-address @var{symbol}
20904 Describe where @var{symbol} is stored.
20906 @subsubheading @value{GDBN} Command
20908 The corresponding @value{GDBN} command is @samp{info address}.
20910 @subsubheading Example
20914 @subheading The @code{-symbol-info-file} Command
20915 @findex -symbol-info-file
20917 @subsubheading Synopsis
20923 Show the file for the symbol.
20925 @subsubheading @value{GDBN} Command
20927 There's no equivalent @value{GDBN} command. @code{gdbtk} has
20928 @samp{gdb_find_file}.
20930 @subsubheading Example
20934 @subheading The @code{-symbol-info-function} Command
20935 @findex -symbol-info-function
20937 @subsubheading Synopsis
20940 -symbol-info-function
20943 Show which function the symbol lives in.
20945 @subsubheading @value{GDBN} Command
20947 @samp{gdb_get_function} in @code{gdbtk}.
20949 @subsubheading Example
20953 @subheading The @code{-symbol-info-line} Command
20954 @findex -symbol-info-line
20956 @subsubheading Synopsis
20962 Show the core addresses of the code for a source line.
20964 @subsubheading @value{GDBN} Command
20966 The corresponding @value{GDBN} command is @samp{info line}.
20967 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
20969 @subsubheading Example
20973 @subheading The @code{-symbol-info-symbol} Command
20974 @findex -symbol-info-symbol
20976 @subsubheading Synopsis
20979 -symbol-info-symbol @var{addr}
20982 Describe what symbol is at location @var{addr}.
20984 @subsubheading @value{GDBN} Command
20986 The corresponding @value{GDBN} command is @samp{info symbol}.
20988 @subsubheading Example
20992 @subheading The @code{-symbol-list-functions} Command
20993 @findex -symbol-list-functions
20995 @subsubheading Synopsis
20998 -symbol-list-functions
21001 List the functions in the executable.
21003 @subsubheading @value{GDBN} Command
21005 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
21006 @samp{gdb_search} in @code{gdbtk}.
21008 @subsubheading Example
21012 @subheading The @code{-symbol-list-lines} Command
21013 @findex -symbol-list-lines
21015 @subsubheading Synopsis
21018 -symbol-list-lines @var{filename}
21021 Print the list of lines that contain code and their associated program
21022 addresses for the given source filename. The entries are sorted in
21023 ascending PC order.
21025 @subsubheading @value{GDBN} Command
21027 There is no corresponding @value{GDBN} command.
21029 @subsubheading Example
21032 -symbol-list-lines basics.c
21033 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
21038 @subheading The @code{-symbol-list-types} Command
21039 @findex -symbol-list-types
21041 @subsubheading Synopsis
21047 List all the type names.
21049 @subsubheading @value{GDBN} Command
21051 The corresponding commands are @samp{info types} in @value{GDBN},
21052 @samp{gdb_search} in @code{gdbtk}.
21054 @subsubheading Example
21058 @subheading The @code{-symbol-list-variables} Command
21059 @findex -symbol-list-variables
21061 @subsubheading Synopsis
21064 -symbol-list-variables
21067 List all the global and static variable names.
21069 @subsubheading @value{GDBN} Command
21071 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
21073 @subsubheading Example
21077 @subheading The @code{-symbol-locate} Command
21078 @findex -symbol-locate
21080 @subsubheading Synopsis
21086 @subsubheading @value{GDBN} Command
21088 @samp{gdb_loc} in @code{gdbtk}.
21090 @subsubheading Example
21094 @subheading The @code{-symbol-type} Command
21095 @findex -symbol-type
21097 @subsubheading Synopsis
21100 -symbol-type @var{variable}
21103 Show type of @var{variable}.
21105 @subsubheading @value{GDBN} Command
21107 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
21108 @samp{gdb_obj_variable}.
21110 @subsubheading Example
21114 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21115 @node GDB/MI File Commands
21116 @section @sc{gdb/mi} File Commands
21118 This section describes the GDB/MI commands to specify executable file names
21119 and to read in and obtain symbol table information.
21121 @subheading The @code{-file-exec-and-symbols} Command
21122 @findex -file-exec-and-symbols
21124 @subsubheading Synopsis
21127 -file-exec-and-symbols @var{file}
21130 Specify the executable file to be debugged. This file is the one from
21131 which the symbol table is also read. If no file is specified, the
21132 command clears the executable and symbol information. If breakpoints
21133 are set when using this command with no arguments, @value{GDBN} will produce
21134 error messages. Otherwise, no output is produced, except a completion
21137 @subsubheading @value{GDBN} Command
21139 The corresponding @value{GDBN} command is @samp{file}.
21141 @subsubheading Example
21145 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
21151 @subheading The @code{-file-exec-file} Command
21152 @findex -file-exec-file
21154 @subsubheading Synopsis
21157 -file-exec-file @var{file}
21160 Specify the executable file to be debugged. Unlike
21161 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
21162 from this file. If used without argument, @value{GDBN} clears the information
21163 about the executable file. No output is produced, except a completion
21166 @subsubheading @value{GDBN} Command
21168 The corresponding @value{GDBN} command is @samp{exec-file}.
21170 @subsubheading Example
21174 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
21180 @subheading The @code{-file-list-exec-sections} Command
21181 @findex -file-list-exec-sections
21183 @subsubheading Synopsis
21186 -file-list-exec-sections
21189 List the sections of the current executable file.
21191 @subsubheading @value{GDBN} Command
21193 The @value{GDBN} command @samp{info file} shows, among the rest, the same
21194 information as this command. @code{gdbtk} has a corresponding command
21195 @samp{gdb_load_info}.
21197 @subsubheading Example
21201 @subheading The @code{-file-list-exec-source-file} Command
21202 @findex -file-list-exec-source-file
21204 @subsubheading Synopsis
21207 -file-list-exec-source-file
21210 List the line number, the current source file, and the absolute path
21211 to the current source file for the current executable. The macro
21212 information field has a value of @samp{1} or @samp{0} depending on
21213 whether or not the file includes preprocessor macro information.
21215 @subsubheading @value{GDBN} Command
21217 The @value{GDBN} equivalent is @samp{info source}
21219 @subsubheading Example
21223 123-file-list-exec-source-file
21224 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
21229 @subheading The @code{-file-list-exec-source-files} Command
21230 @findex -file-list-exec-source-files
21232 @subsubheading Synopsis
21235 -file-list-exec-source-files
21238 List the source files for the current executable.
21240 It will always output the filename, but only when @value{GDBN} can find
21241 the absolute file name of a source file, will it output the fullname.
21243 @subsubheading @value{GDBN} Command
21245 The @value{GDBN} equivalent is @samp{info sources}.
21246 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
21248 @subsubheading Example
21251 -file-list-exec-source-files
21253 @{file=foo.c,fullname=/home/foo.c@},
21254 @{file=/home/bar.c,fullname=/home/bar.c@},
21255 @{file=gdb_could_not_find_fullpath.c@}]
21259 @subheading The @code{-file-list-shared-libraries} Command
21260 @findex -file-list-shared-libraries
21262 @subsubheading Synopsis
21265 -file-list-shared-libraries
21268 List the shared libraries in the program.
21270 @subsubheading @value{GDBN} Command
21272 The corresponding @value{GDBN} command is @samp{info shared}.
21274 @subsubheading Example
21278 @subheading The @code{-file-list-symbol-files} Command
21279 @findex -file-list-symbol-files
21281 @subsubheading Synopsis
21284 -file-list-symbol-files
21289 @subsubheading @value{GDBN} Command
21291 The corresponding @value{GDBN} command is @samp{info file} (part of it).
21293 @subsubheading Example
21297 @subheading The @code{-file-symbol-file} Command
21298 @findex -file-symbol-file
21300 @subsubheading Synopsis
21303 -file-symbol-file @var{file}
21306 Read symbol table info from the specified @var{file} argument. When
21307 used without arguments, clears @value{GDBN}'s symbol table info. No output is
21308 produced, except for a completion notification.
21310 @subsubheading @value{GDBN} Command
21312 The corresponding @value{GDBN} command is @samp{symbol-file}.
21314 @subsubheading Example
21318 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
21324 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21325 @node GDB/MI Memory Overlay Commands
21326 @section @sc{gdb/mi} Memory Overlay Commands
21328 The memory overlay commands are not implemented.
21330 @c @subheading -overlay-auto
21332 @c @subheading -overlay-list-mapping-state
21334 @c @subheading -overlay-list-overlays
21336 @c @subheading -overlay-map
21338 @c @subheading -overlay-off
21340 @c @subheading -overlay-on
21342 @c @subheading -overlay-unmap
21344 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21345 @node GDB/MI Signal Handling Commands
21346 @section @sc{gdb/mi} Signal Handling Commands
21348 Signal handling commands are not implemented.
21350 @c @subheading -signal-handle
21352 @c @subheading -signal-list-handle-actions
21354 @c @subheading -signal-list-signal-types
21358 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21359 @node GDB/MI Target Manipulation
21360 @section @sc{gdb/mi} Target Manipulation Commands
21363 @subheading The @code{-target-attach} Command
21364 @findex -target-attach
21366 @subsubheading Synopsis
21369 -target-attach @var{pid} | @var{file}
21372 Attach to a process @var{pid} or a file @var{file} outside of @value{GDBN}.
21374 @subsubheading @value{GDBN} Command
21376 The corresponding @value{GDBN} command is @samp{attach}.
21378 @subsubheading Example
21382 @subheading The @code{-target-compare-sections} Command
21383 @findex -target-compare-sections
21385 @subsubheading Synopsis
21388 -target-compare-sections [ @var{section} ]
21391 Compare data of section @var{section} on target to the exec file.
21392 Without the argument, all sections are compared.
21394 @subsubheading @value{GDBN} Command
21396 The @value{GDBN} equivalent is @samp{compare-sections}.
21398 @subsubheading Example
21402 @subheading The @code{-target-detach} Command
21403 @findex -target-detach
21405 @subsubheading Synopsis
21411 Detach from the remote target which normally resumes its execution.
21414 @subsubheading @value{GDBN} Command
21416 The corresponding @value{GDBN} command is @samp{detach}.
21418 @subsubheading Example
21428 @subheading The @code{-target-disconnect} Command
21429 @findex -target-disconnect
21431 @subsubheading Synopsis
21437 Disconnect from the remote target. There's no output and the target is
21438 generally not resumed.
21440 @subsubheading @value{GDBN} Command
21442 The corresponding @value{GDBN} command is @samp{disconnect}.
21444 @subsubheading Example
21454 @subheading The @code{-target-download} Command
21455 @findex -target-download
21457 @subsubheading Synopsis
21463 Loads the executable onto the remote target.
21464 It prints out an update message every half second, which includes the fields:
21468 The name of the section.
21470 The size of what has been sent so far for that section.
21472 The size of the section.
21474 The total size of what was sent so far (the current and the previous sections).
21476 The size of the overall executable to download.
21480 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
21481 @sc{gdb/mi} Output Syntax}).
21483 In addition, it prints the name and size of the sections, as they are
21484 downloaded. These messages include the following fields:
21488 The name of the section.
21490 The size of the section.
21492 The size of the overall executable to download.
21496 At the end, a summary is printed.
21498 @subsubheading @value{GDBN} Command
21500 The corresponding @value{GDBN} command is @samp{load}.
21502 @subsubheading Example
21504 Note: each status message appears on a single line. Here the messages
21505 have been broken down so that they can fit onto a page.
21510 +download,@{section=".text",section-size="6668",total-size="9880"@}
21511 +download,@{section=".text",section-sent="512",section-size="6668",
21512 total-sent="512",total-size="9880"@}
21513 +download,@{section=".text",section-sent="1024",section-size="6668",
21514 total-sent="1024",total-size="9880"@}
21515 +download,@{section=".text",section-sent="1536",section-size="6668",
21516 total-sent="1536",total-size="9880"@}
21517 +download,@{section=".text",section-sent="2048",section-size="6668",
21518 total-sent="2048",total-size="9880"@}
21519 +download,@{section=".text",section-sent="2560",section-size="6668",
21520 total-sent="2560",total-size="9880"@}
21521 +download,@{section=".text",section-sent="3072",section-size="6668",
21522 total-sent="3072",total-size="9880"@}
21523 +download,@{section=".text",section-sent="3584",section-size="6668",
21524 total-sent="3584",total-size="9880"@}
21525 +download,@{section=".text",section-sent="4096",section-size="6668",
21526 total-sent="4096",total-size="9880"@}
21527 +download,@{section=".text",section-sent="4608",section-size="6668",
21528 total-sent="4608",total-size="9880"@}
21529 +download,@{section=".text",section-sent="5120",section-size="6668",
21530 total-sent="5120",total-size="9880"@}
21531 +download,@{section=".text",section-sent="5632",section-size="6668",
21532 total-sent="5632",total-size="9880"@}
21533 +download,@{section=".text",section-sent="6144",section-size="6668",
21534 total-sent="6144",total-size="9880"@}
21535 +download,@{section=".text",section-sent="6656",section-size="6668",
21536 total-sent="6656",total-size="9880"@}
21537 +download,@{section=".init",section-size="28",total-size="9880"@}
21538 +download,@{section=".fini",section-size="28",total-size="9880"@}
21539 +download,@{section=".data",section-size="3156",total-size="9880"@}
21540 +download,@{section=".data",section-sent="512",section-size="3156",
21541 total-sent="7236",total-size="9880"@}
21542 +download,@{section=".data",section-sent="1024",section-size="3156",
21543 total-sent="7748",total-size="9880"@}
21544 +download,@{section=".data",section-sent="1536",section-size="3156",
21545 total-sent="8260",total-size="9880"@}
21546 +download,@{section=".data",section-sent="2048",section-size="3156",
21547 total-sent="8772",total-size="9880"@}
21548 +download,@{section=".data",section-sent="2560",section-size="3156",
21549 total-sent="9284",total-size="9880"@}
21550 +download,@{section=".data",section-sent="3072",section-size="3156",
21551 total-sent="9796",total-size="9880"@}
21552 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
21558 @subheading The @code{-target-exec-status} Command
21559 @findex -target-exec-status
21561 @subsubheading Synopsis
21564 -target-exec-status
21567 Provide information on the state of the target (whether it is running or
21568 not, for instance).
21570 @subsubheading @value{GDBN} Command
21572 There's no equivalent @value{GDBN} command.
21574 @subsubheading Example
21578 @subheading The @code{-target-list-available-targets} Command
21579 @findex -target-list-available-targets
21581 @subsubheading Synopsis
21584 -target-list-available-targets
21587 List the possible targets to connect to.
21589 @subsubheading @value{GDBN} Command
21591 The corresponding @value{GDBN} command is @samp{help target}.
21593 @subsubheading Example
21597 @subheading The @code{-target-list-current-targets} Command
21598 @findex -target-list-current-targets
21600 @subsubheading Synopsis
21603 -target-list-current-targets
21606 Describe the current target.
21608 @subsubheading @value{GDBN} Command
21610 The corresponding information is printed by @samp{info file} (among
21613 @subsubheading Example
21617 @subheading The @code{-target-list-parameters} Command
21618 @findex -target-list-parameters
21620 @subsubheading Synopsis
21623 -target-list-parameters
21628 @subsubheading @value{GDBN} Command
21632 @subsubheading Example
21636 @subheading The @code{-target-select} Command
21637 @findex -target-select
21639 @subsubheading Synopsis
21642 -target-select @var{type} @var{parameters @dots{}}
21645 Connect @value{GDBN} to the remote target. This command takes two args:
21649 The type of target, for instance @samp{async}, @samp{remote}, etc.
21650 @item @var{parameters}
21651 Device names, host names and the like. @xref{Target Commands, ,
21652 Commands for Managing Targets}, for more details.
21655 The output is a connection notification, followed by the address at
21656 which the target program is, in the following form:
21659 ^connected,addr="@var{address}",func="@var{function name}",
21660 args=[@var{arg list}]
21663 @subsubheading @value{GDBN} Command
21665 The corresponding @value{GDBN} command is @samp{target}.
21667 @subsubheading Example
21671 -target-select async /dev/ttya
21672 ^connected,addr="0xfe00a300",func="??",args=[]
21676 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21677 @node GDB/MI File Transfer Commands
21678 @section @sc{gdb/mi} File Transfer Commands
21681 @subheading The @code{-target-file-put} Command
21682 @findex -target-file-put
21684 @subsubheading Synopsis
21687 -target-file-put @var{hostfile} @var{targetfile}
21690 Copy file @var{hostfile} from the host system (the machine running
21691 @value{GDBN}) to @var{targetfile} on the target system.
21693 @subsubheading @value{GDBN} Command
21695 The corresponding @value{GDBN} command is @samp{remote put}.
21697 @subsubheading Example
21701 -target-file-put localfile remotefile
21707 @subheading The @code{-target-file-put} Command
21708 @findex -target-file-get
21710 @subsubheading Synopsis
21713 -target-file-get @var{targetfile} @var{hostfile}
21716 Copy file @var{targetfile} from the target system to @var{hostfile}
21717 on the host system.
21719 @subsubheading @value{GDBN} Command
21721 The corresponding @value{GDBN} command is @samp{remote get}.
21723 @subsubheading Example
21727 -target-file-get remotefile localfile
21733 @subheading The @code{-target-file-delete} Command
21734 @findex -target-file-delete
21736 @subsubheading Synopsis
21739 -target-file-delete @var{targetfile}
21742 Delete @var{targetfile} from the target system.
21744 @subsubheading @value{GDBN} Command
21746 The corresponding @value{GDBN} command is @samp{remote delete}.
21748 @subsubheading Example
21752 -target-file-delete remotefile
21758 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21759 @node GDB/MI Miscellaneous Commands
21760 @section Miscellaneous @sc{gdb/mi} Commands
21762 @c @subheading -gdb-complete
21764 @subheading The @code{-gdb-exit} Command
21767 @subsubheading Synopsis
21773 Exit @value{GDBN} immediately.
21775 @subsubheading @value{GDBN} Command
21777 Approximately corresponds to @samp{quit}.
21779 @subsubheading Example
21788 @subheading The @code{-exec-abort} Command
21789 @findex -exec-abort
21791 @subsubheading Synopsis
21797 Kill the inferior running program.
21799 @subsubheading @value{GDBN} Command
21801 The corresponding @value{GDBN} command is @samp{kill}.
21803 @subsubheading Example
21807 @subheading The @code{-gdb-set} Command
21810 @subsubheading Synopsis
21816 Set an internal @value{GDBN} variable.
21817 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
21819 @subsubheading @value{GDBN} Command
21821 The corresponding @value{GDBN} command is @samp{set}.
21823 @subsubheading Example
21833 @subheading The @code{-gdb-show} Command
21836 @subsubheading Synopsis
21842 Show the current value of a @value{GDBN} variable.
21844 @subsubheading @value{GDBN} Command
21846 The corresponding @value{GDBN} command is @samp{show}.
21848 @subsubheading Example
21857 @c @subheading -gdb-source
21860 @subheading The @code{-gdb-version} Command
21861 @findex -gdb-version
21863 @subsubheading Synopsis
21869 Show version information for @value{GDBN}. Used mostly in testing.
21871 @subsubheading @value{GDBN} Command
21873 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
21874 default shows this information when you start an interactive session.
21876 @subsubheading Example
21878 @c This example modifies the actual output from GDB to avoid overfull
21884 ~Copyright 2000 Free Software Foundation, Inc.
21885 ~GDB is free software, covered by the GNU General Public License, and
21886 ~you are welcome to change it and/or distribute copies of it under
21887 ~ certain conditions.
21888 ~Type "show copying" to see the conditions.
21889 ~There is absolutely no warranty for GDB. Type "show warranty" for
21891 ~This GDB was configured as
21892 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
21897 @subheading The @code{-list-features} Command
21898 @findex -list-features
21900 Returns a list of particular features of the MI protocol that
21901 this version of gdb implements. A feature can be a command,
21902 or a new field in an output of some command, or even an
21903 important bugfix. While a frontend can sometimes detect presence
21904 of a feature at runtime, it is easier to perform detection at debugger
21907 The command returns a list of strings, with each string naming an
21908 available feature. Each returned string is just a name, it does not
21909 have any internal structure. The list of possible feature names
21915 (gdb) -list-features
21916 ^done,result=["feature1","feature2"]
21919 The current list of features is:
21923 @samp{frozen-varobjs}---indicates presence of the
21924 @code{-var-set-frozen} command, as well as possible presense of the
21925 @code{frozen} field in the output of @code{-varobj-create}.
21927 @samp{pending-breakpoints}---indicates presence of the @code{-f}
21928 option to the @code{-break-insert} command.
21930 @samp{thread-info}---indicates presence of the @code{-thread-info} command.
21934 @subheading The @code{-interpreter-exec} Command
21935 @findex -interpreter-exec
21937 @subheading Synopsis
21940 -interpreter-exec @var{interpreter} @var{command}
21942 @anchor{-interpreter-exec}
21944 Execute the specified @var{command} in the given @var{interpreter}.
21946 @subheading @value{GDBN} Command
21948 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
21950 @subheading Example
21954 -interpreter-exec console "break main"
21955 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
21956 &"During symbol reading, bad structure-type format.\n"
21957 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
21962 @subheading The @code{-inferior-tty-set} Command
21963 @findex -inferior-tty-set
21965 @subheading Synopsis
21968 -inferior-tty-set /dev/pts/1
21971 Set terminal for future runs of the program being debugged.
21973 @subheading @value{GDBN} Command
21975 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
21977 @subheading Example
21981 -inferior-tty-set /dev/pts/1
21986 @subheading The @code{-inferior-tty-show} Command
21987 @findex -inferior-tty-show
21989 @subheading Synopsis
21995 Show terminal for future runs of program being debugged.
21997 @subheading @value{GDBN} Command
21999 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
22001 @subheading Example
22005 -inferior-tty-set /dev/pts/1
22009 ^done,inferior_tty_terminal="/dev/pts/1"
22013 @subheading The @code{-enable-timings} Command
22014 @findex -enable-timings
22016 @subheading Synopsis
22019 -enable-timings [yes | no]
22022 Toggle the printing of the wallclock, user and system times for an MI
22023 command as a field in its output. This command is to help frontend
22024 developers optimize the performance of their code. No argument is
22025 equivalent to @samp{yes}.
22027 @subheading @value{GDBN} Command
22031 @subheading Example
22039 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22040 addr="0x080484ed",func="main",file="myprog.c",
22041 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
22042 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
22050 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
22051 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
22052 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
22053 fullname="/home/nickrob/myprog.c",line="73"@}
22058 @chapter @value{GDBN} Annotations
22060 This chapter describes annotations in @value{GDBN}. Annotations were
22061 designed to interface @value{GDBN} to graphical user interfaces or other
22062 similar programs which want to interact with @value{GDBN} at a
22063 relatively high level.
22065 The annotation mechanism has largely been superseded by @sc{gdb/mi}
22069 This is Edition @value{EDITION}, @value{DATE}.
22073 * Annotations Overview:: What annotations are; the general syntax.
22074 * Server Prefix:: Issuing a command without affecting user state.
22075 * Prompting:: Annotations marking @value{GDBN}'s need for input.
22076 * Errors:: Annotations for error messages.
22077 * Invalidation:: Some annotations describe things now invalid.
22078 * Annotations for Running::
22079 Whether the program is running, how it stopped, etc.
22080 * Source Annotations:: Annotations describing source code.
22083 @node Annotations Overview
22084 @section What is an Annotation?
22085 @cindex annotations
22087 Annotations start with a newline character, two @samp{control-z}
22088 characters, and the name of the annotation. If there is no additional
22089 information associated with this annotation, the name of the annotation
22090 is followed immediately by a newline. If there is additional
22091 information, the name of the annotation is followed by a space, the
22092 additional information, and a newline. The additional information
22093 cannot contain newline characters.
22095 Any output not beginning with a newline and two @samp{control-z}
22096 characters denotes literal output from @value{GDBN}. Currently there is
22097 no need for @value{GDBN} to output a newline followed by two
22098 @samp{control-z} characters, but if there was such a need, the
22099 annotations could be extended with an @samp{escape} annotation which
22100 means those three characters as output.
22102 The annotation @var{level}, which is specified using the
22103 @option{--annotate} command line option (@pxref{Mode Options}), controls
22104 how much information @value{GDBN} prints together with its prompt,
22105 values of expressions, source lines, and other types of output. Level 0
22106 is for no annotations, level 1 is for use when @value{GDBN} is run as a
22107 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
22108 for programs that control @value{GDBN}, and level 2 annotations have
22109 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
22110 Interface, annotate, GDB's Obsolete Annotations}).
22113 @kindex set annotate
22114 @item set annotate @var{level}
22115 The @value{GDBN} command @code{set annotate} sets the level of
22116 annotations to the specified @var{level}.
22118 @item show annotate
22119 @kindex show annotate
22120 Show the current annotation level.
22123 This chapter describes level 3 annotations.
22125 A simple example of starting up @value{GDBN} with annotations is:
22128 $ @kbd{gdb --annotate=3}
22130 Copyright 2003 Free Software Foundation, Inc.
22131 GDB is free software, covered by the GNU General Public License,
22132 and you are welcome to change it and/or distribute copies of it
22133 under certain conditions.
22134 Type "show copying" to see the conditions.
22135 There is absolutely no warranty for GDB. Type "show warranty"
22137 This GDB was configured as "i386-pc-linux-gnu"
22148 Here @samp{quit} is input to @value{GDBN}; the rest is output from
22149 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
22150 denotes a @samp{control-z} character) are annotations; the rest is
22151 output from @value{GDBN}.
22153 @node Server Prefix
22154 @section The Server Prefix
22155 @cindex server prefix
22157 If you prefix a command with @samp{server } then it will not affect
22158 the command history, nor will it affect @value{GDBN}'s notion of which
22159 command to repeat if @key{RET} is pressed on a line by itself. This
22160 means that commands can be run behind a user's back by a front-end in
22161 a transparent manner.
22163 The server prefix does not affect the recording of values into the value
22164 history; to print a value without recording it into the value history,
22165 use the @code{output} command instead of the @code{print} command.
22168 @section Annotation for @value{GDBN} Input
22170 @cindex annotations for prompts
22171 When @value{GDBN} prompts for input, it annotates this fact so it is possible
22172 to know when to send output, when the output from a given command is
22175 Different kinds of input each have a different @dfn{input type}. Each
22176 input type has three annotations: a @code{pre-} annotation, which
22177 denotes the beginning of any prompt which is being output, a plain
22178 annotation, which denotes the end of the prompt, and then a @code{post-}
22179 annotation which denotes the end of any echo which may (or may not) be
22180 associated with the input. For example, the @code{prompt} input type
22181 features the following annotations:
22189 The input types are
22192 @findex pre-prompt annotation
22193 @findex prompt annotation
22194 @findex post-prompt annotation
22196 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
22198 @findex pre-commands annotation
22199 @findex commands annotation
22200 @findex post-commands annotation
22202 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
22203 command. The annotations are repeated for each command which is input.
22205 @findex pre-overload-choice annotation
22206 @findex overload-choice annotation
22207 @findex post-overload-choice annotation
22208 @item overload-choice
22209 When @value{GDBN} wants the user to select between various overloaded functions.
22211 @findex pre-query annotation
22212 @findex query annotation
22213 @findex post-query annotation
22215 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
22217 @findex pre-prompt-for-continue annotation
22218 @findex prompt-for-continue annotation
22219 @findex post-prompt-for-continue annotation
22220 @item prompt-for-continue
22221 When @value{GDBN} is asking the user to press return to continue. Note: Don't
22222 expect this to work well; instead use @code{set height 0} to disable
22223 prompting. This is because the counting of lines is buggy in the
22224 presence of annotations.
22229 @cindex annotations for errors, warnings and interrupts
22231 @findex quit annotation
22236 This annotation occurs right before @value{GDBN} responds to an interrupt.
22238 @findex error annotation
22243 This annotation occurs right before @value{GDBN} responds to an error.
22245 Quit and error annotations indicate that any annotations which @value{GDBN} was
22246 in the middle of may end abruptly. For example, if a
22247 @code{value-history-begin} annotation is followed by a @code{error}, one
22248 cannot expect to receive the matching @code{value-history-end}. One
22249 cannot expect not to receive it either, however; an error annotation
22250 does not necessarily mean that @value{GDBN} is immediately returning all the way
22253 @findex error-begin annotation
22254 A quit or error annotation may be preceded by
22260 Any output between that and the quit or error annotation is the error
22263 Warning messages are not yet annotated.
22264 @c If we want to change that, need to fix warning(), type_error(),
22265 @c range_error(), and possibly other places.
22268 @section Invalidation Notices
22270 @cindex annotations for invalidation messages
22271 The following annotations say that certain pieces of state may have
22275 @findex frames-invalid annotation
22276 @item ^Z^Zframes-invalid
22278 The frames (for example, output from the @code{backtrace} command) may
22281 @findex breakpoints-invalid annotation
22282 @item ^Z^Zbreakpoints-invalid
22284 The breakpoints may have changed. For example, the user just added or
22285 deleted a breakpoint.
22288 @node Annotations for Running
22289 @section Running the Program
22290 @cindex annotations for running programs
22292 @findex starting annotation
22293 @findex stopping annotation
22294 When the program starts executing due to a @value{GDBN} command such as
22295 @code{step} or @code{continue},
22301 is output. When the program stops,
22307 is output. Before the @code{stopped} annotation, a variety of
22308 annotations describe how the program stopped.
22311 @findex exited annotation
22312 @item ^Z^Zexited @var{exit-status}
22313 The program exited, and @var{exit-status} is the exit status (zero for
22314 successful exit, otherwise nonzero).
22316 @findex signalled annotation
22317 @findex signal-name annotation
22318 @findex signal-name-end annotation
22319 @findex signal-string annotation
22320 @findex signal-string-end annotation
22321 @item ^Z^Zsignalled
22322 The program exited with a signal. After the @code{^Z^Zsignalled}, the
22323 annotation continues:
22329 ^Z^Zsignal-name-end
22333 ^Z^Zsignal-string-end
22338 where @var{name} is the name of the signal, such as @code{SIGILL} or
22339 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
22340 as @code{Illegal Instruction} or @code{Segmentation fault}.
22341 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
22342 user's benefit and have no particular format.
22344 @findex signal annotation
22346 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
22347 just saying that the program received the signal, not that it was
22348 terminated with it.
22350 @findex breakpoint annotation
22351 @item ^Z^Zbreakpoint @var{number}
22352 The program hit breakpoint number @var{number}.
22354 @findex watchpoint annotation
22355 @item ^Z^Zwatchpoint @var{number}
22356 The program hit watchpoint number @var{number}.
22359 @node Source Annotations
22360 @section Displaying Source
22361 @cindex annotations for source display
22363 @findex source annotation
22364 The following annotation is used instead of displaying source code:
22367 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
22370 where @var{filename} is an absolute file name indicating which source
22371 file, @var{line} is the line number within that file (where 1 is the
22372 first line in the file), @var{character} is the character position
22373 within the file (where 0 is the first character in the file) (for most
22374 debug formats this will necessarily point to the beginning of a line),
22375 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
22376 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
22377 @var{addr} is the address in the target program associated with the
22378 source which is being displayed. @var{addr} is in the form @samp{0x}
22379 followed by one or more lowercase hex digits (note that this does not
22380 depend on the language).
22383 @chapter Reporting Bugs in @value{GDBN}
22384 @cindex bugs in @value{GDBN}
22385 @cindex reporting bugs in @value{GDBN}
22387 Your bug reports play an essential role in making @value{GDBN} reliable.
22389 Reporting a bug may help you by bringing a solution to your problem, or it
22390 may not. But in any case the principal function of a bug report is to help
22391 the entire community by making the next version of @value{GDBN} work better. Bug
22392 reports are your contribution to the maintenance of @value{GDBN}.
22394 In order for a bug report to serve its purpose, you must include the
22395 information that enables us to fix the bug.
22398 * Bug Criteria:: Have you found a bug?
22399 * Bug Reporting:: How to report bugs
22403 @section Have You Found a Bug?
22404 @cindex bug criteria
22406 If you are not sure whether you have found a bug, here are some guidelines:
22409 @cindex fatal signal
22410 @cindex debugger crash
22411 @cindex crash of debugger
22413 If the debugger gets a fatal signal, for any input whatever, that is a
22414 @value{GDBN} bug. Reliable debuggers never crash.
22416 @cindex error on valid input
22418 If @value{GDBN} produces an error message for valid input, that is a
22419 bug. (Note that if you're cross debugging, the problem may also be
22420 somewhere in the connection to the target.)
22422 @cindex invalid input
22424 If @value{GDBN} does not produce an error message for invalid input,
22425 that is a bug. However, you should note that your idea of
22426 ``invalid input'' might be our idea of ``an extension'' or ``support
22427 for traditional practice''.
22430 If you are an experienced user of debugging tools, your suggestions
22431 for improvement of @value{GDBN} are welcome in any case.
22434 @node Bug Reporting
22435 @section How to Report Bugs
22436 @cindex bug reports
22437 @cindex @value{GDBN} bugs, reporting
22439 A number of companies and individuals offer support for @sc{gnu} products.
22440 If you obtained @value{GDBN} from a support organization, we recommend you
22441 contact that organization first.
22443 You can find contact information for many support companies and
22444 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
22446 @c should add a web page ref...
22448 In any event, we also recommend that you submit bug reports for
22449 @value{GDBN}. The preferred method is to submit them directly using
22450 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
22451 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
22454 @strong{Do not send bug reports to @samp{info-gdb}, or to
22455 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
22456 not want to receive bug reports. Those that do have arranged to receive
22459 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
22460 serves as a repeater. The mailing list and the newsgroup carry exactly
22461 the same messages. Often people think of posting bug reports to the
22462 newsgroup instead of mailing them. This appears to work, but it has one
22463 problem which can be crucial: a newsgroup posting often lacks a mail
22464 path back to the sender. Thus, if we need to ask for more information,
22465 we may be unable to reach you. For this reason, it is better to send
22466 bug reports to the mailing list.
22468 The fundamental principle of reporting bugs usefully is this:
22469 @strong{report all the facts}. If you are not sure whether to state a
22470 fact or leave it out, state it!
22472 Often people omit facts because they think they know what causes the
22473 problem and assume that some details do not matter. Thus, you might
22474 assume that the name of the variable you use in an example does not matter.
22475 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
22476 stray memory reference which happens to fetch from the location where that
22477 name is stored in memory; perhaps, if the name were different, the contents
22478 of that location would fool the debugger into doing the right thing despite
22479 the bug. Play it safe and give a specific, complete example. That is the
22480 easiest thing for you to do, and the most helpful.
22482 Keep in mind that the purpose of a bug report is to enable us to fix the
22483 bug. It may be that the bug has been reported previously, but neither
22484 you nor we can know that unless your bug report is complete and
22487 Sometimes people give a few sketchy facts and ask, ``Does this ring a
22488 bell?'' Those bug reports are useless, and we urge everyone to
22489 @emph{refuse to respond to them} except to chide the sender to report
22492 To enable us to fix the bug, you should include all these things:
22496 The version of @value{GDBN}. @value{GDBN} announces it if you start
22497 with no arguments; you can also print it at any time using @code{show
22500 Without this, we will not know whether there is any point in looking for
22501 the bug in the current version of @value{GDBN}.
22504 The type of machine you are using, and the operating system name and
22508 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
22509 ``@value{GCC}--2.8.1''.
22512 What compiler (and its version) was used to compile the program you are
22513 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
22514 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
22515 to get this information; for other compilers, see the documentation for
22519 The command arguments you gave the compiler to compile your example and
22520 observe the bug. For example, did you use @samp{-O}? To guarantee
22521 you will not omit something important, list them all. A copy of the
22522 Makefile (or the output from make) is sufficient.
22524 If we were to try to guess the arguments, we would probably guess wrong
22525 and then we might not encounter the bug.
22528 A complete input script, and all necessary source files, that will
22532 A description of what behavior you observe that you believe is
22533 incorrect. For example, ``It gets a fatal signal.''
22535 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
22536 will certainly notice it. But if the bug is incorrect output, we might
22537 not notice unless it is glaringly wrong. You might as well not give us
22538 a chance to make a mistake.
22540 Even if the problem you experience is a fatal signal, you should still
22541 say so explicitly. Suppose something strange is going on, such as, your
22542 copy of @value{GDBN} is out of synch, or you have encountered a bug in
22543 the C library on your system. (This has happened!) Your copy might
22544 crash and ours would not. If you told us to expect a crash, then when
22545 ours fails to crash, we would know that the bug was not happening for
22546 us. If you had not told us to expect a crash, then we would not be able
22547 to draw any conclusion from our observations.
22550 @cindex recording a session script
22551 To collect all this information, you can use a session recording program
22552 such as @command{script}, which is available on many Unix systems.
22553 Just run your @value{GDBN} session inside @command{script} and then
22554 include the @file{typescript} file with your bug report.
22556 Another way to record a @value{GDBN} session is to run @value{GDBN}
22557 inside Emacs and then save the entire buffer to a file.
22560 If you wish to suggest changes to the @value{GDBN} source, send us context
22561 diffs. If you even discuss something in the @value{GDBN} source, refer to
22562 it by context, not by line number.
22564 The line numbers in our development sources will not match those in your
22565 sources. Your line numbers would convey no useful information to us.
22569 Here are some things that are not necessary:
22573 A description of the envelope of the bug.
22575 Often people who encounter a bug spend a lot of time investigating
22576 which changes to the input file will make the bug go away and which
22577 changes will not affect it.
22579 This is often time consuming and not very useful, because the way we
22580 will find the bug is by running a single example under the debugger
22581 with breakpoints, not by pure deduction from a series of examples.
22582 We recommend that you save your time for something else.
22584 Of course, if you can find a simpler example to report @emph{instead}
22585 of the original one, that is a convenience for us. Errors in the
22586 output will be easier to spot, running under the debugger will take
22587 less time, and so on.
22589 However, simplification is not vital; if you do not want to do this,
22590 report the bug anyway and send us the entire test case you used.
22593 A patch for the bug.
22595 A patch for the bug does help us if it is a good one. But do not omit
22596 the necessary information, such as the test case, on the assumption that
22597 a patch is all we need. We might see problems with your patch and decide
22598 to fix the problem another way, or we might not understand it at all.
22600 Sometimes with a program as complicated as @value{GDBN} it is very hard to
22601 construct an example that will make the program follow a certain path
22602 through the code. If you do not send us the example, we will not be able
22603 to construct one, so we will not be able to verify that the bug is fixed.
22605 And if we cannot understand what bug you are trying to fix, or why your
22606 patch should be an improvement, we will not install it. A test case will
22607 help us to understand.
22610 A guess about what the bug is or what it depends on.
22612 Such guesses are usually wrong. Even we cannot guess right about such
22613 things without first using the debugger to find the facts.
22616 @c The readline documentation is distributed with the readline code
22617 @c and consists of the two following files:
22619 @c inc-hist.texinfo
22620 @c Use -I with makeinfo to point to the appropriate directory,
22621 @c environment var TEXINPUTS with TeX.
22622 @include rluser.texi
22623 @include inc-hist.texinfo
22626 @node Formatting Documentation
22627 @appendix Formatting Documentation
22629 @cindex @value{GDBN} reference card
22630 @cindex reference card
22631 The @value{GDBN} 4 release includes an already-formatted reference card, ready
22632 for printing with PostScript or Ghostscript, in the @file{gdb}
22633 subdirectory of the main source directory@footnote{In
22634 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
22635 release.}. If you can use PostScript or Ghostscript with your printer,
22636 you can print the reference card immediately with @file{refcard.ps}.
22638 The release also includes the source for the reference card. You
22639 can format it, using @TeX{}, by typing:
22645 The @value{GDBN} reference card is designed to print in @dfn{landscape}
22646 mode on US ``letter'' size paper;
22647 that is, on a sheet 11 inches wide by 8.5 inches
22648 high. You will need to specify this form of printing as an option to
22649 your @sc{dvi} output program.
22651 @cindex documentation
22653 All the documentation for @value{GDBN} comes as part of the machine-readable
22654 distribution. The documentation is written in Texinfo format, which is
22655 a documentation system that uses a single source file to produce both
22656 on-line information and a printed manual. You can use one of the Info
22657 formatting commands to create the on-line version of the documentation
22658 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
22660 @value{GDBN} includes an already formatted copy of the on-line Info
22661 version of this manual in the @file{gdb} subdirectory. The main Info
22662 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
22663 subordinate files matching @samp{gdb.info*} in the same directory. If
22664 necessary, you can print out these files, or read them with any editor;
22665 but they are easier to read using the @code{info} subsystem in @sc{gnu}
22666 Emacs or the standalone @code{info} program, available as part of the
22667 @sc{gnu} Texinfo distribution.
22669 If you want to format these Info files yourself, you need one of the
22670 Info formatting programs, such as @code{texinfo-format-buffer} or
22673 If you have @code{makeinfo} installed, and are in the top level
22674 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
22675 version @value{GDBVN}), you can make the Info file by typing:
22682 If you want to typeset and print copies of this manual, you need @TeX{},
22683 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
22684 Texinfo definitions file.
22686 @TeX{} is a typesetting program; it does not print files directly, but
22687 produces output files called @sc{dvi} files. To print a typeset
22688 document, you need a program to print @sc{dvi} files. If your system
22689 has @TeX{} installed, chances are it has such a program. The precise
22690 command to use depends on your system; @kbd{lpr -d} is common; another
22691 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
22692 require a file name without any extension or a @samp{.dvi} extension.
22694 @TeX{} also requires a macro definitions file called
22695 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
22696 written in Texinfo format. On its own, @TeX{} cannot either read or
22697 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
22698 and is located in the @file{gdb-@var{version-number}/texinfo}
22701 If you have @TeX{} and a @sc{dvi} printer program installed, you can
22702 typeset and print this manual. First switch to the @file{gdb}
22703 subdirectory of the main source directory (for example, to
22704 @file{gdb-@value{GDBVN}/gdb}) and type:
22710 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
22712 @node Installing GDB
22713 @appendix Installing @value{GDBN}
22714 @cindex installation
22717 * Requirements:: Requirements for building @value{GDBN}
22718 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
22719 * Separate Objdir:: Compiling @value{GDBN} in another directory
22720 * Config Names:: Specifying names for hosts and targets
22721 * Configure Options:: Summary of options for configure
22725 @section Requirements for Building @value{GDBN}
22726 @cindex building @value{GDBN}, requirements for
22728 Building @value{GDBN} requires various tools and packages to be available.
22729 Other packages will be used only if they are found.
22731 @heading Tools/Packages Necessary for Building @value{GDBN}
22733 @item ISO C90 compiler
22734 @value{GDBN} is written in ISO C90. It should be buildable with any
22735 working C90 compiler, e.g.@: GCC.
22739 @heading Tools/Packages Optional for Building @value{GDBN}
22743 @value{GDBN} can use the Expat XML parsing library. This library may be
22744 included with your operating system distribution; if it is not, you
22745 can get the latest version from @url{http://expat.sourceforge.net}.
22746 The @file{configure} script will search for this library in several
22747 standard locations; if it is installed in an unusual path, you can
22748 use the @option{--with-libexpat-prefix} option to specify its location.
22754 Remote protocol memory maps (@pxref{Memory Map Format})
22756 Target descriptions (@pxref{Target Descriptions})
22758 Remote shared library lists (@pxref{Library List Format})
22760 MS-Windows shared libraries (@pxref{Shared Libraries})
22764 @cindex compressed debug sections
22765 @value{GDBN} will use the @samp{zlib} library, if available, to read
22766 compressed debug sections. Some linkers, such as GNU gold, are capable
22767 of producing binaries with compressed debug sections. If @value{GDBN}
22768 is compiled with @samp{zlib}, it will be able to read the debug
22769 information in such binaries.
22771 The @samp{zlib} library is likely included with your operating system
22772 distribution; if it is not, you can get the latest version from
22773 @url{http://zlib.net}.
22777 @node Running Configure
22778 @section Invoking the @value{GDBN} @file{configure} Script
22779 @cindex configuring @value{GDBN}
22780 @value{GDBN} comes with a @file{configure} script that automates the process
22781 of preparing @value{GDBN} for installation; you can then use @code{make} to
22782 build the @code{gdb} program.
22784 @c irrelevant in info file; it's as current as the code it lives with.
22785 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
22786 look at the @file{README} file in the sources; we may have improved the
22787 installation procedures since publishing this manual.}
22790 The @value{GDBN} distribution includes all the source code you need for
22791 @value{GDBN} in a single directory, whose name is usually composed by
22792 appending the version number to @samp{gdb}.
22794 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
22795 @file{gdb-@value{GDBVN}} directory. That directory contains:
22798 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
22799 script for configuring @value{GDBN} and all its supporting libraries
22801 @item gdb-@value{GDBVN}/gdb
22802 the source specific to @value{GDBN} itself
22804 @item gdb-@value{GDBVN}/bfd
22805 source for the Binary File Descriptor library
22807 @item gdb-@value{GDBVN}/include
22808 @sc{gnu} include files
22810 @item gdb-@value{GDBVN}/libiberty
22811 source for the @samp{-liberty} free software library
22813 @item gdb-@value{GDBVN}/opcodes
22814 source for the library of opcode tables and disassemblers
22816 @item gdb-@value{GDBVN}/readline
22817 source for the @sc{gnu} command-line interface
22819 @item gdb-@value{GDBVN}/glob
22820 source for the @sc{gnu} filename pattern-matching subroutine
22822 @item gdb-@value{GDBVN}/mmalloc
22823 source for the @sc{gnu} memory-mapped malloc package
22826 The simplest way to configure and build @value{GDBN} is to run @file{configure}
22827 from the @file{gdb-@var{version-number}} source directory, which in
22828 this example is the @file{gdb-@value{GDBVN}} directory.
22830 First switch to the @file{gdb-@var{version-number}} source directory
22831 if you are not already in it; then run @file{configure}. Pass the
22832 identifier for the platform on which @value{GDBN} will run as an
22838 cd gdb-@value{GDBVN}
22839 ./configure @var{host}
22844 where @var{host} is an identifier such as @samp{sun4} or
22845 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
22846 (You can often leave off @var{host}; @file{configure} tries to guess the
22847 correct value by examining your system.)
22849 Running @samp{configure @var{host}} and then running @code{make} builds the
22850 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
22851 libraries, then @code{gdb} itself. The configured source files, and the
22852 binaries, are left in the corresponding source directories.
22855 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
22856 system does not recognize this automatically when you run a different
22857 shell, you may need to run @code{sh} on it explicitly:
22860 sh configure @var{host}
22863 If you run @file{configure} from a directory that contains source
22864 directories for multiple libraries or programs, such as the
22865 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
22867 creates configuration files for every directory level underneath (unless
22868 you tell it not to, with the @samp{--norecursion} option).
22870 You should run the @file{configure} script from the top directory in the
22871 source tree, the @file{gdb-@var{version-number}} directory. If you run
22872 @file{configure} from one of the subdirectories, you will configure only
22873 that subdirectory. That is usually not what you want. In particular,
22874 if you run the first @file{configure} from the @file{gdb} subdirectory
22875 of the @file{gdb-@var{version-number}} directory, you will omit the
22876 configuration of @file{bfd}, @file{readline}, and other sibling
22877 directories of the @file{gdb} subdirectory. This leads to build errors
22878 about missing include files such as @file{bfd/bfd.h}.
22880 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
22881 However, you should make sure that the shell on your path (named by
22882 the @samp{SHELL} environment variable) is publicly readable. Remember
22883 that @value{GDBN} uses the shell to start your program---some systems refuse to
22884 let @value{GDBN} debug child processes whose programs are not readable.
22886 @node Separate Objdir
22887 @section Compiling @value{GDBN} in Another Directory
22889 If you want to run @value{GDBN} versions for several host or target machines,
22890 you need a different @code{gdb} compiled for each combination of
22891 host and target. @file{configure} is designed to make this easy by
22892 allowing you to generate each configuration in a separate subdirectory,
22893 rather than in the source directory. If your @code{make} program
22894 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
22895 @code{make} in each of these directories builds the @code{gdb}
22896 program specified there.
22898 To build @code{gdb} in a separate directory, run @file{configure}
22899 with the @samp{--srcdir} option to specify where to find the source.
22900 (You also need to specify a path to find @file{configure}
22901 itself from your working directory. If the path to @file{configure}
22902 would be the same as the argument to @samp{--srcdir}, you can leave out
22903 the @samp{--srcdir} option; it is assumed.)
22905 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
22906 separate directory for a Sun 4 like this:
22910 cd gdb-@value{GDBVN}
22913 ../gdb-@value{GDBVN}/configure sun4
22918 When @file{configure} builds a configuration using a remote source
22919 directory, it creates a tree for the binaries with the same structure
22920 (and using the same names) as the tree under the source directory. In
22921 the example, you'd find the Sun 4 library @file{libiberty.a} in the
22922 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
22923 @file{gdb-sun4/gdb}.
22925 Make sure that your path to the @file{configure} script has just one
22926 instance of @file{gdb} in it. If your path to @file{configure} looks
22927 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
22928 one subdirectory of @value{GDBN}, not the whole package. This leads to
22929 build errors about missing include files such as @file{bfd/bfd.h}.
22931 One popular reason to build several @value{GDBN} configurations in separate
22932 directories is to configure @value{GDBN} for cross-compiling (where
22933 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
22934 programs that run on another machine---the @dfn{target}).
22935 You specify a cross-debugging target by
22936 giving the @samp{--target=@var{target}} option to @file{configure}.
22938 When you run @code{make} to build a program or library, you must run
22939 it in a configured directory---whatever directory you were in when you
22940 called @file{configure} (or one of its subdirectories).
22942 The @code{Makefile} that @file{configure} generates in each source
22943 directory also runs recursively. If you type @code{make} in a source
22944 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
22945 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
22946 will build all the required libraries, and then build GDB.
22948 When you have multiple hosts or targets configured in separate
22949 directories, you can run @code{make} on them in parallel (for example,
22950 if they are NFS-mounted on each of the hosts); they will not interfere
22954 @section Specifying Names for Hosts and Targets
22956 The specifications used for hosts and targets in the @file{configure}
22957 script are based on a three-part naming scheme, but some short predefined
22958 aliases are also supported. The full naming scheme encodes three pieces
22959 of information in the following pattern:
22962 @var{architecture}-@var{vendor}-@var{os}
22965 For example, you can use the alias @code{sun4} as a @var{host} argument,
22966 or as the value for @var{target} in a @code{--target=@var{target}}
22967 option. The equivalent full name is @samp{sparc-sun-sunos4}.
22969 The @file{configure} script accompanying @value{GDBN} does not provide
22970 any query facility to list all supported host and target names or
22971 aliases. @file{configure} calls the Bourne shell script
22972 @code{config.sub} to map abbreviations to full names; you can read the
22973 script, if you wish, or you can use it to test your guesses on
22974 abbreviations---for example:
22977 % sh config.sub i386-linux
22979 % sh config.sub alpha-linux
22980 alpha-unknown-linux-gnu
22981 % sh config.sub hp9k700
22983 % sh config.sub sun4
22984 sparc-sun-sunos4.1.1
22985 % sh config.sub sun3
22986 m68k-sun-sunos4.1.1
22987 % sh config.sub i986v
22988 Invalid configuration `i986v': machine `i986v' not recognized
22992 @code{config.sub} is also distributed in the @value{GDBN} source
22993 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
22995 @node Configure Options
22996 @section @file{configure} Options
22998 Here is a summary of the @file{configure} options and arguments that
22999 are most often useful for building @value{GDBN}. @file{configure} also has
23000 several other options not listed here. @inforef{What Configure
23001 Does,,configure.info}, for a full explanation of @file{configure}.
23004 configure @r{[}--help@r{]}
23005 @r{[}--prefix=@var{dir}@r{]}
23006 @r{[}--exec-prefix=@var{dir}@r{]}
23007 @r{[}--srcdir=@var{dirname}@r{]}
23008 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
23009 @r{[}--target=@var{target}@r{]}
23014 You may introduce options with a single @samp{-} rather than
23015 @samp{--} if you prefer; but you may abbreviate option names if you use
23020 Display a quick summary of how to invoke @file{configure}.
23022 @item --prefix=@var{dir}
23023 Configure the source to install programs and files under directory
23026 @item --exec-prefix=@var{dir}
23027 Configure the source to install programs under directory
23030 @c avoid splitting the warning from the explanation:
23032 @item --srcdir=@var{dirname}
23033 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
23034 @code{make} that implements the @code{VPATH} feature.}@*
23035 Use this option to make configurations in directories separate from the
23036 @value{GDBN} source directories. Among other things, you can use this to
23037 build (or maintain) several configurations simultaneously, in separate
23038 directories. @file{configure} writes configuration-specific files in
23039 the current directory, but arranges for them to use the source in the
23040 directory @var{dirname}. @file{configure} creates directories under
23041 the working directory in parallel to the source directories below
23044 @item --norecursion
23045 Configure only the directory level where @file{configure} is executed; do not
23046 propagate configuration to subdirectories.
23048 @item --target=@var{target}
23049 Configure @value{GDBN} for cross-debugging programs running on the specified
23050 @var{target}. Without this option, @value{GDBN} is configured to debug
23051 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
23053 There is no convenient way to generate a list of all available targets.
23055 @item @var{host} @dots{}
23056 Configure @value{GDBN} to run on the specified @var{host}.
23058 There is no convenient way to generate a list of all available hosts.
23061 There are many other options available as well, but they are generally
23062 needed for special purposes only.
23064 @node Maintenance Commands
23065 @appendix Maintenance Commands
23066 @cindex maintenance commands
23067 @cindex internal commands
23069 In addition to commands intended for @value{GDBN} users, @value{GDBN}
23070 includes a number of commands intended for @value{GDBN} developers,
23071 that are not documented elsewhere in this manual. These commands are
23072 provided here for reference. (For commands that turn on debugging
23073 messages, see @ref{Debugging Output}.)
23076 @kindex maint agent
23077 @item maint agent @var{expression}
23078 Translate the given @var{expression} into remote agent bytecodes.
23079 This command is useful for debugging the Agent Expression mechanism
23080 (@pxref{Agent Expressions}).
23082 @kindex maint info breakpoints
23083 @item @anchor{maint info breakpoints}maint info breakpoints
23084 Using the same format as @samp{info breakpoints}, display both the
23085 breakpoints you've set explicitly, and those @value{GDBN} is using for
23086 internal purposes. Internal breakpoints are shown with negative
23087 breakpoint numbers. The type column identifies what kind of breakpoint
23092 Normal, explicitly set breakpoint.
23095 Normal, explicitly set watchpoint.
23098 Internal breakpoint, used to handle correctly stepping through
23099 @code{longjmp} calls.
23101 @item longjmp resume
23102 Internal breakpoint at the target of a @code{longjmp}.
23105 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
23108 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
23111 Shared library events.
23115 @kindex maint check-symtabs
23116 @item maint check-symtabs
23117 Check the consistency of psymtabs and symtabs.
23119 @kindex maint cplus first_component
23120 @item maint cplus first_component @var{name}
23121 Print the first C@t{++} class/namespace component of @var{name}.
23123 @kindex maint cplus namespace
23124 @item maint cplus namespace
23125 Print the list of possible C@t{++} namespaces.
23127 @kindex maint demangle
23128 @item maint demangle @var{name}
23129 Demangle a C@t{++} or Objective-C mangled @var{name}.
23131 @kindex maint deprecate
23132 @kindex maint undeprecate
23133 @cindex deprecated commands
23134 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
23135 @itemx maint undeprecate @var{command}
23136 Deprecate or undeprecate the named @var{command}. Deprecated commands
23137 cause @value{GDBN} to issue a warning when you use them. The optional
23138 argument @var{replacement} says which newer command should be used in
23139 favor of the deprecated one; if it is given, @value{GDBN} will mention
23140 the replacement as part of the warning.
23142 @kindex maint dump-me
23143 @item maint dump-me
23144 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
23145 Cause a fatal signal in the debugger and force it to dump its core.
23146 This is supported only on systems which support aborting a program
23147 with the @code{SIGQUIT} signal.
23149 @kindex maint internal-error
23150 @kindex maint internal-warning
23151 @item maint internal-error @r{[}@var{message-text}@r{]}
23152 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
23153 Cause @value{GDBN} to call the internal function @code{internal_error}
23154 or @code{internal_warning} and hence behave as though an internal error
23155 or internal warning has been detected. In addition to reporting the
23156 internal problem, these functions give the user the opportunity to
23157 either quit @value{GDBN} or create a core file of the current
23158 @value{GDBN} session.
23160 These commands take an optional parameter @var{message-text} that is
23161 used as the text of the error or warning message.
23163 Here's an example of using @code{internal-error}:
23166 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
23167 @dots{}/maint.c:121: internal-error: testing, 1, 2
23168 A problem internal to GDB has been detected. Further
23169 debugging may prove unreliable.
23170 Quit this debugging session? (y or n) @kbd{n}
23171 Create a core file? (y or n) @kbd{n}
23175 @kindex maint packet
23176 @item maint packet @var{text}
23177 If @value{GDBN} is talking to an inferior via the serial protocol,
23178 then this command sends the string @var{text} to the inferior, and
23179 displays the response packet. @value{GDBN} supplies the initial
23180 @samp{$} character, the terminating @samp{#} character, and the
23183 @kindex maint print architecture
23184 @item maint print architecture @r{[}@var{file}@r{]}
23185 Print the entire architecture configuration. The optional argument
23186 @var{file} names the file where the output goes.
23188 @kindex maint print c-tdesc
23189 @item maint print c-tdesc
23190 Print the current target description (@pxref{Target Descriptions}) as
23191 a C source file. The created source file can be used in @value{GDBN}
23192 when an XML parser is not available to parse the description.
23194 @kindex maint print dummy-frames
23195 @item maint print dummy-frames
23196 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
23199 (@value{GDBP}) @kbd{b add}
23201 (@value{GDBP}) @kbd{print add(2,3)}
23202 Breakpoint 2, add (a=2, b=3) at @dots{}
23204 The program being debugged stopped while in a function called from GDB.
23206 (@value{GDBP}) @kbd{maint print dummy-frames}
23207 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
23208 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
23209 call_lo=0x01014000 call_hi=0x01014001
23213 Takes an optional file parameter.
23215 @kindex maint print registers
23216 @kindex maint print raw-registers
23217 @kindex maint print cooked-registers
23218 @kindex maint print register-groups
23219 @item maint print registers @r{[}@var{file}@r{]}
23220 @itemx maint print raw-registers @r{[}@var{file}@r{]}
23221 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
23222 @itemx maint print register-groups @r{[}@var{file}@r{]}
23223 Print @value{GDBN}'s internal register data structures.
23225 The command @code{maint print raw-registers} includes the contents of
23226 the raw register cache; the command @code{maint print cooked-registers}
23227 includes the (cooked) value of all registers; and the command
23228 @code{maint print register-groups} includes the groups that each
23229 register is a member of. @xref{Registers,, Registers, gdbint,
23230 @value{GDBN} Internals}.
23232 These commands take an optional parameter, a file name to which to
23233 write the information.
23235 @kindex maint print reggroups
23236 @item maint print reggroups @r{[}@var{file}@r{]}
23237 Print @value{GDBN}'s internal register group data structures. The
23238 optional argument @var{file} tells to what file to write the
23241 The register groups info looks like this:
23244 (@value{GDBP}) @kbd{maint print reggroups}
23257 This command forces @value{GDBN} to flush its internal register cache.
23259 @kindex maint print objfiles
23260 @cindex info for known object files
23261 @item maint print objfiles
23262 Print a dump of all known object files. For each object file, this
23263 command prints its name, address in memory, and all of its psymtabs
23266 @kindex maint print statistics
23267 @cindex bcache statistics
23268 @item maint print statistics
23269 This command prints, for each object file in the program, various data
23270 about that object file followed by the byte cache (@dfn{bcache})
23271 statistics for the object file. The objfile data includes the number
23272 of minimal, partial, full, and stabs symbols, the number of types
23273 defined by the objfile, the number of as yet unexpanded psym tables,
23274 the number of line tables and string tables, and the amount of memory
23275 used by the various tables. The bcache statistics include the counts,
23276 sizes, and counts of duplicates of all and unique objects, max,
23277 average, and median entry size, total memory used and its overhead and
23278 savings, and various measures of the hash table size and chain
23281 @kindex maint print target-stack
23282 @cindex target stack description
23283 @item maint print target-stack
23284 A @dfn{target} is an interface between the debugger and a particular
23285 kind of file or process. Targets can be stacked in @dfn{strata},
23286 so that more than one target can potentially respond to a request.
23287 In particular, memory accesses will walk down the stack of targets
23288 until they find a target that is interested in handling that particular
23291 This command prints a short description of each layer that was pushed on
23292 the @dfn{target stack}, starting from the top layer down to the bottom one.
23294 @kindex maint print type
23295 @cindex type chain of a data type
23296 @item maint print type @var{expr}
23297 Print the type chain for a type specified by @var{expr}. The argument
23298 can be either a type name or a symbol. If it is a symbol, the type of
23299 that symbol is described. The type chain produced by this command is
23300 a recursive definition of the data type as stored in @value{GDBN}'s
23301 data structures, including its flags and contained types.
23303 @kindex maint set dwarf2 max-cache-age
23304 @kindex maint show dwarf2 max-cache-age
23305 @item maint set dwarf2 max-cache-age
23306 @itemx maint show dwarf2 max-cache-age
23307 Control the DWARF 2 compilation unit cache.
23309 @cindex DWARF 2 compilation units cache
23310 In object files with inter-compilation-unit references, such as those
23311 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
23312 reader needs to frequently refer to previously read compilation units.
23313 This setting controls how long a compilation unit will remain in the
23314 cache if it is not referenced. A higher limit means that cached
23315 compilation units will be stored in memory longer, and more total
23316 memory will be used. Setting it to zero disables caching, which will
23317 slow down @value{GDBN} startup, but reduce memory consumption.
23319 @kindex maint set profile
23320 @kindex maint show profile
23321 @cindex profiling GDB
23322 @item maint set profile
23323 @itemx maint show profile
23324 Control profiling of @value{GDBN}.
23326 Profiling will be disabled until you use the @samp{maint set profile}
23327 command to enable it. When you enable profiling, the system will begin
23328 collecting timing and execution count data; when you disable profiling or
23329 exit @value{GDBN}, the results will be written to a log file. Remember that
23330 if you use profiling, @value{GDBN} will overwrite the profiling log file
23331 (often called @file{gmon.out}). If you have a record of important profiling
23332 data in a @file{gmon.out} file, be sure to move it to a safe location.
23334 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
23335 compiled with the @samp{-pg} compiler option.
23337 @kindex maint set linux-async
23338 @kindex maint show linux-async
23339 @cindex asynchronous support
23340 @item maint set linux-async
23341 @itemx maint show linux-async
23342 Control the GNU/Linux native asynchronous support of @value{GDBN}.
23344 GNU/Linux native asynchronous support will be disabled until you use
23345 the @samp{maint set linux-async} command to enable it.
23347 @kindex maint show-debug-regs
23348 @cindex x86 hardware debug registers
23349 @item maint show-debug-regs
23350 Control whether to show variables that mirror the x86 hardware debug
23351 registers. Use @code{ON} to enable, @code{OFF} to disable. If
23352 enabled, the debug registers values are shown when @value{GDBN} inserts or
23353 removes a hardware breakpoint or watchpoint, and when the inferior
23354 triggers a hardware-assisted breakpoint or watchpoint.
23356 @kindex maint space
23357 @cindex memory used by commands
23359 Control whether to display memory usage for each command. If set to a
23360 nonzero value, @value{GDBN} will display how much memory each command
23361 took, following the command's own output. This can also be requested
23362 by invoking @value{GDBN} with the @option{--statistics} command-line
23363 switch (@pxref{Mode Options}).
23366 @cindex time of command execution
23368 Control whether to display the execution time for each command. If
23369 set to a nonzero value, @value{GDBN} will display how much time it
23370 took to execute each command, following the command's own output.
23371 This can also be requested by invoking @value{GDBN} with the
23372 @option{--statistics} command-line switch (@pxref{Mode Options}).
23374 @kindex maint translate-address
23375 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
23376 Find the symbol stored at the location specified by the address
23377 @var{addr} and an optional section name @var{section}. If found,
23378 @value{GDBN} prints the name of the closest symbol and an offset from
23379 the symbol's location to the specified address. This is similar to
23380 the @code{info address} command (@pxref{Symbols}), except that this
23381 command also allows to find symbols in other sections.
23385 The following command is useful for non-interactive invocations of
23386 @value{GDBN}, such as in the test suite.
23389 @item set watchdog @var{nsec}
23390 @kindex set watchdog
23391 @cindex watchdog timer
23392 @cindex timeout for commands
23393 Set the maximum number of seconds @value{GDBN} will wait for the
23394 target operation to finish. If this time expires, @value{GDBN}
23395 reports and error and the command is aborted.
23397 @item show watchdog
23398 Show the current setting of the target wait timeout.
23401 @node Remote Protocol
23402 @appendix @value{GDBN} Remote Serial Protocol
23407 * Stop Reply Packets::
23408 * General Query Packets::
23409 * Register Packet Format::
23410 * Tracepoint Packets::
23411 * Host I/O Packets::
23414 * File-I/O Remote Protocol Extension::
23415 * Library List Format::
23416 * Memory Map Format::
23422 There may be occasions when you need to know something about the
23423 protocol---for example, if there is only one serial port to your target
23424 machine, you might want your program to do something special if it
23425 recognizes a packet meant for @value{GDBN}.
23427 In the examples below, @samp{->} and @samp{<-} are used to indicate
23428 transmitted and received data, respectively.
23430 @cindex protocol, @value{GDBN} remote serial
23431 @cindex serial protocol, @value{GDBN} remote
23432 @cindex remote serial protocol
23433 All @value{GDBN} commands and responses (other than acknowledgments) are
23434 sent as a @var{packet}. A @var{packet} is introduced with the character
23435 @samp{$}, the actual @var{packet-data}, and the terminating character
23436 @samp{#} followed by a two-digit @var{checksum}:
23439 @code{$}@var{packet-data}@code{#}@var{checksum}
23443 @cindex checksum, for @value{GDBN} remote
23445 The two-digit @var{checksum} is computed as the modulo 256 sum of all
23446 characters between the leading @samp{$} and the trailing @samp{#} (an
23447 eight bit unsigned checksum).
23449 Implementors should note that prior to @value{GDBN} 5.0 the protocol
23450 specification also included an optional two-digit @var{sequence-id}:
23453 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
23456 @cindex sequence-id, for @value{GDBN} remote
23458 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
23459 has never output @var{sequence-id}s. Stubs that handle packets added
23460 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
23462 @cindex acknowledgment, for @value{GDBN} remote
23463 When either the host or the target machine receives a packet, the first
23464 response expected is an acknowledgment: either @samp{+} (to indicate
23465 the package was received correctly) or @samp{-} (to request
23469 -> @code{$}@var{packet-data}@code{#}@var{checksum}
23474 The host (@value{GDBN}) sends @var{command}s, and the target (the
23475 debugging stub incorporated in your program) sends a @var{response}. In
23476 the case of step and continue @var{command}s, the response is only sent
23477 when the operation has completed (the target has again stopped).
23479 @var{packet-data} consists of a sequence of characters with the
23480 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
23483 @cindex remote protocol, field separator
23484 Fields within the packet should be separated using @samp{,} @samp{;} or
23485 @samp{:}. Except where otherwise noted all numbers are represented in
23486 @sc{hex} with leading zeros suppressed.
23488 Implementors should note that prior to @value{GDBN} 5.0, the character
23489 @samp{:} could not appear as the third character in a packet (as it
23490 would potentially conflict with the @var{sequence-id}).
23492 @cindex remote protocol, binary data
23493 @anchor{Binary Data}
23494 Binary data in most packets is encoded either as two hexadecimal
23495 digits per byte of binary data. This allowed the traditional remote
23496 protocol to work over connections which were only seven-bit clean.
23497 Some packets designed more recently assume an eight-bit clean
23498 connection, and use a more efficient encoding to send and receive
23501 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
23502 as an escape character. Any escaped byte is transmitted as the escape
23503 character followed by the original character XORed with @code{0x20}.
23504 For example, the byte @code{0x7d} would be transmitted as the two
23505 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
23506 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
23507 @samp{@}}) must always be escaped. Responses sent by the stub
23508 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
23509 is not interpreted as the start of a run-length encoded sequence
23512 Response @var{data} can be run-length encoded to save space.
23513 Run-length encoding replaces runs of identical characters with one
23514 instance of the repeated character, followed by a @samp{*} and a
23515 repeat count. The repeat count is itself sent encoded, to avoid
23516 binary characters in @var{data}: a value of @var{n} is sent as
23517 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
23518 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
23519 code 32) for a repeat count of 3. (This is because run-length
23520 encoding starts to win for counts 3 or more.) Thus, for example,
23521 @samp{0* } is a run-length encoding of ``0000'': the space character
23522 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
23525 The printable characters @samp{#} and @samp{$} or with a numeric value
23526 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
23527 seven repeats (@samp{$}) can be expanded using a repeat count of only
23528 five (@samp{"}). For example, @samp{00000000} can be encoded as
23531 The error response returned for some packets includes a two character
23532 error number. That number is not well defined.
23534 @cindex empty response, for unsupported packets
23535 For any @var{command} not supported by the stub, an empty response
23536 (@samp{$#00}) should be returned. That way it is possible to extend the
23537 protocol. A newer @value{GDBN} can tell if a packet is supported based
23540 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
23541 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
23547 The following table provides a complete list of all currently defined
23548 @var{command}s and their corresponding response @var{data}.
23549 @xref{File-I/O Remote Protocol Extension}, for details about the File
23550 I/O extension of the remote protocol.
23552 Each packet's description has a template showing the packet's overall
23553 syntax, followed by an explanation of the packet's meaning. We
23554 include spaces in some of the templates for clarity; these are not
23555 part of the packet's syntax. No @value{GDBN} packet uses spaces to
23556 separate its components. For example, a template like @samp{foo
23557 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
23558 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
23559 @var{baz}. @value{GDBN} does not transmit a space character between the
23560 @samp{foo} and the @var{bar}, or between the @var{bar} and the
23563 Note that all packet forms beginning with an upper- or lower-case
23564 letter, other than those described here, are reserved for future use.
23566 Here are the packet descriptions.
23571 @cindex @samp{!} packet
23572 @anchor{extended mode}
23573 Enable extended mode. In extended mode, the remote server is made
23574 persistent. The @samp{R} packet is used to restart the program being
23580 The remote target both supports and has enabled extended mode.
23584 @cindex @samp{?} packet
23585 Indicate the reason the target halted. The reply is the same as for
23589 @xref{Stop Reply Packets}, for the reply specifications.
23591 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
23592 @cindex @samp{A} packet
23593 Initialized @code{argv[]} array passed into program. @var{arglen}
23594 specifies the number of bytes in the hex encoded byte stream
23595 @var{arg}. See @code{gdbserver} for more details.
23600 The arguments were set.
23606 @cindex @samp{b} packet
23607 (Don't use this packet; its behavior is not well-defined.)
23608 Change the serial line speed to @var{baud}.
23610 JTC: @emph{When does the transport layer state change? When it's
23611 received, or after the ACK is transmitted. In either case, there are
23612 problems if the command or the acknowledgment packet is dropped.}
23614 Stan: @emph{If people really wanted to add something like this, and get
23615 it working for the first time, they ought to modify ser-unix.c to send
23616 some kind of out-of-band message to a specially-setup stub and have the
23617 switch happen "in between" packets, so that from remote protocol's point
23618 of view, nothing actually happened.}
23620 @item B @var{addr},@var{mode}
23621 @cindex @samp{B} packet
23622 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
23623 breakpoint at @var{addr}.
23625 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
23626 (@pxref{insert breakpoint or watchpoint packet}).
23628 @item c @r{[}@var{addr}@r{]}
23629 @cindex @samp{c} packet
23630 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
23631 resume at current address.
23634 @xref{Stop Reply Packets}, for the reply specifications.
23636 @item C @var{sig}@r{[};@var{addr}@r{]}
23637 @cindex @samp{C} packet
23638 Continue with signal @var{sig} (hex signal number). If
23639 @samp{;@var{addr}} is omitted, resume at same address.
23642 @xref{Stop Reply Packets}, for the reply specifications.
23645 @cindex @samp{d} packet
23648 Don't use this packet; instead, define a general set packet
23649 (@pxref{General Query Packets}).
23652 @cindex @samp{D} packet
23653 Detach @value{GDBN} from the remote system. Sent to the remote target
23654 before @value{GDBN} disconnects via the @code{detach} command.
23664 @item F @var{RC},@var{EE},@var{CF};@var{XX}
23665 @cindex @samp{F} packet
23666 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
23667 This is part of the File-I/O protocol extension. @xref{File-I/O
23668 Remote Protocol Extension}, for the specification.
23671 @anchor{read registers packet}
23672 @cindex @samp{g} packet
23673 Read general registers.
23677 @item @var{XX@dots{}}
23678 Each byte of register data is described by two hex digits. The bytes
23679 with the register are transmitted in target byte order. The size of
23680 each register and their position within the @samp{g} packet are
23681 determined by the @value{GDBN} internal gdbarch functions
23682 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
23683 specification of several standard @samp{g} packets is specified below.
23688 @item G @var{XX@dots{}}
23689 @cindex @samp{G} packet
23690 Write general registers. @xref{read registers packet}, for a
23691 description of the @var{XX@dots{}} data.
23701 @item H @var{c} @var{t}
23702 @cindex @samp{H} packet
23703 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
23704 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
23705 should be @samp{c} for step and continue operations, @samp{g} for other
23706 operations. The thread designator @var{t} may be @samp{-1}, meaning all
23707 the threads, a thread number, or @samp{0} which means pick any thread.
23718 @c 'H': How restrictive (or permissive) is the thread model. If a
23719 @c thread is selected and stopped, are other threads allowed
23720 @c to continue to execute? As I mentioned above, I think the
23721 @c semantics of each command when a thread is selected must be
23722 @c described. For example:
23724 @c 'g': If the stub supports threads and a specific thread is
23725 @c selected, returns the register block from that thread;
23726 @c otherwise returns current registers.
23728 @c 'G' If the stub supports threads and a specific thread is
23729 @c selected, sets the registers of the register block of
23730 @c that thread; otherwise sets current registers.
23732 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
23733 @anchor{cycle step packet}
23734 @cindex @samp{i} packet
23735 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
23736 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
23737 step starting at that address.
23740 @cindex @samp{I} packet
23741 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
23745 @cindex @samp{k} packet
23748 FIXME: @emph{There is no description of how to operate when a specific
23749 thread context has been selected (i.e.@: does 'k' kill only that
23752 @item m @var{addr},@var{length}
23753 @cindex @samp{m} packet
23754 Read @var{length} bytes of memory starting at address @var{addr}.
23755 Note that @var{addr} may not be aligned to any particular boundary.
23757 The stub need not use any particular size or alignment when gathering
23758 data from memory for the response; even if @var{addr} is word-aligned
23759 and @var{length} is a multiple of the word size, the stub is free to
23760 use byte accesses, or not. For this reason, this packet may not be
23761 suitable for accessing memory-mapped I/O devices.
23762 @cindex alignment of remote memory accesses
23763 @cindex size of remote memory accesses
23764 @cindex memory, alignment and size of remote accesses
23768 @item @var{XX@dots{}}
23769 Memory contents; each byte is transmitted as a two-digit hexadecimal
23770 number. The reply may contain fewer bytes than requested if the
23771 server was able to read only part of the region of memory.
23776 @item M @var{addr},@var{length}:@var{XX@dots{}}
23777 @cindex @samp{M} packet
23778 Write @var{length} bytes of memory starting at address @var{addr}.
23779 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
23780 hexadecimal number.
23787 for an error (this includes the case where only part of the data was
23792 @cindex @samp{p} packet
23793 Read the value of register @var{n}; @var{n} is in hex.
23794 @xref{read registers packet}, for a description of how the returned
23795 register value is encoded.
23799 @item @var{XX@dots{}}
23800 the register's value
23804 Indicating an unrecognized @var{query}.
23807 @item P @var{n@dots{}}=@var{r@dots{}}
23808 @anchor{write register packet}
23809 @cindex @samp{P} packet
23810 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
23811 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
23812 digits for each byte in the register (target byte order).
23822 @item q @var{name} @var{params}@dots{}
23823 @itemx Q @var{name} @var{params}@dots{}
23824 @cindex @samp{q} packet
23825 @cindex @samp{Q} packet
23826 General query (@samp{q}) and set (@samp{Q}). These packets are
23827 described fully in @ref{General Query Packets}.
23830 @cindex @samp{r} packet
23831 Reset the entire system.
23833 Don't use this packet; use the @samp{R} packet instead.
23836 @cindex @samp{R} packet
23837 Restart the program being debugged. @var{XX}, while needed, is ignored.
23838 This packet is only available in extended mode (@pxref{extended mode}).
23840 The @samp{R} packet has no reply.
23842 @item s @r{[}@var{addr}@r{]}
23843 @cindex @samp{s} packet
23844 Single step. @var{addr} is the address at which to resume. If
23845 @var{addr} is omitted, resume at same address.
23848 @xref{Stop Reply Packets}, for the reply specifications.
23850 @item S @var{sig}@r{[};@var{addr}@r{]}
23851 @anchor{step with signal packet}
23852 @cindex @samp{S} packet
23853 Step with signal. This is analogous to the @samp{C} packet, but
23854 requests a single-step, rather than a normal resumption of execution.
23857 @xref{Stop Reply Packets}, for the reply specifications.
23859 @item t @var{addr}:@var{PP},@var{MM}
23860 @cindex @samp{t} packet
23861 Search backwards starting at address @var{addr} for a match with pattern
23862 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
23863 @var{addr} must be at least 3 digits.
23866 @cindex @samp{T} packet
23867 Find out if the thread XX is alive.
23872 thread is still alive
23878 Packets starting with @samp{v} are identified by a multi-letter name,
23879 up to the first @samp{;} or @samp{?} (or the end of the packet).
23881 @item vAttach;@var{pid}
23882 @cindex @samp{vAttach} packet
23883 Attach to a new process with the specified process ID. @var{pid} is a
23884 hexadecimal integer identifying the process. If the stub is currently
23885 controlling a process, it is killed. The attached process is stopped.
23887 This packet is only available in extended mode (@pxref{extended mode}).
23893 @item @r{Any stop packet}
23894 for success (@pxref{Stop Reply Packets})
23897 @item vCont@r{[};@var{action}@r{[}:@var{tid}@r{]]}@dots{}
23898 @cindex @samp{vCont} packet
23899 Resume the inferior, specifying different actions for each thread.
23900 If an action is specified with no @var{tid}, then it is applied to any
23901 threads that don't have a specific action specified; if no default action is
23902 specified then other threads should remain stopped. Specifying multiple
23903 default actions is an error; specifying no actions is also an error.
23904 Thread IDs are specified in hexadecimal. Currently supported actions are:
23910 Continue with signal @var{sig}. @var{sig} should be two hex digits.
23914 Step with signal @var{sig}. @var{sig} should be two hex digits.
23917 The optional @var{addr} argument normally associated with these packets is
23918 not supported in @samp{vCont}.
23921 @xref{Stop Reply Packets}, for the reply specifications.
23924 @cindex @samp{vCont?} packet
23925 Request a list of actions supported by the @samp{vCont} packet.
23929 @item vCont@r{[};@var{action}@dots{}@r{]}
23930 The @samp{vCont} packet is supported. Each @var{action} is a supported
23931 command in the @samp{vCont} packet.
23933 The @samp{vCont} packet is not supported.
23936 @item vFile:@var{operation}:@var{parameter}@dots{}
23937 @cindex @samp{vFile} packet
23938 Perform a file operation on the target system. For details,
23939 see @ref{Host I/O Packets}.
23941 @item vFlashErase:@var{addr},@var{length}
23942 @cindex @samp{vFlashErase} packet
23943 Direct the stub to erase @var{length} bytes of flash starting at
23944 @var{addr}. The region may enclose any number of flash blocks, but
23945 its start and end must fall on block boundaries, as indicated by the
23946 flash block size appearing in the memory map (@pxref{Memory Map
23947 Format}). @value{GDBN} groups flash memory programming operations
23948 together, and sends a @samp{vFlashDone} request after each group; the
23949 stub is allowed to delay erase operation until the @samp{vFlashDone}
23950 packet is received.
23960 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
23961 @cindex @samp{vFlashWrite} packet
23962 Direct the stub to write data to flash address @var{addr}. The data
23963 is passed in binary form using the same encoding as for the @samp{X}
23964 packet (@pxref{Binary Data}). The memory ranges specified by
23965 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
23966 not overlap, and must appear in order of increasing addresses
23967 (although @samp{vFlashErase} packets for higher addresses may already
23968 have been received; the ordering is guaranteed only between
23969 @samp{vFlashWrite} packets). If a packet writes to an address that was
23970 neither erased by a preceding @samp{vFlashErase} packet nor by some other
23971 target-specific method, the results are unpredictable.
23979 for vFlashWrite addressing non-flash memory
23985 @cindex @samp{vFlashDone} packet
23986 Indicate to the stub that flash programming operation is finished.
23987 The stub is permitted to delay or batch the effects of a group of
23988 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
23989 @samp{vFlashDone} packet is received. The contents of the affected
23990 regions of flash memory are unpredictable until the @samp{vFlashDone}
23991 request is completed.
23993 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
23994 @cindex @samp{vRun} packet
23995 Run the program @var{filename}, passing it each @var{argument} on its
23996 command line. The file and arguments are hex-encoded strings. If
23997 @var{filename} is an empty string, the stub may use a default program
23998 (e.g.@: the last program run). The program is created in the stopped
23999 state. If the stub is currently controlling a process, it is killed.
24001 This packet is only available in extended mode (@pxref{extended mode}).
24007 @item @r{Any stop packet}
24008 for success (@pxref{Stop Reply Packets})
24011 @item X @var{addr},@var{length}:@var{XX@dots{}}
24013 @cindex @samp{X} packet
24014 Write data to memory, where the data is transmitted in binary.
24015 @var{addr} is address, @var{length} is number of bytes,
24016 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
24026 @item z @var{type},@var{addr},@var{length}
24027 @itemx Z @var{type},@var{addr},@var{length}
24028 @anchor{insert breakpoint or watchpoint packet}
24029 @cindex @samp{z} packet
24030 @cindex @samp{Z} packets
24031 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
24032 watchpoint starting at address @var{address} and covering the next
24033 @var{length} bytes.
24035 Each breakpoint and watchpoint packet @var{type} is documented
24038 @emph{Implementation notes: A remote target shall return an empty string
24039 for an unrecognized breakpoint or watchpoint packet @var{type}. A
24040 remote target shall support either both or neither of a given
24041 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
24042 avoid potential problems with duplicate packets, the operations should
24043 be implemented in an idempotent way.}
24045 @item z0,@var{addr},@var{length}
24046 @itemx Z0,@var{addr},@var{length}
24047 @cindex @samp{z0} packet
24048 @cindex @samp{Z0} packet
24049 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
24050 @var{addr} of size @var{length}.
24052 A memory breakpoint is implemented by replacing the instruction at
24053 @var{addr} with a software breakpoint or trap instruction. The
24054 @var{length} is used by targets that indicates the size of the
24055 breakpoint (in bytes) that should be inserted (e.g., the @sc{arm} and
24056 @sc{mips} can insert either a 2 or 4 byte breakpoint).
24058 @emph{Implementation note: It is possible for a target to copy or move
24059 code that contains memory breakpoints (e.g., when implementing
24060 overlays). The behavior of this packet, in the presence of such a
24061 target, is not defined.}
24073 @item z1,@var{addr},@var{length}
24074 @itemx Z1,@var{addr},@var{length}
24075 @cindex @samp{z1} packet
24076 @cindex @samp{Z1} packet
24077 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
24078 address @var{addr} of size @var{length}.
24080 A hardware breakpoint is implemented using a mechanism that is not
24081 dependant on being able to modify the target's memory.
24083 @emph{Implementation note: A hardware breakpoint is not affected by code
24096 @item z2,@var{addr},@var{length}
24097 @itemx Z2,@var{addr},@var{length}
24098 @cindex @samp{z2} packet
24099 @cindex @samp{Z2} packet
24100 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint.
24112 @item z3,@var{addr},@var{length}
24113 @itemx Z3,@var{addr},@var{length}
24114 @cindex @samp{z3} packet
24115 @cindex @samp{Z3} packet
24116 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint.
24128 @item z4,@var{addr},@var{length}
24129 @itemx Z4,@var{addr},@var{length}
24130 @cindex @samp{z4} packet
24131 @cindex @samp{Z4} packet
24132 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint.
24146 @node Stop Reply Packets
24147 @section Stop Reply Packets
24148 @cindex stop reply packets
24150 The @samp{C}, @samp{c}, @samp{S}, @samp{s} and @samp{?} packets can
24151 receive any of the below as a reply. In the case of the @samp{C},
24152 @samp{c}, @samp{S} and @samp{s} packets, that reply is only returned
24153 when the target halts. In the below the exact meaning of @dfn{signal
24154 number} is defined by the header @file{include/gdb/signals.h} in the
24155 @value{GDBN} source code.
24157 As in the description of request packets, we include spaces in the
24158 reply templates for clarity; these are not part of the reply packet's
24159 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
24165 The program received signal number @var{AA} (a two-digit hexadecimal
24166 number). This is equivalent to a @samp{T} response with no
24167 @var{n}:@var{r} pairs.
24169 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
24170 @cindex @samp{T} packet reply
24171 The program received signal number @var{AA} (a two-digit hexadecimal
24172 number). This is equivalent to an @samp{S} response, except that the
24173 @samp{@var{n}:@var{r}} pairs can carry values of important registers
24174 and other information directly in the stop reply packet, reducing
24175 round-trip latency. Single-step and breakpoint traps are reported
24176 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
24180 If @var{n} is a hexadecimal number, it is a register number, and the
24181 corresponding @var{r} gives that register's value. @var{r} is a
24182 series of bytes in target byte order, with each byte given by a
24183 two-digit hex number.
24186 If @var{n} is @samp{thread}, then @var{r} is the thread process ID, in
24190 If @var{n} is a recognized @dfn{stop reason}, it describes a more
24191 specific event that stopped the target. The currently defined stop
24192 reasons are listed below. @var{aa} should be @samp{05}, the trap
24193 signal. At most one stop reason should be present.
24196 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
24197 and go on to the next; this allows us to extend the protocol in the
24201 The currently defined stop reasons are:
24207 The packet indicates a watchpoint hit, and @var{r} is the data address, in
24210 @cindex shared library events, remote reply
24212 The packet indicates that the loaded libraries have changed.
24213 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
24214 list of loaded libraries. @var{r} is ignored.
24218 The process exited, and @var{AA} is the exit status. This is only
24219 applicable to certain targets.
24222 The process terminated with signal @var{AA}.
24224 @item O @var{XX}@dots{}
24225 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
24226 written as the program's console output. This can happen at any time
24227 while the program is running and the debugger should continue to wait
24228 for @samp{W}, @samp{T}, etc.
24230 @item F @var{call-id},@var{parameter}@dots{}
24231 @var{call-id} is the identifier which says which host system call should
24232 be called. This is just the name of the function. Translation into the
24233 correct system call is only applicable as it's defined in @value{GDBN}.
24234 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
24237 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
24238 this very system call.
24240 The target replies with this packet when it expects @value{GDBN} to
24241 call a host system call on behalf of the target. @value{GDBN} replies
24242 with an appropriate @samp{F} packet and keeps up waiting for the next
24243 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
24244 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
24245 Protocol Extension}, for more details.
24249 @node General Query Packets
24250 @section General Query Packets
24251 @cindex remote query requests
24253 Packets starting with @samp{q} are @dfn{general query packets};
24254 packets starting with @samp{Q} are @dfn{general set packets}. General
24255 query and set packets are a semi-unified form for retrieving and
24256 sending information to and from the stub.
24258 The initial letter of a query or set packet is followed by a name
24259 indicating what sort of thing the packet applies to. For example,
24260 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
24261 definitions with the stub. These packet names follow some
24266 The name must not contain commas, colons or semicolons.
24268 Most @value{GDBN} query and set packets have a leading upper case
24271 The names of custom vendor packets should use a company prefix, in
24272 lower case, followed by a period. For example, packets designed at
24273 the Acme Corporation might begin with @samp{qacme.foo} (for querying
24274 foos) or @samp{Qacme.bar} (for setting bars).
24277 The name of a query or set packet should be separated from any
24278 parameters by a @samp{:}; the parameters themselves should be
24279 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
24280 full packet name, and check for a separator or the end of the packet,
24281 in case two packet names share a common prefix. New packets should not begin
24282 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
24283 packets predate these conventions, and have arguments without any terminator
24284 for the packet name; we suspect they are in widespread use in places that
24285 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
24286 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
24289 Like the descriptions of the other packets, each description here
24290 has a template showing the packet's overall syntax, followed by an
24291 explanation of the packet's meaning. We include spaces in some of the
24292 templates for clarity; these are not part of the packet's syntax. No
24293 @value{GDBN} packet uses spaces to separate its components.
24295 Here are the currently defined query and set packets:
24300 @cindex current thread, remote request
24301 @cindex @samp{qC} packet
24302 Return the current thread id.
24307 Where @var{pid} is an unsigned hexadecimal process id.
24308 @item @r{(anything else)}
24309 Any other reply implies the old pid.
24312 @item qCRC:@var{addr},@var{length}
24313 @cindex CRC of memory block, remote request
24314 @cindex @samp{qCRC} packet
24315 Compute the CRC checksum of a block of memory.
24319 An error (such as memory fault)
24320 @item C @var{crc32}
24321 The specified memory region's checksum is @var{crc32}.
24325 @itemx qsThreadInfo
24326 @cindex list active threads, remote request
24327 @cindex @samp{qfThreadInfo} packet
24328 @cindex @samp{qsThreadInfo} packet
24329 Obtain a list of all active thread ids from the target (OS). Since there
24330 may be too many active threads to fit into one reply packet, this query
24331 works iteratively: it may require more than one query/reply sequence to
24332 obtain the entire list of threads. The first query of the sequence will
24333 be the @samp{qfThreadInfo} query; subsequent queries in the
24334 sequence will be the @samp{qsThreadInfo} query.
24336 NOTE: This packet replaces the @samp{qL} query (see below).
24342 @item m @var{id},@var{id}@dots{}
24343 a comma-separated list of thread ids
24345 (lower case letter @samp{L}) denotes end of list.
24348 In response to each query, the target will reply with a list of one or
24349 more thread ids, in big-endian unsigned hex, separated by commas.
24350 @value{GDBN} will respond to each reply with a request for more thread
24351 ids (using the @samp{qs} form of the query), until the target responds
24352 with @samp{l} (lower-case el, for @dfn{last}).
24354 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
24355 @cindex get thread-local storage address, remote request
24356 @cindex @samp{qGetTLSAddr} packet
24357 Fetch the address associated with thread local storage specified
24358 by @var{thread-id}, @var{offset}, and @var{lm}.
24360 @var{thread-id} is the (big endian, hex encoded) thread id associated with the
24361 thread for which to fetch the TLS address.
24363 @var{offset} is the (big endian, hex encoded) offset associated with the
24364 thread local variable. (This offset is obtained from the debug
24365 information associated with the variable.)
24367 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
24368 the load module associated with the thread local storage. For example,
24369 a @sc{gnu}/Linux system will pass the link map address of the shared
24370 object associated with the thread local storage under consideration.
24371 Other operating environments may choose to represent the load module
24372 differently, so the precise meaning of this parameter will vary.
24376 @item @var{XX}@dots{}
24377 Hex encoded (big endian) bytes representing the address of the thread
24378 local storage requested.
24381 An error occurred. @var{nn} are hex digits.
24384 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
24387 @item qL @var{startflag} @var{threadcount} @var{nextthread}
24388 Obtain thread information from RTOS. Where: @var{startflag} (one hex
24389 digit) is one to indicate the first query and zero to indicate a
24390 subsequent query; @var{threadcount} (two hex digits) is the maximum
24391 number of threads the response packet can contain; and @var{nextthread}
24392 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
24393 returned in the response as @var{argthread}.
24395 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
24399 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
24400 Where: @var{count} (two hex digits) is the number of threads being
24401 returned; @var{done} (one hex digit) is zero to indicate more threads
24402 and one indicates no further threads; @var{argthreadid} (eight hex
24403 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
24404 is a sequence of thread IDs from the target. @var{threadid} (eight hex
24405 digits). See @code{remote.c:parse_threadlist_response()}.
24409 @cindex section offsets, remote request
24410 @cindex @samp{qOffsets} packet
24411 Get section offsets that the target used when relocating the downloaded
24416 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
24417 Relocate the @code{Text} section by @var{xxx} from its original address.
24418 Relocate the @code{Data} section by @var{yyy} from its original address.
24419 If the object file format provides segment information (e.g.@: @sc{elf}
24420 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
24421 segments by the supplied offsets.
24423 @emph{Note: while a @code{Bss} offset may be included in the response,
24424 @value{GDBN} ignores this and instead applies the @code{Data} offset
24425 to the @code{Bss} section.}
24427 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
24428 Relocate the first segment of the object file, which conventionally
24429 contains program code, to a starting address of @var{xxx}. If
24430 @samp{DataSeg} is specified, relocate the second segment, which
24431 conventionally contains modifiable data, to a starting address of
24432 @var{yyy}. @value{GDBN} will report an error if the object file
24433 does not contain segment information, or does not contain at least
24434 as many segments as mentioned in the reply. Extra segments are
24435 kept at fixed offsets relative to the last relocated segment.
24438 @item qP @var{mode} @var{threadid}
24439 @cindex thread information, remote request
24440 @cindex @samp{qP} packet
24441 Returns information on @var{threadid}. Where: @var{mode} is a hex
24442 encoded 32 bit mode; @var{threadid} is a hex encoded 64 bit thread ID.
24444 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
24447 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
24449 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
24450 @cindex pass signals to inferior, remote request
24451 @cindex @samp{QPassSignals} packet
24452 @anchor{QPassSignals}
24453 Each listed @var{signal} should be passed directly to the inferior process.
24454 Signals are numbered identically to continue packets and stop replies
24455 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
24456 strictly greater than the previous item. These signals do not need to stop
24457 the inferior, or be reported to @value{GDBN}. All other signals should be
24458 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
24459 combine; any earlier @samp{QPassSignals} list is completely replaced by the
24460 new list. This packet improves performance when using @samp{handle
24461 @var{signal} nostop noprint pass}.
24466 The request succeeded.
24469 An error occurred. @var{nn} are hex digits.
24472 An empty reply indicates that @samp{QPassSignals} is not supported by
24476 Use of this packet is controlled by the @code{set remote pass-signals}
24477 command (@pxref{Remote Configuration, set remote pass-signals}).
24478 This packet is not probed by default; the remote stub must request it,
24479 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
24481 @item qRcmd,@var{command}
24482 @cindex execute remote command, remote request
24483 @cindex @samp{qRcmd} packet
24484 @var{command} (hex encoded) is passed to the local interpreter for
24485 execution. Invalid commands should be reported using the output
24486 string. Before the final result packet, the target may also respond
24487 with a number of intermediate @samp{O@var{output}} console output
24488 packets. @emph{Implementors should note that providing access to a
24489 stubs's interpreter may have security implications}.
24494 A command response with no output.
24496 A command response with the hex encoded output string @var{OUTPUT}.
24498 Indicate a badly formed request.
24500 An empty reply indicates that @samp{qRcmd} is not recognized.
24503 (Note that the @code{qRcmd} packet's name is separated from the
24504 command by a @samp{,}, not a @samp{:}, contrary to the naming
24505 conventions above. Please don't use this packet as a model for new
24508 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
24509 @cindex supported packets, remote query
24510 @cindex features of the remote protocol
24511 @cindex @samp{qSupported} packet
24512 @anchor{qSupported}
24513 Tell the remote stub about features supported by @value{GDBN}, and
24514 query the stub for features it supports. This packet allows
24515 @value{GDBN} and the remote stub to take advantage of each others'
24516 features. @samp{qSupported} also consolidates multiple feature probes
24517 at startup, to improve @value{GDBN} performance---a single larger
24518 packet performs better than multiple smaller probe packets on
24519 high-latency links. Some features may enable behavior which must not
24520 be on by default, e.g.@: because it would confuse older clients or
24521 stubs. Other features may describe packets which could be
24522 automatically probed for, but are not. These features must be
24523 reported before @value{GDBN} will use them. This ``default
24524 unsupported'' behavior is not appropriate for all packets, but it
24525 helps to keep the initial connection time under control with new
24526 versions of @value{GDBN} which support increasing numbers of packets.
24530 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
24531 The stub supports or does not support each returned @var{stubfeature},
24532 depending on the form of each @var{stubfeature} (see below for the
24535 An empty reply indicates that @samp{qSupported} is not recognized,
24536 or that no features needed to be reported to @value{GDBN}.
24539 The allowed forms for each feature (either a @var{gdbfeature} in the
24540 @samp{qSupported} packet, or a @var{stubfeature} in the response)
24544 @item @var{name}=@var{value}
24545 The remote protocol feature @var{name} is supported, and associated
24546 with the specified @var{value}. The format of @var{value} depends
24547 on the feature, but it must not include a semicolon.
24549 The remote protocol feature @var{name} is supported, and does not
24550 need an associated value.
24552 The remote protocol feature @var{name} is not supported.
24554 The remote protocol feature @var{name} may be supported, and
24555 @value{GDBN} should auto-detect support in some other way when it is
24556 needed. This form will not be used for @var{gdbfeature} notifications,
24557 but may be used for @var{stubfeature} responses.
24560 Whenever the stub receives a @samp{qSupported} request, the
24561 supplied set of @value{GDBN} features should override any previous
24562 request. This allows @value{GDBN} to put the stub in a known
24563 state, even if the stub had previously been communicating with
24564 a different version of @value{GDBN}.
24566 No values of @var{gdbfeature} (for the packet sent by @value{GDBN})
24567 are defined yet. Stubs should ignore any unknown values for
24568 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
24569 packet supports receiving packets of unlimited length (earlier
24570 versions of @value{GDBN} may reject overly long responses). Values
24571 for @var{gdbfeature} may be defined in the future to let the stub take
24572 advantage of new features in @value{GDBN}, e.g.@: incompatible
24573 improvements in the remote protocol---support for unlimited length
24574 responses would be a @var{gdbfeature} example, if it were not implied by
24575 the @samp{qSupported} query. The stub's reply should be independent
24576 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
24577 describes all the features it supports, and then the stub replies with
24578 all the features it supports.
24580 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
24581 responses, as long as each response uses one of the standard forms.
24583 Some features are flags. A stub which supports a flag feature
24584 should respond with a @samp{+} form response. Other features
24585 require values, and the stub should respond with an @samp{=}
24588 Each feature has a default value, which @value{GDBN} will use if
24589 @samp{qSupported} is not available or if the feature is not mentioned
24590 in the @samp{qSupported} response. The default values are fixed; a
24591 stub is free to omit any feature responses that match the defaults.
24593 Not all features can be probed, but for those which can, the probing
24594 mechanism is useful: in some cases, a stub's internal
24595 architecture may not allow the protocol layer to know some information
24596 about the underlying target in advance. This is especially common in
24597 stubs which may be configured for multiple targets.
24599 These are the currently defined stub features and their properties:
24601 @multitable @columnfractions 0.35 0.2 0.12 0.2
24602 @c NOTE: The first row should be @headitem, but we do not yet require
24603 @c a new enough version of Texinfo (4.7) to use @headitem.
24605 @tab Value Required
24609 @item @samp{PacketSize}
24614 @item @samp{qXfer:auxv:read}
24619 @item @samp{qXfer:features:read}
24624 @item @samp{qXfer:libraries:read}
24629 @item @samp{qXfer:memory-map:read}
24634 @item @samp{qXfer:spu:read}
24639 @item @samp{qXfer:spu:write}
24644 @item @samp{QPassSignals}
24651 These are the currently defined stub features, in more detail:
24654 @cindex packet size, remote protocol
24655 @item PacketSize=@var{bytes}
24656 The remote stub can accept packets up to at least @var{bytes} in
24657 length. @value{GDBN} will send packets up to this size for bulk
24658 transfers, and will never send larger packets. This is a limit on the
24659 data characters in the packet, including the frame and checksum.
24660 There is no trailing NUL byte in a remote protocol packet; if the stub
24661 stores packets in a NUL-terminated format, it should allow an extra
24662 byte in its buffer for the NUL. If this stub feature is not supported,
24663 @value{GDBN} guesses based on the size of the @samp{g} packet response.
24665 @item qXfer:auxv:read
24666 The remote stub understands the @samp{qXfer:auxv:read} packet
24667 (@pxref{qXfer auxiliary vector read}).
24669 @item qXfer:features:read
24670 The remote stub understands the @samp{qXfer:features:read} packet
24671 (@pxref{qXfer target description read}).
24673 @item qXfer:libraries:read
24674 The remote stub understands the @samp{qXfer:libraries:read} packet
24675 (@pxref{qXfer library list read}).
24677 @item qXfer:memory-map:read
24678 The remote stub understands the @samp{qXfer:memory-map:read} packet
24679 (@pxref{qXfer memory map read}).
24681 @item qXfer:spu:read
24682 The remote stub understands the @samp{qXfer:spu:read} packet
24683 (@pxref{qXfer spu read}).
24685 @item qXfer:spu:write
24686 The remote stub understands the @samp{qXfer:spu:write} packet
24687 (@pxref{qXfer spu write}).
24690 The remote stub understands the @samp{QPassSignals} packet
24691 (@pxref{QPassSignals}).
24696 @cindex symbol lookup, remote request
24697 @cindex @samp{qSymbol} packet
24698 Notify the target that @value{GDBN} is prepared to serve symbol lookup
24699 requests. Accept requests from the target for the values of symbols.
24704 The target does not need to look up any (more) symbols.
24705 @item qSymbol:@var{sym_name}
24706 The target requests the value of symbol @var{sym_name} (hex encoded).
24707 @value{GDBN} may provide the value by using the
24708 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
24712 @item qSymbol:@var{sym_value}:@var{sym_name}
24713 Set the value of @var{sym_name} to @var{sym_value}.
24715 @var{sym_name} (hex encoded) is the name of a symbol whose value the
24716 target has previously requested.
24718 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
24719 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
24725 The target does not need to look up any (more) symbols.
24726 @item qSymbol:@var{sym_name}
24727 The target requests the value of a new symbol @var{sym_name} (hex
24728 encoded). @value{GDBN} will continue to supply the values of symbols
24729 (if available), until the target ceases to request them.
24734 @xref{Tracepoint Packets}.
24736 @item qThreadExtraInfo,@var{id}
24737 @cindex thread attributes info, remote request
24738 @cindex @samp{qThreadExtraInfo} packet
24739 Obtain a printable string description of a thread's attributes from
24740 the target OS. @var{id} is a thread-id in big-endian hex. This
24741 string may contain anything that the target OS thinks is interesting
24742 for @value{GDBN} to tell the user about the thread. The string is
24743 displayed in @value{GDBN}'s @code{info threads} display. Some
24744 examples of possible thread extra info strings are @samp{Runnable}, or
24745 @samp{Blocked on Mutex}.
24749 @item @var{XX}@dots{}
24750 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
24751 comprising the printable string containing the extra information about
24752 the thread's attributes.
24755 (Note that the @code{qThreadExtraInfo} packet's name is separated from
24756 the command by a @samp{,}, not a @samp{:}, contrary to the naming
24757 conventions above. Please don't use this packet as a model for new
24765 @xref{Tracepoint Packets}.
24767 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
24768 @cindex read special object, remote request
24769 @cindex @samp{qXfer} packet
24770 @anchor{qXfer read}
24771 Read uninterpreted bytes from the target's special data area
24772 identified by the keyword @var{object}. Request @var{length} bytes
24773 starting at @var{offset} bytes into the data. The content and
24774 encoding of @var{annex} is specific to @var{object}; it can supply
24775 additional details about what data to access.
24777 Here are the specific requests of this form defined so far. All
24778 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
24779 formats, listed below.
24782 @item qXfer:auxv:read::@var{offset},@var{length}
24783 @anchor{qXfer auxiliary vector read}
24784 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
24785 auxiliary vector}. Note @var{annex} must be empty.
24787 This packet is not probed by default; the remote stub must request it,
24788 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
24790 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
24791 @anchor{qXfer target description read}
24792 Access the @dfn{target description}. @xref{Target Descriptions}. The
24793 annex specifies which XML document to access. The main description is
24794 always loaded from the @samp{target.xml} annex.
24796 This packet is not probed by default; the remote stub must request it,
24797 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
24799 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
24800 @anchor{qXfer library list read}
24801 Access the target's list of loaded libraries. @xref{Library List Format}.
24802 The annex part of the generic @samp{qXfer} packet must be empty
24803 (@pxref{qXfer read}).
24805 Targets which maintain a list of libraries in the program's memory do
24806 not need to implement this packet; it is designed for platforms where
24807 the operating system manages the list of loaded libraries.
24809 This packet is not probed by default; the remote stub must request it,
24810 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
24812 @item qXfer:memory-map:read::@var{offset},@var{length}
24813 @anchor{qXfer memory map read}
24814 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
24815 annex part of the generic @samp{qXfer} packet must be empty
24816 (@pxref{qXfer read}).
24818 This packet is not probed by default; the remote stub must request it,
24819 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
24821 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
24822 @anchor{qXfer spu read}
24823 Read contents of an @code{spufs} file on the target system. The
24824 annex specifies which file to read; it must be of the form
24825 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
24826 in the target process, and @var{name} identifes the @code{spufs} file
24827 in that context to be accessed.
24829 This packet is not probed by default; the remote stub must request it,
24830 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
24836 Data @var{data} (@pxref{Binary Data}) has been read from the
24837 target. There may be more data at a higher address (although
24838 it is permitted to return @samp{m} even for the last valid
24839 block of data, as long as at least one byte of data was read).
24840 @var{data} may have fewer bytes than the @var{length} in the
24844 Data @var{data} (@pxref{Binary Data}) has been read from the target.
24845 There is no more data to be read. @var{data} may have fewer bytes
24846 than the @var{length} in the request.
24849 The @var{offset} in the request is at the end of the data.
24850 There is no more data to be read.
24853 The request was malformed, or @var{annex} was invalid.
24856 The offset was invalid, or there was an error encountered reading the data.
24857 @var{nn} is a hex-encoded @code{errno} value.
24860 An empty reply indicates the @var{object} string was not recognized by
24861 the stub, or that the object does not support reading.
24864 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
24865 @cindex write data into object, remote request
24866 Write uninterpreted bytes into the target's special data area
24867 identified by the keyword @var{object}, starting at @var{offset} bytes
24868 into the data. @var{data}@dots{} is the binary-encoded data
24869 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
24870 is specific to @var{object}; it can supply additional details about what data
24873 Here are the specific requests of this form defined so far. All
24874 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
24875 formats, listed below.
24878 @item qXfer:@var{spu}:write:@var{annex}:@var{offset}:@var{data}@dots{}
24879 @anchor{qXfer spu write}
24880 Write @var{data} to an @code{spufs} file on the target system. The
24881 annex specifies which file to write; it must be of the form
24882 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
24883 in the target process, and @var{name} identifes the @code{spufs} file
24884 in that context to be accessed.
24886 This packet is not probed by default; the remote stub must request it,
24887 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
24893 @var{nn} (hex encoded) is the number of bytes written.
24894 This may be fewer bytes than supplied in the request.
24897 The request was malformed, or @var{annex} was invalid.
24900 The offset was invalid, or there was an error encountered writing the data.
24901 @var{nn} is a hex-encoded @code{errno} value.
24904 An empty reply indicates the @var{object} string was not
24905 recognized by the stub, or that the object does not support writing.
24908 @item qXfer:@var{object}:@var{operation}:@dots{}
24909 Requests of this form may be added in the future. When a stub does
24910 not recognize the @var{object} keyword, or its support for
24911 @var{object} does not recognize the @var{operation} keyword, the stub
24912 must respond with an empty packet.
24916 @node Register Packet Format
24917 @section Register Packet Format
24919 The following @code{g}/@code{G} packets have previously been defined.
24920 In the below, some thirty-two bit registers are transferred as
24921 sixty-four bits. Those registers should be zero/sign extended (which?)
24922 to fill the space allocated. Register bytes are transferred in target
24923 byte order. The two nibbles within a register byte are transferred
24924 most-significant - least-significant.
24930 All registers are transferred as thirty-two bit quantities in the order:
24931 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
24932 registers; fsr; fir; fp.
24936 All registers are transferred as sixty-four bit quantities (including
24937 thirty-two bit registers such as @code{sr}). The ordering is the same
24942 @node Tracepoint Packets
24943 @section Tracepoint Packets
24944 @cindex tracepoint packets
24945 @cindex packets, tracepoint
24947 Here we describe the packets @value{GDBN} uses to implement
24948 tracepoints (@pxref{Tracepoints}).
24952 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}@r{[}-@r{]}
24953 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
24954 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
24955 the tracepoint is disabled. @var{step} is the tracepoint's step
24956 count, and @var{pass} is its pass count. If the trailing @samp{-} is
24957 present, further @samp{QTDP} packets will follow to specify this
24958 tracepoint's actions.
24963 The packet was understood and carried out.
24965 The packet was not recognized.
24968 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
24969 Define actions to be taken when a tracepoint is hit. @var{n} and
24970 @var{addr} must be the same as in the initial @samp{QTDP} packet for
24971 this tracepoint. This packet may only be sent immediately after
24972 another @samp{QTDP} packet that ended with a @samp{-}. If the
24973 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
24974 specifying more actions for this tracepoint.
24976 In the series of action packets for a given tracepoint, at most one
24977 can have an @samp{S} before its first @var{action}. If such a packet
24978 is sent, it and the following packets define ``while-stepping''
24979 actions. Any prior packets define ordinary actions --- that is, those
24980 taken when the tracepoint is first hit. If no action packet has an
24981 @samp{S}, then all the packets in the series specify ordinary
24982 tracepoint actions.
24984 The @samp{@var{action}@dots{}} portion of the packet is a series of
24985 actions, concatenated without separators. Each action has one of the
24991 Collect the registers whose bits are set in @var{mask}. @var{mask} is
24992 a hexadecimal number whose @var{i}'th bit is set if register number
24993 @var{i} should be collected. (The least significant bit is numbered
24994 zero.) Note that @var{mask} may be any number of digits long; it may
24995 not fit in a 32-bit word.
24997 @item M @var{basereg},@var{offset},@var{len}
24998 Collect @var{len} bytes of memory starting at the address in register
24999 number @var{basereg}, plus @var{offset}. If @var{basereg} is
25000 @samp{-1}, then the range has a fixed address: @var{offset} is the
25001 address of the lowest byte to collect. The @var{basereg},
25002 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
25003 values (the @samp{-1} value for @var{basereg} is a special case).
25005 @item X @var{len},@var{expr}
25006 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
25007 it directs. @var{expr} is an agent expression, as described in
25008 @ref{Agent Expressions}. Each byte of the expression is encoded as a
25009 two-digit hex number in the packet; @var{len} is the number of bytes
25010 in the expression (and thus one-half the number of hex digits in the
25015 Any number of actions may be packed together in a single @samp{QTDP}
25016 packet, as long as the packet does not exceed the maximum packet
25017 length (400 bytes, for many stubs). There may be only one @samp{R}
25018 action per tracepoint, and it must precede any @samp{M} or @samp{X}
25019 actions. Any registers referred to by @samp{M} and @samp{X} actions
25020 must be collected by a preceding @samp{R} action. (The
25021 ``while-stepping'' actions are treated as if they were attached to a
25022 separate tracepoint, as far as these restrictions are concerned.)
25027 The packet was understood and carried out.
25029 The packet was not recognized.
25032 @item QTFrame:@var{n}
25033 Select the @var{n}'th tracepoint frame from the buffer, and use the
25034 register and memory contents recorded there to answer subsequent
25035 request packets from @value{GDBN}.
25037 A successful reply from the stub indicates that the stub has found the
25038 requested frame. The response is a series of parts, concatenated
25039 without separators, describing the frame we selected. Each part has
25040 one of the following forms:
25044 The selected frame is number @var{n} in the trace frame buffer;
25045 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
25046 was no frame matching the criteria in the request packet.
25049 The selected trace frame records a hit of tracepoint number @var{t};
25050 @var{t} is a hexadecimal number.
25054 @item QTFrame:pc:@var{addr}
25055 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
25056 currently selected frame whose PC is @var{addr};
25057 @var{addr} is a hexadecimal number.
25059 @item QTFrame:tdp:@var{t}
25060 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
25061 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
25062 is a hexadecimal number.
25064 @item QTFrame:range:@var{start}:@var{end}
25065 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
25066 currently selected frame whose PC is between @var{start} (inclusive)
25067 and @var{end} (exclusive); @var{start} and @var{end} are hexadecimal
25070 @item QTFrame:outside:@var{start}:@var{end}
25071 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
25072 frame @emph{outside} the given range of addresses.
25075 Begin the tracepoint experiment. Begin collecting data from tracepoint
25076 hits in the trace frame buffer.
25079 End the tracepoint experiment. Stop collecting trace frames.
25082 Clear the table of tracepoints, and empty the trace frame buffer.
25084 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
25085 Establish the given ranges of memory as ``transparent''. The stub
25086 will answer requests for these ranges from memory's current contents,
25087 if they were not collected as part of the tracepoint hit.
25089 @value{GDBN} uses this to mark read-only regions of memory, like those
25090 containing program code. Since these areas never change, they should
25091 still have the same contents they did when the tracepoint was hit, so
25092 there's no reason for the stub to refuse to provide their contents.
25095 Ask the stub if there is a trace experiment running right now.
25100 There is no trace experiment running.
25102 There is a trace experiment running.
25108 @node Host I/O Packets
25109 @section Host I/O Packets
25110 @cindex Host I/O, remote protocol
25111 @cindex file transfer, remote protocol
25113 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
25114 operations on the far side of a remote link. For example, Host I/O is
25115 used to upload and download files to a remote target with its own
25116 filesystem. Host I/O uses the same constant values and data structure
25117 layout as the target-initiated File-I/O protocol. However, the
25118 Host I/O packets are structured differently. The target-initiated
25119 protocol relies on target memory to store parameters and buffers.
25120 Host I/O requests are initiated by @value{GDBN}, and the
25121 target's memory is not involved. @xref{File-I/O Remote Protocol
25122 Extension}, for more details on the target-initiated protocol.
25124 The Host I/O request packets all encode a single operation along with
25125 its arguments. They have this format:
25129 @item vFile:@var{operation}: @var{parameter}@dots{}
25130 @var{operation} is the name of the particular request; the target
25131 should compare the entire packet name up to the second colon when checking
25132 for a supported operation. The format of @var{parameter} depends on
25133 the operation. Numbers are always passed in hexadecimal. Negative
25134 numbers have an explicit minus sign (i.e.@: two's complement is not
25135 used). Strings (e.g.@: filenames) are encoded as a series of
25136 hexadecimal bytes. The last argument to a system call may be a
25137 buffer of escaped binary data (@pxref{Binary Data}).
25141 The valid responses to Host I/O packets are:
25145 @item F @var{result} [, @var{errno}] [; @var{attachment}]
25146 @var{result} is the integer value returned by this operation, usually
25147 non-negative for success and -1 for errors. If an error has occured,
25148 @var{errno} will be included in the result. @var{errno} will have a
25149 value defined by the File-I/O protocol (@pxref{Errno Values}). For
25150 operations which return data, @var{attachment} supplies the data as a
25151 binary buffer. Binary buffers in response packets are escaped in the
25152 normal way (@pxref{Binary Data}). See the individual packet
25153 documentation for the interpretation of @var{result} and
25157 An empty response indicates that this operation is not recognized.
25161 These are the supported Host I/O operations:
25164 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
25165 Open a file at @var{pathname} and return a file descriptor for it, or
25166 return -1 if an error occurs. @var{pathname} is a string,
25167 @var{flags} is an integer indicating a mask of open flags
25168 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
25169 of mode bits to use if the file is created (@pxref{mode_t Values}).
25170 @xref{open}, for details of the open flags and mode values.
25172 @item vFile:close: @var{fd}
25173 Close the open file corresponding to @var{fd} and return 0, or
25174 -1 if an error occurs.
25176 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
25177 Read data from the open file corresponding to @var{fd}. Up to
25178 @var{count} bytes will be read from the file, starting at @var{offset}
25179 relative to the start of the file. The target may read fewer bytes;
25180 common reasons include packet size limits and an end-of-file
25181 condition. The number of bytes read is returned. Zero should only be
25182 returned for a successful read at the end of the file, or if
25183 @var{count} was zero.
25185 The data read should be returned as a binary attachment on success.
25186 If zero bytes were read, the response should include an empty binary
25187 attachment (i.e.@: a trailing semicolon). The return value is the
25188 number of target bytes read; the binary attachment may be longer if
25189 some characters were escaped.
25191 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
25192 Write @var{data} (a binary buffer) to the open file corresponding
25193 to @var{fd}. Start the write at @var{offset} from the start of the
25194 file. Unlike many @code{write} system calls, there is no
25195 separate @var{count} argument; the length of @var{data} in the
25196 packet is used. @samp{vFile:write} returns the number of bytes written,
25197 which may be shorter than the length of @var{data}, or -1 if an
25200 @item vFile:unlink: @var{pathname}
25201 Delete the file at @var{pathname} on the target. Return 0,
25202 or -1 if an error occurs. @var{pathname} is a string.
25207 @section Interrupts
25208 @cindex interrupts (remote protocol)
25210 When a program on the remote target is running, @value{GDBN} may
25211 attempt to interrupt it by sending a @samp{Ctrl-C} or a @code{BREAK},
25212 control of which is specified via @value{GDBN}'s @samp{remotebreak}
25213 setting (@pxref{set remotebreak}).
25215 The precise meaning of @code{BREAK} is defined by the transport
25216 mechanism and may, in fact, be undefined. @value{GDBN} does
25217 not currently define a @code{BREAK} mechanism for any of the network
25220 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
25221 transport mechanisms. It is represented by sending the single byte
25222 @code{0x03} without any of the usual packet overhead described in
25223 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
25224 transmitted as part of a packet, it is considered to be packet data
25225 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
25226 (@pxref{X packet}), used for binary downloads, may include an unescaped
25227 @code{0x03} as part of its packet.
25229 Stubs are not required to recognize these interrupt mechanisms and the
25230 precise meaning associated with receipt of the interrupt is
25231 implementation defined. If the stub is successful at interrupting the
25232 running program, it is expected that it will send one of the Stop
25233 Reply Packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
25234 of successfully stopping the program. Interrupts received while the
25235 program is stopped will be discarded.
25240 Example sequence of a target being re-started. Notice how the restart
25241 does not get any direct output:
25246 @emph{target restarts}
25249 <- @code{T001:1234123412341234}
25253 Example sequence of a target being stepped by a single instruction:
25256 -> @code{G1445@dots{}}
25261 <- @code{T001:1234123412341234}
25265 <- @code{1455@dots{}}
25269 @node File-I/O Remote Protocol Extension
25270 @section File-I/O Remote Protocol Extension
25271 @cindex File-I/O remote protocol extension
25274 * File-I/O Overview::
25275 * Protocol Basics::
25276 * The F Request Packet::
25277 * The F Reply Packet::
25278 * The Ctrl-C Message::
25280 * List of Supported Calls::
25281 * Protocol-specific Representation of Datatypes::
25283 * File-I/O Examples::
25286 @node File-I/O Overview
25287 @subsection File-I/O Overview
25288 @cindex file-i/o overview
25290 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
25291 target to use the host's file system and console I/O to perform various
25292 system calls. System calls on the target system are translated into a
25293 remote protocol packet to the host system, which then performs the needed
25294 actions and returns a response packet to the target system.
25295 This simulates file system operations even on targets that lack file systems.
25297 The protocol is defined to be independent of both the host and target systems.
25298 It uses its own internal representation of datatypes and values. Both
25299 @value{GDBN} and the target's @value{GDBN} stub are responsible for
25300 translating the system-dependent value representations into the internal
25301 protocol representations when data is transmitted.
25303 The communication is synchronous. A system call is possible only when
25304 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
25305 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
25306 the target is stopped to allow deterministic access to the target's
25307 memory. Therefore File-I/O is not interruptible by target signals. On
25308 the other hand, it is possible to interrupt File-I/O by a user interrupt
25309 (@samp{Ctrl-C}) within @value{GDBN}.
25311 The target's request to perform a host system call does not finish
25312 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
25313 after finishing the system call, the target returns to continuing the
25314 previous activity (continue, step). No additional continue or step
25315 request from @value{GDBN} is required.
25318 (@value{GDBP}) continue
25319 <- target requests 'system call X'
25320 target is stopped, @value{GDBN} executes system call
25321 -> @value{GDBN} returns result
25322 ... target continues, @value{GDBN} returns to wait for the target
25323 <- target hits breakpoint and sends a Txx packet
25326 The protocol only supports I/O on the console and to regular files on
25327 the host file system. Character or block special devices, pipes,
25328 named pipes, sockets or any other communication method on the host
25329 system are not supported by this protocol.
25331 @node Protocol Basics
25332 @subsection Protocol Basics
25333 @cindex protocol basics, file-i/o
25335 The File-I/O protocol uses the @code{F} packet as the request as well
25336 as reply packet. Since a File-I/O system call can only occur when
25337 @value{GDBN} is waiting for a response from the continuing or stepping target,
25338 the File-I/O request is a reply that @value{GDBN} has to expect as a result
25339 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
25340 This @code{F} packet contains all information needed to allow @value{GDBN}
25341 to call the appropriate host system call:
25345 A unique identifier for the requested system call.
25348 All parameters to the system call. Pointers are given as addresses
25349 in the target memory address space. Pointers to strings are given as
25350 pointer/length pair. Numerical values are given as they are.
25351 Numerical control flags are given in a protocol-specific representation.
25355 At this point, @value{GDBN} has to perform the following actions.
25359 If the parameters include pointer values to data needed as input to a
25360 system call, @value{GDBN} requests this data from the target with a
25361 standard @code{m} packet request. This additional communication has to be
25362 expected by the target implementation and is handled as any other @code{m}
25366 @value{GDBN} translates all value from protocol representation to host
25367 representation as needed. Datatypes are coerced into the host types.
25370 @value{GDBN} calls the system call.
25373 It then coerces datatypes back to protocol representation.
25376 If the system call is expected to return data in buffer space specified
25377 by pointer parameters to the call, the data is transmitted to the
25378 target using a @code{M} or @code{X} packet. This packet has to be expected
25379 by the target implementation and is handled as any other @code{M} or @code{X}
25384 Eventually @value{GDBN} replies with another @code{F} packet which contains all
25385 necessary information for the target to continue. This at least contains
25392 @code{errno}, if has been changed by the system call.
25399 After having done the needed type and value coercion, the target continues
25400 the latest continue or step action.
25402 @node The F Request Packet
25403 @subsection The @code{F} Request Packet
25404 @cindex file-i/o request packet
25405 @cindex @code{F} request packet
25407 The @code{F} request packet has the following format:
25410 @item F@var{call-id},@var{parameter@dots{}}
25412 @var{call-id} is the identifier to indicate the host system call to be called.
25413 This is just the name of the function.
25415 @var{parameter@dots{}} are the parameters to the system call.
25416 Parameters are hexadecimal integer values, either the actual values in case
25417 of scalar datatypes, pointers to target buffer space in case of compound
25418 datatypes and unspecified memory areas, or pointer/length pairs in case
25419 of string parameters. These are appended to the @var{call-id} as a
25420 comma-delimited list. All values are transmitted in ASCII
25421 string representation, pointer/length pairs separated by a slash.
25427 @node The F Reply Packet
25428 @subsection The @code{F} Reply Packet
25429 @cindex file-i/o reply packet
25430 @cindex @code{F} reply packet
25432 The @code{F} reply packet has the following format:
25436 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
25438 @var{retcode} is the return code of the system call as hexadecimal value.
25440 @var{errno} is the @code{errno} set by the call, in protocol-specific
25442 This parameter can be omitted if the call was successful.
25444 @var{Ctrl-C flag} is only sent if the user requested a break. In this
25445 case, @var{errno} must be sent as well, even if the call was successful.
25446 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
25453 or, if the call was interrupted before the host call has been performed:
25460 assuming 4 is the protocol-specific representation of @code{EINTR}.
25465 @node The Ctrl-C Message
25466 @subsection The @samp{Ctrl-C} Message
25467 @cindex ctrl-c message, in file-i/o protocol
25469 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
25470 reply packet (@pxref{The F Reply Packet}),
25471 the target should behave as if it had
25472 gotten a break message. The meaning for the target is ``system call
25473 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
25474 (as with a break message) and return to @value{GDBN} with a @code{T02}
25477 It's important for the target to know in which
25478 state the system call was interrupted. There are two possible cases:
25482 The system call hasn't been performed on the host yet.
25485 The system call on the host has been finished.
25489 These two states can be distinguished by the target by the value of the
25490 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
25491 call hasn't been performed. This is equivalent to the @code{EINTR} handling
25492 on POSIX systems. In any other case, the target may presume that the
25493 system call has been finished --- successfully or not --- and should behave
25494 as if the break message arrived right after the system call.
25496 @value{GDBN} must behave reliably. If the system call has not been called
25497 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
25498 @code{errno} in the packet. If the system call on the host has been finished
25499 before the user requests a break, the full action must be finished by
25500 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
25501 The @code{F} packet may only be sent when either nothing has happened
25502 or the full action has been completed.
25505 @subsection Console I/O
25506 @cindex console i/o as part of file-i/o
25508 By default and if not explicitly closed by the target system, the file
25509 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
25510 on the @value{GDBN} console is handled as any other file output operation
25511 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
25512 by @value{GDBN} so that after the target read request from file descriptor
25513 0 all following typing is buffered until either one of the following
25518 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
25520 system call is treated as finished.
25523 The user presses @key{RET}. This is treated as end of input with a trailing
25527 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
25528 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
25532 If the user has typed more characters than fit in the buffer given to
25533 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
25534 either another @code{read(0, @dots{})} is requested by the target, or debugging
25535 is stopped at the user's request.
25538 @node List of Supported Calls
25539 @subsection List of Supported Calls
25540 @cindex list of supported file-i/o calls
25557 @unnumberedsubsubsec open
25558 @cindex open, file-i/o system call
25563 int open(const char *pathname, int flags);
25564 int open(const char *pathname, int flags, mode_t mode);
25568 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
25571 @var{flags} is the bitwise @code{OR} of the following values:
25575 If the file does not exist it will be created. The host
25576 rules apply as far as file ownership and time stamps
25580 When used with @code{O_CREAT}, if the file already exists it is
25581 an error and open() fails.
25584 If the file already exists and the open mode allows
25585 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
25586 truncated to zero length.
25589 The file is opened in append mode.
25592 The file is opened for reading only.
25595 The file is opened for writing only.
25598 The file is opened for reading and writing.
25602 Other bits are silently ignored.
25606 @var{mode} is the bitwise @code{OR} of the following values:
25610 User has read permission.
25613 User has write permission.
25616 Group has read permission.
25619 Group has write permission.
25622 Others have read permission.
25625 Others have write permission.
25629 Other bits are silently ignored.
25632 @item Return value:
25633 @code{open} returns the new file descriptor or -1 if an error
25640 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
25643 @var{pathname} refers to a directory.
25646 The requested access is not allowed.
25649 @var{pathname} was too long.
25652 A directory component in @var{pathname} does not exist.
25655 @var{pathname} refers to a device, pipe, named pipe or socket.
25658 @var{pathname} refers to a file on a read-only filesystem and
25659 write access was requested.
25662 @var{pathname} is an invalid pointer value.
25665 No space on device to create the file.
25668 The process already has the maximum number of files open.
25671 The limit on the total number of files open on the system
25675 The call was interrupted by the user.
25681 @unnumberedsubsubsec close
25682 @cindex close, file-i/o system call
25691 @samp{Fclose,@var{fd}}
25693 @item Return value:
25694 @code{close} returns zero on success, or -1 if an error occurred.
25700 @var{fd} isn't a valid open file descriptor.
25703 The call was interrupted by the user.
25709 @unnumberedsubsubsec read
25710 @cindex read, file-i/o system call
25715 int read(int fd, void *buf, unsigned int count);
25719 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
25721 @item Return value:
25722 On success, the number of bytes read is returned.
25723 Zero indicates end of file. If count is zero, read
25724 returns zero as well. On error, -1 is returned.
25730 @var{fd} is not a valid file descriptor or is not open for
25734 @var{bufptr} is an invalid pointer value.
25737 The call was interrupted by the user.
25743 @unnumberedsubsubsec write
25744 @cindex write, file-i/o system call
25749 int write(int fd, const void *buf, unsigned int count);
25753 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
25755 @item Return value:
25756 On success, the number of bytes written are returned.
25757 Zero indicates nothing was written. On error, -1
25764 @var{fd} is not a valid file descriptor or is not open for
25768 @var{bufptr} is an invalid pointer value.
25771 An attempt was made to write a file that exceeds the
25772 host-specific maximum file size allowed.
25775 No space on device to write the data.
25778 The call was interrupted by the user.
25784 @unnumberedsubsubsec lseek
25785 @cindex lseek, file-i/o system call
25790 long lseek (int fd, long offset, int flag);
25794 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
25796 @var{flag} is one of:
25800 The offset is set to @var{offset} bytes.
25803 The offset is set to its current location plus @var{offset}
25807 The offset is set to the size of the file plus @var{offset}
25811 @item Return value:
25812 On success, the resulting unsigned offset in bytes from
25813 the beginning of the file is returned. Otherwise, a
25814 value of -1 is returned.
25820 @var{fd} is not a valid open file descriptor.
25823 @var{fd} is associated with the @value{GDBN} console.
25826 @var{flag} is not a proper value.
25829 The call was interrupted by the user.
25835 @unnumberedsubsubsec rename
25836 @cindex rename, file-i/o system call
25841 int rename(const char *oldpath, const char *newpath);
25845 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
25847 @item Return value:
25848 On success, zero is returned. On error, -1 is returned.
25854 @var{newpath} is an existing directory, but @var{oldpath} is not a
25858 @var{newpath} is a non-empty directory.
25861 @var{oldpath} or @var{newpath} is a directory that is in use by some
25865 An attempt was made to make a directory a subdirectory
25869 A component used as a directory in @var{oldpath} or new
25870 path is not a directory. Or @var{oldpath} is a directory
25871 and @var{newpath} exists but is not a directory.
25874 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
25877 No access to the file or the path of the file.
25881 @var{oldpath} or @var{newpath} was too long.
25884 A directory component in @var{oldpath} or @var{newpath} does not exist.
25887 The file is on a read-only filesystem.
25890 The device containing the file has no room for the new
25894 The call was interrupted by the user.
25900 @unnumberedsubsubsec unlink
25901 @cindex unlink, file-i/o system call
25906 int unlink(const char *pathname);
25910 @samp{Funlink,@var{pathnameptr}/@var{len}}
25912 @item Return value:
25913 On success, zero is returned. On error, -1 is returned.
25919 No access to the file or the path of the file.
25922 The system does not allow unlinking of directories.
25925 The file @var{pathname} cannot be unlinked because it's
25926 being used by another process.
25929 @var{pathnameptr} is an invalid pointer value.
25932 @var{pathname} was too long.
25935 A directory component in @var{pathname} does not exist.
25938 A component of the path is not a directory.
25941 The file is on a read-only filesystem.
25944 The call was interrupted by the user.
25950 @unnumberedsubsubsec stat/fstat
25951 @cindex fstat, file-i/o system call
25952 @cindex stat, file-i/o system call
25957 int stat(const char *pathname, struct stat *buf);
25958 int fstat(int fd, struct stat *buf);
25962 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
25963 @samp{Ffstat,@var{fd},@var{bufptr}}
25965 @item Return value:
25966 On success, zero is returned. On error, -1 is returned.
25972 @var{fd} is not a valid open file.
25975 A directory component in @var{pathname} does not exist or the
25976 path is an empty string.
25979 A component of the path is not a directory.
25982 @var{pathnameptr} is an invalid pointer value.
25985 No access to the file or the path of the file.
25988 @var{pathname} was too long.
25991 The call was interrupted by the user.
25997 @unnumberedsubsubsec gettimeofday
25998 @cindex gettimeofday, file-i/o system call
26003 int gettimeofday(struct timeval *tv, void *tz);
26007 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
26009 @item Return value:
26010 On success, 0 is returned, -1 otherwise.
26016 @var{tz} is a non-NULL pointer.
26019 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
26025 @unnumberedsubsubsec isatty
26026 @cindex isatty, file-i/o system call
26031 int isatty(int fd);
26035 @samp{Fisatty,@var{fd}}
26037 @item Return value:
26038 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
26044 The call was interrupted by the user.
26049 Note that the @code{isatty} call is treated as a special case: it returns
26050 1 to the target if the file descriptor is attached
26051 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
26052 would require implementing @code{ioctl} and would be more complex than
26057 @unnumberedsubsubsec system
26058 @cindex system, file-i/o system call
26063 int system(const char *command);
26067 @samp{Fsystem,@var{commandptr}/@var{len}}
26069 @item Return value:
26070 If @var{len} is zero, the return value indicates whether a shell is
26071 available. A zero return value indicates a shell is not available.
26072 For non-zero @var{len}, the value returned is -1 on error and the
26073 return status of the command otherwise. Only the exit status of the
26074 command is returned, which is extracted from the host's @code{system}
26075 return value by calling @code{WEXITSTATUS(retval)}. In case
26076 @file{/bin/sh} could not be executed, 127 is returned.
26082 The call was interrupted by the user.
26087 @value{GDBN} takes over the full task of calling the necessary host calls
26088 to perform the @code{system} call. The return value of @code{system} on
26089 the host is simplified before it's returned
26090 to the target. Any termination signal information from the child process
26091 is discarded, and the return value consists
26092 entirely of the exit status of the called command.
26094 Due to security concerns, the @code{system} call is by default refused
26095 by @value{GDBN}. The user has to allow this call explicitly with the
26096 @code{set remote system-call-allowed 1} command.
26099 @item set remote system-call-allowed
26100 @kindex set remote system-call-allowed
26101 Control whether to allow the @code{system} calls in the File I/O
26102 protocol for the remote target. The default is zero (disabled).
26104 @item show remote system-call-allowed
26105 @kindex show remote system-call-allowed
26106 Show whether the @code{system} calls are allowed in the File I/O
26110 @node Protocol-specific Representation of Datatypes
26111 @subsection Protocol-specific Representation of Datatypes
26112 @cindex protocol-specific representation of datatypes, in file-i/o protocol
26115 * Integral Datatypes::
26117 * Memory Transfer::
26122 @node Integral Datatypes
26123 @unnumberedsubsubsec Integral Datatypes
26124 @cindex integral datatypes, in file-i/o protocol
26126 The integral datatypes used in the system calls are @code{int},
26127 @code{unsigned int}, @code{long}, @code{unsigned long},
26128 @code{mode_t}, and @code{time_t}.
26130 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
26131 implemented as 32 bit values in this protocol.
26133 @code{long} and @code{unsigned long} are implemented as 64 bit types.
26135 @xref{Limits}, for corresponding MIN and MAX values (similar to those
26136 in @file{limits.h}) to allow range checking on host and target.
26138 @code{time_t} datatypes are defined as seconds since the Epoch.
26140 All integral datatypes transferred as part of a memory read or write of a
26141 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
26144 @node Pointer Values
26145 @unnumberedsubsubsec Pointer Values
26146 @cindex pointer values, in file-i/o protocol
26148 Pointers to target data are transmitted as they are. An exception
26149 is made for pointers to buffers for which the length isn't
26150 transmitted as part of the function call, namely strings. Strings
26151 are transmitted as a pointer/length pair, both as hex values, e.g.@:
26158 which is a pointer to data of length 18 bytes at position 0x1aaf.
26159 The length is defined as the full string length in bytes, including
26160 the trailing null byte. For example, the string @code{"hello world"}
26161 at address 0x123456 is transmitted as
26167 @node Memory Transfer
26168 @unnumberedsubsubsec Memory Transfer
26169 @cindex memory transfer, in file-i/o protocol
26171 Structured data which is transferred using a memory read or write (for
26172 example, a @code{struct stat}) is expected to be in a protocol-specific format
26173 with all scalar multibyte datatypes being big endian. Translation to
26174 this representation needs to be done both by the target before the @code{F}
26175 packet is sent, and by @value{GDBN} before
26176 it transfers memory to the target. Transferred pointers to structured
26177 data should point to the already-coerced data at any time.
26181 @unnumberedsubsubsec struct stat
26182 @cindex struct stat, in file-i/o protocol
26184 The buffer of type @code{struct stat} used by the target and @value{GDBN}
26185 is defined as follows:
26189 unsigned int st_dev; /* device */
26190 unsigned int st_ino; /* inode */
26191 mode_t st_mode; /* protection */
26192 unsigned int st_nlink; /* number of hard links */
26193 unsigned int st_uid; /* user ID of owner */
26194 unsigned int st_gid; /* group ID of owner */
26195 unsigned int st_rdev; /* device type (if inode device) */
26196 unsigned long st_size; /* total size, in bytes */
26197 unsigned long st_blksize; /* blocksize for filesystem I/O */
26198 unsigned long st_blocks; /* number of blocks allocated */
26199 time_t st_atime; /* time of last access */
26200 time_t st_mtime; /* time of last modification */
26201 time_t st_ctime; /* time of last change */
26205 The integral datatypes conform to the definitions given in the
26206 appropriate section (see @ref{Integral Datatypes}, for details) so this
26207 structure is of size 64 bytes.
26209 The values of several fields have a restricted meaning and/or
26215 A value of 0 represents a file, 1 the console.
26218 No valid meaning for the target. Transmitted unchanged.
26221 Valid mode bits are described in @ref{Constants}. Any other
26222 bits have currently no meaning for the target.
26227 No valid meaning for the target. Transmitted unchanged.
26232 These values have a host and file system dependent
26233 accuracy. Especially on Windows hosts, the file system may not
26234 support exact timing values.
26237 The target gets a @code{struct stat} of the above representation and is
26238 responsible for coercing it to the target representation before
26241 Note that due to size differences between the host, target, and protocol
26242 representations of @code{struct stat} members, these members could eventually
26243 get truncated on the target.
26245 @node struct timeval
26246 @unnumberedsubsubsec struct timeval
26247 @cindex struct timeval, in file-i/o protocol
26249 The buffer of type @code{struct timeval} used by the File-I/O protocol
26250 is defined as follows:
26254 time_t tv_sec; /* second */
26255 long tv_usec; /* microsecond */
26259 The integral datatypes conform to the definitions given in the
26260 appropriate section (see @ref{Integral Datatypes}, for details) so this
26261 structure is of size 8 bytes.
26264 @subsection Constants
26265 @cindex constants, in file-i/o protocol
26267 The following values are used for the constants inside of the
26268 protocol. @value{GDBN} and target are responsible for translating these
26269 values before and after the call as needed.
26280 @unnumberedsubsubsec Open Flags
26281 @cindex open flags, in file-i/o protocol
26283 All values are given in hexadecimal representation.
26295 @node mode_t Values
26296 @unnumberedsubsubsec mode_t Values
26297 @cindex mode_t values, in file-i/o protocol
26299 All values are given in octal representation.
26316 @unnumberedsubsubsec Errno Values
26317 @cindex errno values, in file-i/o protocol
26319 All values are given in decimal representation.
26344 @code{EUNKNOWN} is used as a fallback error value if a host system returns
26345 any error value not in the list of supported error numbers.
26348 @unnumberedsubsubsec Lseek Flags
26349 @cindex lseek flags, in file-i/o protocol
26358 @unnumberedsubsubsec Limits
26359 @cindex limits, in file-i/o protocol
26361 All values are given in decimal representation.
26364 INT_MIN -2147483648
26366 UINT_MAX 4294967295
26367 LONG_MIN -9223372036854775808
26368 LONG_MAX 9223372036854775807
26369 ULONG_MAX 18446744073709551615
26372 @node File-I/O Examples
26373 @subsection File-I/O Examples
26374 @cindex file-i/o examples
26376 Example sequence of a write call, file descriptor 3, buffer is at target
26377 address 0x1234, 6 bytes should be written:
26380 <- @code{Fwrite,3,1234,6}
26381 @emph{request memory read from target}
26384 @emph{return "6 bytes written"}
26388 Example sequence of a read call, file descriptor 3, buffer is at target
26389 address 0x1234, 6 bytes should be read:
26392 <- @code{Fread,3,1234,6}
26393 @emph{request memory write to target}
26394 -> @code{X1234,6:XXXXXX}
26395 @emph{return "6 bytes read"}
26399 Example sequence of a read call, call fails on the host due to invalid
26400 file descriptor (@code{EBADF}):
26403 <- @code{Fread,3,1234,6}
26407 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
26411 <- @code{Fread,3,1234,6}
26416 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
26420 <- @code{Fread,3,1234,6}
26421 -> @code{X1234,6:XXXXXX}
26425 @node Library List Format
26426 @section Library List Format
26427 @cindex library list format, remote protocol
26429 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
26430 same process as your application to manage libraries. In this case,
26431 @value{GDBN} can use the loader's symbol table and normal memory
26432 operations to maintain a list of shared libraries. On other
26433 platforms, the operating system manages loaded libraries.
26434 @value{GDBN} can not retrieve the list of currently loaded libraries
26435 through memory operations, so it uses the @samp{qXfer:libraries:read}
26436 packet (@pxref{qXfer library list read}) instead. The remote stub
26437 queries the target's operating system and reports which libraries
26440 The @samp{qXfer:libraries:read} packet returns an XML document which
26441 lists loaded libraries and their offsets. Each library has an
26442 associated name and one or more segment or section base addresses,
26443 which report where the library was loaded in memory.
26445 For the common case of libraries that are fully linked binaries, the
26446 library should have a list of segments. If the target supports
26447 dynamic linking of a relocatable object file, its library XML element
26448 should instead include a list of allocated sections. The segment or
26449 section bases are start addresses, not relocation offsets; they do not
26450 depend on the library's link-time base addresses.
26452 @value{GDBN} must be linked with the Expat library to support XML
26453 library lists. @xref{Expat}.
26455 A simple memory map, with one loaded library relocated by a single
26456 offset, looks like this:
26460 <library name="/lib/libc.so.6">
26461 <segment address="0x10000000"/>
26466 Another simple memory map, with one loaded library with three
26467 allocated sections (.text, .data, .bss), looks like this:
26471 <library name="sharedlib.o">
26472 <section address="0x10000000"/>
26473 <section address="0x20000000"/>
26474 <section address="0x30000000"/>
26479 The format of a library list is described by this DTD:
26482 <!-- library-list: Root element with versioning -->
26483 <!ELEMENT library-list (library)*>
26484 <!ATTLIST library-list version CDATA #FIXED "1.0">
26485 <!ELEMENT library (segment*, section*)>
26486 <!ATTLIST library name CDATA #REQUIRED>
26487 <!ELEMENT segment EMPTY>
26488 <!ATTLIST segment address CDATA #REQUIRED>
26489 <!ELEMENT section EMPTY>
26490 <!ATTLIST section address CDATA #REQUIRED>
26493 In addition, segments and section descriptors cannot be mixed within a
26494 single library element, and you must supply at least one segment or
26495 section for each library.
26497 @node Memory Map Format
26498 @section Memory Map Format
26499 @cindex memory map format
26501 To be able to write into flash memory, @value{GDBN} needs to obtain a
26502 memory map from the target. This section describes the format of the
26505 The memory map is obtained using the @samp{qXfer:memory-map:read}
26506 (@pxref{qXfer memory map read}) packet and is an XML document that
26507 lists memory regions.
26509 @value{GDBN} must be linked with the Expat library to support XML
26510 memory maps. @xref{Expat}.
26512 The top-level structure of the document is shown below:
26515 <?xml version="1.0"?>
26516 <!DOCTYPE memory-map
26517 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
26518 "http://sourceware.org/gdb/gdb-memory-map.dtd">
26524 Each region can be either:
26529 A region of RAM starting at @var{addr} and extending for @var{length}
26533 <memory type="ram" start="@var{addr}" length="@var{length}"/>
26538 A region of read-only memory:
26541 <memory type="rom" start="@var{addr}" length="@var{length}"/>
26546 A region of flash memory, with erasure blocks @var{blocksize}
26550 <memory type="flash" start="@var{addr}" length="@var{length}">
26551 <property name="blocksize">@var{blocksize}</property>
26557 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
26558 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
26559 packets to write to addresses in such ranges.
26561 The formal DTD for memory map format is given below:
26564 <!-- ................................................... -->
26565 <!-- Memory Map XML DTD ................................ -->
26566 <!-- File: memory-map.dtd .............................. -->
26567 <!-- .................................... .............. -->
26568 <!-- memory-map.dtd -->
26569 <!-- memory-map: Root element with versioning -->
26570 <!ELEMENT memory-map (memory | property)>
26571 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
26572 <!ELEMENT memory (property)>
26573 <!-- memory: Specifies a memory region,
26574 and its type, or device. -->
26575 <!ATTLIST memory type CDATA #REQUIRED
26576 start CDATA #REQUIRED
26577 length CDATA #REQUIRED
26578 device CDATA #IMPLIED>
26579 <!-- property: Generic attribute tag -->
26580 <!ELEMENT property (#PCDATA | property)*>
26581 <!ATTLIST property name CDATA #REQUIRED>
26584 @include agentexpr.texi
26586 @node Target Descriptions
26587 @appendix Target Descriptions
26588 @cindex target descriptions
26590 @strong{Warning:} target descriptions are still under active development,
26591 and the contents and format may change between @value{GDBN} releases.
26592 The format is expected to stabilize in the future.
26594 One of the challenges of using @value{GDBN} to debug embedded systems
26595 is that there are so many minor variants of each processor
26596 architecture in use. It is common practice for vendors to start with
26597 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
26598 and then make changes to adapt it to a particular market niche. Some
26599 architectures have hundreds of variants, available from dozens of
26600 vendors. This leads to a number of problems:
26604 With so many different customized processors, it is difficult for
26605 the @value{GDBN} maintainers to keep up with the changes.
26607 Since individual variants may have short lifetimes or limited
26608 audiences, it may not be worthwhile to carry information about every
26609 variant in the @value{GDBN} source tree.
26611 When @value{GDBN} does support the architecture of the embedded system
26612 at hand, the task of finding the correct architecture name to give the
26613 @command{set architecture} command can be error-prone.
26616 To address these problems, the @value{GDBN} remote protocol allows a
26617 target system to not only identify itself to @value{GDBN}, but to
26618 actually describe its own features. This lets @value{GDBN} support
26619 processor variants it has never seen before --- to the extent that the
26620 descriptions are accurate, and that @value{GDBN} understands them.
26622 @value{GDBN} must be linked with the Expat library to support XML
26623 target descriptions. @xref{Expat}.
26626 * Retrieving Descriptions:: How descriptions are fetched from a target.
26627 * Target Description Format:: The contents of a target description.
26628 * Predefined Target Types:: Standard types available for target
26630 * Standard Target Features:: Features @value{GDBN} knows about.
26633 @node Retrieving Descriptions
26634 @section Retrieving Descriptions
26636 Target descriptions can be read from the target automatically, or
26637 specified by the user manually. The default behavior is to read the
26638 description from the target. @value{GDBN} retrieves it via the remote
26639 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
26640 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
26641 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
26642 XML document, of the form described in @ref{Target Description
26645 Alternatively, you can specify a file to read for the target description.
26646 If a file is set, the target will not be queried. The commands to
26647 specify a file are:
26650 @cindex set tdesc filename
26651 @item set tdesc filename @var{path}
26652 Read the target description from @var{path}.
26654 @cindex unset tdesc filename
26655 @item unset tdesc filename
26656 Do not read the XML target description from a file. @value{GDBN}
26657 will use the description supplied by the current target.
26659 @cindex show tdesc filename
26660 @item show tdesc filename
26661 Show the filename to read for a target description, if any.
26665 @node Target Description Format
26666 @section Target Description Format
26667 @cindex target descriptions, XML format
26669 A target description annex is an @uref{http://www.w3.org/XML/, XML}
26670 document which complies with the Document Type Definition provided in
26671 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
26672 means you can use generally available tools like @command{xmllint} to
26673 check that your feature descriptions are well-formed and valid.
26674 However, to help people unfamiliar with XML write descriptions for
26675 their targets, we also describe the grammar here.
26677 Target descriptions can identify the architecture of the remote target
26678 and (for some architectures) provide information about custom register
26679 sets. @value{GDBN} can use this information to autoconfigure for your
26680 target, or to warn you if you connect to an unsupported target.
26682 Here is a simple target description:
26685 <target version="1.0">
26686 <architecture>i386:x86-64</architecture>
26691 This minimal description only says that the target uses
26692 the x86-64 architecture.
26694 A target description has the following overall form, with [ ] marking
26695 optional elements and @dots{} marking repeatable elements. The elements
26696 are explained further below.
26699 <?xml version="1.0"?>
26700 <!DOCTYPE target SYSTEM "gdb-target.dtd">
26701 <target version="1.0">
26702 @r{[}@var{architecture}@r{]}
26703 @r{[}@var{feature}@dots{}@r{]}
26708 The description is generally insensitive to whitespace and line
26709 breaks, under the usual common-sense rules. The XML version
26710 declaration and document type declaration can generally be omitted
26711 (@value{GDBN} does not require them), but specifying them may be
26712 useful for XML validation tools. The @samp{version} attribute for
26713 @samp{<target>} may also be omitted, but we recommend
26714 including it; if future versions of @value{GDBN} use an incompatible
26715 revision of @file{gdb-target.dtd}, they will detect and report
26716 the version mismatch.
26718 @subsection Inclusion
26719 @cindex target descriptions, inclusion
26722 @cindex <xi:include>
26725 It can sometimes be valuable to split a target description up into
26726 several different annexes, either for organizational purposes, or to
26727 share files between different possible target descriptions. You can
26728 divide a description into multiple files by replacing any element of
26729 the target description with an inclusion directive of the form:
26732 <xi:include href="@var{document}"/>
26736 When @value{GDBN} encounters an element of this form, it will retrieve
26737 the named XML @var{document}, and replace the inclusion directive with
26738 the contents of that document. If the current description was read
26739 using @samp{qXfer}, then so will be the included document;
26740 @var{document} will be interpreted as the name of an annex. If the
26741 current description was read from a file, @value{GDBN} will look for
26742 @var{document} as a file in the same directory where it found the
26743 original description.
26745 @subsection Architecture
26746 @cindex <architecture>
26748 An @samp{<architecture>} element has this form:
26751 <architecture>@var{arch}</architecture>
26754 @var{arch} is an architecture name from the same selection
26755 accepted by @code{set architecture} (@pxref{Targets, ,Specifying a
26756 Debugging Target}).
26758 @subsection Features
26761 Each @samp{<feature>} describes some logical portion of the target
26762 system. Features are currently used to describe available CPU
26763 registers and the types of their contents. A @samp{<feature>} element
26767 <feature name="@var{name}">
26768 @r{[}@var{type}@dots{}@r{]}
26774 Each feature's name should be unique within the description. The name
26775 of a feature does not matter unless @value{GDBN} has some special
26776 knowledge of the contents of that feature; if it does, the feature
26777 should have its standard name. @xref{Standard Target Features}.
26781 Any register's value is a collection of bits which @value{GDBN} must
26782 interpret. The default interpretation is a two's complement integer,
26783 but other types can be requested by name in the register description.
26784 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
26785 Target Types}), and the description can define additional composite types.
26787 Each type element must have an @samp{id} attribute, which gives
26788 a unique (within the containing @samp{<feature>}) name to the type.
26789 Types must be defined before they are used.
26792 Some targets offer vector registers, which can be treated as arrays
26793 of scalar elements. These types are written as @samp{<vector>} elements,
26794 specifying the array element type, @var{type}, and the number of elements,
26798 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
26802 If a register's value is usefully viewed in multiple ways, define it
26803 with a union type containing the useful representations. The
26804 @samp{<union>} element contains one or more @samp{<field>} elements,
26805 each of which has a @var{name} and a @var{type}:
26808 <union id="@var{id}">
26809 <field name="@var{name}" type="@var{type}"/>
26814 @subsection Registers
26817 Each register is represented as an element with this form:
26820 <reg name="@var{name}"
26821 bitsize="@var{size}"
26822 @r{[}regnum="@var{num}"@r{]}
26823 @r{[}save-restore="@var{save-restore}"@r{]}
26824 @r{[}type="@var{type}"@r{]}
26825 @r{[}group="@var{group}"@r{]}/>
26829 The components are as follows:
26834 The register's name; it must be unique within the target description.
26837 The register's size, in bits.
26840 The register's number. If omitted, a register's number is one greater
26841 than that of the previous register (either in the current feature or in
26842 a preceeding feature); the first register in the target description
26843 defaults to zero. This register number is used to read or write
26844 the register; e.g.@: it is used in the remote @code{p} and @code{P}
26845 packets, and registers appear in the @code{g} and @code{G} packets
26846 in order of increasing register number.
26849 Whether the register should be preserved across inferior function
26850 calls; this must be either @code{yes} or @code{no}. The default is
26851 @code{yes}, which is appropriate for most registers except for
26852 some system control registers; this is not related to the target's
26856 The type of the register. @var{type} may be a predefined type, a type
26857 defined in the current feature, or one of the special types @code{int}
26858 and @code{float}. @code{int} is an integer type of the correct size
26859 for @var{bitsize}, and @code{float} is a floating point type (in the
26860 architecture's normal floating point format) of the correct size for
26861 @var{bitsize}. The default is @code{int}.
26864 The register group to which this register belongs. @var{group} must
26865 be either @code{general}, @code{float}, or @code{vector}. If no
26866 @var{group} is specified, @value{GDBN} will not display the register
26867 in @code{info registers}.
26871 @node Predefined Target Types
26872 @section Predefined Target Types
26873 @cindex target descriptions, predefined types
26875 Type definitions in the self-description can build up composite types
26876 from basic building blocks, but can not define fundamental types. Instead,
26877 standard identifiers are provided by @value{GDBN} for the fundamental
26878 types. The currently supported types are:
26887 Signed integer types holding the specified number of bits.
26894 Unsigned integer types holding the specified number of bits.
26898 Pointers to unspecified code and data. The program counter and
26899 any dedicated return address register may be marked as code
26900 pointers; printing a code pointer converts it into a symbolic
26901 address. The stack pointer and any dedicated address registers
26902 may be marked as data pointers.
26905 Single precision IEEE floating point.
26908 Double precision IEEE floating point.
26911 The 12-byte extended precision format used by ARM FPA registers.
26915 @node Standard Target Features
26916 @section Standard Target Features
26917 @cindex target descriptions, standard features
26919 A target description must contain either no registers or all the
26920 target's registers. If the description contains no registers, then
26921 @value{GDBN} will assume a default register layout, selected based on
26922 the architecture. If the description contains any registers, the
26923 default layout will not be used; the standard registers must be
26924 described in the target description, in such a way that @value{GDBN}
26925 can recognize them.
26927 This is accomplished by giving specific names to feature elements
26928 which contain standard registers. @value{GDBN} will look for features
26929 with those names and verify that they contain the expected registers;
26930 if any known feature is missing required registers, or if any required
26931 feature is missing, @value{GDBN} will reject the target
26932 description. You can add additional registers to any of the
26933 standard features --- @value{GDBN} will display them just as if
26934 they were added to an unrecognized feature.
26936 This section lists the known features and their expected contents.
26937 Sample XML documents for these features are included in the
26938 @value{GDBN} source tree, in the directory @file{gdb/features}.
26940 Names recognized by @value{GDBN} should include the name of the
26941 company or organization which selected the name, and the overall
26942 architecture to which the feature applies; so e.g.@: the feature
26943 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
26945 The names of registers are not case sensitive for the purpose
26946 of recognizing standard features, but @value{GDBN} will only display
26947 registers using the capitalization used in the description.
26953 * PowerPC Features::
26958 @subsection ARM Features
26959 @cindex target descriptions, ARM features
26961 The @samp{org.gnu.gdb.arm.core} feature is required for ARM targets.
26962 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
26963 @samp{lr}, @samp{pc}, and @samp{cpsr}.
26965 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
26966 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
26968 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
26969 it should contain at least registers @samp{wR0} through @samp{wR15} and
26970 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
26971 @samp{wCSSF}, and @samp{wCASF} registers are optional.
26973 @node MIPS Features
26974 @subsection MIPS Features
26975 @cindex target descriptions, MIPS features
26977 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
26978 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
26979 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
26982 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
26983 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
26984 registers. They may be 32-bit or 64-bit depending on the target.
26986 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
26987 it may be optional in a future version of @value{GDBN}. It should
26988 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
26989 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
26991 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
26992 contain a single register, @samp{restart}, which is used by the
26993 Linux kernel to control restartable syscalls.
26995 @node M68K Features
26996 @subsection M68K Features
26997 @cindex target descriptions, M68K features
27000 @item @samp{org.gnu.gdb.m68k.core}
27001 @itemx @samp{org.gnu.gdb.coldfire.core}
27002 @itemx @samp{org.gnu.gdb.fido.core}
27003 One of those features must be always present.
27004 The feature that is present determines which flavor of m86k is
27005 used. The feature that is present should contain registers
27006 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
27007 @samp{sp}, @samp{ps} and @samp{pc}.
27009 @item @samp{org.gnu.gdb.coldfire.fp}
27010 This feature is optional. If present, it should contain registers
27011 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
27015 @node PowerPC Features
27016 @subsection PowerPC Features
27017 @cindex target descriptions, PowerPC features
27019 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
27020 targets. It should contain registers @samp{r0} through @samp{r31},
27021 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
27022 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
27024 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
27025 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
27027 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
27028 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
27031 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
27032 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
27033 @samp{spefscr}. SPE targets should provide 32-bit registers in
27034 @samp{org.gnu.gdb.power.core} and provide the upper halves in
27035 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
27036 these to present registers @samp{ev0} through @samp{ev31} to the
27051 % I think something like @colophon should be in texinfo. In the
27053 \long\def\colophon{\hbox to0pt{}\vfill
27054 \centerline{The body of this manual is set in}
27055 \centerline{\fontname\tenrm,}
27056 \centerline{with headings in {\bf\fontname\tenbf}}
27057 \centerline{and examples in {\tt\fontname\tentt}.}
27058 \centerline{{\it\fontname\tenit\/},}
27059 \centerline{{\bf\fontname\tenbf}, and}
27060 \centerline{{\sl\fontname\tensl\/}}
27061 \centerline{are used for emphasis.}\vfill}
27063 % Blame: doc@cygnus.com, 1991.