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, 2007, 2008, 2009
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 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996,
48 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009
49 Free Software Foundation, Inc.
51 Permission is granted to copy, distribute and/or modify this document
52 under the terms of the GNU Free Documentation License, Version 1.1 or
53 any later version published by the Free Software Foundation; with the
54 Invariant Sections being ``Free Software'' and ``Free Software Needs
55 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
56 and with the Back-Cover Texts as in (a) below.
58 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
59 this GNU Manual. Buying copies from GNU Press supports the FSF in
60 developing GNU and promoting software freedom.''
64 This file documents the @sc{gnu} debugger @value{GDBN}.
66 This is the @value{EDITION} Edition, of @cite{Debugging with
67 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
68 @ifset VERSION_PACKAGE
69 @value{VERSION_PACKAGE}
71 Version @value{GDBVN}.
77 @title Debugging with @value{GDBN}
78 @subtitle The @sc{gnu} Source-Level Debugger
80 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
81 @ifset VERSION_PACKAGE
83 @subtitle @value{VERSION_PACKAGE}
85 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
89 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
90 \hfill {\it Debugging with @value{GDBN}}\par
91 \hfill \TeX{}info \texinfoversion\par
95 @vskip 0pt plus 1filll
96 Published by the Free Software Foundation @*
97 51 Franklin Street, Fifth Floor,
98 Boston, MA 02110-1301, USA@*
103 This edition of the GDB manual is dedicated to the memory of Fred
104 Fish. Fred was a long-standing contributor to GDB and to Free
105 software in general. We will miss him.
110 @node Top, Summary, (dir), (dir)
112 @top Debugging with @value{GDBN}
114 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
116 This is the @value{EDITION} Edition, for @value{GDBN}
117 @ifset VERSION_PACKAGE
118 @value{VERSION_PACKAGE}
120 Version @value{GDBVN}.
122 Copyright (C) 1988-2009 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 * Reverse Execution:: Running programs backward
137 * Stack:: Examining the stack
138 * Source:: Examining source files
139 * Data:: Examining data
140 * Macros:: Preprocessor Macros
141 * Tracepoints:: Debugging remote targets non-intrusively
142 * Overlays:: Debugging programs that use overlays
144 * Languages:: Using @value{GDBN} with different languages
146 * Symbols:: Examining the symbol table
147 * Altering:: Altering execution
148 * GDB Files:: @value{GDBN} files
149 * Targets:: Specifying a debugging target
150 * Remote Debugging:: Debugging remote programs
151 * Configurations:: Configuration-specific information
152 * Controlling GDB:: Controlling @value{GDBN}
153 * Extending GDB:: Extending @value{GDBN}
154 * Interpreters:: Command Interpreters
155 * TUI:: @value{GDBN} Text User Interface
156 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
157 * GDB/MI:: @value{GDBN}'s Machine Interface.
158 * Annotations:: @value{GDBN}'s annotation interface.
160 * GDB Bugs:: Reporting bugs in @value{GDBN}
162 * Command Line Editing:: Command Line Editing
163 * Using History Interactively:: Using History Interactively
164 * Formatting Documentation:: How to format and print @value{GDBN} documentation
165 * Installing GDB:: Installing GDB
166 * Maintenance Commands:: Maintenance Commands
167 * Remote Protocol:: GDB Remote Serial Protocol
168 * Agent Expressions:: The GDB Agent Expression Mechanism
169 * Target Descriptions:: How targets can describe themselves to
171 * Operating System Information:: Getting additional information from
173 * Copying:: GNU General Public License says
174 how you can copy and share GDB
175 * GNU Free Documentation License:: The license for this documentation
184 @unnumbered Summary of @value{GDBN}
186 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
187 going on ``inside'' another program while it executes---or what another
188 program was doing at the moment it crashed.
190 @value{GDBN} can do four main kinds of things (plus other things in support of
191 these) to help you catch bugs in the act:
195 Start your program, specifying anything that might affect its behavior.
198 Make your program stop on specified conditions.
201 Examine what has happened, when your program has stopped.
204 Change things in your program, so you can experiment with correcting the
205 effects of one bug and go on to learn about another.
208 You can use @value{GDBN} to debug programs written in C and C@t{++}.
209 For more information, see @ref{Supported Languages,,Supported Languages}.
210 For more information, see @ref{C,,C and C++}.
213 Support for Modula-2 is partial. For information on Modula-2, see
214 @ref{Modula-2,,Modula-2}.
217 Debugging Pascal programs which use sets, subranges, file variables, or
218 nested functions does not currently work. @value{GDBN} does not support
219 entering expressions, printing values, or similar features using Pascal
223 @value{GDBN} can be used to debug programs written in Fortran, although
224 it may be necessary to refer to some variables with a trailing
227 @value{GDBN} can be used to debug programs written in Objective-C,
228 using either the Apple/NeXT or the GNU Objective-C runtime.
231 * Free Software:: Freely redistributable software
232 * Contributors:: Contributors to GDB
236 @unnumberedsec Free Software
238 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
239 General Public License
240 (GPL). The GPL gives you the freedom to copy or adapt a licensed
241 program---but every person getting a copy also gets with it the
242 freedom to modify that copy (which means that they must get access to
243 the source code), and the freedom to distribute further copies.
244 Typical software companies use copyrights to limit your freedoms; the
245 Free Software Foundation uses the GPL to preserve these freedoms.
247 Fundamentally, the General Public License is a license which says that
248 you have these freedoms and that you cannot take these freedoms away
251 @unnumberedsec Free Software Needs Free Documentation
253 The biggest deficiency in the free software community today is not in
254 the software---it is the lack of good free documentation that we can
255 include with the free software. Many of our most important
256 programs do not come with free reference manuals and free introductory
257 texts. Documentation is an essential part of any software package;
258 when an important free software package does not come with a free
259 manual and a free tutorial, that is a major gap. We have many such
262 Consider Perl, for instance. The tutorial manuals that people
263 normally use are non-free. How did this come about? Because the
264 authors of those manuals published them with restrictive terms---no
265 copying, no modification, source files not available---which exclude
266 them from the free software world.
268 That wasn't the first time this sort of thing happened, and it was far
269 from the last. Many times we have heard a GNU user eagerly describe a
270 manual that he is writing, his intended contribution to the community,
271 only to learn that he had ruined everything by signing a publication
272 contract to make it non-free.
274 Free documentation, like free software, is a matter of freedom, not
275 price. The problem with the non-free manual is not that publishers
276 charge a price for printed copies---that in itself is fine. (The Free
277 Software Foundation sells printed copies of manuals, too.) The
278 problem is the restrictions on the use of the manual. Free manuals
279 are available in source code form, and give you permission to copy and
280 modify. Non-free manuals do not allow this.
282 The criteria of freedom for a free manual are roughly the same as for
283 free software. Redistribution (including the normal kinds of
284 commercial redistribution) must be permitted, so that the manual can
285 accompany every copy of the program, both on-line and on paper.
287 Permission for modification of the technical content is crucial too.
288 When people modify the software, adding or changing features, if they
289 are conscientious they will change the manual too---so they can
290 provide accurate and clear documentation for the modified program. A
291 manual that leaves you no choice but to write a new manual to document
292 a changed version of the program is not really available to our
295 Some kinds of limits on the way modification is handled are
296 acceptable. For example, requirements to preserve the original
297 author's copyright notice, the distribution terms, or the list of
298 authors, are ok. It is also no problem to require modified versions
299 to include notice that they were modified. Even entire sections that
300 may not be deleted or changed are acceptable, as long as they deal
301 with nontechnical topics (like this one). These kinds of restrictions
302 are acceptable because they don't obstruct the community's normal use
305 However, it must be possible to modify all the @emph{technical}
306 content of the manual, and then distribute the result in all the usual
307 media, through all the usual channels. Otherwise, the restrictions
308 obstruct the use of the manual, it is not free, and we need another
309 manual to replace it.
311 Please spread the word about this issue. Our community continues to
312 lose manuals to proprietary publishing. If we spread the word that
313 free software needs free reference manuals and free tutorials, perhaps
314 the next person who wants to contribute by writing documentation will
315 realize, before it is too late, that only free manuals contribute to
316 the free software community.
318 If you are writing documentation, please insist on publishing it under
319 the GNU Free Documentation License or another free documentation
320 license. Remember that this decision requires your approval---you
321 don't have to let the publisher decide. Some commercial publishers
322 will use a free license if you insist, but they will not propose the
323 option; it is up to you to raise the issue and say firmly that this is
324 what you want. If the publisher you are dealing with refuses, please
325 try other publishers. If you're not sure whether a proposed license
326 is free, write to @email{licensing@@gnu.org}.
328 You can encourage commercial publishers to sell more free, copylefted
329 manuals and tutorials by buying them, and particularly by buying
330 copies from the publishers that paid for their writing or for major
331 improvements. Meanwhile, try to avoid buying non-free documentation
332 at all. Check the distribution terms of a manual before you buy it,
333 and insist that whoever seeks your business must respect your freedom.
334 Check the history of the book, and try to reward the publishers that
335 have paid or pay the authors to work on it.
337 The Free Software Foundation maintains a list of free documentation
338 published by other publishers, at
339 @url{http://www.fsf.org/doc/other-free-books.html}.
342 @unnumberedsec Contributors to @value{GDBN}
344 Richard Stallman was the original author of @value{GDBN}, and of many
345 other @sc{gnu} programs. Many others have contributed to its
346 development. This section attempts to credit major contributors. One
347 of the virtues of free software is that everyone is free to contribute
348 to it; with regret, we cannot actually acknowledge everyone here. The
349 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
350 blow-by-blow account.
352 Changes much prior to version 2.0 are lost in the mists of time.
355 @emph{Plea:} Additions to this section are particularly welcome. If you
356 or your friends (or enemies, to be evenhanded) have been unfairly
357 omitted from this list, we would like to add your names!
360 So that they may not regard their many labors as thankless, we
361 particularly thank those who shepherded @value{GDBN} through major
363 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
364 Jim Blandy (release 4.18);
365 Jason Molenda (release 4.17);
366 Stan Shebs (release 4.14);
367 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
368 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
369 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
370 Jim Kingdon (releases 3.5, 3.4, and 3.3);
371 and Randy Smith (releases 3.2, 3.1, and 3.0).
373 Richard Stallman, assisted at various times by Peter TerMaat, Chris
374 Hanson, and Richard Mlynarik, handled releases through 2.8.
376 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
377 in @value{GDBN}, with significant additional contributions from Per
378 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
379 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
380 much general update work leading to release 3.0).
382 @value{GDBN} uses the BFD subroutine library to examine multiple
383 object-file formats; BFD was a joint project of David V.
384 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
386 David Johnson wrote the original COFF support; Pace Willison did
387 the original support for encapsulated COFF.
389 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
391 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
392 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
394 Jean-Daniel Fekete contributed Sun 386i support.
395 Chris Hanson improved the HP9000 support.
396 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
397 David Johnson contributed Encore Umax support.
398 Jyrki Kuoppala contributed Altos 3068 support.
399 Jeff Law contributed HP PA and SOM support.
400 Keith Packard contributed NS32K support.
401 Doug Rabson contributed Acorn Risc Machine support.
402 Bob Rusk contributed Harris Nighthawk CX-UX support.
403 Chris Smith contributed Convex support (and Fortran debugging).
404 Jonathan Stone contributed Pyramid support.
405 Michael Tiemann contributed SPARC support.
406 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
407 Pace Willison contributed Intel 386 support.
408 Jay Vosburgh contributed Symmetry support.
409 Marko Mlinar contributed OpenRISC 1000 support.
411 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
413 Rich Schaefer and Peter Schauer helped with support of SunOS shared
416 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
417 about several machine instruction sets.
419 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
420 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
421 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
422 and RDI targets, respectively.
424 Brian Fox is the author of the readline libraries providing
425 command-line editing and command history.
427 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
428 Modula-2 support, and contributed the Languages chapter of this manual.
430 Fred Fish wrote most of the support for Unix System Vr4.
431 He also enhanced the command-completion support to cover C@t{++} overloaded
434 Hitachi America (now Renesas America), Ltd. sponsored the support for
435 H8/300, H8/500, and Super-H processors.
437 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
439 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
442 Toshiba sponsored the support for the TX39 Mips processor.
444 Matsushita sponsored the support for the MN10200 and MN10300 processors.
446 Fujitsu sponsored the support for SPARClite and FR30 processors.
448 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
451 Michael Snyder added support for tracepoints.
453 Stu Grossman wrote gdbserver.
455 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
456 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
458 The following people at the Hewlett-Packard Company contributed
459 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
460 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
461 compiler, and the Text User Interface (nee Terminal User Interface):
462 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
463 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
464 provided HP-specific information in this manual.
466 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
467 Robert Hoehne made significant contributions to the DJGPP port.
469 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
470 development since 1991. Cygnus engineers who have worked on @value{GDBN}
471 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
472 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
473 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
474 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
475 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
476 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
477 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
478 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
479 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
480 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
481 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
482 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
483 Zuhn have made contributions both large and small.
485 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
486 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
488 Jim Blandy added support for preprocessor macros, while working for Red
491 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
492 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
493 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
494 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
495 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
496 with the migration of old architectures to this new framework.
498 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
499 unwinder framework, this consisting of a fresh new design featuring
500 frame IDs, independent frame sniffers, and the sentinel frame. Mark
501 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
502 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
503 trad unwinders. The architecture-specific changes, each involving a
504 complete rewrite of the architecture's frame code, were carried out by
505 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
506 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
507 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
508 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
511 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
512 Tensilica, Inc.@: contributed support for Xtensa processors. Others
513 who have worked on the Xtensa port of @value{GDBN} in the past include
514 Steve Tjiang, John Newlin, and Scott Foehner.
517 @chapter A Sample @value{GDBN} Session
519 You can use this manual at your leisure to read all about @value{GDBN}.
520 However, a handful of commands are enough to get started using the
521 debugger. This chapter illustrates those commands.
524 In this sample session, we emphasize user input like this: @b{input},
525 to make it easier to pick out from the surrounding output.
528 @c FIXME: this example may not be appropriate for some configs, where
529 @c FIXME...primary interest is in remote use.
531 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
532 processor) exhibits the following bug: sometimes, when we change its
533 quote strings from the default, the commands used to capture one macro
534 definition within another stop working. In the following short @code{m4}
535 session, we define a macro @code{foo} which expands to @code{0000}; we
536 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
537 same thing. However, when we change the open quote string to
538 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
539 procedure fails to define a new synonym @code{baz}:
548 @b{define(bar,defn(`foo'))}
552 @b{changequote(<QUOTE>,<UNQUOTE>)}
554 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
557 m4: End of input: 0: fatal error: EOF in string
561 Let us use @value{GDBN} to try to see what is going on.
564 $ @b{@value{GDBP} m4}
565 @c FIXME: this falsifies the exact text played out, to permit smallbook
566 @c FIXME... format to come out better.
567 @value{GDBN} is free software and you are welcome to distribute copies
568 of it under certain conditions; type "show copying" to see
570 There is absolutely no warranty for @value{GDBN}; type "show warranty"
573 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
578 @value{GDBN} reads only enough symbol data to know where to find the
579 rest when needed; as a result, the first prompt comes up very quickly.
580 We now tell @value{GDBN} to use a narrower display width than usual, so
581 that examples fit in this manual.
584 (@value{GDBP}) @b{set width 70}
588 We need to see how the @code{m4} built-in @code{changequote} works.
589 Having looked at the source, we know the relevant subroutine is
590 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
591 @code{break} command.
594 (@value{GDBP}) @b{break m4_changequote}
595 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
599 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
600 control; as long as control does not reach the @code{m4_changequote}
601 subroutine, the program runs as usual:
604 (@value{GDBP}) @b{run}
605 Starting program: /work/Editorial/gdb/gnu/m4/m4
613 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
614 suspends execution of @code{m4}, displaying information about the
615 context where it stops.
618 @b{changequote(<QUOTE>,<UNQUOTE>)}
620 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
622 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
626 Now we use the command @code{n} (@code{next}) to advance execution to
627 the next line of the current function.
631 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
636 @code{set_quotes} looks like a promising subroutine. We can go into it
637 by using the command @code{s} (@code{step}) instead of @code{next}.
638 @code{step} goes to the next line to be executed in @emph{any}
639 subroutine, so it steps into @code{set_quotes}.
643 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
645 530 if (lquote != def_lquote)
649 The display that shows the subroutine where @code{m4} is now
650 suspended (and its arguments) is called a stack frame display. It
651 shows a summary of the stack. We can use the @code{backtrace}
652 command (which can also be spelled @code{bt}), to see where we are
653 in the stack as a whole: the @code{backtrace} command displays a
654 stack frame for each active subroutine.
657 (@value{GDBP}) @b{bt}
658 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
660 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
662 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
663 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
665 #4 0x79dc in expand_input () at macro.c:40
666 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
670 We step through a few more lines to see what happens. The first two
671 times, we can use @samp{s}; the next two times we use @code{n} to avoid
672 falling into the @code{xstrdup} subroutine.
676 0x3b5c 532 if (rquote != def_rquote)
678 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
679 def_lquote : xstrdup(lq);
681 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
684 538 len_lquote = strlen(rquote);
688 The last line displayed looks a little odd; we can examine the variables
689 @code{lquote} and @code{rquote} to see if they are in fact the new left
690 and right quotes we specified. We use the command @code{p}
691 (@code{print}) to see their values.
694 (@value{GDBP}) @b{p lquote}
695 $1 = 0x35d40 "<QUOTE>"
696 (@value{GDBP}) @b{p rquote}
697 $2 = 0x35d50 "<UNQUOTE>"
701 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
702 To look at some context, we can display ten lines of source
703 surrounding the current line with the @code{l} (@code{list}) command.
709 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
711 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
714 538 len_lquote = strlen(rquote);
715 539 len_rquote = strlen(lquote);
722 Let us step past the two lines that set @code{len_lquote} and
723 @code{len_rquote}, and then examine the values of those variables.
727 539 len_rquote = strlen(lquote);
730 (@value{GDBP}) @b{p len_lquote}
732 (@value{GDBP}) @b{p len_rquote}
737 That certainly looks wrong, assuming @code{len_lquote} and
738 @code{len_rquote} are meant to be the lengths of @code{lquote} and
739 @code{rquote} respectively. We can set them to better values using
740 the @code{p} command, since it can print the value of
741 any expression---and that expression can include subroutine calls and
745 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
747 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
752 Is that enough to fix the problem of using the new quotes with the
753 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
754 executing with the @code{c} (@code{continue}) command, and then try the
755 example that caused trouble initially:
761 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
768 Success! The new quotes now work just as well as the default ones. The
769 problem seems to have been just the two typos defining the wrong
770 lengths. We allow @code{m4} exit by giving it an EOF as input:
774 Program exited normally.
778 The message @samp{Program exited normally.} is from @value{GDBN}; it
779 indicates @code{m4} has finished executing. We can end our @value{GDBN}
780 session with the @value{GDBN} @code{quit} command.
783 (@value{GDBP}) @b{quit}
787 @chapter Getting In and Out of @value{GDBN}
789 This chapter discusses how to start @value{GDBN}, and how to get out of it.
793 type @samp{@value{GDBP}} to start @value{GDBN}.
795 type @kbd{quit} or @kbd{Ctrl-d} to exit.
799 * Invoking GDB:: How to start @value{GDBN}
800 * Quitting GDB:: How to quit @value{GDBN}
801 * Shell Commands:: How to use shell commands inside @value{GDBN}
802 * Logging Output:: How to log @value{GDBN}'s output to a file
806 @section Invoking @value{GDBN}
808 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
809 @value{GDBN} reads commands from the terminal until you tell it to exit.
811 You can also run @code{@value{GDBP}} with a variety of arguments and options,
812 to specify more of your debugging environment at the outset.
814 The command-line options described here are designed
815 to cover a variety of situations; in some environments, some of these
816 options may effectively be unavailable.
818 The most usual way to start @value{GDBN} is with one argument,
819 specifying an executable program:
822 @value{GDBP} @var{program}
826 You can also start with both an executable program and a core file
830 @value{GDBP} @var{program} @var{core}
833 You can, instead, specify a process ID as a second argument, if you want
834 to debug a running process:
837 @value{GDBP} @var{program} 1234
841 would attach @value{GDBN} to process @code{1234} (unless you also have a file
842 named @file{1234}; @value{GDBN} does check for a core file first).
844 Taking advantage of the second command-line argument requires a fairly
845 complete operating system; when you use @value{GDBN} as a remote
846 debugger attached to a bare board, there may not be any notion of
847 ``process'', and there is often no way to get a core dump. @value{GDBN}
848 will warn you if it is unable to attach or to read core dumps.
850 You can optionally have @code{@value{GDBP}} pass any arguments after the
851 executable file to the inferior using @code{--args}. This option stops
854 @value{GDBP} --args gcc -O2 -c foo.c
856 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
857 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
859 You can run @code{@value{GDBP}} without printing the front material, which describes
860 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
867 You can further control how @value{GDBN} starts up by using command-line
868 options. @value{GDBN} itself can remind you of the options available.
878 to display all available options and briefly describe their use
879 (@samp{@value{GDBP} -h} is a shorter equivalent).
881 All options and command line arguments you give are processed
882 in sequential order. The order makes a difference when the
883 @samp{-x} option is used.
887 * File Options:: Choosing files
888 * Mode Options:: Choosing modes
889 * Startup:: What @value{GDBN} does during startup
893 @subsection Choosing Files
895 When @value{GDBN} starts, it reads any arguments other than options as
896 specifying an executable file and core file (or process ID). This is
897 the same as if the arguments were specified by the @samp{-se} and
898 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
899 first argument that does not have an associated option flag as
900 equivalent to the @samp{-se} option followed by that argument; and the
901 second argument that does not have an associated option flag, if any, as
902 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
903 If the second argument begins with a decimal digit, @value{GDBN} will
904 first attempt to attach to it as a process, and if that fails, attempt
905 to open it as a corefile. If you have a corefile whose name begins with
906 a digit, you can prevent @value{GDBN} from treating it as a pid by
907 prefixing it with @file{./}, e.g.@: @file{./12345}.
909 If @value{GDBN} has not been configured to included core file support,
910 such as for most embedded targets, then it will complain about a second
911 argument and ignore it.
913 Many options have both long and short forms; both are shown in the
914 following list. @value{GDBN} also recognizes the long forms if you truncate
915 them, so long as enough of the option is present to be unambiguous.
916 (If you prefer, you can flag option arguments with @samp{--} rather
917 than @samp{-}, though we illustrate the more usual convention.)
919 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
920 @c way, both those who look for -foo and --foo in the index, will find
924 @item -symbols @var{file}
926 @cindex @code{--symbols}
928 Read symbol table from file @var{file}.
930 @item -exec @var{file}
932 @cindex @code{--exec}
934 Use file @var{file} as the executable file to execute when appropriate,
935 and for examining pure data in conjunction with a core dump.
939 Read symbol table from file @var{file} and use it as the executable
942 @item -core @var{file}
944 @cindex @code{--core}
946 Use file @var{file} as a core dump to examine.
948 @item -pid @var{number}
949 @itemx -p @var{number}
952 Connect to process ID @var{number}, as with the @code{attach} command.
954 @item -command @var{file}
956 @cindex @code{--command}
958 Execute @value{GDBN} commands from file @var{file}. @xref{Command
959 Files,, Command files}.
961 @item -eval-command @var{command}
962 @itemx -ex @var{command}
963 @cindex @code{--eval-command}
965 Execute a single @value{GDBN} command.
967 This option may be used multiple times to call multiple commands. It may
968 also be interleaved with @samp{-command} as required.
971 @value{GDBP} -ex 'target sim' -ex 'load' \
972 -x setbreakpoints -ex 'run' a.out
975 @item -directory @var{directory}
976 @itemx -d @var{directory}
977 @cindex @code{--directory}
979 Add @var{directory} to the path to search for source and script files.
983 @cindex @code{--readnow}
985 Read each symbol file's entire symbol table immediately, rather than
986 the default, which is to read it incrementally as it is needed.
987 This makes startup slower, but makes future operations faster.
992 @subsection Choosing Modes
994 You can run @value{GDBN} in various alternative modes---for example, in
995 batch mode or quiet mode.
1002 Do not execute commands found in any initialization files. Normally,
1003 @value{GDBN} executes the commands in these files after all the command
1004 options and arguments have been processed. @xref{Command Files,,Command
1010 @cindex @code{--quiet}
1011 @cindex @code{--silent}
1013 ``Quiet''. Do not print the introductory and copyright messages. These
1014 messages are also suppressed in batch mode.
1017 @cindex @code{--batch}
1018 Run in batch mode. Exit with status @code{0} after processing all the
1019 command files specified with @samp{-x} (and all commands from
1020 initialization files, if not inhibited with @samp{-n}). Exit with
1021 nonzero status if an error occurs in executing the @value{GDBN} commands
1022 in the command files.
1024 Batch mode may be useful for running @value{GDBN} as a filter, for
1025 example to download and run a program on another computer; in order to
1026 make this more useful, the message
1029 Program exited normally.
1033 (which is ordinarily issued whenever a program running under
1034 @value{GDBN} control terminates) is not issued when running in batch
1038 @cindex @code{--batch-silent}
1039 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1040 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1041 unaffected). This is much quieter than @samp{-silent} and would be useless
1042 for an interactive session.
1044 This is particularly useful when using targets that give @samp{Loading section}
1045 messages, for example.
1047 Note that targets that give their output via @value{GDBN}, as opposed to
1048 writing directly to @code{stdout}, will also be made silent.
1050 @item -return-child-result
1051 @cindex @code{--return-child-result}
1052 The return code from @value{GDBN} will be the return code from the child
1053 process (the process being debugged), with the following exceptions:
1057 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1058 internal error. In this case the exit code is the same as it would have been
1059 without @samp{-return-child-result}.
1061 The user quits with an explicit value. E.g., @samp{quit 1}.
1063 The child process never runs, or is not allowed to terminate, in which case
1064 the exit code will be -1.
1067 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1068 when @value{GDBN} is being used as a remote program loader or simulator
1073 @cindex @code{--nowindows}
1075 ``No windows''. If @value{GDBN} comes with a graphical user interface
1076 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1077 interface. If no GUI is available, this option has no effect.
1081 @cindex @code{--windows}
1083 If @value{GDBN} includes a GUI, then this option requires it to be
1086 @item -cd @var{directory}
1088 Run @value{GDBN} using @var{directory} as its working directory,
1089 instead of the current directory.
1093 @cindex @code{--fullname}
1095 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1096 subprocess. It tells @value{GDBN} to output the full file name and line
1097 number in a standard, recognizable fashion each time a stack frame is
1098 displayed (which includes each time your program stops). This
1099 recognizable format looks like two @samp{\032} characters, followed by
1100 the file name, line number and character position separated by colons,
1101 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1102 @samp{\032} characters as a signal to display the source code for the
1106 @cindex @code{--epoch}
1107 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1108 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1109 routines so as to allow Epoch to display values of expressions in a
1112 @item -annotate @var{level}
1113 @cindex @code{--annotate}
1114 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1115 effect is identical to using @samp{set annotate @var{level}}
1116 (@pxref{Annotations}). The annotation @var{level} controls how much
1117 information @value{GDBN} prints together with its prompt, values of
1118 expressions, source lines, and other types of output. Level 0 is the
1119 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1120 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1121 that control @value{GDBN}, and level 2 has been deprecated.
1123 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1127 @cindex @code{--args}
1128 Change interpretation of command line so that arguments following the
1129 executable file are passed as command line arguments to the inferior.
1130 This option stops option processing.
1132 @item -baud @var{bps}
1134 @cindex @code{--baud}
1136 Set the line speed (baud rate or bits per second) of any serial
1137 interface used by @value{GDBN} for remote debugging.
1139 @item -l @var{timeout}
1141 Set the timeout (in seconds) of any communication used by @value{GDBN}
1142 for remote debugging.
1144 @item -tty @var{device}
1145 @itemx -t @var{device}
1146 @cindex @code{--tty}
1148 Run using @var{device} for your program's standard input and output.
1149 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1151 @c resolve the situation of these eventually
1153 @cindex @code{--tui}
1154 Activate the @dfn{Text User Interface} when starting. The Text User
1155 Interface manages several text windows on the terminal, showing
1156 source, assembly, registers and @value{GDBN} command outputs
1157 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1158 Text User Interface can be enabled by invoking the program
1159 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1160 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1163 @c @cindex @code{--xdb}
1164 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1165 @c For information, see the file @file{xdb_trans.html}, which is usually
1166 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1169 @item -interpreter @var{interp}
1170 @cindex @code{--interpreter}
1171 Use the interpreter @var{interp} for interface with the controlling
1172 program or device. This option is meant to be set by programs which
1173 communicate with @value{GDBN} using it as a back end.
1174 @xref{Interpreters, , Command Interpreters}.
1176 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1177 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1178 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1179 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1180 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1181 @sc{gdb/mi} interfaces are no longer supported.
1184 @cindex @code{--write}
1185 Open the executable and core files for both reading and writing. This
1186 is equivalent to the @samp{set write on} command inside @value{GDBN}
1190 @cindex @code{--statistics}
1191 This option causes @value{GDBN} to print statistics about time and
1192 memory usage after it completes each command and returns to the prompt.
1195 @cindex @code{--version}
1196 This option causes @value{GDBN} to print its version number and
1197 no-warranty blurb, and exit.
1202 @subsection What @value{GDBN} Does During Startup
1203 @cindex @value{GDBN} startup
1205 Here's the description of what @value{GDBN} does during session startup:
1209 Sets up the command interpreter as specified by the command line
1210 (@pxref{Mode Options, interpreter}).
1214 Reads the @dfn{init file} (if any) in your home directory@footnote{On
1215 DOS/Windows systems, the home directory is the one pointed to by the
1216 @code{HOME} environment variable.} and executes all the commands in
1220 Processes command line options and operands.
1223 Reads and executes the commands from init file (if any) in the current
1224 working directory. This is only done if the current directory is
1225 different from your home directory. Thus, you can have more than one
1226 init file, one generic in your home directory, and another, specific
1227 to the program you are debugging, in the directory where you invoke
1231 Reads command files specified by the @samp{-x} option. @xref{Command
1232 Files}, for more details about @value{GDBN} command files.
1235 Reads the command history recorded in the @dfn{history file}.
1236 @xref{Command History}, for more details about the command history and the
1237 files where @value{GDBN} records it.
1240 Init files use the same syntax as @dfn{command files} (@pxref{Command
1241 Files}) and are processed by @value{GDBN} in the same way. The init
1242 file in your home directory can set options (such as @samp{set
1243 complaints}) that affect subsequent processing of command line options
1244 and operands. Init files are not executed if you use the @samp{-nx}
1245 option (@pxref{Mode Options, ,Choosing Modes}).
1247 @cindex init file name
1248 @cindex @file{.gdbinit}
1249 @cindex @file{gdb.ini}
1250 The @value{GDBN} init files are normally called @file{.gdbinit}.
1251 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1252 the limitations of file names imposed by DOS filesystems. The Windows
1253 ports of @value{GDBN} use the standard name, but if they find a
1254 @file{gdb.ini} file, they warn you about that and suggest to rename
1255 the file to the standard name.
1259 @section Quitting @value{GDBN}
1260 @cindex exiting @value{GDBN}
1261 @cindex leaving @value{GDBN}
1264 @kindex quit @r{[}@var{expression}@r{]}
1265 @kindex q @r{(@code{quit})}
1266 @item quit @r{[}@var{expression}@r{]}
1268 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1269 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1270 do not supply @var{expression}, @value{GDBN} will terminate normally;
1271 otherwise it will terminate using the result of @var{expression} as the
1276 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1277 terminates the action of any @value{GDBN} command that is in progress and
1278 returns to @value{GDBN} command level. It is safe to type the interrupt
1279 character at any time because @value{GDBN} does not allow it to take effect
1280 until a time when it is safe.
1282 If you have been using @value{GDBN} to control an attached process or
1283 device, you can release it with the @code{detach} command
1284 (@pxref{Attach, ,Debugging an Already-running Process}).
1286 @node Shell Commands
1287 @section Shell Commands
1289 If you need to execute occasional shell commands during your
1290 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1291 just use the @code{shell} command.
1295 @cindex shell escape
1296 @item shell @var{command string}
1297 Invoke a standard shell to execute @var{command string}.
1298 If it exists, the environment variable @code{SHELL} determines which
1299 shell to run. Otherwise @value{GDBN} uses the default shell
1300 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1303 The utility @code{make} is often needed in development environments.
1304 You do not have to use the @code{shell} command for this purpose in
1309 @cindex calling make
1310 @item make @var{make-args}
1311 Execute the @code{make} program with the specified
1312 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1315 @node Logging Output
1316 @section Logging Output
1317 @cindex logging @value{GDBN} output
1318 @cindex save @value{GDBN} output to a file
1320 You may want to save the output of @value{GDBN} commands to a file.
1321 There are several commands to control @value{GDBN}'s logging.
1325 @item set logging on
1327 @item set logging off
1329 @cindex logging file name
1330 @item set logging file @var{file}
1331 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1332 @item set logging overwrite [on|off]
1333 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1334 you want @code{set logging on} to overwrite the logfile instead.
1335 @item set logging redirect [on|off]
1336 By default, @value{GDBN} output will go to both the terminal and the logfile.
1337 Set @code{redirect} if you want output to go only to the log file.
1338 @kindex show logging
1340 Show the current values of the logging settings.
1344 @chapter @value{GDBN} Commands
1346 You can abbreviate a @value{GDBN} command to the first few letters of the command
1347 name, if that abbreviation is unambiguous; and you can repeat certain
1348 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1349 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1350 show you the alternatives available, if there is more than one possibility).
1353 * Command Syntax:: How to give commands to @value{GDBN}
1354 * Completion:: Command completion
1355 * Help:: How to ask @value{GDBN} for help
1358 @node Command Syntax
1359 @section Command Syntax
1361 A @value{GDBN} command is a single line of input. There is no limit on
1362 how long it can be. It starts with a command name, which is followed by
1363 arguments whose meaning depends on the command name. For example, the
1364 command @code{step} accepts an argument which is the number of times to
1365 step, as in @samp{step 5}. You can also use the @code{step} command
1366 with no arguments. Some commands do not allow any arguments.
1368 @cindex abbreviation
1369 @value{GDBN} command names may always be truncated if that abbreviation is
1370 unambiguous. Other possible command abbreviations are listed in the
1371 documentation for individual commands. In some cases, even ambiguous
1372 abbreviations are allowed; for example, @code{s} is specially defined as
1373 equivalent to @code{step} even though there are other commands whose
1374 names start with @code{s}. You can test abbreviations by using them as
1375 arguments to the @code{help} command.
1377 @cindex repeating commands
1378 @kindex RET @r{(repeat last command)}
1379 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1380 repeat the previous command. Certain commands (for example, @code{run})
1381 will not repeat this way; these are commands whose unintentional
1382 repetition might cause trouble and which you are unlikely to want to
1383 repeat. User-defined commands can disable this feature; see
1384 @ref{Define, dont-repeat}.
1386 The @code{list} and @code{x} commands, when you repeat them with
1387 @key{RET}, construct new arguments rather than repeating
1388 exactly as typed. This permits easy scanning of source or memory.
1390 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1391 output, in a way similar to the common utility @code{more}
1392 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1393 @key{RET} too many in this situation, @value{GDBN} disables command
1394 repetition after any command that generates this sort of display.
1396 @kindex # @r{(a comment)}
1398 Any text from a @kbd{#} to the end of the line is a comment; it does
1399 nothing. This is useful mainly in command files (@pxref{Command
1400 Files,,Command Files}).
1402 @cindex repeating command sequences
1403 @kindex Ctrl-o @r{(operate-and-get-next)}
1404 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1405 commands. This command accepts the current line, like @key{RET}, and
1406 then fetches the next line relative to the current line from the history
1410 @section Command Completion
1413 @cindex word completion
1414 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1415 only one possibility; it can also show you what the valid possibilities
1416 are for the next word in a command, at any time. This works for @value{GDBN}
1417 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1419 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1420 of a word. If there is only one possibility, @value{GDBN} fills in the
1421 word, and waits for you to finish the command (or press @key{RET} to
1422 enter it). For example, if you type
1424 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1425 @c complete accuracy in these examples; space introduced for clarity.
1426 @c If texinfo enhancements make it unnecessary, it would be nice to
1427 @c replace " @key" by "@key" in the following...
1429 (@value{GDBP}) info bre @key{TAB}
1433 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1434 the only @code{info} subcommand beginning with @samp{bre}:
1437 (@value{GDBP}) info breakpoints
1441 You can either press @key{RET} at this point, to run the @code{info
1442 breakpoints} command, or backspace and enter something else, if
1443 @samp{breakpoints} does not look like the command you expected. (If you
1444 were sure you wanted @code{info breakpoints} in the first place, you
1445 might as well just type @key{RET} immediately after @samp{info bre},
1446 to exploit command abbreviations rather than command completion).
1448 If there is more than one possibility for the next word when you press
1449 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1450 characters and try again, or just press @key{TAB} a second time;
1451 @value{GDBN} displays all the possible completions for that word. For
1452 example, you might want to set a breakpoint on a subroutine whose name
1453 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1454 just sounds the bell. Typing @key{TAB} again displays all the
1455 function names in your program that begin with those characters, for
1459 (@value{GDBP}) b make_ @key{TAB}
1460 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1461 make_a_section_from_file make_environ
1462 make_abs_section make_function_type
1463 make_blockvector make_pointer_type
1464 make_cleanup make_reference_type
1465 make_command make_symbol_completion_list
1466 (@value{GDBP}) b make_
1470 After displaying the available possibilities, @value{GDBN} copies your
1471 partial input (@samp{b make_} in the example) so you can finish the
1474 If you just want to see the list of alternatives in the first place, you
1475 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1476 means @kbd{@key{META} ?}. You can type this either by holding down a
1477 key designated as the @key{META} shift on your keyboard (if there is
1478 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1480 @cindex quotes in commands
1481 @cindex completion of quoted strings
1482 Sometimes the string you need, while logically a ``word'', may contain
1483 parentheses or other characters that @value{GDBN} normally excludes from
1484 its notion of a word. To permit word completion to work in this
1485 situation, you may enclose words in @code{'} (single quote marks) in
1486 @value{GDBN} commands.
1488 The most likely situation where you might need this is in typing the
1489 name of a C@t{++} function. This is because C@t{++} allows function
1490 overloading (multiple definitions of the same function, distinguished
1491 by argument type). For example, when you want to set a breakpoint you
1492 may need to distinguish whether you mean the version of @code{name}
1493 that takes an @code{int} parameter, @code{name(int)}, or the version
1494 that takes a @code{float} parameter, @code{name(float)}. To use the
1495 word-completion facilities in this situation, type a single quote
1496 @code{'} at the beginning of the function name. This alerts
1497 @value{GDBN} that it may need to consider more information than usual
1498 when you press @key{TAB} or @kbd{M-?} to request word completion:
1501 (@value{GDBP}) b 'bubble( @kbd{M-?}
1502 bubble(double,double) bubble(int,int)
1503 (@value{GDBP}) b 'bubble(
1506 In some cases, @value{GDBN} can tell that completing a name requires using
1507 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1508 completing as much as it can) if you do not type the quote in the first
1512 (@value{GDBP}) b bub @key{TAB}
1513 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1514 (@value{GDBP}) b 'bubble(
1518 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1519 you have not yet started typing the argument list when you ask for
1520 completion on an overloaded symbol.
1522 For more information about overloaded functions, see @ref{C Plus Plus
1523 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1524 overload-resolution off} to disable overload resolution;
1525 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1527 @cindex completion of structure field names
1528 @cindex structure field name completion
1529 @cindex completion of union field names
1530 @cindex union field name completion
1531 When completing in an expression which looks up a field in a
1532 structure, @value{GDBN} also tries@footnote{The completer can be
1533 confused by certain kinds of invalid expressions. Also, it only
1534 examines the static type of the expression, not the dynamic type.} to
1535 limit completions to the field names available in the type of the
1539 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1540 magic to_delete to_fputs to_put to_rewind
1541 to_data to_flush to_isatty to_read to_write
1545 This is because the @code{gdb_stdout} is a variable of the type
1546 @code{struct ui_file} that is defined in @value{GDBN} sources as
1553 ui_file_flush_ftype *to_flush;
1554 ui_file_write_ftype *to_write;
1555 ui_file_fputs_ftype *to_fputs;
1556 ui_file_read_ftype *to_read;
1557 ui_file_delete_ftype *to_delete;
1558 ui_file_isatty_ftype *to_isatty;
1559 ui_file_rewind_ftype *to_rewind;
1560 ui_file_put_ftype *to_put;
1567 @section Getting Help
1568 @cindex online documentation
1571 You can always ask @value{GDBN} itself for information on its commands,
1572 using the command @code{help}.
1575 @kindex h @r{(@code{help})}
1578 You can use @code{help} (abbreviated @code{h}) with no arguments to
1579 display a short list of named classes of commands:
1583 List of classes of commands:
1585 aliases -- Aliases of other commands
1586 breakpoints -- Making program stop at certain points
1587 data -- Examining data
1588 files -- Specifying and examining files
1589 internals -- Maintenance commands
1590 obscure -- Obscure features
1591 running -- Running the program
1592 stack -- Examining the stack
1593 status -- Status inquiries
1594 support -- Support facilities
1595 tracepoints -- Tracing of program execution without
1596 stopping the program
1597 user-defined -- User-defined commands
1599 Type "help" followed by a class name for a list of
1600 commands in that class.
1601 Type "help" followed by command name for full
1603 Command name abbreviations are allowed if unambiguous.
1606 @c the above line break eliminates huge line overfull...
1608 @item help @var{class}
1609 Using one of the general help classes as an argument, you can get a
1610 list of the individual commands in that class. For example, here is the
1611 help display for the class @code{status}:
1614 (@value{GDBP}) help status
1619 @c Line break in "show" line falsifies real output, but needed
1620 @c to fit in smallbook page size.
1621 info -- Generic command for showing things
1622 about the program being debugged
1623 show -- Generic command for showing things
1626 Type "help" followed by command name for full
1628 Command name abbreviations are allowed if unambiguous.
1632 @item help @var{command}
1633 With a command name as @code{help} argument, @value{GDBN} displays a
1634 short paragraph on how to use that command.
1637 @item apropos @var{args}
1638 The @code{apropos} command searches through all of the @value{GDBN}
1639 commands, and their documentation, for the regular expression specified in
1640 @var{args}. It prints out all matches found. For example:
1651 set symbol-reloading -- Set dynamic symbol table reloading
1652 multiple times in one run
1653 show symbol-reloading -- Show dynamic symbol table reloading
1654 multiple times in one run
1659 @item complete @var{args}
1660 The @code{complete @var{args}} command lists all the possible completions
1661 for the beginning of a command. Use @var{args} to specify the beginning of the
1662 command you want completed. For example:
1668 @noindent results in:
1679 @noindent This is intended for use by @sc{gnu} Emacs.
1682 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1683 and @code{show} to inquire about the state of your program, or the state
1684 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1685 manual introduces each of them in the appropriate context. The listings
1686 under @code{info} and under @code{show} in the Index point to
1687 all the sub-commands. @xref{Index}.
1692 @kindex i @r{(@code{info})}
1694 This command (abbreviated @code{i}) is for describing the state of your
1695 program. For example, you can show the arguments passed to a function
1696 with @code{info args}, list the registers currently in use with @code{info
1697 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1698 You can get a complete list of the @code{info} sub-commands with
1699 @w{@code{help info}}.
1703 You can assign the result of an expression to an environment variable with
1704 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1705 @code{set prompt $}.
1709 In contrast to @code{info}, @code{show} is for describing the state of
1710 @value{GDBN} itself.
1711 You can change most of the things you can @code{show}, by using the
1712 related command @code{set}; for example, you can control what number
1713 system is used for displays with @code{set radix}, or simply inquire
1714 which is currently in use with @code{show radix}.
1717 To display all the settable parameters and their current
1718 values, you can use @code{show} with no arguments; you may also use
1719 @code{info set}. Both commands produce the same display.
1720 @c FIXME: "info set" violates the rule that "info" is for state of
1721 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1722 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1726 Here are three miscellaneous @code{show} subcommands, all of which are
1727 exceptional in lacking corresponding @code{set} commands:
1730 @kindex show version
1731 @cindex @value{GDBN} version number
1733 Show what version of @value{GDBN} is running. You should include this
1734 information in @value{GDBN} bug-reports. If multiple versions of
1735 @value{GDBN} are in use at your site, you may need to determine which
1736 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1737 commands are introduced, and old ones may wither away. Also, many
1738 system vendors ship variant versions of @value{GDBN}, and there are
1739 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1740 The version number is the same as the one announced when you start
1743 @kindex show copying
1744 @kindex info copying
1745 @cindex display @value{GDBN} copyright
1748 Display information about permission for copying @value{GDBN}.
1750 @kindex show warranty
1751 @kindex info warranty
1753 @itemx info warranty
1754 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1755 if your version of @value{GDBN} comes with one.
1760 @chapter Running Programs Under @value{GDBN}
1762 When you run a program under @value{GDBN}, you must first generate
1763 debugging information when you compile it.
1765 You may start @value{GDBN} with its arguments, if any, in an environment
1766 of your choice. If you are doing native debugging, you may redirect
1767 your program's input and output, debug an already running process, or
1768 kill a child process.
1771 * Compilation:: Compiling for debugging
1772 * Starting:: Starting your program
1773 * Arguments:: Your program's arguments
1774 * Environment:: Your program's environment
1776 * Working Directory:: Your program's working directory
1777 * Input/Output:: Your program's input and output
1778 * Attach:: Debugging an already-running process
1779 * Kill Process:: Killing the child process
1781 * Inferiors:: Debugging multiple inferiors
1782 * Threads:: Debugging programs with multiple threads
1783 * Processes:: Debugging programs with multiple processes
1784 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1788 @section Compiling for Debugging
1790 In order to debug a program effectively, you need to generate
1791 debugging information when you compile it. This debugging information
1792 is stored in the object file; it describes the data type of each
1793 variable or function and the correspondence between source line numbers
1794 and addresses in the executable code.
1796 To request debugging information, specify the @samp{-g} option when you run
1799 Programs that are to be shipped to your customers are compiled with
1800 optimizations, using the @samp{-O} compiler option. However, many
1801 compilers are unable to handle the @samp{-g} and @samp{-O} options
1802 together. Using those compilers, you cannot generate optimized
1803 executables containing debugging information.
1805 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1806 without @samp{-O}, making it possible to debug optimized code. We
1807 recommend that you @emph{always} use @samp{-g} whenever you compile a
1808 program. You may think your program is correct, but there is no sense
1809 in pushing your luck.
1811 @cindex optimized code, debugging
1812 @cindex debugging optimized code
1813 When you debug a program compiled with @samp{-g -O}, remember that the
1814 optimizer is rearranging your code; the debugger shows you what is
1815 really there. Do not be too surprised when the execution path does not
1816 exactly match your source file! An extreme example: if you define a
1817 variable, but never use it, @value{GDBN} never sees that
1818 variable---because the compiler optimizes it out of existence.
1820 Some things do not work as well with @samp{-g -O} as with just
1821 @samp{-g}, particularly on machines with instruction scheduling. If in
1822 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1823 please report it to us as a bug (including a test case!).
1824 @xref{Variables}, for more information about debugging optimized code.
1826 Older versions of the @sc{gnu} C compiler permitted a variant option
1827 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1828 format; if your @sc{gnu} C compiler has this option, do not use it.
1830 @value{GDBN} knows about preprocessor macros and can show you their
1831 expansion (@pxref{Macros}). Most compilers do not include information
1832 about preprocessor macros in the debugging information if you specify
1833 the @option{-g} flag alone, because this information is rather large.
1834 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1835 provides macro information if you specify the options
1836 @option{-gdwarf-2} and @option{-g3}; the former option requests
1837 debugging information in the Dwarf 2 format, and the latter requests
1838 ``extra information''. In the future, we hope to find more compact
1839 ways to represent macro information, so that it can be included with
1844 @section Starting your Program
1850 @kindex r @r{(@code{run})}
1853 Use the @code{run} command to start your program under @value{GDBN}.
1854 You must first specify the program name (except on VxWorks) with an
1855 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1856 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1857 (@pxref{Files, ,Commands to Specify Files}).
1861 If you are running your program in an execution environment that
1862 supports processes, @code{run} creates an inferior process and makes
1863 that process run your program. In some environments without processes,
1864 @code{run} jumps to the start of your program. Other targets,
1865 like @samp{remote}, are always running. If you get an error
1866 message like this one:
1869 The "remote" target does not support "run".
1870 Try "help target" or "continue".
1874 then use @code{continue} to run your program. You may need @code{load}
1875 first (@pxref{load}).
1877 The execution of a program is affected by certain information it
1878 receives from its superior. @value{GDBN} provides ways to specify this
1879 information, which you must do @emph{before} starting your program. (You
1880 can change it after starting your program, but such changes only affect
1881 your program the next time you start it.) This information may be
1882 divided into four categories:
1885 @item The @emph{arguments.}
1886 Specify the arguments to give your program as the arguments of the
1887 @code{run} command. If a shell is available on your target, the shell
1888 is used to pass the arguments, so that you may use normal conventions
1889 (such as wildcard expansion or variable substitution) in describing
1891 In Unix systems, you can control which shell is used with the
1892 @code{SHELL} environment variable.
1893 @xref{Arguments, ,Your Program's Arguments}.
1895 @item The @emph{environment.}
1896 Your program normally inherits its environment from @value{GDBN}, but you can
1897 use the @value{GDBN} commands @code{set environment} and @code{unset
1898 environment} to change parts of the environment that affect
1899 your program. @xref{Environment, ,Your Program's Environment}.
1901 @item The @emph{working directory.}
1902 Your program inherits its working directory from @value{GDBN}. You can set
1903 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1904 @xref{Working Directory, ,Your Program's Working Directory}.
1906 @item The @emph{standard input and output.}
1907 Your program normally uses the same device for standard input and
1908 standard output as @value{GDBN} is using. You can redirect input and output
1909 in the @code{run} command line, or you can use the @code{tty} command to
1910 set a different device for your program.
1911 @xref{Input/Output, ,Your Program's Input and Output}.
1914 @emph{Warning:} While input and output redirection work, you cannot use
1915 pipes to pass the output of the program you are debugging to another
1916 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1920 When you issue the @code{run} command, your program begins to execute
1921 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1922 of how to arrange for your program to stop. Once your program has
1923 stopped, you may call functions in your program, using the @code{print}
1924 or @code{call} commands. @xref{Data, ,Examining Data}.
1926 If the modification time of your symbol file has changed since the last
1927 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1928 table, and reads it again. When it does this, @value{GDBN} tries to retain
1929 your current breakpoints.
1934 @cindex run to main procedure
1935 The name of the main procedure can vary from language to language.
1936 With C or C@t{++}, the main procedure name is always @code{main}, but
1937 other languages such as Ada do not require a specific name for their
1938 main procedure. The debugger provides a convenient way to start the
1939 execution of the program and to stop at the beginning of the main
1940 procedure, depending on the language used.
1942 The @samp{start} command does the equivalent of setting a temporary
1943 breakpoint at the beginning of the main procedure and then invoking
1944 the @samp{run} command.
1946 @cindex elaboration phase
1947 Some programs contain an @dfn{elaboration} phase where some startup code is
1948 executed before the main procedure is called. This depends on the
1949 languages used to write your program. In C@t{++}, for instance,
1950 constructors for static and global objects are executed before
1951 @code{main} is called. It is therefore possible that the debugger stops
1952 before reaching the main procedure. However, the temporary breakpoint
1953 will remain to halt execution.
1955 Specify the arguments to give to your program as arguments to the
1956 @samp{start} command. These arguments will be given verbatim to the
1957 underlying @samp{run} command. Note that the same arguments will be
1958 reused if no argument is provided during subsequent calls to
1959 @samp{start} or @samp{run}.
1961 It is sometimes necessary to debug the program during elaboration. In
1962 these cases, using the @code{start} command would stop the execution of
1963 your program too late, as the program would have already completed the
1964 elaboration phase. Under these circumstances, insert breakpoints in your
1965 elaboration code before running your program.
1967 @kindex set exec-wrapper
1968 @item set exec-wrapper @var{wrapper}
1969 @itemx show exec-wrapper
1970 @itemx unset exec-wrapper
1971 When @samp{exec-wrapper} is set, the specified wrapper is used to
1972 launch programs for debugging. @value{GDBN} starts your program
1973 with a shell command of the form @kbd{exec @var{wrapper}
1974 @var{program}}. Quoting is added to @var{program} and its
1975 arguments, but not to @var{wrapper}, so you should add quotes if
1976 appropriate for your shell. The wrapper runs until it executes
1977 your program, and then @value{GDBN} takes control.
1979 You can use any program that eventually calls @code{execve} with
1980 its arguments as a wrapper. Several standard Unix utilities do
1981 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
1982 with @code{exec "$@@"} will also work.
1984 For example, you can use @code{env} to pass an environment variable to
1985 the debugged program, without setting the variable in your shell's
1989 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
1993 This command is available when debugging locally on most targets, excluding
1994 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
1996 @kindex set disable-randomization
1997 @item set disable-randomization
1998 @itemx set disable-randomization on
1999 This option (enabled by default in @value{GDBN}) will turn off the native
2000 randomization of the virtual address space of the started program. This option
2001 is useful for multiple debugging sessions to make the execution better
2002 reproducible and memory addresses reusable across debugging sessions.
2004 This feature is implemented only on @sc{gnu}/Linux. You can get the same
2008 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2011 @item set disable-randomization off
2012 Leave the behavior of the started executable unchanged. Some bugs rear their
2013 ugly heads only when the program is loaded at certain addresses. If your bug
2014 disappears when you run the program under @value{GDBN}, that might be because
2015 @value{GDBN} by default disables the address randomization on platforms, such
2016 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2017 disable-randomization off} to try to reproduce such elusive bugs.
2019 The virtual address space randomization is implemented only on @sc{gnu}/Linux.
2020 It protects the programs against some kinds of security attacks. In these
2021 cases the attacker needs to know the exact location of a concrete executable
2022 code. Randomizing its location makes it impossible to inject jumps misusing
2023 a code at its expected addresses.
2025 Prelinking shared libraries provides a startup performance advantage but it
2026 makes addresses in these libraries predictable for privileged processes by
2027 having just unprivileged access at the target system. Reading the shared
2028 library binary gives enough information for assembling the malicious code
2029 misusing it. Still even a prelinked shared library can get loaded at a new
2030 random address just requiring the regular relocation process during the
2031 startup. Shared libraries not already prelinked are always loaded at
2032 a randomly chosen address.
2034 Position independent executables (PIE) contain position independent code
2035 similar to the shared libraries and therefore such executables get loaded at
2036 a randomly chosen address upon startup. PIE executables always load even
2037 already prelinked shared libraries at a random address. You can build such
2038 executable using @command{gcc -fPIE -pie}.
2040 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2041 (as long as the randomization is enabled).
2043 @item show disable-randomization
2044 Show the current setting of the explicit disable of the native randomization of
2045 the virtual address space of the started program.
2050 @section Your Program's Arguments
2052 @cindex arguments (to your program)
2053 The arguments to your program can be specified by the arguments of the
2055 They are passed to a shell, which expands wildcard characters and
2056 performs redirection of I/O, and thence to your program. Your
2057 @code{SHELL} environment variable (if it exists) specifies what shell
2058 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2059 the default shell (@file{/bin/sh} on Unix).
2061 On non-Unix systems, the program is usually invoked directly by
2062 @value{GDBN}, which emulates I/O redirection via the appropriate system
2063 calls, and the wildcard characters are expanded by the startup code of
2064 the program, not by the shell.
2066 @code{run} with no arguments uses the same arguments used by the previous
2067 @code{run}, or those set by the @code{set args} command.
2072 Specify the arguments to be used the next time your program is run. If
2073 @code{set args} has no arguments, @code{run} executes your program
2074 with no arguments. Once you have run your program with arguments,
2075 using @code{set args} before the next @code{run} is the only way to run
2076 it again without arguments.
2080 Show the arguments to give your program when it is started.
2084 @section Your Program's Environment
2086 @cindex environment (of your program)
2087 The @dfn{environment} consists of a set of environment variables and
2088 their values. Environment variables conventionally record such things as
2089 your user name, your home directory, your terminal type, and your search
2090 path for programs to run. Usually you set up environment variables with
2091 the shell and they are inherited by all the other programs you run. When
2092 debugging, it can be useful to try running your program with a modified
2093 environment without having to start @value{GDBN} over again.
2097 @item path @var{directory}
2098 Add @var{directory} to the front of the @code{PATH} environment variable
2099 (the search path for executables) that will be passed to your program.
2100 The value of @code{PATH} used by @value{GDBN} does not change.
2101 You may specify several directory names, separated by whitespace or by a
2102 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2103 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2104 is moved to the front, so it is searched sooner.
2106 You can use the string @samp{$cwd} to refer to whatever is the current
2107 working directory at the time @value{GDBN} searches the path. If you
2108 use @samp{.} instead, it refers to the directory where you executed the
2109 @code{path} command. @value{GDBN} replaces @samp{.} in the
2110 @var{directory} argument (with the current path) before adding
2111 @var{directory} to the search path.
2112 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2113 @c document that, since repeating it would be a no-op.
2117 Display the list of search paths for executables (the @code{PATH}
2118 environment variable).
2120 @kindex show environment
2121 @item show environment @r{[}@var{varname}@r{]}
2122 Print the value of environment variable @var{varname} to be given to
2123 your program when it starts. If you do not supply @var{varname},
2124 print the names and values of all environment variables to be given to
2125 your program. You can abbreviate @code{environment} as @code{env}.
2127 @kindex set environment
2128 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2129 Set environment variable @var{varname} to @var{value}. The value
2130 changes for your program only, not for @value{GDBN} itself. @var{value} may
2131 be any string; the values of environment variables are just strings, and
2132 any interpretation is supplied by your program itself. The @var{value}
2133 parameter is optional; if it is eliminated, the variable is set to a
2135 @c "any string" here does not include leading, trailing
2136 @c blanks. Gnu asks: does anyone care?
2138 For example, this command:
2145 tells the debugged program, when subsequently run, that its user is named
2146 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2147 are not actually required.)
2149 @kindex unset environment
2150 @item unset environment @var{varname}
2151 Remove variable @var{varname} from the environment to be passed to your
2152 program. This is different from @samp{set env @var{varname} =};
2153 @code{unset environment} removes the variable from the environment,
2154 rather than assigning it an empty value.
2157 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2159 by your @code{SHELL} environment variable if it exists (or
2160 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2161 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2162 @file{.bashrc} for BASH---any variables you set in that file affect
2163 your program. You may wish to move setting of environment variables to
2164 files that are only run when you sign on, such as @file{.login} or
2167 @node Working Directory
2168 @section Your Program's Working Directory
2170 @cindex working directory (of your program)
2171 Each time you start your program with @code{run}, it inherits its
2172 working directory from the current working directory of @value{GDBN}.
2173 The @value{GDBN} working directory is initially whatever it inherited
2174 from its parent process (typically the shell), but you can specify a new
2175 working directory in @value{GDBN} with the @code{cd} command.
2177 The @value{GDBN} working directory also serves as a default for the commands
2178 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2183 @cindex change working directory
2184 @item cd @var{directory}
2185 Set the @value{GDBN} working directory to @var{directory}.
2189 Print the @value{GDBN} working directory.
2192 It is generally impossible to find the current working directory of
2193 the process being debugged (since a program can change its directory
2194 during its run). If you work on a system where @value{GDBN} is
2195 configured with the @file{/proc} support, you can use the @code{info
2196 proc} command (@pxref{SVR4 Process Information}) to find out the
2197 current working directory of the debuggee.
2200 @section Your Program's Input and Output
2205 By default, the program you run under @value{GDBN} does input and output to
2206 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2207 to its own terminal modes to interact with you, but it records the terminal
2208 modes your program was using and switches back to them when you continue
2209 running your program.
2212 @kindex info terminal
2214 Displays information recorded by @value{GDBN} about the terminal modes your
2218 You can redirect your program's input and/or output using shell
2219 redirection with the @code{run} command. For example,
2226 starts your program, diverting its output to the file @file{outfile}.
2229 @cindex controlling terminal
2230 Another way to specify where your program should do input and output is
2231 with the @code{tty} command. This command accepts a file name as
2232 argument, and causes this file to be the default for future @code{run}
2233 commands. It also resets the controlling terminal for the child
2234 process, for future @code{run} commands. For example,
2241 directs that processes started with subsequent @code{run} commands
2242 default to do input and output on the terminal @file{/dev/ttyb} and have
2243 that as their controlling terminal.
2245 An explicit redirection in @code{run} overrides the @code{tty} command's
2246 effect on the input/output device, but not its effect on the controlling
2249 When you use the @code{tty} command or redirect input in the @code{run}
2250 command, only the input @emph{for your program} is affected. The input
2251 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2252 for @code{set inferior-tty}.
2254 @cindex inferior tty
2255 @cindex set inferior controlling terminal
2256 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2257 display the name of the terminal that will be used for future runs of your
2261 @item set inferior-tty /dev/ttyb
2262 @kindex set inferior-tty
2263 Set the tty for the program being debugged to /dev/ttyb.
2265 @item show inferior-tty
2266 @kindex show inferior-tty
2267 Show the current tty for the program being debugged.
2271 @section Debugging an Already-running Process
2276 @item attach @var{process-id}
2277 This command attaches to a running process---one that was started
2278 outside @value{GDBN}. (@code{info files} shows your active
2279 targets.) The command takes as argument a process ID. The usual way to
2280 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2281 or with the @samp{jobs -l} shell command.
2283 @code{attach} does not repeat if you press @key{RET} a second time after
2284 executing the command.
2287 To use @code{attach}, your program must be running in an environment
2288 which supports processes; for example, @code{attach} does not work for
2289 programs on bare-board targets that lack an operating system. You must
2290 also have permission to send the process a signal.
2292 When you use @code{attach}, the debugger finds the program running in
2293 the process first by looking in the current working directory, then (if
2294 the program is not found) by using the source file search path
2295 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2296 the @code{file} command to load the program. @xref{Files, ,Commands to
2299 The first thing @value{GDBN} does after arranging to debug the specified
2300 process is to stop it. You can examine and modify an attached process
2301 with all the @value{GDBN} commands that are ordinarily available when
2302 you start processes with @code{run}. You can insert breakpoints; you
2303 can step and continue; you can modify storage. If you would rather the
2304 process continue running, you may use the @code{continue} command after
2305 attaching @value{GDBN} to the process.
2310 When you have finished debugging the attached process, you can use the
2311 @code{detach} command to release it from @value{GDBN} control. Detaching
2312 the process continues its execution. After the @code{detach} command,
2313 that process and @value{GDBN} become completely independent once more, and you
2314 are ready to @code{attach} another process or start one with @code{run}.
2315 @code{detach} does not repeat if you press @key{RET} again after
2316 executing the command.
2319 If you exit @value{GDBN} while you have an attached process, you detach
2320 that process. If you use the @code{run} command, you kill that process.
2321 By default, @value{GDBN} asks for confirmation if you try to do either of these
2322 things; you can control whether or not you need to confirm by using the
2323 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2327 @section Killing the Child Process
2332 Kill the child process in which your program is running under @value{GDBN}.
2335 This command is useful if you wish to debug a core dump instead of a
2336 running process. @value{GDBN} ignores any core dump file while your program
2339 On some operating systems, a program cannot be executed outside @value{GDBN}
2340 while you have breakpoints set on it inside @value{GDBN}. You can use the
2341 @code{kill} command in this situation to permit running your program
2342 outside the debugger.
2344 The @code{kill} command is also useful if you wish to recompile and
2345 relink your program, since on many systems it is impossible to modify an
2346 executable file while it is running in a process. In this case, when you
2347 next type @code{run}, @value{GDBN} notices that the file has changed, and
2348 reads the symbol table again (while trying to preserve your current
2349 breakpoint settings).
2352 @section Debugging Multiple Inferiors
2354 Some @value{GDBN} targets are able to run multiple processes created
2355 from a single executable. This can happen, for instance, with an
2356 embedded system reporting back several processes via the remote
2360 @value{GDBN} represents the state of each program execution with an
2361 object called an @dfn{inferior}. An inferior typically corresponds to
2362 a process, but is more general and applies also to targets that do not
2363 have processes. Inferiors may be created before a process runs, and
2364 may (in future) be retained after a process exits. Each run of an
2365 executable creates a new inferior, as does each attachment to an
2366 existing process. Inferiors have unique identifiers that are
2367 different from process ids, and may optionally be named as well.
2368 Usually each inferior will also have its own distinct address space,
2369 although some embedded targets may have several inferiors running in
2370 different parts of a single space.
2372 Each inferior may in turn have multiple threads running in it.
2374 To find out what inferiors exist at any moment, use @code{info inferiors}:
2377 @kindex info inferiors
2378 @item info inferiors
2379 Print a list of all inferiors currently being managed by @value{GDBN}.
2381 @kindex set print inferior-events
2382 @cindex print messages on inferior start and exit
2383 @item set print inferior-events
2384 @itemx set print inferior-events on
2385 @itemx set print inferior-events off
2386 The @code{set print inferior-events} command allows you to enable or
2387 disable printing of messages when @value{GDBN} notices that new
2388 inferiors have started or that inferiors have exited or have been
2389 detached. By default, these messages will not be printed.
2391 @kindex show print inferior-events
2392 @item show print inferior-events
2393 Show whether messages will be printed when @value{GDBN} detects that
2394 inferiors have started, exited or have been detached.
2398 @section Debugging Programs with Multiple Threads
2400 @cindex threads of execution
2401 @cindex multiple threads
2402 @cindex switching threads
2403 In some operating systems, such as HP-UX and Solaris, a single program
2404 may have more than one @dfn{thread} of execution. The precise semantics
2405 of threads differ from one operating system to another, but in general
2406 the threads of a single program are akin to multiple processes---except
2407 that they share one address space (that is, they can all examine and
2408 modify the same variables). On the other hand, each thread has its own
2409 registers and execution stack, and perhaps private memory.
2411 @value{GDBN} provides these facilities for debugging multi-thread
2415 @item automatic notification of new threads
2416 @item @samp{thread @var{threadno}}, a command to switch among threads
2417 @item @samp{info threads}, a command to inquire about existing threads
2418 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2419 a command to apply a command to a list of threads
2420 @item thread-specific breakpoints
2421 @item @samp{set print thread-events}, which controls printing of
2422 messages on thread start and exit.
2426 @emph{Warning:} These facilities are not yet available on every
2427 @value{GDBN} configuration where the operating system supports threads.
2428 If your @value{GDBN} does not support threads, these commands have no
2429 effect. For example, a system without thread support shows no output
2430 from @samp{info threads}, and always rejects the @code{thread} command,
2434 (@value{GDBP}) info threads
2435 (@value{GDBP}) thread 1
2436 Thread ID 1 not known. Use the "info threads" command to
2437 see the IDs of currently known threads.
2439 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2440 @c doesn't support threads"?
2443 @cindex focus of debugging
2444 @cindex current thread
2445 The @value{GDBN} thread debugging facility allows you to observe all
2446 threads while your program runs---but whenever @value{GDBN} takes
2447 control, one thread in particular is always the focus of debugging.
2448 This thread is called the @dfn{current thread}. Debugging commands show
2449 program information from the perspective of the current thread.
2451 @cindex @code{New} @var{systag} message
2452 @cindex thread identifier (system)
2453 @c FIXME-implementors!! It would be more helpful if the [New...] message
2454 @c included GDB's numeric thread handle, so you could just go to that
2455 @c thread without first checking `info threads'.
2456 Whenever @value{GDBN} detects a new thread in your program, it displays
2457 the target system's identification for the thread with a message in the
2458 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2459 whose form varies depending on the particular system. For example, on
2460 @sc{gnu}/Linux, you might see
2463 [New Thread 46912507313328 (LWP 25582)]
2467 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2468 the @var{systag} is simply something like @samp{process 368}, with no
2471 @c FIXME!! (1) Does the [New...] message appear even for the very first
2472 @c thread of a program, or does it only appear for the
2473 @c second---i.e.@: when it becomes obvious we have a multithread
2475 @c (2) *Is* there necessarily a first thread always? Or do some
2476 @c multithread systems permit starting a program with multiple
2477 @c threads ab initio?
2479 @cindex thread number
2480 @cindex thread identifier (GDB)
2481 For debugging purposes, @value{GDBN} associates its own thread
2482 number---always a single integer---with each thread in your program.
2485 @kindex info threads
2487 Display a summary of all threads currently in your
2488 program. @value{GDBN} displays for each thread (in this order):
2492 the thread number assigned by @value{GDBN}
2495 the target system's thread identifier (@var{systag})
2498 the current stack frame summary for that thread
2502 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2503 indicates the current thread.
2507 @c end table here to get a little more width for example
2510 (@value{GDBP}) info threads
2511 3 process 35 thread 27 0x34e5 in sigpause ()
2512 2 process 35 thread 23 0x34e5 in sigpause ()
2513 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2519 @cindex debugging multithreaded programs (on HP-UX)
2520 @cindex thread identifier (GDB), on HP-UX
2521 For debugging purposes, @value{GDBN} associates its own thread
2522 number---a small integer assigned in thread-creation order---with each
2523 thread in your program.
2525 @cindex @code{New} @var{systag} message, on HP-UX
2526 @cindex thread identifier (system), on HP-UX
2527 @c FIXME-implementors!! It would be more helpful if the [New...] message
2528 @c included GDB's numeric thread handle, so you could just go to that
2529 @c thread without first checking `info threads'.
2530 Whenever @value{GDBN} detects a new thread in your program, it displays
2531 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2532 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2533 whose form varies depending on the particular system. For example, on
2537 [New thread 2 (system thread 26594)]
2541 when @value{GDBN} notices a new thread.
2544 @kindex info threads (HP-UX)
2546 Display a summary of all threads currently in your
2547 program. @value{GDBN} displays for each thread (in this order):
2550 @item the thread number assigned by @value{GDBN}
2552 @item the target system's thread identifier (@var{systag})
2554 @item the current stack frame summary for that thread
2558 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2559 indicates the current thread.
2563 @c end table here to get a little more width for example
2566 (@value{GDBP}) info threads
2567 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2569 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2570 from /usr/lib/libc.2
2571 1 system thread 27905 0x7b003498 in _brk () \@*
2572 from /usr/lib/libc.2
2575 On Solaris, you can display more information about user threads with a
2576 Solaris-specific command:
2579 @item maint info sol-threads
2580 @kindex maint info sol-threads
2581 @cindex thread info (Solaris)
2582 Display info on Solaris user threads.
2586 @kindex thread @var{threadno}
2587 @item thread @var{threadno}
2588 Make thread number @var{threadno} the current thread. The command
2589 argument @var{threadno} is the internal @value{GDBN} thread number, as
2590 shown in the first field of the @samp{info threads} display.
2591 @value{GDBN} responds by displaying the system identifier of the thread
2592 you selected, and its current stack frame summary:
2595 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2596 (@value{GDBP}) thread 2
2597 [Switching to process 35 thread 23]
2598 0x34e5 in sigpause ()
2602 As with the @samp{[New @dots{}]} message, the form of the text after
2603 @samp{Switching to} depends on your system's conventions for identifying
2606 @kindex thread apply
2607 @cindex apply command to several threads
2608 @item thread apply [@var{threadno}] [@var{all}] @var{command}
2609 The @code{thread apply} command allows you to apply the named
2610 @var{command} to one or more threads. Specify the numbers of the
2611 threads that you want affected with the command argument
2612 @var{threadno}. It can be a single thread number, one of the numbers
2613 shown in the first field of the @samp{info threads} display; or it
2614 could be a range of thread numbers, as in @code{2-4}. To apply a
2615 command to all threads, type @kbd{thread apply all @var{command}}.
2617 @kindex set print thread-events
2618 @cindex print messages on thread start and exit
2619 @item set print thread-events
2620 @itemx set print thread-events on
2621 @itemx set print thread-events off
2622 The @code{set print thread-events} command allows you to enable or
2623 disable printing of messages when @value{GDBN} notices that new threads have
2624 started or that threads have exited. By default, these messages will
2625 be printed if detection of these events is supported by the target.
2626 Note that these messages cannot be disabled on all targets.
2628 @kindex show print thread-events
2629 @item show print thread-events
2630 Show whether messages will be printed when @value{GDBN} detects that threads
2631 have started and exited.
2634 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2635 more information about how @value{GDBN} behaves when you stop and start
2636 programs with multiple threads.
2638 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2639 watchpoints in programs with multiple threads.
2642 @section Debugging Programs with Multiple Processes
2644 @cindex fork, debugging programs which call
2645 @cindex multiple processes
2646 @cindex processes, multiple
2647 On most systems, @value{GDBN} has no special support for debugging
2648 programs which create additional processes using the @code{fork}
2649 function. When a program forks, @value{GDBN} will continue to debug the
2650 parent process and the child process will run unimpeded. If you have
2651 set a breakpoint in any code which the child then executes, the child
2652 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2653 will cause it to terminate.
2655 However, if you want to debug the child process there is a workaround
2656 which isn't too painful. Put a call to @code{sleep} in the code which
2657 the child process executes after the fork. It may be useful to sleep
2658 only if a certain environment variable is set, or a certain file exists,
2659 so that the delay need not occur when you don't want to run @value{GDBN}
2660 on the child. While the child is sleeping, use the @code{ps} program to
2661 get its process ID. Then tell @value{GDBN} (a new invocation of
2662 @value{GDBN} if you are also debugging the parent process) to attach to
2663 the child process (@pxref{Attach}). From that point on you can debug
2664 the child process just like any other process which you attached to.
2666 On some systems, @value{GDBN} provides support for debugging programs that
2667 create additional processes using the @code{fork} or @code{vfork} functions.
2668 Currently, the only platforms with this feature are HP-UX (11.x and later
2669 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2671 By default, when a program forks, @value{GDBN} will continue to debug
2672 the parent process and the child process will run unimpeded.
2674 If you want to follow the child process instead of the parent process,
2675 use the command @w{@code{set follow-fork-mode}}.
2678 @kindex set follow-fork-mode
2679 @item set follow-fork-mode @var{mode}
2680 Set the debugger response to a program call of @code{fork} or
2681 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2682 process. The @var{mode} argument can be:
2686 The original process is debugged after a fork. The child process runs
2687 unimpeded. This is the default.
2690 The new process is debugged after a fork. The parent process runs
2695 @kindex show follow-fork-mode
2696 @item show follow-fork-mode
2697 Display the current debugger response to a @code{fork} or @code{vfork} call.
2700 @cindex debugging multiple processes
2701 On Linux, if you want to debug both the parent and child processes, use the
2702 command @w{@code{set detach-on-fork}}.
2705 @kindex set detach-on-fork
2706 @item set detach-on-fork @var{mode}
2707 Tells gdb whether to detach one of the processes after a fork, or
2708 retain debugger control over them both.
2712 The child process (or parent process, depending on the value of
2713 @code{follow-fork-mode}) will be detached and allowed to run
2714 independently. This is the default.
2717 Both processes will be held under the control of @value{GDBN}.
2718 One process (child or parent, depending on the value of
2719 @code{follow-fork-mode}) is debugged as usual, while the other
2724 @kindex show detach-on-fork
2725 @item show detach-on-fork
2726 Show whether detach-on-fork mode is on/off.
2729 If you choose to set @samp{detach-on-fork} mode off, then
2730 @value{GDBN} will retain control of all forked processes (including
2731 nested forks). You can list the forked processes under the control of
2732 @value{GDBN} by using the @w{@code{info forks}} command, and switch
2733 from one fork to another by using the @w{@code{fork}} command.
2738 Print a list of all forked processes under the control of @value{GDBN}.
2739 The listing will include a fork id, a process id, and the current
2740 position (program counter) of the process.
2742 @kindex fork @var{fork-id}
2743 @item fork @var{fork-id}
2744 Make fork number @var{fork-id} the current process. The argument
2745 @var{fork-id} is the internal fork number assigned by @value{GDBN},
2746 as shown in the first field of the @samp{info forks} display.
2748 @kindex process @var{process-id}
2749 @item process @var{process-id}
2750 Make process number @var{process-id} the current process. The
2751 argument @var{process-id} must be one that is listed in the output of
2756 To quit debugging one of the forked processes, you can either detach
2757 from it by using the @w{@code{detach fork}} command (allowing it to
2758 run independently), or delete (and kill) it using the
2759 @w{@code{delete fork}} command.
2762 @kindex detach fork @var{fork-id}
2763 @item detach fork @var{fork-id}
2764 Detach from the process identified by @value{GDBN} fork number
2765 @var{fork-id}, and remove it from the fork list. The process will be
2766 allowed to run independently.
2768 @kindex delete fork @var{fork-id}
2769 @item delete fork @var{fork-id}
2770 Kill the process identified by @value{GDBN} fork number @var{fork-id},
2771 and remove it from the fork list.
2775 If you ask to debug a child process and a @code{vfork} is followed by an
2776 @code{exec}, @value{GDBN} executes the new target up to the first
2777 breakpoint in the new target. If you have a breakpoint set on
2778 @code{main} in your original program, the breakpoint will also be set on
2779 the child process's @code{main}.
2781 When a child process is spawned by @code{vfork}, you cannot debug the
2782 child or parent until an @code{exec} call completes.
2784 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2785 call executes, the new target restarts. To restart the parent process,
2786 use the @code{file} command with the parent executable name as its
2789 You can use the @code{catch} command to make @value{GDBN} stop whenever
2790 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2791 Catchpoints, ,Setting Catchpoints}.
2793 @node Checkpoint/Restart
2794 @section Setting a @emph{Bookmark} to Return to Later
2799 @cindex snapshot of a process
2800 @cindex rewind program state
2802 On certain operating systems@footnote{Currently, only
2803 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
2804 program's state, called a @dfn{checkpoint}, and come back to it
2807 Returning to a checkpoint effectively undoes everything that has
2808 happened in the program since the @code{checkpoint} was saved. This
2809 includes changes in memory, registers, and even (within some limits)
2810 system state. Effectively, it is like going back in time to the
2811 moment when the checkpoint was saved.
2813 Thus, if you're stepping thru a program and you think you're
2814 getting close to the point where things go wrong, you can save
2815 a checkpoint. Then, if you accidentally go too far and miss
2816 the critical statement, instead of having to restart your program
2817 from the beginning, you can just go back to the checkpoint and
2818 start again from there.
2820 This can be especially useful if it takes a lot of time or
2821 steps to reach the point where you think the bug occurs.
2823 To use the @code{checkpoint}/@code{restart} method of debugging:
2828 Save a snapshot of the debugged program's current execution state.
2829 The @code{checkpoint} command takes no arguments, but each checkpoint
2830 is assigned a small integer id, similar to a breakpoint id.
2832 @kindex info checkpoints
2833 @item info checkpoints
2834 List the checkpoints that have been saved in the current debugging
2835 session. For each checkpoint, the following information will be
2842 @item Source line, or label
2845 @kindex restart @var{checkpoint-id}
2846 @item restart @var{checkpoint-id}
2847 Restore the program state that was saved as checkpoint number
2848 @var{checkpoint-id}. All program variables, registers, stack frames
2849 etc.@: will be returned to the values that they had when the checkpoint
2850 was saved. In essence, gdb will ``wind back the clock'' to the point
2851 in time when the checkpoint was saved.
2853 Note that breakpoints, @value{GDBN} variables, command history etc.
2854 are not affected by restoring a checkpoint. In general, a checkpoint
2855 only restores things that reside in the program being debugged, not in
2858 @kindex delete checkpoint @var{checkpoint-id}
2859 @item delete checkpoint @var{checkpoint-id}
2860 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
2864 Returning to a previously saved checkpoint will restore the user state
2865 of the program being debugged, plus a significant subset of the system
2866 (OS) state, including file pointers. It won't ``un-write'' data from
2867 a file, but it will rewind the file pointer to the previous location,
2868 so that the previously written data can be overwritten. For files
2869 opened in read mode, the pointer will also be restored so that the
2870 previously read data can be read again.
2872 Of course, characters that have been sent to a printer (or other
2873 external device) cannot be ``snatched back'', and characters received
2874 from eg.@: a serial device can be removed from internal program buffers,
2875 but they cannot be ``pushed back'' into the serial pipeline, ready to
2876 be received again. Similarly, the actual contents of files that have
2877 been changed cannot be restored (at this time).
2879 However, within those constraints, you actually can ``rewind'' your
2880 program to a previously saved point in time, and begin debugging it
2881 again --- and you can change the course of events so as to debug a
2882 different execution path this time.
2884 @cindex checkpoints and process id
2885 Finally, there is one bit of internal program state that will be
2886 different when you return to a checkpoint --- the program's process
2887 id. Each checkpoint will have a unique process id (or @var{pid}),
2888 and each will be different from the program's original @var{pid}.
2889 If your program has saved a local copy of its process id, this could
2890 potentially pose a problem.
2892 @subsection A Non-obvious Benefit of Using Checkpoints
2894 On some systems such as @sc{gnu}/Linux, address space randomization
2895 is performed on new processes for security reasons. This makes it
2896 difficult or impossible to set a breakpoint, or watchpoint, on an
2897 absolute address if you have to restart the program, since the
2898 absolute location of a symbol will change from one execution to the
2901 A checkpoint, however, is an @emph{identical} copy of a process.
2902 Therefore if you create a checkpoint at (eg.@:) the start of main,
2903 and simply return to that checkpoint instead of restarting the
2904 process, you can avoid the effects of address randomization and
2905 your symbols will all stay in the same place.
2908 @chapter Stopping and Continuing
2910 The principal purposes of using a debugger are so that you can stop your
2911 program before it terminates; or so that, if your program runs into
2912 trouble, you can investigate and find out why.
2914 Inside @value{GDBN}, your program may stop for any of several reasons,
2915 such as a signal, a breakpoint, or reaching a new line after a
2916 @value{GDBN} command such as @code{step}. You may then examine and
2917 change variables, set new breakpoints or remove old ones, and then
2918 continue execution. Usually, the messages shown by @value{GDBN} provide
2919 ample explanation of the status of your program---but you can also
2920 explicitly request this information at any time.
2923 @kindex info program
2925 Display information about the status of your program: whether it is
2926 running or not, what process it is, and why it stopped.
2930 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2931 * Continuing and Stepping:: Resuming execution
2933 * Thread Stops:: Stopping and starting multi-thread programs
2937 @section Breakpoints, Watchpoints, and Catchpoints
2940 A @dfn{breakpoint} makes your program stop whenever a certain point in
2941 the program is reached. For each breakpoint, you can add conditions to
2942 control in finer detail whether your program stops. You can set
2943 breakpoints with the @code{break} command and its variants (@pxref{Set
2944 Breaks, ,Setting Breakpoints}), to specify the place where your program
2945 should stop by line number, function name or exact address in the
2948 On some systems, you can set breakpoints in shared libraries before
2949 the executable is run. There is a minor limitation on HP-UX systems:
2950 you must wait until the executable is run in order to set breakpoints
2951 in shared library routines that are not called directly by the program
2952 (for example, routines that are arguments in a @code{pthread_create}
2956 @cindex data breakpoints
2957 @cindex memory tracing
2958 @cindex breakpoint on memory address
2959 @cindex breakpoint on variable modification
2960 A @dfn{watchpoint} is a special breakpoint that stops your program
2961 when the value of an expression changes. The expression may be a value
2962 of a variable, or it could involve values of one or more variables
2963 combined by operators, such as @samp{a + b}. This is sometimes called
2964 @dfn{data breakpoints}. You must use a different command to set
2965 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
2966 from that, you can manage a watchpoint like any other breakpoint: you
2967 enable, disable, and delete both breakpoints and watchpoints using the
2970 You can arrange to have values from your program displayed automatically
2971 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2975 @cindex breakpoint on events
2976 A @dfn{catchpoint} is another special breakpoint that stops your program
2977 when a certain kind of event occurs, such as the throwing of a C@t{++}
2978 exception or the loading of a library. As with watchpoints, you use a
2979 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
2980 Catchpoints}), but aside from that, you can manage a catchpoint like any
2981 other breakpoint. (To stop when your program receives a signal, use the
2982 @code{handle} command; see @ref{Signals, ,Signals}.)
2984 @cindex breakpoint numbers
2985 @cindex numbers for breakpoints
2986 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
2987 catchpoint when you create it; these numbers are successive integers
2988 starting with one. In many of the commands for controlling various
2989 features of breakpoints you use the breakpoint number to say which
2990 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
2991 @dfn{disabled}; if disabled, it has no effect on your program until you
2994 @cindex breakpoint ranges
2995 @cindex ranges of breakpoints
2996 Some @value{GDBN} commands accept a range of breakpoints on which to
2997 operate. A breakpoint range is either a single breakpoint number, like
2998 @samp{5}, or two such numbers, in increasing order, separated by a
2999 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3000 all breakpoints in that range are operated on.
3003 * Set Breaks:: Setting breakpoints
3004 * Set Watchpoints:: Setting watchpoints
3005 * Set Catchpoints:: Setting catchpoints
3006 * Delete Breaks:: Deleting breakpoints
3007 * Disabling:: Disabling breakpoints
3008 * Conditions:: Break conditions
3009 * Break Commands:: Breakpoint command lists
3010 * Error in Breakpoints:: ``Cannot insert breakpoints''
3011 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3015 @subsection Setting Breakpoints
3017 @c FIXME LMB what does GDB do if no code on line of breakpt?
3018 @c consider in particular declaration with/without initialization.
3020 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3023 @kindex b @r{(@code{break})}
3024 @vindex $bpnum@r{, convenience variable}
3025 @cindex latest breakpoint
3026 Breakpoints are set with the @code{break} command (abbreviated
3027 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3028 number of the breakpoint you've set most recently; see @ref{Convenience
3029 Vars,, Convenience Variables}, for a discussion of what you can do with
3030 convenience variables.
3033 @item break @var{location}
3034 Set a breakpoint at the given @var{location}, which can specify a
3035 function name, a line number, or an address of an instruction.
3036 (@xref{Specify Location}, for a list of all the possible ways to
3037 specify a @var{location}.) The breakpoint will stop your program just
3038 before it executes any of the code in the specified @var{location}.
3040 When using source languages that permit overloading of symbols, such as
3041 C@t{++}, a function name may refer to more than one possible place to break.
3042 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3046 When called without any arguments, @code{break} sets a breakpoint at
3047 the next instruction to be executed in the selected stack frame
3048 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3049 innermost, this makes your program stop as soon as control
3050 returns to that frame. This is similar to the effect of a
3051 @code{finish} command in the frame inside the selected frame---except
3052 that @code{finish} does not leave an active breakpoint. If you use
3053 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3054 the next time it reaches the current location; this may be useful
3057 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3058 least one instruction has been executed. If it did not do this, you
3059 would be unable to proceed past a breakpoint without first disabling the
3060 breakpoint. This rule applies whether or not the breakpoint already
3061 existed when your program stopped.
3063 @item break @dots{} if @var{cond}
3064 Set a breakpoint with condition @var{cond}; evaluate the expression
3065 @var{cond} each time the breakpoint is reached, and stop only if the
3066 value is nonzero---that is, if @var{cond} evaluates as true.
3067 @samp{@dots{}} stands for one of the possible arguments described
3068 above (or no argument) specifying where to break. @xref{Conditions,
3069 ,Break Conditions}, for more information on breakpoint conditions.
3072 @item tbreak @var{args}
3073 Set a breakpoint enabled only for one stop. @var{args} are the
3074 same as for the @code{break} command, and the breakpoint is set in the same
3075 way, but the breakpoint is automatically deleted after the first time your
3076 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3079 @cindex hardware breakpoints
3080 @item hbreak @var{args}
3081 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3082 @code{break} command and the breakpoint is set in the same way, but the
3083 breakpoint requires hardware support and some target hardware may not
3084 have this support. The main purpose of this is EPROM/ROM code
3085 debugging, so you can set a breakpoint at an instruction without
3086 changing the instruction. This can be used with the new trap-generation
3087 provided by SPARClite DSU and most x86-based targets. These targets
3088 will generate traps when a program accesses some data or instruction
3089 address that is assigned to the debug registers. However the hardware
3090 breakpoint registers can take a limited number of breakpoints. For
3091 example, on the DSU, only two data breakpoints can be set at a time, and
3092 @value{GDBN} will reject this command if more than two are used. Delete
3093 or disable unused hardware breakpoints before setting new ones
3094 (@pxref{Disabling, ,Disabling Breakpoints}).
3095 @xref{Conditions, ,Break Conditions}.
3096 For remote targets, you can restrict the number of hardware
3097 breakpoints @value{GDBN} will use, see @ref{set remote
3098 hardware-breakpoint-limit}.
3101 @item thbreak @var{args}
3102 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3103 are the same as for the @code{hbreak} command and the breakpoint is set in
3104 the same way. However, like the @code{tbreak} command,
3105 the breakpoint is automatically deleted after the
3106 first time your program stops there. Also, like the @code{hbreak}
3107 command, the breakpoint requires hardware support and some target hardware
3108 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3109 See also @ref{Conditions, ,Break Conditions}.
3112 @cindex regular expression
3113 @cindex breakpoints in functions matching a regexp
3114 @cindex set breakpoints in many functions
3115 @item rbreak @var{regex}
3116 Set breakpoints on all functions matching the regular expression
3117 @var{regex}. This command sets an unconditional breakpoint on all
3118 matches, printing a list of all breakpoints it set. Once these
3119 breakpoints are set, they are treated just like the breakpoints set with
3120 the @code{break} command. You can delete them, disable them, or make
3121 them conditional the same way as any other breakpoint.
3123 The syntax of the regular expression is the standard one used with tools
3124 like @file{grep}. Note that this is different from the syntax used by
3125 shells, so for instance @code{foo*} matches all functions that include
3126 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3127 @code{.*} leading and trailing the regular expression you supply, so to
3128 match only functions that begin with @code{foo}, use @code{^foo}.
3130 @cindex non-member C@t{++} functions, set breakpoint in
3131 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3132 breakpoints on overloaded functions that are not members of any special
3135 @cindex set breakpoints on all functions
3136 The @code{rbreak} command can be used to set breakpoints in
3137 @strong{all} the functions in a program, like this:
3140 (@value{GDBP}) rbreak .
3143 @kindex info breakpoints
3144 @cindex @code{$_} and @code{info breakpoints}
3145 @item info breakpoints @r{[}@var{n}@r{]}
3146 @itemx info break @r{[}@var{n}@r{]}
3147 @itemx info watchpoints @r{[}@var{n}@r{]}
3148 Print a table of all breakpoints, watchpoints, and catchpoints set and
3149 not deleted. Optional argument @var{n} means print information only
3150 about the specified breakpoint (or watchpoint or catchpoint). For
3151 each breakpoint, following columns are printed:
3154 @item Breakpoint Numbers
3156 Breakpoint, watchpoint, or catchpoint.
3158 Whether the breakpoint is marked to be disabled or deleted when hit.
3159 @item Enabled or Disabled
3160 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3161 that are not enabled.
3163 Where the breakpoint is in your program, as a memory address. For a
3164 pending breakpoint whose address is not yet known, this field will
3165 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3166 library that has the symbol or line referred by breakpoint is loaded.
3167 See below for details. A breakpoint with several locations will
3168 have @samp{<MULTIPLE>} in this field---see below for details.
3170 Where the breakpoint is in the source for your program, as a file and
3171 line number. For a pending breakpoint, the original string passed to
3172 the breakpoint command will be listed as it cannot be resolved until
3173 the appropriate shared library is loaded in the future.
3177 If a breakpoint is conditional, @code{info break} shows the condition on
3178 the line following the affected breakpoint; breakpoint commands, if any,
3179 are listed after that. A pending breakpoint is allowed to have a condition
3180 specified for it. The condition is not parsed for validity until a shared
3181 library is loaded that allows the pending breakpoint to resolve to a
3185 @code{info break} with a breakpoint
3186 number @var{n} as argument lists only that breakpoint. The
3187 convenience variable @code{$_} and the default examining-address for
3188 the @code{x} command are set to the address of the last breakpoint
3189 listed (@pxref{Memory, ,Examining Memory}).
3192 @code{info break} displays a count of the number of times the breakpoint
3193 has been hit. This is especially useful in conjunction with the
3194 @code{ignore} command. You can ignore a large number of breakpoint
3195 hits, look at the breakpoint info to see how many times the breakpoint
3196 was hit, and then run again, ignoring one less than that number. This
3197 will get you quickly to the last hit of that breakpoint.
3200 @value{GDBN} allows you to set any number of breakpoints at the same place in
3201 your program. There is nothing silly or meaningless about this. When
3202 the breakpoints are conditional, this is even useful
3203 (@pxref{Conditions, ,Break Conditions}).
3205 @cindex multiple locations, breakpoints
3206 @cindex breakpoints, multiple locations
3207 It is possible that a breakpoint corresponds to several locations
3208 in your program. Examples of this situation are:
3212 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3213 instances of the function body, used in different cases.
3216 For a C@t{++} template function, a given line in the function can
3217 correspond to any number of instantiations.
3220 For an inlined function, a given source line can correspond to
3221 several places where that function is inlined.
3224 In all those cases, @value{GDBN} will insert a breakpoint at all
3225 the relevant locations@footnote{
3226 As of this writing, multiple-location breakpoints work only if there's
3227 line number information for all the locations. This means that they
3228 will generally not work in system libraries, unless you have debug
3229 info with line numbers for them.}.
3231 A breakpoint with multiple locations is displayed in the breakpoint
3232 table using several rows---one header row, followed by one row for
3233 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3234 address column. The rows for individual locations contain the actual
3235 addresses for locations, and show the functions to which those
3236 locations belong. The number column for a location is of the form
3237 @var{breakpoint-number}.@var{location-number}.
3242 Num Type Disp Enb Address What
3243 1 breakpoint keep y <MULTIPLE>
3245 breakpoint already hit 1 time
3246 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3247 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3250 Each location can be individually enabled or disabled by passing
3251 @var{breakpoint-number}.@var{location-number} as argument to the
3252 @code{enable} and @code{disable} commands. Note that you cannot
3253 delete the individual locations from the list, you can only delete the
3254 entire list of locations that belong to their parent breakpoint (with
3255 the @kbd{delete @var{num}} command, where @var{num} is the number of
3256 the parent breakpoint, 1 in the above example). Disabling or enabling
3257 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3258 that belong to that breakpoint.
3260 @cindex pending breakpoints
3261 It's quite common to have a breakpoint inside a shared library.
3262 Shared libraries can be loaded and unloaded explicitly,
3263 and possibly repeatedly, as the program is executed. To support
3264 this use case, @value{GDBN} updates breakpoint locations whenever
3265 any shared library is loaded or unloaded. Typically, you would
3266 set a breakpoint in a shared library at the beginning of your
3267 debugging session, when the library is not loaded, and when the
3268 symbols from the library are not available. When you try to set
3269 breakpoint, @value{GDBN} will ask you if you want to set
3270 a so called @dfn{pending breakpoint}---breakpoint whose address
3271 is not yet resolved.
3273 After the program is run, whenever a new shared library is loaded,
3274 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3275 shared library contains the symbol or line referred to by some
3276 pending breakpoint, that breakpoint is resolved and becomes an
3277 ordinary breakpoint. When a library is unloaded, all breakpoints
3278 that refer to its symbols or source lines become pending again.
3280 This logic works for breakpoints with multiple locations, too. For
3281 example, if you have a breakpoint in a C@t{++} template function, and
3282 a newly loaded shared library has an instantiation of that template,
3283 a new location is added to the list of locations for the breakpoint.
3285 Except for having unresolved address, pending breakpoints do not
3286 differ from regular breakpoints. You can set conditions or commands,
3287 enable and disable them and perform other breakpoint operations.
3289 @value{GDBN} provides some additional commands for controlling what
3290 happens when the @samp{break} command cannot resolve breakpoint
3291 address specification to an address:
3293 @kindex set breakpoint pending
3294 @kindex show breakpoint pending
3296 @item set breakpoint pending auto
3297 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3298 location, it queries you whether a pending breakpoint should be created.
3300 @item set breakpoint pending on
3301 This indicates that an unrecognized breakpoint location should automatically
3302 result in a pending breakpoint being created.
3304 @item set breakpoint pending off
3305 This indicates that pending breakpoints are not to be created. Any
3306 unrecognized breakpoint location results in an error. This setting does
3307 not affect any pending breakpoints previously created.
3309 @item show breakpoint pending
3310 Show the current behavior setting for creating pending breakpoints.
3313 The settings above only affect the @code{break} command and its
3314 variants. Once breakpoint is set, it will be automatically updated
3315 as shared libraries are loaded and unloaded.
3317 @cindex automatic hardware breakpoints
3318 For some targets, @value{GDBN} can automatically decide if hardware or
3319 software breakpoints should be used, depending on whether the
3320 breakpoint address is read-only or read-write. This applies to
3321 breakpoints set with the @code{break} command as well as to internal
3322 breakpoints set by commands like @code{next} and @code{finish}. For
3323 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3326 You can control this automatic behaviour with the following commands::
3328 @kindex set breakpoint auto-hw
3329 @kindex show breakpoint auto-hw
3331 @item set breakpoint auto-hw on
3332 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3333 will try to use the target memory map to decide if software or hardware
3334 breakpoint must be used.
3336 @item set breakpoint auto-hw off
3337 This indicates @value{GDBN} should not automatically select breakpoint
3338 type. If the target provides a memory map, @value{GDBN} will warn when
3339 trying to set software breakpoint at a read-only address.
3342 @value{GDBN} normally implements breakpoints by replacing the program code
3343 at the breakpoint address with a special instruction, which, when
3344 executed, given control to the debugger. By default, the program
3345 code is so modified only when the program is resumed. As soon as
3346 the program stops, @value{GDBN} restores the original instructions. This
3347 behaviour guards against leaving breakpoints inserted in the
3348 target should gdb abrubptly disconnect. However, with slow remote
3349 targets, inserting and removing breakpoint can reduce the performance.
3350 This behavior can be controlled with the following commands::
3352 @kindex set breakpoint always-inserted
3353 @kindex show breakpoint always-inserted
3355 @item set breakpoint always-inserted off
3356 All breakpoints, including newly added by the user, are inserted in
3357 the target only when the target is resumed. All breakpoints are
3358 removed from the target when it stops.
3360 @item set breakpoint always-inserted on
3361 Causes all breakpoints to be inserted in the target at all times. If
3362 the user adds a new breakpoint, or changes an existing breakpoint, the
3363 breakpoints in the target are updated immediately. A breakpoint is
3364 removed from the target only when breakpoint itself is removed.
3366 @cindex non-stop mode, and @code{breakpoint always-inserted}
3367 @item set breakpoint always-inserted auto
3368 This is the default mode. If @value{GDBN} is controlling the inferior
3369 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3370 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3371 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3372 @code{breakpoint always-inserted} mode is off.
3375 @cindex negative breakpoint numbers
3376 @cindex internal @value{GDBN} breakpoints
3377 @value{GDBN} itself sometimes sets breakpoints in your program for
3378 special purposes, such as proper handling of @code{longjmp} (in C
3379 programs). These internal breakpoints are assigned negative numbers,
3380 starting with @code{-1}; @samp{info breakpoints} does not display them.
3381 You can see these breakpoints with the @value{GDBN} maintenance command
3382 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3385 @node Set Watchpoints
3386 @subsection Setting Watchpoints
3388 @cindex setting watchpoints
3389 You can use a watchpoint to stop execution whenever the value of an
3390 expression changes, without having to predict a particular place where
3391 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3392 The expression may be as simple as the value of a single variable, or
3393 as complex as many variables combined by operators. Examples include:
3397 A reference to the value of a single variable.
3400 An address cast to an appropriate data type. For example,
3401 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3402 address (assuming an @code{int} occupies 4 bytes).
3405 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3406 expression can use any operators valid in the program's native
3407 language (@pxref{Languages}).
3410 You can set a watchpoint on an expression even if the expression can
3411 not be evaluated yet. For instance, you can set a watchpoint on
3412 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3413 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3414 the expression produces a valid value. If the expression becomes
3415 valid in some other way than changing a variable (e.g.@: if the memory
3416 pointed to by @samp{*global_ptr} becomes readable as the result of a
3417 @code{malloc} call), @value{GDBN} may not stop until the next time
3418 the expression changes.
3420 @cindex software watchpoints
3421 @cindex hardware watchpoints
3422 Depending on your system, watchpoints may be implemented in software or
3423 hardware. @value{GDBN} does software watchpointing by single-stepping your
3424 program and testing the variable's value each time, which is hundreds of
3425 times slower than normal execution. (But this may still be worth it, to
3426 catch errors where you have no clue what part of your program is the
3429 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3430 x86-based targets, @value{GDBN} includes support for hardware
3431 watchpoints, which do not slow down the running of your program.
3435 @item watch @var{expr} @r{[}thread @var{threadnum}@r{]}
3436 Set a watchpoint for an expression. @value{GDBN} will break when the
3437 expression @var{expr} is written into by the program and its value
3438 changes. The simplest (and the most popular) use of this command is
3439 to watch the value of a single variable:
3442 (@value{GDBP}) watch foo
3445 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3446 clause, @value{GDBN} breaks only when the thread identified by
3447 @var{threadnum} changes the value of @var{expr}. If any other threads
3448 change the value of @var{expr}, @value{GDBN} will not break. Note
3449 that watchpoints restricted to a single thread in this way only work
3450 with Hardware Watchpoints.
3453 @item rwatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3454 Set a watchpoint that will break when the value of @var{expr} is read
3458 @item awatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3459 Set a watchpoint that will break when @var{expr} is either read from
3460 or written into by the program.
3462 @kindex info watchpoints @r{[}@var{n}@r{]}
3463 @item info watchpoints
3464 This command prints a list of watchpoints, breakpoints, and catchpoints;
3465 it is the same as @code{info break} (@pxref{Set Breaks}).
3468 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3469 watchpoints execute very quickly, and the debugger reports a change in
3470 value at the exact instruction where the change occurs. If @value{GDBN}
3471 cannot set a hardware watchpoint, it sets a software watchpoint, which
3472 executes more slowly and reports the change in value at the next
3473 @emph{statement}, not the instruction, after the change occurs.
3475 @cindex use only software watchpoints
3476 You can force @value{GDBN} to use only software watchpoints with the
3477 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3478 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3479 the underlying system supports them. (Note that hardware-assisted
3480 watchpoints that were set @emph{before} setting
3481 @code{can-use-hw-watchpoints} to zero will still use the hardware
3482 mechanism of watching expression values.)
3485 @item set can-use-hw-watchpoints
3486 @kindex set can-use-hw-watchpoints
3487 Set whether or not to use hardware watchpoints.
3489 @item show can-use-hw-watchpoints
3490 @kindex show can-use-hw-watchpoints
3491 Show the current mode of using hardware watchpoints.
3494 For remote targets, you can restrict the number of hardware
3495 watchpoints @value{GDBN} will use, see @ref{set remote
3496 hardware-breakpoint-limit}.
3498 When you issue the @code{watch} command, @value{GDBN} reports
3501 Hardware watchpoint @var{num}: @var{expr}
3505 if it was able to set a hardware watchpoint.
3507 Currently, the @code{awatch} and @code{rwatch} commands can only set
3508 hardware watchpoints, because accesses to data that don't change the
3509 value of the watched expression cannot be detected without examining
3510 every instruction as it is being executed, and @value{GDBN} does not do
3511 that currently. If @value{GDBN} finds that it is unable to set a
3512 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3513 will print a message like this:
3516 Expression cannot be implemented with read/access watchpoint.
3519 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3520 data type of the watched expression is wider than what a hardware
3521 watchpoint on the target machine can handle. For example, some systems
3522 can only watch regions that are up to 4 bytes wide; on such systems you
3523 cannot set hardware watchpoints for an expression that yields a
3524 double-precision floating-point number (which is typically 8 bytes
3525 wide). As a work-around, it might be possible to break the large region
3526 into a series of smaller ones and watch them with separate watchpoints.
3528 If you set too many hardware watchpoints, @value{GDBN} might be unable
3529 to insert all of them when you resume the execution of your program.
3530 Since the precise number of active watchpoints is unknown until such
3531 time as the program is about to be resumed, @value{GDBN} might not be
3532 able to warn you about this when you set the watchpoints, and the
3533 warning will be printed only when the program is resumed:
3536 Hardware watchpoint @var{num}: Could not insert watchpoint
3540 If this happens, delete or disable some of the watchpoints.
3542 Watching complex expressions that reference many variables can also
3543 exhaust the resources available for hardware-assisted watchpoints.
3544 That's because @value{GDBN} needs to watch every variable in the
3545 expression with separately allocated resources.
3547 If you call a function interactively using @code{print} or @code{call},
3548 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3549 kind of breakpoint or the call completes.
3551 @value{GDBN} automatically deletes watchpoints that watch local
3552 (automatic) variables, or expressions that involve such variables, when
3553 they go out of scope, that is, when the execution leaves the block in
3554 which these variables were defined. In particular, when the program
3555 being debugged terminates, @emph{all} local variables go out of scope,
3556 and so only watchpoints that watch global variables remain set. If you
3557 rerun the program, you will need to set all such watchpoints again. One
3558 way of doing that would be to set a code breakpoint at the entry to the
3559 @code{main} function and when it breaks, set all the watchpoints.
3561 @cindex watchpoints and threads
3562 @cindex threads and watchpoints
3563 In multi-threaded programs, watchpoints will detect changes to the
3564 watched expression from every thread.
3567 @emph{Warning:} In multi-threaded programs, software watchpoints
3568 have only limited usefulness. If @value{GDBN} creates a software
3569 watchpoint, it can only watch the value of an expression @emph{in a
3570 single thread}. If you are confident that the expression can only
3571 change due to the current thread's activity (and if you are also
3572 confident that no other thread can become current), then you can use
3573 software watchpoints as usual. However, @value{GDBN} may not notice
3574 when a non-current thread's activity changes the expression. (Hardware
3575 watchpoints, in contrast, watch an expression in all threads.)
3578 @xref{set remote hardware-watchpoint-limit}.
3580 @node Set Catchpoints
3581 @subsection Setting Catchpoints
3582 @cindex catchpoints, setting
3583 @cindex exception handlers
3584 @cindex event handling
3586 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3587 kinds of program events, such as C@t{++} exceptions or the loading of a
3588 shared library. Use the @code{catch} command to set a catchpoint.
3592 @item catch @var{event}
3593 Stop when @var{event} occurs. @var{event} can be any of the following:
3596 @cindex stop on C@t{++} exceptions
3597 The throwing of a C@t{++} exception.
3600 The catching of a C@t{++} exception.
3603 @cindex Ada exception catching
3604 @cindex catch Ada exceptions
3605 An Ada exception being raised. If an exception name is specified
3606 at the end of the command (eg @code{catch exception Program_Error}),
3607 the debugger will stop only when this specific exception is raised.
3608 Otherwise, the debugger stops execution when any Ada exception is raised.
3610 When inserting an exception catchpoint on a user-defined exception whose
3611 name is identical to one of the exceptions defined by the language, the
3612 fully qualified name must be used as the exception name. Otherwise,
3613 @value{GDBN} will assume that it should stop on the pre-defined exception
3614 rather than the user-defined one. For instance, assuming an exception
3615 called @code{Constraint_Error} is defined in package @code{Pck}, then
3616 the command to use to catch such exceptions is @kbd{catch exception
3617 Pck.Constraint_Error}.
3619 @item exception unhandled
3620 An exception that was raised but is not handled by the program.
3623 A failed Ada assertion.
3626 @cindex break on fork/exec
3627 A call to @code{exec}. This is currently only available for HP-UX
3631 A call to @code{fork}. This is currently only available for HP-UX
3635 A call to @code{vfork}. This is currently only available for HP-UX
3640 @item tcatch @var{event}
3641 Set a catchpoint that is enabled only for one stop. The catchpoint is
3642 automatically deleted after the first time the event is caught.
3646 Use the @code{info break} command to list the current catchpoints.
3648 There are currently some limitations to C@t{++} exception handling
3649 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
3653 If you call a function interactively, @value{GDBN} normally returns
3654 control to you when the function has finished executing. If the call
3655 raises an exception, however, the call may bypass the mechanism that
3656 returns control to you and cause your program either to abort or to
3657 simply continue running until it hits a breakpoint, catches a signal
3658 that @value{GDBN} is listening for, or exits. This is the case even if
3659 you set a catchpoint for the exception; catchpoints on exceptions are
3660 disabled within interactive calls.
3663 You cannot raise an exception interactively.
3666 You cannot install an exception handler interactively.
3669 @cindex raise exceptions
3670 Sometimes @code{catch} is not the best way to debug exception handling:
3671 if you need to know exactly where an exception is raised, it is better to
3672 stop @emph{before} the exception handler is called, since that way you
3673 can see the stack before any unwinding takes place. If you set a
3674 breakpoint in an exception handler instead, it may not be easy to find
3675 out where the exception was raised.
3677 To stop just before an exception handler is called, you need some
3678 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
3679 raised by calling a library function named @code{__raise_exception}
3680 which has the following ANSI C interface:
3683 /* @var{addr} is where the exception identifier is stored.
3684 @var{id} is the exception identifier. */
3685 void __raise_exception (void **addr, void *id);
3689 To make the debugger catch all exceptions before any stack
3690 unwinding takes place, set a breakpoint on @code{__raise_exception}
3691 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
3693 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
3694 that depends on the value of @var{id}, you can stop your program when
3695 a specific exception is raised. You can use multiple conditional
3696 breakpoints to stop your program when any of a number of exceptions are
3701 @subsection Deleting Breakpoints
3703 @cindex clearing breakpoints, watchpoints, catchpoints
3704 @cindex deleting breakpoints, watchpoints, catchpoints
3705 It is often necessary to eliminate a breakpoint, watchpoint, or
3706 catchpoint once it has done its job and you no longer want your program
3707 to stop there. This is called @dfn{deleting} the breakpoint. A
3708 breakpoint that has been deleted no longer exists; it is forgotten.
3710 With the @code{clear} command you can delete breakpoints according to
3711 where they are in your program. With the @code{delete} command you can
3712 delete individual breakpoints, watchpoints, or catchpoints by specifying
3713 their breakpoint numbers.
3715 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
3716 automatically ignores breakpoints on the first instruction to be executed
3717 when you continue execution without changing the execution address.
3722 Delete any breakpoints at the next instruction to be executed in the
3723 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
3724 the innermost frame is selected, this is a good way to delete a
3725 breakpoint where your program just stopped.
3727 @item clear @var{location}
3728 Delete any breakpoints set at the specified @var{location}.
3729 @xref{Specify Location}, for the various forms of @var{location}; the
3730 most useful ones are listed below:
3733 @item clear @var{function}
3734 @itemx clear @var{filename}:@var{function}
3735 Delete any breakpoints set at entry to the named @var{function}.
3737 @item clear @var{linenum}
3738 @itemx clear @var{filename}:@var{linenum}
3739 Delete any breakpoints set at or within the code of the specified
3740 @var{linenum} of the specified @var{filename}.
3743 @cindex delete breakpoints
3745 @kindex d @r{(@code{delete})}
3746 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3747 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
3748 ranges specified as arguments. If no argument is specified, delete all
3749 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
3750 confirm off}). You can abbreviate this command as @code{d}.
3754 @subsection Disabling Breakpoints
3756 @cindex enable/disable a breakpoint
3757 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
3758 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
3759 it had been deleted, but remembers the information on the breakpoint so
3760 that you can @dfn{enable} it again later.
3762 You disable and enable breakpoints, watchpoints, and catchpoints with
3763 the @code{enable} and @code{disable} commands, optionally specifying one
3764 or more breakpoint numbers as arguments. Use @code{info break} or
3765 @code{info watch} to print a list of breakpoints, watchpoints, and
3766 catchpoints if you do not know which numbers to use.
3768 Disabling and enabling a breakpoint that has multiple locations
3769 affects all of its locations.
3771 A breakpoint, watchpoint, or catchpoint can have any of four different
3772 states of enablement:
3776 Enabled. The breakpoint stops your program. A breakpoint set
3777 with the @code{break} command starts out in this state.
3779 Disabled. The breakpoint has no effect on your program.
3781 Enabled once. The breakpoint stops your program, but then becomes
3784 Enabled for deletion. The breakpoint stops your program, but
3785 immediately after it does so it is deleted permanently. A breakpoint
3786 set with the @code{tbreak} command starts out in this state.
3789 You can use the following commands to enable or disable breakpoints,
3790 watchpoints, and catchpoints:
3794 @kindex dis @r{(@code{disable})}
3795 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3796 Disable the specified breakpoints---or all breakpoints, if none are
3797 listed. A disabled breakpoint has no effect but is not forgotten. All
3798 options such as ignore-counts, conditions and commands are remembered in
3799 case the breakpoint is enabled again later. You may abbreviate
3800 @code{disable} as @code{dis}.
3803 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3804 Enable the specified breakpoints (or all defined breakpoints). They
3805 become effective once again in stopping your program.
3807 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
3808 Enable the specified breakpoints temporarily. @value{GDBN} disables any
3809 of these breakpoints immediately after stopping your program.
3811 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
3812 Enable the specified breakpoints to work once, then die. @value{GDBN}
3813 deletes any of these breakpoints as soon as your program stops there.
3814 Breakpoints set by the @code{tbreak} command start out in this state.
3817 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
3818 @c confusing: tbreak is also initially enabled.
3819 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
3820 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
3821 subsequently, they become disabled or enabled only when you use one of
3822 the commands above. (The command @code{until} can set and delete a
3823 breakpoint of its own, but it does not change the state of your other
3824 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
3828 @subsection Break Conditions
3829 @cindex conditional breakpoints
3830 @cindex breakpoint conditions
3832 @c FIXME what is scope of break condition expr? Context where wanted?
3833 @c in particular for a watchpoint?
3834 The simplest sort of breakpoint breaks every time your program reaches a
3835 specified place. You can also specify a @dfn{condition} for a
3836 breakpoint. A condition is just a Boolean expression in your
3837 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
3838 a condition evaluates the expression each time your program reaches it,
3839 and your program stops only if the condition is @emph{true}.
3841 This is the converse of using assertions for program validation; in that
3842 situation, you want to stop when the assertion is violated---that is,
3843 when the condition is false. In C, if you want to test an assertion expressed
3844 by the condition @var{assert}, you should set the condition
3845 @samp{! @var{assert}} on the appropriate breakpoint.
3847 Conditions are also accepted for watchpoints; you may not need them,
3848 since a watchpoint is inspecting the value of an expression anyhow---but
3849 it might be simpler, say, to just set a watchpoint on a variable name,
3850 and specify a condition that tests whether the new value is an interesting
3853 Break conditions can have side effects, and may even call functions in
3854 your program. This can be useful, for example, to activate functions
3855 that log program progress, or to use your own print functions to
3856 format special data structures. The effects are completely predictable
3857 unless there is another enabled breakpoint at the same address. (In
3858 that case, @value{GDBN} might see the other breakpoint first and stop your
3859 program without checking the condition of this one.) Note that
3860 breakpoint commands are usually more convenient and flexible than break
3862 purpose of performing side effects when a breakpoint is reached
3863 (@pxref{Break Commands, ,Breakpoint Command Lists}).
3865 Break conditions can be specified when a breakpoint is set, by using
3866 @samp{if} in the arguments to the @code{break} command. @xref{Set
3867 Breaks, ,Setting Breakpoints}. They can also be changed at any time
3868 with the @code{condition} command.
3870 You can also use the @code{if} keyword with the @code{watch} command.
3871 The @code{catch} command does not recognize the @code{if} keyword;
3872 @code{condition} is the only way to impose a further condition on a
3877 @item condition @var{bnum} @var{expression}
3878 Specify @var{expression} as the break condition for breakpoint,
3879 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
3880 breakpoint @var{bnum} stops your program only if the value of
3881 @var{expression} is true (nonzero, in C). When you use
3882 @code{condition}, @value{GDBN} checks @var{expression} immediately for
3883 syntactic correctness, and to determine whether symbols in it have
3884 referents in the context of your breakpoint. If @var{expression} uses
3885 symbols not referenced in the context of the breakpoint, @value{GDBN}
3886 prints an error message:
3889 No symbol "foo" in current context.
3894 not actually evaluate @var{expression} at the time the @code{condition}
3895 command (or a command that sets a breakpoint with a condition, like
3896 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
3898 @item condition @var{bnum}
3899 Remove the condition from breakpoint number @var{bnum}. It becomes
3900 an ordinary unconditional breakpoint.
3903 @cindex ignore count (of breakpoint)
3904 A special case of a breakpoint condition is to stop only when the
3905 breakpoint has been reached a certain number of times. This is so
3906 useful that there is a special way to do it, using the @dfn{ignore
3907 count} of the breakpoint. Every breakpoint has an ignore count, which
3908 is an integer. Most of the time, the ignore count is zero, and
3909 therefore has no effect. But if your program reaches a breakpoint whose
3910 ignore count is positive, then instead of stopping, it just decrements
3911 the ignore count by one and continues. As a result, if the ignore count
3912 value is @var{n}, the breakpoint does not stop the next @var{n} times
3913 your program reaches it.
3917 @item ignore @var{bnum} @var{count}
3918 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
3919 The next @var{count} times the breakpoint is reached, your program's
3920 execution does not stop; other than to decrement the ignore count, @value{GDBN}
3923 To make the breakpoint stop the next time it is reached, specify
3926 When you use @code{continue} to resume execution of your program from a
3927 breakpoint, you can specify an ignore count directly as an argument to
3928 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
3929 Stepping,,Continuing and Stepping}.
3931 If a breakpoint has a positive ignore count and a condition, the
3932 condition is not checked. Once the ignore count reaches zero,
3933 @value{GDBN} resumes checking the condition.
3935 You could achieve the effect of the ignore count with a condition such
3936 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
3937 is decremented each time. @xref{Convenience Vars, ,Convenience
3941 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
3944 @node Break Commands
3945 @subsection Breakpoint Command Lists
3947 @cindex breakpoint commands
3948 You can give any breakpoint (or watchpoint or catchpoint) a series of
3949 commands to execute when your program stops due to that breakpoint. For
3950 example, you might want to print the values of certain expressions, or
3951 enable other breakpoints.
3955 @kindex end@r{ (breakpoint commands)}
3956 @item commands @r{[}@var{bnum}@r{]}
3957 @itemx @dots{} @var{command-list} @dots{}
3959 Specify a list of commands for breakpoint number @var{bnum}. The commands
3960 themselves appear on the following lines. Type a line containing just
3961 @code{end} to terminate the commands.
3963 To remove all commands from a breakpoint, type @code{commands} and
3964 follow it immediately with @code{end}; that is, give no commands.
3966 With no @var{bnum} argument, @code{commands} refers to the last
3967 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
3968 recently encountered).
3971 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
3972 disabled within a @var{command-list}.
3974 You can use breakpoint commands to start your program up again. Simply
3975 use the @code{continue} command, or @code{step}, or any other command
3976 that resumes execution.
3978 Any other commands in the command list, after a command that resumes
3979 execution, are ignored. This is because any time you resume execution
3980 (even with a simple @code{next} or @code{step}), you may encounter
3981 another breakpoint---which could have its own command list, leading to
3982 ambiguities about which list to execute.
3985 If the first command you specify in a command list is @code{silent}, the
3986 usual message about stopping at a breakpoint is not printed. This may
3987 be desirable for breakpoints that are to print a specific message and
3988 then continue. If none of the remaining commands print anything, you
3989 see no sign that the breakpoint was reached. @code{silent} is
3990 meaningful only at the beginning of a breakpoint command list.
3992 The commands @code{echo}, @code{output}, and @code{printf} allow you to
3993 print precisely controlled output, and are often useful in silent
3994 breakpoints. @xref{Output, ,Commands for Controlled Output}.
3996 For example, here is how you could use breakpoint commands to print the
3997 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4003 printf "x is %d\n",x
4008 One application for breakpoint commands is to compensate for one bug so
4009 you can test for another. Put a breakpoint just after the erroneous line
4010 of code, give it a condition to detect the case in which something
4011 erroneous has been done, and give it commands to assign correct values
4012 to any variables that need them. End with the @code{continue} command
4013 so that your program does not stop, and start with the @code{silent}
4014 command so that no output is produced. Here is an example:
4025 @c @ifclear BARETARGET
4026 @node Error in Breakpoints
4027 @subsection ``Cannot insert breakpoints''
4029 If you request too many active hardware-assisted breakpoints and
4030 watchpoints, you will see this error message:
4032 @c FIXME: the precise wording of this message may change; the relevant
4033 @c source change is not committed yet (Sep 3, 1999).
4035 Stopped; cannot insert breakpoints.
4036 You may have requested too many hardware breakpoints and watchpoints.
4040 This message is printed when you attempt to resume the program, since
4041 only then @value{GDBN} knows exactly how many hardware breakpoints and
4042 watchpoints it needs to insert.
4044 When this message is printed, you need to disable or remove some of the
4045 hardware-assisted breakpoints and watchpoints, and then continue.
4047 @node Breakpoint-related Warnings
4048 @subsection ``Breakpoint address adjusted...''
4049 @cindex breakpoint address adjusted
4051 Some processor architectures place constraints on the addresses at
4052 which breakpoints may be placed. For architectures thus constrained,
4053 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4054 with the constraints dictated by the architecture.
4056 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4057 a VLIW architecture in which a number of RISC-like instructions may be
4058 bundled together for parallel execution. The FR-V architecture
4059 constrains the location of a breakpoint instruction within such a
4060 bundle to the instruction with the lowest address. @value{GDBN}
4061 honors this constraint by adjusting a breakpoint's address to the
4062 first in the bundle.
4064 It is not uncommon for optimized code to have bundles which contain
4065 instructions from different source statements, thus it may happen that
4066 a breakpoint's address will be adjusted from one source statement to
4067 another. Since this adjustment may significantly alter @value{GDBN}'s
4068 breakpoint related behavior from what the user expects, a warning is
4069 printed when the breakpoint is first set and also when the breakpoint
4072 A warning like the one below is printed when setting a breakpoint
4073 that's been subject to address adjustment:
4076 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4079 Such warnings are printed both for user settable and @value{GDBN}'s
4080 internal breakpoints. If you see one of these warnings, you should
4081 verify that a breakpoint set at the adjusted address will have the
4082 desired affect. If not, the breakpoint in question may be removed and
4083 other breakpoints may be set which will have the desired behavior.
4084 E.g., it may be sufficient to place the breakpoint at a later
4085 instruction. A conditional breakpoint may also be useful in some
4086 cases to prevent the breakpoint from triggering too often.
4088 @value{GDBN} will also issue a warning when stopping at one of these
4089 adjusted breakpoints:
4092 warning: Breakpoint 1 address previously adjusted from 0x00010414
4096 When this warning is encountered, it may be too late to take remedial
4097 action except in cases where the breakpoint is hit earlier or more
4098 frequently than expected.
4100 @node Continuing and Stepping
4101 @section Continuing and Stepping
4105 @cindex resuming execution
4106 @dfn{Continuing} means resuming program execution until your program
4107 completes normally. In contrast, @dfn{stepping} means executing just
4108 one more ``step'' of your program, where ``step'' may mean either one
4109 line of source code, or one machine instruction (depending on what
4110 particular command you use). Either when continuing or when stepping,
4111 your program may stop even sooner, due to a breakpoint or a signal. (If
4112 it stops due to a signal, you may want to use @code{handle}, or use
4113 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4117 @kindex c @r{(@code{continue})}
4118 @kindex fg @r{(resume foreground execution)}
4119 @item continue @r{[}@var{ignore-count}@r{]}
4120 @itemx c @r{[}@var{ignore-count}@r{]}
4121 @itemx fg @r{[}@var{ignore-count}@r{]}
4122 Resume program execution, at the address where your program last stopped;
4123 any breakpoints set at that address are bypassed. The optional argument
4124 @var{ignore-count} allows you to specify a further number of times to
4125 ignore a breakpoint at this location; its effect is like that of
4126 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4128 The argument @var{ignore-count} is meaningful only when your program
4129 stopped due to a breakpoint. At other times, the argument to
4130 @code{continue} is ignored.
4132 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4133 debugged program is deemed to be the foreground program) are provided
4134 purely for convenience, and have exactly the same behavior as
4138 To resume execution at a different place, you can use @code{return}
4139 (@pxref{Returning, ,Returning from a Function}) to go back to the
4140 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4141 Different Address}) to go to an arbitrary location in your program.
4143 A typical technique for using stepping is to set a breakpoint
4144 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4145 beginning of the function or the section of your program where a problem
4146 is believed to lie, run your program until it stops at that breakpoint,
4147 and then step through the suspect area, examining the variables that are
4148 interesting, until you see the problem happen.
4152 @kindex s @r{(@code{step})}
4154 Continue running your program until control reaches a different source
4155 line, then stop it and return control to @value{GDBN}. This command is
4156 abbreviated @code{s}.
4159 @c "without debugging information" is imprecise; actually "without line
4160 @c numbers in the debugging information". (gcc -g1 has debugging info but
4161 @c not line numbers). But it seems complex to try to make that
4162 @c distinction here.
4163 @emph{Warning:} If you use the @code{step} command while control is
4164 within a function that was compiled without debugging information,
4165 execution proceeds until control reaches a function that does have
4166 debugging information. Likewise, it will not step into a function which
4167 is compiled without debugging information. To step through functions
4168 without debugging information, use the @code{stepi} command, described
4172 The @code{step} command only stops at the first instruction of a source
4173 line. This prevents the multiple stops that could otherwise occur in
4174 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4175 to stop if a function that has debugging information is called within
4176 the line. In other words, @code{step} @emph{steps inside} any functions
4177 called within the line.
4179 Also, the @code{step} command only enters a function if there is line
4180 number information for the function. Otherwise it acts like the
4181 @code{next} command. This avoids problems when using @code{cc -gl}
4182 on MIPS machines. Previously, @code{step} entered subroutines if there
4183 was any debugging information about the routine.
4185 @item step @var{count}
4186 Continue running as in @code{step}, but do so @var{count} times. If a
4187 breakpoint is reached, or a signal not related to stepping occurs before
4188 @var{count} steps, stepping stops right away.
4191 @kindex n @r{(@code{next})}
4192 @item next @r{[}@var{count}@r{]}
4193 Continue to the next source line in the current (innermost) stack frame.
4194 This is similar to @code{step}, but function calls that appear within
4195 the line of code are executed without stopping. Execution stops when
4196 control reaches a different line of code at the original stack level
4197 that was executing when you gave the @code{next} command. This command
4198 is abbreviated @code{n}.
4200 An argument @var{count} is a repeat count, as for @code{step}.
4203 @c FIX ME!! Do we delete this, or is there a way it fits in with
4204 @c the following paragraph? --- Vctoria
4206 @c @code{next} within a function that lacks debugging information acts like
4207 @c @code{step}, but any function calls appearing within the code of the
4208 @c function are executed without stopping.
4210 The @code{next} command only stops at the first instruction of a
4211 source line. This prevents multiple stops that could otherwise occur in
4212 @code{switch} statements, @code{for} loops, etc.
4214 @kindex set step-mode
4216 @cindex functions without line info, and stepping
4217 @cindex stepping into functions with no line info
4218 @itemx set step-mode on
4219 The @code{set step-mode on} command causes the @code{step} command to
4220 stop at the first instruction of a function which contains no debug line
4221 information rather than stepping over it.
4223 This is useful in cases where you may be interested in inspecting the
4224 machine instructions of a function which has no symbolic info and do not
4225 want @value{GDBN} to automatically skip over this function.
4227 @item set step-mode off
4228 Causes the @code{step} command to step over any functions which contains no
4229 debug information. This is the default.
4231 @item show step-mode
4232 Show whether @value{GDBN} will stop in or step over functions without
4233 source line debug information.
4236 @kindex fin @r{(@code{finish})}
4238 Continue running until just after function in the selected stack frame
4239 returns. Print the returned value (if any). This command can be
4240 abbreviated as @code{fin}.
4242 Contrast this with the @code{return} command (@pxref{Returning,
4243 ,Returning from a Function}).
4246 @kindex u @r{(@code{until})}
4247 @cindex run until specified location
4250 Continue running until a source line past the current line, in the
4251 current stack frame, is reached. This command is used to avoid single
4252 stepping through a loop more than once. It is like the @code{next}
4253 command, except that when @code{until} encounters a jump, it
4254 automatically continues execution until the program counter is greater
4255 than the address of the jump.
4257 This means that when you reach the end of a loop after single stepping
4258 though it, @code{until} makes your program continue execution until it
4259 exits the loop. In contrast, a @code{next} command at the end of a loop
4260 simply steps back to the beginning of the loop, which forces you to step
4261 through the next iteration.
4263 @code{until} always stops your program if it attempts to exit the current
4266 @code{until} may produce somewhat counterintuitive results if the order
4267 of machine code does not match the order of the source lines. For
4268 example, in the following excerpt from a debugging session, the @code{f}
4269 (@code{frame}) command shows that execution is stopped at line
4270 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4274 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4276 (@value{GDBP}) until
4277 195 for ( ; argc > 0; NEXTARG) @{
4280 This happened because, for execution efficiency, the compiler had
4281 generated code for the loop closure test at the end, rather than the
4282 start, of the loop---even though the test in a C @code{for}-loop is
4283 written before the body of the loop. The @code{until} command appeared
4284 to step back to the beginning of the loop when it advanced to this
4285 expression; however, it has not really gone to an earlier
4286 statement---not in terms of the actual machine code.
4288 @code{until} with no argument works by means of single
4289 instruction stepping, and hence is slower than @code{until} with an
4292 @item until @var{location}
4293 @itemx u @var{location}
4294 Continue running your program until either the specified location is
4295 reached, or the current stack frame returns. @var{location} is any of
4296 the forms described in @ref{Specify Location}.
4297 This form of the command uses temporary breakpoints, and
4298 hence is quicker than @code{until} without an argument. The specified
4299 location is actually reached only if it is in the current frame. This
4300 implies that @code{until} can be used to skip over recursive function
4301 invocations. For instance in the code below, if the current location is
4302 line @code{96}, issuing @code{until 99} will execute the program up to
4303 line @code{99} in the same invocation of factorial, i.e., after the inner
4304 invocations have returned.
4307 94 int factorial (int value)
4309 96 if (value > 1) @{
4310 97 value *= factorial (value - 1);
4317 @kindex advance @var{location}
4318 @itemx advance @var{location}
4319 Continue running the program up to the given @var{location}. An argument is
4320 required, which should be of one of the forms described in
4321 @ref{Specify Location}.
4322 Execution will also stop upon exit from the current stack
4323 frame. This command is similar to @code{until}, but @code{advance} will
4324 not skip over recursive function calls, and the target location doesn't
4325 have to be in the same frame as the current one.
4329 @kindex si @r{(@code{stepi})}
4331 @itemx stepi @var{arg}
4333 Execute one machine instruction, then stop and return to the debugger.
4335 It is often useful to do @samp{display/i $pc} when stepping by machine
4336 instructions. This makes @value{GDBN} automatically display the next
4337 instruction to be executed, each time your program stops. @xref{Auto
4338 Display,, Automatic Display}.
4340 An argument is a repeat count, as in @code{step}.
4344 @kindex ni @r{(@code{nexti})}
4346 @itemx nexti @var{arg}
4348 Execute one machine instruction, but if it is a function call,
4349 proceed until the function returns.
4351 An argument is a repeat count, as in @code{next}.
4358 A signal is an asynchronous event that can happen in a program. The
4359 operating system defines the possible kinds of signals, and gives each
4360 kind a name and a number. For example, in Unix @code{SIGINT} is the
4361 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4362 @code{SIGSEGV} is the signal a program gets from referencing a place in
4363 memory far away from all the areas in use; @code{SIGALRM} occurs when
4364 the alarm clock timer goes off (which happens only if your program has
4365 requested an alarm).
4367 @cindex fatal signals
4368 Some signals, including @code{SIGALRM}, are a normal part of the
4369 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4370 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4371 program has not specified in advance some other way to handle the signal.
4372 @code{SIGINT} does not indicate an error in your program, but it is normally
4373 fatal so it can carry out the purpose of the interrupt: to kill the program.
4375 @value{GDBN} has the ability to detect any occurrence of a signal in your
4376 program. You can tell @value{GDBN} in advance what to do for each kind of
4379 @cindex handling signals
4380 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4381 @code{SIGALRM} be silently passed to your program
4382 (so as not to interfere with their role in the program's functioning)
4383 but to stop your program immediately whenever an error signal happens.
4384 You can change these settings with the @code{handle} command.
4387 @kindex info signals
4391 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4392 handle each one. You can use this to see the signal numbers of all
4393 the defined types of signals.
4395 @item info signals @var{sig}
4396 Similar, but print information only about the specified signal number.
4398 @code{info handle} is an alias for @code{info signals}.
4401 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4402 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4403 can be the number of a signal or its name (with or without the
4404 @samp{SIG} at the beginning); a list of signal numbers of the form
4405 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4406 known signals. Optional arguments @var{keywords}, described below,
4407 say what change to make.
4411 The keywords allowed by the @code{handle} command can be abbreviated.
4412 Their full names are:
4416 @value{GDBN} should not stop your program when this signal happens. It may
4417 still print a message telling you that the signal has come in.
4420 @value{GDBN} should stop your program when this signal happens. This implies
4421 the @code{print} keyword as well.
4424 @value{GDBN} should print a message when this signal happens.
4427 @value{GDBN} should not mention the occurrence of the signal at all. This
4428 implies the @code{nostop} keyword as well.
4432 @value{GDBN} should allow your program to see this signal; your program
4433 can handle the signal, or else it may terminate if the signal is fatal
4434 and not handled. @code{pass} and @code{noignore} are synonyms.
4438 @value{GDBN} should not allow your program to see this signal.
4439 @code{nopass} and @code{ignore} are synonyms.
4443 When a signal stops your program, the signal is not visible to the
4445 continue. Your program sees the signal then, if @code{pass} is in
4446 effect for the signal in question @emph{at that time}. In other words,
4447 after @value{GDBN} reports a signal, you can use the @code{handle}
4448 command with @code{pass} or @code{nopass} to control whether your
4449 program sees that signal when you continue.
4451 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4452 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4453 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4456 You can also use the @code{signal} command to prevent your program from
4457 seeing a signal, or cause it to see a signal it normally would not see,
4458 or to give it any signal at any time. For example, if your program stopped
4459 due to some sort of memory reference error, you might store correct
4460 values into the erroneous variables and continue, hoping to see more
4461 execution; but your program would probably terminate immediately as
4462 a result of the fatal signal once it saw the signal. To prevent this,
4463 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4467 @section Stopping and Starting Multi-thread Programs
4469 @cindex stopped threads
4470 @cindex threads, stopped
4472 @cindex continuing threads
4473 @cindex threads, continuing
4475 @value{GDBN} supports debugging programs with multiple threads
4476 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
4477 are two modes of controlling execution of your program within the
4478 debugger. In the default mode, referred to as @dfn{all-stop mode},
4479 when any thread in your program stops (for example, at a breakpoint
4480 or while being stepped), all other threads in the program are also stopped by
4481 @value{GDBN}. On some targets, @value{GDBN} also supports
4482 @dfn{non-stop mode}, in which other threads can continue to run freely while
4483 you examine the stopped thread in the debugger.
4486 * All-Stop Mode:: All threads stop when GDB takes control
4487 * Non-Stop Mode:: Other threads continue to execute
4488 * Background Execution:: Running your program asynchronously
4489 * Thread-Specific Breakpoints:: Controlling breakpoints
4490 * Interrupted System Calls:: GDB may interfere with system calls
4494 @subsection All-Stop Mode
4496 @cindex all-stop mode
4498 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
4499 @emph{all} threads of execution stop, not just the current thread. This
4500 allows you to examine the overall state of the program, including
4501 switching between threads, without worrying that things may change
4504 Conversely, whenever you restart the program, @emph{all} threads start
4505 executing. @emph{This is true even when single-stepping} with commands
4506 like @code{step} or @code{next}.
4508 In particular, @value{GDBN} cannot single-step all threads in lockstep.
4509 Since thread scheduling is up to your debugging target's operating
4510 system (not controlled by @value{GDBN}), other threads may
4511 execute more than one statement while the current thread completes a
4512 single step. Moreover, in general other threads stop in the middle of a
4513 statement, rather than at a clean statement boundary, when the program
4516 You might even find your program stopped in another thread after
4517 continuing or even single-stepping. This happens whenever some other
4518 thread runs into a breakpoint, a signal, or an exception before the
4519 first thread completes whatever you requested.
4521 @cindex automatic thread selection
4522 @cindex switching threads automatically
4523 @cindex threads, automatic switching
4524 Whenever @value{GDBN} stops your program, due to a breakpoint or a
4525 signal, it automatically selects the thread where that breakpoint or
4526 signal happened. @value{GDBN} alerts you to the context switch with a
4527 message such as @samp{[Switching to Thread @var{n}]} to identify the
4530 On some OSes, you can modify @value{GDBN}'s default behavior by
4531 locking the OS scheduler to allow only a single thread to run.
4534 @item set scheduler-locking @var{mode}
4535 @cindex scheduler locking mode
4536 @cindex lock scheduler
4537 Set the scheduler locking mode. If it is @code{off}, then there is no
4538 locking and any thread may run at any time. If @code{on}, then only the
4539 current thread may run when the inferior is resumed. The @code{step}
4540 mode optimizes for single-stepping; it prevents other threads
4541 from preempting the current thread while you are stepping, so that
4542 the focus of debugging does not change unexpectedly.
4543 Other threads only rarely (or never) get a chance to run
4544 when you step. They are more likely to run when you @samp{next} over a
4545 function call, and they are completely free to run when you use commands
4546 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
4547 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
4548 the current thread away from the thread that you are debugging.
4550 @item show scheduler-locking
4551 Display the current scheduler locking mode.
4555 @subsection Non-Stop Mode
4557 @cindex non-stop mode
4559 @c This section is really only a place-holder, and needs to be expanded
4560 @c with more details.
4562 For some multi-threaded targets, @value{GDBN} supports an optional
4563 mode of operation in which you can examine stopped program threads in
4564 the debugger while other threads continue to execute freely. This
4565 minimizes intrusion when debugging live systems, such as programs
4566 where some threads have real-time constraints or must continue to
4567 respond to external events. This is referred to as @dfn{non-stop} mode.
4569 In non-stop mode, when a thread stops to report a debugging event,
4570 @emph{only} that thread is stopped; @value{GDBN} does not stop other
4571 threads as well, in contrast to the all-stop mode behavior. Additionally,
4572 execution commands such as @code{continue} and @code{step} apply by default
4573 only to the current thread in non-stop mode, rather than all threads as
4574 in all-stop mode. This allows you to control threads explicitly in
4575 ways that are not possible in all-stop mode --- for example, stepping
4576 one thread while allowing others to run freely, stepping
4577 one thread while holding all others stopped, or stepping several threads
4578 independently and simultaneously.
4580 To enter non-stop mode, use this sequence of commands before you run
4581 or attach to your program:
4584 # Enable the async interface.
4587 # If using the CLI, pagination breaks non-stop.
4590 # Finally, turn it on!
4594 You can use these commands to manipulate the non-stop mode setting:
4597 @kindex set non-stop
4598 @item set non-stop on
4599 Enable selection of non-stop mode.
4600 @item set non-stop off
4601 Disable selection of non-stop mode.
4602 @kindex show non-stop
4604 Show the current non-stop enablement setting.
4607 Note these commands only reflect whether non-stop mode is enabled,
4608 not whether the currently-executing program is being run in non-stop mode.
4609 In particular, the @code{set non-stop} preference is only consulted when
4610 @value{GDBN} starts or connects to the target program, and it is generally
4611 not possible to switch modes once debugging has started. Furthermore,
4612 since not all targets support non-stop mode, even when you have enabled
4613 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
4616 In non-stop mode, all execution commands apply only to the current thread
4617 by default. That is, @code{continue} only continues one thread.
4618 To continue all threads, issue @code{continue -a} or @code{c -a}.
4620 You can use @value{GDBN}'s background execution commands
4621 (@pxref{Background Execution}) to run some threads in the background
4622 while you continue to examine or step others from @value{GDBN}.
4623 The MI execution commands (@pxref{GDB/MI Program Execution}) are
4624 always executed asynchronously in non-stop mode.
4626 Suspending execution is done with the @code{interrupt} command when
4627 running in the background, or @kbd{Ctrl-c} during foreground execution.
4628 In all-stop mode, this stops the whole process;
4629 but in non-stop mode the interrupt applies only to the current thread.
4630 To stop the whole program, use @code{interrupt -a}.
4632 Other execution commands do not currently support the @code{-a} option.
4634 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
4635 that thread current, as it does in all-stop mode. This is because the
4636 thread stop notifications are asynchronous with respect to @value{GDBN}'s
4637 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
4638 changed to a different thread just as you entered a command to operate on the
4639 previously current thread.
4641 @node Background Execution
4642 @subsection Background Execution
4644 @cindex foreground execution
4645 @cindex background execution
4646 @cindex asynchronous execution
4647 @cindex execution, foreground, background and asynchronous
4649 @value{GDBN}'s execution commands have two variants: the normal
4650 foreground (synchronous) behavior, and a background
4651 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
4652 the program to report that some thread has stopped before prompting for
4653 another command. In background execution, @value{GDBN} immediately gives
4654 a command prompt so that you can issue other commands while your program runs.
4656 To specify background execution, add a @code{&} to the command. For example,
4657 the background form of the @code{continue} command is @code{continue&}, or
4658 just @code{c&}. The execution commands that accept background execution
4664 @xref{Starting, , Starting your Program}.
4668 @xref{Attach, , Debugging an Already-running Process}.
4672 @xref{Continuing and Stepping, step}.
4676 @xref{Continuing and Stepping, stepi}.
4680 @xref{Continuing and Stepping, next}.
4684 @xref{Continuing and Stepping, nexti}.
4688 @xref{Continuing and Stepping, continue}.
4692 @xref{Continuing and Stepping, finish}.
4696 @xref{Continuing and Stepping, until}.
4700 Background execution is especially useful in conjunction with non-stop
4701 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
4702 However, you can also use these commands in the normal all-stop mode with
4703 the restriction that you cannot issue another execution command until the
4704 previous one finishes. Examples of commands that are valid in all-stop
4705 mode while the program is running include @code{help} and @code{info break}.
4707 You can interrupt your program while it is running in the background by
4708 using the @code{interrupt} command.
4715 Suspend execution of the running program. In all-stop mode,
4716 @code{interrupt} stops the whole process, but in non-stop mode, it stops
4717 only the current thread. To stop the whole program in non-stop mode,
4718 use @code{interrupt -a}.
4721 You may need to explicitly enable async mode before you can use background
4722 execution commands, with the @code{set target-async 1} command. If the
4723 target doesn't support async mode, @value{GDBN} issues an error message
4724 if you attempt to use the background execution commands.
4726 @node Thread-Specific Breakpoints
4727 @subsection Thread-Specific Breakpoints
4729 When your program has multiple threads (@pxref{Threads,, Debugging
4730 Programs with Multiple Threads}), you can choose whether to set
4731 breakpoints on all threads, or on a particular thread.
4734 @cindex breakpoints and threads
4735 @cindex thread breakpoints
4736 @kindex break @dots{} thread @var{threadno}
4737 @item break @var{linespec} thread @var{threadno}
4738 @itemx break @var{linespec} thread @var{threadno} if @dots{}
4739 @var{linespec} specifies source lines; there are several ways of
4740 writing them (@pxref{Specify Location}), but the effect is always to
4741 specify some source line.
4743 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
4744 to specify that you only want @value{GDBN} to stop the program when a
4745 particular thread reaches this breakpoint. @var{threadno} is one of the
4746 numeric thread identifiers assigned by @value{GDBN}, shown in the first
4747 column of the @samp{info threads} display.
4749 If you do not specify @samp{thread @var{threadno}} when you set a
4750 breakpoint, the breakpoint applies to @emph{all} threads of your
4753 You can use the @code{thread} qualifier on conditional breakpoints as
4754 well; in this case, place @samp{thread @var{threadno}} before the
4755 breakpoint condition, like this:
4758 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
4763 @node Interrupted System Calls
4764 @subsection Interrupted System Calls
4766 @cindex thread breakpoints and system calls
4767 @cindex system calls and thread breakpoints
4768 @cindex premature return from system calls
4769 There is an unfortunate side effect when using @value{GDBN} to debug
4770 multi-threaded programs. If one thread stops for a
4771 breakpoint, or for some other reason, and another thread is blocked in a
4772 system call, then the system call may return prematurely. This is a
4773 consequence of the interaction between multiple threads and the signals
4774 that @value{GDBN} uses to implement breakpoints and other events that
4777 To handle this problem, your program should check the return value of
4778 each system call and react appropriately. This is good programming
4781 For example, do not write code like this:
4787 The call to @code{sleep} will return early if a different thread stops
4788 at a breakpoint or for some other reason.
4790 Instead, write this:
4795 unslept = sleep (unslept);
4798 A system call is allowed to return early, so the system is still
4799 conforming to its specification. But @value{GDBN} does cause your
4800 multi-threaded program to behave differently than it would without
4803 Also, @value{GDBN} uses internal breakpoints in the thread library to
4804 monitor certain events such as thread creation and thread destruction.
4805 When such an event happens, a system call in another thread may return
4806 prematurely, even though your program does not appear to stop.
4809 @node Reverse Execution
4810 @chapter Running programs backward
4811 @cindex reverse execution
4812 @cindex running programs backward
4814 When you are debugging a program, it is not unusual to realize that
4815 you have gone too far, and some event of interest has already happened.
4816 If the target environment supports it, @value{GDBN} can allow you to
4817 ``rewind'' the program by running it backward.
4819 A target environment that supports reverse execution should be able
4820 to ``undo'' the changes in machine state that have taken place as the
4821 program was executing normally. Variables, registers etc.@: should
4822 revert to their previous values. Obviously this requires a great
4823 deal of sophistication on the part of the target environment; not
4824 all target environments can support reverse execution.
4826 When a program is executed in reverse, the instructions that
4827 have most recently been executed are ``un-executed'', in reverse
4828 order. The program counter runs backward, following the previous
4829 thread of execution in reverse. As each instruction is ``un-executed'',
4830 the values of memory and/or registers that were changed by that
4831 instruction are reverted to their previous states. After executing
4832 a piece of source code in reverse, all side effects of that code
4833 should be ``undone'', and all variables should be returned to their
4834 prior values@footnote{
4835 Note that some side effects are easier to undo than others. For instance,
4836 memory and registers are relatively easy, but device I/O is hard. Some
4837 targets may be able undo things like device I/O, and some may not.
4839 The contract between @value{GDBN} and the reverse executing target
4840 requires only that the target do something reasonable when
4841 @value{GDBN} tells it to execute backwards, and then report the
4842 results back to @value{GDBN}. Whatever the target reports back to
4843 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
4844 assumes that the memory and registers that the target reports are in a
4845 consistant state, but @value{GDBN} accepts whatever it is given.
4848 If you are debugging in a target environment that supports
4849 reverse execution, @value{GDBN} provides the following commands.
4852 @kindex reverse-continue
4853 @kindex rc @r{(@code{reverse-continue})}
4854 @item reverse-continue @r{[}@var{ignore-count}@r{]}
4855 @itemx rc @r{[}@var{ignore-count}@r{]}
4856 Beginning at the point where your program last stopped, start executing
4857 in reverse. Reverse execution will stop for breakpoints and synchronous
4858 exceptions (signals), just like normal execution. Behavior of
4859 asynchronous signals depends on the target environment.
4861 @kindex reverse-step
4862 @kindex rs @r{(@code{step})}
4863 @item reverse-step @r{[}@var{count}@r{]}
4864 Run the program backward until control reaches the start of a
4865 different source line; then stop it, and return control to @value{GDBN}.
4867 Like the @code{step} command, @code{reverse-step} will only stop
4868 at the beginning of a source line. It ``un-executes'' the previously
4869 executed source line. If the previous source line included calls to
4870 debuggable functions, @code{reverse-step} will step (backward) into
4871 the called function, stopping at the beginning of the @emph{last}
4872 statement in the called function (typically a return statement).
4874 Also, as with the @code{step} command, if non-debuggable functions are
4875 called, @code{reverse-step} will run thru them backward without stopping.
4877 @kindex reverse-stepi
4878 @kindex rsi @r{(@code{reverse-stepi})}
4879 @item reverse-stepi @r{[}@var{count}@r{]}
4880 Reverse-execute one machine instruction. Note that the instruction
4881 to be reverse-executed is @emph{not} the one pointed to by the program
4882 counter, but the instruction executed prior to that one. For instance,
4883 if the last instruction was a jump, @code{reverse-stepi} will take you
4884 back from the destination of the jump to the jump instruction itself.
4886 @kindex reverse-next
4887 @kindex rn @r{(@code{reverse-next})}
4888 @item reverse-next @r{[}@var{count}@r{]}
4889 Run backward to the beginning of the previous line executed in
4890 the current (innermost) stack frame. If the line contains function
4891 calls, they will be ``un-executed'' without stopping. Starting from
4892 the first line of a function, @code{reverse-next} will take you back
4893 to the caller of that function, @emph{before} the function was called,
4894 just as the normal @code{next} command would take you from the last
4895 line of a function back to its return to its caller
4896 @footnote{Unles the code is too heavily optimized.}.
4898 @kindex reverse-nexti
4899 @kindex rni @r{(@code{reverse-nexti})}
4900 @item reverse-nexti @r{[}@var{count}@r{]}
4901 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
4902 in reverse, except that called functions are ``un-executed'' atomically.
4903 That is, if the previously executed instruction was a return from
4904 another instruction, @code{reverse-nexti} will continue to execute
4905 in reverse until the call to that function (from the current stack
4908 @kindex reverse-finish
4909 @item reverse-finish
4910 Just as the @code{finish} command takes you to the point where the
4911 current function returns, @code{reverse-finish} takes you to the point
4912 where it was called. Instead of ending up at the end of the current
4913 function invocation, you end up at the beginning.
4915 @kindex set exec-direction
4916 @item set exec-direction
4917 Set the direction of target execution.
4918 @itemx set exec-direction reverse
4919 @cindex execute forward or backward in time
4920 @value{GDBN} will perform all execution commands in reverse, until the
4921 exec-direction mode is changed to ``forward''. Affected commands include
4922 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
4923 command cannot be used in reverse mode.
4924 @item set exec-direction forward
4925 @value{GDBN} will perform all execution commands in the normal fashion.
4926 This is the default.
4931 @chapter Examining the Stack
4933 When your program has stopped, the first thing you need to know is where it
4934 stopped and how it got there.
4937 Each time your program performs a function call, information about the call
4939 That information includes the location of the call in your program,
4940 the arguments of the call,
4941 and the local variables of the function being called.
4942 The information is saved in a block of data called a @dfn{stack frame}.
4943 The stack frames are allocated in a region of memory called the @dfn{call
4946 When your program stops, the @value{GDBN} commands for examining the
4947 stack allow you to see all of this information.
4949 @cindex selected frame
4950 One of the stack frames is @dfn{selected} by @value{GDBN} and many
4951 @value{GDBN} commands refer implicitly to the selected frame. In
4952 particular, whenever you ask @value{GDBN} for the value of a variable in
4953 your program, the value is found in the selected frame. There are
4954 special @value{GDBN} commands to select whichever frame you are
4955 interested in. @xref{Selection, ,Selecting a Frame}.
4957 When your program stops, @value{GDBN} automatically selects the
4958 currently executing frame and describes it briefly, similar to the
4959 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
4962 * Frames:: Stack frames
4963 * Backtrace:: Backtraces
4964 * Selection:: Selecting a frame
4965 * Frame Info:: Information on a frame
4970 @section Stack Frames
4972 @cindex frame, definition
4974 The call stack is divided up into contiguous pieces called @dfn{stack
4975 frames}, or @dfn{frames} for short; each frame is the data associated
4976 with one call to one function. The frame contains the arguments given
4977 to the function, the function's local variables, and the address at
4978 which the function is executing.
4980 @cindex initial frame
4981 @cindex outermost frame
4982 @cindex innermost frame
4983 When your program is started, the stack has only one frame, that of the
4984 function @code{main}. This is called the @dfn{initial} frame or the
4985 @dfn{outermost} frame. Each time a function is called, a new frame is
4986 made. Each time a function returns, the frame for that function invocation
4987 is eliminated. If a function is recursive, there can be many frames for
4988 the same function. The frame for the function in which execution is
4989 actually occurring is called the @dfn{innermost} frame. This is the most
4990 recently created of all the stack frames that still exist.
4992 @cindex frame pointer
4993 Inside your program, stack frames are identified by their addresses. A
4994 stack frame consists of many bytes, each of which has its own address; each
4995 kind of computer has a convention for choosing one byte whose
4996 address serves as the address of the frame. Usually this address is kept
4997 in a register called the @dfn{frame pointer register}
4998 (@pxref{Registers, $fp}) while execution is going on in that frame.
5000 @cindex frame number
5001 @value{GDBN} assigns numbers to all existing stack frames, starting with
5002 zero for the innermost frame, one for the frame that called it,
5003 and so on upward. These numbers do not really exist in your program;
5004 they are assigned by @value{GDBN} to give you a way of designating stack
5005 frames in @value{GDBN} commands.
5007 @c The -fomit-frame-pointer below perennially causes hbox overflow
5008 @c underflow problems.
5009 @cindex frameless execution
5010 Some compilers provide a way to compile functions so that they operate
5011 without stack frames. (For example, the @value{NGCC} option
5013 @samp{-fomit-frame-pointer}
5015 generates functions without a frame.)
5016 This is occasionally done with heavily used library functions to save
5017 the frame setup time. @value{GDBN} has limited facilities for dealing
5018 with these function invocations. If the innermost function invocation
5019 has no stack frame, @value{GDBN} nevertheless regards it as though
5020 it had a separate frame, which is numbered zero as usual, allowing
5021 correct tracing of the function call chain. However, @value{GDBN} has
5022 no provision for frameless functions elsewhere in the stack.
5025 @kindex frame@r{, command}
5026 @cindex current stack frame
5027 @item frame @var{args}
5028 The @code{frame} command allows you to move from one stack frame to another,
5029 and to print the stack frame you select. @var{args} may be either the
5030 address of the frame or the stack frame number. Without an argument,
5031 @code{frame} prints the current stack frame.
5033 @kindex select-frame
5034 @cindex selecting frame silently
5036 The @code{select-frame} command allows you to move from one stack frame
5037 to another without printing the frame. This is the silent version of
5045 @cindex call stack traces
5046 A backtrace is a summary of how your program got where it is. It shows one
5047 line per frame, for many frames, starting with the currently executing
5048 frame (frame zero), followed by its caller (frame one), and on up the
5053 @kindex bt @r{(@code{backtrace})}
5056 Print a backtrace of the entire stack: one line per frame for all
5057 frames in the stack.
5059 You can stop the backtrace at any time by typing the system interrupt
5060 character, normally @kbd{Ctrl-c}.
5062 @item backtrace @var{n}
5064 Similar, but print only the innermost @var{n} frames.
5066 @item backtrace -@var{n}
5068 Similar, but print only the outermost @var{n} frames.
5070 @item backtrace full
5072 @itemx bt full @var{n}
5073 @itemx bt full -@var{n}
5074 Print the values of the local variables also. @var{n} specifies the
5075 number of frames to print, as described above.
5080 The names @code{where} and @code{info stack} (abbreviated @code{info s})
5081 are additional aliases for @code{backtrace}.
5083 @cindex multiple threads, backtrace
5084 In a multi-threaded program, @value{GDBN} by default shows the
5085 backtrace only for the current thread. To display the backtrace for
5086 several or all of the threads, use the command @code{thread apply}
5087 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
5088 apply all backtrace}, @value{GDBN} will display the backtrace for all
5089 the threads; this is handy when you debug a core dump of a
5090 multi-threaded program.
5092 Each line in the backtrace shows the frame number and the function name.
5093 The program counter value is also shown---unless you use @code{set
5094 print address off}. The backtrace also shows the source file name and
5095 line number, as well as the arguments to the function. The program
5096 counter value is omitted if it is at the beginning of the code for that
5099 Here is an example of a backtrace. It was made with the command
5100 @samp{bt 3}, so it shows the innermost three frames.
5104 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5106 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
5107 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
5109 (More stack frames follow...)
5114 The display for frame zero does not begin with a program counter
5115 value, indicating that your program has stopped at the beginning of the
5116 code for line @code{993} of @code{builtin.c}.
5118 @cindex value optimized out, in backtrace
5119 @cindex function call arguments, optimized out
5120 If your program was compiled with optimizations, some compilers will
5121 optimize away arguments passed to functions if those arguments are
5122 never used after the call. Such optimizations generate code that
5123 passes arguments through registers, but doesn't store those arguments
5124 in the stack frame. @value{GDBN} has no way of displaying such
5125 arguments in stack frames other than the innermost one. Here's what
5126 such a backtrace might look like:
5130 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5132 #1 0x6e38 in expand_macro (sym=<value optimized out>) at macro.c:242
5133 #2 0x6840 in expand_token (obs=0x0, t=<value optimized out>, td=0xf7fffb08)
5135 (More stack frames follow...)
5140 The values of arguments that were not saved in their stack frames are
5141 shown as @samp{<value optimized out>}.
5143 If you need to display the values of such optimized-out arguments,
5144 either deduce that from other variables whose values depend on the one
5145 you are interested in, or recompile without optimizations.
5147 @cindex backtrace beyond @code{main} function
5148 @cindex program entry point
5149 @cindex startup code, and backtrace
5150 Most programs have a standard user entry point---a place where system
5151 libraries and startup code transition into user code. For C this is
5152 @code{main}@footnote{
5153 Note that embedded programs (the so-called ``free-standing''
5154 environment) are not required to have a @code{main} function as the
5155 entry point. They could even have multiple entry points.}.
5156 When @value{GDBN} finds the entry function in a backtrace
5157 it will terminate the backtrace, to avoid tracing into highly
5158 system-specific (and generally uninteresting) code.
5160 If you need to examine the startup code, or limit the number of levels
5161 in a backtrace, you can change this behavior:
5164 @item set backtrace past-main
5165 @itemx set backtrace past-main on
5166 @kindex set backtrace
5167 Backtraces will continue past the user entry point.
5169 @item set backtrace past-main off
5170 Backtraces will stop when they encounter the user entry point. This is the
5173 @item show backtrace past-main
5174 @kindex show backtrace
5175 Display the current user entry point backtrace policy.
5177 @item set backtrace past-entry
5178 @itemx set backtrace past-entry on
5179 Backtraces will continue past the internal entry point of an application.
5180 This entry point is encoded by the linker when the application is built,
5181 and is likely before the user entry point @code{main} (or equivalent) is called.
5183 @item set backtrace past-entry off
5184 Backtraces will stop when they encounter the internal entry point of an
5185 application. This is the default.
5187 @item show backtrace past-entry
5188 Display the current internal entry point backtrace policy.
5190 @item set backtrace limit @var{n}
5191 @itemx set backtrace limit 0
5192 @cindex backtrace limit
5193 Limit the backtrace to @var{n} levels. A value of zero means
5196 @item show backtrace limit
5197 Display the current limit on backtrace levels.
5201 @section Selecting a Frame
5203 Most commands for examining the stack and other data in your program work on
5204 whichever stack frame is selected at the moment. Here are the commands for
5205 selecting a stack frame; all of them finish by printing a brief description
5206 of the stack frame just selected.
5209 @kindex frame@r{, selecting}
5210 @kindex f @r{(@code{frame})}
5213 Select frame number @var{n}. Recall that frame zero is the innermost
5214 (currently executing) frame, frame one is the frame that called the
5215 innermost one, and so on. The highest-numbered frame is the one for
5218 @item frame @var{addr}
5220 Select the frame at address @var{addr}. This is useful mainly if the
5221 chaining of stack frames has been damaged by a bug, making it
5222 impossible for @value{GDBN} to assign numbers properly to all frames. In
5223 addition, this can be useful when your program has multiple stacks and
5224 switches between them.
5226 On the SPARC architecture, @code{frame} needs two addresses to
5227 select an arbitrary frame: a frame pointer and a stack pointer.
5229 On the MIPS and Alpha architecture, it needs two addresses: a stack
5230 pointer and a program counter.
5232 On the 29k architecture, it needs three addresses: a register stack
5233 pointer, a program counter, and a memory stack pointer.
5237 Move @var{n} frames up the stack. For positive numbers @var{n}, this
5238 advances toward the outermost frame, to higher frame numbers, to frames
5239 that have existed longer. @var{n} defaults to one.
5242 @kindex do @r{(@code{down})}
5244 Move @var{n} frames down the stack. For positive numbers @var{n}, this
5245 advances toward the innermost frame, to lower frame numbers, to frames
5246 that were created more recently. @var{n} defaults to one. You may
5247 abbreviate @code{down} as @code{do}.
5250 All of these commands end by printing two lines of output describing the
5251 frame. The first line shows the frame number, the function name, the
5252 arguments, and the source file and line number of execution in that
5253 frame. The second line shows the text of that source line.
5261 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
5263 10 read_input_file (argv[i]);
5267 After such a printout, the @code{list} command with no arguments
5268 prints ten lines centered on the point of execution in the frame.
5269 You can also edit the program at the point of execution with your favorite
5270 editing program by typing @code{edit}.
5271 @xref{List, ,Printing Source Lines},
5275 @kindex down-silently
5277 @item up-silently @var{n}
5278 @itemx down-silently @var{n}
5279 These two commands are variants of @code{up} and @code{down},
5280 respectively; they differ in that they do their work silently, without
5281 causing display of the new frame. They are intended primarily for use
5282 in @value{GDBN} command scripts, where the output might be unnecessary and
5287 @section Information About a Frame
5289 There are several other commands to print information about the selected
5295 When used without any argument, this command does not change which
5296 frame is selected, but prints a brief description of the currently
5297 selected stack frame. It can be abbreviated @code{f}. With an
5298 argument, this command is used to select a stack frame.
5299 @xref{Selection, ,Selecting a Frame}.
5302 @kindex info f @r{(@code{info frame})}
5305 This command prints a verbose description of the selected stack frame,
5310 the address of the frame
5312 the address of the next frame down (called by this frame)
5314 the address of the next frame up (caller of this frame)
5316 the language in which the source code corresponding to this frame is written
5318 the address of the frame's arguments
5320 the address of the frame's local variables
5322 the program counter saved in it (the address of execution in the caller frame)
5324 which registers were saved in the frame
5327 @noindent The verbose description is useful when
5328 something has gone wrong that has made the stack format fail to fit
5329 the usual conventions.
5331 @item info frame @var{addr}
5332 @itemx info f @var{addr}
5333 Print a verbose description of the frame at address @var{addr}, without
5334 selecting that frame. The selected frame remains unchanged by this
5335 command. This requires the same kind of address (more than one for some
5336 architectures) that you specify in the @code{frame} command.
5337 @xref{Selection, ,Selecting a Frame}.
5341 Print the arguments of the selected frame, each on a separate line.
5345 Print the local variables of the selected frame, each on a separate
5346 line. These are all variables (declared either static or automatic)
5347 accessible at the point of execution of the selected frame.
5350 @cindex catch exceptions, list active handlers
5351 @cindex exception handlers, how to list
5353 Print a list of all the exception handlers that are active in the
5354 current stack frame at the current point of execution. To see other
5355 exception handlers, visit the associated frame (using the @code{up},
5356 @code{down}, or @code{frame} commands); then type @code{info catch}.
5357 @xref{Set Catchpoints, , Setting Catchpoints}.
5363 @chapter Examining Source Files
5365 @value{GDBN} can print parts of your program's source, since the debugging
5366 information recorded in the program tells @value{GDBN} what source files were
5367 used to build it. When your program stops, @value{GDBN} spontaneously prints
5368 the line where it stopped. Likewise, when you select a stack frame
5369 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
5370 execution in that frame has stopped. You can print other portions of
5371 source files by explicit command.
5373 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
5374 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
5375 @value{GDBN} under @sc{gnu} Emacs}.
5378 * List:: Printing source lines
5379 * Specify Location:: How to specify code locations
5380 * Edit:: Editing source files
5381 * Search:: Searching source files
5382 * Source Path:: Specifying source directories
5383 * Machine Code:: Source and machine code
5387 @section Printing Source Lines
5390 @kindex l @r{(@code{list})}
5391 To print lines from a source file, use the @code{list} command
5392 (abbreviated @code{l}). By default, ten lines are printed.
5393 There are several ways to specify what part of the file you want to
5394 print; see @ref{Specify Location}, for the full list.
5396 Here are the forms of the @code{list} command most commonly used:
5399 @item list @var{linenum}
5400 Print lines centered around line number @var{linenum} in the
5401 current source file.
5403 @item list @var{function}
5404 Print lines centered around the beginning of function
5408 Print more lines. If the last lines printed were printed with a
5409 @code{list} command, this prints lines following the last lines
5410 printed; however, if the last line printed was a solitary line printed
5411 as part of displaying a stack frame (@pxref{Stack, ,Examining the
5412 Stack}), this prints lines centered around that line.
5415 Print lines just before the lines last printed.
5418 @cindex @code{list}, how many lines to display
5419 By default, @value{GDBN} prints ten source lines with any of these forms of
5420 the @code{list} command. You can change this using @code{set listsize}:
5423 @kindex set listsize
5424 @item set listsize @var{count}
5425 Make the @code{list} command display @var{count} source lines (unless
5426 the @code{list} argument explicitly specifies some other number).
5428 @kindex show listsize
5430 Display the number of lines that @code{list} prints.
5433 Repeating a @code{list} command with @key{RET} discards the argument,
5434 so it is equivalent to typing just @code{list}. This is more useful
5435 than listing the same lines again. An exception is made for an
5436 argument of @samp{-}; that argument is preserved in repetition so that
5437 each repetition moves up in the source file.
5439 In general, the @code{list} command expects you to supply zero, one or two
5440 @dfn{linespecs}. Linespecs specify source lines; there are several ways
5441 of writing them (@pxref{Specify Location}), but the effect is always
5442 to specify some source line.
5444 Here is a complete description of the possible arguments for @code{list}:
5447 @item list @var{linespec}
5448 Print lines centered around the line specified by @var{linespec}.
5450 @item list @var{first},@var{last}
5451 Print lines from @var{first} to @var{last}. Both arguments are
5452 linespecs. When a @code{list} command has two linespecs, and the
5453 source file of the second linespec is omitted, this refers to
5454 the same source file as the first linespec.
5456 @item list ,@var{last}
5457 Print lines ending with @var{last}.
5459 @item list @var{first},
5460 Print lines starting with @var{first}.
5463 Print lines just after the lines last printed.
5466 Print lines just before the lines last printed.
5469 As described in the preceding table.
5472 @node Specify Location
5473 @section Specifying a Location
5474 @cindex specifying location
5477 Several @value{GDBN} commands accept arguments that specify a location
5478 of your program's code. Since @value{GDBN} is a source-level
5479 debugger, a location usually specifies some line in the source code;
5480 for that reason, locations are also known as @dfn{linespecs}.
5482 Here are all the different ways of specifying a code location that
5483 @value{GDBN} understands:
5487 Specifies the line number @var{linenum} of the current source file.
5490 @itemx +@var{offset}
5491 Specifies the line @var{offset} lines before or after the @dfn{current
5492 line}. For the @code{list} command, the current line is the last one
5493 printed; for the breakpoint commands, this is the line at which
5494 execution stopped in the currently selected @dfn{stack frame}
5495 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
5496 used as the second of the two linespecs in a @code{list} command,
5497 this specifies the line @var{offset} lines up or down from the first
5500 @item @var{filename}:@var{linenum}
5501 Specifies the line @var{linenum} in the source file @var{filename}.
5503 @item @var{function}
5504 Specifies the line that begins the body of the function @var{function}.
5505 For example, in C, this is the line with the open brace.
5507 @item @var{filename}:@var{function}
5508 Specifies the line that begins the body of the function @var{function}
5509 in the file @var{filename}. You only need the file name with a
5510 function name to avoid ambiguity when there are identically named
5511 functions in different source files.
5513 @item *@var{address}
5514 Specifies the program address @var{address}. For line-oriented
5515 commands, such as @code{list} and @code{edit}, this specifies a source
5516 line that contains @var{address}. For @code{break} and other
5517 breakpoint oriented commands, this can be used to set breakpoints in
5518 parts of your program which do not have debugging information or
5521 Here @var{address} may be any expression valid in the current working
5522 language (@pxref{Languages, working language}) that specifies a code
5523 address. In addition, as a convenience, @value{GDBN} extends the
5524 semantics of expressions used in locations to cover the situations
5525 that frequently happen during debugging. Here are the various forms
5529 @item @var{expression}
5530 Any expression valid in the current working language.
5532 @item @var{funcaddr}
5533 An address of a function or procedure derived from its name. In C,
5534 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
5535 simply the function's name @var{function} (and actually a special case
5536 of a valid expression). In Pascal and Modula-2, this is
5537 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
5538 (although the Pascal form also works).
5540 This form specifies the address of the function's first instruction,
5541 before the stack frame and arguments have been set up.
5543 @item '@var{filename}'::@var{funcaddr}
5544 Like @var{funcaddr} above, but also specifies the name of the source
5545 file explicitly. This is useful if the name of the function does not
5546 specify the function unambiguously, e.g., if there are several
5547 functions with identical names in different source files.
5554 @section Editing Source Files
5555 @cindex editing source files
5558 @kindex e @r{(@code{edit})}
5559 To edit the lines in a source file, use the @code{edit} command.
5560 The editing program of your choice
5561 is invoked with the current line set to
5562 the active line in the program.
5563 Alternatively, there are several ways to specify what part of the file you
5564 want to print if you want to see other parts of the program:
5567 @item edit @var{location}
5568 Edit the source file specified by @code{location}. Editing starts at
5569 that @var{location}, e.g., at the specified source line of the
5570 specified file. @xref{Specify Location}, for all the possible forms
5571 of the @var{location} argument; here are the forms of the @code{edit}
5572 command most commonly used:
5575 @item edit @var{number}
5576 Edit the current source file with @var{number} as the active line number.
5578 @item edit @var{function}
5579 Edit the file containing @var{function} at the beginning of its definition.
5584 @subsection Choosing your Editor
5585 You can customize @value{GDBN} to use any editor you want
5587 The only restriction is that your editor (say @code{ex}), recognizes the
5588 following command-line syntax:
5590 ex +@var{number} file
5592 The optional numeric value +@var{number} specifies the number of the line in
5593 the file where to start editing.}.
5594 By default, it is @file{@value{EDITOR}}, but you can change this
5595 by setting the environment variable @code{EDITOR} before using
5596 @value{GDBN}. For example, to configure @value{GDBN} to use the
5597 @code{vi} editor, you could use these commands with the @code{sh} shell:
5603 or in the @code{csh} shell,
5605 setenv EDITOR /usr/bin/vi
5610 @section Searching Source Files
5611 @cindex searching source files
5613 There are two commands for searching through the current source file for a
5618 @kindex forward-search
5619 @item forward-search @var{regexp}
5620 @itemx search @var{regexp}
5621 The command @samp{forward-search @var{regexp}} checks each line,
5622 starting with the one following the last line listed, for a match for
5623 @var{regexp}. It lists the line that is found. You can use the
5624 synonym @samp{search @var{regexp}} or abbreviate the command name as
5627 @kindex reverse-search
5628 @item reverse-search @var{regexp}
5629 The command @samp{reverse-search @var{regexp}} checks each line, starting
5630 with the one before the last line listed and going backward, for a match
5631 for @var{regexp}. It lists the line that is found. You can abbreviate
5632 this command as @code{rev}.
5636 @section Specifying Source Directories
5639 @cindex directories for source files
5640 Executable programs sometimes do not record the directories of the source
5641 files from which they were compiled, just the names. Even when they do,
5642 the directories could be moved between the compilation and your debugging
5643 session. @value{GDBN} has a list of directories to search for source files;
5644 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
5645 it tries all the directories in the list, in the order they are present
5646 in the list, until it finds a file with the desired name.
5648 For example, suppose an executable references the file
5649 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
5650 @file{/mnt/cross}. The file is first looked up literally; if this
5651 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
5652 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
5653 message is printed. @value{GDBN} does not look up the parts of the
5654 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
5655 Likewise, the subdirectories of the source path are not searched: if
5656 the source path is @file{/mnt/cross}, and the binary refers to
5657 @file{foo.c}, @value{GDBN} would not find it under
5658 @file{/mnt/cross/usr/src/foo-1.0/lib}.
5660 Plain file names, relative file names with leading directories, file
5661 names containing dots, etc.@: are all treated as described above; for
5662 instance, if the source path is @file{/mnt/cross}, and the source file
5663 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
5664 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
5665 that---@file{/mnt/cross/foo.c}.
5667 Note that the executable search path is @emph{not} used to locate the
5670 Whenever you reset or rearrange the source path, @value{GDBN} clears out
5671 any information it has cached about where source files are found and where
5672 each line is in the file.
5676 When you start @value{GDBN}, its source path includes only @samp{cdir}
5677 and @samp{cwd}, in that order.
5678 To add other directories, use the @code{directory} command.
5680 The search path is used to find both program source files and @value{GDBN}
5681 script files (read using the @samp{-command} option and @samp{source} command).
5683 In addition to the source path, @value{GDBN} provides a set of commands
5684 that manage a list of source path substitution rules. A @dfn{substitution
5685 rule} specifies how to rewrite source directories stored in the program's
5686 debug information in case the sources were moved to a different
5687 directory between compilation and debugging. A rule is made of
5688 two strings, the first specifying what needs to be rewritten in
5689 the path, and the second specifying how it should be rewritten.
5690 In @ref{set substitute-path}, we name these two parts @var{from} and
5691 @var{to} respectively. @value{GDBN} does a simple string replacement
5692 of @var{from} with @var{to} at the start of the directory part of the
5693 source file name, and uses that result instead of the original file
5694 name to look up the sources.
5696 Using the previous example, suppose the @file{foo-1.0} tree has been
5697 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
5698 @value{GDBN} to replace @file{/usr/src} in all source path names with
5699 @file{/mnt/cross}. The first lookup will then be
5700 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
5701 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
5702 substitution rule, use the @code{set substitute-path} command
5703 (@pxref{set substitute-path}).
5705 To avoid unexpected substitution results, a rule is applied only if the
5706 @var{from} part of the directory name ends at a directory separator.
5707 For instance, a rule substituting @file{/usr/source} into
5708 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
5709 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
5710 is applied only at the beginning of the directory name, this rule will
5711 not be applied to @file{/root/usr/source/baz.c} either.
5713 In many cases, you can achieve the same result using the @code{directory}
5714 command. However, @code{set substitute-path} can be more efficient in
5715 the case where the sources are organized in a complex tree with multiple
5716 subdirectories. With the @code{directory} command, you need to add each
5717 subdirectory of your project. If you moved the entire tree while
5718 preserving its internal organization, then @code{set substitute-path}
5719 allows you to direct the debugger to all the sources with one single
5722 @code{set substitute-path} is also more than just a shortcut command.
5723 The source path is only used if the file at the original location no
5724 longer exists. On the other hand, @code{set substitute-path} modifies
5725 the debugger behavior to look at the rewritten location instead. So, if
5726 for any reason a source file that is not relevant to your executable is
5727 located at the original location, a substitution rule is the only
5728 method available to point @value{GDBN} at the new location.
5731 @item directory @var{dirname} @dots{}
5732 @item dir @var{dirname} @dots{}
5733 Add directory @var{dirname} to the front of the source path. Several
5734 directory names may be given to this command, separated by @samp{:}
5735 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
5736 part of absolute file names) or
5737 whitespace. You may specify a directory that is already in the source
5738 path; this moves it forward, so @value{GDBN} searches it sooner.
5742 @vindex $cdir@r{, convenience variable}
5743 @vindex $cwd@r{, convenience variable}
5744 @cindex compilation directory
5745 @cindex current directory
5746 @cindex working directory
5747 @cindex directory, current
5748 @cindex directory, compilation
5749 You can use the string @samp{$cdir} to refer to the compilation
5750 directory (if one is recorded), and @samp{$cwd} to refer to the current
5751 working directory. @samp{$cwd} is not the same as @samp{.}---the former
5752 tracks the current working directory as it changes during your @value{GDBN}
5753 session, while the latter is immediately expanded to the current
5754 directory at the time you add an entry to the source path.
5757 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
5759 @c RET-repeat for @code{directory} is explicitly disabled, but since
5760 @c repeating it would be a no-op we do not say that. (thanks to RMS)
5762 @item show directories
5763 @kindex show directories
5764 Print the source path: show which directories it contains.
5766 @anchor{set substitute-path}
5767 @item set substitute-path @var{from} @var{to}
5768 @kindex set substitute-path
5769 Define a source path substitution rule, and add it at the end of the
5770 current list of existing substitution rules. If a rule with the same
5771 @var{from} was already defined, then the old rule is also deleted.
5773 For example, if the file @file{/foo/bar/baz.c} was moved to
5774 @file{/mnt/cross/baz.c}, then the command
5777 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
5781 will tell @value{GDBN} to replace @samp{/usr/src} with
5782 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
5783 @file{baz.c} even though it was moved.
5785 In the case when more than one substitution rule have been defined,
5786 the rules are evaluated one by one in the order where they have been
5787 defined. The first one matching, if any, is selected to perform
5790 For instance, if we had entered the following commands:
5793 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
5794 (@value{GDBP}) set substitute-path /usr/src /mnt/src
5798 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
5799 @file{/mnt/include/defs.h} by using the first rule. However, it would
5800 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
5801 @file{/mnt/src/lib/foo.c}.
5804 @item unset substitute-path [path]
5805 @kindex unset substitute-path
5806 If a path is specified, search the current list of substitution rules
5807 for a rule that would rewrite that path. Delete that rule if found.
5808 A warning is emitted by the debugger if no rule could be found.
5810 If no path is specified, then all substitution rules are deleted.
5812 @item show substitute-path [path]
5813 @kindex show substitute-path
5814 If a path is specified, then print the source path substitution rule
5815 which would rewrite that path, if any.
5817 If no path is specified, then print all existing source path substitution
5822 If your source path is cluttered with directories that are no longer of
5823 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
5824 versions of source. You can correct the situation as follows:
5828 Use @code{directory} with no argument to reset the source path to its default value.
5831 Use @code{directory} with suitable arguments to reinstall the
5832 directories you want in the source path. You can add all the
5833 directories in one command.
5837 @section Source and Machine Code
5838 @cindex source line and its code address
5840 You can use the command @code{info line} to map source lines to program
5841 addresses (and vice versa), and the command @code{disassemble} to display
5842 a range of addresses as machine instructions. When run under @sc{gnu} Emacs
5843 mode, the @code{info line} command causes the arrow to point to the
5844 line specified. Also, @code{info line} prints addresses in symbolic form as
5849 @item info line @var{linespec}
5850 Print the starting and ending addresses of the compiled code for
5851 source line @var{linespec}. You can specify source lines in any of
5852 the ways documented in @ref{Specify Location}.
5855 For example, we can use @code{info line} to discover the location of
5856 the object code for the first line of function
5857 @code{m4_changequote}:
5859 @c FIXME: I think this example should also show the addresses in
5860 @c symbolic form, as they usually would be displayed.
5862 (@value{GDBP}) info line m4_changequote
5863 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
5867 @cindex code address and its source line
5868 We can also inquire (using @code{*@var{addr}} as the form for
5869 @var{linespec}) what source line covers a particular address:
5871 (@value{GDBP}) info line *0x63ff
5872 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
5875 @cindex @code{$_} and @code{info line}
5876 @cindex @code{x} command, default address
5877 @kindex x@r{(examine), and} info line
5878 After @code{info line}, the default address for the @code{x} command
5879 is changed to the starting address of the line, so that @samp{x/i} is
5880 sufficient to begin examining the machine code (@pxref{Memory,
5881 ,Examining Memory}). Also, this address is saved as the value of the
5882 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
5887 @cindex assembly instructions
5888 @cindex instructions, assembly
5889 @cindex machine instructions
5890 @cindex listing machine instructions
5892 @itemx disassemble /m
5893 This specialized command dumps a range of memory as machine
5894 instructions. It can also print mixed source+disassembly by specifying
5895 the @code{/m} modifier.
5896 The default memory range is the function surrounding the
5897 program counter of the selected frame. A single argument to this
5898 command is a program counter value; @value{GDBN} dumps the function
5899 surrounding this value. Two arguments specify a range of addresses
5900 (first inclusive, second exclusive) to dump.
5903 The following example shows the disassembly of a range of addresses of
5904 HP PA-RISC 2.0 code:
5907 (@value{GDBP}) disas 0x32c4 0x32e4
5908 Dump of assembler code from 0x32c4 to 0x32e4:
5909 0x32c4 <main+204>: addil 0,dp
5910 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
5911 0x32cc <main+212>: ldil 0x3000,r31
5912 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
5913 0x32d4 <main+220>: ldo 0(r31),rp
5914 0x32d8 <main+224>: addil -0x800,dp
5915 0x32dc <main+228>: ldo 0x588(r1),r26
5916 0x32e0 <main+232>: ldil 0x3000,r31
5917 End of assembler dump.
5920 Here is an example showing mixed source+assembly for Intel x86:
5923 (@value{GDBP}) disas /m main
5924 Dump of assembler code for function main:
5926 0x08048330 <main+0>: push %ebp
5927 0x08048331 <main+1>: mov %esp,%ebp
5928 0x08048333 <main+3>: sub $0x8,%esp
5929 0x08048336 <main+6>: and $0xfffffff0,%esp
5930 0x08048339 <main+9>: sub $0x10,%esp
5932 6 printf ("Hello.\n");
5933 0x0804833c <main+12>: movl $0x8048440,(%esp)
5934 0x08048343 <main+19>: call 0x8048284 <puts@@plt>
5938 0x08048348 <main+24>: mov $0x0,%eax
5939 0x0804834d <main+29>: leave
5940 0x0804834e <main+30>: ret
5942 End of assembler dump.
5945 Some architectures have more than one commonly-used set of instruction
5946 mnemonics or other syntax.
5948 For programs that were dynamically linked and use shared libraries,
5949 instructions that call functions or branch to locations in the shared
5950 libraries might show a seemingly bogus location---it's actually a
5951 location of the relocation table. On some architectures, @value{GDBN}
5952 might be able to resolve these to actual function names.
5955 @kindex set disassembly-flavor
5956 @cindex Intel disassembly flavor
5957 @cindex AT&T disassembly flavor
5958 @item set disassembly-flavor @var{instruction-set}
5959 Select the instruction set to use when disassembling the
5960 program via the @code{disassemble} or @code{x/i} commands.
5962 Currently this command is only defined for the Intel x86 family. You
5963 can set @var{instruction-set} to either @code{intel} or @code{att}.
5964 The default is @code{att}, the AT&T flavor used by default by Unix
5965 assemblers for x86-based targets.
5967 @kindex show disassembly-flavor
5968 @item show disassembly-flavor
5969 Show the current setting of the disassembly flavor.
5974 @chapter Examining Data
5976 @cindex printing data
5977 @cindex examining data
5980 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
5981 @c document because it is nonstandard... Under Epoch it displays in a
5982 @c different window or something like that.
5983 The usual way to examine data in your program is with the @code{print}
5984 command (abbreviated @code{p}), or its synonym @code{inspect}. It
5985 evaluates and prints the value of an expression of the language your
5986 program is written in (@pxref{Languages, ,Using @value{GDBN} with
5987 Different Languages}).
5990 @item print @var{expr}
5991 @itemx print /@var{f} @var{expr}
5992 @var{expr} is an expression (in the source language). By default the
5993 value of @var{expr} is printed in a format appropriate to its data type;
5994 you can choose a different format by specifying @samp{/@var{f}}, where
5995 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
5999 @itemx print /@var{f}
6000 @cindex reprint the last value
6001 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
6002 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
6003 conveniently inspect the same value in an alternative format.
6006 A more low-level way of examining data is with the @code{x} command.
6007 It examines data in memory at a specified address and prints it in a
6008 specified format. @xref{Memory, ,Examining Memory}.
6010 If you are interested in information about types, or about how the
6011 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
6012 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
6016 * Expressions:: Expressions
6017 * Ambiguous Expressions:: Ambiguous Expressions
6018 * Variables:: Program variables
6019 * Arrays:: Artificial arrays
6020 * Output Formats:: Output formats
6021 * Memory:: Examining memory
6022 * Auto Display:: Automatic display
6023 * Print Settings:: Print settings
6024 * Value History:: Value history
6025 * Convenience Vars:: Convenience variables
6026 * Registers:: Registers
6027 * Floating Point Hardware:: Floating point hardware
6028 * Vector Unit:: Vector Unit
6029 * OS Information:: Auxiliary data provided by operating system
6030 * Memory Region Attributes:: Memory region attributes
6031 * Dump/Restore Files:: Copy between memory and a file
6032 * Core File Generation:: Cause a program dump its core
6033 * Character Sets:: Debugging programs that use a different
6034 character set than GDB does
6035 * Caching Remote Data:: Data caching for remote targets
6036 * Searching Memory:: Searching memory for a sequence of bytes
6040 @section Expressions
6043 @code{print} and many other @value{GDBN} commands accept an expression and
6044 compute its value. Any kind of constant, variable or operator defined
6045 by the programming language you are using is valid in an expression in
6046 @value{GDBN}. This includes conditional expressions, function calls,
6047 casts, and string constants. It also includes preprocessor macros, if
6048 you compiled your program to include this information; see
6051 @cindex arrays in expressions
6052 @value{GDBN} supports array constants in expressions input by
6053 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
6054 you can use the command @code{print @{1, 2, 3@}} to create an array
6055 of three integers. If you pass an array to a function or assign it
6056 to a program variable, @value{GDBN} copies the array to memory that
6057 is @code{malloc}ed in the target program.
6059 Because C is so widespread, most of the expressions shown in examples in
6060 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
6061 Languages}, for information on how to use expressions in other
6064 In this section, we discuss operators that you can use in @value{GDBN}
6065 expressions regardless of your programming language.
6067 @cindex casts, in expressions
6068 Casts are supported in all languages, not just in C, because it is so
6069 useful to cast a number into a pointer in order to examine a structure
6070 at that address in memory.
6071 @c FIXME: casts supported---Mod2 true?
6073 @value{GDBN} supports these operators, in addition to those common
6074 to programming languages:
6078 @samp{@@} is a binary operator for treating parts of memory as arrays.
6079 @xref{Arrays, ,Artificial Arrays}, for more information.
6082 @samp{::} allows you to specify a variable in terms of the file or
6083 function where it is defined. @xref{Variables, ,Program Variables}.
6085 @cindex @{@var{type}@}
6086 @cindex type casting memory
6087 @cindex memory, viewing as typed object
6088 @cindex casts, to view memory
6089 @item @{@var{type}@} @var{addr}
6090 Refers to an object of type @var{type} stored at address @var{addr} in
6091 memory. @var{addr} may be any expression whose value is an integer or
6092 pointer (but parentheses are required around binary operators, just as in
6093 a cast). This construct is allowed regardless of what kind of data is
6094 normally supposed to reside at @var{addr}.
6097 @node Ambiguous Expressions
6098 @section Ambiguous Expressions
6099 @cindex ambiguous expressions
6101 Expressions can sometimes contain some ambiguous elements. For instance,
6102 some programming languages (notably Ada, C@t{++} and Objective-C) permit
6103 a single function name to be defined several times, for application in
6104 different contexts. This is called @dfn{overloading}. Another example
6105 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
6106 templates and is typically instantiated several times, resulting in
6107 the same function name being defined in different contexts.
6109 In some cases and depending on the language, it is possible to adjust
6110 the expression to remove the ambiguity. For instance in C@t{++}, you
6111 can specify the signature of the function you want to break on, as in
6112 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
6113 qualified name of your function often makes the expression unambiguous
6116 When an ambiguity that needs to be resolved is detected, the debugger
6117 has the capability to display a menu of numbered choices for each
6118 possibility, and then waits for the selection with the prompt @samp{>}.
6119 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
6120 aborts the current command. If the command in which the expression was
6121 used allows more than one choice to be selected, the next option in the
6122 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
6125 For example, the following session excerpt shows an attempt to set a
6126 breakpoint at the overloaded symbol @code{String::after}.
6127 We choose three particular definitions of that function name:
6129 @c FIXME! This is likely to change to show arg type lists, at least
6132 (@value{GDBP}) b String::after
6135 [2] file:String.cc; line number:867
6136 [3] file:String.cc; line number:860
6137 [4] file:String.cc; line number:875
6138 [5] file:String.cc; line number:853
6139 [6] file:String.cc; line number:846
6140 [7] file:String.cc; line number:735
6142 Breakpoint 1 at 0xb26c: file String.cc, line 867.
6143 Breakpoint 2 at 0xb344: file String.cc, line 875.
6144 Breakpoint 3 at 0xafcc: file String.cc, line 846.
6145 Multiple breakpoints were set.
6146 Use the "delete" command to delete unwanted
6153 @kindex set multiple-symbols
6154 @item set multiple-symbols @var{mode}
6155 @cindex multiple-symbols menu
6157 This option allows you to adjust the debugger behavior when an expression
6160 By default, @var{mode} is set to @code{all}. If the command with which
6161 the expression is used allows more than one choice, then @value{GDBN}
6162 automatically selects all possible choices. For instance, inserting
6163 a breakpoint on a function using an ambiguous name results in a breakpoint
6164 inserted on each possible match. However, if a unique choice must be made,
6165 then @value{GDBN} uses the menu to help you disambiguate the expression.
6166 For instance, printing the address of an overloaded function will result
6167 in the use of the menu.
6169 When @var{mode} is set to @code{ask}, the debugger always uses the menu
6170 when an ambiguity is detected.
6172 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
6173 an error due to the ambiguity and the command is aborted.
6175 @kindex show multiple-symbols
6176 @item show multiple-symbols
6177 Show the current value of the @code{multiple-symbols} setting.
6181 @section Program Variables
6183 The most common kind of expression to use is the name of a variable
6186 Variables in expressions are understood in the selected stack frame
6187 (@pxref{Selection, ,Selecting a Frame}); they must be either:
6191 global (or file-static)
6198 visible according to the scope rules of the
6199 programming language from the point of execution in that frame
6202 @noindent This means that in the function
6217 you can examine and use the variable @code{a} whenever your program is
6218 executing within the function @code{foo}, but you can only use or
6219 examine the variable @code{b} while your program is executing inside
6220 the block where @code{b} is declared.
6222 @cindex variable name conflict
6223 There is an exception: you can refer to a variable or function whose
6224 scope is a single source file even if the current execution point is not
6225 in this file. But it is possible to have more than one such variable or
6226 function with the same name (in different source files). If that
6227 happens, referring to that name has unpredictable effects. If you wish,
6228 you can specify a static variable in a particular function or file,
6229 using the colon-colon (@code{::}) notation:
6231 @cindex colon-colon, context for variables/functions
6233 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
6234 @cindex @code{::}, context for variables/functions
6237 @var{file}::@var{variable}
6238 @var{function}::@var{variable}
6242 Here @var{file} or @var{function} is the name of the context for the
6243 static @var{variable}. In the case of file names, you can use quotes to
6244 make sure @value{GDBN} parses the file name as a single word---for example,
6245 to print a global value of @code{x} defined in @file{f2.c}:
6248 (@value{GDBP}) p 'f2.c'::x
6251 @cindex C@t{++} scope resolution
6252 This use of @samp{::} is very rarely in conflict with the very similar
6253 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
6254 scope resolution operator in @value{GDBN} expressions.
6255 @c FIXME: Um, so what happens in one of those rare cases where it's in
6258 @cindex wrong values
6259 @cindex variable values, wrong
6260 @cindex function entry/exit, wrong values of variables
6261 @cindex optimized code, wrong values of variables
6263 @emph{Warning:} Occasionally, a local variable may appear to have the
6264 wrong value at certain points in a function---just after entry to a new
6265 scope, and just before exit.
6267 You may see this problem when you are stepping by machine instructions.
6268 This is because, on most machines, it takes more than one instruction to
6269 set up a stack frame (including local variable definitions); if you are
6270 stepping by machine instructions, variables may appear to have the wrong
6271 values until the stack frame is completely built. On exit, it usually
6272 also takes more than one machine instruction to destroy a stack frame;
6273 after you begin stepping through that group of instructions, local
6274 variable definitions may be gone.
6276 This may also happen when the compiler does significant optimizations.
6277 To be sure of always seeing accurate values, turn off all optimization
6280 @cindex ``No symbol "foo" in current context''
6281 Another possible effect of compiler optimizations is to optimize
6282 unused variables out of existence, or assign variables to registers (as
6283 opposed to memory addresses). Depending on the support for such cases
6284 offered by the debug info format used by the compiler, @value{GDBN}
6285 might not be able to display values for such local variables. If that
6286 happens, @value{GDBN} will print a message like this:
6289 No symbol "foo" in current context.
6292 To solve such problems, either recompile without optimizations, or use a
6293 different debug info format, if the compiler supports several such
6294 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
6295 usually supports the @option{-gstabs+} option. @option{-gstabs+}
6296 produces debug info in a format that is superior to formats such as
6297 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
6298 an effective form for debug info. @xref{Debugging Options,,Options
6299 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
6300 Compiler Collection (GCC)}.
6301 @xref{C, ,C and C@t{++}}, for more information about debug info formats
6302 that are best suited to C@t{++} programs.
6304 If you ask to print an object whose contents are unknown to
6305 @value{GDBN}, e.g., because its data type is not completely specified
6306 by the debug information, @value{GDBN} will say @samp{<incomplete
6307 type>}. @xref{Symbols, incomplete type}, for more about this.
6309 Strings are identified as arrays of @code{char} values without specified
6310 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
6311 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
6312 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
6313 defines literal string type @code{"char"} as @code{char} without a sign.
6318 signed char var1[] = "A";
6321 You get during debugging
6326 $2 = @{65 'A', 0 '\0'@}
6330 @section Artificial Arrays
6332 @cindex artificial array
6334 @kindex @@@r{, referencing memory as an array}
6335 It is often useful to print out several successive objects of the
6336 same type in memory; a section of an array, or an array of
6337 dynamically determined size for which only a pointer exists in the
6340 You can do this by referring to a contiguous span of memory as an
6341 @dfn{artificial array}, using the binary operator @samp{@@}. The left
6342 operand of @samp{@@} should be the first element of the desired array
6343 and be an individual object. The right operand should be the desired length
6344 of the array. The result is an array value whose elements are all of
6345 the type of the left argument. The first element is actually the left
6346 argument; the second element comes from bytes of memory immediately
6347 following those that hold the first element, and so on. Here is an
6348 example. If a program says
6351 int *array = (int *) malloc (len * sizeof (int));
6355 you can print the contents of @code{array} with
6361 The left operand of @samp{@@} must reside in memory. Array values made
6362 with @samp{@@} in this way behave just like other arrays in terms of
6363 subscripting, and are coerced to pointers when used in expressions.
6364 Artificial arrays most often appear in expressions via the value history
6365 (@pxref{Value History, ,Value History}), after printing one out.
6367 Another way to create an artificial array is to use a cast.
6368 This re-interprets a value as if it were an array.
6369 The value need not be in memory:
6371 (@value{GDBP}) p/x (short[2])0x12345678
6372 $1 = @{0x1234, 0x5678@}
6375 As a convenience, if you leave the array length out (as in
6376 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
6377 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
6379 (@value{GDBP}) p/x (short[])0x12345678
6380 $2 = @{0x1234, 0x5678@}
6383 Sometimes the artificial array mechanism is not quite enough; in
6384 moderately complex data structures, the elements of interest may not
6385 actually be adjacent---for example, if you are interested in the values
6386 of pointers in an array. One useful work-around in this situation is
6387 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
6388 Variables}) as a counter in an expression that prints the first
6389 interesting value, and then repeat that expression via @key{RET}. For
6390 instance, suppose you have an array @code{dtab} of pointers to
6391 structures, and you are interested in the values of a field @code{fv}
6392 in each structure. Here is an example of what you might type:
6402 @node Output Formats
6403 @section Output Formats
6405 @cindex formatted output
6406 @cindex output formats
6407 By default, @value{GDBN} prints a value according to its data type. Sometimes
6408 this is not what you want. For example, you might want to print a number
6409 in hex, or a pointer in decimal. Or you might want to view data in memory
6410 at a certain address as a character string or as an instruction. To do
6411 these things, specify an @dfn{output format} when you print a value.
6413 The simplest use of output formats is to say how to print a value
6414 already computed. This is done by starting the arguments of the
6415 @code{print} command with a slash and a format letter. The format
6416 letters supported are:
6420 Regard the bits of the value as an integer, and print the integer in
6424 Print as integer in signed decimal.
6427 Print as integer in unsigned decimal.
6430 Print as integer in octal.
6433 Print as integer in binary. The letter @samp{t} stands for ``two''.
6434 @footnote{@samp{b} cannot be used because these format letters are also
6435 used with the @code{x} command, where @samp{b} stands for ``byte'';
6436 see @ref{Memory,,Examining Memory}.}
6439 @cindex unknown address, locating
6440 @cindex locate address
6441 Print as an address, both absolute in hexadecimal and as an offset from
6442 the nearest preceding symbol. You can use this format used to discover
6443 where (in what function) an unknown address is located:
6446 (@value{GDBP}) p/a 0x54320
6447 $3 = 0x54320 <_initialize_vx+396>
6451 The command @code{info symbol 0x54320} yields similar results.
6452 @xref{Symbols, info symbol}.
6455 Regard as an integer and print it as a character constant. This
6456 prints both the numerical value and its character representation. The
6457 character representation is replaced with the octal escape @samp{\nnn}
6458 for characters outside the 7-bit @sc{ascii} range.
6460 Without this format, @value{GDBN} displays @code{char},
6461 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
6462 constants. Single-byte members of vectors are displayed as integer
6466 Regard the bits of the value as a floating point number and print
6467 using typical floating point syntax.
6470 @cindex printing strings
6471 @cindex printing byte arrays
6472 Regard as a string, if possible. With this format, pointers to single-byte
6473 data are displayed as null-terminated strings and arrays of single-byte data
6474 are displayed as fixed-length strings. Other values are displayed in their
6477 Without this format, @value{GDBN} displays pointers to and arrays of
6478 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
6479 strings. Single-byte members of a vector are displayed as an integer
6483 For example, to print the program counter in hex (@pxref{Registers}), type
6490 Note that no space is required before the slash; this is because command
6491 names in @value{GDBN} cannot contain a slash.
6493 To reprint the last value in the value history with a different format,
6494 you can use the @code{print} command with just a format and no
6495 expression. For example, @samp{p/x} reprints the last value in hex.
6498 @section Examining Memory
6500 You can use the command @code{x} (for ``examine'') to examine memory in
6501 any of several formats, independently of your program's data types.
6503 @cindex examining memory
6505 @kindex x @r{(examine memory)}
6506 @item x/@var{nfu} @var{addr}
6509 Use the @code{x} command to examine memory.
6512 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
6513 much memory to display and how to format it; @var{addr} is an
6514 expression giving the address where you want to start displaying memory.
6515 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
6516 Several commands set convenient defaults for @var{addr}.
6519 @item @var{n}, the repeat count
6520 The repeat count is a decimal integer; the default is 1. It specifies
6521 how much memory (counting by units @var{u}) to display.
6522 @c This really is **decimal**; unaffected by 'set radix' as of GDB
6525 @item @var{f}, the display format
6526 The display format is one of the formats used by @code{print}
6527 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
6528 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
6529 The default is @samp{x} (hexadecimal) initially. The default changes
6530 each time you use either @code{x} or @code{print}.
6532 @item @var{u}, the unit size
6533 The unit size is any of
6539 Halfwords (two bytes).
6541 Words (four bytes). This is the initial default.
6543 Giant words (eight bytes).
6546 Each time you specify a unit size with @code{x}, that size becomes the
6547 default unit the next time you use @code{x}. (For the @samp{s} and
6548 @samp{i} formats, the unit size is ignored and is normally not written.)
6550 @item @var{addr}, starting display address
6551 @var{addr} is the address where you want @value{GDBN} to begin displaying
6552 memory. The expression need not have a pointer value (though it may);
6553 it is always interpreted as an integer address of a byte of memory.
6554 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
6555 @var{addr} is usually just after the last address examined---but several
6556 other commands also set the default address: @code{info breakpoints} (to
6557 the address of the last breakpoint listed), @code{info line} (to the
6558 starting address of a line), and @code{print} (if you use it to display
6559 a value from memory).
6562 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
6563 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
6564 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
6565 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
6566 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
6568 Since the letters indicating unit sizes are all distinct from the
6569 letters specifying output formats, you do not have to remember whether
6570 unit size or format comes first; either order works. The output
6571 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
6572 (However, the count @var{n} must come first; @samp{wx4} does not work.)
6574 Even though the unit size @var{u} is ignored for the formats @samp{s}
6575 and @samp{i}, you might still want to use a count @var{n}; for example,
6576 @samp{3i} specifies that you want to see three machine instructions,
6577 including any operands. For convenience, especially when used with
6578 the @code{display} command, the @samp{i} format also prints branch delay
6579 slot instructions, if any, beyond the count specified, which immediately
6580 follow the last instruction that is within the count. The command
6581 @code{disassemble} gives an alternative way of inspecting machine
6582 instructions; see @ref{Machine Code,,Source and Machine Code}.
6584 All the defaults for the arguments to @code{x} are designed to make it
6585 easy to continue scanning memory with minimal specifications each time
6586 you use @code{x}. For example, after you have inspected three machine
6587 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
6588 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
6589 the repeat count @var{n} is used again; the other arguments default as
6590 for successive uses of @code{x}.
6592 @cindex @code{$_}, @code{$__}, and value history
6593 The addresses and contents printed by the @code{x} command are not saved
6594 in the value history because there is often too much of them and they
6595 would get in the way. Instead, @value{GDBN} makes these values available for
6596 subsequent use in expressions as values of the convenience variables
6597 @code{$_} and @code{$__}. After an @code{x} command, the last address
6598 examined is available for use in expressions in the convenience variable
6599 @code{$_}. The contents of that address, as examined, are available in
6600 the convenience variable @code{$__}.
6602 If the @code{x} command has a repeat count, the address and contents saved
6603 are from the last memory unit printed; this is not the same as the last
6604 address printed if several units were printed on the last line of output.
6606 @cindex remote memory comparison
6607 @cindex verify remote memory image
6608 When you are debugging a program running on a remote target machine
6609 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
6610 remote machine's memory against the executable file you downloaded to
6611 the target. The @code{compare-sections} command is provided for such
6615 @kindex compare-sections
6616 @item compare-sections @r{[}@var{section-name}@r{]}
6617 Compare the data of a loadable section @var{section-name} in the
6618 executable file of the program being debugged with the same section in
6619 the remote machine's memory, and report any mismatches. With no
6620 arguments, compares all loadable sections. This command's
6621 availability depends on the target's support for the @code{"qCRC"}
6626 @section Automatic Display
6627 @cindex automatic display
6628 @cindex display of expressions
6630 If you find that you want to print the value of an expression frequently
6631 (to see how it changes), you might want to add it to the @dfn{automatic
6632 display list} so that @value{GDBN} prints its value each time your program stops.
6633 Each expression added to the list is given a number to identify it;
6634 to remove an expression from the list, you specify that number.
6635 The automatic display looks like this:
6639 3: bar[5] = (struct hack *) 0x3804
6643 This display shows item numbers, expressions and their current values. As with
6644 displays you request manually using @code{x} or @code{print}, you can
6645 specify the output format you prefer; in fact, @code{display} decides
6646 whether to use @code{print} or @code{x} depending your format
6647 specification---it uses @code{x} if you specify either the @samp{i}
6648 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
6652 @item display @var{expr}
6653 Add the expression @var{expr} to the list of expressions to display
6654 each time your program stops. @xref{Expressions, ,Expressions}.
6656 @code{display} does not repeat if you press @key{RET} again after using it.
6658 @item display/@var{fmt} @var{expr}
6659 For @var{fmt} specifying only a display format and not a size or
6660 count, add the expression @var{expr} to the auto-display list but
6661 arrange to display it each time in the specified format @var{fmt}.
6662 @xref{Output Formats,,Output Formats}.
6664 @item display/@var{fmt} @var{addr}
6665 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
6666 number of units, add the expression @var{addr} as a memory address to
6667 be examined each time your program stops. Examining means in effect
6668 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
6671 For example, @samp{display/i $pc} can be helpful, to see the machine
6672 instruction about to be executed each time execution stops (@samp{$pc}
6673 is a common name for the program counter; @pxref{Registers, ,Registers}).
6676 @kindex delete display
6678 @item undisplay @var{dnums}@dots{}
6679 @itemx delete display @var{dnums}@dots{}
6680 Remove item numbers @var{dnums} from the list of expressions to display.
6682 @code{undisplay} does not repeat if you press @key{RET} after using it.
6683 (Otherwise you would just get the error @samp{No display number @dots{}}.)
6685 @kindex disable display
6686 @item disable display @var{dnums}@dots{}
6687 Disable the display of item numbers @var{dnums}. A disabled display
6688 item is not printed automatically, but is not forgotten. It may be
6689 enabled again later.
6691 @kindex enable display
6692 @item enable display @var{dnums}@dots{}
6693 Enable display of item numbers @var{dnums}. It becomes effective once
6694 again in auto display of its expression, until you specify otherwise.
6697 Display the current values of the expressions on the list, just as is
6698 done when your program stops.
6700 @kindex info display
6702 Print the list of expressions previously set up to display
6703 automatically, each one with its item number, but without showing the
6704 values. This includes disabled expressions, which are marked as such.
6705 It also includes expressions which would not be displayed right now
6706 because they refer to automatic variables not currently available.
6709 @cindex display disabled out of scope
6710 If a display expression refers to local variables, then it does not make
6711 sense outside the lexical context for which it was set up. Such an
6712 expression is disabled when execution enters a context where one of its
6713 variables is not defined. For example, if you give the command
6714 @code{display last_char} while inside a function with an argument
6715 @code{last_char}, @value{GDBN} displays this argument while your program
6716 continues to stop inside that function. When it stops elsewhere---where
6717 there is no variable @code{last_char}---the display is disabled
6718 automatically. The next time your program stops where @code{last_char}
6719 is meaningful, you can enable the display expression once again.
6721 @node Print Settings
6722 @section Print Settings
6724 @cindex format options
6725 @cindex print settings
6726 @value{GDBN} provides the following ways to control how arrays, structures,
6727 and symbols are printed.
6730 These settings are useful for debugging programs in any language:
6734 @item set print address
6735 @itemx set print address on
6736 @cindex print/don't print memory addresses
6737 @value{GDBN} prints memory addresses showing the location of stack
6738 traces, structure values, pointer values, breakpoints, and so forth,
6739 even when it also displays the contents of those addresses. The default
6740 is @code{on}. For example, this is what a stack frame display looks like with
6741 @code{set print address on}:
6746 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
6748 530 if (lquote != def_lquote)
6752 @item set print address off
6753 Do not print addresses when displaying their contents. For example,
6754 this is the same stack frame displayed with @code{set print address off}:
6758 (@value{GDBP}) set print addr off
6760 #0 set_quotes (lq="<<", rq=">>") at input.c:530
6761 530 if (lquote != def_lquote)
6765 You can use @samp{set print address off} to eliminate all machine
6766 dependent displays from the @value{GDBN} interface. For example, with
6767 @code{print address off}, you should get the same text for backtraces on
6768 all machines---whether or not they involve pointer arguments.
6771 @item show print address
6772 Show whether or not addresses are to be printed.
6775 When @value{GDBN} prints a symbolic address, it normally prints the
6776 closest earlier symbol plus an offset. If that symbol does not uniquely
6777 identify the address (for example, it is a name whose scope is a single
6778 source file), you may need to clarify. One way to do this is with
6779 @code{info line}, for example @samp{info line *0x4537}. Alternately,
6780 you can set @value{GDBN} to print the source file and line number when
6781 it prints a symbolic address:
6784 @item set print symbol-filename on
6785 @cindex source file and line of a symbol
6786 @cindex symbol, source file and line
6787 Tell @value{GDBN} to print the source file name and line number of a
6788 symbol in the symbolic form of an address.
6790 @item set print symbol-filename off
6791 Do not print source file name and line number of a symbol. This is the
6794 @item show print symbol-filename
6795 Show whether or not @value{GDBN} will print the source file name and
6796 line number of a symbol in the symbolic form of an address.
6799 Another situation where it is helpful to show symbol filenames and line
6800 numbers is when disassembling code; @value{GDBN} shows you the line
6801 number and source file that corresponds to each instruction.
6803 Also, you may wish to see the symbolic form only if the address being
6804 printed is reasonably close to the closest earlier symbol:
6807 @item set print max-symbolic-offset @var{max-offset}
6808 @cindex maximum value for offset of closest symbol
6809 Tell @value{GDBN} to only display the symbolic form of an address if the
6810 offset between the closest earlier symbol and the address is less than
6811 @var{max-offset}. The default is 0, which tells @value{GDBN}
6812 to always print the symbolic form of an address if any symbol precedes it.
6814 @item show print max-symbolic-offset
6815 Ask how large the maximum offset is that @value{GDBN} prints in a
6819 @cindex wild pointer, interpreting
6820 @cindex pointer, finding referent
6821 If you have a pointer and you are not sure where it points, try
6822 @samp{set print symbol-filename on}. Then you can determine the name
6823 and source file location of the variable where it points, using
6824 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
6825 For example, here @value{GDBN} shows that a variable @code{ptt} points
6826 at another variable @code{t}, defined in @file{hi2.c}:
6829 (@value{GDBP}) set print symbol-filename on
6830 (@value{GDBP}) p/a ptt
6831 $4 = 0xe008 <t in hi2.c>
6835 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
6836 does not show the symbol name and filename of the referent, even with
6837 the appropriate @code{set print} options turned on.
6840 Other settings control how different kinds of objects are printed:
6843 @item set print array
6844 @itemx set print array on
6845 @cindex pretty print arrays
6846 Pretty print arrays. This format is more convenient to read,
6847 but uses more space. The default is off.
6849 @item set print array off
6850 Return to compressed format for arrays.
6852 @item show print array
6853 Show whether compressed or pretty format is selected for displaying
6856 @cindex print array indexes
6857 @item set print array-indexes
6858 @itemx set print array-indexes on
6859 Print the index of each element when displaying arrays. May be more
6860 convenient to locate a given element in the array or quickly find the
6861 index of a given element in that printed array. The default is off.
6863 @item set print array-indexes off
6864 Stop printing element indexes when displaying arrays.
6866 @item show print array-indexes
6867 Show whether the index of each element is printed when displaying
6870 @item set print elements @var{number-of-elements}
6871 @cindex number of array elements to print
6872 @cindex limit on number of printed array elements
6873 Set a limit on how many elements of an array @value{GDBN} will print.
6874 If @value{GDBN} is printing a large array, it stops printing after it has
6875 printed the number of elements set by the @code{set print elements} command.
6876 This limit also applies to the display of strings.
6877 When @value{GDBN} starts, this limit is set to 200.
6878 Setting @var{number-of-elements} to zero means that the printing is unlimited.
6880 @item show print elements
6881 Display the number of elements of a large array that @value{GDBN} will print.
6882 If the number is 0, then the printing is unlimited.
6884 @item set print frame-arguments @var{value}
6885 @cindex printing frame argument values
6886 @cindex print all frame argument values
6887 @cindex print frame argument values for scalars only
6888 @cindex do not print frame argument values
6889 This command allows to control how the values of arguments are printed
6890 when the debugger prints a frame (@pxref{Frames}). The possible
6895 The values of all arguments are printed. This is the default.
6898 Print the value of an argument only if it is a scalar. The value of more
6899 complex arguments such as arrays, structures, unions, etc, is replaced
6900 by @code{@dots{}}. Here is an example where only scalar arguments are shown:
6903 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
6908 None of the argument values are printed. Instead, the value of each argument
6909 is replaced by @code{@dots{}}. In this case, the example above now becomes:
6912 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
6917 By default, all argument values are always printed. But this command
6918 can be useful in several cases. For instance, it can be used to reduce
6919 the amount of information printed in each frame, making the backtrace
6920 more readable. Also, this command can be used to improve performance
6921 when displaying Ada frames, because the computation of large arguments
6922 can sometimes be CPU-intensive, especiallly in large applications.
6923 Setting @code{print frame-arguments} to @code{scalars} or @code{none}
6924 avoids this computation, thus speeding up the display of each Ada frame.
6926 @item show print frame-arguments
6927 Show how the value of arguments should be displayed when printing a frame.
6929 @item set print repeats
6930 @cindex repeated array elements
6931 Set the threshold for suppressing display of repeated array
6932 elements. When the number of consecutive identical elements of an
6933 array exceeds the threshold, @value{GDBN} prints the string
6934 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
6935 identical repetitions, instead of displaying the identical elements
6936 themselves. Setting the threshold to zero will cause all elements to
6937 be individually printed. The default threshold is 10.
6939 @item show print repeats
6940 Display the current threshold for printing repeated identical
6943 @item set print null-stop
6944 @cindex @sc{null} elements in arrays
6945 Cause @value{GDBN} to stop printing the characters of an array when the first
6946 @sc{null} is encountered. This is useful when large arrays actually
6947 contain only short strings.
6950 @item show print null-stop
6951 Show whether @value{GDBN} stops printing an array on the first
6952 @sc{null} character.
6954 @item set print pretty on
6955 @cindex print structures in indented form
6956 @cindex indentation in structure display
6957 Cause @value{GDBN} to print structures in an indented format with one member
6958 per line, like this:
6973 @item set print pretty off
6974 Cause @value{GDBN} to print structures in a compact format, like this:
6978 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
6979 meat = 0x54 "Pork"@}
6984 This is the default format.
6986 @item show print pretty
6987 Show which format @value{GDBN} is using to print structures.
6989 @item set print sevenbit-strings on
6990 @cindex eight-bit characters in strings
6991 @cindex octal escapes in strings
6992 Print using only seven-bit characters; if this option is set,
6993 @value{GDBN} displays any eight-bit characters (in strings or
6994 character values) using the notation @code{\}@var{nnn}. This setting is
6995 best if you are working in English (@sc{ascii}) and you use the
6996 high-order bit of characters as a marker or ``meta'' bit.
6998 @item set print sevenbit-strings off
6999 Print full eight-bit characters. This allows the use of more
7000 international character sets, and is the default.
7002 @item show print sevenbit-strings
7003 Show whether or not @value{GDBN} is printing only seven-bit characters.
7005 @item set print union on
7006 @cindex unions in structures, printing
7007 Tell @value{GDBN} to print unions which are contained in structures
7008 and other unions. This is the default setting.
7010 @item set print union off
7011 Tell @value{GDBN} not to print unions which are contained in
7012 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
7015 @item show print union
7016 Ask @value{GDBN} whether or not it will print unions which are contained in
7017 structures and other unions.
7019 For example, given the declarations
7022 typedef enum @{Tree, Bug@} Species;
7023 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
7024 typedef enum @{Caterpillar, Cocoon, Butterfly@}
7035 struct thing foo = @{Tree, @{Acorn@}@};
7039 with @code{set print union on} in effect @samp{p foo} would print
7042 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
7046 and with @code{set print union off} in effect it would print
7049 $1 = @{it = Tree, form = @{...@}@}
7053 @code{set print union} affects programs written in C-like languages
7059 These settings are of interest when debugging C@t{++} programs:
7062 @cindex demangling C@t{++} names
7063 @item set print demangle
7064 @itemx set print demangle on
7065 Print C@t{++} names in their source form rather than in the encoded
7066 (``mangled'') form passed to the assembler and linker for type-safe
7067 linkage. The default is on.
7069 @item show print demangle
7070 Show whether C@t{++} names are printed in mangled or demangled form.
7072 @item set print asm-demangle
7073 @itemx set print asm-demangle on
7074 Print C@t{++} names in their source form rather than their mangled form, even
7075 in assembler code printouts such as instruction disassemblies.
7078 @item show print asm-demangle
7079 Show whether C@t{++} names in assembly listings are printed in mangled
7082 @cindex C@t{++} symbol decoding style
7083 @cindex symbol decoding style, C@t{++}
7084 @kindex set demangle-style
7085 @item set demangle-style @var{style}
7086 Choose among several encoding schemes used by different compilers to
7087 represent C@t{++} names. The choices for @var{style} are currently:
7091 Allow @value{GDBN} to choose a decoding style by inspecting your program.
7094 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
7095 This is the default.
7098 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
7101 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
7104 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
7105 @strong{Warning:} this setting alone is not sufficient to allow
7106 debugging @code{cfront}-generated executables. @value{GDBN} would
7107 require further enhancement to permit that.
7110 If you omit @var{style}, you will see a list of possible formats.
7112 @item show demangle-style
7113 Display the encoding style currently in use for decoding C@t{++} symbols.
7115 @item set print object
7116 @itemx set print object on
7117 @cindex derived type of an object, printing
7118 @cindex display derived types
7119 When displaying a pointer to an object, identify the @emph{actual}
7120 (derived) type of the object rather than the @emph{declared} type, using
7121 the virtual function table.
7123 @item set print object off
7124 Display only the declared type of objects, without reference to the
7125 virtual function table. This is the default setting.
7127 @item show print object
7128 Show whether actual, or declared, object types are displayed.
7130 @item set print static-members
7131 @itemx set print static-members on
7132 @cindex static members of C@t{++} objects
7133 Print static members when displaying a C@t{++} object. The default is on.
7135 @item set print static-members off
7136 Do not print static members when displaying a C@t{++} object.
7138 @item show print static-members
7139 Show whether C@t{++} static members are printed or not.
7141 @item set print pascal_static-members
7142 @itemx set print pascal_static-members on
7143 @cindex static members of Pascal objects
7144 @cindex Pascal objects, static members display
7145 Print static members when displaying a Pascal object. The default is on.
7147 @item set print pascal_static-members off
7148 Do not print static members when displaying a Pascal object.
7150 @item show print pascal_static-members
7151 Show whether Pascal static members are printed or not.
7153 @c These don't work with HP ANSI C++ yet.
7154 @item set print vtbl
7155 @itemx set print vtbl on
7156 @cindex pretty print C@t{++} virtual function tables
7157 @cindex virtual functions (C@t{++}) display
7158 @cindex VTBL display
7159 Pretty print C@t{++} virtual function tables. The default is off.
7160 (The @code{vtbl} commands do not work on programs compiled with the HP
7161 ANSI C@t{++} compiler (@code{aCC}).)
7163 @item set print vtbl off
7164 Do not pretty print C@t{++} virtual function tables.
7166 @item show print vtbl
7167 Show whether C@t{++} virtual function tables are pretty printed, or not.
7171 @section Value History
7173 @cindex value history
7174 @cindex history of values printed by @value{GDBN}
7175 Values printed by the @code{print} command are saved in the @value{GDBN}
7176 @dfn{value history}. This allows you to refer to them in other expressions.
7177 Values are kept until the symbol table is re-read or discarded
7178 (for example with the @code{file} or @code{symbol-file} commands).
7179 When the symbol table changes, the value history is discarded,
7180 since the values may contain pointers back to the types defined in the
7185 @cindex history number
7186 The values printed are given @dfn{history numbers} by which you can
7187 refer to them. These are successive integers starting with one.
7188 @code{print} shows you the history number assigned to a value by
7189 printing @samp{$@var{num} = } before the value; here @var{num} is the
7192 To refer to any previous value, use @samp{$} followed by the value's
7193 history number. The way @code{print} labels its output is designed to
7194 remind you of this. Just @code{$} refers to the most recent value in
7195 the history, and @code{$$} refers to the value before that.
7196 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
7197 is the value just prior to @code{$$}, @code{$$1} is equivalent to
7198 @code{$$}, and @code{$$0} is equivalent to @code{$}.
7200 For example, suppose you have just printed a pointer to a structure and
7201 want to see the contents of the structure. It suffices to type
7207 If you have a chain of structures where the component @code{next} points
7208 to the next one, you can print the contents of the next one with this:
7215 You can print successive links in the chain by repeating this
7216 command---which you can do by just typing @key{RET}.
7218 Note that the history records values, not expressions. If the value of
7219 @code{x} is 4 and you type these commands:
7227 then the value recorded in the value history by the @code{print} command
7228 remains 4 even though the value of @code{x} has changed.
7233 Print the last ten values in the value history, with their item numbers.
7234 This is like @samp{p@ $$9} repeated ten times, except that @code{show
7235 values} does not change the history.
7237 @item show values @var{n}
7238 Print ten history values centered on history item number @var{n}.
7241 Print ten history values just after the values last printed. If no more
7242 values are available, @code{show values +} produces no display.
7245 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
7246 same effect as @samp{show values +}.
7248 @node Convenience Vars
7249 @section Convenience Variables
7251 @cindex convenience variables
7252 @cindex user-defined variables
7253 @value{GDBN} provides @dfn{convenience variables} that you can use within
7254 @value{GDBN} to hold on to a value and refer to it later. These variables
7255 exist entirely within @value{GDBN}; they are not part of your program, and
7256 setting a convenience variable has no direct effect on further execution
7257 of your program. That is why you can use them freely.
7259 Convenience variables are prefixed with @samp{$}. Any name preceded by
7260 @samp{$} can be used for a convenience variable, unless it is one of
7261 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
7262 (Value history references, in contrast, are @emph{numbers} preceded
7263 by @samp{$}. @xref{Value History, ,Value History}.)
7265 You can save a value in a convenience variable with an assignment
7266 expression, just as you would set a variable in your program.
7270 set $foo = *object_ptr
7274 would save in @code{$foo} the value contained in the object pointed to by
7277 Using a convenience variable for the first time creates it, but its
7278 value is @code{void} until you assign a new value. You can alter the
7279 value with another assignment at any time.
7281 Convenience variables have no fixed types. You can assign a convenience
7282 variable any type of value, including structures and arrays, even if
7283 that variable already has a value of a different type. The convenience
7284 variable, when used as an expression, has the type of its current value.
7287 @kindex show convenience
7288 @cindex show all user variables
7289 @item show convenience
7290 Print a list of convenience variables used so far, and their values.
7291 Abbreviated @code{show conv}.
7293 @kindex init-if-undefined
7294 @cindex convenience variables, initializing
7295 @item init-if-undefined $@var{variable} = @var{expression}
7296 Set a convenience variable if it has not already been set. This is useful
7297 for user-defined commands that keep some state. It is similar, in concept,
7298 to using local static variables with initializers in C (except that
7299 convenience variables are global). It can also be used to allow users to
7300 override default values used in a command script.
7302 If the variable is already defined then the expression is not evaluated so
7303 any side-effects do not occur.
7306 One of the ways to use a convenience variable is as a counter to be
7307 incremented or a pointer to be advanced. For example, to print
7308 a field from successive elements of an array of structures:
7312 print bar[$i++]->contents
7316 Repeat that command by typing @key{RET}.
7318 Some convenience variables are created automatically by @value{GDBN} and given
7319 values likely to be useful.
7322 @vindex $_@r{, convenience variable}
7324 The variable @code{$_} is automatically set by the @code{x} command to
7325 the last address examined (@pxref{Memory, ,Examining Memory}). Other
7326 commands which provide a default address for @code{x} to examine also
7327 set @code{$_} to that address; these commands include @code{info line}
7328 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
7329 except when set by the @code{x} command, in which case it is a pointer
7330 to the type of @code{$__}.
7332 @vindex $__@r{, convenience variable}
7334 The variable @code{$__} is automatically set by the @code{x} command
7335 to the value found in the last address examined. Its type is chosen
7336 to match the format in which the data was printed.
7339 @vindex $_exitcode@r{, convenience variable}
7340 The variable @code{$_exitcode} is automatically set to the exit code when
7341 the program being debugged terminates.
7344 On HP-UX systems, if you refer to a function or variable name that
7345 begins with a dollar sign, @value{GDBN} searches for a user or system
7346 name first, before it searches for a convenience variable.
7352 You can refer to machine register contents, in expressions, as variables
7353 with names starting with @samp{$}. The names of registers are different
7354 for each machine; use @code{info registers} to see the names used on
7358 @kindex info registers
7359 @item info registers
7360 Print the names and values of all registers except floating-point
7361 and vector registers (in the selected stack frame).
7363 @kindex info all-registers
7364 @cindex floating point registers
7365 @item info all-registers
7366 Print the names and values of all registers, including floating-point
7367 and vector registers (in the selected stack frame).
7369 @item info registers @var{regname} @dots{}
7370 Print the @dfn{relativized} value of each specified register @var{regname}.
7371 As discussed in detail below, register values are normally relative to
7372 the selected stack frame. @var{regname} may be any register name valid on
7373 the machine you are using, with or without the initial @samp{$}.
7376 @cindex stack pointer register
7377 @cindex program counter register
7378 @cindex process status register
7379 @cindex frame pointer register
7380 @cindex standard registers
7381 @value{GDBN} has four ``standard'' register names that are available (in
7382 expressions) on most machines---whenever they do not conflict with an
7383 architecture's canonical mnemonics for registers. The register names
7384 @code{$pc} and @code{$sp} are used for the program counter register and
7385 the stack pointer. @code{$fp} is used for a register that contains a
7386 pointer to the current stack frame, and @code{$ps} is used for a
7387 register that contains the processor status. For example,
7388 you could print the program counter in hex with
7395 or print the instruction to be executed next with
7402 or add four to the stack pointer@footnote{This is a way of removing
7403 one word from the stack, on machines where stacks grow downward in
7404 memory (most machines, nowadays). This assumes that the innermost
7405 stack frame is selected; setting @code{$sp} is not allowed when other
7406 stack frames are selected. To pop entire frames off the stack,
7407 regardless of machine architecture, use @code{return};
7408 see @ref{Returning, ,Returning from a Function}.} with
7414 Whenever possible, these four standard register names are available on
7415 your machine even though the machine has different canonical mnemonics,
7416 so long as there is no conflict. The @code{info registers} command
7417 shows the canonical names. For example, on the SPARC, @code{info
7418 registers} displays the processor status register as @code{$psr} but you
7419 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
7420 is an alias for the @sc{eflags} register.
7422 @value{GDBN} always considers the contents of an ordinary register as an
7423 integer when the register is examined in this way. Some machines have
7424 special registers which can hold nothing but floating point; these
7425 registers are considered to have floating point values. There is no way
7426 to refer to the contents of an ordinary register as floating point value
7427 (although you can @emph{print} it as a floating point value with
7428 @samp{print/f $@var{regname}}).
7430 Some registers have distinct ``raw'' and ``virtual'' data formats. This
7431 means that the data format in which the register contents are saved by
7432 the operating system is not the same one that your program normally
7433 sees. For example, the registers of the 68881 floating point
7434 coprocessor are always saved in ``extended'' (raw) format, but all C
7435 programs expect to work with ``double'' (virtual) format. In such
7436 cases, @value{GDBN} normally works with the virtual format only (the format
7437 that makes sense for your program), but the @code{info registers} command
7438 prints the data in both formats.
7440 @cindex SSE registers (x86)
7441 @cindex MMX registers (x86)
7442 Some machines have special registers whose contents can be interpreted
7443 in several different ways. For example, modern x86-based machines
7444 have SSE and MMX registers that can hold several values packed
7445 together in several different formats. @value{GDBN} refers to such
7446 registers in @code{struct} notation:
7449 (@value{GDBP}) print $xmm1
7451 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
7452 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
7453 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
7454 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
7455 v4_int32 = @{0, 20657912, 11, 13@},
7456 v2_int64 = @{88725056443645952, 55834574859@},
7457 uint128 = 0x0000000d0000000b013b36f800000000
7462 To set values of such registers, you need to tell @value{GDBN} which
7463 view of the register you wish to change, as if you were assigning
7464 value to a @code{struct} member:
7467 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
7470 Normally, register values are relative to the selected stack frame
7471 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
7472 value that the register would contain if all stack frames farther in
7473 were exited and their saved registers restored. In order to see the
7474 true contents of hardware registers, you must select the innermost
7475 frame (with @samp{frame 0}).
7477 However, @value{GDBN} must deduce where registers are saved, from the machine
7478 code generated by your compiler. If some registers are not saved, or if
7479 @value{GDBN} is unable to locate the saved registers, the selected stack
7480 frame makes no difference.
7482 @node Floating Point Hardware
7483 @section Floating Point Hardware
7484 @cindex floating point
7486 Depending on the configuration, @value{GDBN} may be able to give
7487 you more information about the status of the floating point hardware.
7492 Display hardware-dependent information about the floating
7493 point unit. The exact contents and layout vary depending on the
7494 floating point chip. Currently, @samp{info float} is supported on
7495 the ARM and x86 machines.
7499 @section Vector Unit
7502 Depending on the configuration, @value{GDBN} may be able to give you
7503 more information about the status of the vector unit.
7508 Display information about the vector unit. The exact contents and
7509 layout vary depending on the hardware.
7512 @node OS Information
7513 @section Operating System Auxiliary Information
7514 @cindex OS information
7516 @value{GDBN} provides interfaces to useful OS facilities that can help
7517 you debug your program.
7519 @cindex @code{ptrace} system call
7520 @cindex @code{struct user} contents
7521 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
7522 machines), it interfaces with the inferior via the @code{ptrace}
7523 system call. The operating system creates a special sata structure,
7524 called @code{struct user}, for this interface. You can use the
7525 command @code{info udot} to display the contents of this data
7531 Display the contents of the @code{struct user} maintained by the OS
7532 kernel for the program being debugged. @value{GDBN} displays the
7533 contents of @code{struct user} as a list of hex numbers, similar to
7534 the @code{examine} command.
7537 @cindex auxiliary vector
7538 @cindex vector, auxiliary
7539 Some operating systems supply an @dfn{auxiliary vector} to programs at
7540 startup. This is akin to the arguments and environment that you
7541 specify for a program, but contains a system-dependent variety of
7542 binary values that tell system libraries important details about the
7543 hardware, operating system, and process. Each value's purpose is
7544 identified by an integer tag; the meanings are well-known but system-specific.
7545 Depending on the configuration and operating system facilities,
7546 @value{GDBN} may be able to show you this information. For remote
7547 targets, this functionality may further depend on the remote stub's
7548 support of the @samp{qXfer:auxv:read} packet, see
7549 @ref{qXfer auxiliary vector read}.
7554 Display the auxiliary vector of the inferior, which can be either a
7555 live process or a core dump file. @value{GDBN} prints each tag value
7556 numerically, and also shows names and text descriptions for recognized
7557 tags. Some values in the vector are numbers, some bit masks, and some
7558 pointers to strings or other data. @value{GDBN} displays each value in the
7559 most appropriate form for a recognized tag, and in hexadecimal for
7560 an unrecognized tag.
7563 On some targets, @value{GDBN} can access operating-system-specific information
7564 and display it to user, without interpretation. For remote targets,
7565 this functionality depends on the remote stub's support of the
7566 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
7569 @kindex info os processes
7570 @item info os processes
7571 Display the list of processes on the target. For each process,
7572 @value{GDBN} prints the process identifier, the name of the user, and
7573 the command corresponding to the process.
7576 @node Memory Region Attributes
7577 @section Memory Region Attributes
7578 @cindex memory region attributes
7580 @dfn{Memory region attributes} allow you to describe special handling
7581 required by regions of your target's memory. @value{GDBN} uses
7582 attributes to determine whether to allow certain types of memory
7583 accesses; whether to use specific width accesses; and whether to cache
7584 target memory. By default the description of memory regions is
7585 fetched from the target (if the current target supports this), but the
7586 user can override the fetched regions.
7588 Defined memory regions can be individually enabled and disabled. When a
7589 memory region is disabled, @value{GDBN} uses the default attributes when
7590 accessing memory in that region. Similarly, if no memory regions have
7591 been defined, @value{GDBN} uses the default attributes when accessing
7594 When a memory region is defined, it is given a number to identify it;
7595 to enable, disable, or remove a memory region, you specify that number.
7599 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
7600 Define a memory region bounded by @var{lower} and @var{upper} with
7601 attributes @var{attributes}@dots{}, and add it to the list of regions
7602 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
7603 case: it is treated as the target's maximum memory address.
7604 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
7607 Discard any user changes to the memory regions and use target-supplied
7608 regions, if available, or no regions if the target does not support.
7611 @item delete mem @var{nums}@dots{}
7612 Remove memory regions @var{nums}@dots{} from the list of regions
7613 monitored by @value{GDBN}.
7616 @item disable mem @var{nums}@dots{}
7617 Disable monitoring of memory regions @var{nums}@dots{}.
7618 A disabled memory region is not forgotten.
7619 It may be enabled again later.
7622 @item enable mem @var{nums}@dots{}
7623 Enable monitoring of memory regions @var{nums}@dots{}.
7627 Print a table of all defined memory regions, with the following columns
7631 @item Memory Region Number
7632 @item Enabled or Disabled.
7633 Enabled memory regions are marked with @samp{y}.
7634 Disabled memory regions are marked with @samp{n}.
7637 The address defining the inclusive lower bound of the memory region.
7640 The address defining the exclusive upper bound of the memory region.
7643 The list of attributes set for this memory region.
7648 @subsection Attributes
7650 @subsubsection Memory Access Mode
7651 The access mode attributes set whether @value{GDBN} may make read or
7652 write accesses to a memory region.
7654 While these attributes prevent @value{GDBN} from performing invalid
7655 memory accesses, they do nothing to prevent the target system, I/O DMA,
7656 etc.@: from accessing memory.
7660 Memory is read only.
7662 Memory is write only.
7664 Memory is read/write. This is the default.
7667 @subsubsection Memory Access Size
7668 The access size attribute tells @value{GDBN} to use specific sized
7669 accesses in the memory region. Often memory mapped device registers
7670 require specific sized accesses. If no access size attribute is
7671 specified, @value{GDBN} may use accesses of any size.
7675 Use 8 bit memory accesses.
7677 Use 16 bit memory accesses.
7679 Use 32 bit memory accesses.
7681 Use 64 bit memory accesses.
7684 @c @subsubsection Hardware/Software Breakpoints
7685 @c The hardware/software breakpoint attributes set whether @value{GDBN}
7686 @c will use hardware or software breakpoints for the internal breakpoints
7687 @c used by the step, next, finish, until, etc. commands.
7691 @c Always use hardware breakpoints
7692 @c @item swbreak (default)
7695 @subsubsection Data Cache
7696 The data cache attributes set whether @value{GDBN} will cache target
7697 memory. While this generally improves performance by reducing debug
7698 protocol overhead, it can lead to incorrect results because @value{GDBN}
7699 does not know about volatile variables or memory mapped device
7704 Enable @value{GDBN} to cache target memory.
7706 Disable @value{GDBN} from caching target memory. This is the default.
7709 @subsection Memory Access Checking
7710 @value{GDBN} can be instructed to refuse accesses to memory that is
7711 not explicitly described. This can be useful if accessing such
7712 regions has undesired effects for a specific target, or to provide
7713 better error checking. The following commands control this behaviour.
7716 @kindex set mem inaccessible-by-default
7717 @item set mem inaccessible-by-default [on|off]
7718 If @code{on} is specified, make @value{GDBN} treat memory not
7719 explicitly described by the memory ranges as non-existent and refuse accesses
7720 to such memory. The checks are only performed if there's at least one
7721 memory range defined. If @code{off} is specified, make @value{GDBN}
7722 treat the memory not explicitly described by the memory ranges as RAM.
7723 The default value is @code{on}.
7724 @kindex show mem inaccessible-by-default
7725 @item show mem inaccessible-by-default
7726 Show the current handling of accesses to unknown memory.
7730 @c @subsubsection Memory Write Verification
7731 @c The memory write verification attributes set whether @value{GDBN}
7732 @c will re-reads data after each write to verify the write was successful.
7736 @c @item noverify (default)
7739 @node Dump/Restore Files
7740 @section Copy Between Memory and a File
7741 @cindex dump/restore files
7742 @cindex append data to a file
7743 @cindex dump data to a file
7744 @cindex restore data from a file
7746 You can use the commands @code{dump}, @code{append}, and
7747 @code{restore} to copy data between target memory and a file. The
7748 @code{dump} and @code{append} commands write data to a file, and the
7749 @code{restore} command reads data from a file back into the inferior's
7750 memory. Files may be in binary, Motorola S-record, Intel hex, or
7751 Tektronix Hex format; however, @value{GDBN} can only append to binary
7757 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
7758 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
7759 Dump the contents of memory from @var{start_addr} to @var{end_addr},
7760 or the value of @var{expr}, to @var{filename} in the given format.
7762 The @var{format} parameter may be any one of:
7769 Motorola S-record format.
7771 Tektronix Hex format.
7774 @value{GDBN} uses the same definitions of these formats as the
7775 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
7776 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
7780 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
7781 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
7782 Append the contents of memory from @var{start_addr} to @var{end_addr},
7783 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
7784 (@value{GDBN} can only append data to files in raw binary form.)
7787 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
7788 Restore the contents of file @var{filename} into memory. The
7789 @code{restore} command can automatically recognize any known @sc{bfd}
7790 file format, except for raw binary. To restore a raw binary file you
7791 must specify the optional keyword @code{binary} after the filename.
7793 If @var{bias} is non-zero, its value will be added to the addresses
7794 contained in the file. Binary files always start at address zero, so
7795 they will be restored at address @var{bias}. Other bfd files have
7796 a built-in location; they will be restored at offset @var{bias}
7799 If @var{start} and/or @var{end} are non-zero, then only data between
7800 file offset @var{start} and file offset @var{end} will be restored.
7801 These offsets are relative to the addresses in the file, before
7802 the @var{bias} argument is applied.
7806 @node Core File Generation
7807 @section How to Produce a Core File from Your Program
7808 @cindex dump core from inferior
7810 A @dfn{core file} or @dfn{core dump} is a file that records the memory
7811 image of a running process and its process status (register values
7812 etc.). Its primary use is post-mortem debugging of a program that
7813 crashed while it ran outside a debugger. A program that crashes
7814 automatically produces a core file, unless this feature is disabled by
7815 the user. @xref{Files}, for information on invoking @value{GDBN} in
7816 the post-mortem debugging mode.
7818 Occasionally, you may wish to produce a core file of the program you
7819 are debugging in order to preserve a snapshot of its state.
7820 @value{GDBN} has a special command for that.
7824 @kindex generate-core-file
7825 @item generate-core-file [@var{file}]
7826 @itemx gcore [@var{file}]
7827 Produce a core dump of the inferior process. The optional argument
7828 @var{file} specifies the file name where to put the core dump. If not
7829 specified, the file name defaults to @file{core.@var{pid}}, where
7830 @var{pid} is the inferior process ID.
7832 Note that this command is implemented only for some systems (as of
7833 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
7836 @node Character Sets
7837 @section Character Sets
7838 @cindex character sets
7840 @cindex translating between character sets
7841 @cindex host character set
7842 @cindex target character set
7844 If the program you are debugging uses a different character set to
7845 represent characters and strings than the one @value{GDBN} uses itself,
7846 @value{GDBN} can automatically translate between the character sets for
7847 you. The character set @value{GDBN} uses we call the @dfn{host
7848 character set}; the one the inferior program uses we call the
7849 @dfn{target character set}.
7851 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
7852 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
7853 remote protocol (@pxref{Remote Debugging}) to debug a program
7854 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
7855 then the host character set is Latin-1, and the target character set is
7856 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
7857 target-charset EBCDIC-US}, then @value{GDBN} translates between
7858 @sc{ebcdic} and Latin 1 as you print character or string values, or use
7859 character and string literals in expressions.
7861 @value{GDBN} has no way to automatically recognize which character set
7862 the inferior program uses; you must tell it, using the @code{set
7863 target-charset} command, described below.
7865 Here are the commands for controlling @value{GDBN}'s character set
7869 @item set target-charset @var{charset}
7870 @kindex set target-charset
7871 Set the current target character set to @var{charset}. We list the
7872 character set names @value{GDBN} recognizes below, but if you type
7873 @code{set target-charset} followed by @key{TAB}@key{TAB}, @value{GDBN} will
7874 list the target character sets it supports.
7878 @item set host-charset @var{charset}
7879 @kindex set host-charset
7880 Set the current host character set to @var{charset}.
7882 By default, @value{GDBN} uses a host character set appropriate to the
7883 system it is running on; you can override that default using the
7884 @code{set host-charset} command.
7886 @value{GDBN} can only use certain character sets as its host character
7887 set. We list the character set names @value{GDBN} recognizes below, and
7888 indicate which can be host character sets, but if you type
7889 @code{set target-charset} followed by @key{TAB}@key{TAB}, @value{GDBN} will
7890 list the host character sets it supports.
7892 @item set charset @var{charset}
7894 Set the current host and target character sets to @var{charset}. As
7895 above, if you type @code{set charset} followed by @key{TAB}@key{TAB},
7896 @value{GDBN} will list the name of the character sets that can be used
7897 for both host and target.
7901 @kindex show charset
7902 Show the names of the current host and target charsets.
7904 @itemx show host-charset
7905 @kindex show host-charset
7906 Show the name of the current host charset.
7908 @itemx show target-charset
7909 @kindex show target-charset
7910 Show the name of the current target charset.
7914 @value{GDBN} currently includes support for the following character
7920 @cindex ASCII character set
7921 Seven-bit U.S. @sc{ascii}. @value{GDBN} can use this as its host
7925 @cindex ISO 8859-1 character set
7926 @cindex ISO Latin 1 character set
7927 The ISO Latin 1 character set. This extends @sc{ascii} with accented
7928 characters needed for French, German, and Spanish. @value{GDBN} can use
7929 this as its host character set.
7933 @cindex EBCDIC character set
7934 @cindex IBM1047 character set
7935 Variants of the @sc{ebcdic} character set, used on some of IBM's
7936 mainframe operating systems. (@sc{gnu}/Linux on the S/390 uses U.S. @sc{ascii}.)
7937 @value{GDBN} cannot use these as its host character set.
7941 Note that these are all single-byte character sets. More work inside
7942 @value{GDBN} is needed to support multi-byte or variable-width character
7943 encodings, like the UTF-8 and UCS-2 encodings of Unicode.
7945 Here is an example of @value{GDBN}'s character set support in action.
7946 Assume that the following source code has been placed in the file
7947 @file{charset-test.c}:
7953 = @{72, 101, 108, 108, 111, 44, 32, 119,
7954 111, 114, 108, 100, 33, 10, 0@};
7955 char ibm1047_hello[]
7956 = @{200, 133, 147, 147, 150, 107, 64, 166,
7957 150, 153, 147, 132, 90, 37, 0@};
7961 printf ("Hello, world!\n");
7965 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
7966 containing the string @samp{Hello, world!} followed by a newline,
7967 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
7969 We compile the program, and invoke the debugger on it:
7972 $ gcc -g charset-test.c -o charset-test
7973 $ gdb -nw charset-test
7974 GNU gdb 2001-12-19-cvs
7975 Copyright 2001 Free Software Foundation, Inc.
7980 We can use the @code{show charset} command to see what character sets
7981 @value{GDBN} is currently using to interpret and display characters and
7985 (@value{GDBP}) show charset
7986 The current host and target character set is `ISO-8859-1'.
7990 For the sake of printing this manual, let's use @sc{ascii} as our
7991 initial character set:
7993 (@value{GDBP}) set charset ASCII
7994 (@value{GDBP}) show charset
7995 The current host and target character set is `ASCII'.
7999 Let's assume that @sc{ascii} is indeed the correct character set for our
8000 host system --- in other words, let's assume that if @value{GDBN} prints
8001 characters using the @sc{ascii} character set, our terminal will display
8002 them properly. Since our current target character set is also
8003 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
8006 (@value{GDBP}) print ascii_hello
8007 $1 = 0x401698 "Hello, world!\n"
8008 (@value{GDBP}) print ascii_hello[0]
8013 @value{GDBN} uses the target character set for character and string
8014 literals you use in expressions:
8017 (@value{GDBP}) print '+'
8022 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
8025 @value{GDBN} relies on the user to tell it which character set the
8026 target program uses. If we print @code{ibm1047_hello} while our target
8027 character set is still @sc{ascii}, we get jibberish:
8030 (@value{GDBP}) print ibm1047_hello
8031 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
8032 (@value{GDBP}) print ibm1047_hello[0]
8037 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
8038 @value{GDBN} tells us the character sets it supports:
8041 (@value{GDBP}) set target-charset
8042 ASCII EBCDIC-US IBM1047 ISO-8859-1
8043 (@value{GDBP}) set target-charset
8046 We can select @sc{ibm1047} as our target character set, and examine the
8047 program's strings again. Now the @sc{ascii} string is wrong, but
8048 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
8049 target character set, @sc{ibm1047}, to the host character set,
8050 @sc{ascii}, and they display correctly:
8053 (@value{GDBP}) set target-charset IBM1047
8054 (@value{GDBP}) show charset
8055 The current host character set is `ASCII'.
8056 The current target character set is `IBM1047'.
8057 (@value{GDBP}) print ascii_hello
8058 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
8059 (@value{GDBP}) print ascii_hello[0]
8061 (@value{GDBP}) print ibm1047_hello
8062 $8 = 0x4016a8 "Hello, world!\n"
8063 (@value{GDBP}) print ibm1047_hello[0]
8068 As above, @value{GDBN} uses the target character set for character and
8069 string literals you use in expressions:
8072 (@value{GDBP}) print '+'
8077 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
8080 @node Caching Remote Data
8081 @section Caching Data of Remote Targets
8082 @cindex caching data of remote targets
8084 @value{GDBN} can cache data exchanged between the debugger and a
8085 remote target (@pxref{Remote Debugging}). Such caching generally improves
8086 performance, because it reduces the overhead of the remote protocol by
8087 bundling memory reads and writes into large chunks. Unfortunately,
8088 @value{GDBN} does not currently know anything about volatile
8089 registers, and thus data caching will produce incorrect results when
8090 volatile registers are in use.
8093 @kindex set remotecache
8094 @item set remotecache on
8095 @itemx set remotecache off
8096 Set caching state for remote targets. When @code{ON}, use data
8097 caching. By default, this option is @code{OFF}.
8099 @kindex show remotecache
8100 @item show remotecache
8101 Show the current state of data caching for remote targets.
8105 Print the information about the data cache performance. The
8106 information displayed includes: the dcache width and depth; and for
8107 each cache line, how many times it was referenced, and its data and
8108 state (invalid, dirty, valid). This command is useful for debugging
8109 the data cache operation.
8112 @node Searching Memory
8113 @section Search Memory
8114 @cindex searching memory
8116 Memory can be searched for a particular sequence of bytes with the
8117 @code{find} command.
8121 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8122 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8123 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
8124 etc. The search begins at address @var{start_addr} and continues for either
8125 @var{len} bytes or through to @var{end_addr} inclusive.
8128 @var{s} and @var{n} are optional parameters.
8129 They may be specified in either order, apart or together.
8132 @item @var{s}, search query size
8133 The size of each search query value.
8139 halfwords (two bytes)
8143 giant words (eight bytes)
8146 All values are interpreted in the current language.
8147 This means, for example, that if the current source language is C/C@t{++}
8148 then searching for the string ``hello'' includes the trailing '\0'.
8150 If the value size is not specified, it is taken from the
8151 value's type in the current language.
8152 This is useful when one wants to specify the search
8153 pattern as a mixture of types.
8154 Note that this means, for example, that in the case of C-like languages
8155 a search for an untyped 0x42 will search for @samp{(int) 0x42}
8156 which is typically four bytes.
8158 @item @var{n}, maximum number of finds
8159 The maximum number of matches to print. The default is to print all finds.
8162 You can use strings as search values. Quote them with double-quotes
8164 The string value is copied into the search pattern byte by byte,
8165 regardless of the endianness of the target and the size specification.
8167 The address of each match found is printed as well as a count of the
8168 number of matches found.
8170 The address of the last value found is stored in convenience variable
8172 A count of the number of matches is stored in @samp{$numfound}.
8174 For example, if stopped at the @code{printf} in this function:
8180 static char hello[] = "hello-hello";
8181 static struct @{ char c; short s; int i; @}
8182 __attribute__ ((packed)) mixed
8183 = @{ 'c', 0x1234, 0x87654321 @};
8184 printf ("%s\n", hello);
8189 you get during debugging:
8192 (gdb) find &hello[0], +sizeof(hello), "hello"
8193 0x804956d <hello.1620+6>
8195 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
8196 0x8049567 <hello.1620>
8197 0x804956d <hello.1620+6>
8199 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
8200 0x8049567 <hello.1620>
8202 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
8203 0x8049560 <mixed.1625>
8205 (gdb) print $numfound
8208 $2 = (void *) 0x8049560
8212 @chapter C Preprocessor Macros
8214 Some languages, such as C and C@t{++}, provide a way to define and invoke
8215 ``preprocessor macros'' which expand into strings of tokens.
8216 @value{GDBN} can evaluate expressions containing macro invocations, show
8217 the result of macro expansion, and show a macro's definition, including
8218 where it was defined.
8220 You may need to compile your program specially to provide @value{GDBN}
8221 with information about preprocessor macros. Most compilers do not
8222 include macros in their debugging information, even when you compile
8223 with the @option{-g} flag. @xref{Compilation}.
8225 A program may define a macro at one point, remove that definition later,
8226 and then provide a different definition after that. Thus, at different
8227 points in the program, a macro may have different definitions, or have
8228 no definition at all. If there is a current stack frame, @value{GDBN}
8229 uses the macros in scope at that frame's source code line. Otherwise,
8230 @value{GDBN} uses the macros in scope at the current listing location;
8233 Whenever @value{GDBN} evaluates an expression, it always expands any
8234 macro invocations present in the expression. @value{GDBN} also provides
8235 the following commands for working with macros explicitly.
8239 @kindex macro expand
8240 @cindex macro expansion, showing the results of preprocessor
8241 @cindex preprocessor macro expansion, showing the results of
8242 @cindex expanding preprocessor macros
8243 @item macro expand @var{expression}
8244 @itemx macro exp @var{expression}
8245 Show the results of expanding all preprocessor macro invocations in
8246 @var{expression}. Since @value{GDBN} simply expands macros, but does
8247 not parse the result, @var{expression} need not be a valid expression;
8248 it can be any string of tokens.
8251 @item macro expand-once @var{expression}
8252 @itemx macro exp1 @var{expression}
8253 @cindex expand macro once
8254 @i{(This command is not yet implemented.)} Show the results of
8255 expanding those preprocessor macro invocations that appear explicitly in
8256 @var{expression}. Macro invocations appearing in that expansion are
8257 left unchanged. This command allows you to see the effect of a
8258 particular macro more clearly, without being confused by further
8259 expansions. Since @value{GDBN} simply expands macros, but does not
8260 parse the result, @var{expression} need not be a valid expression; it
8261 can be any string of tokens.
8264 @cindex macro definition, showing
8265 @cindex definition, showing a macro's
8266 @item info macro @var{macro}
8267 Show the definition of the macro named @var{macro}, and describe the
8268 source location where that definition was established.
8270 @kindex macro define
8271 @cindex user-defined macros
8272 @cindex defining macros interactively
8273 @cindex macros, user-defined
8274 @item macro define @var{macro} @var{replacement-list}
8275 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
8276 Introduce a definition for a preprocessor macro named @var{macro},
8277 invocations of which are replaced by the tokens given in
8278 @var{replacement-list}. The first form of this command defines an
8279 ``object-like'' macro, which takes no arguments; the second form
8280 defines a ``function-like'' macro, which takes the arguments given in
8283 A definition introduced by this command is in scope in every
8284 expression evaluated in @value{GDBN}, until it is removed with the
8285 @code{macro undef} command, described below. The definition overrides
8286 all definitions for @var{macro} present in the program being debugged,
8287 as well as any previous user-supplied definition.
8290 @item macro undef @var{macro}
8291 Remove any user-supplied definition for the macro named @var{macro}.
8292 This command only affects definitions provided with the @code{macro
8293 define} command, described above; it cannot remove definitions present
8294 in the program being debugged.
8298 List all the macros defined using the @code{macro define} command.
8301 @cindex macros, example of debugging with
8302 Here is a transcript showing the above commands in action. First, we
8303 show our source files:
8311 #define ADD(x) (M + x)
8316 printf ("Hello, world!\n");
8318 printf ("We're so creative.\n");
8320 printf ("Goodbye, world!\n");
8327 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
8328 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
8329 compiler includes information about preprocessor macros in the debugging
8333 $ gcc -gdwarf-2 -g3 sample.c -o sample
8337 Now, we start @value{GDBN} on our sample program:
8341 GNU gdb 2002-05-06-cvs
8342 Copyright 2002 Free Software Foundation, Inc.
8343 GDB is free software, @dots{}
8347 We can expand macros and examine their definitions, even when the
8348 program is not running. @value{GDBN} uses the current listing position
8349 to decide which macro definitions are in scope:
8352 (@value{GDBP}) list main
8355 5 #define ADD(x) (M + x)
8360 10 printf ("Hello, world!\n");
8362 12 printf ("We're so creative.\n");
8363 (@value{GDBP}) info macro ADD
8364 Defined at /home/jimb/gdb/macros/play/sample.c:5
8365 #define ADD(x) (M + x)
8366 (@value{GDBP}) info macro Q
8367 Defined at /home/jimb/gdb/macros/play/sample.h:1
8368 included at /home/jimb/gdb/macros/play/sample.c:2
8370 (@value{GDBP}) macro expand ADD(1)
8371 expands to: (42 + 1)
8372 (@value{GDBP}) macro expand-once ADD(1)
8373 expands to: once (M + 1)
8377 In the example above, note that @code{macro expand-once} expands only
8378 the macro invocation explicit in the original text --- the invocation of
8379 @code{ADD} --- but does not expand the invocation of the macro @code{M},
8380 which was introduced by @code{ADD}.
8382 Once the program is running, @value{GDBN} uses the macro definitions in
8383 force at the source line of the current stack frame:
8386 (@value{GDBP}) break main
8387 Breakpoint 1 at 0x8048370: file sample.c, line 10.
8389 Starting program: /home/jimb/gdb/macros/play/sample
8391 Breakpoint 1, main () at sample.c:10
8392 10 printf ("Hello, world!\n");
8396 At line 10, the definition of the macro @code{N} at line 9 is in force:
8399 (@value{GDBP}) info macro N
8400 Defined at /home/jimb/gdb/macros/play/sample.c:9
8402 (@value{GDBP}) macro expand N Q M
8404 (@value{GDBP}) print N Q M
8409 As we step over directives that remove @code{N}'s definition, and then
8410 give it a new definition, @value{GDBN} finds the definition (or lack
8411 thereof) in force at each point:
8416 12 printf ("We're so creative.\n");
8417 (@value{GDBP}) info macro N
8418 The symbol `N' has no definition as a C/C++ preprocessor macro
8419 at /home/jimb/gdb/macros/play/sample.c:12
8422 14 printf ("Goodbye, world!\n");
8423 (@value{GDBP}) info macro N
8424 Defined at /home/jimb/gdb/macros/play/sample.c:13
8426 (@value{GDBP}) macro expand N Q M
8427 expands to: 1729 < 42
8428 (@value{GDBP}) print N Q M
8435 @chapter Tracepoints
8436 @c This chapter is based on the documentation written by Michael
8437 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
8440 In some applications, it is not feasible for the debugger to interrupt
8441 the program's execution long enough for the developer to learn
8442 anything helpful about its behavior. If the program's correctness
8443 depends on its real-time behavior, delays introduced by a debugger
8444 might cause the program to change its behavior drastically, or perhaps
8445 fail, even when the code itself is correct. It is useful to be able
8446 to observe the program's behavior without interrupting it.
8448 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
8449 specify locations in the program, called @dfn{tracepoints}, and
8450 arbitrary expressions to evaluate when those tracepoints are reached.
8451 Later, using the @code{tfind} command, you can examine the values
8452 those expressions had when the program hit the tracepoints. The
8453 expressions may also denote objects in memory---structures or arrays,
8454 for example---whose values @value{GDBN} should record; while visiting
8455 a particular tracepoint, you may inspect those objects as if they were
8456 in memory at that moment. However, because @value{GDBN} records these
8457 values without interacting with you, it can do so quickly and
8458 unobtrusively, hopefully not disturbing the program's behavior.
8460 The tracepoint facility is currently available only for remote
8461 targets. @xref{Targets}. In addition, your remote target must know
8462 how to collect trace data. This functionality is implemented in the
8463 remote stub; however, none of the stubs distributed with @value{GDBN}
8464 support tracepoints as of this writing. The format of the remote
8465 packets used to implement tracepoints are described in @ref{Tracepoint
8468 This chapter describes the tracepoint commands and features.
8472 * Analyze Collected Data::
8473 * Tracepoint Variables::
8476 @node Set Tracepoints
8477 @section Commands to Set Tracepoints
8479 Before running such a @dfn{trace experiment}, an arbitrary number of
8480 tracepoints can be set. Like a breakpoint (@pxref{Set Breaks}), a
8481 tracepoint has a number assigned to it by @value{GDBN}. Like with
8482 breakpoints, tracepoint numbers are successive integers starting from
8483 one. Many of the commands associated with tracepoints take the
8484 tracepoint number as their argument, to identify which tracepoint to
8487 For each tracepoint, you can specify, in advance, some arbitrary set
8488 of data that you want the target to collect in the trace buffer when
8489 it hits that tracepoint. The collected data can include registers,
8490 local variables, or global data. Later, you can use @value{GDBN}
8491 commands to examine the values these data had at the time the
8494 This section describes commands to set tracepoints and associated
8495 conditions and actions.
8498 * Create and Delete Tracepoints::
8499 * Enable and Disable Tracepoints::
8500 * Tracepoint Passcounts::
8501 * Tracepoint Actions::
8502 * Listing Tracepoints::
8503 * Starting and Stopping Trace Experiments::
8506 @node Create and Delete Tracepoints
8507 @subsection Create and Delete Tracepoints
8510 @cindex set tracepoint
8513 The @code{trace} command is very similar to the @code{break} command.
8514 Its argument can be a source line, a function name, or an address in
8515 the target program. @xref{Set Breaks}. The @code{trace} command
8516 defines a tracepoint, which is a point in the target program where the
8517 debugger will briefly stop, collect some data, and then allow the
8518 program to continue. Setting a tracepoint or changing its commands
8519 doesn't take effect until the next @code{tstart} command; thus, you
8520 cannot change the tracepoint attributes once a trace experiment is
8523 Here are some examples of using the @code{trace} command:
8526 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
8528 (@value{GDBP}) @b{trace +2} // 2 lines forward
8530 (@value{GDBP}) @b{trace my_function} // first source line of function
8532 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
8534 (@value{GDBP}) @b{trace *0x2117c4} // an address
8538 You can abbreviate @code{trace} as @code{tr}.
8541 @cindex last tracepoint number
8542 @cindex recent tracepoint number
8543 @cindex tracepoint number
8544 The convenience variable @code{$tpnum} records the tracepoint number
8545 of the most recently set tracepoint.
8547 @kindex delete tracepoint
8548 @cindex tracepoint deletion
8549 @item delete tracepoint @r{[}@var{num}@r{]}
8550 Permanently delete one or more tracepoints. With no argument, the
8551 default is to delete all tracepoints.
8556 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
8558 (@value{GDBP}) @b{delete trace} // remove all tracepoints
8562 You can abbreviate this command as @code{del tr}.
8565 @node Enable and Disable Tracepoints
8566 @subsection Enable and Disable Tracepoints
8569 @kindex disable tracepoint
8570 @item disable tracepoint @r{[}@var{num}@r{]}
8571 Disable tracepoint @var{num}, or all tracepoints if no argument
8572 @var{num} is given. A disabled tracepoint will have no effect during
8573 the next trace experiment, but it is not forgotten. You can re-enable
8574 a disabled tracepoint using the @code{enable tracepoint} command.
8576 @kindex enable tracepoint
8577 @item enable tracepoint @r{[}@var{num}@r{]}
8578 Enable tracepoint @var{num}, or all tracepoints. The enabled
8579 tracepoints will become effective the next time a trace experiment is
8583 @node Tracepoint Passcounts
8584 @subsection Tracepoint Passcounts
8588 @cindex tracepoint pass count
8589 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
8590 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
8591 automatically stop a trace experiment. If a tracepoint's passcount is
8592 @var{n}, then the trace experiment will be automatically stopped on
8593 the @var{n}'th time that tracepoint is hit. If the tracepoint number
8594 @var{num} is not specified, the @code{passcount} command sets the
8595 passcount of the most recently defined tracepoint. If no passcount is
8596 given, the trace experiment will run until stopped explicitly by the
8602 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
8603 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
8605 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
8606 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
8607 (@value{GDBP}) @b{trace foo}
8608 (@value{GDBP}) @b{pass 3}
8609 (@value{GDBP}) @b{trace bar}
8610 (@value{GDBP}) @b{pass 2}
8611 (@value{GDBP}) @b{trace baz}
8612 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
8613 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
8614 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
8615 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
8619 @node Tracepoint Actions
8620 @subsection Tracepoint Action Lists
8624 @cindex tracepoint actions
8625 @item actions @r{[}@var{num}@r{]}
8626 This command will prompt for a list of actions to be taken when the
8627 tracepoint is hit. If the tracepoint number @var{num} is not
8628 specified, this command sets the actions for the one that was most
8629 recently defined (so that you can define a tracepoint and then say
8630 @code{actions} without bothering about its number). You specify the
8631 actions themselves on the following lines, one action at a time, and
8632 terminate the actions list with a line containing just @code{end}. So
8633 far, the only defined actions are @code{collect} and
8634 @code{while-stepping}.
8636 @cindex remove actions from a tracepoint
8637 To remove all actions from a tracepoint, type @samp{actions @var{num}}
8638 and follow it immediately with @samp{end}.
8641 (@value{GDBP}) @b{collect @var{data}} // collect some data
8643 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
8645 (@value{GDBP}) @b{end} // signals the end of actions.
8648 In the following example, the action list begins with @code{collect}
8649 commands indicating the things to be collected when the tracepoint is
8650 hit. Then, in order to single-step and collect additional data
8651 following the tracepoint, a @code{while-stepping} command is used,
8652 followed by the list of things to be collected while stepping. The
8653 @code{while-stepping} command is terminated by its own separate
8654 @code{end} command. Lastly, the action list is terminated by an
8658 (@value{GDBP}) @b{trace foo}
8659 (@value{GDBP}) @b{actions}
8660 Enter actions for tracepoint 1, one per line:
8669 @kindex collect @r{(tracepoints)}
8670 @item collect @var{expr1}, @var{expr2}, @dots{}
8671 Collect values of the given expressions when the tracepoint is hit.
8672 This command accepts a comma-separated list of any valid expressions.
8673 In addition to global, static, or local variables, the following
8674 special arguments are supported:
8678 collect all registers
8681 collect all function arguments
8684 collect all local variables.
8687 You can give several consecutive @code{collect} commands, each one
8688 with a single argument, or one @code{collect} command with several
8689 arguments separated by commas: the effect is the same.
8691 The command @code{info scope} (@pxref{Symbols, info scope}) is
8692 particularly useful for figuring out what data to collect.
8694 @kindex while-stepping @r{(tracepoints)}
8695 @item while-stepping @var{n}
8696 Perform @var{n} single-step traces after the tracepoint, collecting
8697 new data at each step. The @code{while-stepping} command is
8698 followed by the list of what to collect while stepping (followed by
8699 its own @code{end} command):
8703 > collect $regs, myglobal
8709 You may abbreviate @code{while-stepping} as @code{ws} or
8713 @node Listing Tracepoints
8714 @subsection Listing Tracepoints
8717 @kindex info tracepoints
8719 @cindex information about tracepoints
8720 @item info tracepoints @r{[}@var{num}@r{]}
8721 Display information about the tracepoint @var{num}. If you don't specify
8722 a tracepoint number, displays information about all the tracepoints
8723 defined so far. For each tracepoint, the following information is
8730 whether it is enabled or disabled
8734 its passcount as given by the @code{passcount @var{n}} command
8736 its step count as given by the @code{while-stepping @var{n}} command
8738 where in the source files is the tracepoint set
8740 its action list as given by the @code{actions} command
8744 (@value{GDBP}) @b{info trace}
8745 Num Enb Address PassC StepC What
8746 1 y 0x002117c4 0 0 <gdb_asm>
8747 2 y 0x0020dc64 0 0 in g_test at g_test.c:1375
8748 3 y 0x0020b1f4 0 0 in get_data at ../foo.c:41
8753 This command can be abbreviated @code{info tp}.
8756 @node Starting and Stopping Trace Experiments
8757 @subsection Starting and Stopping Trace Experiments
8761 @cindex start a new trace experiment
8762 @cindex collected data discarded
8764 This command takes no arguments. It starts the trace experiment, and
8765 begins collecting data. This has the side effect of discarding all
8766 the data collected in the trace buffer during the previous trace
8770 @cindex stop a running trace experiment
8772 This command takes no arguments. It ends the trace experiment, and
8773 stops collecting data.
8775 @strong{Note}: a trace experiment and data collection may stop
8776 automatically if any tracepoint's passcount is reached
8777 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
8780 @cindex status of trace data collection
8781 @cindex trace experiment, status of
8783 This command displays the status of the current trace data
8787 Here is an example of the commands we described so far:
8790 (@value{GDBP}) @b{trace gdb_c_test}
8791 (@value{GDBP}) @b{actions}
8792 Enter actions for tracepoint #1, one per line.
8793 > collect $regs,$locals,$args
8798 (@value{GDBP}) @b{tstart}
8799 [time passes @dots{}]
8800 (@value{GDBP}) @b{tstop}
8804 @node Analyze Collected Data
8805 @section Using the Collected Data
8807 After the tracepoint experiment ends, you use @value{GDBN} commands
8808 for examining the trace data. The basic idea is that each tracepoint
8809 collects a trace @dfn{snapshot} every time it is hit and another
8810 snapshot every time it single-steps. All these snapshots are
8811 consecutively numbered from zero and go into a buffer, and you can
8812 examine them later. The way you examine them is to @dfn{focus} on a
8813 specific trace snapshot. When the remote stub is focused on a trace
8814 snapshot, it will respond to all @value{GDBN} requests for memory and
8815 registers by reading from the buffer which belongs to that snapshot,
8816 rather than from @emph{real} memory or registers of the program being
8817 debugged. This means that @strong{all} @value{GDBN} commands
8818 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
8819 behave as if we were currently debugging the program state as it was
8820 when the tracepoint occurred. Any requests for data that are not in
8821 the buffer will fail.
8824 * tfind:: How to select a trace snapshot
8825 * tdump:: How to display all data for a snapshot
8826 * save-tracepoints:: How to save tracepoints for a future run
8830 @subsection @code{tfind @var{n}}
8833 @cindex select trace snapshot
8834 @cindex find trace snapshot
8835 The basic command for selecting a trace snapshot from the buffer is
8836 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
8837 counting from zero. If no argument @var{n} is given, the next
8838 snapshot is selected.
8840 Here are the various forms of using the @code{tfind} command.
8844 Find the first snapshot in the buffer. This is a synonym for
8845 @code{tfind 0} (since 0 is the number of the first snapshot).
8848 Stop debugging trace snapshots, resume @emph{live} debugging.
8851 Same as @samp{tfind none}.
8854 No argument means find the next trace snapshot.
8857 Find the previous trace snapshot before the current one. This permits
8858 retracing earlier steps.
8860 @item tfind tracepoint @var{num}
8861 Find the next snapshot associated with tracepoint @var{num}. Search
8862 proceeds forward from the last examined trace snapshot. If no
8863 argument @var{num} is given, it means find the next snapshot collected
8864 for the same tracepoint as the current snapshot.
8866 @item tfind pc @var{addr}
8867 Find the next snapshot associated with the value @var{addr} of the
8868 program counter. Search proceeds forward from the last examined trace
8869 snapshot. If no argument @var{addr} is given, it means find the next
8870 snapshot with the same value of PC as the current snapshot.
8872 @item tfind outside @var{addr1}, @var{addr2}
8873 Find the next snapshot whose PC is outside the given range of
8876 @item tfind range @var{addr1}, @var{addr2}
8877 Find the next snapshot whose PC is between @var{addr1} and
8878 @var{addr2}. @c FIXME: Is the range inclusive or exclusive?
8880 @item tfind line @r{[}@var{file}:@r{]}@var{n}
8881 Find the next snapshot associated with the source line @var{n}. If
8882 the optional argument @var{file} is given, refer to line @var{n} in
8883 that source file. Search proceeds forward from the last examined
8884 trace snapshot. If no argument @var{n} is given, it means find the
8885 next line other than the one currently being examined; thus saying
8886 @code{tfind line} repeatedly can appear to have the same effect as
8887 stepping from line to line in a @emph{live} debugging session.
8890 The default arguments for the @code{tfind} commands are specifically
8891 designed to make it easy to scan through the trace buffer. For
8892 instance, @code{tfind} with no argument selects the next trace
8893 snapshot, and @code{tfind -} with no argument selects the previous
8894 trace snapshot. So, by giving one @code{tfind} command, and then
8895 simply hitting @key{RET} repeatedly you can examine all the trace
8896 snapshots in order. Or, by saying @code{tfind -} and then hitting
8897 @key{RET} repeatedly you can examine the snapshots in reverse order.
8898 The @code{tfind line} command with no argument selects the snapshot
8899 for the next source line executed. The @code{tfind pc} command with
8900 no argument selects the next snapshot with the same program counter
8901 (PC) as the current frame. The @code{tfind tracepoint} command with
8902 no argument selects the next trace snapshot collected by the same
8903 tracepoint as the current one.
8905 In addition to letting you scan through the trace buffer manually,
8906 these commands make it easy to construct @value{GDBN} scripts that
8907 scan through the trace buffer and print out whatever collected data
8908 you are interested in. Thus, if we want to examine the PC, FP, and SP
8909 registers from each trace frame in the buffer, we can say this:
8912 (@value{GDBP}) @b{tfind start}
8913 (@value{GDBP}) @b{while ($trace_frame != -1)}
8914 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
8915 $trace_frame, $pc, $sp, $fp
8919 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
8920 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
8921 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
8922 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
8923 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
8924 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
8925 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
8926 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
8927 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
8928 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
8929 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
8932 Or, if we want to examine the variable @code{X} at each source line in
8936 (@value{GDBP}) @b{tfind start}
8937 (@value{GDBP}) @b{while ($trace_frame != -1)}
8938 > printf "Frame %d, X == %d\n", $trace_frame, X
8948 @subsection @code{tdump}
8950 @cindex dump all data collected at tracepoint
8951 @cindex tracepoint data, display
8953 This command takes no arguments. It prints all the data collected at
8954 the current trace snapshot.
8957 (@value{GDBP}) @b{trace 444}
8958 (@value{GDBP}) @b{actions}
8959 Enter actions for tracepoint #2, one per line:
8960 > collect $regs, $locals, $args, gdb_long_test
8963 (@value{GDBP}) @b{tstart}
8965 (@value{GDBP}) @b{tfind line 444}
8966 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
8968 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
8970 (@value{GDBP}) @b{tdump}
8971 Data collected at tracepoint 2, trace frame 1:
8972 d0 0xc4aa0085 -995491707
8976 d4 0x71aea3d 119204413
8981 a1 0x3000668 50333288
8984 a4 0x3000698 50333336
8986 fp 0x30bf3c 0x30bf3c
8987 sp 0x30bf34 0x30bf34
8989 pc 0x20b2c8 0x20b2c8
8993 p = 0x20e5b4 "gdb-test"
9000 gdb_long_test = 17 '\021'
9005 @node save-tracepoints
9006 @subsection @code{save-tracepoints @var{filename}}
9007 @kindex save-tracepoints
9008 @cindex save tracepoints for future sessions
9010 This command saves all current tracepoint definitions together with
9011 their actions and passcounts, into a file @file{@var{filename}}
9012 suitable for use in a later debugging session. To read the saved
9013 tracepoint definitions, use the @code{source} command (@pxref{Command
9016 @node Tracepoint Variables
9017 @section Convenience Variables for Tracepoints
9018 @cindex tracepoint variables
9019 @cindex convenience variables for tracepoints
9022 @vindex $trace_frame
9023 @item (int) $trace_frame
9024 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
9025 snapshot is selected.
9028 @item (int) $tracepoint
9029 The tracepoint for the current trace snapshot.
9032 @item (int) $trace_line
9033 The line number for the current trace snapshot.
9036 @item (char []) $trace_file
9037 The source file for the current trace snapshot.
9040 @item (char []) $trace_func
9041 The name of the function containing @code{$tracepoint}.
9044 Note: @code{$trace_file} is not suitable for use in @code{printf},
9045 use @code{output} instead.
9047 Here's a simple example of using these convenience variables for
9048 stepping through all the trace snapshots and printing some of their
9052 (@value{GDBP}) @b{tfind start}
9054 (@value{GDBP}) @b{while $trace_frame != -1}
9055 > output $trace_file
9056 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
9062 @chapter Debugging Programs That Use Overlays
9065 If your program is too large to fit completely in your target system's
9066 memory, you can sometimes use @dfn{overlays} to work around this
9067 problem. @value{GDBN} provides some support for debugging programs that
9071 * How Overlays Work:: A general explanation of overlays.
9072 * Overlay Commands:: Managing overlays in @value{GDBN}.
9073 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
9074 mapped by asking the inferior.
9075 * Overlay Sample Program:: A sample program using overlays.
9078 @node How Overlays Work
9079 @section How Overlays Work
9080 @cindex mapped overlays
9081 @cindex unmapped overlays
9082 @cindex load address, overlay's
9083 @cindex mapped address
9084 @cindex overlay area
9086 Suppose you have a computer whose instruction address space is only 64
9087 kilobytes long, but which has much more memory which can be accessed by
9088 other means: special instructions, segment registers, or memory
9089 management hardware, for example. Suppose further that you want to
9090 adapt a program which is larger than 64 kilobytes to run on this system.
9092 One solution is to identify modules of your program which are relatively
9093 independent, and need not call each other directly; call these modules
9094 @dfn{overlays}. Separate the overlays from the main program, and place
9095 their machine code in the larger memory. Place your main program in
9096 instruction memory, but leave at least enough space there to hold the
9097 largest overlay as well.
9099 Now, to call a function located in an overlay, you must first copy that
9100 overlay's machine code from the large memory into the space set aside
9101 for it in the instruction memory, and then jump to its entry point
9104 @c NB: In the below the mapped area's size is greater or equal to the
9105 @c size of all overlays. This is intentional to remind the developer
9106 @c that overlays don't necessarily need to be the same size.
9110 Data Instruction Larger
9111 Address Space Address Space Address Space
9112 +-----------+ +-----------+ +-----------+
9114 +-----------+ +-----------+ +-----------+<-- overlay 1
9115 | program | | main | .----| overlay 1 | load address
9116 | variables | | program | | +-----------+
9117 | and heap | | | | | |
9118 +-----------+ | | | +-----------+<-- overlay 2
9119 | | +-----------+ | | | load address
9120 +-----------+ | | | .-| overlay 2 |
9122 mapped --->+-----------+ | | +-----------+
9124 | overlay | <-' | | |
9125 | area | <---' +-----------+<-- overlay 3
9126 | | <---. | | load address
9127 +-----------+ `--| overlay 3 |
9134 @anchor{A code overlay}A code overlay
9138 The diagram (@pxref{A code overlay}) shows a system with separate data
9139 and instruction address spaces. To map an overlay, the program copies
9140 its code from the larger address space to the instruction address space.
9141 Since the overlays shown here all use the same mapped address, only one
9142 may be mapped at a time. For a system with a single address space for
9143 data and instructions, the diagram would be similar, except that the
9144 program variables and heap would share an address space with the main
9145 program and the overlay area.
9147 An overlay loaded into instruction memory and ready for use is called a
9148 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
9149 instruction memory. An overlay not present (or only partially present)
9150 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
9151 is its address in the larger memory. The mapped address is also called
9152 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
9153 called the @dfn{load memory address}, or @dfn{LMA}.
9155 Unfortunately, overlays are not a completely transparent way to adapt a
9156 program to limited instruction memory. They introduce a new set of
9157 global constraints you must keep in mind as you design your program:
9162 Before calling or returning to a function in an overlay, your program
9163 must make sure that overlay is actually mapped. Otherwise, the call or
9164 return will transfer control to the right address, but in the wrong
9165 overlay, and your program will probably crash.
9168 If the process of mapping an overlay is expensive on your system, you
9169 will need to choose your overlays carefully to minimize their effect on
9170 your program's performance.
9173 The executable file you load onto your system must contain each
9174 overlay's instructions, appearing at the overlay's load address, not its
9175 mapped address. However, each overlay's instructions must be relocated
9176 and its symbols defined as if the overlay were at its mapped address.
9177 You can use GNU linker scripts to specify different load and relocation
9178 addresses for pieces of your program; see @ref{Overlay Description,,,
9179 ld.info, Using ld: the GNU linker}.
9182 The procedure for loading executable files onto your system must be able
9183 to load their contents into the larger address space as well as the
9184 instruction and data spaces.
9188 The overlay system described above is rather simple, and could be
9189 improved in many ways:
9194 If your system has suitable bank switch registers or memory management
9195 hardware, you could use those facilities to make an overlay's load area
9196 contents simply appear at their mapped address in instruction space.
9197 This would probably be faster than copying the overlay to its mapped
9198 area in the usual way.
9201 If your overlays are small enough, you could set aside more than one
9202 overlay area, and have more than one overlay mapped at a time.
9205 You can use overlays to manage data, as well as instructions. In
9206 general, data overlays are even less transparent to your design than
9207 code overlays: whereas code overlays only require care when you call or
9208 return to functions, data overlays require care every time you access
9209 the data. Also, if you change the contents of a data overlay, you
9210 must copy its contents back out to its load address before you can copy a
9211 different data overlay into the same mapped area.
9216 @node Overlay Commands
9217 @section Overlay Commands
9219 To use @value{GDBN}'s overlay support, each overlay in your program must
9220 correspond to a separate section of the executable file. The section's
9221 virtual memory address and load memory address must be the overlay's
9222 mapped and load addresses. Identifying overlays with sections allows
9223 @value{GDBN} to determine the appropriate address of a function or
9224 variable, depending on whether the overlay is mapped or not.
9226 @value{GDBN}'s overlay commands all start with the word @code{overlay};
9227 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
9232 Disable @value{GDBN}'s overlay support. When overlay support is
9233 disabled, @value{GDBN} assumes that all functions and variables are
9234 always present at their mapped addresses. By default, @value{GDBN}'s
9235 overlay support is disabled.
9237 @item overlay manual
9238 @cindex manual overlay debugging
9239 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
9240 relies on you to tell it which overlays are mapped, and which are not,
9241 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
9242 commands described below.
9244 @item overlay map-overlay @var{overlay}
9245 @itemx overlay map @var{overlay}
9246 @cindex map an overlay
9247 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
9248 be the name of the object file section containing the overlay. When an
9249 overlay is mapped, @value{GDBN} assumes it can find the overlay's
9250 functions and variables at their mapped addresses. @value{GDBN} assumes
9251 that any other overlays whose mapped ranges overlap that of
9252 @var{overlay} are now unmapped.
9254 @item overlay unmap-overlay @var{overlay}
9255 @itemx overlay unmap @var{overlay}
9256 @cindex unmap an overlay
9257 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
9258 must be the name of the object file section containing the overlay.
9259 When an overlay is unmapped, @value{GDBN} assumes it can find the
9260 overlay's functions and variables at their load addresses.
9263 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
9264 consults a data structure the overlay manager maintains in the inferior
9265 to see which overlays are mapped. For details, see @ref{Automatic
9268 @item overlay load-target
9270 @cindex reloading the overlay table
9271 Re-read the overlay table from the inferior. Normally, @value{GDBN}
9272 re-reads the table @value{GDBN} automatically each time the inferior
9273 stops, so this command should only be necessary if you have changed the
9274 overlay mapping yourself using @value{GDBN}. This command is only
9275 useful when using automatic overlay debugging.
9277 @item overlay list-overlays
9279 @cindex listing mapped overlays
9280 Display a list of the overlays currently mapped, along with their mapped
9281 addresses, load addresses, and sizes.
9285 Normally, when @value{GDBN} prints a code address, it includes the name
9286 of the function the address falls in:
9289 (@value{GDBP}) print main
9290 $3 = @{int ()@} 0x11a0 <main>
9293 When overlay debugging is enabled, @value{GDBN} recognizes code in
9294 unmapped overlays, and prints the names of unmapped functions with
9295 asterisks around them. For example, if @code{foo} is a function in an
9296 unmapped overlay, @value{GDBN} prints it this way:
9299 (@value{GDBP}) overlay list
9300 No sections are mapped.
9301 (@value{GDBP}) print foo
9302 $5 = @{int (int)@} 0x100000 <*foo*>
9305 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
9309 (@value{GDBP}) overlay list
9310 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
9311 mapped at 0x1016 - 0x104a
9312 (@value{GDBP}) print foo
9313 $6 = @{int (int)@} 0x1016 <foo>
9316 When overlay debugging is enabled, @value{GDBN} can find the correct
9317 address for functions and variables in an overlay, whether or not the
9318 overlay is mapped. This allows most @value{GDBN} commands, like
9319 @code{break} and @code{disassemble}, to work normally, even on unmapped
9320 code. However, @value{GDBN}'s breakpoint support has some limitations:
9324 @cindex breakpoints in overlays
9325 @cindex overlays, setting breakpoints in
9326 You can set breakpoints in functions in unmapped overlays, as long as
9327 @value{GDBN} can write to the overlay at its load address.
9329 @value{GDBN} can not set hardware or simulator-based breakpoints in
9330 unmapped overlays. However, if you set a breakpoint at the end of your
9331 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
9332 you are using manual overlay management), @value{GDBN} will re-set its
9333 breakpoints properly.
9337 @node Automatic Overlay Debugging
9338 @section Automatic Overlay Debugging
9339 @cindex automatic overlay debugging
9341 @value{GDBN} can automatically track which overlays are mapped and which
9342 are not, given some simple co-operation from the overlay manager in the
9343 inferior. If you enable automatic overlay debugging with the
9344 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
9345 looks in the inferior's memory for certain variables describing the
9346 current state of the overlays.
9348 Here are the variables your overlay manager must define to support
9349 @value{GDBN}'s automatic overlay debugging:
9353 @item @code{_ovly_table}:
9354 This variable must be an array of the following structures:
9359 /* The overlay's mapped address. */
9362 /* The size of the overlay, in bytes. */
9365 /* The overlay's load address. */
9368 /* Non-zero if the overlay is currently mapped;
9370 unsigned long mapped;
9374 @item @code{_novlys}:
9375 This variable must be a four-byte signed integer, holding the total
9376 number of elements in @code{_ovly_table}.
9380 To decide whether a particular overlay is mapped or not, @value{GDBN}
9381 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
9382 @code{lma} members equal the VMA and LMA of the overlay's section in the
9383 executable file. When @value{GDBN} finds a matching entry, it consults
9384 the entry's @code{mapped} member to determine whether the overlay is
9387 In addition, your overlay manager may define a function called
9388 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
9389 will silently set a breakpoint there. If the overlay manager then
9390 calls this function whenever it has changed the overlay table, this
9391 will enable @value{GDBN} to accurately keep track of which overlays
9392 are in program memory, and update any breakpoints that may be set
9393 in overlays. This will allow breakpoints to work even if the
9394 overlays are kept in ROM or other non-writable memory while they
9395 are not being executed.
9397 @node Overlay Sample Program
9398 @section Overlay Sample Program
9399 @cindex overlay example program
9401 When linking a program which uses overlays, you must place the overlays
9402 at their load addresses, while relocating them to run at their mapped
9403 addresses. To do this, you must write a linker script (@pxref{Overlay
9404 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
9405 since linker scripts are specific to a particular host system, target
9406 architecture, and target memory layout, this manual cannot provide
9407 portable sample code demonstrating @value{GDBN}'s overlay support.
9409 However, the @value{GDBN} source distribution does contain an overlaid
9410 program, with linker scripts for a few systems, as part of its test
9411 suite. The program consists of the following files from
9412 @file{gdb/testsuite/gdb.base}:
9416 The main program file.
9418 A simple overlay manager, used by @file{overlays.c}.
9423 Overlay modules, loaded and used by @file{overlays.c}.
9426 Linker scripts for linking the test program on the @code{d10v-elf}
9427 and @code{m32r-elf} targets.
9430 You can build the test program using the @code{d10v-elf} GCC
9431 cross-compiler like this:
9434 $ d10v-elf-gcc -g -c overlays.c
9435 $ d10v-elf-gcc -g -c ovlymgr.c
9436 $ d10v-elf-gcc -g -c foo.c
9437 $ d10v-elf-gcc -g -c bar.c
9438 $ d10v-elf-gcc -g -c baz.c
9439 $ d10v-elf-gcc -g -c grbx.c
9440 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
9441 baz.o grbx.o -Wl,-Td10v.ld -o overlays
9444 The build process is identical for any other architecture, except that
9445 you must substitute the appropriate compiler and linker script for the
9446 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
9450 @chapter Using @value{GDBN} with Different Languages
9453 Although programming languages generally have common aspects, they are
9454 rarely expressed in the same manner. For instance, in ANSI C,
9455 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
9456 Modula-2, it is accomplished by @code{p^}. Values can also be
9457 represented (and displayed) differently. Hex numbers in C appear as
9458 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
9460 @cindex working language
9461 Language-specific information is built into @value{GDBN} for some languages,
9462 allowing you to express operations like the above in your program's
9463 native language, and allowing @value{GDBN} to output values in a manner
9464 consistent with the syntax of your program's native language. The
9465 language you use to build expressions is called the @dfn{working
9469 * Setting:: Switching between source languages
9470 * Show:: Displaying the language
9471 * Checks:: Type and range checks
9472 * Supported Languages:: Supported languages
9473 * Unsupported Languages:: Unsupported languages
9477 @section Switching Between Source Languages
9479 There are two ways to control the working language---either have @value{GDBN}
9480 set it automatically, or select it manually yourself. You can use the
9481 @code{set language} command for either purpose. On startup, @value{GDBN}
9482 defaults to setting the language automatically. The working language is
9483 used to determine how expressions you type are interpreted, how values
9486 In addition to the working language, every source file that
9487 @value{GDBN} knows about has its own working language. For some object
9488 file formats, the compiler might indicate which language a particular
9489 source file is in. However, most of the time @value{GDBN} infers the
9490 language from the name of the file. The language of a source file
9491 controls whether C@t{++} names are demangled---this way @code{backtrace} can
9492 show each frame appropriately for its own language. There is no way to
9493 set the language of a source file from within @value{GDBN}, but you can
9494 set the language associated with a filename extension. @xref{Show, ,
9495 Displaying the Language}.
9497 This is most commonly a problem when you use a program, such
9498 as @code{cfront} or @code{f2c}, that generates C but is written in
9499 another language. In that case, make the
9500 program use @code{#line} directives in its C output; that way
9501 @value{GDBN} will know the correct language of the source code of the original
9502 program, and will display that source code, not the generated C code.
9505 * Filenames:: Filename extensions and languages.
9506 * Manually:: Setting the working language manually
9507 * Automatically:: Having @value{GDBN} infer the source language
9511 @subsection List of Filename Extensions and Languages
9513 If a source file name ends in one of the following extensions, then
9514 @value{GDBN} infers that its language is the one indicated.
9535 Objective-C source file
9542 Modula-2 source file
9546 Assembler source file. This actually behaves almost like C, but
9547 @value{GDBN} does not skip over function prologues when stepping.
9550 In addition, you may set the language associated with a filename
9551 extension. @xref{Show, , Displaying the Language}.
9554 @subsection Setting the Working Language
9556 If you allow @value{GDBN} to set the language automatically,
9557 expressions are interpreted the same way in your debugging session and
9560 @kindex set language
9561 If you wish, you may set the language manually. To do this, issue the
9562 command @samp{set language @var{lang}}, where @var{lang} is the name of
9564 @code{c} or @code{modula-2}.
9565 For a list of the supported languages, type @samp{set language}.
9567 Setting the language manually prevents @value{GDBN} from updating the working
9568 language automatically. This can lead to confusion if you try
9569 to debug a program when the working language is not the same as the
9570 source language, when an expression is acceptable to both
9571 languages---but means different things. For instance, if the current
9572 source file were written in C, and @value{GDBN} was parsing Modula-2, a
9580 might not have the effect you intended. In C, this means to add
9581 @code{b} and @code{c} and place the result in @code{a}. The result
9582 printed would be the value of @code{a}. In Modula-2, this means to compare
9583 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
9586 @subsection Having @value{GDBN} Infer the Source Language
9588 To have @value{GDBN} set the working language automatically, use
9589 @samp{set language local} or @samp{set language auto}. @value{GDBN}
9590 then infers the working language. That is, when your program stops in a
9591 frame (usually by encountering a breakpoint), @value{GDBN} sets the
9592 working language to the language recorded for the function in that
9593 frame. If the language for a frame is unknown (that is, if the function
9594 or block corresponding to the frame was defined in a source file that
9595 does not have a recognized extension), the current working language is
9596 not changed, and @value{GDBN} issues a warning.
9598 This may not seem necessary for most programs, which are written
9599 entirely in one source language. However, program modules and libraries
9600 written in one source language can be used by a main program written in
9601 a different source language. Using @samp{set language auto} in this
9602 case frees you from having to set the working language manually.
9605 @section Displaying the Language
9607 The following commands help you find out which language is the
9608 working language, and also what language source files were written in.
9612 @kindex show language
9613 Display the current working language. This is the
9614 language you can use with commands such as @code{print} to
9615 build and compute expressions that may involve variables in your program.
9618 @kindex info frame@r{, show the source language}
9619 Display the source language for this frame. This language becomes the
9620 working language if you use an identifier from this frame.
9621 @xref{Frame Info, ,Information about a Frame}, to identify the other
9622 information listed here.
9625 @kindex info source@r{, show the source language}
9626 Display the source language of this source file.
9627 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
9628 information listed here.
9631 In unusual circumstances, you may have source files with extensions
9632 not in the standard list. You can then set the extension associated
9633 with a language explicitly:
9636 @item set extension-language @var{ext} @var{language}
9637 @kindex set extension-language
9638 Tell @value{GDBN} that source files with extension @var{ext} are to be
9639 assumed as written in the source language @var{language}.
9641 @item info extensions
9642 @kindex info extensions
9643 List all the filename extensions and the associated languages.
9647 @section Type and Range Checking
9650 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
9651 checking are included, but they do not yet have any effect. This
9652 section documents the intended facilities.
9654 @c FIXME remove warning when type/range code added
9656 Some languages are designed to guard you against making seemingly common
9657 errors through a series of compile- and run-time checks. These include
9658 checking the type of arguments to functions and operators, and making
9659 sure mathematical overflows are caught at run time. Checks such as
9660 these help to ensure a program's correctness once it has been compiled
9661 by eliminating type mismatches, and providing active checks for range
9662 errors when your program is running.
9664 @value{GDBN} can check for conditions like the above if you wish.
9665 Although @value{GDBN} does not check the statements in your program,
9666 it can check expressions entered directly into @value{GDBN} for
9667 evaluation via the @code{print} command, for example. As with the
9668 working language, @value{GDBN} can also decide whether or not to check
9669 automatically based on your program's source language.
9670 @xref{Supported Languages, ,Supported Languages}, for the default
9671 settings of supported languages.
9674 * Type Checking:: An overview of type checking
9675 * Range Checking:: An overview of range checking
9678 @cindex type checking
9679 @cindex checks, type
9681 @subsection An Overview of Type Checking
9683 Some languages, such as Modula-2, are strongly typed, meaning that the
9684 arguments to operators and functions have to be of the correct type,
9685 otherwise an error occurs. These checks prevent type mismatch
9686 errors from ever causing any run-time problems. For example,
9694 The second example fails because the @code{CARDINAL} 1 is not
9695 type-compatible with the @code{REAL} 2.3.
9697 For the expressions you use in @value{GDBN} commands, you can tell the
9698 @value{GDBN} type checker to skip checking;
9699 to treat any mismatches as errors and abandon the expression;
9700 or to only issue warnings when type mismatches occur,
9701 but evaluate the expression anyway. When you choose the last of
9702 these, @value{GDBN} evaluates expressions like the second example above, but
9703 also issues a warning.
9705 Even if you turn type checking off, there may be other reasons
9706 related to type that prevent @value{GDBN} from evaluating an expression.
9707 For instance, @value{GDBN} does not know how to add an @code{int} and
9708 a @code{struct foo}. These particular type errors have nothing to do
9709 with the language in use, and usually arise from expressions, such as
9710 the one described above, which make little sense to evaluate anyway.
9712 Each language defines to what degree it is strict about type. For
9713 instance, both Modula-2 and C require the arguments to arithmetical
9714 operators to be numbers. In C, enumerated types and pointers can be
9715 represented as numbers, so that they are valid arguments to mathematical
9716 operators. @xref{Supported Languages, ,Supported Languages}, for further
9717 details on specific languages.
9719 @value{GDBN} provides some additional commands for controlling the type checker:
9721 @kindex set check type
9722 @kindex show check type
9724 @item set check type auto
9725 Set type checking on or off based on the current working language.
9726 @xref{Supported Languages, ,Supported Languages}, for the default settings for
9729 @item set check type on
9730 @itemx set check type off
9731 Set type checking on or off, overriding the default setting for the
9732 current working language. Issue a warning if the setting does not
9733 match the language default. If any type mismatches occur in
9734 evaluating an expression while type checking is on, @value{GDBN} prints a
9735 message and aborts evaluation of the expression.
9737 @item set check type warn
9738 Cause the type checker to issue warnings, but to always attempt to
9739 evaluate the expression. Evaluating the expression may still
9740 be impossible for other reasons. For example, @value{GDBN} cannot add
9741 numbers and structures.
9744 Show the current setting of the type checker, and whether or not @value{GDBN}
9745 is setting it automatically.
9748 @cindex range checking
9749 @cindex checks, range
9750 @node Range Checking
9751 @subsection An Overview of Range Checking
9753 In some languages (such as Modula-2), it is an error to exceed the
9754 bounds of a type; this is enforced with run-time checks. Such range
9755 checking is meant to ensure program correctness by making sure
9756 computations do not overflow, or indices on an array element access do
9757 not exceed the bounds of the array.
9759 For expressions you use in @value{GDBN} commands, you can tell
9760 @value{GDBN} to treat range errors in one of three ways: ignore them,
9761 always treat them as errors and abandon the expression, or issue
9762 warnings but evaluate the expression anyway.
9764 A range error can result from numerical overflow, from exceeding an
9765 array index bound, or when you type a constant that is not a member
9766 of any type. Some languages, however, do not treat overflows as an
9767 error. In many implementations of C, mathematical overflow causes the
9768 result to ``wrap around'' to lower values---for example, if @var{m} is
9769 the largest integer value, and @var{s} is the smallest, then
9772 @var{m} + 1 @result{} @var{s}
9775 This, too, is specific to individual languages, and in some cases
9776 specific to individual compilers or machines. @xref{Supported Languages, ,
9777 Supported Languages}, for further details on specific languages.
9779 @value{GDBN} provides some additional commands for controlling the range checker:
9781 @kindex set check range
9782 @kindex show check range
9784 @item set check range auto
9785 Set range checking on or off based on the current working language.
9786 @xref{Supported Languages, ,Supported Languages}, for the default settings for
9789 @item set check range on
9790 @itemx set check range off
9791 Set range checking on or off, overriding the default setting for the
9792 current working language. A warning is issued if the setting does not
9793 match the language default. If a range error occurs and range checking is on,
9794 then a message is printed and evaluation of the expression is aborted.
9796 @item set check range warn
9797 Output messages when the @value{GDBN} range checker detects a range error,
9798 but attempt to evaluate the expression anyway. Evaluating the
9799 expression may still be impossible for other reasons, such as accessing
9800 memory that the process does not own (a typical example from many Unix
9804 Show the current setting of the range checker, and whether or not it is
9805 being set automatically by @value{GDBN}.
9808 @node Supported Languages
9809 @section Supported Languages
9811 @value{GDBN} supports C, C@t{++}, Objective-C, Fortran, Java, Pascal,
9812 assembly, Modula-2, and Ada.
9813 @c This is false ...
9814 Some @value{GDBN} features may be used in expressions regardless of the
9815 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
9816 and the @samp{@{type@}addr} construct (@pxref{Expressions,
9817 ,Expressions}) can be used with the constructs of any supported
9820 The following sections detail to what degree each source language is
9821 supported by @value{GDBN}. These sections are not meant to be language
9822 tutorials or references, but serve only as a reference guide to what the
9823 @value{GDBN} expression parser accepts, and what input and output
9824 formats should look like for different languages. There are many good
9825 books written on each of these languages; please look to these for a
9826 language reference or tutorial.
9830 * Objective-C:: Objective-C
9833 * Modula-2:: Modula-2
9838 @subsection C and C@t{++}
9840 @cindex C and C@t{++}
9841 @cindex expressions in C or C@t{++}
9843 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
9844 to both languages. Whenever this is the case, we discuss those languages
9848 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
9849 @cindex @sc{gnu} C@t{++}
9850 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
9851 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
9852 effectively, you must compile your C@t{++} programs with a supported
9853 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
9854 compiler (@code{aCC}).
9856 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
9857 format; if it doesn't work on your system, try the stabs+ debugging
9858 format. You can select those formats explicitly with the @code{g++}
9859 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
9860 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
9861 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
9864 * C Operators:: C and C@t{++} operators
9865 * C Constants:: C and C@t{++} constants
9866 * C Plus Plus Expressions:: C@t{++} expressions
9867 * C Defaults:: Default settings for C and C@t{++}
9868 * C Checks:: C and C@t{++} type and range checks
9869 * Debugging C:: @value{GDBN} and C
9870 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
9871 * Decimal Floating Point:: Numbers in Decimal Floating Point format
9875 @subsubsection C and C@t{++} Operators
9877 @cindex C and C@t{++} operators
9879 Operators must be defined on values of specific types. For instance,
9880 @code{+} is defined on numbers, but not on structures. Operators are
9881 often defined on groups of types.
9883 For the purposes of C and C@t{++}, the following definitions hold:
9888 @emph{Integral types} include @code{int} with any of its storage-class
9889 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
9892 @emph{Floating-point types} include @code{float}, @code{double}, and
9893 @code{long double} (if supported by the target platform).
9896 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
9899 @emph{Scalar types} include all of the above.
9904 The following operators are supported. They are listed here
9905 in order of increasing precedence:
9909 The comma or sequencing operator. Expressions in a comma-separated list
9910 are evaluated from left to right, with the result of the entire
9911 expression being the last expression evaluated.
9914 Assignment. The value of an assignment expression is the value
9915 assigned. Defined on scalar types.
9918 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
9919 and translated to @w{@code{@var{a} = @var{a op b}}}.
9920 @w{@code{@var{op}=}} and @code{=} have the same precedence.
9921 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
9922 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
9925 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
9926 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
9930 Logical @sc{or}. Defined on integral types.
9933 Logical @sc{and}. Defined on integral types.
9936 Bitwise @sc{or}. Defined on integral types.
9939 Bitwise exclusive-@sc{or}. Defined on integral types.
9942 Bitwise @sc{and}. Defined on integral types.
9945 Equality and inequality. Defined on scalar types. The value of these
9946 expressions is 0 for false and non-zero for true.
9948 @item <@r{, }>@r{, }<=@r{, }>=
9949 Less than, greater than, less than or equal, greater than or equal.
9950 Defined on scalar types. The value of these expressions is 0 for false
9951 and non-zero for true.
9954 left shift, and right shift. Defined on integral types.
9957 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
9960 Addition and subtraction. Defined on integral types, floating-point types and
9963 @item *@r{, }/@r{, }%
9964 Multiplication, division, and modulus. Multiplication and division are
9965 defined on integral and floating-point types. Modulus is defined on
9969 Increment and decrement. When appearing before a variable, the
9970 operation is performed before the variable is used in an expression;
9971 when appearing after it, the variable's value is used before the
9972 operation takes place.
9975 Pointer dereferencing. Defined on pointer types. Same precedence as
9979 Address operator. Defined on variables. Same precedence as @code{++}.
9981 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
9982 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
9983 to examine the address
9984 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
9988 Negative. Defined on integral and floating-point types. Same
9989 precedence as @code{++}.
9992 Logical negation. Defined on integral types. Same precedence as
9996 Bitwise complement operator. Defined on integral types. Same precedence as
10001 Structure member, and pointer-to-structure member. For convenience,
10002 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
10003 pointer based on the stored type information.
10004 Defined on @code{struct} and @code{union} data.
10007 Dereferences of pointers to members.
10010 Array indexing. @code{@var{a}[@var{i}]} is defined as
10011 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
10014 Function parameter list. Same precedence as @code{->}.
10017 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
10018 and @code{class} types.
10021 Doubled colons also represent the @value{GDBN} scope operator
10022 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
10026 If an operator is redefined in the user code, @value{GDBN} usually
10027 attempts to invoke the redefined version instead of using the operator's
10028 predefined meaning.
10031 @subsubsection C and C@t{++} Constants
10033 @cindex C and C@t{++} constants
10035 @value{GDBN} allows you to express the constants of C and C@t{++} in the
10040 Integer constants are a sequence of digits. Octal constants are
10041 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
10042 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
10043 @samp{l}, specifying that the constant should be treated as a
10047 Floating point constants are a sequence of digits, followed by a decimal
10048 point, followed by a sequence of digits, and optionally followed by an
10049 exponent. An exponent is of the form:
10050 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
10051 sequence of digits. The @samp{+} is optional for positive exponents.
10052 A floating-point constant may also end with a letter @samp{f} or
10053 @samp{F}, specifying that the constant should be treated as being of
10054 the @code{float} (as opposed to the default @code{double}) type; or with
10055 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
10059 Enumerated constants consist of enumerated identifiers, or their
10060 integral equivalents.
10063 Character constants are a single character surrounded by single quotes
10064 (@code{'}), or a number---the ordinal value of the corresponding character
10065 (usually its @sc{ascii} value). Within quotes, the single character may
10066 be represented by a letter or by @dfn{escape sequences}, which are of
10067 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
10068 of the character's ordinal value; or of the form @samp{\@var{x}}, where
10069 @samp{@var{x}} is a predefined special character---for example,
10070 @samp{\n} for newline.
10073 String constants are a sequence of character constants surrounded by
10074 double quotes (@code{"}). Any valid character constant (as described
10075 above) may appear. Double quotes within the string must be preceded by
10076 a backslash, so for instance @samp{"a\"b'c"} is a string of five
10080 Pointer constants are an integral value. You can also write pointers
10081 to constants using the C operator @samp{&}.
10084 Array constants are comma-separated lists surrounded by braces @samp{@{}
10085 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
10086 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
10087 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
10090 @node C Plus Plus Expressions
10091 @subsubsection C@t{++} Expressions
10093 @cindex expressions in C@t{++}
10094 @value{GDBN} expression handling can interpret most C@t{++} expressions.
10096 @cindex debugging C@t{++} programs
10097 @cindex C@t{++} compilers
10098 @cindex debug formats and C@t{++}
10099 @cindex @value{NGCC} and C@t{++}
10101 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
10102 proper compiler and the proper debug format. Currently, @value{GDBN}
10103 works best when debugging C@t{++} code that is compiled with
10104 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
10105 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
10106 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
10107 stabs+ as their default debug format, so you usually don't need to
10108 specify a debug format explicitly. Other compilers and/or debug formats
10109 are likely to work badly or not at all when using @value{GDBN} to debug
10115 @cindex member functions
10117 Member function calls are allowed; you can use expressions like
10120 count = aml->GetOriginal(x, y)
10123 @vindex this@r{, inside C@t{++} member functions}
10124 @cindex namespace in C@t{++}
10126 While a member function is active (in the selected stack frame), your
10127 expressions have the same namespace available as the member function;
10128 that is, @value{GDBN} allows implicit references to the class instance
10129 pointer @code{this} following the same rules as C@t{++}.
10131 @cindex call overloaded functions
10132 @cindex overloaded functions, calling
10133 @cindex type conversions in C@t{++}
10135 You can call overloaded functions; @value{GDBN} resolves the function
10136 call to the right definition, with some restrictions. @value{GDBN} does not
10137 perform overload resolution involving user-defined type conversions,
10138 calls to constructors, or instantiations of templates that do not exist
10139 in the program. It also cannot handle ellipsis argument lists or
10142 It does perform integral conversions and promotions, floating-point
10143 promotions, arithmetic conversions, pointer conversions, conversions of
10144 class objects to base classes, and standard conversions such as those of
10145 functions or arrays to pointers; it requires an exact match on the
10146 number of function arguments.
10148 Overload resolution is always performed, unless you have specified
10149 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
10150 ,@value{GDBN} Features for C@t{++}}.
10152 You must specify @code{set overload-resolution off} in order to use an
10153 explicit function signature to call an overloaded function, as in
10155 p 'foo(char,int)'('x', 13)
10158 The @value{GDBN} command-completion facility can simplify this;
10159 see @ref{Completion, ,Command Completion}.
10161 @cindex reference declarations
10163 @value{GDBN} understands variables declared as C@t{++} references; you can use
10164 them in expressions just as you do in C@t{++} source---they are automatically
10167 In the parameter list shown when @value{GDBN} displays a frame, the values of
10168 reference variables are not displayed (unlike other variables); this
10169 avoids clutter, since references are often used for large structures.
10170 The @emph{address} of a reference variable is always shown, unless
10171 you have specified @samp{set print address off}.
10174 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
10175 expressions can use it just as expressions in your program do. Since
10176 one scope may be defined in another, you can use @code{::} repeatedly if
10177 necessary, for example in an expression like
10178 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
10179 resolving name scope by reference to source files, in both C and C@t{++}
10180 debugging (@pxref{Variables, ,Program Variables}).
10183 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
10184 calling virtual functions correctly, printing out virtual bases of
10185 objects, calling functions in a base subobject, casting objects, and
10186 invoking user-defined operators.
10189 @subsubsection C and C@t{++} Defaults
10191 @cindex C and C@t{++} defaults
10193 If you allow @value{GDBN} to set type and range checking automatically, they
10194 both default to @code{off} whenever the working language changes to
10195 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
10196 selects the working language.
10198 If you allow @value{GDBN} to set the language automatically, it
10199 recognizes source files whose names end with @file{.c}, @file{.C}, or
10200 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
10201 these files, it sets the working language to C or C@t{++}.
10202 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
10203 for further details.
10205 @c Type checking is (a) primarily motivated by Modula-2, and (b)
10206 @c unimplemented. If (b) changes, it might make sense to let this node
10207 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
10210 @subsubsection C and C@t{++} Type and Range Checks
10212 @cindex C and C@t{++} checks
10214 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
10215 is not used. However, if you turn type checking on, @value{GDBN}
10216 considers two variables type equivalent if:
10220 The two variables are structured and have the same structure, union, or
10224 The two variables have the same type name, or types that have been
10225 declared equivalent through @code{typedef}.
10228 @c leaving this out because neither J Gilmore nor R Pesch understand it.
10231 The two @code{struct}, @code{union}, or @code{enum} variables are
10232 declared in the same declaration. (Note: this may not be true for all C
10237 Range checking, if turned on, is done on mathematical operations. Array
10238 indices are not checked, since they are often used to index a pointer
10239 that is not itself an array.
10242 @subsubsection @value{GDBN} and C
10244 The @code{set print union} and @code{show print union} commands apply to
10245 the @code{union} type. When set to @samp{on}, any @code{union} that is
10246 inside a @code{struct} or @code{class} is also printed. Otherwise, it
10247 appears as @samp{@{...@}}.
10249 The @code{@@} operator aids in the debugging of dynamic arrays, formed
10250 with pointers and a memory allocation function. @xref{Expressions,
10253 @node Debugging C Plus Plus
10254 @subsubsection @value{GDBN} Features for C@t{++}
10256 @cindex commands for C@t{++}
10258 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
10259 designed specifically for use with C@t{++}. Here is a summary:
10262 @cindex break in overloaded functions
10263 @item @r{breakpoint menus}
10264 When you want a breakpoint in a function whose name is overloaded,
10265 @value{GDBN} has the capability to display a menu of possible breakpoint
10266 locations to help you specify which function definition you want.
10267 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
10269 @cindex overloading in C@t{++}
10270 @item rbreak @var{regex}
10271 Setting breakpoints using regular expressions is helpful for setting
10272 breakpoints on overloaded functions that are not members of any special
10274 @xref{Set Breaks, ,Setting Breakpoints}.
10276 @cindex C@t{++} exception handling
10279 Debug C@t{++} exception handling using these commands. @xref{Set
10280 Catchpoints, , Setting Catchpoints}.
10282 @cindex inheritance
10283 @item ptype @var{typename}
10284 Print inheritance relationships as well as other information for type
10286 @xref{Symbols, ,Examining the Symbol Table}.
10288 @cindex C@t{++} symbol display
10289 @item set print demangle
10290 @itemx show print demangle
10291 @itemx set print asm-demangle
10292 @itemx show print asm-demangle
10293 Control whether C@t{++} symbols display in their source form, both when
10294 displaying code as C@t{++} source and when displaying disassemblies.
10295 @xref{Print Settings, ,Print Settings}.
10297 @item set print object
10298 @itemx show print object
10299 Choose whether to print derived (actual) or declared types of objects.
10300 @xref{Print Settings, ,Print Settings}.
10302 @item set print vtbl
10303 @itemx show print vtbl
10304 Control the format for printing virtual function tables.
10305 @xref{Print Settings, ,Print Settings}.
10306 (The @code{vtbl} commands do not work on programs compiled with the HP
10307 ANSI C@t{++} compiler (@code{aCC}).)
10309 @kindex set overload-resolution
10310 @cindex overloaded functions, overload resolution
10311 @item set overload-resolution on
10312 Enable overload resolution for C@t{++} expression evaluation. The default
10313 is on. For overloaded functions, @value{GDBN} evaluates the arguments
10314 and searches for a function whose signature matches the argument types,
10315 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
10316 Expressions, ,C@t{++} Expressions}, for details).
10317 If it cannot find a match, it emits a message.
10319 @item set overload-resolution off
10320 Disable overload resolution for C@t{++} expression evaluation. For
10321 overloaded functions that are not class member functions, @value{GDBN}
10322 chooses the first function of the specified name that it finds in the
10323 symbol table, whether or not its arguments are of the correct type. For
10324 overloaded functions that are class member functions, @value{GDBN}
10325 searches for a function whose signature @emph{exactly} matches the
10328 @kindex show overload-resolution
10329 @item show overload-resolution
10330 Show the current setting of overload resolution.
10332 @item @r{Overloaded symbol names}
10333 You can specify a particular definition of an overloaded symbol, using
10334 the same notation that is used to declare such symbols in C@t{++}: type
10335 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
10336 also use the @value{GDBN} command-line word completion facilities to list the
10337 available choices, or to finish the type list for you.
10338 @xref{Completion,, Command Completion}, for details on how to do this.
10341 @node Decimal Floating Point
10342 @subsubsection Decimal Floating Point format
10343 @cindex decimal floating point format
10345 @value{GDBN} can examine, set and perform computations with numbers in
10346 decimal floating point format, which in the C language correspond to the
10347 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
10348 specified by the extension to support decimal floating-point arithmetic.
10350 There are two encodings in use, depending on the architecture: BID (Binary
10351 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
10352 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
10355 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
10356 to manipulate decimal floating point numbers, it is not possible to convert
10357 (using a cast, for example) integers wider than 32-bit to decimal float.
10359 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
10360 point computations, error checking in decimal float operations ignores
10361 underflow, overflow and divide by zero exceptions.
10363 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
10364 to inspect @code{_Decimal128} values stored in floating point registers. See
10365 @ref{PowerPC,,PowerPC} for more details.
10368 @subsection Objective-C
10370 @cindex Objective-C
10371 This section provides information about some commands and command
10372 options that are useful for debugging Objective-C code. See also
10373 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
10374 few more commands specific to Objective-C support.
10377 * Method Names in Commands::
10378 * The Print Command with Objective-C::
10381 @node Method Names in Commands
10382 @subsubsection Method Names in Commands
10384 The following commands have been extended to accept Objective-C method
10385 names as line specifications:
10387 @kindex clear@r{, and Objective-C}
10388 @kindex break@r{, and Objective-C}
10389 @kindex info line@r{, and Objective-C}
10390 @kindex jump@r{, and Objective-C}
10391 @kindex list@r{, and Objective-C}
10395 @item @code{info line}
10400 A fully qualified Objective-C method name is specified as
10403 -[@var{Class} @var{methodName}]
10406 where the minus sign is used to indicate an instance method and a
10407 plus sign (not shown) is used to indicate a class method. The class
10408 name @var{Class} and method name @var{methodName} are enclosed in
10409 brackets, similar to the way messages are specified in Objective-C
10410 source code. For example, to set a breakpoint at the @code{create}
10411 instance method of class @code{Fruit} in the program currently being
10415 break -[Fruit create]
10418 To list ten program lines around the @code{initialize} class method,
10422 list +[NSText initialize]
10425 In the current version of @value{GDBN}, the plus or minus sign is
10426 required. In future versions of @value{GDBN}, the plus or minus
10427 sign will be optional, but you can use it to narrow the search. It
10428 is also possible to specify just a method name:
10434 You must specify the complete method name, including any colons. If
10435 your program's source files contain more than one @code{create} method,
10436 you'll be presented with a numbered list of classes that implement that
10437 method. Indicate your choice by number, or type @samp{0} to exit if
10440 As another example, to clear a breakpoint established at the
10441 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
10444 clear -[NSWindow makeKeyAndOrderFront:]
10447 @node The Print Command with Objective-C
10448 @subsubsection The Print Command With Objective-C
10449 @cindex Objective-C, print objects
10450 @kindex print-object
10451 @kindex po @r{(@code{print-object})}
10453 The print command has also been extended to accept methods. For example:
10456 print -[@var{object} hash]
10459 @cindex print an Objective-C object description
10460 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
10462 will tell @value{GDBN} to send the @code{hash} message to @var{object}
10463 and print the result. Also, an additional command has been added,
10464 @code{print-object} or @code{po} for short, which is meant to print
10465 the description of an object. However, this command may only work
10466 with certain Objective-C libraries that have a particular hook
10467 function, @code{_NSPrintForDebugger}, defined.
10470 @subsection Fortran
10471 @cindex Fortran-specific support in @value{GDBN}
10473 @value{GDBN} can be used to debug programs written in Fortran, but it
10474 currently supports only the features of Fortran 77 language.
10476 @cindex trailing underscore, in Fortran symbols
10477 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
10478 among them) append an underscore to the names of variables and
10479 functions. When you debug programs compiled by those compilers, you
10480 will need to refer to variables and functions with a trailing
10484 * Fortran Operators:: Fortran operators and expressions
10485 * Fortran Defaults:: Default settings for Fortran
10486 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
10489 @node Fortran Operators
10490 @subsubsection Fortran Operators and Expressions
10492 @cindex Fortran operators and expressions
10494 Operators must be defined on values of specific types. For instance,
10495 @code{+} is defined on numbers, but not on characters or other non-
10496 arithmetic types. Operators are often defined on groups of types.
10500 The exponentiation operator. It raises the first operand to the power
10504 The range operator. Normally used in the form of array(low:high) to
10505 represent a section of array.
10508 The access component operator. Normally used to access elements in derived
10509 types. Also suitable for unions. As unions aren't part of regular Fortran,
10510 this can only happen when accessing a register that uses a gdbarch-defined
10514 @node Fortran Defaults
10515 @subsubsection Fortran Defaults
10517 @cindex Fortran Defaults
10519 Fortran symbols are usually case-insensitive, so @value{GDBN} by
10520 default uses case-insensitive matches for Fortran symbols. You can
10521 change that with the @samp{set case-insensitive} command, see
10522 @ref{Symbols}, for the details.
10524 @node Special Fortran Commands
10525 @subsubsection Special Fortran Commands
10527 @cindex Special Fortran commands
10529 @value{GDBN} has some commands to support Fortran-specific features,
10530 such as displaying common blocks.
10533 @cindex @code{COMMON} blocks, Fortran
10534 @kindex info common
10535 @item info common @r{[}@var{common-name}@r{]}
10536 This command prints the values contained in the Fortran @code{COMMON}
10537 block whose name is @var{common-name}. With no argument, the names of
10538 all @code{COMMON} blocks visible at the current program location are
10545 @cindex Pascal support in @value{GDBN}, limitations
10546 Debugging Pascal programs which use sets, subranges, file variables, or
10547 nested functions does not currently work. @value{GDBN} does not support
10548 entering expressions, printing values, or similar features using Pascal
10551 The Pascal-specific command @code{set print pascal_static-members}
10552 controls whether static members of Pascal objects are displayed.
10553 @xref{Print Settings, pascal_static-members}.
10556 @subsection Modula-2
10558 @cindex Modula-2, @value{GDBN} support
10560 The extensions made to @value{GDBN} to support Modula-2 only support
10561 output from the @sc{gnu} Modula-2 compiler (which is currently being
10562 developed). Other Modula-2 compilers are not currently supported, and
10563 attempting to debug executables produced by them is most likely
10564 to give an error as @value{GDBN} reads in the executable's symbol
10567 @cindex expressions in Modula-2
10569 * M2 Operators:: Built-in operators
10570 * Built-In Func/Proc:: Built-in functions and procedures
10571 * M2 Constants:: Modula-2 constants
10572 * M2 Types:: Modula-2 types
10573 * M2 Defaults:: Default settings for Modula-2
10574 * Deviations:: Deviations from standard Modula-2
10575 * M2 Checks:: Modula-2 type and range checks
10576 * M2 Scope:: The scope operators @code{::} and @code{.}
10577 * GDB/M2:: @value{GDBN} and Modula-2
10581 @subsubsection Operators
10582 @cindex Modula-2 operators
10584 Operators must be defined on values of specific types. For instance,
10585 @code{+} is defined on numbers, but not on structures. Operators are
10586 often defined on groups of types. For the purposes of Modula-2, the
10587 following definitions hold:
10592 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
10596 @emph{Character types} consist of @code{CHAR} and its subranges.
10599 @emph{Floating-point types} consist of @code{REAL}.
10602 @emph{Pointer types} consist of anything declared as @code{POINTER TO
10606 @emph{Scalar types} consist of all of the above.
10609 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
10612 @emph{Boolean types} consist of @code{BOOLEAN}.
10616 The following operators are supported, and appear in order of
10617 increasing precedence:
10621 Function argument or array index separator.
10624 Assignment. The value of @var{var} @code{:=} @var{value} is
10628 Less than, greater than on integral, floating-point, or enumerated
10632 Less than or equal to, greater than or equal to
10633 on integral, floating-point and enumerated types, or set inclusion on
10634 set types. Same precedence as @code{<}.
10636 @item =@r{, }<>@r{, }#
10637 Equality and two ways of expressing inequality, valid on scalar types.
10638 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
10639 available for inequality, since @code{#} conflicts with the script
10643 Set membership. Defined on set types and the types of their members.
10644 Same precedence as @code{<}.
10647 Boolean disjunction. Defined on boolean types.
10650 Boolean conjunction. Defined on boolean types.
10653 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
10656 Addition and subtraction on integral and floating-point types, or union
10657 and difference on set types.
10660 Multiplication on integral and floating-point types, or set intersection
10664 Division on floating-point types, or symmetric set difference on set
10665 types. Same precedence as @code{*}.
10668 Integer division and remainder. Defined on integral types. Same
10669 precedence as @code{*}.
10672 Negative. Defined on @code{INTEGER} and @code{REAL} data.
10675 Pointer dereferencing. Defined on pointer types.
10678 Boolean negation. Defined on boolean types. Same precedence as
10682 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
10683 precedence as @code{^}.
10686 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
10689 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
10693 @value{GDBN} and Modula-2 scope operators.
10697 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
10698 treats the use of the operator @code{IN}, or the use of operators
10699 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
10700 @code{<=}, and @code{>=} on sets as an error.
10704 @node Built-In Func/Proc
10705 @subsubsection Built-in Functions and Procedures
10706 @cindex Modula-2 built-ins
10708 Modula-2 also makes available several built-in procedures and functions.
10709 In describing these, the following metavariables are used:
10714 represents an @code{ARRAY} variable.
10717 represents a @code{CHAR} constant or variable.
10720 represents a variable or constant of integral type.
10723 represents an identifier that belongs to a set. Generally used in the
10724 same function with the metavariable @var{s}. The type of @var{s} should
10725 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
10728 represents a variable or constant of integral or floating-point type.
10731 represents a variable or constant of floating-point type.
10737 represents a variable.
10740 represents a variable or constant of one of many types. See the
10741 explanation of the function for details.
10744 All Modula-2 built-in procedures also return a result, described below.
10748 Returns the absolute value of @var{n}.
10751 If @var{c} is a lower case letter, it returns its upper case
10752 equivalent, otherwise it returns its argument.
10755 Returns the character whose ordinal value is @var{i}.
10758 Decrements the value in the variable @var{v} by one. Returns the new value.
10760 @item DEC(@var{v},@var{i})
10761 Decrements the value in the variable @var{v} by @var{i}. Returns the
10764 @item EXCL(@var{m},@var{s})
10765 Removes the element @var{m} from the set @var{s}. Returns the new
10768 @item FLOAT(@var{i})
10769 Returns the floating point equivalent of the integer @var{i}.
10771 @item HIGH(@var{a})
10772 Returns the index of the last member of @var{a}.
10775 Increments the value in the variable @var{v} by one. Returns the new value.
10777 @item INC(@var{v},@var{i})
10778 Increments the value in the variable @var{v} by @var{i}. Returns the
10781 @item INCL(@var{m},@var{s})
10782 Adds the element @var{m} to the set @var{s} if it is not already
10783 there. Returns the new set.
10786 Returns the maximum value of the type @var{t}.
10789 Returns the minimum value of the type @var{t}.
10792 Returns boolean TRUE if @var{i} is an odd number.
10795 Returns the ordinal value of its argument. For example, the ordinal
10796 value of a character is its @sc{ascii} value (on machines supporting the
10797 @sc{ascii} character set). @var{x} must be of an ordered type, which include
10798 integral, character and enumerated types.
10800 @item SIZE(@var{x})
10801 Returns the size of its argument. @var{x} can be a variable or a type.
10803 @item TRUNC(@var{r})
10804 Returns the integral part of @var{r}.
10806 @item TSIZE(@var{x})
10807 Returns the size of its argument. @var{x} can be a variable or a type.
10809 @item VAL(@var{t},@var{i})
10810 Returns the member of the type @var{t} whose ordinal value is @var{i}.
10814 @emph{Warning:} Sets and their operations are not yet supported, so
10815 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
10819 @cindex Modula-2 constants
10821 @subsubsection Constants
10823 @value{GDBN} allows you to express the constants of Modula-2 in the following
10829 Integer constants are simply a sequence of digits. When used in an
10830 expression, a constant is interpreted to be type-compatible with the
10831 rest of the expression. Hexadecimal integers are specified by a
10832 trailing @samp{H}, and octal integers by a trailing @samp{B}.
10835 Floating point constants appear as a sequence of digits, followed by a
10836 decimal point and another sequence of digits. An optional exponent can
10837 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
10838 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
10839 digits of the floating point constant must be valid decimal (base 10)
10843 Character constants consist of a single character enclosed by a pair of
10844 like quotes, either single (@code{'}) or double (@code{"}). They may
10845 also be expressed by their ordinal value (their @sc{ascii} value, usually)
10846 followed by a @samp{C}.
10849 String constants consist of a sequence of characters enclosed by a
10850 pair of like quotes, either single (@code{'}) or double (@code{"}).
10851 Escape sequences in the style of C are also allowed. @xref{C
10852 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
10856 Enumerated constants consist of an enumerated identifier.
10859 Boolean constants consist of the identifiers @code{TRUE} and
10863 Pointer constants consist of integral values only.
10866 Set constants are not yet supported.
10870 @subsubsection Modula-2 Types
10871 @cindex Modula-2 types
10873 Currently @value{GDBN} can print the following data types in Modula-2
10874 syntax: array types, record types, set types, pointer types, procedure
10875 types, enumerated types, subrange types and base types. You can also
10876 print the contents of variables declared using these type.
10877 This section gives a number of simple source code examples together with
10878 sample @value{GDBN} sessions.
10880 The first example contains the following section of code:
10889 and you can request @value{GDBN} to interrogate the type and value of
10890 @code{r} and @code{s}.
10893 (@value{GDBP}) print s
10895 (@value{GDBP}) ptype s
10897 (@value{GDBP}) print r
10899 (@value{GDBP}) ptype r
10904 Likewise if your source code declares @code{s} as:
10908 s: SET ['A'..'Z'] ;
10912 then you may query the type of @code{s} by:
10915 (@value{GDBP}) ptype s
10916 type = SET ['A'..'Z']
10920 Note that at present you cannot interactively manipulate set
10921 expressions using the debugger.
10923 The following example shows how you might declare an array in Modula-2
10924 and how you can interact with @value{GDBN} to print its type and contents:
10928 s: ARRAY [-10..10] OF CHAR ;
10932 (@value{GDBP}) ptype s
10933 ARRAY [-10..10] OF CHAR
10936 Note that the array handling is not yet complete and although the type
10937 is printed correctly, expression handling still assumes that all
10938 arrays have a lower bound of zero and not @code{-10} as in the example
10941 Here are some more type related Modula-2 examples:
10945 colour = (blue, red, yellow, green) ;
10946 t = [blue..yellow] ;
10954 The @value{GDBN} interaction shows how you can query the data type
10955 and value of a variable.
10958 (@value{GDBP}) print s
10960 (@value{GDBP}) ptype t
10961 type = [blue..yellow]
10965 In this example a Modula-2 array is declared and its contents
10966 displayed. Observe that the contents are written in the same way as
10967 their @code{C} counterparts.
10971 s: ARRAY [1..5] OF CARDINAL ;
10977 (@value{GDBP}) print s
10978 $1 = @{1, 0, 0, 0, 0@}
10979 (@value{GDBP}) ptype s
10980 type = ARRAY [1..5] OF CARDINAL
10983 The Modula-2 language interface to @value{GDBN} also understands
10984 pointer types as shown in this example:
10988 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
10995 and you can request that @value{GDBN} describes the type of @code{s}.
10998 (@value{GDBP}) ptype s
10999 type = POINTER TO ARRAY [1..5] OF CARDINAL
11002 @value{GDBN} handles compound types as we can see in this example.
11003 Here we combine array types, record types, pointer types and subrange
11014 myarray = ARRAY myrange OF CARDINAL ;
11015 myrange = [-2..2] ;
11017 s: POINTER TO ARRAY myrange OF foo ;
11021 and you can ask @value{GDBN} to describe the type of @code{s} as shown
11025 (@value{GDBP}) ptype s
11026 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
11029 f3 : ARRAY [-2..2] OF CARDINAL;
11034 @subsubsection Modula-2 Defaults
11035 @cindex Modula-2 defaults
11037 If type and range checking are set automatically by @value{GDBN}, they
11038 both default to @code{on} whenever the working language changes to
11039 Modula-2. This happens regardless of whether you or @value{GDBN}
11040 selected the working language.
11042 If you allow @value{GDBN} to set the language automatically, then entering
11043 code compiled from a file whose name ends with @file{.mod} sets the
11044 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
11045 Infer the Source Language}, for further details.
11048 @subsubsection Deviations from Standard Modula-2
11049 @cindex Modula-2, deviations from
11051 A few changes have been made to make Modula-2 programs easier to debug.
11052 This is done primarily via loosening its type strictness:
11056 Unlike in standard Modula-2, pointer constants can be formed by
11057 integers. This allows you to modify pointer variables during
11058 debugging. (In standard Modula-2, the actual address contained in a
11059 pointer variable is hidden from you; it can only be modified
11060 through direct assignment to another pointer variable or expression that
11061 returned a pointer.)
11064 C escape sequences can be used in strings and characters to represent
11065 non-printable characters. @value{GDBN} prints out strings with these
11066 escape sequences embedded. Single non-printable characters are
11067 printed using the @samp{CHR(@var{nnn})} format.
11070 The assignment operator (@code{:=}) returns the value of its right-hand
11074 All built-in procedures both modify @emph{and} return their argument.
11078 @subsubsection Modula-2 Type and Range Checks
11079 @cindex Modula-2 checks
11082 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
11085 @c FIXME remove warning when type/range checks added
11087 @value{GDBN} considers two Modula-2 variables type equivalent if:
11091 They are of types that have been declared equivalent via a @code{TYPE
11092 @var{t1} = @var{t2}} statement
11095 They have been declared on the same line. (Note: This is true of the
11096 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
11099 As long as type checking is enabled, any attempt to combine variables
11100 whose types are not equivalent is an error.
11102 Range checking is done on all mathematical operations, assignment, array
11103 index bounds, and all built-in functions and procedures.
11106 @subsubsection The Scope Operators @code{::} and @code{.}
11108 @cindex @code{.}, Modula-2 scope operator
11109 @cindex colon, doubled as scope operator
11111 @vindex colon-colon@r{, in Modula-2}
11112 @c Info cannot handle :: but TeX can.
11115 @vindex ::@r{, in Modula-2}
11118 There are a few subtle differences between the Modula-2 scope operator
11119 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
11124 @var{module} . @var{id}
11125 @var{scope} :: @var{id}
11129 where @var{scope} is the name of a module or a procedure,
11130 @var{module} the name of a module, and @var{id} is any declared
11131 identifier within your program, except another module.
11133 Using the @code{::} operator makes @value{GDBN} search the scope
11134 specified by @var{scope} for the identifier @var{id}. If it is not
11135 found in the specified scope, then @value{GDBN} searches all scopes
11136 enclosing the one specified by @var{scope}.
11138 Using the @code{.} operator makes @value{GDBN} search the current scope for
11139 the identifier specified by @var{id} that was imported from the
11140 definition module specified by @var{module}. With this operator, it is
11141 an error if the identifier @var{id} was not imported from definition
11142 module @var{module}, or if @var{id} is not an identifier in
11146 @subsubsection @value{GDBN} and Modula-2
11148 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
11149 Five subcommands of @code{set print} and @code{show print} apply
11150 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
11151 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
11152 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
11153 analogue in Modula-2.
11155 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
11156 with any language, is not useful with Modula-2. Its
11157 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
11158 created in Modula-2 as they can in C or C@t{++}. However, because an
11159 address can be specified by an integral constant, the construct
11160 @samp{@{@var{type}@}@var{adrexp}} is still useful.
11162 @cindex @code{#} in Modula-2
11163 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
11164 interpreted as the beginning of a comment. Use @code{<>} instead.
11170 The extensions made to @value{GDBN} for Ada only support
11171 output from the @sc{gnu} Ada (GNAT) compiler.
11172 Other Ada compilers are not currently supported, and
11173 attempting to debug executables produced by them is most likely
11177 @cindex expressions in Ada
11179 * Ada Mode Intro:: General remarks on the Ada syntax
11180 and semantics supported by Ada mode
11182 * Omissions from Ada:: Restrictions on the Ada expression syntax.
11183 * Additions to Ada:: Extensions of the Ada expression syntax.
11184 * Stopping Before Main Program:: Debugging the program during elaboration.
11185 * Ada Tasks:: Listing and setting breakpoints in tasks.
11186 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
11187 * Ada Glitches:: Known peculiarities of Ada mode.
11190 @node Ada Mode Intro
11191 @subsubsection Introduction
11192 @cindex Ada mode, general
11194 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
11195 syntax, with some extensions.
11196 The philosophy behind the design of this subset is
11200 That @value{GDBN} should provide basic literals and access to operations for
11201 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
11202 leaving more sophisticated computations to subprograms written into the
11203 program (which therefore may be called from @value{GDBN}).
11206 That type safety and strict adherence to Ada language restrictions
11207 are not particularly important to the @value{GDBN} user.
11210 That brevity is important to the @value{GDBN} user.
11213 Thus, for brevity, the debugger acts as if all names declared in
11214 user-written packages are directly visible, even if they are not visible
11215 according to Ada rules, thus making it unnecessary to fully qualify most
11216 names with their packages, regardless of context. Where this causes
11217 ambiguity, @value{GDBN} asks the user's intent.
11219 The debugger will start in Ada mode if it detects an Ada main program.
11220 As for other languages, it will enter Ada mode when stopped in a program that
11221 was translated from an Ada source file.
11223 While in Ada mode, you may use `@t{--}' for comments. This is useful
11224 mostly for documenting command files. The standard @value{GDBN} comment
11225 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
11226 middle (to allow based literals).
11228 The debugger supports limited overloading. Given a subprogram call in which
11229 the function symbol has multiple definitions, it will use the number of
11230 actual parameters and some information about their types to attempt to narrow
11231 the set of definitions. It also makes very limited use of context, preferring
11232 procedures to functions in the context of the @code{call} command, and
11233 functions to procedures elsewhere.
11235 @node Omissions from Ada
11236 @subsubsection Omissions from Ada
11237 @cindex Ada, omissions from
11239 Here are the notable omissions from the subset:
11243 Only a subset of the attributes are supported:
11247 @t{'First}, @t{'Last}, and @t{'Length}
11248 on array objects (not on types and subtypes).
11251 @t{'Min} and @t{'Max}.
11254 @t{'Pos} and @t{'Val}.
11260 @t{'Range} on array objects (not subtypes), but only as the right
11261 operand of the membership (@code{in}) operator.
11264 @t{'Access}, @t{'Unchecked_Access}, and
11265 @t{'Unrestricted_Access} (a GNAT extension).
11273 @code{Characters.Latin_1} are not available and
11274 concatenation is not implemented. Thus, escape characters in strings are
11275 not currently available.
11278 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
11279 equality of representations. They will generally work correctly
11280 for strings and arrays whose elements have integer or enumeration types.
11281 They may not work correctly for arrays whose element
11282 types have user-defined equality, for arrays of real values
11283 (in particular, IEEE-conformant floating point, because of negative
11284 zeroes and NaNs), and for arrays whose elements contain unused bits with
11285 indeterminate values.
11288 The other component-by-component array operations (@code{and}, @code{or},
11289 @code{xor}, @code{not}, and relational tests other than equality)
11290 are not implemented.
11293 @cindex array aggregates (Ada)
11294 @cindex record aggregates (Ada)
11295 @cindex aggregates (Ada)
11296 There is limited support for array and record aggregates. They are
11297 permitted only on the right sides of assignments, as in these examples:
11300 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
11301 (@value{GDBP}) set An_Array := (1, others => 0)
11302 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
11303 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
11304 (@value{GDBP}) set A_Record := (1, "Peter", True);
11305 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
11309 discriminant's value by assigning an aggregate has an
11310 undefined effect if that discriminant is used within the record.
11311 However, you can first modify discriminants by directly assigning to
11312 them (which normally would not be allowed in Ada), and then performing an
11313 aggregate assignment. For example, given a variable @code{A_Rec}
11314 declared to have a type such as:
11317 type Rec (Len : Small_Integer := 0) is record
11319 Vals : IntArray (1 .. Len);
11323 you can assign a value with a different size of @code{Vals} with two
11327 (@value{GDBP}) set A_Rec.Len := 4
11328 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
11331 As this example also illustrates, @value{GDBN} is very loose about the usual
11332 rules concerning aggregates. You may leave out some of the
11333 components of an array or record aggregate (such as the @code{Len}
11334 component in the assignment to @code{A_Rec} above); they will retain their
11335 original values upon assignment. You may freely use dynamic values as
11336 indices in component associations. You may even use overlapping or
11337 redundant component associations, although which component values are
11338 assigned in such cases is not defined.
11341 Calls to dispatching subprograms are not implemented.
11344 The overloading algorithm is much more limited (i.e., less selective)
11345 than that of real Ada. It makes only limited use of the context in
11346 which a subexpression appears to resolve its meaning, and it is much
11347 looser in its rules for allowing type matches. As a result, some
11348 function calls will be ambiguous, and the user will be asked to choose
11349 the proper resolution.
11352 The @code{new} operator is not implemented.
11355 Entry calls are not implemented.
11358 Aside from printing, arithmetic operations on the native VAX floating-point
11359 formats are not supported.
11362 It is not possible to slice a packed array.
11365 The names @code{True} and @code{False}, when not part of a qualified name,
11366 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
11368 Should your program
11369 redefine these names in a package or procedure (at best a dubious practice),
11370 you will have to use fully qualified names to access their new definitions.
11373 @node Additions to Ada
11374 @subsubsection Additions to Ada
11375 @cindex Ada, deviations from
11377 As it does for other languages, @value{GDBN} makes certain generic
11378 extensions to Ada (@pxref{Expressions}):
11382 If the expression @var{E} is a variable residing in memory (typically
11383 a local variable or array element) and @var{N} is a positive integer,
11384 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
11385 @var{N}-1 adjacent variables following it in memory as an array. In
11386 Ada, this operator is generally not necessary, since its prime use is
11387 in displaying parts of an array, and slicing will usually do this in
11388 Ada. However, there are occasional uses when debugging programs in
11389 which certain debugging information has been optimized away.
11392 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
11393 appears in function or file @var{B}.'' When @var{B} is a file name,
11394 you must typically surround it in single quotes.
11397 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
11398 @var{type} that appears at address @var{addr}.''
11401 A name starting with @samp{$} is a convenience variable
11402 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
11405 In addition, @value{GDBN} provides a few other shortcuts and outright
11406 additions specific to Ada:
11410 The assignment statement is allowed as an expression, returning
11411 its right-hand operand as its value. Thus, you may enter
11414 (@value{GDBP}) set x := y + 3
11415 (@value{GDBP}) print A(tmp := y + 1)
11419 The semicolon is allowed as an ``operator,'' returning as its value
11420 the value of its right-hand operand.
11421 This allows, for example,
11422 complex conditional breaks:
11425 (@value{GDBP}) break f
11426 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
11430 Rather than use catenation and symbolic character names to introduce special
11431 characters into strings, one may instead use a special bracket notation,
11432 which is also used to print strings. A sequence of characters of the form
11433 @samp{["@var{XX}"]} within a string or character literal denotes the
11434 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
11435 sequence of characters @samp{["""]} also denotes a single quotation mark
11436 in strings. For example,
11438 "One line.["0a"]Next line.["0a"]"
11441 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
11445 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
11446 @t{'Max} is optional (and is ignored in any case). For example, it is valid
11450 (@value{GDBP}) print 'max(x, y)
11454 When printing arrays, @value{GDBN} uses positional notation when the
11455 array has a lower bound of 1, and uses a modified named notation otherwise.
11456 For example, a one-dimensional array of three integers with a lower bound
11457 of 3 might print as
11464 That is, in contrast to valid Ada, only the first component has a @code{=>}
11468 You may abbreviate attributes in expressions with any unique,
11469 multi-character subsequence of
11470 their names (an exact match gets preference).
11471 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
11472 in place of @t{a'length}.
11475 @cindex quoting Ada internal identifiers
11476 Since Ada is case-insensitive, the debugger normally maps identifiers you type
11477 to lower case. The GNAT compiler uses upper-case characters for
11478 some of its internal identifiers, which are normally of no interest to users.
11479 For the rare occasions when you actually have to look at them,
11480 enclose them in angle brackets to avoid the lower-case mapping.
11483 (@value{GDBP}) print <JMPBUF_SAVE>[0]
11487 Printing an object of class-wide type or dereferencing an
11488 access-to-class-wide value will display all the components of the object's
11489 specific type (as indicated by its run-time tag). Likewise, component
11490 selection on such a value will operate on the specific type of the
11495 @node Stopping Before Main Program
11496 @subsubsection Stopping at the Very Beginning
11498 @cindex breakpointing Ada elaboration code
11499 It is sometimes necessary to debug the program during elaboration, and
11500 before reaching the main procedure.
11501 As defined in the Ada Reference
11502 Manual, the elaboration code is invoked from a procedure called
11503 @code{adainit}. To run your program up to the beginning of
11504 elaboration, simply use the following two commands:
11505 @code{tbreak adainit} and @code{run}.
11508 @subsubsection Extensions for Ada Tasks
11509 @cindex Ada, tasking
11511 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
11512 @value{GDBN} provides the following task-related commands:
11517 This command shows a list of current Ada tasks, as in the following example:
11524 (@value{GDBP}) info tasks
11525 ID TID P-ID Pri State Name
11526 1 8088000 0 15 Child Activation Wait main_task
11527 2 80a4000 1 15 Accept Statement b
11528 3 809a800 1 15 Child Activation Wait a
11529 * 4 80ae800 3 15 Running c
11534 In this listing, the asterisk before the last task indicates it to be the
11535 task currently being inspected.
11539 Represents @value{GDBN}'s internal task number.
11545 The parent's task ID (@value{GDBN}'s internal task number).
11548 The base priority of the task.
11551 Current state of the task.
11555 The task has been created but has not been activated. It cannot be
11559 The task currently running.
11562 The task is not blocked for any reason known to Ada. (It may be waiting
11563 for a mutex, though.) It is conceptually "executing" in normal mode.
11566 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
11567 that were waiting on terminate alternatives have been awakened and have
11568 terminated themselves.
11570 @item Child Activation Wait
11571 The task is waiting for created tasks to complete activation.
11573 @item Accept Statement
11574 The task is waiting on an accept or selective wait statement.
11576 @item Waiting on entry call
11577 The task is waiting on an entry call.
11579 @item Async Select Wait
11580 The task is waiting to start the abortable part of an asynchronous
11584 The task is waiting on a select statement with only a delay
11587 @item Child Termination Wait
11588 The task is sleeping having completed a master within itself, and is
11589 waiting for the tasks dependent on that master to become terminated or
11590 waiting on a terminate Phase.
11592 @item Wait Child in Term Alt
11593 The task is sleeping waiting for tasks on terminate alternatives to
11594 finish terminating.
11596 @item Accepting RV with @var{taskno}
11597 The task is accepting a rendez-vous with the task @var{taskno}.
11601 Name of the task in the program.
11605 @kindex info task @var{taskno}
11606 @item info task @var{taskno}
11607 This command shows detailled informations on the specified task, as in
11608 the following example:
11613 (@value{GDBP}) info tasks
11614 ID TID P-ID Pri State Name
11615 1 8077880 0 15 Child Activation Wait main_task
11616 * 2 807c468 1 15 Running task_1
11617 (@value{GDBP}) info task 2
11618 Ada Task: 0x807c468
11621 Parent: 1 (main_task)
11627 @kindex task@r{ (Ada)}
11628 @cindex current Ada task ID
11629 This command prints the ID of the current task.
11635 (@value{GDBP}) info tasks
11636 ID TID P-ID Pri State Name
11637 1 8077870 0 15 Child Activation Wait main_task
11638 * 2 807c458 1 15 Running t
11639 (@value{GDBP}) task
11640 [Current task is 2]
11643 @item task @var{taskno}
11644 @cindex Ada task switching
11645 This command is like the @code{thread @var{threadno}}
11646 command (@pxref{Threads}). It switches the context of debugging
11647 from the current task to the given task.
11653 (@value{GDBP}) info tasks
11654 ID TID P-ID Pri State Name
11655 1 8077870 0 15 Child Activation Wait main_task
11656 * 2 807c458 1 15 Running t
11657 (@value{GDBP}) task 1
11658 [Switching to task 1]
11659 #0 0x8067726 in pthread_cond_wait ()
11661 #0 0x8067726 in pthread_cond_wait ()
11662 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
11663 #2 0x805cb63 in system.task_primitives.operations.sleep ()
11664 #3 0x806153e in system.tasking.stages.activate_tasks ()
11665 #4 0x804aacc in un () at un.adb:5
11670 @node Ada Tasks and Core Files
11671 @subsubsection Tasking Support when Debugging Core Files
11672 @cindex Ada tasking and core file debugging
11674 When inspecting a core file, as opposed to debugging a live program,
11675 tasking support may be limited or even unavailable, depending on
11676 the platform being used.
11677 For instance, on x86-linux, the list of tasks is available, but task
11678 switching is not supported. On Tru64, however, task switching will work
11681 On certain platforms, including Tru64, the debugger needs to perform some
11682 memory writes in order to provide Ada tasking support. When inspecting
11683 a core file, this means that the core file must be opened with read-write
11684 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
11685 Under these circumstances, you should make a backup copy of the core
11686 file before inspecting it with @value{GDBN}.
11689 @subsubsection Known Peculiarities of Ada Mode
11690 @cindex Ada, problems
11692 Besides the omissions listed previously (@pxref{Omissions from Ada}),
11693 we know of several problems with and limitations of Ada mode in
11695 some of which will be fixed with planned future releases of the debugger
11696 and the GNU Ada compiler.
11700 Currently, the debugger
11701 has insufficient information to determine whether certain pointers represent
11702 pointers to objects or the objects themselves.
11703 Thus, the user may have to tack an extra @code{.all} after an expression
11704 to get it printed properly.
11707 Static constants that the compiler chooses not to materialize as objects in
11708 storage are invisible to the debugger.
11711 Named parameter associations in function argument lists are ignored (the
11712 argument lists are treated as positional).
11715 Many useful library packages are currently invisible to the debugger.
11718 Fixed-point arithmetic, conversions, input, and output is carried out using
11719 floating-point arithmetic, and may give results that only approximate those on
11723 The GNAT compiler never generates the prefix @code{Standard} for any of
11724 the standard symbols defined by the Ada language. @value{GDBN} knows about
11725 this: it will strip the prefix from names when you use it, and will never
11726 look for a name you have so qualified among local symbols, nor match against
11727 symbols in other packages or subprograms. If you have
11728 defined entities anywhere in your program other than parameters and
11729 local variables whose simple names match names in @code{Standard},
11730 GNAT's lack of qualification here can cause confusion. When this happens,
11731 you can usually resolve the confusion
11732 by qualifying the problematic names with package
11733 @code{Standard} explicitly.
11736 @node Unsupported Languages
11737 @section Unsupported Languages
11739 @cindex unsupported languages
11740 @cindex minimal language
11741 In addition to the other fully-supported programming languages,
11742 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
11743 It does not represent a real programming language, but provides a set
11744 of capabilities close to what the C or assembly languages provide.
11745 This should allow most simple operations to be performed while debugging
11746 an application that uses a language currently not supported by @value{GDBN}.
11748 If the language is set to @code{auto}, @value{GDBN} will automatically
11749 select this language if the current frame corresponds to an unsupported
11753 @chapter Examining the Symbol Table
11755 The commands described in this chapter allow you to inquire about the
11756 symbols (names of variables, functions and types) defined in your
11757 program. This information is inherent in the text of your program and
11758 does not change as your program executes. @value{GDBN} finds it in your
11759 program's symbol table, in the file indicated when you started @value{GDBN}
11760 (@pxref{File Options, ,Choosing Files}), or by one of the
11761 file-management commands (@pxref{Files, ,Commands to Specify Files}).
11763 @cindex symbol names
11764 @cindex names of symbols
11765 @cindex quoting names
11766 Occasionally, you may need to refer to symbols that contain unusual
11767 characters, which @value{GDBN} ordinarily treats as word delimiters. The
11768 most frequent case is in referring to static variables in other
11769 source files (@pxref{Variables,,Program Variables}). File names
11770 are recorded in object files as debugging symbols, but @value{GDBN} would
11771 ordinarily parse a typical file name, like @file{foo.c}, as the three words
11772 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
11773 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
11780 looks up the value of @code{x} in the scope of the file @file{foo.c}.
11783 @cindex case-insensitive symbol names
11784 @cindex case sensitivity in symbol names
11785 @kindex set case-sensitive
11786 @item set case-sensitive on
11787 @itemx set case-sensitive off
11788 @itemx set case-sensitive auto
11789 Normally, when @value{GDBN} looks up symbols, it matches their names
11790 with case sensitivity determined by the current source language.
11791 Occasionally, you may wish to control that. The command @code{set
11792 case-sensitive} lets you do that by specifying @code{on} for
11793 case-sensitive matches or @code{off} for case-insensitive ones. If
11794 you specify @code{auto}, case sensitivity is reset to the default
11795 suitable for the source language. The default is case-sensitive
11796 matches for all languages except for Fortran, for which the default is
11797 case-insensitive matches.
11799 @kindex show case-sensitive
11800 @item show case-sensitive
11801 This command shows the current setting of case sensitivity for symbols
11804 @kindex info address
11805 @cindex address of a symbol
11806 @item info address @var{symbol}
11807 Describe where the data for @var{symbol} is stored. For a register
11808 variable, this says which register it is kept in. For a non-register
11809 local variable, this prints the stack-frame offset at which the variable
11812 Note the contrast with @samp{print &@var{symbol}}, which does not work
11813 at all for a register variable, and for a stack local variable prints
11814 the exact address of the current instantiation of the variable.
11816 @kindex info symbol
11817 @cindex symbol from address
11818 @cindex closest symbol and offset for an address
11819 @item info symbol @var{addr}
11820 Print the name of a symbol which is stored at the address @var{addr}.
11821 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
11822 nearest symbol and an offset from it:
11825 (@value{GDBP}) info symbol 0x54320
11826 _initialize_vx + 396 in section .text
11830 This is the opposite of the @code{info address} command. You can use
11831 it to find out the name of a variable or a function given its address.
11833 For dynamically linked executables, the name of executable or shared
11834 library containing the symbol is also printed:
11837 (@value{GDBP}) info symbol 0x400225
11838 _start + 5 in section .text of /tmp/a.out
11839 (@value{GDBP}) info symbol 0x2aaaac2811cf
11840 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
11844 @item whatis [@var{arg}]
11845 Print the data type of @var{arg}, which can be either an expression or
11846 a data type. With no argument, print the data type of @code{$}, the
11847 last value in the value history. If @var{arg} is an expression, it is
11848 not actually evaluated, and any side-effecting operations (such as
11849 assignments or function calls) inside it do not take place. If
11850 @var{arg} is a type name, it may be the name of a type or typedef, or
11851 for C code it may have the form @samp{class @var{class-name}},
11852 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
11853 @samp{enum @var{enum-tag}}.
11854 @xref{Expressions, ,Expressions}.
11857 @item ptype [@var{arg}]
11858 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
11859 detailed description of the type, instead of just the name of the type.
11860 @xref{Expressions, ,Expressions}.
11862 For example, for this variable declaration:
11865 struct complex @{double real; double imag;@} v;
11869 the two commands give this output:
11873 (@value{GDBP}) whatis v
11874 type = struct complex
11875 (@value{GDBP}) ptype v
11876 type = struct complex @{
11884 As with @code{whatis}, using @code{ptype} without an argument refers to
11885 the type of @code{$}, the last value in the value history.
11887 @cindex incomplete type
11888 Sometimes, programs use opaque data types or incomplete specifications
11889 of complex data structure. If the debug information included in the
11890 program does not allow @value{GDBN} to display a full declaration of
11891 the data type, it will say @samp{<incomplete type>}. For example,
11892 given these declarations:
11896 struct foo *fooptr;
11900 but no definition for @code{struct foo} itself, @value{GDBN} will say:
11903 (@value{GDBP}) ptype foo
11904 $1 = <incomplete type>
11908 ``Incomplete type'' is C terminology for data types that are not
11909 completely specified.
11912 @item info types @var{regexp}
11914 Print a brief description of all types whose names match the regular
11915 expression @var{regexp} (or all types in your program, if you supply
11916 no argument). Each complete typename is matched as though it were a
11917 complete line; thus, @samp{i type value} gives information on all
11918 types in your program whose names include the string @code{value}, but
11919 @samp{i type ^value$} gives information only on types whose complete
11920 name is @code{value}.
11922 This command differs from @code{ptype} in two ways: first, like
11923 @code{whatis}, it does not print a detailed description; second, it
11924 lists all source files where a type is defined.
11927 @cindex local variables
11928 @item info scope @var{location}
11929 List all the variables local to a particular scope. This command
11930 accepts a @var{location} argument---a function name, a source line, or
11931 an address preceded by a @samp{*}, and prints all the variables local
11932 to the scope defined by that location. (@xref{Specify Location}, for
11933 details about supported forms of @var{location}.) For example:
11936 (@value{GDBP}) @b{info scope command_line_handler}
11937 Scope for command_line_handler:
11938 Symbol rl is an argument at stack/frame offset 8, length 4.
11939 Symbol linebuffer is in static storage at address 0x150a18, length 4.
11940 Symbol linelength is in static storage at address 0x150a1c, length 4.
11941 Symbol p is a local variable in register $esi, length 4.
11942 Symbol p1 is a local variable in register $ebx, length 4.
11943 Symbol nline is a local variable in register $edx, length 4.
11944 Symbol repeat is a local variable at frame offset -8, length 4.
11948 This command is especially useful for determining what data to collect
11949 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
11952 @kindex info source
11954 Show information about the current source file---that is, the source file for
11955 the function containing the current point of execution:
11958 the name of the source file, and the directory containing it,
11960 the directory it was compiled in,
11962 its length, in lines,
11964 which programming language it is written in,
11966 whether the executable includes debugging information for that file, and
11967 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
11969 whether the debugging information includes information about
11970 preprocessor macros.
11974 @kindex info sources
11976 Print the names of all source files in your program for which there is
11977 debugging information, organized into two lists: files whose symbols
11978 have already been read, and files whose symbols will be read when needed.
11980 @kindex info functions
11981 @item info functions
11982 Print the names and data types of all defined functions.
11984 @item info functions @var{regexp}
11985 Print the names and data types of all defined functions
11986 whose names contain a match for regular expression @var{regexp}.
11987 Thus, @samp{info fun step} finds all functions whose names
11988 include @code{step}; @samp{info fun ^step} finds those whose names
11989 start with @code{step}. If a function name contains characters
11990 that conflict with the regular expression language (e.g.@:
11991 @samp{operator*()}), they may be quoted with a backslash.
11993 @kindex info variables
11994 @item info variables
11995 Print the names and data types of all variables that are declared
11996 outside of functions (i.e.@: excluding local variables).
11998 @item info variables @var{regexp}
11999 Print the names and data types of all variables (except for local
12000 variables) whose names contain a match for regular expression
12003 @kindex info classes
12004 @cindex Objective-C, classes and selectors
12006 @itemx info classes @var{regexp}
12007 Display all Objective-C classes in your program, or
12008 (with the @var{regexp} argument) all those matching a particular regular
12011 @kindex info selectors
12012 @item info selectors
12013 @itemx info selectors @var{regexp}
12014 Display all Objective-C selectors in your program, or
12015 (with the @var{regexp} argument) all those matching a particular regular
12019 This was never implemented.
12020 @kindex info methods
12022 @itemx info methods @var{regexp}
12023 The @code{info methods} command permits the user to examine all defined
12024 methods within C@t{++} program, or (with the @var{regexp} argument) a
12025 specific set of methods found in the various C@t{++} classes. Many
12026 C@t{++} classes provide a large number of methods. Thus, the output
12027 from the @code{ptype} command can be overwhelming and hard to use. The
12028 @code{info-methods} command filters the methods, printing only those
12029 which match the regular-expression @var{regexp}.
12032 @cindex reloading symbols
12033 Some systems allow individual object files that make up your program to
12034 be replaced without stopping and restarting your program. For example,
12035 in VxWorks you can simply recompile a defective object file and keep on
12036 running. If you are running on one of these systems, you can allow
12037 @value{GDBN} to reload the symbols for automatically relinked modules:
12040 @kindex set symbol-reloading
12041 @item set symbol-reloading on
12042 Replace symbol definitions for the corresponding source file when an
12043 object file with a particular name is seen again.
12045 @item set symbol-reloading off
12046 Do not replace symbol definitions when encountering object files of the
12047 same name more than once. This is the default state; if you are not
12048 running on a system that permits automatic relinking of modules, you
12049 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
12050 may discard symbols when linking large programs, that may contain
12051 several modules (from different directories or libraries) with the same
12054 @kindex show symbol-reloading
12055 @item show symbol-reloading
12056 Show the current @code{on} or @code{off} setting.
12059 @cindex opaque data types
12060 @kindex set opaque-type-resolution
12061 @item set opaque-type-resolution on
12062 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
12063 declared as a pointer to a @code{struct}, @code{class}, or
12064 @code{union}---for example, @code{struct MyType *}---that is used in one
12065 source file although the full declaration of @code{struct MyType} is in
12066 another source file. The default is on.
12068 A change in the setting of this subcommand will not take effect until
12069 the next time symbols for a file are loaded.
12071 @item set opaque-type-resolution off
12072 Tell @value{GDBN} not to resolve opaque types. In this case, the type
12073 is printed as follows:
12075 @{<no data fields>@}
12078 @kindex show opaque-type-resolution
12079 @item show opaque-type-resolution
12080 Show whether opaque types are resolved or not.
12082 @kindex set print symbol-loading
12083 @cindex print messages when symbols are loaded
12084 @item set print symbol-loading
12085 @itemx set print symbol-loading on
12086 @itemx set print symbol-loading off
12087 The @code{set print symbol-loading} command allows you to enable or
12088 disable printing of messages when @value{GDBN} loads symbols.
12089 By default, these messages will be printed, and normally this is what
12090 you want. Disabling these messages is useful when debugging applications
12091 with lots of shared libraries where the quantity of output can be more
12092 annoying than useful.
12094 @kindex show print symbol-loading
12095 @item show print symbol-loading
12096 Show whether messages will be printed when @value{GDBN} loads symbols.
12098 @kindex maint print symbols
12099 @cindex symbol dump
12100 @kindex maint print psymbols
12101 @cindex partial symbol dump
12102 @item maint print symbols @var{filename}
12103 @itemx maint print psymbols @var{filename}
12104 @itemx maint print msymbols @var{filename}
12105 Write a dump of debugging symbol data into the file @var{filename}.
12106 These commands are used to debug the @value{GDBN} symbol-reading code. Only
12107 symbols with debugging data are included. If you use @samp{maint print
12108 symbols}, @value{GDBN} includes all the symbols for which it has already
12109 collected full details: that is, @var{filename} reflects symbols for
12110 only those files whose symbols @value{GDBN} has read. You can use the
12111 command @code{info sources} to find out which files these are. If you
12112 use @samp{maint print psymbols} instead, the dump shows information about
12113 symbols that @value{GDBN} only knows partially---that is, symbols defined in
12114 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
12115 @samp{maint print msymbols} dumps just the minimal symbol information
12116 required for each object file from which @value{GDBN} has read some symbols.
12117 @xref{Files, ,Commands to Specify Files}, for a discussion of how
12118 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
12120 @kindex maint info symtabs
12121 @kindex maint info psymtabs
12122 @cindex listing @value{GDBN}'s internal symbol tables
12123 @cindex symbol tables, listing @value{GDBN}'s internal
12124 @cindex full symbol tables, listing @value{GDBN}'s internal
12125 @cindex partial symbol tables, listing @value{GDBN}'s internal
12126 @item maint info symtabs @r{[} @var{regexp} @r{]}
12127 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
12129 List the @code{struct symtab} or @code{struct partial_symtab}
12130 structures whose names match @var{regexp}. If @var{regexp} is not
12131 given, list them all. The output includes expressions which you can
12132 copy into a @value{GDBN} debugging this one to examine a particular
12133 structure in more detail. For example:
12136 (@value{GDBP}) maint info psymtabs dwarf2read
12137 @{ objfile /home/gnu/build/gdb/gdb
12138 ((struct objfile *) 0x82e69d0)
12139 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
12140 ((struct partial_symtab *) 0x8474b10)
12143 text addresses 0x814d3c8 -- 0x8158074
12144 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
12145 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
12146 dependencies (none)
12149 (@value{GDBP}) maint info symtabs
12153 We see that there is one partial symbol table whose filename contains
12154 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
12155 and we see that @value{GDBN} has not read in any symtabs yet at all.
12156 If we set a breakpoint on a function, that will cause @value{GDBN} to
12157 read the symtab for the compilation unit containing that function:
12160 (@value{GDBP}) break dwarf2_psymtab_to_symtab
12161 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
12163 (@value{GDBP}) maint info symtabs
12164 @{ objfile /home/gnu/build/gdb/gdb
12165 ((struct objfile *) 0x82e69d0)
12166 @{ symtab /home/gnu/src/gdb/dwarf2read.c
12167 ((struct symtab *) 0x86c1f38)
12170 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
12171 linetable ((struct linetable *) 0x8370fa0)
12172 debugformat DWARF 2
12181 @chapter Altering Execution
12183 Once you think you have found an error in your program, you might want to
12184 find out for certain whether correcting the apparent error would lead to
12185 correct results in the rest of the run. You can find the answer by
12186 experiment, using the @value{GDBN} features for altering execution of the
12189 For example, you can store new values into variables or memory
12190 locations, give your program a signal, restart it at a different
12191 address, or even return prematurely from a function.
12194 * Assignment:: Assignment to variables
12195 * Jumping:: Continuing at a different address
12196 * Signaling:: Giving your program a signal
12197 * Returning:: Returning from a function
12198 * Calling:: Calling your program's functions
12199 * Patching:: Patching your program
12203 @section Assignment to Variables
12206 @cindex setting variables
12207 To alter the value of a variable, evaluate an assignment expression.
12208 @xref{Expressions, ,Expressions}. For example,
12215 stores the value 4 into the variable @code{x}, and then prints the
12216 value of the assignment expression (which is 4).
12217 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
12218 information on operators in supported languages.
12220 @kindex set variable
12221 @cindex variables, setting
12222 If you are not interested in seeing the value of the assignment, use the
12223 @code{set} command instead of the @code{print} command. @code{set} is
12224 really the same as @code{print} except that the expression's value is
12225 not printed and is not put in the value history (@pxref{Value History,
12226 ,Value History}). The expression is evaluated only for its effects.
12228 If the beginning of the argument string of the @code{set} command
12229 appears identical to a @code{set} subcommand, use the @code{set
12230 variable} command instead of just @code{set}. This command is identical
12231 to @code{set} except for its lack of subcommands. For example, if your
12232 program has a variable @code{width}, you get an error if you try to set
12233 a new value with just @samp{set width=13}, because @value{GDBN} has the
12234 command @code{set width}:
12237 (@value{GDBP}) whatis width
12239 (@value{GDBP}) p width
12241 (@value{GDBP}) set width=47
12242 Invalid syntax in expression.
12246 The invalid expression, of course, is @samp{=47}. In
12247 order to actually set the program's variable @code{width}, use
12250 (@value{GDBP}) set var width=47
12253 Because the @code{set} command has many subcommands that can conflict
12254 with the names of program variables, it is a good idea to use the
12255 @code{set variable} command instead of just @code{set}. For example, if
12256 your program has a variable @code{g}, you run into problems if you try
12257 to set a new value with just @samp{set g=4}, because @value{GDBN} has
12258 the command @code{set gnutarget}, abbreviated @code{set g}:
12262 (@value{GDBP}) whatis g
12266 (@value{GDBP}) set g=4
12270 The program being debugged has been started already.
12271 Start it from the beginning? (y or n) y
12272 Starting program: /home/smith/cc_progs/a.out
12273 "/home/smith/cc_progs/a.out": can't open to read symbols:
12274 Invalid bfd target.
12275 (@value{GDBP}) show g
12276 The current BFD target is "=4".
12281 The program variable @code{g} did not change, and you silently set the
12282 @code{gnutarget} to an invalid value. In order to set the variable
12286 (@value{GDBP}) set var g=4
12289 @value{GDBN} allows more implicit conversions in assignments than C; you can
12290 freely store an integer value into a pointer variable or vice versa,
12291 and you can convert any structure to any other structure that is the
12292 same length or shorter.
12293 @comment FIXME: how do structs align/pad in these conversions?
12294 @comment /doc@cygnus.com 18dec1990
12296 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
12297 construct to generate a value of specified type at a specified address
12298 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
12299 to memory location @code{0x83040} as an integer (which implies a certain size
12300 and representation in memory), and
12303 set @{int@}0x83040 = 4
12307 stores the value 4 into that memory location.
12310 @section Continuing at a Different Address
12312 Ordinarily, when you continue your program, you do so at the place where
12313 it stopped, with the @code{continue} command. You can instead continue at
12314 an address of your own choosing, with the following commands:
12318 @item jump @var{linespec}
12319 @itemx jump @var{location}
12320 Resume execution at line @var{linespec} or at address given by
12321 @var{location}. Execution stops again immediately if there is a
12322 breakpoint there. @xref{Specify Location}, for a description of the
12323 different forms of @var{linespec} and @var{location}. It is common
12324 practice to use the @code{tbreak} command in conjunction with
12325 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
12327 The @code{jump} command does not change the current stack frame, or
12328 the stack pointer, or the contents of any memory location or any
12329 register other than the program counter. If line @var{linespec} is in
12330 a different function from the one currently executing, the results may
12331 be bizarre if the two functions expect different patterns of arguments or
12332 of local variables. For this reason, the @code{jump} command requests
12333 confirmation if the specified line is not in the function currently
12334 executing. However, even bizarre results are predictable if you are
12335 well acquainted with the machine-language code of your program.
12338 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
12339 On many systems, you can get much the same effect as the @code{jump}
12340 command by storing a new value into the register @code{$pc}. The
12341 difference is that this does not start your program running; it only
12342 changes the address of where it @emph{will} run when you continue. For
12350 makes the next @code{continue} command or stepping command execute at
12351 address @code{0x485}, rather than at the address where your program stopped.
12352 @xref{Continuing and Stepping, ,Continuing and Stepping}.
12354 The most common occasion to use the @code{jump} command is to back
12355 up---perhaps with more breakpoints set---over a portion of a program
12356 that has already executed, in order to examine its execution in more
12361 @section Giving your Program a Signal
12362 @cindex deliver a signal to a program
12366 @item signal @var{signal}
12367 Resume execution where your program stopped, but immediately give it the
12368 signal @var{signal}. @var{signal} can be the name or the number of a
12369 signal. For example, on many systems @code{signal 2} and @code{signal
12370 SIGINT} are both ways of sending an interrupt signal.
12372 Alternatively, if @var{signal} is zero, continue execution without
12373 giving a signal. This is useful when your program stopped on account of
12374 a signal and would ordinary see the signal when resumed with the
12375 @code{continue} command; @samp{signal 0} causes it to resume without a
12378 @code{signal} does not repeat when you press @key{RET} a second time
12379 after executing the command.
12383 Invoking the @code{signal} command is not the same as invoking the
12384 @code{kill} utility from the shell. Sending a signal with @code{kill}
12385 causes @value{GDBN} to decide what to do with the signal depending on
12386 the signal handling tables (@pxref{Signals}). The @code{signal} command
12387 passes the signal directly to your program.
12391 @section Returning from a Function
12394 @cindex returning from a function
12397 @itemx return @var{expression}
12398 You can cancel execution of a function call with the @code{return}
12399 command. If you give an
12400 @var{expression} argument, its value is used as the function's return
12404 When you use @code{return}, @value{GDBN} discards the selected stack frame
12405 (and all frames within it). You can think of this as making the
12406 discarded frame return prematurely. If you wish to specify a value to
12407 be returned, give that value as the argument to @code{return}.
12409 This pops the selected stack frame (@pxref{Selection, ,Selecting a
12410 Frame}), and any other frames inside of it, leaving its caller as the
12411 innermost remaining frame. That frame becomes selected. The
12412 specified value is stored in the registers used for returning values
12415 The @code{return} command does not resume execution; it leaves the
12416 program stopped in the state that would exist if the function had just
12417 returned. In contrast, the @code{finish} command (@pxref{Continuing
12418 and Stepping, ,Continuing and Stepping}) resumes execution until the
12419 selected stack frame returns naturally.
12422 @section Calling Program Functions
12425 @cindex calling functions
12426 @cindex inferior functions, calling
12427 @item print @var{expr}
12428 Evaluate the expression @var{expr} and display the resulting value.
12429 @var{expr} may include calls to functions in the program being
12433 @item call @var{expr}
12434 Evaluate the expression @var{expr} without displaying @code{void}
12437 You can use this variant of the @code{print} command if you want to
12438 execute a function from your program that does not return anything
12439 (a.k.a.@: @dfn{a void function}), but without cluttering the output
12440 with @code{void} returned values that @value{GDBN} will otherwise
12441 print. If the result is not void, it is printed and saved in the
12445 It is possible for the function you call via the @code{print} or
12446 @code{call} command to generate a signal (e.g., if there's a bug in
12447 the function, or if you passed it incorrect arguments). What happens
12448 in that case is controlled by the @code{set unwindonsignal} command.
12451 @item set unwindonsignal
12452 @kindex set unwindonsignal
12453 @cindex unwind stack in called functions
12454 @cindex call dummy stack unwinding
12455 Set unwinding of the stack if a signal is received while in a function
12456 that @value{GDBN} called in the program being debugged. If set to on,
12457 @value{GDBN} unwinds the stack it created for the call and restores
12458 the context to what it was before the call. If set to off (the
12459 default), @value{GDBN} stops in the frame where the signal was
12462 @item show unwindonsignal
12463 @kindex show unwindonsignal
12464 Show the current setting of stack unwinding in the functions called by
12468 @cindex weak alias functions
12469 Sometimes, a function you wish to call is actually a @dfn{weak alias}
12470 for another function. In such case, @value{GDBN} might not pick up
12471 the type information, including the types of the function arguments,
12472 which causes @value{GDBN} to call the inferior function incorrectly.
12473 As a result, the called function will function erroneously and may
12474 even crash. A solution to that is to use the name of the aliased
12478 @section Patching Programs
12480 @cindex patching binaries
12481 @cindex writing into executables
12482 @cindex writing into corefiles
12484 By default, @value{GDBN} opens the file containing your program's
12485 executable code (or the corefile) read-only. This prevents accidental
12486 alterations to machine code; but it also prevents you from intentionally
12487 patching your program's binary.
12489 If you'd like to be able to patch the binary, you can specify that
12490 explicitly with the @code{set write} command. For example, you might
12491 want to turn on internal debugging flags, or even to make emergency
12497 @itemx set write off
12498 If you specify @samp{set write on}, @value{GDBN} opens executable and
12499 core files for both reading and writing; if you specify @kbd{set write
12500 off} (the default), @value{GDBN} opens them read-only.
12502 If you have already loaded a file, you must load it again (using the
12503 @code{exec-file} or @code{core-file} command) after changing @code{set
12504 write}, for your new setting to take effect.
12508 Display whether executable files and core files are opened for writing
12509 as well as reading.
12513 @chapter @value{GDBN} Files
12515 @value{GDBN} needs to know the file name of the program to be debugged,
12516 both in order to read its symbol table and in order to start your
12517 program. To debug a core dump of a previous run, you must also tell
12518 @value{GDBN} the name of the core dump file.
12521 * Files:: Commands to specify files
12522 * Separate Debug Files:: Debugging information in separate files
12523 * Symbol Errors:: Errors reading symbol files
12527 @section Commands to Specify Files
12529 @cindex symbol table
12530 @cindex core dump file
12532 You may want to specify executable and core dump file names. The usual
12533 way to do this is at start-up time, using the arguments to
12534 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
12535 Out of @value{GDBN}}).
12537 Occasionally it is necessary to change to a different file during a
12538 @value{GDBN} session. Or you may run @value{GDBN} and forget to
12539 specify a file you want to use. Or you are debugging a remote target
12540 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
12541 Program}). In these situations the @value{GDBN} commands to specify
12542 new files are useful.
12545 @cindex executable file
12547 @item file @var{filename}
12548 Use @var{filename} as the program to be debugged. It is read for its
12549 symbols and for the contents of pure memory. It is also the program
12550 executed when you use the @code{run} command. If you do not specify a
12551 directory and the file is not found in the @value{GDBN} working directory,
12552 @value{GDBN} uses the environment variable @code{PATH} as a list of
12553 directories to search, just as the shell does when looking for a program
12554 to run. You can change the value of this variable, for both @value{GDBN}
12555 and your program, using the @code{path} command.
12557 @cindex unlinked object files
12558 @cindex patching object files
12559 You can load unlinked object @file{.o} files into @value{GDBN} using
12560 the @code{file} command. You will not be able to ``run'' an object
12561 file, but you can disassemble functions and inspect variables. Also,
12562 if the underlying BFD functionality supports it, you could use
12563 @kbd{gdb -write} to patch object files using this technique. Note
12564 that @value{GDBN} can neither interpret nor modify relocations in this
12565 case, so branches and some initialized variables will appear to go to
12566 the wrong place. But this feature is still handy from time to time.
12569 @code{file} with no argument makes @value{GDBN} discard any information it
12570 has on both executable file and the symbol table.
12573 @item exec-file @r{[} @var{filename} @r{]}
12574 Specify that the program to be run (but not the symbol table) is found
12575 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
12576 if necessary to locate your program. Omitting @var{filename} means to
12577 discard information on the executable file.
12579 @kindex symbol-file
12580 @item symbol-file @r{[} @var{filename} @r{]}
12581 Read symbol table information from file @var{filename}. @code{PATH} is
12582 searched when necessary. Use the @code{file} command to get both symbol
12583 table and program to run from the same file.
12585 @code{symbol-file} with no argument clears out @value{GDBN} information on your
12586 program's symbol table.
12588 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
12589 some breakpoints and auto-display expressions. This is because they may
12590 contain pointers to the internal data recording symbols and data types,
12591 which are part of the old symbol table data being discarded inside
12594 @code{symbol-file} does not repeat if you press @key{RET} again after
12597 When @value{GDBN} is configured for a particular environment, it
12598 understands debugging information in whatever format is the standard
12599 generated for that environment; you may use either a @sc{gnu} compiler, or
12600 other compilers that adhere to the local conventions.
12601 Best results are usually obtained from @sc{gnu} compilers; for example,
12602 using @code{@value{NGCC}} you can generate debugging information for
12605 For most kinds of object files, with the exception of old SVR3 systems
12606 using COFF, the @code{symbol-file} command does not normally read the
12607 symbol table in full right away. Instead, it scans the symbol table
12608 quickly to find which source files and which symbols are present. The
12609 details are read later, one source file at a time, as they are needed.
12611 The purpose of this two-stage reading strategy is to make @value{GDBN}
12612 start up faster. For the most part, it is invisible except for
12613 occasional pauses while the symbol table details for a particular source
12614 file are being read. (The @code{set verbose} command can turn these
12615 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
12616 Warnings and Messages}.)
12618 We have not implemented the two-stage strategy for COFF yet. When the
12619 symbol table is stored in COFF format, @code{symbol-file} reads the
12620 symbol table data in full right away. Note that ``stabs-in-COFF''
12621 still does the two-stage strategy, since the debug info is actually
12625 @cindex reading symbols immediately
12626 @cindex symbols, reading immediately
12627 @item symbol-file @var{filename} @r{[} -readnow @r{]}
12628 @itemx file @var{filename} @r{[} -readnow @r{]}
12629 You can override the @value{GDBN} two-stage strategy for reading symbol
12630 tables by using the @samp{-readnow} option with any of the commands that
12631 load symbol table information, if you want to be sure @value{GDBN} has the
12632 entire symbol table available.
12634 @c FIXME: for now no mention of directories, since this seems to be in
12635 @c flux. 13mar1992 status is that in theory GDB would look either in
12636 @c current dir or in same dir as myprog; but issues like competing
12637 @c GDB's, or clutter in system dirs, mean that in practice right now
12638 @c only current dir is used. FFish says maybe a special GDB hierarchy
12639 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
12643 @item core-file @r{[}@var{filename}@r{]}
12645 Specify the whereabouts of a core dump file to be used as the ``contents
12646 of memory''. Traditionally, core files contain only some parts of the
12647 address space of the process that generated them; @value{GDBN} can access the
12648 executable file itself for other parts.
12650 @code{core-file} with no argument specifies that no core file is
12653 Note that the core file is ignored when your program is actually running
12654 under @value{GDBN}. So, if you have been running your program and you
12655 wish to debug a core file instead, you must kill the subprocess in which
12656 the program is running. To do this, use the @code{kill} command
12657 (@pxref{Kill Process, ,Killing the Child Process}).
12659 @kindex add-symbol-file
12660 @cindex dynamic linking
12661 @item add-symbol-file @var{filename} @var{address}
12662 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
12663 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
12664 The @code{add-symbol-file} command reads additional symbol table
12665 information from the file @var{filename}. You would use this command
12666 when @var{filename} has been dynamically loaded (by some other means)
12667 into the program that is running. @var{address} should be the memory
12668 address at which the file has been loaded; @value{GDBN} cannot figure
12669 this out for itself. You can additionally specify an arbitrary number
12670 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
12671 section name and base address for that section. You can specify any
12672 @var{address} as an expression.
12674 The symbol table of the file @var{filename} is added to the symbol table
12675 originally read with the @code{symbol-file} command. You can use the
12676 @code{add-symbol-file} command any number of times; the new symbol data
12677 thus read keeps adding to the old. To discard all old symbol data
12678 instead, use the @code{symbol-file} command without any arguments.
12680 @cindex relocatable object files, reading symbols from
12681 @cindex object files, relocatable, reading symbols from
12682 @cindex reading symbols from relocatable object files
12683 @cindex symbols, reading from relocatable object files
12684 @cindex @file{.o} files, reading symbols from
12685 Although @var{filename} is typically a shared library file, an
12686 executable file, or some other object file which has been fully
12687 relocated for loading into a process, you can also load symbolic
12688 information from relocatable @file{.o} files, as long as:
12692 the file's symbolic information refers only to linker symbols defined in
12693 that file, not to symbols defined by other object files,
12695 every section the file's symbolic information refers to has actually
12696 been loaded into the inferior, as it appears in the file, and
12698 you can determine the address at which every section was loaded, and
12699 provide these to the @code{add-symbol-file} command.
12703 Some embedded operating systems, like Sun Chorus and VxWorks, can load
12704 relocatable files into an already running program; such systems
12705 typically make the requirements above easy to meet. However, it's
12706 important to recognize that many native systems use complex link
12707 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
12708 assembly, for example) that make the requirements difficult to meet. In
12709 general, one cannot assume that using @code{add-symbol-file} to read a
12710 relocatable object file's symbolic information will have the same effect
12711 as linking the relocatable object file into the program in the normal
12714 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
12716 @kindex add-symbol-file-from-memory
12717 @cindex @code{syscall DSO}
12718 @cindex load symbols from memory
12719 @item add-symbol-file-from-memory @var{address}
12720 Load symbols from the given @var{address} in a dynamically loaded
12721 object file whose image is mapped directly into the inferior's memory.
12722 For example, the Linux kernel maps a @code{syscall DSO} into each
12723 process's address space; this DSO provides kernel-specific code for
12724 some system calls. The argument can be any expression whose
12725 evaluation yields the address of the file's shared object file header.
12726 For this command to work, you must have used @code{symbol-file} or
12727 @code{exec-file} commands in advance.
12729 @kindex add-shared-symbol-files
12731 @item add-shared-symbol-files @var{library-file}
12732 @itemx assf @var{library-file}
12733 The @code{add-shared-symbol-files} command can currently be used only
12734 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
12735 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
12736 @value{GDBN} automatically looks for shared libraries, however if
12737 @value{GDBN} does not find yours, you can invoke
12738 @code{add-shared-symbol-files}. It takes one argument: the shared
12739 library's file name. @code{assf} is a shorthand alias for
12740 @code{add-shared-symbol-files}.
12743 @item section @var{section} @var{addr}
12744 The @code{section} command changes the base address of the named
12745 @var{section} of the exec file to @var{addr}. This can be used if the
12746 exec file does not contain section addresses, (such as in the
12747 @code{a.out} format), or when the addresses specified in the file
12748 itself are wrong. Each section must be changed separately. The
12749 @code{info files} command, described below, lists all the sections and
12753 @kindex info target
12756 @code{info files} and @code{info target} are synonymous; both print the
12757 current target (@pxref{Targets, ,Specifying a Debugging Target}),
12758 including the names of the executable and core dump files currently in
12759 use by @value{GDBN}, and the files from which symbols were loaded. The
12760 command @code{help target} lists all possible targets rather than
12763 @kindex maint info sections
12764 @item maint info sections
12765 Another command that can give you extra information about program sections
12766 is @code{maint info sections}. In addition to the section information
12767 displayed by @code{info files}, this command displays the flags and file
12768 offset of each section in the executable and core dump files. In addition,
12769 @code{maint info sections} provides the following command options (which
12770 may be arbitrarily combined):
12774 Display sections for all loaded object files, including shared libraries.
12775 @item @var{sections}
12776 Display info only for named @var{sections}.
12777 @item @var{section-flags}
12778 Display info only for sections for which @var{section-flags} are true.
12779 The section flags that @value{GDBN} currently knows about are:
12782 Section will have space allocated in the process when loaded.
12783 Set for all sections except those containing debug information.
12785 Section will be loaded from the file into the child process memory.
12786 Set for pre-initialized code and data, clear for @code{.bss} sections.
12788 Section needs to be relocated before loading.
12790 Section cannot be modified by the child process.
12792 Section contains executable code only.
12794 Section contains data only (no executable code).
12796 Section will reside in ROM.
12798 Section contains data for constructor/destructor lists.
12800 Section is not empty.
12802 An instruction to the linker to not output the section.
12803 @item COFF_SHARED_LIBRARY
12804 A notification to the linker that the section contains
12805 COFF shared library information.
12807 Section contains common symbols.
12810 @kindex set trust-readonly-sections
12811 @cindex read-only sections
12812 @item set trust-readonly-sections on
12813 Tell @value{GDBN} that readonly sections in your object file
12814 really are read-only (i.e.@: that their contents will not change).
12815 In that case, @value{GDBN} can fetch values from these sections
12816 out of the object file, rather than from the target program.
12817 For some targets (notably embedded ones), this can be a significant
12818 enhancement to debugging performance.
12820 The default is off.
12822 @item set trust-readonly-sections off
12823 Tell @value{GDBN} not to trust readonly sections. This means that
12824 the contents of the section might change while the program is running,
12825 and must therefore be fetched from the target when needed.
12827 @item show trust-readonly-sections
12828 Show the current setting of trusting readonly sections.
12831 All file-specifying commands allow both absolute and relative file names
12832 as arguments. @value{GDBN} always converts the file name to an absolute file
12833 name and remembers it that way.
12835 @cindex shared libraries
12836 @anchor{Shared Libraries}
12837 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
12838 and IBM RS/6000 AIX shared libraries.
12840 On MS-Windows @value{GDBN} must be linked with the Expat library to support
12841 shared libraries. @xref{Expat}.
12843 @value{GDBN} automatically loads symbol definitions from shared libraries
12844 when you use the @code{run} command, or when you examine a core file.
12845 (Before you issue the @code{run} command, @value{GDBN} does not understand
12846 references to a function in a shared library, however---unless you are
12847 debugging a core file).
12849 On HP-UX, if the program loads a library explicitly, @value{GDBN}
12850 automatically loads the symbols at the time of the @code{shl_load} call.
12852 @c FIXME: some @value{GDBN} release may permit some refs to undef
12853 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
12854 @c FIXME...lib; check this from time to time when updating manual
12856 There are times, however, when you may wish to not automatically load
12857 symbol definitions from shared libraries, such as when they are
12858 particularly large or there are many of them.
12860 To control the automatic loading of shared library symbols, use the
12864 @kindex set auto-solib-add
12865 @item set auto-solib-add @var{mode}
12866 If @var{mode} is @code{on}, symbols from all shared object libraries
12867 will be loaded automatically when the inferior begins execution, you
12868 attach to an independently started inferior, or when the dynamic linker
12869 informs @value{GDBN} that a new library has been loaded. If @var{mode}
12870 is @code{off}, symbols must be loaded manually, using the
12871 @code{sharedlibrary} command. The default value is @code{on}.
12873 @cindex memory used for symbol tables
12874 If your program uses lots of shared libraries with debug info that
12875 takes large amounts of memory, you can decrease the @value{GDBN}
12876 memory footprint by preventing it from automatically loading the
12877 symbols from shared libraries. To that end, type @kbd{set
12878 auto-solib-add off} before running the inferior, then load each
12879 library whose debug symbols you do need with @kbd{sharedlibrary
12880 @var{regexp}}, where @var{regexp} is a regular expression that matches
12881 the libraries whose symbols you want to be loaded.
12883 @kindex show auto-solib-add
12884 @item show auto-solib-add
12885 Display the current autoloading mode.
12888 @cindex load shared library
12889 To explicitly load shared library symbols, use the @code{sharedlibrary}
12893 @kindex info sharedlibrary
12896 @itemx info sharedlibrary
12897 Print the names of the shared libraries which are currently loaded.
12899 @kindex sharedlibrary
12901 @item sharedlibrary @var{regex}
12902 @itemx share @var{regex}
12903 Load shared object library symbols for files matching a
12904 Unix regular expression.
12905 As with files loaded automatically, it only loads shared libraries
12906 required by your program for a core file or after typing @code{run}. If
12907 @var{regex} is omitted all shared libraries required by your program are
12910 @item nosharedlibrary
12911 @kindex nosharedlibrary
12912 @cindex unload symbols from shared libraries
12913 Unload all shared object library symbols. This discards all symbols
12914 that have been loaded from all shared libraries. Symbols from shared
12915 libraries that were loaded by explicit user requests are not
12919 Sometimes you may wish that @value{GDBN} stops and gives you control
12920 when any of shared library events happen. Use the @code{set
12921 stop-on-solib-events} command for this:
12924 @item set stop-on-solib-events
12925 @kindex set stop-on-solib-events
12926 This command controls whether @value{GDBN} should give you control
12927 when the dynamic linker notifies it about some shared library event.
12928 The most common event of interest is loading or unloading of a new
12931 @item show stop-on-solib-events
12932 @kindex show stop-on-solib-events
12933 Show whether @value{GDBN} stops and gives you control when shared
12934 library events happen.
12937 Shared libraries are also supported in many cross or remote debugging
12938 configurations. @value{GDBN} needs to have access to the target's libraries;
12939 this can be accomplished either by providing copies of the libraries
12940 on the host system, or by asking @value{GDBN} to automatically retrieve the
12941 libraries from the target. If copies of the target libraries are
12942 provided, they need to be the same as the target libraries, although the
12943 copies on the target can be stripped as long as the copies on the host are
12946 @cindex where to look for shared libraries
12947 For remote debugging, you need to tell @value{GDBN} where the target
12948 libraries are, so that it can load the correct copies---otherwise, it
12949 may try to load the host's libraries. @value{GDBN} has two variables
12950 to specify the search directories for target libraries.
12953 @cindex prefix for shared library file names
12954 @cindex system root, alternate
12955 @kindex set solib-absolute-prefix
12956 @kindex set sysroot
12957 @item set sysroot @var{path}
12958 Use @var{path} as the system root for the program being debugged. Any
12959 absolute shared library paths will be prefixed with @var{path}; many
12960 runtime loaders store the absolute paths to the shared library in the
12961 target program's memory. If you use @code{set sysroot} to find shared
12962 libraries, they need to be laid out in the same way that they are on
12963 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
12966 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
12967 retrieve the target libraries from the remote system. This is only
12968 supported when using a remote target that supports the @code{remote get}
12969 command (@pxref{File Transfer,,Sending files to a remote system}).
12970 The part of @var{path} following the initial @file{remote:}
12971 (if present) is used as system root prefix on the remote file system.
12972 @footnote{If you want to specify a local system root using a directory
12973 that happens to be named @file{remote:}, you need to use some equivalent
12974 variant of the name like @file{./remote:}.}
12976 The @code{set solib-absolute-prefix} command is an alias for @code{set
12979 @cindex default system root
12980 @cindex @samp{--with-sysroot}
12981 You can set the default system root by using the configure-time
12982 @samp{--with-sysroot} option. If the system root is inside
12983 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
12984 @samp{--exec-prefix}), then the default system root will be updated
12985 automatically if the installed @value{GDBN} is moved to a new
12988 @kindex show sysroot
12990 Display the current shared library prefix.
12992 @kindex set solib-search-path
12993 @item set solib-search-path @var{path}
12994 If this variable is set, @var{path} is a colon-separated list of
12995 directories to search for shared libraries. @samp{solib-search-path}
12996 is used after @samp{sysroot} fails to locate the library, or if the
12997 path to the library is relative instead of absolute. If you want to
12998 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
12999 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
13000 finding your host's libraries. @samp{sysroot} is preferred; setting
13001 it to a nonexistent directory may interfere with automatic loading
13002 of shared library symbols.
13004 @kindex show solib-search-path
13005 @item show solib-search-path
13006 Display the current shared library search path.
13010 @node Separate Debug Files
13011 @section Debugging Information in Separate Files
13012 @cindex separate debugging information files
13013 @cindex debugging information in separate files
13014 @cindex @file{.debug} subdirectories
13015 @cindex debugging information directory, global
13016 @cindex global debugging information directory
13017 @cindex build ID, and separate debugging files
13018 @cindex @file{.build-id} directory
13020 @value{GDBN} allows you to put a program's debugging information in a
13021 file separate from the executable itself, in a way that allows
13022 @value{GDBN} to find and load the debugging information automatically.
13023 Since debugging information can be very large---sometimes larger
13024 than the executable code itself---some systems distribute debugging
13025 information for their executables in separate files, which users can
13026 install only when they need to debug a problem.
13028 @value{GDBN} supports two ways of specifying the separate debug info
13033 The executable contains a @dfn{debug link} that specifies the name of
13034 the separate debug info file. The separate debug file's name is
13035 usually @file{@var{executable}.debug}, where @var{executable} is the
13036 name of the corresponding executable file without leading directories
13037 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
13038 debug link specifies a CRC32 checksum for the debug file, which
13039 @value{GDBN} uses to validate that the executable and the debug file
13040 came from the same build.
13043 The executable contains a @dfn{build ID}, a unique bit string that is
13044 also present in the corresponding debug info file. (This is supported
13045 only on some operating systems, notably those which use the ELF format
13046 for binary files and the @sc{gnu} Binutils.) For more details about
13047 this feature, see the description of the @option{--build-id}
13048 command-line option in @ref{Options, , Command Line Options, ld.info,
13049 The GNU Linker}. The debug info file's name is not specified
13050 explicitly by the build ID, but can be computed from the build ID, see
13054 Depending on the way the debug info file is specified, @value{GDBN}
13055 uses two different methods of looking for the debug file:
13059 For the ``debug link'' method, @value{GDBN} looks up the named file in
13060 the directory of the executable file, then in a subdirectory of that
13061 directory named @file{.debug}, and finally under the global debug
13062 directory, in a subdirectory whose name is identical to the leading
13063 directories of the executable's absolute file name.
13066 For the ``build ID'' method, @value{GDBN} looks in the
13067 @file{.build-id} subdirectory of the global debug directory for a file
13068 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
13069 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
13070 are the rest of the bit string. (Real build ID strings are 32 or more
13071 hex characters, not 10.)
13074 So, for example, suppose you ask @value{GDBN} to debug
13075 @file{/usr/bin/ls}, which has a debug link that specifies the
13076 file @file{ls.debug}, and a build ID whose value in hex is
13077 @code{abcdef1234}. If the global debug directory is
13078 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
13079 debug information files, in the indicated order:
13083 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
13085 @file{/usr/bin/ls.debug}
13087 @file{/usr/bin/.debug/ls.debug}
13089 @file{/usr/lib/debug/usr/bin/ls.debug}.
13092 You can set the global debugging info directory's name, and view the
13093 name @value{GDBN} is currently using.
13097 @kindex set debug-file-directory
13098 @item set debug-file-directory @var{directory}
13099 Set the directory which @value{GDBN} searches for separate debugging
13100 information files to @var{directory}.
13102 @kindex show debug-file-directory
13103 @item show debug-file-directory
13104 Show the directory @value{GDBN} searches for separate debugging
13109 @cindex @code{.gnu_debuglink} sections
13110 @cindex debug link sections
13111 A debug link is a special section of the executable file named
13112 @code{.gnu_debuglink}. The section must contain:
13116 A filename, with any leading directory components removed, followed by
13119 zero to three bytes of padding, as needed to reach the next four-byte
13120 boundary within the section, and
13122 a four-byte CRC checksum, stored in the same endianness used for the
13123 executable file itself. The checksum is computed on the debugging
13124 information file's full contents by the function given below, passing
13125 zero as the @var{crc} argument.
13128 Any executable file format can carry a debug link, as long as it can
13129 contain a section named @code{.gnu_debuglink} with the contents
13132 @cindex @code{.note.gnu.build-id} sections
13133 @cindex build ID sections
13134 The build ID is a special section in the executable file (and in other
13135 ELF binary files that @value{GDBN} may consider). This section is
13136 often named @code{.note.gnu.build-id}, but that name is not mandatory.
13137 It contains unique identification for the built files---the ID remains
13138 the same across multiple builds of the same build tree. The default
13139 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
13140 content for the build ID string. The same section with an identical
13141 value is present in the original built binary with symbols, in its
13142 stripped variant, and in the separate debugging information file.
13144 The debugging information file itself should be an ordinary
13145 executable, containing a full set of linker symbols, sections, and
13146 debugging information. The sections of the debugging information file
13147 should have the same names, addresses, and sizes as the original file,
13148 but they need not contain any data---much like a @code{.bss} section
13149 in an ordinary executable.
13151 The @sc{gnu} binary utilities (Binutils) package includes the
13152 @samp{objcopy} utility that can produce
13153 the separated executable / debugging information file pairs using the
13154 following commands:
13157 @kbd{objcopy --only-keep-debug foo foo.debug}
13162 These commands remove the debugging
13163 information from the executable file @file{foo} and place it in the file
13164 @file{foo.debug}. You can use the first, second or both methods to link the
13169 The debug link method needs the following additional command to also leave
13170 behind a debug link in @file{foo}:
13173 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
13176 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
13177 a version of the @code{strip} command such that the command @kbd{strip foo -f
13178 foo.debug} has the same functionality as the two @code{objcopy} commands and
13179 the @code{ln -s} command above, together.
13182 Build ID gets embedded into the main executable using @code{ld --build-id} or
13183 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
13184 compatibility fixes for debug files separation are present in @sc{gnu} binary
13185 utilities (Binutils) package since version 2.18.
13190 Since there are many different ways to compute CRC's for the debug
13191 link (different polynomials, reversals, byte ordering, etc.), the
13192 simplest way to describe the CRC used in @code{.gnu_debuglink}
13193 sections is to give the complete code for a function that computes it:
13195 @kindex gnu_debuglink_crc32
13198 gnu_debuglink_crc32 (unsigned long crc,
13199 unsigned char *buf, size_t len)
13201 static const unsigned long crc32_table[256] =
13203 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
13204 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
13205 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
13206 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
13207 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
13208 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
13209 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
13210 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
13211 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
13212 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
13213 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
13214 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
13215 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
13216 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
13217 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
13218 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
13219 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
13220 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
13221 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
13222 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
13223 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
13224 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
13225 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
13226 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
13227 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
13228 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
13229 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
13230 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
13231 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
13232 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
13233 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
13234 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
13235 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
13236 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
13237 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
13238 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
13239 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
13240 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
13241 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
13242 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
13243 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
13244 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
13245 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
13246 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
13247 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
13248 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
13249 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
13250 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
13251 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
13252 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
13253 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
13256 unsigned char *end;
13258 crc = ~crc & 0xffffffff;
13259 for (end = buf + len; buf < end; ++buf)
13260 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
13261 return ~crc & 0xffffffff;
13266 This computation does not apply to the ``build ID'' method.
13269 @node Symbol Errors
13270 @section Errors Reading Symbol Files
13272 While reading a symbol file, @value{GDBN} occasionally encounters problems,
13273 such as symbol types it does not recognize, or known bugs in compiler
13274 output. By default, @value{GDBN} does not notify you of such problems, since
13275 they are relatively common and primarily of interest to people
13276 debugging compilers. If you are interested in seeing information
13277 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
13278 only one message about each such type of problem, no matter how many
13279 times the problem occurs; or you can ask @value{GDBN} to print more messages,
13280 to see how many times the problems occur, with the @code{set
13281 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
13284 The messages currently printed, and their meanings, include:
13287 @item inner block not inside outer block in @var{symbol}
13289 The symbol information shows where symbol scopes begin and end
13290 (such as at the start of a function or a block of statements). This
13291 error indicates that an inner scope block is not fully contained
13292 in its outer scope blocks.
13294 @value{GDBN} circumvents the problem by treating the inner block as if it had
13295 the same scope as the outer block. In the error message, @var{symbol}
13296 may be shown as ``@code{(don't know)}'' if the outer block is not a
13299 @item block at @var{address} out of order
13301 The symbol information for symbol scope blocks should occur in
13302 order of increasing addresses. This error indicates that it does not
13305 @value{GDBN} does not circumvent this problem, and has trouble
13306 locating symbols in the source file whose symbols it is reading. (You
13307 can often determine what source file is affected by specifying
13308 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
13311 @item bad block start address patched
13313 The symbol information for a symbol scope block has a start address
13314 smaller than the address of the preceding source line. This is known
13315 to occur in the SunOS 4.1.1 (and earlier) C compiler.
13317 @value{GDBN} circumvents the problem by treating the symbol scope block as
13318 starting on the previous source line.
13320 @item bad string table offset in symbol @var{n}
13323 Symbol number @var{n} contains a pointer into the string table which is
13324 larger than the size of the string table.
13326 @value{GDBN} circumvents the problem by considering the symbol to have the
13327 name @code{foo}, which may cause other problems if many symbols end up
13330 @item unknown symbol type @code{0x@var{nn}}
13332 The symbol information contains new data types that @value{GDBN} does
13333 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
13334 uncomprehended information, in hexadecimal.
13336 @value{GDBN} circumvents the error by ignoring this symbol information.
13337 This usually allows you to debug your program, though certain symbols
13338 are not accessible. If you encounter such a problem and feel like
13339 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
13340 on @code{complain}, then go up to the function @code{read_dbx_symtab}
13341 and examine @code{*bufp} to see the symbol.
13343 @item stub type has NULL name
13345 @value{GDBN} could not find the full definition for a struct or class.
13347 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
13348 The symbol information for a C@t{++} member function is missing some
13349 information that recent versions of the compiler should have output for
13352 @item info mismatch between compiler and debugger
13354 @value{GDBN} could not parse a type specification output by the compiler.
13359 @chapter Specifying a Debugging Target
13361 @cindex debugging target
13362 A @dfn{target} is the execution environment occupied by your program.
13364 Often, @value{GDBN} runs in the same host environment as your program;
13365 in that case, the debugging target is specified as a side effect when
13366 you use the @code{file} or @code{core} commands. When you need more
13367 flexibility---for example, running @value{GDBN} on a physically separate
13368 host, or controlling a standalone system over a serial port or a
13369 realtime system over a TCP/IP connection---you can use the @code{target}
13370 command to specify one of the target types configured for @value{GDBN}
13371 (@pxref{Target Commands, ,Commands for Managing Targets}).
13373 @cindex target architecture
13374 It is possible to build @value{GDBN} for several different @dfn{target
13375 architectures}. When @value{GDBN} is built like that, you can choose
13376 one of the available architectures with the @kbd{set architecture}
13380 @kindex set architecture
13381 @kindex show architecture
13382 @item set architecture @var{arch}
13383 This command sets the current target architecture to @var{arch}. The
13384 value of @var{arch} can be @code{"auto"}, in addition to one of the
13385 supported architectures.
13387 @item show architecture
13388 Show the current target architecture.
13390 @item set processor
13392 @kindex set processor
13393 @kindex show processor
13394 These are alias commands for, respectively, @code{set architecture}
13395 and @code{show architecture}.
13399 * Active Targets:: Active targets
13400 * Target Commands:: Commands for managing targets
13401 * Byte Order:: Choosing target byte order
13404 @node Active Targets
13405 @section Active Targets
13407 @cindex stacking targets
13408 @cindex active targets
13409 @cindex multiple targets
13411 There are three classes of targets: processes, core files, and
13412 executable files. @value{GDBN} can work concurrently on up to three
13413 active targets, one in each class. This allows you to (for example)
13414 start a process and inspect its activity without abandoning your work on
13417 For example, if you execute @samp{gdb a.out}, then the executable file
13418 @code{a.out} is the only active target. If you designate a core file as
13419 well---presumably from a prior run that crashed and coredumped---then
13420 @value{GDBN} has two active targets and uses them in tandem, looking
13421 first in the corefile target, then in the executable file, to satisfy
13422 requests for memory addresses. (Typically, these two classes of target
13423 are complementary, since core files contain only a program's
13424 read-write memory---variables and so on---plus machine status, while
13425 executable files contain only the program text and initialized data.)
13427 When you type @code{run}, your executable file becomes an active process
13428 target as well. When a process target is active, all @value{GDBN}
13429 commands requesting memory addresses refer to that target; addresses in
13430 an active core file or executable file target are obscured while the
13431 process target is active.
13433 Use the @code{core-file} and @code{exec-file} commands to select a new
13434 core file or executable target (@pxref{Files, ,Commands to Specify
13435 Files}). To specify as a target a process that is already running, use
13436 the @code{attach} command (@pxref{Attach, ,Debugging an Already-running
13439 @node Target Commands
13440 @section Commands for Managing Targets
13443 @item target @var{type} @var{parameters}
13444 Connects the @value{GDBN} host environment to a target machine or
13445 process. A target is typically a protocol for talking to debugging
13446 facilities. You use the argument @var{type} to specify the type or
13447 protocol of the target machine.
13449 Further @var{parameters} are interpreted by the target protocol, but
13450 typically include things like device names or host names to connect
13451 with, process numbers, and baud rates.
13453 The @code{target} command does not repeat if you press @key{RET} again
13454 after executing the command.
13456 @kindex help target
13458 Displays the names of all targets available. To display targets
13459 currently selected, use either @code{info target} or @code{info files}
13460 (@pxref{Files, ,Commands to Specify Files}).
13462 @item help target @var{name}
13463 Describe a particular target, including any parameters necessary to
13466 @kindex set gnutarget
13467 @item set gnutarget @var{args}
13468 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
13469 knows whether it is reading an @dfn{executable},
13470 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
13471 with the @code{set gnutarget} command. Unlike most @code{target} commands,
13472 with @code{gnutarget} the @code{target} refers to a program, not a machine.
13475 @emph{Warning:} To specify a file format with @code{set gnutarget},
13476 you must know the actual BFD name.
13480 @xref{Files, , Commands to Specify Files}.
13482 @kindex show gnutarget
13483 @item show gnutarget
13484 Use the @code{show gnutarget} command to display what file format
13485 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
13486 @value{GDBN} will determine the file format for each file automatically,
13487 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
13490 @cindex common targets
13491 Here are some common targets (available, or not, depending on the GDB
13496 @item target exec @var{program}
13497 @cindex executable file target
13498 An executable file. @samp{target exec @var{program}} is the same as
13499 @samp{exec-file @var{program}}.
13501 @item target core @var{filename}
13502 @cindex core dump file target
13503 A core dump file. @samp{target core @var{filename}} is the same as
13504 @samp{core-file @var{filename}}.
13506 @item target remote @var{medium}
13507 @cindex remote target
13508 A remote system connected to @value{GDBN} via a serial line or network
13509 connection. This command tells @value{GDBN} to use its own remote
13510 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
13512 For example, if you have a board connected to @file{/dev/ttya} on the
13513 machine running @value{GDBN}, you could say:
13516 target remote /dev/ttya
13519 @code{target remote} supports the @code{load} command. This is only
13520 useful if you have some other way of getting the stub to the target
13521 system, and you can put it somewhere in memory where it won't get
13522 clobbered by the download.
13525 @cindex built-in simulator target
13526 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
13534 works; however, you cannot assume that a specific memory map, device
13535 drivers, or even basic I/O is available, although some simulators do
13536 provide these. For info about any processor-specific simulator details,
13537 see the appropriate section in @ref{Embedded Processors, ,Embedded
13542 Some configurations may include these targets as well:
13546 @item target nrom @var{dev}
13547 @cindex NetROM ROM emulator target
13548 NetROM ROM emulator. This target only supports downloading.
13552 Different targets are available on different configurations of @value{GDBN};
13553 your configuration may have more or fewer targets.
13555 Many remote targets require you to download the executable's code once
13556 you've successfully established a connection. You may wish to control
13557 various aspects of this process.
13562 @kindex set hash@r{, for remote monitors}
13563 @cindex hash mark while downloading
13564 This command controls whether a hash mark @samp{#} is displayed while
13565 downloading a file to the remote monitor. If on, a hash mark is
13566 displayed after each S-record is successfully downloaded to the
13570 @kindex show hash@r{, for remote monitors}
13571 Show the current status of displaying the hash mark.
13573 @item set debug monitor
13574 @kindex set debug monitor
13575 @cindex display remote monitor communications
13576 Enable or disable display of communications messages between
13577 @value{GDBN} and the remote monitor.
13579 @item show debug monitor
13580 @kindex show debug monitor
13581 Show the current status of displaying communications between
13582 @value{GDBN} and the remote monitor.
13587 @kindex load @var{filename}
13588 @item load @var{filename}
13590 Depending on what remote debugging facilities are configured into
13591 @value{GDBN}, the @code{load} command may be available. Where it exists, it
13592 is meant to make @var{filename} (an executable) available for debugging
13593 on the remote system---by downloading, or dynamic linking, for example.
13594 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
13595 the @code{add-symbol-file} command.
13597 If your @value{GDBN} does not have a @code{load} command, attempting to
13598 execute it gets the error message ``@code{You can't do that when your
13599 target is @dots{}}''
13601 The file is loaded at whatever address is specified in the executable.
13602 For some object file formats, you can specify the load address when you
13603 link the program; for other formats, like a.out, the object file format
13604 specifies a fixed address.
13605 @c FIXME! This would be a good place for an xref to the GNU linker doc.
13607 Depending on the remote side capabilities, @value{GDBN} may be able to
13608 load programs into flash memory.
13610 @code{load} does not repeat if you press @key{RET} again after using it.
13614 @section Choosing Target Byte Order
13616 @cindex choosing target byte order
13617 @cindex target byte order
13619 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
13620 offer the ability to run either big-endian or little-endian byte
13621 orders. Usually the executable or symbol will include a bit to
13622 designate the endian-ness, and you will not need to worry about
13623 which to use. However, you may still find it useful to adjust
13624 @value{GDBN}'s idea of processor endian-ness manually.
13628 @item set endian big
13629 Instruct @value{GDBN} to assume the target is big-endian.
13631 @item set endian little
13632 Instruct @value{GDBN} to assume the target is little-endian.
13634 @item set endian auto
13635 Instruct @value{GDBN} to use the byte order associated with the
13639 Display @value{GDBN}'s current idea of the target byte order.
13643 Note that these commands merely adjust interpretation of symbolic
13644 data on the host, and that they have absolutely no effect on the
13648 @node Remote Debugging
13649 @chapter Debugging Remote Programs
13650 @cindex remote debugging
13652 If you are trying to debug a program running on a machine that cannot run
13653 @value{GDBN} in the usual way, it is often useful to use remote debugging.
13654 For example, you might use remote debugging on an operating system kernel,
13655 or on a small system which does not have a general purpose operating system
13656 powerful enough to run a full-featured debugger.
13658 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
13659 to make this work with particular debugging targets. In addition,
13660 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
13661 but not specific to any particular target system) which you can use if you
13662 write the remote stubs---the code that runs on the remote system to
13663 communicate with @value{GDBN}.
13665 Other remote targets may be available in your
13666 configuration of @value{GDBN}; use @code{help target} to list them.
13669 * Connecting:: Connecting to a remote target
13670 * File Transfer:: Sending files to a remote system
13671 * Server:: Using the gdbserver program
13672 * Remote Configuration:: Remote configuration
13673 * Remote Stub:: Implementing a remote stub
13677 @section Connecting to a Remote Target
13679 On the @value{GDBN} host machine, you will need an unstripped copy of
13680 your program, since @value{GDBN} needs symbol and debugging information.
13681 Start up @value{GDBN} as usual, using the name of the local copy of your
13682 program as the first argument.
13684 @cindex @code{target remote}
13685 @value{GDBN} can communicate with the target over a serial line, or
13686 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
13687 each case, @value{GDBN} uses the same protocol for debugging your
13688 program; only the medium carrying the debugging packets varies. The
13689 @code{target remote} command establishes a connection to the target.
13690 Its arguments indicate which medium to use:
13694 @item target remote @var{serial-device}
13695 @cindex serial line, @code{target remote}
13696 Use @var{serial-device} to communicate with the target. For example,
13697 to use a serial line connected to the device named @file{/dev/ttyb}:
13700 target remote /dev/ttyb
13703 If you're using a serial line, you may want to give @value{GDBN} the
13704 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
13705 (@pxref{Remote Configuration, set remotebaud}) before the
13706 @code{target} command.
13708 @item target remote @code{@var{host}:@var{port}}
13709 @itemx target remote @code{tcp:@var{host}:@var{port}}
13710 @cindex @acronym{TCP} port, @code{target remote}
13711 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
13712 The @var{host} may be either a host name or a numeric @acronym{IP}
13713 address; @var{port} must be a decimal number. The @var{host} could be
13714 the target machine itself, if it is directly connected to the net, or
13715 it might be a terminal server which in turn has a serial line to the
13718 For example, to connect to port 2828 on a terminal server named
13722 target remote manyfarms:2828
13725 If your remote target is actually running on the same machine as your
13726 debugger session (e.g.@: a simulator for your target running on the
13727 same host), you can omit the hostname. For example, to connect to
13728 port 1234 on your local machine:
13731 target remote :1234
13735 Note that the colon is still required here.
13737 @item target remote @code{udp:@var{host}:@var{port}}
13738 @cindex @acronym{UDP} port, @code{target remote}
13739 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
13740 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
13743 target remote udp:manyfarms:2828
13746 When using a @acronym{UDP} connection for remote debugging, you should
13747 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
13748 can silently drop packets on busy or unreliable networks, which will
13749 cause havoc with your debugging session.
13751 @item target remote | @var{command}
13752 @cindex pipe, @code{target remote} to
13753 Run @var{command} in the background and communicate with it using a
13754 pipe. The @var{command} is a shell command, to be parsed and expanded
13755 by the system's command shell, @code{/bin/sh}; it should expect remote
13756 protocol packets on its standard input, and send replies on its
13757 standard output. You could use this to run a stand-alone simulator
13758 that speaks the remote debugging protocol, to make net connections
13759 using programs like @code{ssh}, or for other similar tricks.
13761 If @var{command} closes its standard output (perhaps by exiting),
13762 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
13763 program has already exited, this will have no effect.)
13767 Once the connection has been established, you can use all the usual
13768 commands to examine and change data. The remote program is already
13769 running; you can use @kbd{step} and @kbd{continue}, and you do not
13770 need to use @kbd{run}.
13772 @cindex interrupting remote programs
13773 @cindex remote programs, interrupting
13774 Whenever @value{GDBN} is waiting for the remote program, if you type the
13775 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
13776 program. This may or may not succeed, depending in part on the hardware
13777 and the serial drivers the remote system uses. If you type the
13778 interrupt character once again, @value{GDBN} displays this prompt:
13781 Interrupted while waiting for the program.
13782 Give up (and stop debugging it)? (y or n)
13785 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
13786 (If you decide you want to try again later, you can use @samp{target
13787 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
13788 goes back to waiting.
13791 @kindex detach (remote)
13793 When you have finished debugging the remote program, you can use the
13794 @code{detach} command to release it from @value{GDBN} control.
13795 Detaching from the target normally resumes its execution, but the results
13796 will depend on your particular remote stub. After the @code{detach}
13797 command, @value{GDBN} is free to connect to another target.
13801 The @code{disconnect} command behaves like @code{detach}, except that
13802 the target is generally not resumed. It will wait for @value{GDBN}
13803 (this instance or another one) to connect and continue debugging. After
13804 the @code{disconnect} command, @value{GDBN} is again free to connect to
13807 @cindex send command to remote monitor
13808 @cindex extend @value{GDBN} for remote targets
13809 @cindex add new commands for external monitor
13811 @item monitor @var{cmd}
13812 This command allows you to send arbitrary commands directly to the
13813 remote monitor. Since @value{GDBN} doesn't care about the commands it
13814 sends like this, this command is the way to extend @value{GDBN}---you
13815 can add new commands that only the external monitor will understand
13819 @node File Transfer
13820 @section Sending files to a remote system
13821 @cindex remote target, file transfer
13822 @cindex file transfer
13823 @cindex sending files to remote systems
13825 Some remote targets offer the ability to transfer files over the same
13826 connection used to communicate with @value{GDBN}. This is convenient
13827 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
13828 running @code{gdbserver} over a network interface. For other targets,
13829 e.g.@: embedded devices with only a single serial port, this may be
13830 the only way to upload or download files.
13832 Not all remote targets support these commands.
13836 @item remote put @var{hostfile} @var{targetfile}
13837 Copy file @var{hostfile} from the host system (the machine running
13838 @value{GDBN}) to @var{targetfile} on the target system.
13841 @item remote get @var{targetfile} @var{hostfile}
13842 Copy file @var{targetfile} from the target system to @var{hostfile}
13843 on the host system.
13845 @kindex remote delete
13846 @item remote delete @var{targetfile}
13847 Delete @var{targetfile} from the target system.
13852 @section Using the @code{gdbserver} Program
13855 @cindex remote connection without stubs
13856 @code{gdbserver} is a control program for Unix-like systems, which
13857 allows you to connect your program with a remote @value{GDBN} via
13858 @code{target remote}---but without linking in the usual debugging stub.
13860 @code{gdbserver} is not a complete replacement for the debugging stubs,
13861 because it requires essentially the same operating-system facilities
13862 that @value{GDBN} itself does. In fact, a system that can run
13863 @code{gdbserver} to connect to a remote @value{GDBN} could also run
13864 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
13865 because it is a much smaller program than @value{GDBN} itself. It is
13866 also easier to port than all of @value{GDBN}, so you may be able to get
13867 started more quickly on a new system by using @code{gdbserver}.
13868 Finally, if you develop code for real-time systems, you may find that
13869 the tradeoffs involved in real-time operation make it more convenient to
13870 do as much development work as possible on another system, for example
13871 by cross-compiling. You can use @code{gdbserver} to make a similar
13872 choice for debugging.
13874 @value{GDBN} and @code{gdbserver} communicate via either a serial line
13875 or a TCP connection, using the standard @value{GDBN} remote serial
13879 @emph{Warning:} @code{gdbserver} does not have any built-in security.
13880 Do not run @code{gdbserver} connected to any public network; a
13881 @value{GDBN} connection to @code{gdbserver} provides access to the
13882 target system with the same privileges as the user running
13886 @subsection Running @code{gdbserver}
13887 @cindex arguments, to @code{gdbserver}
13889 Run @code{gdbserver} on the target system. You need a copy of the
13890 program you want to debug, including any libraries it requires.
13891 @code{gdbserver} does not need your program's symbol table, so you can
13892 strip the program if necessary to save space. @value{GDBN} on the host
13893 system does all the symbol handling.
13895 To use the server, you must tell it how to communicate with @value{GDBN};
13896 the name of your program; and the arguments for your program. The usual
13900 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
13903 @var{comm} is either a device name (to use a serial line) or a TCP
13904 hostname and portnumber. For example, to debug Emacs with the argument
13905 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
13909 target> gdbserver /dev/com1 emacs foo.txt
13912 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
13915 To use a TCP connection instead of a serial line:
13918 target> gdbserver host:2345 emacs foo.txt
13921 The only difference from the previous example is the first argument,
13922 specifying that you are communicating with the host @value{GDBN} via
13923 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
13924 expect a TCP connection from machine @samp{host} to local TCP port 2345.
13925 (Currently, the @samp{host} part is ignored.) You can choose any number
13926 you want for the port number as long as it does not conflict with any
13927 TCP ports already in use on the target system (for example, @code{23} is
13928 reserved for @code{telnet}).@footnote{If you choose a port number that
13929 conflicts with another service, @code{gdbserver} prints an error message
13930 and exits.} You must use the same port number with the host @value{GDBN}
13931 @code{target remote} command.
13933 @subsubsection Attaching to a Running Program
13935 On some targets, @code{gdbserver} can also attach to running programs.
13936 This is accomplished via the @code{--attach} argument. The syntax is:
13939 target> gdbserver --attach @var{comm} @var{pid}
13942 @var{pid} is the process ID of a currently running process. It isn't necessary
13943 to point @code{gdbserver} at a binary for the running process.
13946 @cindex attach to a program by name
13947 You can debug processes by name instead of process ID if your target has the
13948 @code{pidof} utility:
13951 target> gdbserver --attach @var{comm} `pidof @var{program}`
13954 In case more than one copy of @var{program} is running, or @var{program}
13955 has multiple threads, most versions of @code{pidof} support the
13956 @code{-s} option to only return the first process ID.
13958 @subsubsection Multi-Process Mode for @code{gdbserver}
13959 @cindex gdbserver, multiple processes
13960 @cindex multiple processes with gdbserver
13962 When you connect to @code{gdbserver} using @code{target remote},
13963 @code{gdbserver} debugs the specified program only once. When the
13964 program exits, or you detach from it, @value{GDBN} closes the connection
13965 and @code{gdbserver} exits.
13967 If you connect using @kbd{target extended-remote}, @code{gdbserver}
13968 enters multi-process mode. When the debugged program exits, or you
13969 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
13970 though no program is running. The @code{run} and @code{attach}
13971 commands instruct @code{gdbserver} to run or attach to a new program.
13972 The @code{run} command uses @code{set remote exec-file} (@pxref{set
13973 remote exec-file}) to select the program to run. Command line
13974 arguments are supported, except for wildcard expansion and I/O
13975 redirection (@pxref{Arguments}).
13977 To start @code{gdbserver} without supplying an initial command to run
13978 or process ID to attach, use the @option{--multi} command line option.
13979 Then you can connect using @kbd{target extended-remote} and start
13980 the program you want to debug.
13982 @code{gdbserver} does not automatically exit in multi-process mode.
13983 You can terminate it by using @code{monitor exit}
13984 (@pxref{Monitor Commands for gdbserver}).
13986 @subsubsection Other Command-Line Arguments for @code{gdbserver}
13988 The @option{--debug} option tells @code{gdbserver} to display extra
13989 status information about the debugging process. The
13990 @option{--remote-debug} option tells @code{gdbserver} to display
13991 remote protocol debug output. These options are intended for
13992 @code{gdbserver} development and for bug reports to the developers.
13994 The @option{--wrapper} option specifies a wrapper to launch programs
13995 for debugging. The option should be followed by the name of the
13996 wrapper, then any command-line arguments to pass to the wrapper, then
13997 @kbd{--} indicating the end of the wrapper arguments.
13999 @code{gdbserver} runs the specified wrapper program with a combined
14000 command line including the wrapper arguments, then the name of the
14001 program to debug, then any arguments to the program. The wrapper
14002 runs until it executes your program, and then @value{GDBN} gains control.
14004 You can use any program that eventually calls @code{execve} with
14005 its arguments as a wrapper. Several standard Unix utilities do
14006 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
14007 with @code{exec "$@@"} will also work.
14009 For example, you can use @code{env} to pass an environment variable to
14010 the debugged program, without setting the variable in @code{gdbserver}'s
14014 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
14017 @subsection Connecting to @code{gdbserver}
14019 Run @value{GDBN} on the host system.
14021 First make sure you have the necessary symbol files. Load symbols for
14022 your application using the @code{file} command before you connect. Use
14023 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
14024 was compiled with the correct sysroot using @code{--with-sysroot}).
14026 The symbol file and target libraries must exactly match the executable
14027 and libraries on the target, with one exception: the files on the host
14028 system should not be stripped, even if the files on the target system
14029 are. Mismatched or missing files will lead to confusing results
14030 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
14031 files may also prevent @code{gdbserver} from debugging multi-threaded
14034 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
14035 For TCP connections, you must start up @code{gdbserver} prior to using
14036 the @code{target remote} command. Otherwise you may get an error whose
14037 text depends on the host system, but which usually looks something like
14038 @samp{Connection refused}. Don't use the @code{load}
14039 command in @value{GDBN} when using @code{gdbserver}, since the program is
14040 already on the target.
14042 @subsection Monitor Commands for @code{gdbserver}
14043 @cindex monitor commands, for @code{gdbserver}
14044 @anchor{Monitor Commands for gdbserver}
14046 During a @value{GDBN} session using @code{gdbserver}, you can use the
14047 @code{monitor} command to send special requests to @code{gdbserver}.
14048 Here are the available commands.
14052 List the available monitor commands.
14054 @item monitor set debug 0
14055 @itemx monitor set debug 1
14056 Disable or enable general debugging messages.
14058 @item monitor set remote-debug 0
14059 @itemx monitor set remote-debug 1
14060 Disable or enable specific debugging messages associated with the remote
14061 protocol (@pxref{Remote Protocol}).
14064 Tell gdbserver to exit immediately. This command should be followed by
14065 @code{disconnect} to close the debugging session. @code{gdbserver} will
14066 detach from any attached processes and kill any processes it created.
14067 Use @code{monitor exit} to terminate @code{gdbserver} at the end
14068 of a multi-process mode debug session.
14072 @node Remote Configuration
14073 @section Remote Configuration
14076 @kindex show remote
14077 This section documents the configuration options available when
14078 debugging remote programs. For the options related to the File I/O
14079 extensions of the remote protocol, see @ref{system,
14080 system-call-allowed}.
14083 @item set remoteaddresssize @var{bits}
14084 @cindex address size for remote targets
14085 @cindex bits in remote address
14086 Set the maximum size of address in a memory packet to the specified
14087 number of bits. @value{GDBN} will mask off the address bits above
14088 that number, when it passes addresses to the remote target. The
14089 default value is the number of bits in the target's address.
14091 @item show remoteaddresssize
14092 Show the current value of remote address size in bits.
14094 @item set remotebaud @var{n}
14095 @cindex baud rate for remote targets
14096 Set the baud rate for the remote serial I/O to @var{n} baud. The
14097 value is used to set the speed of the serial port used for debugging
14100 @item show remotebaud
14101 Show the current speed of the remote connection.
14103 @item set remotebreak
14104 @cindex interrupt remote programs
14105 @cindex BREAK signal instead of Ctrl-C
14106 @anchor{set remotebreak}
14107 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
14108 when you type @kbd{Ctrl-c} to interrupt the program running
14109 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
14110 character instead. The default is off, since most remote systems
14111 expect to see @samp{Ctrl-C} as the interrupt signal.
14113 @item show remotebreak
14114 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
14115 interrupt the remote program.
14117 @item set remoteflow on
14118 @itemx set remoteflow off
14119 @kindex set remoteflow
14120 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
14121 on the serial port used to communicate to the remote target.
14123 @item show remoteflow
14124 @kindex show remoteflow
14125 Show the current setting of hardware flow control.
14127 @item set remotelogbase @var{base}
14128 Set the base (a.k.a.@: radix) of logging serial protocol
14129 communications to @var{base}. Supported values of @var{base} are:
14130 @code{ascii}, @code{octal}, and @code{hex}. The default is
14133 @item show remotelogbase
14134 Show the current setting of the radix for logging remote serial
14137 @item set remotelogfile @var{file}
14138 @cindex record serial communications on file
14139 Record remote serial communications on the named @var{file}. The
14140 default is not to record at all.
14142 @item show remotelogfile.
14143 Show the current setting of the file name on which to record the
14144 serial communications.
14146 @item set remotetimeout @var{num}
14147 @cindex timeout for serial communications
14148 @cindex remote timeout
14149 Set the timeout limit to wait for the remote target to respond to
14150 @var{num} seconds. The default is 2 seconds.
14152 @item show remotetimeout
14153 Show the current number of seconds to wait for the remote target
14156 @cindex limit hardware breakpoints and watchpoints
14157 @cindex remote target, limit break- and watchpoints
14158 @anchor{set remote hardware-watchpoint-limit}
14159 @anchor{set remote hardware-breakpoint-limit}
14160 @item set remote hardware-watchpoint-limit @var{limit}
14161 @itemx set remote hardware-breakpoint-limit @var{limit}
14162 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
14163 watchpoints. A limit of -1, the default, is treated as unlimited.
14165 @item set remote exec-file @var{filename}
14166 @itemx show remote exec-file
14167 @anchor{set remote exec-file}
14168 @cindex executable file, for remote target
14169 Select the file used for @code{run} with @code{target
14170 extended-remote}. This should be set to a filename valid on the
14171 target system. If it is not set, the target will use a default
14172 filename (e.g.@: the last program run).
14176 @item set tcp auto-retry on
14177 @cindex auto-retry, for remote TCP target
14178 Enable auto-retry for remote TCP connections. This is useful if the remote
14179 debugging agent is launched in parallel with @value{GDBN}; there is a race
14180 condition because the agent may not become ready to accept the connection
14181 before @value{GDBN} attempts to connect. When auto-retry is
14182 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
14183 to establish the connection using the timeout specified by
14184 @code{set tcp connect-timeout}.
14186 @item set tcp auto-retry off
14187 Do not auto-retry failed TCP connections.
14189 @item show tcp auto-retry
14190 Show the current auto-retry setting.
14192 @item set tcp connect-timeout @var{seconds}
14193 @cindex connection timeout, for remote TCP target
14194 @cindex timeout, for remote target connection
14195 Set the timeout for establishing a TCP connection to the remote target to
14196 @var{seconds}. The timeout affects both polling to retry failed connections
14197 (enabled by @code{set tcp auto-retry on}) and waiting for connections
14198 that are merely slow to complete, and represents an approximate cumulative
14201 @item show tcp connect-timeout
14202 Show the current connection timeout setting.
14205 @cindex remote packets, enabling and disabling
14206 The @value{GDBN} remote protocol autodetects the packets supported by
14207 your debugging stub. If you need to override the autodetection, you
14208 can use these commands to enable or disable individual packets. Each
14209 packet can be set to @samp{on} (the remote target supports this
14210 packet), @samp{off} (the remote target does not support this packet),
14211 or @samp{auto} (detect remote target support for this packet). They
14212 all default to @samp{auto}. For more information about each packet,
14213 see @ref{Remote Protocol}.
14215 During normal use, you should not have to use any of these commands.
14216 If you do, that may be a bug in your remote debugging stub, or a bug
14217 in @value{GDBN}. You may want to report the problem to the
14218 @value{GDBN} developers.
14220 For each packet @var{name}, the command to enable or disable the
14221 packet is @code{set remote @var{name}-packet}. The available settings
14224 @multitable @columnfractions 0.28 0.32 0.25
14227 @tab Related Features
14229 @item @code{fetch-register}
14231 @tab @code{info registers}
14233 @item @code{set-register}
14237 @item @code{binary-download}
14239 @tab @code{load}, @code{set}
14241 @item @code{read-aux-vector}
14242 @tab @code{qXfer:auxv:read}
14243 @tab @code{info auxv}
14245 @item @code{symbol-lookup}
14246 @tab @code{qSymbol}
14247 @tab Detecting multiple threads
14249 @item @code{attach}
14250 @tab @code{vAttach}
14253 @item @code{verbose-resume}
14255 @tab Stepping or resuming multiple threads
14261 @item @code{software-breakpoint}
14265 @item @code{hardware-breakpoint}
14269 @item @code{write-watchpoint}
14273 @item @code{read-watchpoint}
14277 @item @code{access-watchpoint}
14281 @item @code{target-features}
14282 @tab @code{qXfer:features:read}
14283 @tab @code{set architecture}
14285 @item @code{library-info}
14286 @tab @code{qXfer:libraries:read}
14287 @tab @code{info sharedlibrary}
14289 @item @code{memory-map}
14290 @tab @code{qXfer:memory-map:read}
14291 @tab @code{info mem}
14293 @item @code{read-spu-object}
14294 @tab @code{qXfer:spu:read}
14295 @tab @code{info spu}
14297 @item @code{write-spu-object}
14298 @tab @code{qXfer:spu:write}
14299 @tab @code{info spu}
14301 @item @code{get-thread-local-@*storage-address}
14302 @tab @code{qGetTLSAddr}
14303 @tab Displaying @code{__thread} variables
14305 @item @code{search-memory}
14306 @tab @code{qSearch:memory}
14309 @item @code{supported-packets}
14310 @tab @code{qSupported}
14311 @tab Remote communications parameters
14313 @item @code{pass-signals}
14314 @tab @code{QPassSignals}
14315 @tab @code{handle @var{signal}}
14317 @item @code{hostio-close-packet}
14318 @tab @code{vFile:close}
14319 @tab @code{remote get}, @code{remote put}
14321 @item @code{hostio-open-packet}
14322 @tab @code{vFile:open}
14323 @tab @code{remote get}, @code{remote put}
14325 @item @code{hostio-pread-packet}
14326 @tab @code{vFile:pread}
14327 @tab @code{remote get}, @code{remote put}
14329 @item @code{hostio-pwrite-packet}
14330 @tab @code{vFile:pwrite}
14331 @tab @code{remote get}, @code{remote put}
14333 @item @code{hostio-unlink-packet}
14334 @tab @code{vFile:unlink}
14335 @tab @code{remote delete}
14337 @item @code{noack-packet}
14338 @tab @code{QStartNoAckMode}
14339 @tab Packet acknowledgment
14341 @item @code{osdata}
14342 @tab @code{qXfer:osdata:read}
14343 @tab @code{info os}
14347 @section Implementing a Remote Stub
14349 @cindex debugging stub, example
14350 @cindex remote stub, example
14351 @cindex stub example, remote debugging
14352 The stub files provided with @value{GDBN} implement the target side of the
14353 communication protocol, and the @value{GDBN} side is implemented in the
14354 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
14355 these subroutines to communicate, and ignore the details. (If you're
14356 implementing your own stub file, you can still ignore the details: start
14357 with one of the existing stub files. @file{sparc-stub.c} is the best
14358 organized, and therefore the easiest to read.)
14360 @cindex remote serial debugging, overview
14361 To debug a program running on another machine (the debugging
14362 @dfn{target} machine), you must first arrange for all the usual
14363 prerequisites for the program to run by itself. For example, for a C
14368 A startup routine to set up the C runtime environment; these usually
14369 have a name like @file{crt0}. The startup routine may be supplied by
14370 your hardware supplier, or you may have to write your own.
14373 A C subroutine library to support your program's
14374 subroutine calls, notably managing input and output.
14377 A way of getting your program to the other machine---for example, a
14378 download program. These are often supplied by the hardware
14379 manufacturer, but you may have to write your own from hardware
14383 The next step is to arrange for your program to use a serial port to
14384 communicate with the machine where @value{GDBN} is running (the @dfn{host}
14385 machine). In general terms, the scheme looks like this:
14389 @value{GDBN} already understands how to use this protocol; when everything
14390 else is set up, you can simply use the @samp{target remote} command
14391 (@pxref{Targets,,Specifying a Debugging Target}).
14393 @item On the target,
14394 you must link with your program a few special-purpose subroutines that
14395 implement the @value{GDBN} remote serial protocol. The file containing these
14396 subroutines is called a @dfn{debugging stub}.
14398 On certain remote targets, you can use an auxiliary program
14399 @code{gdbserver} instead of linking a stub into your program.
14400 @xref{Server,,Using the @code{gdbserver} Program}, for details.
14403 The debugging stub is specific to the architecture of the remote
14404 machine; for example, use @file{sparc-stub.c} to debug programs on
14407 @cindex remote serial stub list
14408 These working remote stubs are distributed with @value{GDBN}:
14413 @cindex @file{i386-stub.c}
14416 For Intel 386 and compatible architectures.
14419 @cindex @file{m68k-stub.c}
14420 @cindex Motorola 680x0
14422 For Motorola 680x0 architectures.
14425 @cindex @file{sh-stub.c}
14428 For Renesas SH architectures.
14431 @cindex @file{sparc-stub.c}
14433 For @sc{sparc} architectures.
14435 @item sparcl-stub.c
14436 @cindex @file{sparcl-stub.c}
14439 For Fujitsu @sc{sparclite} architectures.
14443 The @file{README} file in the @value{GDBN} distribution may list other
14444 recently added stubs.
14447 * Stub Contents:: What the stub can do for you
14448 * Bootstrapping:: What you must do for the stub
14449 * Debug Session:: Putting it all together
14452 @node Stub Contents
14453 @subsection What the Stub Can Do for You
14455 @cindex remote serial stub
14456 The debugging stub for your architecture supplies these three
14460 @item set_debug_traps
14461 @findex set_debug_traps
14462 @cindex remote serial stub, initialization
14463 This routine arranges for @code{handle_exception} to run when your
14464 program stops. You must call this subroutine explicitly near the
14465 beginning of your program.
14467 @item handle_exception
14468 @findex handle_exception
14469 @cindex remote serial stub, main routine
14470 This is the central workhorse, but your program never calls it
14471 explicitly---the setup code arranges for @code{handle_exception} to
14472 run when a trap is triggered.
14474 @code{handle_exception} takes control when your program stops during
14475 execution (for example, on a breakpoint), and mediates communications
14476 with @value{GDBN} on the host machine. This is where the communications
14477 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
14478 representative on the target machine. It begins by sending summary
14479 information on the state of your program, then continues to execute,
14480 retrieving and transmitting any information @value{GDBN} needs, until you
14481 execute a @value{GDBN} command that makes your program resume; at that point,
14482 @code{handle_exception} returns control to your own code on the target
14486 @cindex @code{breakpoint} subroutine, remote
14487 Use this auxiliary subroutine to make your program contain a
14488 breakpoint. Depending on the particular situation, this may be the only
14489 way for @value{GDBN} to get control. For instance, if your target
14490 machine has some sort of interrupt button, you won't need to call this;
14491 pressing the interrupt button transfers control to
14492 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
14493 simply receiving characters on the serial port may also trigger a trap;
14494 again, in that situation, you don't need to call @code{breakpoint} from
14495 your own program---simply running @samp{target remote} from the host
14496 @value{GDBN} session gets control.
14498 Call @code{breakpoint} if none of these is true, or if you simply want
14499 to make certain your program stops at a predetermined point for the
14500 start of your debugging session.
14503 @node Bootstrapping
14504 @subsection What You Must Do for the Stub
14506 @cindex remote stub, support routines
14507 The debugging stubs that come with @value{GDBN} are set up for a particular
14508 chip architecture, but they have no information about the rest of your
14509 debugging target machine.
14511 First of all you need to tell the stub how to communicate with the
14515 @item int getDebugChar()
14516 @findex getDebugChar
14517 Write this subroutine to read a single character from the serial port.
14518 It may be identical to @code{getchar} for your target system; a
14519 different name is used to allow you to distinguish the two if you wish.
14521 @item void putDebugChar(int)
14522 @findex putDebugChar
14523 Write this subroutine to write a single character to the serial port.
14524 It may be identical to @code{putchar} for your target system; a
14525 different name is used to allow you to distinguish the two if you wish.
14528 @cindex control C, and remote debugging
14529 @cindex interrupting remote targets
14530 If you want @value{GDBN} to be able to stop your program while it is
14531 running, you need to use an interrupt-driven serial driver, and arrange
14532 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
14533 character). That is the character which @value{GDBN} uses to tell the
14534 remote system to stop.
14536 Getting the debugging target to return the proper status to @value{GDBN}
14537 probably requires changes to the standard stub; one quick and dirty way
14538 is to just execute a breakpoint instruction (the ``dirty'' part is that
14539 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
14541 Other routines you need to supply are:
14544 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
14545 @findex exceptionHandler
14546 Write this function to install @var{exception_address} in the exception
14547 handling tables. You need to do this because the stub does not have any
14548 way of knowing what the exception handling tables on your target system
14549 are like (for example, the processor's table might be in @sc{rom},
14550 containing entries which point to a table in @sc{ram}).
14551 @var{exception_number} is the exception number which should be changed;
14552 its meaning is architecture-dependent (for example, different numbers
14553 might represent divide by zero, misaligned access, etc). When this
14554 exception occurs, control should be transferred directly to
14555 @var{exception_address}, and the processor state (stack, registers,
14556 and so on) should be just as it is when a processor exception occurs. So if
14557 you want to use a jump instruction to reach @var{exception_address}, it
14558 should be a simple jump, not a jump to subroutine.
14560 For the 386, @var{exception_address} should be installed as an interrupt
14561 gate so that interrupts are masked while the handler runs. The gate
14562 should be at privilege level 0 (the most privileged level). The
14563 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
14564 help from @code{exceptionHandler}.
14566 @item void flush_i_cache()
14567 @findex flush_i_cache
14568 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
14569 instruction cache, if any, on your target machine. If there is no
14570 instruction cache, this subroutine may be a no-op.
14572 On target machines that have instruction caches, @value{GDBN} requires this
14573 function to make certain that the state of your program is stable.
14577 You must also make sure this library routine is available:
14580 @item void *memset(void *, int, int)
14582 This is the standard library function @code{memset} that sets an area of
14583 memory to a known value. If you have one of the free versions of
14584 @code{libc.a}, @code{memset} can be found there; otherwise, you must
14585 either obtain it from your hardware manufacturer, or write your own.
14588 If you do not use the GNU C compiler, you may need other standard
14589 library subroutines as well; this varies from one stub to another,
14590 but in general the stubs are likely to use any of the common library
14591 subroutines which @code{@value{NGCC}} generates as inline code.
14594 @node Debug Session
14595 @subsection Putting it All Together
14597 @cindex remote serial debugging summary
14598 In summary, when your program is ready to debug, you must follow these
14603 Make sure you have defined the supporting low-level routines
14604 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
14606 @code{getDebugChar}, @code{putDebugChar},
14607 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
14611 Insert these lines near the top of your program:
14619 For the 680x0 stub only, you need to provide a variable called
14620 @code{exceptionHook}. Normally you just use:
14623 void (*exceptionHook)() = 0;
14627 but if before calling @code{set_debug_traps}, you set it to point to a
14628 function in your program, that function is called when
14629 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
14630 error). The function indicated by @code{exceptionHook} is called with
14631 one parameter: an @code{int} which is the exception number.
14634 Compile and link together: your program, the @value{GDBN} debugging stub for
14635 your target architecture, and the supporting subroutines.
14638 Make sure you have a serial connection between your target machine and
14639 the @value{GDBN} host, and identify the serial port on the host.
14642 @c The "remote" target now provides a `load' command, so we should
14643 @c document that. FIXME.
14644 Download your program to your target machine (or get it there by
14645 whatever means the manufacturer provides), and start it.
14648 Start @value{GDBN} on the host, and connect to the target
14649 (@pxref{Connecting,,Connecting to a Remote Target}).
14653 @node Configurations
14654 @chapter Configuration-Specific Information
14656 While nearly all @value{GDBN} commands are available for all native and
14657 cross versions of the debugger, there are some exceptions. This chapter
14658 describes things that are only available in certain configurations.
14660 There are three major categories of configurations: native
14661 configurations, where the host and target are the same, embedded
14662 operating system configurations, which are usually the same for several
14663 different processor architectures, and bare embedded processors, which
14664 are quite different from each other.
14669 * Embedded Processors::
14676 This section describes details specific to particular native
14681 * BSD libkvm Interface:: Debugging BSD kernel memory images
14682 * SVR4 Process Information:: SVR4 process information
14683 * DJGPP Native:: Features specific to the DJGPP port
14684 * Cygwin Native:: Features specific to the Cygwin port
14685 * Hurd Native:: Features specific to @sc{gnu} Hurd
14686 * Neutrino:: Features specific to QNX Neutrino
14687 * Darwin:: Features specific to Darwin
14693 On HP-UX systems, if you refer to a function or variable name that
14694 begins with a dollar sign, @value{GDBN} searches for a user or system
14695 name first, before it searches for a convenience variable.
14698 @node BSD libkvm Interface
14699 @subsection BSD libkvm Interface
14702 @cindex kernel memory image
14703 @cindex kernel crash dump
14705 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
14706 interface that provides a uniform interface for accessing kernel virtual
14707 memory images, including live systems and crash dumps. @value{GDBN}
14708 uses this interface to allow you to debug live kernels and kernel crash
14709 dumps on many native BSD configurations. This is implemented as a
14710 special @code{kvm} debugging target. For debugging a live system, load
14711 the currently running kernel into @value{GDBN} and connect to the
14715 (@value{GDBP}) @b{target kvm}
14718 For debugging crash dumps, provide the file name of the crash dump as an
14722 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
14725 Once connected to the @code{kvm} target, the following commands are
14731 Set current context from the @dfn{Process Control Block} (PCB) address.
14734 Set current context from proc address. This command isn't available on
14735 modern FreeBSD systems.
14738 @node SVR4 Process Information
14739 @subsection SVR4 Process Information
14741 @cindex examine process image
14742 @cindex process info via @file{/proc}
14744 Many versions of SVR4 and compatible systems provide a facility called
14745 @samp{/proc} that can be used to examine the image of a running
14746 process using file-system subroutines. If @value{GDBN} is configured
14747 for an operating system with this facility, the command @code{info
14748 proc} is available to report information about the process running
14749 your program, or about any process running on your system. @code{info
14750 proc} works only on SVR4 systems that include the @code{procfs} code.
14751 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
14752 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
14758 @itemx info proc @var{process-id}
14759 Summarize available information about any running process. If a
14760 process ID is specified by @var{process-id}, display information about
14761 that process; otherwise display information about the program being
14762 debugged. The summary includes the debugged process ID, the command
14763 line used to invoke it, its current working directory, and its
14764 executable file's absolute file name.
14766 On some systems, @var{process-id} can be of the form
14767 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
14768 within a process. If the optional @var{pid} part is missing, it means
14769 a thread from the process being debugged (the leading @samp{/} still
14770 needs to be present, or else @value{GDBN} will interpret the number as
14771 a process ID rather than a thread ID).
14773 @item info proc mappings
14774 @cindex memory address space mappings
14775 Report the memory address space ranges accessible in the program, with
14776 information on whether the process has read, write, or execute access
14777 rights to each range. On @sc{gnu}/Linux systems, each memory range
14778 includes the object file which is mapped to that range, instead of the
14779 memory access rights to that range.
14781 @item info proc stat
14782 @itemx info proc status
14783 @cindex process detailed status information
14784 These subcommands are specific to @sc{gnu}/Linux systems. They show
14785 the process-related information, including the user ID and group ID;
14786 how many threads are there in the process; its virtual memory usage;
14787 the signals that are pending, blocked, and ignored; its TTY; its
14788 consumption of system and user time; its stack size; its @samp{nice}
14789 value; etc. For more information, see the @samp{proc} man page
14790 (type @kbd{man 5 proc} from your shell prompt).
14792 @item info proc all
14793 Show all the information about the process described under all of the
14794 above @code{info proc} subcommands.
14797 @comment These sub-options of 'info proc' were not included when
14798 @comment procfs.c was re-written. Keep their descriptions around
14799 @comment against the day when someone finds the time to put them back in.
14800 @kindex info proc times
14801 @item info proc times
14802 Starting time, user CPU time, and system CPU time for your program and
14805 @kindex info proc id
14807 Report on the process IDs related to your program: its own process ID,
14808 the ID of its parent, the process group ID, and the session ID.
14811 @item set procfs-trace
14812 @kindex set procfs-trace
14813 @cindex @code{procfs} API calls
14814 This command enables and disables tracing of @code{procfs} API calls.
14816 @item show procfs-trace
14817 @kindex show procfs-trace
14818 Show the current state of @code{procfs} API call tracing.
14820 @item set procfs-file @var{file}
14821 @kindex set procfs-file
14822 Tell @value{GDBN} to write @code{procfs} API trace to the named
14823 @var{file}. @value{GDBN} appends the trace info to the previous
14824 contents of the file. The default is to display the trace on the
14827 @item show procfs-file
14828 @kindex show procfs-file
14829 Show the file to which @code{procfs} API trace is written.
14831 @item proc-trace-entry
14832 @itemx proc-trace-exit
14833 @itemx proc-untrace-entry
14834 @itemx proc-untrace-exit
14835 @kindex proc-trace-entry
14836 @kindex proc-trace-exit
14837 @kindex proc-untrace-entry
14838 @kindex proc-untrace-exit
14839 These commands enable and disable tracing of entries into and exits
14840 from the @code{syscall} interface.
14843 @kindex info pidlist
14844 @cindex process list, QNX Neutrino
14845 For QNX Neutrino only, this command displays the list of all the
14846 processes and all the threads within each process.
14849 @kindex info meminfo
14850 @cindex mapinfo list, QNX Neutrino
14851 For QNX Neutrino only, this command displays the list of all mapinfos.
14855 @subsection Features for Debugging @sc{djgpp} Programs
14856 @cindex @sc{djgpp} debugging
14857 @cindex native @sc{djgpp} debugging
14858 @cindex MS-DOS-specific commands
14861 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
14862 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
14863 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
14864 top of real-mode DOS systems and their emulations.
14866 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
14867 defines a few commands specific to the @sc{djgpp} port. This
14868 subsection describes those commands.
14873 This is a prefix of @sc{djgpp}-specific commands which print
14874 information about the target system and important OS structures.
14877 @cindex MS-DOS system info
14878 @cindex free memory information (MS-DOS)
14879 @item info dos sysinfo
14880 This command displays assorted information about the underlying
14881 platform: the CPU type and features, the OS version and flavor, the
14882 DPMI version, and the available conventional and DPMI memory.
14887 @cindex segment descriptor tables
14888 @cindex descriptor tables display
14890 @itemx info dos ldt
14891 @itemx info dos idt
14892 These 3 commands display entries from, respectively, Global, Local,
14893 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
14894 tables are data structures which store a descriptor for each segment
14895 that is currently in use. The segment's selector is an index into a
14896 descriptor table; the table entry for that index holds the
14897 descriptor's base address and limit, and its attributes and access
14900 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
14901 segment (used for both data and the stack), and a DOS segment (which
14902 allows access to DOS/BIOS data structures and absolute addresses in
14903 conventional memory). However, the DPMI host will usually define
14904 additional segments in order to support the DPMI environment.
14906 @cindex garbled pointers
14907 These commands allow to display entries from the descriptor tables.
14908 Without an argument, all entries from the specified table are
14909 displayed. An argument, which should be an integer expression, means
14910 display a single entry whose index is given by the argument. For
14911 example, here's a convenient way to display information about the
14912 debugged program's data segment:
14915 @exdent @code{(@value{GDBP}) info dos ldt $ds}
14916 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
14920 This comes in handy when you want to see whether a pointer is outside
14921 the data segment's limit (i.e.@: @dfn{garbled}).
14923 @cindex page tables display (MS-DOS)
14925 @itemx info dos pte
14926 These two commands display entries from, respectively, the Page
14927 Directory and the Page Tables. Page Directories and Page Tables are
14928 data structures which control how virtual memory addresses are mapped
14929 into physical addresses. A Page Table includes an entry for every
14930 page of memory that is mapped into the program's address space; there
14931 may be several Page Tables, each one holding up to 4096 entries. A
14932 Page Directory has up to 4096 entries, one each for every Page Table
14933 that is currently in use.
14935 Without an argument, @kbd{info dos pde} displays the entire Page
14936 Directory, and @kbd{info dos pte} displays all the entries in all of
14937 the Page Tables. An argument, an integer expression, given to the
14938 @kbd{info dos pde} command means display only that entry from the Page
14939 Directory table. An argument given to the @kbd{info dos pte} command
14940 means display entries from a single Page Table, the one pointed to by
14941 the specified entry in the Page Directory.
14943 @cindex direct memory access (DMA) on MS-DOS
14944 These commands are useful when your program uses @dfn{DMA} (Direct
14945 Memory Access), which needs physical addresses to program the DMA
14948 These commands are supported only with some DPMI servers.
14950 @cindex physical address from linear address
14951 @item info dos address-pte @var{addr}
14952 This command displays the Page Table entry for a specified linear
14953 address. The argument @var{addr} is a linear address which should
14954 already have the appropriate segment's base address added to it,
14955 because this command accepts addresses which may belong to @emph{any}
14956 segment. For example, here's how to display the Page Table entry for
14957 the page where a variable @code{i} is stored:
14960 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
14961 @exdent @code{Page Table entry for address 0x11a00d30:}
14962 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
14966 This says that @code{i} is stored at offset @code{0xd30} from the page
14967 whose physical base address is @code{0x02698000}, and shows all the
14968 attributes of that page.
14970 Note that you must cast the addresses of variables to a @code{char *},
14971 since otherwise the value of @code{__djgpp_base_address}, the base
14972 address of all variables and functions in a @sc{djgpp} program, will
14973 be added using the rules of C pointer arithmetics: if @code{i} is
14974 declared an @code{int}, @value{GDBN} will add 4 times the value of
14975 @code{__djgpp_base_address} to the address of @code{i}.
14977 Here's another example, it displays the Page Table entry for the
14981 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
14982 @exdent @code{Page Table entry for address 0x29110:}
14983 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
14987 (The @code{+ 3} offset is because the transfer buffer's address is the
14988 3rd member of the @code{_go32_info_block} structure.) The output
14989 clearly shows that this DPMI server maps the addresses in conventional
14990 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
14991 linear (@code{0x29110}) addresses are identical.
14993 This command is supported only with some DPMI servers.
14996 @cindex DOS serial data link, remote debugging
14997 In addition to native debugging, the DJGPP port supports remote
14998 debugging via a serial data link. The following commands are specific
14999 to remote serial debugging in the DJGPP port of @value{GDBN}.
15002 @kindex set com1base
15003 @kindex set com1irq
15004 @kindex set com2base
15005 @kindex set com2irq
15006 @kindex set com3base
15007 @kindex set com3irq
15008 @kindex set com4base
15009 @kindex set com4irq
15010 @item set com1base @var{addr}
15011 This command sets the base I/O port address of the @file{COM1} serial
15014 @item set com1irq @var{irq}
15015 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
15016 for the @file{COM1} serial port.
15018 There are similar commands @samp{set com2base}, @samp{set com3irq},
15019 etc.@: for setting the port address and the @code{IRQ} lines for the
15022 @kindex show com1base
15023 @kindex show com1irq
15024 @kindex show com2base
15025 @kindex show com2irq
15026 @kindex show com3base
15027 @kindex show com3irq
15028 @kindex show com4base
15029 @kindex show com4irq
15030 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
15031 display the current settings of the base address and the @code{IRQ}
15032 lines used by the COM ports.
15035 @kindex info serial
15036 @cindex DOS serial port status
15037 This command prints the status of the 4 DOS serial ports. For each
15038 port, it prints whether it's active or not, its I/O base address and
15039 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
15040 counts of various errors encountered so far.
15044 @node Cygwin Native
15045 @subsection Features for Debugging MS Windows PE Executables
15046 @cindex MS Windows debugging
15047 @cindex native Cygwin debugging
15048 @cindex Cygwin-specific commands
15050 @value{GDBN} supports native debugging of MS Windows programs, including
15051 DLLs with and without symbolic debugging information. There are various
15052 additional Cygwin-specific commands, described in this section.
15053 Working with DLLs that have no debugging symbols is described in
15054 @ref{Non-debug DLL Symbols}.
15059 This is a prefix of MS Windows-specific commands which print
15060 information about the target system and important OS structures.
15062 @item info w32 selector
15063 This command displays information returned by
15064 the Win32 API @code{GetThreadSelectorEntry} function.
15065 It takes an optional argument that is evaluated to
15066 a long value to give the information about this given selector.
15067 Without argument, this command displays information
15068 about the six segment registers.
15072 This is a Cygwin-specific alias of @code{info shared}.
15074 @kindex dll-symbols
15076 This command loads symbols from a dll similarly to
15077 add-sym command but without the need to specify a base address.
15079 @kindex set cygwin-exceptions
15080 @cindex debugging the Cygwin DLL
15081 @cindex Cygwin DLL, debugging
15082 @item set cygwin-exceptions @var{mode}
15083 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
15084 happen inside the Cygwin DLL. If @var{mode} is @code{off},
15085 @value{GDBN} will delay recognition of exceptions, and may ignore some
15086 exceptions which seem to be caused by internal Cygwin DLL
15087 ``bookkeeping''. This option is meant primarily for debugging the
15088 Cygwin DLL itself; the default value is @code{off} to avoid annoying
15089 @value{GDBN} users with false @code{SIGSEGV} signals.
15091 @kindex show cygwin-exceptions
15092 @item show cygwin-exceptions
15093 Displays whether @value{GDBN} will break on exceptions that happen
15094 inside the Cygwin DLL itself.
15096 @kindex set new-console
15097 @item set new-console @var{mode}
15098 If @var{mode} is @code{on} the debuggee will
15099 be started in a new console on next start.
15100 If @var{mode} is @code{off}i, the debuggee will
15101 be started in the same console as the debugger.
15103 @kindex show new-console
15104 @item show new-console
15105 Displays whether a new console is used
15106 when the debuggee is started.
15108 @kindex set new-group
15109 @item set new-group @var{mode}
15110 This boolean value controls whether the debuggee should
15111 start a new group or stay in the same group as the debugger.
15112 This affects the way the Windows OS handles
15115 @kindex show new-group
15116 @item show new-group
15117 Displays current value of new-group boolean.
15119 @kindex set debugevents
15120 @item set debugevents
15121 This boolean value adds debug output concerning kernel events related
15122 to the debuggee seen by the debugger. This includes events that
15123 signal thread and process creation and exit, DLL loading and
15124 unloading, console interrupts, and debugging messages produced by the
15125 Windows @code{OutputDebugString} API call.
15127 @kindex set debugexec
15128 @item set debugexec
15129 This boolean value adds debug output concerning execute events
15130 (such as resume thread) seen by the debugger.
15132 @kindex set debugexceptions
15133 @item set debugexceptions
15134 This boolean value adds debug output concerning exceptions in the
15135 debuggee seen by the debugger.
15137 @kindex set debugmemory
15138 @item set debugmemory
15139 This boolean value adds debug output concerning debuggee memory reads
15140 and writes by the debugger.
15144 This boolean values specifies whether the debuggee is called
15145 via a shell or directly (default value is on).
15149 Displays if the debuggee will be started with a shell.
15154 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
15157 @node Non-debug DLL Symbols
15158 @subsubsection Support for DLLs without Debugging Symbols
15159 @cindex DLLs with no debugging symbols
15160 @cindex Minimal symbols and DLLs
15162 Very often on windows, some of the DLLs that your program relies on do
15163 not include symbolic debugging information (for example,
15164 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
15165 symbols in a DLL, it relies on the minimal amount of symbolic
15166 information contained in the DLL's export table. This section
15167 describes working with such symbols, known internally to @value{GDBN} as
15168 ``minimal symbols''.
15170 Note that before the debugged program has started execution, no DLLs
15171 will have been loaded. The easiest way around this problem is simply to
15172 start the program --- either by setting a breakpoint or letting the
15173 program run once to completion. It is also possible to force
15174 @value{GDBN} to load a particular DLL before starting the executable ---
15175 see the shared library information in @ref{Files}, or the
15176 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
15177 explicitly loading symbols from a DLL with no debugging information will
15178 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
15179 which may adversely affect symbol lookup performance.
15181 @subsubsection DLL Name Prefixes
15183 In keeping with the naming conventions used by the Microsoft debugging
15184 tools, DLL export symbols are made available with a prefix based on the
15185 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
15186 also entered into the symbol table, so @code{CreateFileA} is often
15187 sufficient. In some cases there will be name clashes within a program
15188 (particularly if the executable itself includes full debugging symbols)
15189 necessitating the use of the fully qualified name when referring to the
15190 contents of the DLL. Use single-quotes around the name to avoid the
15191 exclamation mark (``!'') being interpreted as a language operator.
15193 Note that the internal name of the DLL may be all upper-case, even
15194 though the file name of the DLL is lower-case, or vice-versa. Since
15195 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
15196 some confusion. If in doubt, try the @code{info functions} and
15197 @code{info variables} commands or even @code{maint print msymbols}
15198 (@pxref{Symbols}). Here's an example:
15201 (@value{GDBP}) info function CreateFileA
15202 All functions matching regular expression "CreateFileA":
15204 Non-debugging symbols:
15205 0x77e885f4 CreateFileA
15206 0x77e885f4 KERNEL32!CreateFileA
15210 (@value{GDBP}) info function !
15211 All functions matching regular expression "!":
15213 Non-debugging symbols:
15214 0x6100114c cygwin1!__assert
15215 0x61004034 cygwin1!_dll_crt0@@0
15216 0x61004240 cygwin1!dll_crt0(per_process *)
15220 @subsubsection Working with Minimal Symbols
15222 Symbols extracted from a DLL's export table do not contain very much
15223 type information. All that @value{GDBN} can do is guess whether a symbol
15224 refers to a function or variable depending on the linker section that
15225 contains the symbol. Also note that the actual contents of the memory
15226 contained in a DLL are not available unless the program is running. This
15227 means that you cannot examine the contents of a variable or disassemble
15228 a function within a DLL without a running program.
15230 Variables are generally treated as pointers and dereferenced
15231 automatically. For this reason, it is often necessary to prefix a
15232 variable name with the address-of operator (``&'') and provide explicit
15233 type information in the command. Here's an example of the type of
15237 (@value{GDBP}) print 'cygwin1!__argv'
15242 (@value{GDBP}) x 'cygwin1!__argv'
15243 0x10021610: "\230y\""
15246 And two possible solutions:
15249 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
15250 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
15254 (@value{GDBP}) x/2x &'cygwin1!__argv'
15255 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
15256 (@value{GDBP}) x/x 0x10021608
15257 0x10021608: 0x0022fd98
15258 (@value{GDBP}) x/s 0x0022fd98
15259 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
15262 Setting a break point within a DLL is possible even before the program
15263 starts execution. However, under these circumstances, @value{GDBN} can't
15264 examine the initial instructions of the function in order to skip the
15265 function's frame set-up code. You can work around this by using ``*&''
15266 to set the breakpoint at a raw memory address:
15269 (@value{GDBP}) break *&'python22!PyOS_Readline'
15270 Breakpoint 1 at 0x1e04eff0
15273 The author of these extensions is not entirely convinced that setting a
15274 break point within a shared DLL like @file{kernel32.dll} is completely
15278 @subsection Commands Specific to @sc{gnu} Hurd Systems
15279 @cindex @sc{gnu} Hurd debugging
15281 This subsection describes @value{GDBN} commands specific to the
15282 @sc{gnu} Hurd native debugging.
15287 @kindex set signals@r{, Hurd command}
15288 @kindex set sigs@r{, Hurd command}
15289 This command toggles the state of inferior signal interception by
15290 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
15291 affected by this command. @code{sigs} is a shorthand alias for
15296 @kindex show signals@r{, Hurd command}
15297 @kindex show sigs@r{, Hurd command}
15298 Show the current state of intercepting inferior's signals.
15300 @item set signal-thread
15301 @itemx set sigthread
15302 @kindex set signal-thread
15303 @kindex set sigthread
15304 This command tells @value{GDBN} which thread is the @code{libc} signal
15305 thread. That thread is run when a signal is delivered to a running
15306 process. @code{set sigthread} is the shorthand alias of @code{set
15309 @item show signal-thread
15310 @itemx show sigthread
15311 @kindex show signal-thread
15312 @kindex show sigthread
15313 These two commands show which thread will run when the inferior is
15314 delivered a signal.
15317 @kindex set stopped@r{, Hurd command}
15318 This commands tells @value{GDBN} that the inferior process is stopped,
15319 as with the @code{SIGSTOP} signal. The stopped process can be
15320 continued by delivering a signal to it.
15323 @kindex show stopped@r{, Hurd command}
15324 This command shows whether @value{GDBN} thinks the debuggee is
15327 @item set exceptions
15328 @kindex set exceptions@r{, Hurd command}
15329 Use this command to turn off trapping of exceptions in the inferior.
15330 When exception trapping is off, neither breakpoints nor
15331 single-stepping will work. To restore the default, set exception
15334 @item show exceptions
15335 @kindex show exceptions@r{, Hurd command}
15336 Show the current state of trapping exceptions in the inferior.
15338 @item set task pause
15339 @kindex set task@r{, Hurd commands}
15340 @cindex task attributes (@sc{gnu} Hurd)
15341 @cindex pause current task (@sc{gnu} Hurd)
15342 This command toggles task suspension when @value{GDBN} has control.
15343 Setting it to on takes effect immediately, and the task is suspended
15344 whenever @value{GDBN} gets control. Setting it to off will take
15345 effect the next time the inferior is continued. If this option is set
15346 to off, you can use @code{set thread default pause on} or @code{set
15347 thread pause on} (see below) to pause individual threads.
15349 @item show task pause
15350 @kindex show task@r{, Hurd commands}
15351 Show the current state of task suspension.
15353 @item set task detach-suspend-count
15354 @cindex task suspend count
15355 @cindex detach from task, @sc{gnu} Hurd
15356 This command sets the suspend count the task will be left with when
15357 @value{GDBN} detaches from it.
15359 @item show task detach-suspend-count
15360 Show the suspend count the task will be left with when detaching.
15362 @item set task exception-port
15363 @itemx set task excp
15364 @cindex task exception port, @sc{gnu} Hurd
15365 This command sets the task exception port to which @value{GDBN} will
15366 forward exceptions. The argument should be the value of the @dfn{send
15367 rights} of the task. @code{set task excp} is a shorthand alias.
15369 @item set noninvasive
15370 @cindex noninvasive task options
15371 This command switches @value{GDBN} to a mode that is the least
15372 invasive as far as interfering with the inferior is concerned. This
15373 is the same as using @code{set task pause}, @code{set exceptions}, and
15374 @code{set signals} to values opposite to the defaults.
15376 @item info send-rights
15377 @itemx info receive-rights
15378 @itemx info port-rights
15379 @itemx info port-sets
15380 @itemx info dead-names
15383 @cindex send rights, @sc{gnu} Hurd
15384 @cindex receive rights, @sc{gnu} Hurd
15385 @cindex port rights, @sc{gnu} Hurd
15386 @cindex port sets, @sc{gnu} Hurd
15387 @cindex dead names, @sc{gnu} Hurd
15388 These commands display information about, respectively, send rights,
15389 receive rights, port rights, port sets, and dead names of a task.
15390 There are also shorthand aliases: @code{info ports} for @code{info
15391 port-rights} and @code{info psets} for @code{info port-sets}.
15393 @item set thread pause
15394 @kindex set thread@r{, Hurd command}
15395 @cindex thread properties, @sc{gnu} Hurd
15396 @cindex pause current thread (@sc{gnu} Hurd)
15397 This command toggles current thread suspension when @value{GDBN} has
15398 control. Setting it to on takes effect immediately, and the current
15399 thread is suspended whenever @value{GDBN} gets control. Setting it to
15400 off will take effect the next time the inferior is continued.
15401 Normally, this command has no effect, since when @value{GDBN} has
15402 control, the whole task is suspended. However, if you used @code{set
15403 task pause off} (see above), this command comes in handy to suspend
15404 only the current thread.
15406 @item show thread pause
15407 @kindex show thread@r{, Hurd command}
15408 This command shows the state of current thread suspension.
15410 @item set thread run
15411 This command sets whether the current thread is allowed to run.
15413 @item show thread run
15414 Show whether the current thread is allowed to run.
15416 @item set thread detach-suspend-count
15417 @cindex thread suspend count, @sc{gnu} Hurd
15418 @cindex detach from thread, @sc{gnu} Hurd
15419 This command sets the suspend count @value{GDBN} will leave on a
15420 thread when detaching. This number is relative to the suspend count
15421 found by @value{GDBN} when it notices the thread; use @code{set thread
15422 takeover-suspend-count} to force it to an absolute value.
15424 @item show thread detach-suspend-count
15425 Show the suspend count @value{GDBN} will leave on the thread when
15428 @item set thread exception-port
15429 @itemx set thread excp
15430 Set the thread exception port to which to forward exceptions. This
15431 overrides the port set by @code{set task exception-port} (see above).
15432 @code{set thread excp} is the shorthand alias.
15434 @item set thread takeover-suspend-count
15435 Normally, @value{GDBN}'s thread suspend counts are relative to the
15436 value @value{GDBN} finds when it notices each thread. This command
15437 changes the suspend counts to be absolute instead.
15439 @item set thread default
15440 @itemx show thread default
15441 @cindex thread default settings, @sc{gnu} Hurd
15442 Each of the above @code{set thread} commands has a @code{set thread
15443 default} counterpart (e.g., @code{set thread default pause}, @code{set
15444 thread default exception-port}, etc.). The @code{thread default}
15445 variety of commands sets the default thread properties for all
15446 threads; you can then change the properties of individual threads with
15447 the non-default commands.
15452 @subsection QNX Neutrino
15453 @cindex QNX Neutrino
15455 @value{GDBN} provides the following commands specific to the QNX
15459 @item set debug nto-debug
15460 @kindex set debug nto-debug
15461 When set to on, enables debugging messages specific to the QNX
15464 @item show debug nto-debug
15465 @kindex show debug nto-debug
15466 Show the current state of QNX Neutrino messages.
15473 @value{GDBN} provides the following commands specific to the Darwin target:
15476 @item set debug darwin @var{num}
15477 @kindex set debug darwin
15478 When set to a non zero value, enables debugging messages specific to
15479 the Darwin support. Higher values produce more verbose output.
15481 @item show debug darwin
15482 @kindex show debug darwin
15483 Show the current state of Darwin messages.
15485 @item set debug mach-o @var{num}
15486 @kindex set debug mach-o
15487 When set to a non zero value, enables debugging messages while
15488 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
15489 file format used on Darwin for object and executable files.) Higher
15490 values produce more verbose output. This is a command to diagnose
15491 problems internal to @value{GDBN} and should not be needed in normal
15494 @item show debug mach-o
15495 @kindex show debug mach-o
15496 Show the current state of Mach-O file messages.
15498 @item set mach-exceptions on
15499 @itemx set mach-exceptions off
15500 @kindex set mach-exceptions
15501 On Darwin, faults are first reported as a Mach exception and are then
15502 mapped to a Posix signal. Use this command to turn on trapping of
15503 Mach exceptions in the inferior. This might be sometimes useful to
15504 better understand the cause of a fault. The default is off.
15506 @item show mach-exceptions
15507 @kindex show mach-exceptions
15508 Show the current state of exceptions trapping.
15513 @section Embedded Operating Systems
15515 This section describes configurations involving the debugging of
15516 embedded operating systems that are available for several different
15520 * VxWorks:: Using @value{GDBN} with VxWorks
15523 @value{GDBN} includes the ability to debug programs running on
15524 various real-time operating systems.
15527 @subsection Using @value{GDBN} with VxWorks
15533 @kindex target vxworks
15534 @item target vxworks @var{machinename}
15535 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
15536 is the target system's machine name or IP address.
15540 On VxWorks, @code{load} links @var{filename} dynamically on the
15541 current target system as well as adding its symbols in @value{GDBN}.
15543 @value{GDBN} enables developers to spawn and debug tasks running on networked
15544 VxWorks targets from a Unix host. Already-running tasks spawned from
15545 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
15546 both the Unix host and on the VxWorks target. The program
15547 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
15548 installed with the name @code{vxgdb}, to distinguish it from a
15549 @value{GDBN} for debugging programs on the host itself.)
15552 @item VxWorks-timeout @var{args}
15553 @kindex vxworks-timeout
15554 All VxWorks-based targets now support the option @code{vxworks-timeout}.
15555 This option is set by the user, and @var{args} represents the number of
15556 seconds @value{GDBN} waits for responses to rpc's. You might use this if
15557 your VxWorks target is a slow software simulator or is on the far side
15558 of a thin network line.
15561 The following information on connecting to VxWorks was current when
15562 this manual was produced; newer releases of VxWorks may use revised
15565 @findex INCLUDE_RDB
15566 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
15567 to include the remote debugging interface routines in the VxWorks
15568 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
15569 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
15570 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
15571 source debugging task @code{tRdbTask} when VxWorks is booted. For more
15572 information on configuring and remaking VxWorks, see the manufacturer's
15574 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
15576 Once you have included @file{rdb.a} in your VxWorks system image and set
15577 your Unix execution search path to find @value{GDBN}, you are ready to
15578 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
15579 @code{vxgdb}, depending on your installation).
15581 @value{GDBN} comes up showing the prompt:
15588 * VxWorks Connection:: Connecting to VxWorks
15589 * VxWorks Download:: VxWorks download
15590 * VxWorks Attach:: Running tasks
15593 @node VxWorks Connection
15594 @subsubsection Connecting to VxWorks
15596 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
15597 network. To connect to a target whose host name is ``@code{tt}'', type:
15600 (vxgdb) target vxworks tt
15604 @value{GDBN} displays messages like these:
15607 Attaching remote machine across net...
15612 @value{GDBN} then attempts to read the symbol tables of any object modules
15613 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
15614 these files by searching the directories listed in the command search
15615 path (@pxref{Environment, ,Your Program's Environment}); if it fails
15616 to find an object file, it displays a message such as:
15619 prog.o: No such file or directory.
15622 When this happens, add the appropriate directory to the search path with
15623 the @value{GDBN} command @code{path}, and execute the @code{target}
15626 @node VxWorks Download
15627 @subsubsection VxWorks Download
15629 @cindex download to VxWorks
15630 If you have connected to the VxWorks target and you want to debug an
15631 object that has not yet been loaded, you can use the @value{GDBN}
15632 @code{load} command to download a file from Unix to VxWorks
15633 incrementally. The object file given as an argument to the @code{load}
15634 command is actually opened twice: first by the VxWorks target in order
15635 to download the code, then by @value{GDBN} in order to read the symbol
15636 table. This can lead to problems if the current working directories on
15637 the two systems differ. If both systems have NFS mounted the same
15638 filesystems, you can avoid these problems by using absolute paths.
15639 Otherwise, it is simplest to set the working directory on both systems
15640 to the directory in which the object file resides, and then to reference
15641 the file by its name, without any path. For instance, a program
15642 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
15643 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
15644 program, type this on VxWorks:
15647 -> cd "@var{vxpath}/vw/demo/rdb"
15651 Then, in @value{GDBN}, type:
15654 (vxgdb) cd @var{hostpath}/vw/demo/rdb
15655 (vxgdb) load prog.o
15658 @value{GDBN} displays a response similar to this:
15661 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
15664 You can also use the @code{load} command to reload an object module
15665 after editing and recompiling the corresponding source file. Note that
15666 this makes @value{GDBN} delete all currently-defined breakpoints,
15667 auto-displays, and convenience variables, and to clear the value
15668 history. (This is necessary in order to preserve the integrity of
15669 debugger's data structures that reference the target system's symbol
15672 @node VxWorks Attach
15673 @subsubsection Running Tasks
15675 @cindex running VxWorks tasks
15676 You can also attach to an existing task using the @code{attach} command as
15680 (vxgdb) attach @var{task}
15684 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
15685 or suspended when you attach to it. Running tasks are suspended at
15686 the time of attachment.
15688 @node Embedded Processors
15689 @section Embedded Processors
15691 This section goes into details specific to particular embedded
15694 @cindex send command to simulator
15695 Whenever a specific embedded processor has a simulator, @value{GDBN}
15696 allows to send an arbitrary command to the simulator.
15699 @item sim @var{command}
15700 @kindex sim@r{, a command}
15701 Send an arbitrary @var{command} string to the simulator. Consult the
15702 documentation for the specific simulator in use for information about
15703 acceptable commands.
15709 * M32R/D:: Renesas M32R/D
15710 * M68K:: Motorola M68K
15711 * MIPS Embedded:: MIPS Embedded
15712 * OpenRISC 1000:: OpenRisc 1000
15713 * PA:: HP PA Embedded
15714 * PowerPC Embedded:: PowerPC Embedded
15715 * Sparclet:: Tsqware Sparclet
15716 * Sparclite:: Fujitsu Sparclite
15717 * Z8000:: Zilog Z8000
15720 * Super-H:: Renesas Super-H
15729 @item target rdi @var{dev}
15730 ARM Angel monitor, via RDI library interface to ADP protocol. You may
15731 use this target to communicate with both boards running the Angel
15732 monitor, or with the EmbeddedICE JTAG debug device.
15735 @item target rdp @var{dev}
15740 @value{GDBN} provides the following ARM-specific commands:
15743 @item set arm disassembler
15745 This commands selects from a list of disassembly styles. The
15746 @code{"std"} style is the standard style.
15748 @item show arm disassembler
15750 Show the current disassembly style.
15752 @item set arm apcs32
15753 @cindex ARM 32-bit mode
15754 This command toggles ARM operation mode between 32-bit and 26-bit.
15756 @item show arm apcs32
15757 Display the current usage of the ARM 32-bit mode.
15759 @item set arm fpu @var{fputype}
15760 This command sets the ARM floating-point unit (FPU) type. The
15761 argument @var{fputype} can be one of these:
15765 Determine the FPU type by querying the OS ABI.
15767 Software FPU, with mixed-endian doubles on little-endian ARM
15770 GCC-compiled FPA co-processor.
15772 Software FPU with pure-endian doubles.
15778 Show the current type of the FPU.
15781 This command forces @value{GDBN} to use the specified ABI.
15784 Show the currently used ABI.
15786 @item set arm fallback-mode (arm|thumb|auto)
15787 @value{GDBN} uses the symbol table, when available, to determine
15788 whether instructions are ARM or Thumb. This command controls
15789 @value{GDBN}'s default behavior when the symbol table is not
15790 available. The default is @samp{auto}, which causes @value{GDBN} to
15791 use the current execution mode (from the @code{T} bit in the @code{CPSR}
15794 @item show arm fallback-mode
15795 Show the current fallback instruction mode.
15797 @item set arm force-mode (arm|thumb|auto)
15798 This command overrides use of the symbol table to determine whether
15799 instructions are ARM or Thumb. The default is @samp{auto}, which
15800 causes @value{GDBN} to use the symbol table and then the setting
15801 of @samp{set arm fallback-mode}.
15803 @item show arm force-mode
15804 Show the current forced instruction mode.
15806 @item set debug arm
15807 Toggle whether to display ARM-specific debugging messages from the ARM
15808 target support subsystem.
15810 @item show debug arm
15811 Show whether ARM-specific debugging messages are enabled.
15814 The following commands are available when an ARM target is debugged
15815 using the RDI interface:
15818 @item rdilogfile @r{[}@var{file}@r{]}
15820 @cindex ADP (Angel Debugger Protocol) logging
15821 Set the filename for the ADP (Angel Debugger Protocol) packet log.
15822 With an argument, sets the log file to the specified @var{file}. With
15823 no argument, show the current log file name. The default log file is
15826 @item rdilogenable @r{[}@var{arg}@r{]}
15827 @kindex rdilogenable
15828 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
15829 enables logging, with an argument 0 or @code{"no"} disables it. With
15830 no arguments displays the current setting. When logging is enabled,
15831 ADP packets exchanged between @value{GDBN} and the RDI target device
15832 are logged to a file.
15834 @item set rdiromatzero
15835 @kindex set rdiromatzero
15836 @cindex ROM at zero address, RDI
15837 Tell @value{GDBN} whether the target has ROM at address 0. If on,
15838 vector catching is disabled, so that zero address can be used. If off
15839 (the default), vector catching is enabled. For this command to take
15840 effect, it needs to be invoked prior to the @code{target rdi} command.
15842 @item show rdiromatzero
15843 @kindex show rdiromatzero
15844 Show the current setting of ROM at zero address.
15846 @item set rdiheartbeat
15847 @kindex set rdiheartbeat
15848 @cindex RDI heartbeat
15849 Enable or disable RDI heartbeat packets. It is not recommended to
15850 turn on this option, since it confuses ARM and EPI JTAG interface, as
15851 well as the Angel monitor.
15853 @item show rdiheartbeat
15854 @kindex show rdiheartbeat
15855 Show the setting of RDI heartbeat packets.
15860 @subsection Renesas M32R/D and M32R/SDI
15863 @kindex target m32r
15864 @item target m32r @var{dev}
15865 Renesas M32R/D ROM monitor.
15867 @kindex target m32rsdi
15868 @item target m32rsdi @var{dev}
15869 Renesas M32R SDI server, connected via parallel port to the board.
15872 The following @value{GDBN} commands are specific to the M32R monitor:
15875 @item set download-path @var{path}
15876 @kindex set download-path
15877 @cindex find downloadable @sc{srec} files (M32R)
15878 Set the default path for finding downloadable @sc{srec} files.
15880 @item show download-path
15881 @kindex show download-path
15882 Show the default path for downloadable @sc{srec} files.
15884 @item set board-address @var{addr}
15885 @kindex set board-address
15886 @cindex M32-EVA target board address
15887 Set the IP address for the M32R-EVA target board.
15889 @item show board-address
15890 @kindex show board-address
15891 Show the current IP address of the target board.
15893 @item set server-address @var{addr}
15894 @kindex set server-address
15895 @cindex download server address (M32R)
15896 Set the IP address for the download server, which is the @value{GDBN}'s
15899 @item show server-address
15900 @kindex show server-address
15901 Display the IP address of the download server.
15903 @item upload @r{[}@var{file}@r{]}
15904 @kindex upload@r{, M32R}
15905 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
15906 upload capability. If no @var{file} argument is given, the current
15907 executable file is uploaded.
15909 @item tload @r{[}@var{file}@r{]}
15910 @kindex tload@r{, M32R}
15911 Test the @code{upload} command.
15914 The following commands are available for M32R/SDI:
15919 @cindex reset SDI connection, M32R
15920 This command resets the SDI connection.
15924 This command shows the SDI connection status.
15927 @kindex debug_chaos
15928 @cindex M32R/Chaos debugging
15929 Instructs the remote that M32R/Chaos debugging is to be used.
15931 @item use_debug_dma
15932 @kindex use_debug_dma
15933 Instructs the remote to use the DEBUG_DMA method of accessing memory.
15936 @kindex use_mon_code
15937 Instructs the remote to use the MON_CODE method of accessing memory.
15940 @kindex use_ib_break
15941 Instructs the remote to set breakpoints by IB break.
15943 @item use_dbt_break
15944 @kindex use_dbt_break
15945 Instructs the remote to set breakpoints by DBT.
15951 The Motorola m68k configuration includes ColdFire support, and a
15952 target command for the following ROM monitor.
15956 @kindex target dbug
15957 @item target dbug @var{dev}
15958 dBUG ROM monitor for Motorola ColdFire.
15962 @node MIPS Embedded
15963 @subsection MIPS Embedded
15965 @cindex MIPS boards
15966 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
15967 MIPS board attached to a serial line. This is available when
15968 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
15971 Use these @value{GDBN} commands to specify the connection to your target board:
15974 @item target mips @var{port}
15975 @kindex target mips @var{port}
15976 To run a program on the board, start up @code{@value{GDBP}} with the
15977 name of your program as the argument. To connect to the board, use the
15978 command @samp{target mips @var{port}}, where @var{port} is the name of
15979 the serial port connected to the board. If the program has not already
15980 been downloaded to the board, you may use the @code{load} command to
15981 download it. You can then use all the usual @value{GDBN} commands.
15983 For example, this sequence connects to the target board through a serial
15984 port, and loads and runs a program called @var{prog} through the
15988 host$ @value{GDBP} @var{prog}
15989 @value{GDBN} is free software and @dots{}
15990 (@value{GDBP}) target mips /dev/ttyb
15991 (@value{GDBP}) load @var{prog}
15995 @item target mips @var{hostname}:@var{portnumber}
15996 On some @value{GDBN} host configurations, you can specify a TCP
15997 connection (for instance, to a serial line managed by a terminal
15998 concentrator) instead of a serial port, using the syntax
15999 @samp{@var{hostname}:@var{portnumber}}.
16001 @item target pmon @var{port}
16002 @kindex target pmon @var{port}
16005 @item target ddb @var{port}
16006 @kindex target ddb @var{port}
16007 NEC's DDB variant of PMON for Vr4300.
16009 @item target lsi @var{port}
16010 @kindex target lsi @var{port}
16011 LSI variant of PMON.
16013 @kindex target r3900
16014 @item target r3900 @var{dev}
16015 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
16017 @kindex target array
16018 @item target array @var{dev}
16019 Array Tech LSI33K RAID controller board.
16025 @value{GDBN} also supports these special commands for MIPS targets:
16028 @item set mipsfpu double
16029 @itemx set mipsfpu single
16030 @itemx set mipsfpu none
16031 @itemx set mipsfpu auto
16032 @itemx show mipsfpu
16033 @kindex set mipsfpu
16034 @kindex show mipsfpu
16035 @cindex MIPS remote floating point
16036 @cindex floating point, MIPS remote
16037 If your target board does not support the MIPS floating point
16038 coprocessor, you should use the command @samp{set mipsfpu none} (if you
16039 need this, you may wish to put the command in your @value{GDBN} init
16040 file). This tells @value{GDBN} how to find the return value of
16041 functions which return floating point values. It also allows
16042 @value{GDBN} to avoid saving the floating point registers when calling
16043 functions on the board. If you are using a floating point coprocessor
16044 with only single precision floating point support, as on the @sc{r4650}
16045 processor, use the command @samp{set mipsfpu single}. The default
16046 double precision floating point coprocessor may be selected using
16047 @samp{set mipsfpu double}.
16049 In previous versions the only choices were double precision or no
16050 floating point, so @samp{set mipsfpu on} will select double precision
16051 and @samp{set mipsfpu off} will select no floating point.
16053 As usual, you can inquire about the @code{mipsfpu} variable with
16054 @samp{show mipsfpu}.
16056 @item set timeout @var{seconds}
16057 @itemx set retransmit-timeout @var{seconds}
16058 @itemx show timeout
16059 @itemx show retransmit-timeout
16060 @cindex @code{timeout}, MIPS protocol
16061 @cindex @code{retransmit-timeout}, MIPS protocol
16062 @kindex set timeout
16063 @kindex show timeout
16064 @kindex set retransmit-timeout
16065 @kindex show retransmit-timeout
16066 You can control the timeout used while waiting for a packet, in the MIPS
16067 remote protocol, with the @code{set timeout @var{seconds}} command. The
16068 default is 5 seconds. Similarly, you can control the timeout used while
16069 waiting for an acknowledgment of a packet with the @code{set
16070 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
16071 You can inspect both values with @code{show timeout} and @code{show
16072 retransmit-timeout}. (These commands are @emph{only} available when
16073 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
16075 The timeout set by @code{set timeout} does not apply when @value{GDBN}
16076 is waiting for your program to stop. In that case, @value{GDBN} waits
16077 forever because it has no way of knowing how long the program is going
16078 to run before stopping.
16080 @item set syn-garbage-limit @var{num}
16081 @kindex set syn-garbage-limit@r{, MIPS remote}
16082 @cindex synchronize with remote MIPS target
16083 Limit the maximum number of characters @value{GDBN} should ignore when
16084 it tries to synchronize with the remote target. The default is 10
16085 characters. Setting the limit to -1 means there's no limit.
16087 @item show syn-garbage-limit
16088 @kindex show syn-garbage-limit@r{, MIPS remote}
16089 Show the current limit on the number of characters to ignore when
16090 trying to synchronize with the remote system.
16092 @item set monitor-prompt @var{prompt}
16093 @kindex set monitor-prompt@r{, MIPS remote}
16094 @cindex remote monitor prompt
16095 Tell @value{GDBN} to expect the specified @var{prompt} string from the
16096 remote monitor. The default depends on the target:
16106 @item show monitor-prompt
16107 @kindex show monitor-prompt@r{, MIPS remote}
16108 Show the current strings @value{GDBN} expects as the prompt from the
16111 @item set monitor-warnings
16112 @kindex set monitor-warnings@r{, MIPS remote}
16113 Enable or disable monitor warnings about hardware breakpoints. This
16114 has effect only for the @code{lsi} target. When on, @value{GDBN} will
16115 display warning messages whose codes are returned by the @code{lsi}
16116 PMON monitor for breakpoint commands.
16118 @item show monitor-warnings
16119 @kindex show monitor-warnings@r{, MIPS remote}
16120 Show the current setting of printing monitor warnings.
16122 @item pmon @var{command}
16123 @kindex pmon@r{, MIPS remote}
16124 @cindex send PMON command
16125 This command allows sending an arbitrary @var{command} string to the
16126 monitor. The monitor must be in debug mode for this to work.
16129 @node OpenRISC 1000
16130 @subsection OpenRISC 1000
16131 @cindex OpenRISC 1000
16133 @cindex or1k boards
16134 See OR1k Architecture document (@uref{www.opencores.org}) for more information
16135 about platform and commands.
16139 @kindex target jtag
16140 @item target jtag jtag://@var{host}:@var{port}
16142 Connects to remote JTAG server.
16143 JTAG remote server can be either an or1ksim or JTAG server,
16144 connected via parallel port to the board.
16146 Example: @code{target jtag jtag://localhost:9999}
16149 @item or1ksim @var{command}
16150 If connected to @code{or1ksim} OpenRISC 1000 Architectural
16151 Simulator, proprietary commands can be executed.
16153 @kindex info or1k spr
16154 @item info or1k spr
16155 Displays spr groups.
16157 @item info or1k spr @var{group}
16158 @itemx info or1k spr @var{groupno}
16159 Displays register names in selected group.
16161 @item info or1k spr @var{group} @var{register}
16162 @itemx info or1k spr @var{register}
16163 @itemx info or1k spr @var{groupno} @var{registerno}
16164 @itemx info or1k spr @var{registerno}
16165 Shows information about specified spr register.
16168 @item spr @var{group} @var{register} @var{value}
16169 @itemx spr @var{register @var{value}}
16170 @itemx spr @var{groupno} @var{registerno @var{value}}
16171 @itemx spr @var{registerno @var{value}}
16172 Writes @var{value} to specified spr register.
16175 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
16176 It is very similar to @value{GDBN} trace, except it does not interfere with normal
16177 program execution and is thus much faster. Hardware breakpoints/watchpoint
16178 triggers can be set using:
16181 Load effective address/data
16183 Store effective address/data
16185 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
16190 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
16191 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
16193 @code{htrace} commands:
16194 @cindex OpenRISC 1000 htrace
16197 @item hwatch @var{conditional}
16198 Set hardware watchpoint on combination of Load/Store Effective Address(es)
16199 or Data. For example:
16201 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
16203 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
16207 Display information about current HW trace configuration.
16209 @item htrace trigger @var{conditional}
16210 Set starting criteria for HW trace.
16212 @item htrace qualifier @var{conditional}
16213 Set acquisition qualifier for HW trace.
16215 @item htrace stop @var{conditional}
16216 Set HW trace stopping criteria.
16218 @item htrace record [@var{data}]*
16219 Selects the data to be recorded, when qualifier is met and HW trace was
16222 @item htrace enable
16223 @itemx htrace disable
16224 Enables/disables the HW trace.
16226 @item htrace rewind [@var{filename}]
16227 Clears currently recorded trace data.
16229 If filename is specified, new trace file is made and any newly collected data
16230 will be written there.
16232 @item htrace print [@var{start} [@var{len}]]
16233 Prints trace buffer, using current record configuration.
16235 @item htrace mode continuous
16236 Set continuous trace mode.
16238 @item htrace mode suspend
16239 Set suspend trace mode.
16243 @node PowerPC Embedded
16244 @subsection PowerPC Embedded
16246 @value{GDBN} provides the following PowerPC-specific commands:
16249 @kindex set powerpc
16250 @item set powerpc soft-float
16251 @itemx show powerpc soft-float
16252 Force @value{GDBN} to use (or not use) a software floating point calling
16253 convention. By default, @value{GDBN} selects the calling convention based
16254 on the selected architecture and the provided executable file.
16256 @item set powerpc vector-abi
16257 @itemx show powerpc vector-abi
16258 Force @value{GDBN} to use the specified calling convention for vector
16259 arguments and return values. The valid options are @samp{auto};
16260 @samp{generic}, to avoid vector registers even if they are present;
16261 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
16262 registers. By default, @value{GDBN} selects the calling convention
16263 based on the selected architecture and the provided executable file.
16265 @kindex target dink32
16266 @item target dink32 @var{dev}
16267 DINK32 ROM monitor.
16269 @kindex target ppcbug
16270 @item target ppcbug @var{dev}
16271 @kindex target ppcbug1
16272 @item target ppcbug1 @var{dev}
16273 PPCBUG ROM monitor for PowerPC.
16276 @item target sds @var{dev}
16277 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
16280 @cindex SDS protocol
16281 The following commands specific to the SDS protocol are supported
16285 @item set sdstimeout @var{nsec}
16286 @kindex set sdstimeout
16287 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
16288 default is 2 seconds.
16290 @item show sdstimeout
16291 @kindex show sdstimeout
16292 Show the current value of the SDS timeout.
16294 @item sds @var{command}
16295 @kindex sds@r{, a command}
16296 Send the specified @var{command} string to the SDS monitor.
16301 @subsection HP PA Embedded
16305 @kindex target op50n
16306 @item target op50n @var{dev}
16307 OP50N monitor, running on an OKI HPPA board.
16309 @kindex target w89k
16310 @item target w89k @var{dev}
16311 W89K monitor, running on a Winbond HPPA board.
16316 @subsection Tsqware Sparclet
16320 @value{GDBN} enables developers to debug tasks running on
16321 Sparclet targets from a Unix host.
16322 @value{GDBN} uses code that runs on
16323 both the Unix host and on the Sparclet target. The program
16324 @code{@value{GDBP}} is installed and executed on the Unix host.
16327 @item remotetimeout @var{args}
16328 @kindex remotetimeout
16329 @value{GDBN} supports the option @code{remotetimeout}.
16330 This option is set by the user, and @var{args} represents the number of
16331 seconds @value{GDBN} waits for responses.
16334 @cindex compiling, on Sparclet
16335 When compiling for debugging, include the options @samp{-g} to get debug
16336 information and @samp{-Ttext} to relocate the program to where you wish to
16337 load it on the target. You may also want to add the options @samp{-n} or
16338 @samp{-N} in order to reduce the size of the sections. Example:
16341 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
16344 You can use @code{objdump} to verify that the addresses are what you intended:
16347 sparclet-aout-objdump --headers --syms prog
16350 @cindex running, on Sparclet
16352 your Unix execution search path to find @value{GDBN}, you are ready to
16353 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
16354 (or @code{sparclet-aout-gdb}, depending on your installation).
16356 @value{GDBN} comes up showing the prompt:
16363 * Sparclet File:: Setting the file to debug
16364 * Sparclet Connection:: Connecting to Sparclet
16365 * Sparclet Download:: Sparclet download
16366 * Sparclet Execution:: Running and debugging
16369 @node Sparclet File
16370 @subsubsection Setting File to Debug
16372 The @value{GDBN} command @code{file} lets you choose with program to debug.
16375 (gdbslet) file prog
16379 @value{GDBN} then attempts to read the symbol table of @file{prog}.
16380 @value{GDBN} locates
16381 the file by searching the directories listed in the command search
16383 If the file was compiled with debug information (option @samp{-g}), source
16384 files will be searched as well.
16385 @value{GDBN} locates
16386 the source files by searching the directories listed in the directory search
16387 path (@pxref{Environment, ,Your Program's Environment}).
16389 to find a file, it displays a message such as:
16392 prog: No such file or directory.
16395 When this happens, add the appropriate directories to the search paths with
16396 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
16397 @code{target} command again.
16399 @node Sparclet Connection
16400 @subsubsection Connecting to Sparclet
16402 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
16403 To connect to a target on serial port ``@code{ttya}'', type:
16406 (gdbslet) target sparclet /dev/ttya
16407 Remote target sparclet connected to /dev/ttya
16408 main () at ../prog.c:3
16412 @value{GDBN} displays messages like these:
16418 @node Sparclet Download
16419 @subsubsection Sparclet Download
16421 @cindex download to Sparclet
16422 Once connected to the Sparclet target,
16423 you can use the @value{GDBN}
16424 @code{load} command to download the file from the host to the target.
16425 The file name and load offset should be given as arguments to the @code{load}
16427 Since the file format is aout, the program must be loaded to the starting
16428 address. You can use @code{objdump} to find out what this value is. The load
16429 offset is an offset which is added to the VMA (virtual memory address)
16430 of each of the file's sections.
16431 For instance, if the program
16432 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
16433 and bss at 0x12010170, in @value{GDBN}, type:
16436 (gdbslet) load prog 0x12010000
16437 Loading section .text, size 0xdb0 vma 0x12010000
16440 If the code is loaded at a different address then what the program was linked
16441 to, you may need to use the @code{section} and @code{add-symbol-file} commands
16442 to tell @value{GDBN} where to map the symbol table.
16444 @node Sparclet Execution
16445 @subsubsection Running and Debugging
16447 @cindex running and debugging Sparclet programs
16448 You can now begin debugging the task using @value{GDBN}'s execution control
16449 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
16450 manual for the list of commands.
16454 Breakpoint 1 at 0x12010000: file prog.c, line 3.
16456 Starting program: prog
16457 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
16458 3 char *symarg = 0;
16460 4 char *execarg = "hello!";
16465 @subsection Fujitsu Sparclite
16469 @kindex target sparclite
16470 @item target sparclite @var{dev}
16471 Fujitsu sparclite boards, used only for the purpose of loading.
16472 You must use an additional command to debug the program.
16473 For example: target remote @var{dev} using @value{GDBN} standard
16479 @subsection Zilog Z8000
16482 @cindex simulator, Z8000
16483 @cindex Zilog Z8000 simulator
16485 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
16488 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
16489 unsegmented variant of the Z8000 architecture) or the Z8001 (the
16490 segmented variant). The simulator recognizes which architecture is
16491 appropriate by inspecting the object code.
16494 @item target sim @var{args}
16496 @kindex target sim@r{, with Z8000}
16497 Debug programs on a simulated CPU. If the simulator supports setup
16498 options, specify them via @var{args}.
16502 After specifying this target, you can debug programs for the simulated
16503 CPU in the same style as programs for your host computer; use the
16504 @code{file} command to load a new program image, the @code{run} command
16505 to run your program, and so on.
16507 As well as making available all the usual machine registers
16508 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
16509 additional items of information as specially named registers:
16514 Counts clock-ticks in the simulator.
16517 Counts instructions run in the simulator.
16520 Execution time in 60ths of a second.
16524 You can refer to these values in @value{GDBN} expressions with the usual
16525 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
16526 conditional breakpoint that suspends only after at least 5000
16527 simulated clock ticks.
16530 @subsection Atmel AVR
16533 When configured for debugging the Atmel AVR, @value{GDBN} supports the
16534 following AVR-specific commands:
16537 @item info io_registers
16538 @kindex info io_registers@r{, AVR}
16539 @cindex I/O registers (Atmel AVR)
16540 This command displays information about the AVR I/O registers. For
16541 each register, @value{GDBN} prints its number and value.
16548 When configured for debugging CRIS, @value{GDBN} provides the
16549 following CRIS-specific commands:
16552 @item set cris-version @var{ver}
16553 @cindex CRIS version
16554 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
16555 The CRIS version affects register names and sizes. This command is useful in
16556 case autodetection of the CRIS version fails.
16558 @item show cris-version
16559 Show the current CRIS version.
16561 @item set cris-dwarf2-cfi
16562 @cindex DWARF-2 CFI and CRIS
16563 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
16564 Change to @samp{off} when using @code{gcc-cris} whose version is below
16567 @item show cris-dwarf2-cfi
16568 Show the current state of using DWARF-2 CFI.
16570 @item set cris-mode @var{mode}
16572 Set the current CRIS mode to @var{mode}. It should only be changed when
16573 debugging in guru mode, in which case it should be set to
16574 @samp{guru} (the default is @samp{normal}).
16576 @item show cris-mode
16577 Show the current CRIS mode.
16581 @subsection Renesas Super-H
16584 For the Renesas Super-H processor, @value{GDBN} provides these
16589 @kindex regs@r{, Super-H}
16590 Show the values of all Super-H registers.
16592 @item set sh calling-convention @var{convention}
16593 @kindex set sh calling-convention
16594 Set the calling-convention used when calling functions from @value{GDBN}.
16595 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
16596 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
16597 convention. If the DWARF-2 information of the called function specifies
16598 that the function follows the Renesas calling convention, the function
16599 is called using the Renesas calling convention. If the calling convention
16600 is set to @samp{renesas}, the Renesas calling convention is always used,
16601 regardless of the DWARF-2 information. This can be used to override the
16602 default of @samp{gcc} if debug information is missing, or the compiler
16603 does not emit the DWARF-2 calling convention entry for a function.
16605 @item show sh calling-convention
16606 @kindex show sh calling-convention
16607 Show the current calling convention setting.
16612 @node Architectures
16613 @section Architectures
16615 This section describes characteristics of architectures that affect
16616 all uses of @value{GDBN} with the architecture, both native and cross.
16623 * HPPA:: HP PA architecture
16624 * SPU:: Cell Broadband Engine SPU architecture
16629 @subsection x86 Architecture-specific Issues
16632 @item set struct-convention @var{mode}
16633 @kindex set struct-convention
16634 @cindex struct return convention
16635 @cindex struct/union returned in registers
16636 Set the convention used by the inferior to return @code{struct}s and
16637 @code{union}s from functions to @var{mode}. Possible values of
16638 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
16639 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
16640 are returned on the stack, while @code{"reg"} means that a
16641 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
16642 be returned in a register.
16644 @item show struct-convention
16645 @kindex show struct-convention
16646 Show the current setting of the convention to return @code{struct}s
16655 @kindex set rstack_high_address
16656 @cindex AMD 29K register stack
16657 @cindex register stack, AMD29K
16658 @item set rstack_high_address @var{address}
16659 On AMD 29000 family processors, registers are saved in a separate
16660 @dfn{register stack}. There is no way for @value{GDBN} to determine the
16661 extent of this stack. Normally, @value{GDBN} just assumes that the
16662 stack is ``large enough''. This may result in @value{GDBN} referencing
16663 memory locations that do not exist. If necessary, you can get around
16664 this problem by specifying the ending address of the register stack with
16665 the @code{set rstack_high_address} command. The argument should be an
16666 address, which you probably want to precede with @samp{0x} to specify in
16669 @kindex show rstack_high_address
16670 @item show rstack_high_address
16671 Display the current limit of the register stack, on AMD 29000 family
16679 See the following section.
16684 @cindex stack on Alpha
16685 @cindex stack on MIPS
16686 @cindex Alpha stack
16688 Alpha- and MIPS-based computers use an unusual stack frame, which
16689 sometimes requires @value{GDBN} to search backward in the object code to
16690 find the beginning of a function.
16692 @cindex response time, MIPS debugging
16693 To improve response time (especially for embedded applications, where
16694 @value{GDBN} may be restricted to a slow serial line for this search)
16695 you may want to limit the size of this search, using one of these
16699 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
16700 @item set heuristic-fence-post @var{limit}
16701 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
16702 search for the beginning of a function. A value of @var{0} (the
16703 default) means there is no limit. However, except for @var{0}, the
16704 larger the limit the more bytes @code{heuristic-fence-post} must search
16705 and therefore the longer it takes to run. You should only need to use
16706 this command when debugging a stripped executable.
16708 @item show heuristic-fence-post
16709 Display the current limit.
16713 These commands are available @emph{only} when @value{GDBN} is configured
16714 for debugging programs on Alpha or MIPS processors.
16716 Several MIPS-specific commands are available when debugging MIPS
16720 @item set mips abi @var{arg}
16721 @kindex set mips abi
16722 @cindex set ABI for MIPS
16723 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
16724 values of @var{arg} are:
16728 The default ABI associated with the current binary (this is the
16739 @item show mips abi
16740 @kindex show mips abi
16741 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
16744 @itemx show mipsfpu
16745 @xref{MIPS Embedded, set mipsfpu}.
16747 @item set mips mask-address @var{arg}
16748 @kindex set mips mask-address
16749 @cindex MIPS addresses, masking
16750 This command determines whether the most-significant 32 bits of 64-bit
16751 MIPS addresses are masked off. The argument @var{arg} can be
16752 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
16753 setting, which lets @value{GDBN} determine the correct value.
16755 @item show mips mask-address
16756 @kindex show mips mask-address
16757 Show whether the upper 32 bits of MIPS addresses are masked off or
16760 @item set remote-mips64-transfers-32bit-regs
16761 @kindex set remote-mips64-transfers-32bit-regs
16762 This command controls compatibility with 64-bit MIPS targets that
16763 transfer data in 32-bit quantities. If you have an old MIPS 64 target
16764 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
16765 and 64 bits for other registers, set this option to @samp{on}.
16767 @item show remote-mips64-transfers-32bit-regs
16768 @kindex show remote-mips64-transfers-32bit-regs
16769 Show the current setting of compatibility with older MIPS 64 targets.
16771 @item set debug mips
16772 @kindex set debug mips
16773 This command turns on and off debugging messages for the MIPS-specific
16774 target code in @value{GDBN}.
16776 @item show debug mips
16777 @kindex show debug mips
16778 Show the current setting of MIPS debugging messages.
16784 @cindex HPPA support
16786 When @value{GDBN} is debugging the HP PA architecture, it provides the
16787 following special commands:
16790 @item set debug hppa
16791 @kindex set debug hppa
16792 This command determines whether HPPA architecture-specific debugging
16793 messages are to be displayed.
16795 @item show debug hppa
16796 Show whether HPPA debugging messages are displayed.
16798 @item maint print unwind @var{address}
16799 @kindex maint print unwind@r{, HPPA}
16800 This command displays the contents of the unwind table entry at the
16801 given @var{address}.
16807 @subsection Cell Broadband Engine SPU architecture
16808 @cindex Cell Broadband Engine
16811 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
16812 it provides the following special commands:
16815 @item info spu event
16817 Display SPU event facility status. Shows current event mask
16818 and pending event status.
16820 @item info spu signal
16821 Display SPU signal notification facility status. Shows pending
16822 signal-control word and signal notification mode of both signal
16823 notification channels.
16825 @item info spu mailbox
16826 Display SPU mailbox facility status. Shows all pending entries,
16827 in order of processing, in each of the SPU Write Outbound,
16828 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
16831 Display MFC DMA status. Shows all pending commands in the MFC
16832 DMA queue. For each entry, opcode, tag, class IDs, effective
16833 and local store addresses and transfer size are shown.
16835 @item info spu proxydma
16836 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
16837 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
16838 and local store addresses and transfer size are shown.
16843 @subsection PowerPC
16844 @cindex PowerPC architecture
16846 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
16847 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
16848 numbers stored in the floating point registers. These values must be stored
16849 in two consecutive registers, always starting at an even register like
16850 @code{f0} or @code{f2}.
16852 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
16853 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
16854 @code{f2} and @code{f3} for @code{$dl1} and so on.
16856 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
16857 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
16860 @node Controlling GDB
16861 @chapter Controlling @value{GDBN}
16863 You can alter the way @value{GDBN} interacts with you by using the
16864 @code{set} command. For commands controlling how @value{GDBN} displays
16865 data, see @ref{Print Settings, ,Print Settings}. Other settings are
16870 * Editing:: Command editing
16871 * Command History:: Command history
16872 * Screen Size:: Screen size
16873 * Numbers:: Numbers
16874 * ABI:: Configuring the current ABI
16875 * Messages/Warnings:: Optional warnings and messages
16876 * Debugging Output:: Optional messages about internal happenings
16884 @value{GDBN} indicates its readiness to read a command by printing a string
16885 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
16886 can change the prompt string with the @code{set prompt} command. For
16887 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
16888 the prompt in one of the @value{GDBN} sessions so that you can always tell
16889 which one you are talking to.
16891 @emph{Note:} @code{set prompt} does not add a space for you after the
16892 prompt you set. This allows you to set a prompt which ends in a space
16893 or a prompt that does not.
16897 @item set prompt @var{newprompt}
16898 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
16900 @kindex show prompt
16902 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
16906 @section Command Editing
16908 @cindex command line editing
16910 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
16911 @sc{gnu} library provides consistent behavior for programs which provide a
16912 command line interface to the user. Advantages are @sc{gnu} Emacs-style
16913 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
16914 substitution, and a storage and recall of command history across
16915 debugging sessions.
16917 You may control the behavior of command line editing in @value{GDBN} with the
16918 command @code{set}.
16921 @kindex set editing
16924 @itemx set editing on
16925 Enable command line editing (enabled by default).
16927 @item set editing off
16928 Disable command line editing.
16930 @kindex show editing
16932 Show whether command line editing is enabled.
16935 @xref{Command Line Editing}, for more details about the Readline
16936 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
16937 encouraged to read that chapter.
16939 @node Command History
16940 @section Command History
16941 @cindex command history
16943 @value{GDBN} can keep track of the commands you type during your
16944 debugging sessions, so that you can be certain of precisely what
16945 happened. Use these commands to manage the @value{GDBN} command
16948 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
16949 package, to provide the history facility. @xref{Using History
16950 Interactively}, for the detailed description of the History library.
16952 To issue a command to @value{GDBN} without affecting certain aspects of
16953 the state which is seen by users, prefix it with @samp{server }
16954 (@pxref{Server Prefix}). This
16955 means that this command will not affect the command history, nor will it
16956 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
16957 pressed on a line by itself.
16959 @cindex @code{server}, command prefix
16960 The server prefix does not affect the recording of values into the value
16961 history; to print a value without recording it into the value history,
16962 use the @code{output} command instead of the @code{print} command.
16964 Here is the description of @value{GDBN} commands related to command
16968 @cindex history substitution
16969 @cindex history file
16970 @kindex set history filename
16971 @cindex @env{GDBHISTFILE}, environment variable
16972 @item set history filename @var{fname}
16973 Set the name of the @value{GDBN} command history file to @var{fname}.
16974 This is the file where @value{GDBN} reads an initial command history
16975 list, and where it writes the command history from this session when it
16976 exits. You can access this list through history expansion or through
16977 the history command editing characters listed below. This file defaults
16978 to the value of the environment variable @code{GDBHISTFILE}, or to
16979 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
16982 @cindex save command history
16983 @kindex set history save
16984 @item set history save
16985 @itemx set history save on
16986 Record command history in a file, whose name may be specified with the
16987 @code{set history filename} command. By default, this option is disabled.
16989 @item set history save off
16990 Stop recording command history in a file.
16992 @cindex history size
16993 @kindex set history size
16994 @cindex @env{HISTSIZE}, environment variable
16995 @item set history size @var{size}
16996 Set the number of commands which @value{GDBN} keeps in its history list.
16997 This defaults to the value of the environment variable
16998 @code{HISTSIZE}, or to 256 if this variable is not set.
17001 History expansion assigns special meaning to the character @kbd{!}.
17002 @xref{Event Designators}, for more details.
17004 @cindex history expansion, turn on/off
17005 Since @kbd{!} is also the logical not operator in C, history expansion
17006 is off by default. If you decide to enable history expansion with the
17007 @code{set history expansion on} command, you may sometimes need to
17008 follow @kbd{!} (when it is used as logical not, in an expression) with
17009 a space or a tab to prevent it from being expanded. The readline
17010 history facilities do not attempt substitution on the strings
17011 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
17013 The commands to control history expansion are:
17016 @item set history expansion on
17017 @itemx set history expansion
17018 @kindex set history expansion
17019 Enable history expansion. History expansion is off by default.
17021 @item set history expansion off
17022 Disable history expansion.
17025 @kindex show history
17027 @itemx show history filename
17028 @itemx show history save
17029 @itemx show history size
17030 @itemx show history expansion
17031 These commands display the state of the @value{GDBN} history parameters.
17032 @code{show history} by itself displays all four states.
17037 @kindex show commands
17038 @cindex show last commands
17039 @cindex display command history
17040 @item show commands
17041 Display the last ten commands in the command history.
17043 @item show commands @var{n}
17044 Print ten commands centered on command number @var{n}.
17046 @item show commands +
17047 Print ten commands just after the commands last printed.
17051 @section Screen Size
17052 @cindex size of screen
17053 @cindex pauses in output
17055 Certain commands to @value{GDBN} may produce large amounts of
17056 information output to the screen. To help you read all of it,
17057 @value{GDBN} pauses and asks you for input at the end of each page of
17058 output. Type @key{RET} when you want to continue the output, or @kbd{q}
17059 to discard the remaining output. Also, the screen width setting
17060 determines when to wrap lines of output. Depending on what is being
17061 printed, @value{GDBN} tries to break the line at a readable place,
17062 rather than simply letting it overflow onto the following line.
17064 Normally @value{GDBN} knows the size of the screen from the terminal
17065 driver software. For example, on Unix @value{GDBN} uses the termcap data base
17066 together with the value of the @code{TERM} environment variable and the
17067 @code{stty rows} and @code{stty cols} settings. If this is not correct,
17068 you can override it with the @code{set height} and @code{set
17075 @kindex show height
17076 @item set height @var{lpp}
17078 @itemx set width @var{cpl}
17080 These @code{set} commands specify a screen height of @var{lpp} lines and
17081 a screen width of @var{cpl} characters. The associated @code{show}
17082 commands display the current settings.
17084 If you specify a height of zero lines, @value{GDBN} does not pause during
17085 output no matter how long the output is. This is useful if output is to a
17086 file or to an editor buffer.
17088 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
17089 from wrapping its output.
17091 @item set pagination on
17092 @itemx set pagination off
17093 @kindex set pagination
17094 Turn the output pagination on or off; the default is on. Turning
17095 pagination off is the alternative to @code{set height 0}.
17097 @item show pagination
17098 @kindex show pagination
17099 Show the current pagination mode.
17104 @cindex number representation
17105 @cindex entering numbers
17107 You can always enter numbers in octal, decimal, or hexadecimal in
17108 @value{GDBN} by the usual conventions: octal numbers begin with
17109 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
17110 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
17111 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
17112 10; likewise, the default display for numbers---when no particular
17113 format is specified---is base 10. You can change the default base for
17114 both input and output with the commands described below.
17117 @kindex set input-radix
17118 @item set input-radix @var{base}
17119 Set the default base for numeric input. Supported choices
17120 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
17121 specified either unambiguously or using the current input radix; for
17125 set input-radix 012
17126 set input-radix 10.
17127 set input-radix 0xa
17131 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
17132 leaves the input radix unchanged, no matter what it was, since
17133 @samp{10}, being without any leading or trailing signs of its base, is
17134 interpreted in the current radix. Thus, if the current radix is 16,
17135 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
17138 @kindex set output-radix
17139 @item set output-radix @var{base}
17140 Set the default base for numeric display. Supported choices
17141 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
17142 specified either unambiguously or using the current input radix.
17144 @kindex show input-radix
17145 @item show input-radix
17146 Display the current default base for numeric input.
17148 @kindex show output-radix
17149 @item show output-radix
17150 Display the current default base for numeric display.
17152 @item set radix @r{[}@var{base}@r{]}
17156 These commands set and show the default base for both input and output
17157 of numbers. @code{set radix} sets the radix of input and output to
17158 the same base; without an argument, it resets the radix back to its
17159 default value of 10.
17164 @section Configuring the Current ABI
17166 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
17167 application automatically. However, sometimes you need to override its
17168 conclusions. Use these commands to manage @value{GDBN}'s view of the
17175 One @value{GDBN} configuration can debug binaries for multiple operating
17176 system targets, either via remote debugging or native emulation.
17177 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
17178 but you can override its conclusion using the @code{set osabi} command.
17179 One example where this is useful is in debugging of binaries which use
17180 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
17181 not have the same identifying marks that the standard C library for your
17186 Show the OS ABI currently in use.
17189 With no argument, show the list of registered available OS ABI's.
17191 @item set osabi @var{abi}
17192 Set the current OS ABI to @var{abi}.
17195 @cindex float promotion
17197 Generally, the way that an argument of type @code{float} is passed to a
17198 function depends on whether the function is prototyped. For a prototyped
17199 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
17200 according to the architecture's convention for @code{float}. For unprototyped
17201 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
17202 @code{double} and then passed.
17204 Unfortunately, some forms of debug information do not reliably indicate whether
17205 a function is prototyped. If @value{GDBN} calls a function that is not marked
17206 as prototyped, it consults @kbd{set coerce-float-to-double}.
17209 @kindex set coerce-float-to-double
17210 @item set coerce-float-to-double
17211 @itemx set coerce-float-to-double on
17212 Arguments of type @code{float} will be promoted to @code{double} when passed
17213 to an unprototyped function. This is the default setting.
17215 @item set coerce-float-to-double off
17216 Arguments of type @code{float} will be passed directly to unprototyped
17219 @kindex show coerce-float-to-double
17220 @item show coerce-float-to-double
17221 Show the current setting of promoting @code{float} to @code{double}.
17225 @kindex show cp-abi
17226 @value{GDBN} needs to know the ABI used for your program's C@t{++}
17227 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
17228 used to build your application. @value{GDBN} only fully supports
17229 programs with a single C@t{++} ABI; if your program contains code using
17230 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
17231 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
17232 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
17233 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
17234 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
17235 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
17240 Show the C@t{++} ABI currently in use.
17243 With no argument, show the list of supported C@t{++} ABI's.
17245 @item set cp-abi @var{abi}
17246 @itemx set cp-abi auto
17247 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
17250 @node Messages/Warnings
17251 @section Optional Warnings and Messages
17253 @cindex verbose operation
17254 @cindex optional warnings
17255 By default, @value{GDBN} is silent about its inner workings. If you are
17256 running on a slow machine, you may want to use the @code{set verbose}
17257 command. This makes @value{GDBN} tell you when it does a lengthy
17258 internal operation, so you will not think it has crashed.
17260 Currently, the messages controlled by @code{set verbose} are those
17261 which announce that the symbol table for a source file is being read;
17262 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
17265 @kindex set verbose
17266 @item set verbose on
17267 Enables @value{GDBN} output of certain informational messages.
17269 @item set verbose off
17270 Disables @value{GDBN} output of certain informational messages.
17272 @kindex show verbose
17274 Displays whether @code{set verbose} is on or off.
17277 By default, if @value{GDBN} encounters bugs in the symbol table of an
17278 object file, it is silent; but if you are debugging a compiler, you may
17279 find this information useful (@pxref{Symbol Errors, ,Errors Reading
17284 @kindex set complaints
17285 @item set complaints @var{limit}
17286 Permits @value{GDBN} to output @var{limit} complaints about each type of
17287 unusual symbols before becoming silent about the problem. Set
17288 @var{limit} to zero to suppress all complaints; set it to a large number
17289 to prevent complaints from being suppressed.
17291 @kindex show complaints
17292 @item show complaints
17293 Displays how many symbol complaints @value{GDBN} is permitted to produce.
17297 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
17298 lot of stupid questions to confirm certain commands. For example, if
17299 you try to run a program which is already running:
17303 The program being debugged has been started already.
17304 Start it from the beginning? (y or n)
17307 If you are willing to unflinchingly face the consequences of your own
17308 commands, you can disable this ``feature'':
17312 @kindex set confirm
17314 @cindex confirmation
17315 @cindex stupid questions
17316 @item set confirm off
17317 Disables confirmation requests.
17319 @item set confirm on
17320 Enables confirmation requests (the default).
17322 @kindex show confirm
17324 Displays state of confirmation requests.
17328 @cindex command tracing
17329 If you need to debug user-defined commands or sourced files you may find it
17330 useful to enable @dfn{command tracing}. In this mode each command will be
17331 printed as it is executed, prefixed with one or more @samp{+} symbols, the
17332 quantity denoting the call depth of each command.
17335 @kindex set trace-commands
17336 @cindex command scripts, debugging
17337 @item set trace-commands on
17338 Enable command tracing.
17339 @item set trace-commands off
17340 Disable command tracing.
17341 @item show trace-commands
17342 Display the current state of command tracing.
17345 @node Debugging Output
17346 @section Optional Messages about Internal Happenings
17347 @cindex optional debugging messages
17349 @value{GDBN} has commands that enable optional debugging messages from
17350 various @value{GDBN} subsystems; normally these commands are of
17351 interest to @value{GDBN} maintainers, or when reporting a bug. This
17352 section documents those commands.
17355 @kindex set exec-done-display
17356 @item set exec-done-display
17357 Turns on or off the notification of asynchronous commands'
17358 completion. When on, @value{GDBN} will print a message when an
17359 asynchronous command finishes its execution. The default is off.
17360 @kindex show exec-done-display
17361 @item show exec-done-display
17362 Displays the current setting of asynchronous command completion
17365 @cindex gdbarch debugging info
17366 @cindex architecture debugging info
17367 @item set debug arch
17368 Turns on or off display of gdbarch debugging info. The default is off
17370 @item show debug arch
17371 Displays the current state of displaying gdbarch debugging info.
17372 @item set debug aix-thread
17373 @cindex AIX threads
17374 Display debugging messages about inner workings of the AIX thread
17376 @item show debug aix-thread
17377 Show the current state of AIX thread debugging info display.
17378 @item set debug dwarf2-die
17379 @cindex DWARF2 DIEs
17380 Dump DWARF2 DIEs after they are read in.
17381 The value is the number of nesting levels to print.
17382 A value of zero turns off the display.
17383 @item show debug dwarf2-die
17384 Show the current state of DWARF2 DIE debugging.
17385 @item set debug displaced
17386 @cindex displaced stepping debugging info
17387 Turns on or off display of @value{GDBN} debugging info for the
17388 displaced stepping support. The default is off.
17389 @item show debug displaced
17390 Displays the current state of displaying @value{GDBN} debugging info
17391 related to displaced stepping.
17392 @item set debug event
17393 @cindex event debugging info
17394 Turns on or off display of @value{GDBN} event debugging info. The
17396 @item show debug event
17397 Displays the current state of displaying @value{GDBN} event debugging
17399 @item set debug expression
17400 @cindex expression debugging info
17401 Turns on or off display of debugging info about @value{GDBN}
17402 expression parsing. The default is off.
17403 @item show debug expression
17404 Displays the current state of displaying debugging info about
17405 @value{GDBN} expression parsing.
17406 @item set debug frame
17407 @cindex frame debugging info
17408 Turns on or off display of @value{GDBN} frame debugging info. The
17410 @item show debug frame
17411 Displays the current state of displaying @value{GDBN} frame debugging
17413 @item set debug infrun
17414 @cindex inferior debugging info
17415 Turns on or off display of @value{GDBN} debugging info for running the inferior.
17416 The default is off. @file{infrun.c} contains GDB's runtime state machine used
17417 for implementing operations such as single-stepping the inferior.
17418 @item show debug infrun
17419 Displays the current state of @value{GDBN} inferior debugging.
17420 @item set debug lin-lwp
17421 @cindex @sc{gnu}/Linux LWP debug messages
17422 @cindex Linux lightweight processes
17423 Turns on or off debugging messages from the Linux LWP debug support.
17424 @item show debug lin-lwp
17425 Show the current state of Linux LWP debugging messages.
17426 @item set debug lin-lwp-async
17427 @cindex @sc{gnu}/Linux LWP async debug messages
17428 @cindex Linux lightweight processes
17429 Turns on or off debugging messages from the Linux LWP async debug support.
17430 @item show debug lin-lwp-async
17431 Show the current state of Linux LWP async debugging messages.
17432 @item set debug observer
17433 @cindex observer debugging info
17434 Turns on or off display of @value{GDBN} observer debugging. This
17435 includes info such as the notification of observable events.
17436 @item show debug observer
17437 Displays the current state of observer debugging.
17438 @item set debug overload
17439 @cindex C@t{++} overload debugging info
17440 Turns on or off display of @value{GDBN} C@t{++} overload debugging
17441 info. This includes info such as ranking of functions, etc. The default
17443 @item show debug overload
17444 Displays the current state of displaying @value{GDBN} C@t{++} overload
17446 @cindex packets, reporting on stdout
17447 @cindex serial connections, debugging
17448 @cindex debug remote protocol
17449 @cindex remote protocol debugging
17450 @cindex display remote packets
17451 @item set debug remote
17452 Turns on or off display of reports on all packets sent back and forth across
17453 the serial line to the remote machine. The info is printed on the
17454 @value{GDBN} standard output stream. The default is off.
17455 @item show debug remote
17456 Displays the state of display of remote packets.
17457 @item set debug serial
17458 Turns on or off display of @value{GDBN} serial debugging info. The
17460 @item show debug serial
17461 Displays the current state of displaying @value{GDBN} serial debugging
17463 @item set debug solib-frv
17464 @cindex FR-V shared-library debugging
17465 Turns on or off debugging messages for FR-V shared-library code.
17466 @item show debug solib-frv
17467 Display the current state of FR-V shared-library code debugging
17469 @item set debug target
17470 @cindex target debugging info
17471 Turns on or off display of @value{GDBN} target debugging info. This info
17472 includes what is going on at the target level of GDB, as it happens. The
17473 default is 0. Set it to 1 to track events, and to 2 to also track the
17474 value of large memory transfers. Changes to this flag do not take effect
17475 until the next time you connect to a target or use the @code{run} command.
17476 @item show debug target
17477 Displays the current state of displaying @value{GDBN} target debugging
17479 @item set debug timestamp
17480 @cindex timestampping debugging info
17481 Turns on or off display of timestamps with @value{GDBN} debugging info.
17482 When enabled, seconds and microseconds are displayed before each debugging
17484 @item show debug timestamp
17485 Displays the current state of displaying timestamps with @value{GDBN}
17487 @item set debugvarobj
17488 @cindex variable object debugging info
17489 Turns on or off display of @value{GDBN} variable object debugging
17490 info. The default is off.
17491 @item show debugvarobj
17492 Displays the current state of displaying @value{GDBN} variable object
17494 @item set debug xml
17495 @cindex XML parser debugging
17496 Turns on or off debugging messages for built-in XML parsers.
17497 @item show debug xml
17498 Displays the current state of XML debugging messages.
17501 @node Extending GDB
17502 @chapter Extending @value{GDBN}
17503 @cindex extending GDB
17505 @value{GDBN} provides two mechanisms for extension. The first is based
17506 on composition of @value{GDBN} commands, and the second is based on the
17507 Python scripting language.
17510 * Sequences:: Canned Sequences of Commands
17511 * Python:: Scripting @value{GDBN} using Python
17515 @section Canned Sequences of Commands
17517 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
17518 Command Lists}), @value{GDBN} provides two ways to store sequences of
17519 commands for execution as a unit: user-defined commands and command
17523 * Define:: How to define your own commands
17524 * Hooks:: Hooks for user-defined commands
17525 * Command Files:: How to write scripts of commands to be stored in a file
17526 * Output:: Commands for controlled output
17530 @subsection User-defined Commands
17532 @cindex user-defined command
17533 @cindex arguments, to user-defined commands
17534 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
17535 which you assign a new name as a command. This is done with the
17536 @code{define} command. User commands may accept up to 10 arguments
17537 separated by whitespace. Arguments are accessed within the user command
17538 via @code{$arg0@dots{}$arg9}. A trivial example:
17542 print $arg0 + $arg1 + $arg2
17547 To execute the command use:
17554 This defines the command @code{adder}, which prints the sum of
17555 its three arguments. Note the arguments are text substitutions, so they may
17556 reference variables, use complex expressions, or even perform inferior
17559 @cindex argument count in user-defined commands
17560 @cindex how many arguments (user-defined commands)
17561 In addition, @code{$argc} may be used to find out how many arguments have
17562 been passed. This expands to a number in the range 0@dots{}10.
17567 print $arg0 + $arg1
17570 print $arg0 + $arg1 + $arg2
17578 @item define @var{commandname}
17579 Define a command named @var{commandname}. If there is already a command
17580 by that name, you are asked to confirm that you want to redefine it.
17581 @var{commandname} may be a bare command name consisting of letters,
17582 numbers, dashes, and underscores. It may also start with any predefined
17583 prefix command. For example, @samp{define target my-target} creates
17584 a user-defined @samp{target my-target} command.
17586 The definition of the command is made up of other @value{GDBN} command lines,
17587 which are given following the @code{define} command. The end of these
17588 commands is marked by a line containing @code{end}.
17591 @kindex end@r{ (user-defined commands)}
17592 @item document @var{commandname}
17593 Document the user-defined command @var{commandname}, so that it can be
17594 accessed by @code{help}. The command @var{commandname} must already be
17595 defined. This command reads lines of documentation just as @code{define}
17596 reads the lines of the command definition, ending with @code{end}.
17597 After the @code{document} command is finished, @code{help} on command
17598 @var{commandname} displays the documentation you have written.
17600 You may use the @code{document} command again to change the
17601 documentation of a command. Redefining the command with @code{define}
17602 does not change the documentation.
17604 @kindex dont-repeat
17605 @cindex don't repeat command
17607 Used inside a user-defined command, this tells @value{GDBN} that this
17608 command should not be repeated when the user hits @key{RET}
17609 (@pxref{Command Syntax, repeat last command}).
17611 @kindex help user-defined
17612 @item help user-defined
17613 List all user-defined commands, with the first line of the documentation
17618 @itemx show user @var{commandname}
17619 Display the @value{GDBN} commands used to define @var{commandname} (but
17620 not its documentation). If no @var{commandname} is given, display the
17621 definitions for all user-defined commands.
17623 @cindex infinite recursion in user-defined commands
17624 @kindex show max-user-call-depth
17625 @kindex set max-user-call-depth
17626 @item show max-user-call-depth
17627 @itemx set max-user-call-depth
17628 The value of @code{max-user-call-depth} controls how many recursion
17629 levels are allowed in user-defined commands before @value{GDBN} suspects an
17630 infinite recursion and aborts the command.
17633 In addition to the above commands, user-defined commands frequently
17634 use control flow commands, described in @ref{Command Files}.
17636 When user-defined commands are executed, the
17637 commands of the definition are not printed. An error in any command
17638 stops execution of the user-defined command.
17640 If used interactively, commands that would ask for confirmation proceed
17641 without asking when used inside a user-defined command. Many @value{GDBN}
17642 commands that normally print messages to say what they are doing omit the
17643 messages when used in a user-defined command.
17646 @subsection User-defined Command Hooks
17647 @cindex command hooks
17648 @cindex hooks, for commands
17649 @cindex hooks, pre-command
17652 You may define @dfn{hooks}, which are a special kind of user-defined
17653 command. Whenever you run the command @samp{foo}, if the user-defined
17654 command @samp{hook-foo} exists, it is executed (with no arguments)
17655 before that command.
17657 @cindex hooks, post-command
17659 A hook may also be defined which is run after the command you executed.
17660 Whenever you run the command @samp{foo}, if the user-defined command
17661 @samp{hookpost-foo} exists, it is executed (with no arguments) after
17662 that command. Post-execution hooks may exist simultaneously with
17663 pre-execution hooks, for the same command.
17665 It is valid for a hook to call the command which it hooks. If this
17666 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
17668 @c It would be nice if hookpost could be passed a parameter indicating
17669 @c if the command it hooks executed properly or not. FIXME!
17671 @kindex stop@r{, a pseudo-command}
17672 In addition, a pseudo-command, @samp{stop} exists. Defining
17673 (@samp{hook-stop}) makes the associated commands execute every time
17674 execution stops in your program: before breakpoint commands are run,
17675 displays are printed, or the stack frame is printed.
17677 For example, to ignore @code{SIGALRM} signals while
17678 single-stepping, but treat them normally during normal execution,
17683 handle SIGALRM nopass
17687 handle SIGALRM pass
17690 define hook-continue
17691 handle SIGALRM pass
17695 As a further example, to hook at the beginning and end of the @code{echo}
17696 command, and to add extra text to the beginning and end of the message,
17704 define hookpost-echo
17708 (@value{GDBP}) echo Hello World
17709 <<<---Hello World--->>>
17714 You can define a hook for any single-word command in @value{GDBN}, but
17715 not for command aliases; you should define a hook for the basic command
17716 name, e.g.@: @code{backtrace} rather than @code{bt}.
17717 @c FIXME! So how does Joe User discover whether a command is an alias
17719 You can hook a multi-word command by adding @code{hook-} or
17720 @code{hookpost-} to the last word of the command, e.g.@:
17721 @samp{define target hook-remote} to add a hook to @samp{target remote}.
17723 If an error occurs during the execution of your hook, execution of
17724 @value{GDBN} commands stops and @value{GDBN} issues a prompt
17725 (before the command that you actually typed had a chance to run).
17727 If you try to define a hook which does not match any known command, you
17728 get a warning from the @code{define} command.
17730 @node Command Files
17731 @subsection Command Files
17733 @cindex command files
17734 @cindex scripting commands
17735 A command file for @value{GDBN} is a text file made of lines that are
17736 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
17737 also be included. An empty line in a command file does nothing; it
17738 does not mean to repeat the last command, as it would from the
17741 You can request the execution of a command file with the @code{source}
17746 @cindex execute commands from a file
17747 @item source [@code{-v}] @var{filename}
17748 Execute the command file @var{filename}.
17751 The lines in a command file are generally executed sequentially,
17752 unless the order of execution is changed by one of the
17753 @emph{flow-control commands} described below. The commands are not
17754 printed as they are executed. An error in any command terminates
17755 execution of the command file and control is returned to the console.
17757 @value{GDBN} searches for @var{filename} in the current directory and then
17758 on the search path (specified with the @samp{directory} command).
17760 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
17761 each command as it is executed. The option must be given before
17762 @var{filename}, and is interpreted as part of the filename anywhere else.
17764 Commands that would ask for confirmation if used interactively proceed
17765 without asking when used in a command file. Many @value{GDBN} commands that
17766 normally print messages to say what they are doing omit the messages
17767 when called from command files.
17769 @value{GDBN} also accepts command input from standard input. In this
17770 mode, normal output goes to standard output and error output goes to
17771 standard error. Errors in a command file supplied on standard input do
17772 not terminate execution of the command file---execution continues with
17776 gdb < cmds > log 2>&1
17779 (The syntax above will vary depending on the shell used.) This example
17780 will execute commands from the file @file{cmds}. All output and errors
17781 would be directed to @file{log}.
17783 Since commands stored on command files tend to be more general than
17784 commands typed interactively, they frequently need to deal with
17785 complicated situations, such as different or unexpected values of
17786 variables and symbols, changes in how the program being debugged is
17787 built, etc. @value{GDBN} provides a set of flow-control commands to
17788 deal with these complexities. Using these commands, you can write
17789 complex scripts that loop over data structures, execute commands
17790 conditionally, etc.
17797 This command allows to include in your script conditionally executed
17798 commands. The @code{if} command takes a single argument, which is an
17799 expression to evaluate. It is followed by a series of commands that
17800 are executed only if the expression is true (its value is nonzero).
17801 There can then optionally be an @code{else} line, followed by a series
17802 of commands that are only executed if the expression was false. The
17803 end of the list is marked by a line containing @code{end}.
17807 This command allows to write loops. Its syntax is similar to
17808 @code{if}: the command takes a single argument, which is an expression
17809 to evaluate, and must be followed by the commands to execute, one per
17810 line, terminated by an @code{end}. These commands are called the
17811 @dfn{body} of the loop. The commands in the body of @code{while} are
17812 executed repeatedly as long as the expression evaluates to true.
17816 This command exits the @code{while} loop in whose body it is included.
17817 Execution of the script continues after that @code{while}s @code{end}
17820 @kindex loop_continue
17821 @item loop_continue
17822 This command skips the execution of the rest of the body of commands
17823 in the @code{while} loop in whose body it is included. Execution
17824 branches to the beginning of the @code{while} loop, where it evaluates
17825 the controlling expression.
17827 @kindex end@r{ (if/else/while commands)}
17829 Terminate the block of commands that are the body of @code{if},
17830 @code{else}, or @code{while} flow-control commands.
17835 @subsection Commands for Controlled Output
17837 During the execution of a command file or a user-defined command, normal
17838 @value{GDBN} output is suppressed; the only output that appears is what is
17839 explicitly printed by the commands in the definition. This section
17840 describes three commands useful for generating exactly the output you
17845 @item echo @var{text}
17846 @c I do not consider backslash-space a standard C escape sequence
17847 @c because it is not in ANSI.
17848 Print @var{text}. Nonprinting characters can be included in
17849 @var{text} using C escape sequences, such as @samp{\n} to print a
17850 newline. @strong{No newline is printed unless you specify one.}
17851 In addition to the standard C escape sequences, a backslash followed
17852 by a space stands for a space. This is useful for displaying a
17853 string with spaces at the beginning or the end, since leading and
17854 trailing spaces are otherwise trimmed from all arguments.
17855 To print @samp{@w{ }and foo =@w{ }}, use the command
17856 @samp{echo \@w{ }and foo = \@w{ }}.
17858 A backslash at the end of @var{text} can be used, as in C, to continue
17859 the command onto subsequent lines. For example,
17862 echo This is some text\n\
17863 which is continued\n\
17864 onto several lines.\n
17867 produces the same output as
17870 echo This is some text\n
17871 echo which is continued\n
17872 echo onto several lines.\n
17876 @item output @var{expression}
17877 Print the value of @var{expression} and nothing but that value: no
17878 newlines, no @samp{$@var{nn} = }. The value is not entered in the
17879 value history either. @xref{Expressions, ,Expressions}, for more information
17882 @item output/@var{fmt} @var{expression}
17883 Print the value of @var{expression} in format @var{fmt}. You can use
17884 the same formats as for @code{print}. @xref{Output Formats,,Output
17885 Formats}, for more information.
17888 @item printf @var{template}, @var{expressions}@dots{}
17889 Print the values of one or more @var{expressions} under the control of
17890 the string @var{template}. To print several values, make
17891 @var{expressions} be a comma-separated list of individual expressions,
17892 which may be either numbers or pointers. Their values are printed as
17893 specified by @var{template}, exactly as a C program would do by
17894 executing the code below:
17897 printf (@var{template}, @var{expressions}@dots{});
17900 As in @code{C} @code{printf}, ordinary characters in @var{template}
17901 are printed verbatim, while @dfn{conversion specification} introduced
17902 by the @samp{%} character cause subsequent @var{expressions} to be
17903 evaluated, their values converted and formatted according to type and
17904 style information encoded in the conversion specifications, and then
17907 For example, you can print two values in hex like this:
17910 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
17913 @code{printf} supports all the standard @code{C} conversion
17914 specifications, including the flags and modifiers between the @samp{%}
17915 character and the conversion letter, with the following exceptions:
17919 The argument-ordering modifiers, such as @samp{2$}, are not supported.
17922 The modifier @samp{*} is not supported for specifying precision or
17926 The @samp{'} flag (for separation of digits into groups according to
17927 @code{LC_NUMERIC'}) is not supported.
17930 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
17934 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
17937 The conversion letters @samp{a} and @samp{A} are not supported.
17941 Note that the @samp{ll} type modifier is supported only if the
17942 underlying @code{C} implementation used to build @value{GDBN} supports
17943 the @code{long long int} type, and the @samp{L} type modifier is
17944 supported only if @code{long double} type is available.
17946 As in @code{C}, @code{printf} supports simple backslash-escape
17947 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
17948 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
17949 single character. Octal and hexadecimal escape sequences are not
17952 Additionally, @code{printf} supports conversion specifications for DFP
17953 (@dfn{Decimal Floating Point}) types using the following length modifiers
17954 together with a floating point specifier.
17959 @samp{H} for printing @code{Decimal32} types.
17962 @samp{D} for printing @code{Decimal64} types.
17965 @samp{DD} for printing @code{Decimal128} types.
17968 If the underlying @code{C} implementation used to build @value{GDBN} has
17969 support for the three length modifiers for DFP types, other modifiers
17970 such as width and precision will also be available for @value{GDBN} to use.
17972 In case there is no such @code{C} support, no additional modifiers will be
17973 available and the value will be printed in the standard way.
17975 Here's an example of printing DFP types using the above conversion letters:
17977 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
17983 @section Scripting @value{GDBN} using Python
17984 @cindex python scripting
17985 @cindex scripting with python
17987 You can script @value{GDBN} using the @uref{http://www.python.org/,
17988 Python programming language}. This feature is available only if
17989 @value{GDBN} was configured using @option{--with-python}.
17992 * Python Commands:: Accessing Python from @value{GDBN}.
17993 * Python API:: Accessing @value{GDBN} from Python.
17996 @node Python Commands
17997 @subsection Python Commands
17998 @cindex python commands
17999 @cindex commands to access python
18001 @value{GDBN} provides one command for accessing the Python interpreter,
18002 and one related setting:
18006 @item python @r{[}@var{code}@r{]}
18007 The @code{python} command can be used to evaluate Python code.
18009 If given an argument, the @code{python} command will evaluate the
18010 argument as a Python command. For example:
18013 (@value{GDBP}) python print 23
18017 If you do not provide an argument to @code{python}, it will act as a
18018 multi-line command, like @code{define}. In this case, the Python
18019 script is made up of subsequent command lines, given after the
18020 @code{python} command. This command list is terminated using a line
18021 containing @code{end}. For example:
18024 (@value{GDBP}) python
18026 End with a line saying just "end".
18032 @kindex maint set python print-stack
18033 @item maint set python print-stack
18034 By default, @value{GDBN} will print a stack trace when an error occurs
18035 in a Python script. This can be controlled using @code{maint set
18036 python print-stack}: if @code{on}, the default, then Python stack
18037 printing is enabled; if @code{off}, then Python stack printing is
18042 @subsection Python API
18044 @cindex programming in python
18046 @cindex python stdout
18047 @cindex python pagination
18048 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
18049 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
18050 A Python program which outputs to one of these streams may have its
18051 output interrupted by the user (@pxref{Screen Size}). In this
18052 situation, a Python @code{KeyboardInterrupt} exception is thrown.
18055 * Basic Python:: Basic Python Functions.
18056 * Exception Handling::
18057 * Values From Inferior::
18061 @subsubsection Basic Python
18063 @cindex python functions
18064 @cindex python module
18066 @value{GDBN} introduces a new Python module, named @code{gdb}. All
18067 methods and classes added by @value{GDBN} are placed in this module.
18068 @value{GDBN} automatically @code{import}s the @code{gdb} module for
18069 use in all scripts evaluated by the @code{python} command.
18071 @findex gdb.execute
18072 @defun execute command
18073 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
18074 If a GDB exception happens while @var{command} runs, it is
18075 translated as described in @ref{Exception Handling,,Exception Handling}.
18076 If no exceptions occur, this function returns @code{None}.
18079 @findex gdb.get_parameter
18080 @defun get_parameter parameter
18081 Return the value of a @value{GDBN} parameter. @var{parameter} is a
18082 string naming the parameter to look up; @var{parameter} may contain
18083 spaces if the parameter has a multi-part name. For example,
18084 @samp{print object} is a valid parameter name.
18086 If the named parameter does not exist, this function throws a
18087 @code{RuntimeError}. Otherwise, the parameter's value is converted to
18088 a Python value of the appropriate type, and returned.
18092 @defun write string
18093 Print a string to @value{GDBN}'s paginated standard output stream.
18094 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
18095 call this function.
18100 Flush @value{GDBN}'s paginated standard output stream. Flushing
18101 @code{sys.stdout} or @code{sys.stderr} will automatically call this
18105 @node Exception Handling
18106 @subsubsection Exception Handling
18107 @cindex python exceptions
18108 @cindex exceptions, python
18110 When executing the @code{python} command, Python exceptions
18111 uncaught within the Python code are translated to calls to
18112 @value{GDBN} error-reporting mechanism. If the command that called
18113 @code{python} does not handle the error, @value{GDBN} will
18114 terminate it and print an error message containing the Python
18115 exception name, the associated value, and the Python call stack
18116 backtrace at the point where the exception was raised. Example:
18119 (@value{GDBP}) python print foo
18120 Traceback (most recent call last):
18121 File "<string>", line 1, in <module>
18122 NameError: name 'foo' is not defined
18125 @value{GDBN} errors that happen in @value{GDBN} commands invoked by Python
18126 code are converted to Python @code{RuntimeError} exceptions. User
18127 interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
18128 prompt) is translated to a Python @code{KeyboardInterrupt}
18129 exception. If you catch these exceptions in your Python code, your
18130 exception handler will see @code{RuntimeError} or
18131 @code{KeyboardInterrupt} as the exception type, the @value{GDBN} error
18132 message as its value, and the Python call stack backtrace at the
18133 Python statement closest to where the @value{GDBN} error occured as the
18136 @node Values From Inferior
18137 @subsubsection Values From Inferior
18138 @cindex values from inferior, with Python
18139 @cindex python, working with values from inferior
18141 @cindex @code{gdb.Value}
18142 @value{GDBN} provides values it obtains from the inferior program in
18143 an object of type @code{gdb.Value}. @value{GDBN} uses this object
18144 for its internal bookkeeping of the inferior's values, and for
18145 fetching values when necessary.
18147 Inferior values that are simple scalars can be used directly in
18148 Python expressions that are valid for the value's data type. Here's
18149 an example for an integer or floating-point value @code{some_val}:
18156 As result of this, @code{bar} will also be a @code{gdb.Value} object
18157 whose values are of the same type as those of @code{some_val}.
18159 Inferior values that are structures or instances of some class can
18160 be accessed using the Python @dfn{dictionary syntax}. For example, if
18161 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
18162 can access its @code{foo} element with:
18165 bar = some_val['foo']
18168 Again, @code{bar} will also be a @code{gdb.Value} object.
18170 For pointer data types, @code{gdb.Value} provides a method for
18171 dereferencing the pointer to obtain the object it points to.
18173 @defmethod Value dereference
18174 This method returns a new @code{gdb.Value} object whose contents is
18175 the object pointed to by the pointer. For example, if @code{foo} is
18176 a C pointer to an @code{int}, declared in your C program as
18183 then you can use the corresponding @code{gdb.Value} to access what
18184 @code{foo} points to like this:
18187 bar = foo.dereference ()
18190 The result @code{bar} will be a @code{gdb.Value} object holding the
18191 value pointed to by @code{foo}.
18195 @chapter Command Interpreters
18196 @cindex command interpreters
18198 @value{GDBN} supports multiple command interpreters, and some command
18199 infrastructure to allow users or user interface writers to switch
18200 between interpreters or run commands in other interpreters.
18202 @value{GDBN} currently supports two command interpreters, the console
18203 interpreter (sometimes called the command-line interpreter or @sc{cli})
18204 and the machine interface interpreter (or @sc{gdb/mi}). This manual
18205 describes both of these interfaces in great detail.
18207 By default, @value{GDBN} will start with the console interpreter.
18208 However, the user may choose to start @value{GDBN} with another
18209 interpreter by specifying the @option{-i} or @option{--interpreter}
18210 startup options. Defined interpreters include:
18214 @cindex console interpreter
18215 The traditional console or command-line interpreter. This is the most often
18216 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
18217 @value{GDBN} will use this interpreter.
18220 @cindex mi interpreter
18221 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
18222 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
18223 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
18227 @cindex mi2 interpreter
18228 The current @sc{gdb/mi} interface.
18231 @cindex mi1 interpreter
18232 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
18236 @cindex invoke another interpreter
18237 The interpreter being used by @value{GDBN} may not be dynamically
18238 switched at runtime. Although possible, this could lead to a very
18239 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
18240 enters the command "interpreter-set console" in a console view,
18241 @value{GDBN} would switch to using the console interpreter, rendering
18242 the IDE inoperable!
18244 @kindex interpreter-exec
18245 Although you may only choose a single interpreter at startup, you may execute
18246 commands in any interpreter from the current interpreter using the appropriate
18247 command. If you are running the console interpreter, simply use the
18248 @code{interpreter-exec} command:
18251 interpreter-exec mi "-data-list-register-names"
18254 @sc{gdb/mi} has a similar command, although it is only available in versions of
18255 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
18258 @chapter @value{GDBN} Text User Interface
18260 @cindex Text User Interface
18263 * TUI Overview:: TUI overview
18264 * TUI Keys:: TUI key bindings
18265 * TUI Single Key Mode:: TUI single key mode
18266 * TUI Commands:: TUI-specific commands
18267 * TUI Configuration:: TUI configuration variables
18270 The @value{GDBN} Text User Interface (TUI) is a terminal
18271 interface which uses the @code{curses} library to show the source
18272 file, the assembly output, the program registers and @value{GDBN}
18273 commands in separate text windows. The TUI mode is supported only
18274 on platforms where a suitable version of the @code{curses} library
18277 @pindex @value{GDBTUI}
18278 The TUI mode is enabled by default when you invoke @value{GDBN} as
18279 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
18280 You can also switch in and out of TUI mode while @value{GDBN} runs by
18281 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
18282 @xref{TUI Keys, ,TUI Key Bindings}.
18285 @section TUI Overview
18287 In TUI mode, @value{GDBN} can display several text windows:
18291 This window is the @value{GDBN} command window with the @value{GDBN}
18292 prompt and the @value{GDBN} output. The @value{GDBN} input is still
18293 managed using readline.
18296 The source window shows the source file of the program. The current
18297 line and active breakpoints are displayed in this window.
18300 The assembly window shows the disassembly output of the program.
18303 This window shows the processor registers. Registers are highlighted
18304 when their values change.
18307 The source and assembly windows show the current program position
18308 by highlighting the current line and marking it with a @samp{>} marker.
18309 Breakpoints are indicated with two markers. The first marker
18310 indicates the breakpoint type:
18314 Breakpoint which was hit at least once.
18317 Breakpoint which was never hit.
18320 Hardware breakpoint which was hit at least once.
18323 Hardware breakpoint which was never hit.
18326 The second marker indicates whether the breakpoint is enabled or not:
18330 Breakpoint is enabled.
18333 Breakpoint is disabled.
18336 The source, assembly and register windows are updated when the current
18337 thread changes, when the frame changes, or when the program counter
18340 These windows are not all visible at the same time. The command
18341 window is always visible. The others can be arranged in several
18352 source and assembly,
18355 source and registers, or
18358 assembly and registers.
18361 A status line above the command window shows the following information:
18365 Indicates the current @value{GDBN} target.
18366 (@pxref{Targets, ,Specifying a Debugging Target}).
18369 Gives the current process or thread number.
18370 When no process is being debugged, this field is set to @code{No process}.
18373 Gives the current function name for the selected frame.
18374 The name is demangled if demangling is turned on (@pxref{Print Settings}).
18375 When there is no symbol corresponding to the current program counter,
18376 the string @code{??} is displayed.
18379 Indicates the current line number for the selected frame.
18380 When the current line number is not known, the string @code{??} is displayed.
18383 Indicates the current program counter address.
18387 @section TUI Key Bindings
18388 @cindex TUI key bindings
18390 The TUI installs several key bindings in the readline keymaps
18391 (@pxref{Command Line Editing}). The following key bindings
18392 are installed for both TUI mode and the @value{GDBN} standard mode.
18401 Enter or leave the TUI mode. When leaving the TUI mode,
18402 the curses window management stops and @value{GDBN} operates using
18403 its standard mode, writing on the terminal directly. When reentering
18404 the TUI mode, control is given back to the curses windows.
18405 The screen is then refreshed.
18409 Use a TUI layout with only one window. The layout will
18410 either be @samp{source} or @samp{assembly}. When the TUI mode
18411 is not active, it will switch to the TUI mode.
18413 Think of this key binding as the Emacs @kbd{C-x 1} binding.
18417 Use a TUI layout with at least two windows. When the current
18418 layout already has two windows, the next layout with two windows is used.
18419 When a new layout is chosen, one window will always be common to the
18420 previous layout and the new one.
18422 Think of it as the Emacs @kbd{C-x 2} binding.
18426 Change the active window. The TUI associates several key bindings
18427 (like scrolling and arrow keys) with the active window. This command
18428 gives the focus to the next TUI window.
18430 Think of it as the Emacs @kbd{C-x o} binding.
18434 Switch in and out of the TUI SingleKey mode that binds single
18435 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
18438 The following key bindings only work in the TUI mode:
18443 Scroll the active window one page up.
18447 Scroll the active window one page down.
18451 Scroll the active window one line up.
18455 Scroll the active window one line down.
18459 Scroll the active window one column left.
18463 Scroll the active window one column right.
18467 Refresh the screen.
18470 Because the arrow keys scroll the active window in the TUI mode, they
18471 are not available for their normal use by readline unless the command
18472 window has the focus. When another window is active, you must use
18473 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
18474 and @kbd{C-f} to control the command window.
18476 @node TUI Single Key Mode
18477 @section TUI Single Key Mode
18478 @cindex TUI single key mode
18480 The TUI also provides a @dfn{SingleKey} mode, which binds several
18481 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
18482 switch into this mode, where the following key bindings are used:
18485 @kindex c @r{(SingleKey TUI key)}
18489 @kindex d @r{(SingleKey TUI key)}
18493 @kindex f @r{(SingleKey TUI key)}
18497 @kindex n @r{(SingleKey TUI key)}
18501 @kindex q @r{(SingleKey TUI key)}
18503 exit the SingleKey mode.
18505 @kindex r @r{(SingleKey TUI key)}
18509 @kindex s @r{(SingleKey TUI key)}
18513 @kindex u @r{(SingleKey TUI key)}
18517 @kindex v @r{(SingleKey TUI key)}
18521 @kindex w @r{(SingleKey TUI key)}
18526 Other keys temporarily switch to the @value{GDBN} command prompt.
18527 The key that was pressed is inserted in the editing buffer so that
18528 it is possible to type most @value{GDBN} commands without interaction
18529 with the TUI SingleKey mode. Once the command is entered the TUI
18530 SingleKey mode is restored. The only way to permanently leave
18531 this mode is by typing @kbd{q} or @kbd{C-x s}.
18535 @section TUI-specific Commands
18536 @cindex TUI commands
18538 The TUI has specific commands to control the text windows.
18539 These commands are always available, even when @value{GDBN} is not in
18540 the TUI mode. When @value{GDBN} is in the standard mode, most
18541 of these commands will automatically switch to the TUI mode.
18546 List and give the size of all displayed windows.
18550 Display the next layout.
18553 Display the previous layout.
18556 Display the source window only.
18559 Display the assembly window only.
18562 Display the source and assembly window.
18565 Display the register window together with the source or assembly window.
18569 Make the next window active for scrolling.
18572 Make the previous window active for scrolling.
18575 Make the source window active for scrolling.
18578 Make the assembly window active for scrolling.
18581 Make the register window active for scrolling.
18584 Make the command window active for scrolling.
18588 Refresh the screen. This is similar to typing @kbd{C-L}.
18590 @item tui reg float
18592 Show the floating point registers in the register window.
18594 @item tui reg general
18595 Show the general registers in the register window.
18598 Show the next register group. The list of register groups as well as
18599 their order is target specific. The predefined register groups are the
18600 following: @code{general}, @code{float}, @code{system}, @code{vector},
18601 @code{all}, @code{save}, @code{restore}.
18603 @item tui reg system
18604 Show the system registers in the register window.
18608 Update the source window and the current execution point.
18610 @item winheight @var{name} +@var{count}
18611 @itemx winheight @var{name} -@var{count}
18613 Change the height of the window @var{name} by @var{count}
18614 lines. Positive counts increase the height, while negative counts
18617 @item tabset @var{nchars}
18619 Set the width of tab stops to be @var{nchars} characters.
18622 @node TUI Configuration
18623 @section TUI Configuration Variables
18624 @cindex TUI configuration variables
18626 Several configuration variables control the appearance of TUI windows.
18629 @item set tui border-kind @var{kind}
18630 @kindex set tui border-kind
18631 Select the border appearance for the source, assembly and register windows.
18632 The possible values are the following:
18635 Use a space character to draw the border.
18638 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
18641 Use the Alternate Character Set to draw the border. The border is
18642 drawn using character line graphics if the terminal supports them.
18645 @item set tui border-mode @var{mode}
18646 @kindex set tui border-mode
18647 @itemx set tui active-border-mode @var{mode}
18648 @kindex set tui active-border-mode
18649 Select the display attributes for the borders of the inactive windows
18650 or the active window. The @var{mode} can be one of the following:
18653 Use normal attributes to display the border.
18659 Use reverse video mode.
18662 Use half bright mode.
18664 @item half-standout
18665 Use half bright and standout mode.
18668 Use extra bright or bold mode.
18670 @item bold-standout
18671 Use extra bright or bold and standout mode.
18676 @chapter Using @value{GDBN} under @sc{gnu} Emacs
18679 @cindex @sc{gnu} Emacs
18680 A special interface allows you to use @sc{gnu} Emacs to view (and
18681 edit) the source files for the program you are debugging with
18684 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
18685 executable file you want to debug as an argument. This command starts
18686 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
18687 created Emacs buffer.
18688 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
18690 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
18695 All ``terminal'' input and output goes through an Emacs buffer, called
18698 This applies both to @value{GDBN} commands and their output, and to the input
18699 and output done by the program you are debugging.
18701 This is useful because it means that you can copy the text of previous
18702 commands and input them again; you can even use parts of the output
18705 All the facilities of Emacs' Shell mode are available for interacting
18706 with your program. In particular, you can send signals the usual
18707 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
18711 @value{GDBN} displays source code through Emacs.
18713 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
18714 source file for that frame and puts an arrow (@samp{=>}) at the
18715 left margin of the current line. Emacs uses a separate buffer for
18716 source display, and splits the screen to show both your @value{GDBN} session
18719 Explicit @value{GDBN} @code{list} or search commands still produce output as
18720 usual, but you probably have no reason to use them from Emacs.
18723 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
18724 a graphical mode, enabled by default, which provides further buffers
18725 that can control the execution and describe the state of your program.
18726 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
18728 If you specify an absolute file name when prompted for the @kbd{M-x
18729 gdb} argument, then Emacs sets your current working directory to where
18730 your program resides. If you only specify the file name, then Emacs
18731 sets your current working directory to to the directory associated
18732 with the previous buffer. In this case, @value{GDBN} may find your
18733 program by searching your environment's @code{PATH} variable, but on
18734 some operating systems it might not find the source. So, although the
18735 @value{GDBN} input and output session proceeds normally, the auxiliary
18736 buffer does not display the current source and line of execution.
18738 The initial working directory of @value{GDBN} is printed on the top
18739 line of the GUD buffer and this serves as a default for the commands
18740 that specify files for @value{GDBN} to operate on. @xref{Files,
18741 ,Commands to Specify Files}.
18743 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
18744 need to call @value{GDBN} by a different name (for example, if you
18745 keep several configurations around, with different names) you can
18746 customize the Emacs variable @code{gud-gdb-command-name} to run the
18749 In the GUD buffer, you can use these special Emacs commands in
18750 addition to the standard Shell mode commands:
18754 Describe the features of Emacs' GUD Mode.
18757 Execute to another source line, like the @value{GDBN} @code{step} command; also
18758 update the display window to show the current file and location.
18761 Execute to next source line in this function, skipping all function
18762 calls, like the @value{GDBN} @code{next} command. Then update the display window
18763 to show the current file and location.
18766 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
18767 display window accordingly.
18770 Execute until exit from the selected stack frame, like the @value{GDBN}
18771 @code{finish} command.
18774 Continue execution of your program, like the @value{GDBN} @code{continue}
18778 Go up the number of frames indicated by the numeric argument
18779 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
18780 like the @value{GDBN} @code{up} command.
18783 Go down the number of frames indicated by the numeric argument, like the
18784 @value{GDBN} @code{down} command.
18787 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
18788 tells @value{GDBN} to set a breakpoint on the source line point is on.
18790 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
18791 separate frame which shows a backtrace when the GUD buffer is current.
18792 Move point to any frame in the stack and type @key{RET} to make it
18793 become the current frame and display the associated source in the
18794 source buffer. Alternatively, click @kbd{Mouse-2} to make the
18795 selected frame become the current one. In graphical mode, the
18796 speedbar displays watch expressions.
18798 If you accidentally delete the source-display buffer, an easy way to get
18799 it back is to type the command @code{f} in the @value{GDBN} buffer, to
18800 request a frame display; when you run under Emacs, this recreates
18801 the source buffer if necessary to show you the context of the current
18804 The source files displayed in Emacs are in ordinary Emacs buffers
18805 which are visiting the source files in the usual way. You can edit
18806 the files with these buffers if you wish; but keep in mind that @value{GDBN}
18807 communicates with Emacs in terms of line numbers. If you add or
18808 delete lines from the text, the line numbers that @value{GDBN} knows cease
18809 to correspond properly with the code.
18811 A more detailed description of Emacs' interaction with @value{GDBN} is
18812 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
18815 @c The following dropped because Epoch is nonstandard. Reactivate
18816 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
18818 @kindex Emacs Epoch environment
18822 Version 18 of @sc{gnu} Emacs has a built-in window system
18823 called the @code{epoch}
18824 environment. Users of this environment can use a new command,
18825 @code{inspect} which performs identically to @code{print} except that
18826 each value is printed in its own window.
18831 @chapter The @sc{gdb/mi} Interface
18833 @unnumberedsec Function and Purpose
18835 @cindex @sc{gdb/mi}, its purpose
18836 @sc{gdb/mi} is a line based machine oriented text interface to
18837 @value{GDBN} and is activated by specifying using the
18838 @option{--interpreter} command line option (@pxref{Mode Options}). It
18839 is specifically intended to support the development of systems which
18840 use the debugger as just one small component of a larger system.
18842 This chapter is a specification of the @sc{gdb/mi} interface. It is written
18843 in the form of a reference manual.
18845 Note that @sc{gdb/mi} is still under construction, so some of the
18846 features described below are incomplete and subject to change
18847 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
18849 @unnumberedsec Notation and Terminology
18851 @cindex notational conventions, for @sc{gdb/mi}
18852 This chapter uses the following notation:
18856 @code{|} separates two alternatives.
18859 @code{[ @var{something} ]} indicates that @var{something} is optional:
18860 it may or may not be given.
18863 @code{( @var{group} )*} means that @var{group} inside the parentheses
18864 may repeat zero or more times.
18867 @code{( @var{group} )+} means that @var{group} inside the parentheses
18868 may repeat one or more times.
18871 @code{"@var{string}"} means a literal @var{string}.
18875 @heading Dependencies
18879 * GDB/MI General Design::
18880 * GDB/MI Command Syntax::
18881 * GDB/MI Compatibility with CLI::
18882 * GDB/MI Development and Front Ends::
18883 * GDB/MI Output Records::
18884 * GDB/MI Simple Examples::
18885 * GDB/MI Command Description Format::
18886 * GDB/MI Breakpoint Commands::
18887 * GDB/MI Program Context::
18888 * GDB/MI Thread Commands::
18889 * GDB/MI Program Execution::
18890 * GDB/MI Stack Manipulation::
18891 * GDB/MI Variable Objects::
18892 * GDB/MI Data Manipulation::
18893 * GDB/MI Tracepoint Commands::
18894 * GDB/MI Symbol Query::
18895 * GDB/MI File Commands::
18897 * GDB/MI Kod Commands::
18898 * GDB/MI Memory Overlay Commands::
18899 * GDB/MI Signal Handling Commands::
18901 * GDB/MI Target Manipulation::
18902 * GDB/MI File Transfer Commands::
18903 * GDB/MI Miscellaneous Commands::
18906 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18907 @node GDB/MI General Design
18908 @section @sc{gdb/mi} General Design
18909 @cindex GDB/MI General Design
18911 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
18912 parts---commands sent to @value{GDBN}, responses to those commands
18913 and notifications. Each command results in exactly one response,
18914 indicating either successful completion of the command, or an error.
18915 For the commands that do not resume the target, the response contains the
18916 requested information. For the commands that resume the target, the
18917 response only indicates whether the target was successfully resumed.
18918 Notifications is the mechanism for reporting changes in the state of the
18919 target, or in @value{GDBN} state, that cannot conveniently be associated with
18920 a command and reported as part of that command response.
18922 The important examples of notifications are:
18926 Exec notifications. These are used to report changes in
18927 target state---when a target is resumed, or stopped. It would not
18928 be feasible to include this information in response of resuming
18929 commands, because one resume commands can result in multiple events in
18930 different threads. Also, quite some time may pass before any event
18931 happens in the target, while a frontend needs to know whether the resuming
18932 command itself was successfully executed.
18935 Console output, and status notifications. Console output
18936 notifications are used to report output of CLI commands, as well as
18937 diagnostics for other commands. Status notifications are used to
18938 report the progress of a long-running operation. Naturally, including
18939 this information in command response would mean no output is produced
18940 until the command is finished, which is undesirable.
18943 General notifications. Commands may have various side effects on
18944 the @value{GDBN} or target state beyond their official purpose. For example,
18945 a command may change the selected thread. Although such changes can
18946 be included in command response, using notification allows for more
18947 orthogonal frontend design.
18951 There's no guarantee that whenever an MI command reports an error,
18952 @value{GDBN} or the target are in any specific state, and especially,
18953 the state is not reverted to the state before the MI command was
18954 processed. Therefore, whenever an MI command results in an error,
18955 we recommend that the frontend refreshes all the information shown in
18956 the user interface.
18958 @subsection Context management
18960 In most cases when @value{GDBN} accesses the target, this access is
18961 done in context of a specific thread and frame (@pxref{Frames}).
18962 Often, even when accessing global data, the target requires that a thread
18963 be specified. The CLI interface maintains the selected thread and frame,
18964 and supplies them to target on each command. This is convenient,
18965 because a command line user would not want to specify that information
18966 explicitly on each command, and because user interacts with
18967 @value{GDBN} via a single terminal, so no confusion is possible as
18968 to what thread and frame are the current ones.
18970 In the case of MI, the concept of selected thread and frame is less
18971 useful. First, a frontend can easily remember this information
18972 itself. Second, a graphical frontend can have more than one window,
18973 each one used for debugging a different thread, and the frontend might
18974 want to access additional threads for internal purposes. This
18975 increases the risk that by relying on implicitly selected thread, the
18976 frontend may be operating on a wrong one. Therefore, each MI command
18977 should explicitly specify which thread and frame to operate on. To
18978 make it possible, each MI command accepts the @samp{--thread} and
18979 @samp{--frame} options, the value to each is @value{GDBN} identifier
18980 for thread and frame to operate on.
18982 Usually, each top-level window in a frontend allows the user to select
18983 a thread and a frame, and remembers the user selection for further
18984 operations. However, in some cases @value{GDBN} may suggest that the
18985 current thread be changed. For example, when stopping on a breakpoint
18986 it is reasonable to switch to the thread where breakpoint is hit. For
18987 another example, if the user issues the CLI @samp{thread} command via
18988 the frontend, it is desirable to change the frontend's selected thread to the
18989 one specified by user. @value{GDBN} communicates the suggestion to
18990 change current thread using the @samp{=thread-selected} notification.
18991 No such notification is available for the selected frame at the moment.
18993 Note that historically, MI shares the selected thread with CLI, so
18994 frontends used the @code{-thread-select} to execute commands in the
18995 right context. However, getting this to work right is cumbersome. The
18996 simplest way is for frontend to emit @code{-thread-select} command
18997 before every command. This doubles the number of commands that need
18998 to be sent. The alternative approach is to suppress @code{-thread-select}
18999 if the selected thread in @value{GDBN} is supposed to be identical to the
19000 thread the frontend wants to operate on. However, getting this
19001 optimization right can be tricky. In particular, if the frontend
19002 sends several commands to @value{GDBN}, and one of the commands changes the
19003 selected thread, then the behaviour of subsequent commands will
19004 change. So, a frontend should either wait for response from such
19005 problematic commands, or explicitly add @code{-thread-select} for
19006 all subsequent commands. No frontend is known to do this exactly
19007 right, so it is suggested to just always pass the @samp{--thread} and
19008 @samp{--frame} options.
19010 @subsection Asynchronous command execution and non-stop mode
19012 On some targets, @value{GDBN} is capable of processing MI commands
19013 even while the target is running. This is called @dfn{asynchronous
19014 command execution} (@pxref{Background Execution}). The frontend may
19015 specify a preferrence for asynchronous execution using the
19016 @code{-gdb-set target-async 1} command, which should be emitted before
19017 either running the executable or attaching to the target. After the
19018 frontend has started the executable or attached to the target, it can
19019 find if asynchronous execution is enabled using the
19020 @code{-list-target-features} command.
19022 Even if @value{GDBN} can accept a command while target is running,
19023 many commands that access the target do not work when the target is
19024 running. Therefore, asynchronous command execution is most useful
19025 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
19026 it is possible to examine the state of one thread, while other threads
19029 When a given thread is running, MI commands that try to access the
19030 target in the context of that thread may not work, or may work only on
19031 some targets. In particular, commands that try to operate on thread's
19032 stack will not work, on any target. Commands that read memory, or
19033 modify breakpoints, may work or not work, depending on the target. Note
19034 that even commands that operate on global state, such as @code{print},
19035 @code{set}, and breakpoint commands, still access the target in the
19036 context of a specific thread, so frontend should try to find a
19037 stopped thread and perform the operation on that thread (using the
19038 @samp{--thread} option).
19040 Which commands will work in the context of a running thread is
19041 highly target dependent. However, the two commands
19042 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
19043 to find the state of a thread, will always work.
19045 @subsection Thread groups
19046 @value{GDBN} may be used to debug several processes at the same time.
19047 On some platfroms, @value{GDBN} may support debugging of several
19048 hardware systems, each one having several cores with several different
19049 processes running on each core. This section describes the MI
19050 mechanism to support such debugging scenarios.
19052 The key observation is that regardless of the structure of the
19053 target, MI can have a global list of threads, because most commands that
19054 accept the @samp{--thread} option do not need to know what process that
19055 thread belongs to. Therefore, it is not necessary to introduce
19056 neither additional @samp{--process} option, nor an notion of the
19057 current process in the MI interface. The only strictly new feature
19058 that is required is the ability to find how the threads are grouped
19061 To allow the user to discover such grouping, and to support arbitrary
19062 hierarchy of machines/cores/processes, MI introduces the concept of a
19063 @dfn{thread group}. Thread group is a collection of threads and other
19064 thread groups. A thread group always has a string identifier, a type,
19065 and may have additional attributes specific to the type. A new
19066 command, @code{-list-thread-groups}, returns the list of top-level
19067 thread groups, which correspond to processes that @value{GDBN} is
19068 debugging at the moment. By passing an identifier of a thread group
19069 to the @code{-list-thread-groups} command, it is possible to obtain
19070 the members of specific thread group.
19072 To allow the user to easily discover processes, and other objects, he
19073 wishes to debug, a concept of @dfn{available thread group} is
19074 introduced. Available thread group is an thread group that
19075 @value{GDBN} is not debugging, but that can be attached to, using the
19076 @code{-target-attach} command. The list of available top-level thread
19077 groups can be obtained using @samp{-list-thread-groups --available}.
19078 In general, the content of a thread group may be only retrieved only
19079 after attaching to that thread group.
19081 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19082 @node GDB/MI Command Syntax
19083 @section @sc{gdb/mi} Command Syntax
19086 * GDB/MI Input Syntax::
19087 * GDB/MI Output Syntax::
19090 @node GDB/MI Input Syntax
19091 @subsection @sc{gdb/mi} Input Syntax
19093 @cindex input syntax for @sc{gdb/mi}
19094 @cindex @sc{gdb/mi}, input syntax
19096 @item @var{command} @expansion{}
19097 @code{@var{cli-command} | @var{mi-command}}
19099 @item @var{cli-command} @expansion{}
19100 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
19101 @var{cli-command} is any existing @value{GDBN} CLI command.
19103 @item @var{mi-command} @expansion{}
19104 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
19105 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
19107 @item @var{token} @expansion{}
19108 "any sequence of digits"
19110 @item @var{option} @expansion{}
19111 @code{"-" @var{parameter} [ " " @var{parameter} ]}
19113 @item @var{parameter} @expansion{}
19114 @code{@var{non-blank-sequence} | @var{c-string}}
19116 @item @var{operation} @expansion{}
19117 @emph{any of the operations described in this chapter}
19119 @item @var{non-blank-sequence} @expansion{}
19120 @emph{anything, provided it doesn't contain special characters such as
19121 "-", @var{nl}, """ and of course " "}
19123 @item @var{c-string} @expansion{}
19124 @code{""" @var{seven-bit-iso-c-string-content} """}
19126 @item @var{nl} @expansion{}
19135 The CLI commands are still handled by the @sc{mi} interpreter; their
19136 output is described below.
19139 The @code{@var{token}}, when present, is passed back when the command
19143 Some @sc{mi} commands accept optional arguments as part of the parameter
19144 list. Each option is identified by a leading @samp{-} (dash) and may be
19145 followed by an optional argument parameter. Options occur first in the
19146 parameter list and can be delimited from normal parameters using
19147 @samp{--} (this is useful when some parameters begin with a dash).
19154 We want easy access to the existing CLI syntax (for debugging).
19157 We want it to be easy to spot a @sc{mi} operation.
19160 @node GDB/MI Output Syntax
19161 @subsection @sc{gdb/mi} Output Syntax
19163 @cindex output syntax of @sc{gdb/mi}
19164 @cindex @sc{gdb/mi}, output syntax
19165 The output from @sc{gdb/mi} consists of zero or more out-of-band records
19166 followed, optionally, by a single result record. This result record
19167 is for the most recent command. The sequence of output records is
19168 terminated by @samp{(gdb)}.
19170 If an input command was prefixed with a @code{@var{token}} then the
19171 corresponding output for that command will also be prefixed by that same
19175 @item @var{output} @expansion{}
19176 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
19178 @item @var{result-record} @expansion{}
19179 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
19181 @item @var{out-of-band-record} @expansion{}
19182 @code{@var{async-record} | @var{stream-record}}
19184 @item @var{async-record} @expansion{}
19185 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
19187 @item @var{exec-async-output} @expansion{}
19188 @code{[ @var{token} ] "*" @var{async-output}}
19190 @item @var{status-async-output} @expansion{}
19191 @code{[ @var{token} ] "+" @var{async-output}}
19193 @item @var{notify-async-output} @expansion{}
19194 @code{[ @var{token} ] "=" @var{async-output}}
19196 @item @var{async-output} @expansion{}
19197 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
19199 @item @var{result-class} @expansion{}
19200 @code{"done" | "running" | "connected" | "error" | "exit"}
19202 @item @var{async-class} @expansion{}
19203 @code{"stopped" | @var{others}} (where @var{others} will be added
19204 depending on the needs---this is still in development).
19206 @item @var{result} @expansion{}
19207 @code{ @var{variable} "=" @var{value}}
19209 @item @var{variable} @expansion{}
19210 @code{ @var{string} }
19212 @item @var{value} @expansion{}
19213 @code{ @var{const} | @var{tuple} | @var{list} }
19215 @item @var{const} @expansion{}
19216 @code{@var{c-string}}
19218 @item @var{tuple} @expansion{}
19219 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
19221 @item @var{list} @expansion{}
19222 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
19223 @var{result} ( "," @var{result} )* "]" }
19225 @item @var{stream-record} @expansion{}
19226 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
19228 @item @var{console-stream-output} @expansion{}
19229 @code{"~" @var{c-string}}
19231 @item @var{target-stream-output} @expansion{}
19232 @code{"@@" @var{c-string}}
19234 @item @var{log-stream-output} @expansion{}
19235 @code{"&" @var{c-string}}
19237 @item @var{nl} @expansion{}
19240 @item @var{token} @expansion{}
19241 @emph{any sequence of digits}.
19249 All output sequences end in a single line containing a period.
19252 The @code{@var{token}} is from the corresponding request. Note that
19253 for all async output, while the token is allowed by the grammar and
19254 may be output by future versions of @value{GDBN} for select async
19255 output messages, it is generally omitted. Frontends should treat
19256 all async output as reporting general changes in the state of the
19257 target and there should be no need to associate async output to any
19261 @cindex status output in @sc{gdb/mi}
19262 @var{status-async-output} contains on-going status information about the
19263 progress of a slow operation. It can be discarded. All status output is
19264 prefixed by @samp{+}.
19267 @cindex async output in @sc{gdb/mi}
19268 @var{exec-async-output} contains asynchronous state change on the target
19269 (stopped, started, disappeared). All async output is prefixed by
19273 @cindex notify output in @sc{gdb/mi}
19274 @var{notify-async-output} contains supplementary information that the
19275 client should handle (e.g., a new breakpoint information). All notify
19276 output is prefixed by @samp{=}.
19279 @cindex console output in @sc{gdb/mi}
19280 @var{console-stream-output} is output that should be displayed as is in the
19281 console. It is the textual response to a CLI command. All the console
19282 output is prefixed by @samp{~}.
19285 @cindex target output in @sc{gdb/mi}
19286 @var{target-stream-output} is the output produced by the target program.
19287 All the target output is prefixed by @samp{@@}.
19290 @cindex log output in @sc{gdb/mi}
19291 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
19292 instance messages that should be displayed as part of an error log. All
19293 the log output is prefixed by @samp{&}.
19296 @cindex list output in @sc{gdb/mi}
19297 New @sc{gdb/mi} commands should only output @var{lists} containing
19303 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
19304 details about the various output records.
19306 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19307 @node GDB/MI Compatibility with CLI
19308 @section @sc{gdb/mi} Compatibility with CLI
19310 @cindex compatibility, @sc{gdb/mi} and CLI
19311 @cindex @sc{gdb/mi}, compatibility with CLI
19313 For the developers convenience CLI commands can be entered directly,
19314 but there may be some unexpected behaviour. For example, commands
19315 that query the user will behave as if the user replied yes, breakpoint
19316 command lists are not executed and some CLI commands, such as
19317 @code{if}, @code{when} and @code{define}, prompt for further input with
19318 @samp{>}, which is not valid MI output.
19320 This feature may be removed at some stage in the future and it is
19321 recommended that front ends use the @code{-interpreter-exec} command
19322 (@pxref{-interpreter-exec}).
19324 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19325 @node GDB/MI Development and Front Ends
19326 @section @sc{gdb/mi} Development and Front Ends
19327 @cindex @sc{gdb/mi} development
19329 The application which takes the MI output and presents the state of the
19330 program being debugged to the user is called a @dfn{front end}.
19332 Although @sc{gdb/mi} is still incomplete, it is currently being used
19333 by a variety of front ends to @value{GDBN}. This makes it difficult
19334 to introduce new functionality without breaking existing usage. This
19335 section tries to minimize the problems by describing how the protocol
19338 Some changes in MI need not break a carefully designed front end, and
19339 for these the MI version will remain unchanged. The following is a
19340 list of changes that may occur within one level, so front ends should
19341 parse MI output in a way that can handle them:
19345 New MI commands may be added.
19348 New fields may be added to the output of any MI command.
19351 The range of values for fields with specified values, e.g.,
19352 @code{in_scope} (@pxref{-var-update}) may be extended.
19354 @c The format of field's content e.g type prefix, may change so parse it
19355 @c at your own risk. Yes, in general?
19357 @c The order of fields may change? Shouldn't really matter but it might
19358 @c resolve inconsistencies.
19361 If the changes are likely to break front ends, the MI version level
19362 will be increased by one. This will allow the front end to parse the
19363 output according to the MI version. Apart from mi0, new versions of
19364 @value{GDBN} will not support old versions of MI and it will be the
19365 responsibility of the front end to work with the new one.
19367 @c Starting with mi3, add a new command -mi-version that prints the MI
19370 The best way to avoid unexpected changes in MI that might break your front
19371 end is to make your project known to @value{GDBN} developers and
19372 follow development on @email{gdb@@sourceware.org} and
19373 @email{gdb-patches@@sourceware.org}.
19374 @cindex mailing lists
19376 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19377 @node GDB/MI Output Records
19378 @section @sc{gdb/mi} Output Records
19381 * GDB/MI Result Records::
19382 * GDB/MI Stream Records::
19383 * GDB/MI Async Records::
19384 * GDB/MI Frame Information::
19387 @node GDB/MI Result Records
19388 @subsection @sc{gdb/mi} Result Records
19390 @cindex result records in @sc{gdb/mi}
19391 @cindex @sc{gdb/mi}, result records
19392 In addition to a number of out-of-band notifications, the response to a
19393 @sc{gdb/mi} command includes one of the following result indications:
19397 @item "^done" [ "," @var{results} ]
19398 The synchronous operation was successful, @code{@var{results}} are the return
19403 @c Is this one correct? Should it be an out-of-band notification?
19404 The asynchronous operation was successfully started. The target is
19409 @value{GDBN} has connected to a remote target.
19411 @item "^error" "," @var{c-string}
19413 The operation failed. The @code{@var{c-string}} contains the corresponding
19418 @value{GDBN} has terminated.
19422 @node GDB/MI Stream Records
19423 @subsection @sc{gdb/mi} Stream Records
19425 @cindex @sc{gdb/mi}, stream records
19426 @cindex stream records in @sc{gdb/mi}
19427 @value{GDBN} internally maintains a number of output streams: the console, the
19428 target, and the log. The output intended for each of these streams is
19429 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
19431 Each stream record begins with a unique @dfn{prefix character} which
19432 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
19433 Syntax}). In addition to the prefix, each stream record contains a
19434 @code{@var{string-output}}. This is either raw text (with an implicit new
19435 line) or a quoted C string (which does not contain an implicit newline).
19438 @item "~" @var{string-output}
19439 The console output stream contains text that should be displayed in the
19440 CLI console window. It contains the textual responses to CLI commands.
19442 @item "@@" @var{string-output}
19443 The target output stream contains any textual output from the running
19444 target. This is only present when GDB's event loop is truly
19445 asynchronous, which is currently only the case for remote targets.
19447 @item "&" @var{string-output}
19448 The log stream contains debugging messages being produced by @value{GDBN}'s
19452 @node GDB/MI Async Records
19453 @subsection @sc{gdb/mi} Async Records
19455 @cindex async records in @sc{gdb/mi}
19456 @cindex @sc{gdb/mi}, async records
19457 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
19458 additional changes that have occurred. Those changes can either be a
19459 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
19460 target activity (e.g., target stopped).
19462 The following is the list of possible async records:
19466 @item *running,thread-id="@var{thread}"
19467 The target is now running. The @var{thread} field tells which
19468 specific thread is now running, and can be @samp{all} if all threads
19469 are running. The frontend should assume that no interaction with a
19470 running thread is possible after this notification is produced.
19471 The frontend should not assume that this notification is output
19472 only once for any command. @value{GDBN} may emit this notification
19473 several times, either for different threads, because it cannot resume
19474 all threads together, or even for a single thread, if the thread must
19475 be stepped though some code before letting it run freely.
19477 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}"
19478 The target has stopped. The @var{reason} field can have one of the
19482 @item breakpoint-hit
19483 A breakpoint was reached.
19484 @item watchpoint-trigger
19485 A watchpoint was triggered.
19486 @item read-watchpoint-trigger
19487 A read watchpoint was triggered.
19488 @item access-watchpoint-trigger
19489 An access watchpoint was triggered.
19490 @item function-finished
19491 An -exec-finish or similar CLI command was accomplished.
19492 @item location-reached
19493 An -exec-until or similar CLI command was accomplished.
19494 @item watchpoint-scope
19495 A watchpoint has gone out of scope.
19496 @item end-stepping-range
19497 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
19498 similar CLI command was accomplished.
19499 @item exited-signalled
19500 The inferior exited because of a signal.
19502 The inferior exited.
19503 @item exited-normally
19504 The inferior exited normally.
19505 @item signal-received
19506 A signal was received by the inferior.
19509 The @var{id} field identifies the thread that directly caused the stop
19510 -- for example by hitting a breakpoint. Depending on whether all-stop
19511 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
19512 stop all threads, or only the thread that directly triggered the stop.
19513 If all threads are stopped, the @var{stopped} field will have the
19514 value of @code{"all"}. Otherwise, the value of the @var{stopped}
19515 field will be a list of thread identifiers. Presently, this list will
19516 always include a single thread, but frontend should be prepared to see
19517 several threads in the list.
19519 @item =thread-group-created,id="@var{id}"
19520 @itemx =thread-group-exited,id="@var{id}"
19521 A thread thread group either was attached to, or has exited/detached
19522 from. The @var{id} field contains the @value{GDBN} identifier of the
19525 @item =thread-created,id="@var{id}",group-id="@var{gid}"
19526 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
19527 A thread either was created, or has exited. The @var{id} field
19528 contains the @value{GDBN} identifier of the thread. The @var{gid}
19529 field identifies the thread group this thread belongs to.
19531 @item =thread-selected,id="@var{id}"
19532 Informs that the selected thread was changed as result of the last
19533 command. This notification is not emitted as result of @code{-thread-select}
19534 command but is emitted whenever an MI command that is not documented
19535 to change the selected thread actually changes it. In particular,
19536 invoking, directly or indirectly (via user-defined command), the CLI
19537 @code{thread} command, will generate this notification.
19539 We suggest that in response to this notification, front ends
19540 highlight the selected thread and cause subsequent commands to apply to
19545 @node GDB/MI Frame Information
19546 @subsection @sc{gdb/mi} Frame Information
19548 Response from many MI commands includes an information about stack
19549 frame. This information is a tuple that may have the following
19554 The level of the stack frame. The innermost frame has the level of
19555 zero. This field is always present.
19558 The name of the function corresponding to the frame. This field may
19559 be absent if @value{GDBN} is unable to determine the function name.
19562 The code address for the frame. This field is always present.
19565 The name of the source files that correspond to the frame's code
19566 address. This field may be absent.
19569 The source line corresponding to the frames' code address. This field
19573 The name of the binary file (either executable or shared library) the
19574 corresponds to the frame's code address. This field may be absent.
19579 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19580 @node GDB/MI Simple Examples
19581 @section Simple Examples of @sc{gdb/mi} Interaction
19582 @cindex @sc{gdb/mi}, simple examples
19584 This subsection presents several simple examples of interaction using
19585 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
19586 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
19587 the output received from @sc{gdb/mi}.
19589 Note the line breaks shown in the examples are here only for
19590 readability, they don't appear in the real output.
19592 @subheading Setting a Breakpoint
19594 Setting a breakpoint generates synchronous output which contains detailed
19595 information of the breakpoint.
19598 -> -break-insert main
19599 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
19600 enabled="y",addr="0x08048564",func="main",file="myprog.c",
19601 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
19605 @subheading Program Execution
19607 Program execution generates asynchronous records and MI gives the
19608 reason that execution stopped.
19614 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
19615 frame=@{addr="0x08048564",func="main",
19616 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
19617 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
19622 <- *stopped,reason="exited-normally"
19626 @subheading Quitting @value{GDBN}
19628 Quitting @value{GDBN} just prints the result class @samp{^exit}.
19636 @subheading A Bad Command
19638 Here's what happens if you pass a non-existent command:
19642 <- ^error,msg="Undefined MI command: rubbish"
19647 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19648 @node GDB/MI Command Description Format
19649 @section @sc{gdb/mi} Command Description Format
19651 The remaining sections describe blocks of commands. Each block of
19652 commands is laid out in a fashion similar to this section.
19654 @subheading Motivation
19656 The motivation for this collection of commands.
19658 @subheading Introduction
19660 A brief introduction to this collection of commands as a whole.
19662 @subheading Commands
19664 For each command in the block, the following is described:
19666 @subsubheading Synopsis
19669 -command @var{args}@dots{}
19672 @subsubheading Result
19674 @subsubheading @value{GDBN} Command
19676 The corresponding @value{GDBN} CLI command(s), if any.
19678 @subsubheading Example
19680 Example(s) formatted for readability. Some of the described commands have
19681 not been implemented yet and these are labeled N.A.@: (not available).
19684 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19685 @node GDB/MI Breakpoint Commands
19686 @section @sc{gdb/mi} Breakpoint Commands
19688 @cindex breakpoint commands for @sc{gdb/mi}
19689 @cindex @sc{gdb/mi}, breakpoint commands
19690 This section documents @sc{gdb/mi} commands for manipulating
19693 @subheading The @code{-break-after} Command
19694 @findex -break-after
19696 @subsubheading Synopsis
19699 -break-after @var{number} @var{count}
19702 The breakpoint number @var{number} is not in effect until it has been
19703 hit @var{count} times. To see how this is reflected in the output of
19704 the @samp{-break-list} command, see the description of the
19705 @samp{-break-list} command below.
19707 @subsubheading @value{GDBN} Command
19709 The corresponding @value{GDBN} command is @samp{ignore}.
19711 @subsubheading Example
19716 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
19717 enabled="y",addr="0x000100d0",func="main",file="hello.c",
19718 fullname="/home/foo/hello.c",line="5",times="0"@}
19725 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
19726 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
19727 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
19728 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
19729 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
19730 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
19731 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
19732 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
19733 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
19734 line="5",times="0",ignore="3"@}]@}
19739 @subheading The @code{-break-catch} Command
19740 @findex -break-catch
19742 @subheading The @code{-break-commands} Command
19743 @findex -break-commands
19747 @subheading The @code{-break-condition} Command
19748 @findex -break-condition
19750 @subsubheading Synopsis
19753 -break-condition @var{number} @var{expr}
19756 Breakpoint @var{number} will stop the program only if the condition in
19757 @var{expr} is true. The condition becomes part of the
19758 @samp{-break-list} output (see the description of the @samp{-break-list}
19761 @subsubheading @value{GDBN} Command
19763 The corresponding @value{GDBN} command is @samp{condition}.
19765 @subsubheading Example
19769 -break-condition 1 1
19773 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
19774 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
19775 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
19776 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
19777 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
19778 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
19779 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
19780 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
19781 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
19782 line="5",cond="1",times="0",ignore="3"@}]@}
19786 @subheading The @code{-break-delete} Command
19787 @findex -break-delete
19789 @subsubheading Synopsis
19792 -break-delete ( @var{breakpoint} )+
19795 Delete the breakpoint(s) whose number(s) are specified in the argument
19796 list. This is obviously reflected in the breakpoint list.
19798 @subsubheading @value{GDBN} Command
19800 The corresponding @value{GDBN} command is @samp{delete}.
19802 @subsubheading Example
19810 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
19811 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
19812 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
19813 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
19814 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
19815 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
19816 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
19821 @subheading The @code{-break-disable} Command
19822 @findex -break-disable
19824 @subsubheading Synopsis
19827 -break-disable ( @var{breakpoint} )+
19830 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
19831 break list is now set to @samp{n} for the named @var{breakpoint}(s).
19833 @subsubheading @value{GDBN} Command
19835 The corresponding @value{GDBN} command is @samp{disable}.
19837 @subsubheading Example
19845 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
19846 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
19847 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
19848 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
19849 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
19850 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
19851 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
19852 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
19853 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
19854 line="5",times="0"@}]@}
19858 @subheading The @code{-break-enable} Command
19859 @findex -break-enable
19861 @subsubheading Synopsis
19864 -break-enable ( @var{breakpoint} )+
19867 Enable (previously disabled) @var{breakpoint}(s).
19869 @subsubheading @value{GDBN} Command
19871 The corresponding @value{GDBN} command is @samp{enable}.
19873 @subsubheading Example
19881 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
19882 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
19883 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
19884 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
19885 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
19886 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
19887 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
19888 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
19889 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
19890 line="5",times="0"@}]@}
19894 @subheading The @code{-break-info} Command
19895 @findex -break-info
19897 @subsubheading Synopsis
19900 -break-info @var{breakpoint}
19904 Get information about a single breakpoint.
19906 @subsubheading @value{GDBN} Command
19908 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
19910 @subsubheading Example
19913 @subheading The @code{-break-insert} Command
19914 @findex -break-insert
19916 @subsubheading Synopsis
19919 -break-insert [ -t ] [ -h ] [ -f ]
19920 [ -c @var{condition} ] [ -i @var{ignore-count} ]
19921 [ -p @var{thread} ] [ @var{location} ]
19925 If specified, @var{location}, can be one of:
19932 @item filename:linenum
19933 @item filename:function
19937 The possible optional parameters of this command are:
19941 Insert a temporary breakpoint.
19943 Insert a hardware breakpoint.
19944 @item -c @var{condition}
19945 Make the breakpoint conditional on @var{condition}.
19946 @item -i @var{ignore-count}
19947 Initialize the @var{ignore-count}.
19949 If @var{location} cannot be parsed (for example if it
19950 refers to unknown files or functions), create a pending
19951 breakpoint. Without this flag, @value{GDBN} will report
19952 an error, and won't create a breakpoint, if @var{location}
19956 @subsubheading Result
19958 The result is in the form:
19961 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
19962 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
19963 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
19964 times="@var{times}"@}
19968 where @var{number} is the @value{GDBN} number for this breakpoint,
19969 @var{funcname} is the name of the function where the breakpoint was
19970 inserted, @var{filename} is the name of the source file which contains
19971 this function, @var{lineno} is the source line number within that file
19972 and @var{times} the number of times that the breakpoint has been hit
19973 (always 0 for -break-insert but may be greater for -break-info or -break-list
19974 which use the same output).
19976 Note: this format is open to change.
19977 @c An out-of-band breakpoint instead of part of the result?
19979 @subsubheading @value{GDBN} Command
19981 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
19982 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
19984 @subsubheading Example
19989 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
19990 fullname="/home/foo/recursive2.c,line="4",times="0"@}
19992 -break-insert -t foo
19993 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
19994 fullname="/home/foo/recursive2.c,line="11",times="0"@}
19997 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
19998 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
19999 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
20000 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
20001 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
20002 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
20003 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
20004 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
20005 addr="0x0001072c", func="main",file="recursive2.c",
20006 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
20007 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
20008 addr="0x00010774",func="foo",file="recursive2.c",
20009 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
20011 -break-insert -r foo.*
20012 ~int foo(int, int);
20013 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
20014 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
20018 @subheading The @code{-break-list} Command
20019 @findex -break-list
20021 @subsubheading Synopsis
20027 Displays the list of inserted breakpoints, showing the following fields:
20031 number of the breakpoint
20033 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
20035 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
20038 is the breakpoint enabled or no: @samp{y} or @samp{n}
20040 memory location at which the breakpoint is set
20042 logical location of the breakpoint, expressed by function name, file
20045 number of times the breakpoint has been hit
20048 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
20049 @code{body} field is an empty list.
20051 @subsubheading @value{GDBN} Command
20053 The corresponding @value{GDBN} command is @samp{info break}.
20055 @subsubheading Example
20060 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
20061 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
20062 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
20063 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
20064 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
20065 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
20066 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
20067 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
20068 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
20069 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
20070 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
20071 line="13",times="0"@}]@}
20075 Here's an example of the result when there are no breakpoints:
20080 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
20081 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
20082 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
20083 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
20084 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
20085 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
20086 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
20091 @subheading The @code{-break-watch} Command
20092 @findex -break-watch
20094 @subsubheading Synopsis
20097 -break-watch [ -a | -r ]
20100 Create a watchpoint. With the @samp{-a} option it will create an
20101 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
20102 read from or on a write to the memory location. With the @samp{-r}
20103 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
20104 trigger only when the memory location is accessed for reading. Without
20105 either of the options, the watchpoint created is a regular watchpoint,
20106 i.e., it will trigger when the memory location is accessed for writing.
20107 @xref{Set Watchpoints, , Setting Watchpoints}.
20109 Note that @samp{-break-list} will report a single list of watchpoints and
20110 breakpoints inserted.
20112 @subsubheading @value{GDBN} Command
20114 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
20117 @subsubheading Example
20119 Setting a watchpoint on a variable in the @code{main} function:
20124 ^done,wpt=@{number="2",exp="x"@}
20129 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
20130 value=@{old="-268439212",new="55"@},
20131 frame=@{func="main",args=[],file="recursive2.c",
20132 fullname="/home/foo/bar/recursive2.c",line="5"@}
20136 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
20137 the program execution twice: first for the variable changing value, then
20138 for the watchpoint going out of scope.
20143 ^done,wpt=@{number="5",exp="C"@}
20148 *stopped,reason="watchpoint-trigger",
20149 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
20150 frame=@{func="callee4",args=[],
20151 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20152 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
20157 *stopped,reason="watchpoint-scope",wpnum="5",
20158 frame=@{func="callee3",args=[@{name="strarg",
20159 value="0x11940 \"A string argument.\""@}],
20160 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20161 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
20165 Listing breakpoints and watchpoints, at different points in the program
20166 execution. Note that once the watchpoint goes out of scope, it is
20172 ^done,wpt=@{number="2",exp="C"@}
20175 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
20176 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
20177 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
20178 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
20179 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
20180 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
20181 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
20182 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
20183 addr="0x00010734",func="callee4",
20184 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20185 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
20186 bkpt=@{number="2",type="watchpoint",disp="keep",
20187 enabled="y",addr="",what="C",times="0"@}]@}
20192 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
20193 value=@{old="-276895068",new="3"@},
20194 frame=@{func="callee4",args=[],
20195 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20196 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
20199 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
20200 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
20201 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
20202 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
20203 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
20204 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
20205 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
20206 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
20207 addr="0x00010734",func="callee4",
20208 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20209 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
20210 bkpt=@{number="2",type="watchpoint",disp="keep",
20211 enabled="y",addr="",what="C",times="-5"@}]@}
20215 ^done,reason="watchpoint-scope",wpnum="2",
20216 frame=@{func="callee3",args=[@{name="strarg",
20217 value="0x11940 \"A string argument.\""@}],
20218 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20219 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
20222 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
20223 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
20224 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
20225 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
20226 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
20227 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
20228 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
20229 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
20230 addr="0x00010734",func="callee4",
20231 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20232 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
20237 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20238 @node GDB/MI Program Context
20239 @section @sc{gdb/mi} Program Context
20241 @subheading The @code{-exec-arguments} Command
20242 @findex -exec-arguments
20245 @subsubheading Synopsis
20248 -exec-arguments @var{args}
20251 Set the inferior program arguments, to be used in the next
20254 @subsubheading @value{GDBN} Command
20256 The corresponding @value{GDBN} command is @samp{set args}.
20258 @subsubheading Example
20262 -exec-arguments -v word
20268 @subheading The @code{-exec-show-arguments} Command
20269 @findex -exec-show-arguments
20271 @subsubheading Synopsis
20274 -exec-show-arguments
20277 Print the arguments of the program.
20279 @subsubheading @value{GDBN} Command
20281 The corresponding @value{GDBN} command is @samp{show args}.
20283 @subsubheading Example
20287 @subheading The @code{-environment-cd} Command
20288 @findex -environment-cd
20290 @subsubheading Synopsis
20293 -environment-cd @var{pathdir}
20296 Set @value{GDBN}'s working directory.
20298 @subsubheading @value{GDBN} Command
20300 The corresponding @value{GDBN} command is @samp{cd}.
20302 @subsubheading Example
20306 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
20312 @subheading The @code{-environment-directory} Command
20313 @findex -environment-directory
20315 @subsubheading Synopsis
20318 -environment-directory [ -r ] [ @var{pathdir} ]+
20321 Add directories @var{pathdir} to beginning of search path for source files.
20322 If the @samp{-r} option is used, the search path is reset to the default
20323 search path. If directories @var{pathdir} are supplied in addition to the
20324 @samp{-r} option, the search path is first reset and then addition
20326 Multiple directories may be specified, separated by blanks. Specifying
20327 multiple directories in a single command
20328 results in the directories added to the beginning of the
20329 search path in the same order they were presented in the command.
20330 If blanks are needed as
20331 part of a directory name, double-quotes should be used around
20332 the name. In the command output, the path will show up separated
20333 by the system directory-separator character. The directory-separator
20334 character must not be used
20335 in any directory name.
20336 If no directories are specified, the current search path is displayed.
20338 @subsubheading @value{GDBN} Command
20340 The corresponding @value{GDBN} command is @samp{dir}.
20342 @subsubheading Example
20346 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
20347 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
20349 -environment-directory ""
20350 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
20352 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
20353 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
20355 -environment-directory -r
20356 ^done,source-path="$cdir:$cwd"
20361 @subheading The @code{-environment-path} Command
20362 @findex -environment-path
20364 @subsubheading Synopsis
20367 -environment-path [ -r ] [ @var{pathdir} ]+
20370 Add directories @var{pathdir} to beginning of search path for object files.
20371 If the @samp{-r} option is used, the search path is reset to the original
20372 search path that existed at gdb start-up. If directories @var{pathdir} are
20373 supplied in addition to the
20374 @samp{-r} option, the search path is first reset and then addition
20376 Multiple directories may be specified, separated by blanks. Specifying
20377 multiple directories in a single command
20378 results in the directories added to the beginning of the
20379 search path in the same order they were presented in the command.
20380 If blanks are needed as
20381 part of a directory name, double-quotes should be used around
20382 the name. In the command output, the path will show up separated
20383 by the system directory-separator character. The directory-separator
20384 character must not be used
20385 in any directory name.
20386 If no directories are specified, the current path is displayed.
20389 @subsubheading @value{GDBN} Command
20391 The corresponding @value{GDBN} command is @samp{path}.
20393 @subsubheading Example
20398 ^done,path="/usr/bin"
20400 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
20401 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
20403 -environment-path -r /usr/local/bin
20404 ^done,path="/usr/local/bin:/usr/bin"
20409 @subheading The @code{-environment-pwd} Command
20410 @findex -environment-pwd
20412 @subsubheading Synopsis
20418 Show the current working directory.
20420 @subsubheading @value{GDBN} Command
20422 The corresponding @value{GDBN} command is @samp{pwd}.
20424 @subsubheading Example
20429 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
20433 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20434 @node GDB/MI Thread Commands
20435 @section @sc{gdb/mi} Thread Commands
20438 @subheading The @code{-thread-info} Command
20439 @findex -thread-info
20441 @subsubheading Synopsis
20444 -thread-info [ @var{thread-id} ]
20447 Reports information about either a specific thread, if
20448 the @var{thread-id} parameter is present, or about all
20449 threads. When printing information about all threads,
20450 also reports the current thread.
20452 @subsubheading @value{GDBN} Command
20454 The @samp{info thread} command prints the same information
20457 @subsubheading Example
20462 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
20463 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
20464 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
20465 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
20466 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}],
20467 current-thread-id="1"
20471 The @samp{state} field may have the following values:
20475 The thread is stopped. Frame information is available for stopped
20479 The thread is running. There's no frame information for running
20484 @subheading The @code{-thread-list-ids} Command
20485 @findex -thread-list-ids
20487 @subsubheading Synopsis
20493 Produces a list of the currently known @value{GDBN} thread ids. At the
20494 end of the list it also prints the total number of such threads.
20496 This command is retained for historical reasons, the
20497 @code{-thread-info} command should be used instead.
20499 @subsubheading @value{GDBN} Command
20501 Part of @samp{info threads} supplies the same information.
20503 @subsubheading Example
20505 No threads present, besides the main process:
20510 ^done,thread-ids=@{@},number-of-threads="0"
20520 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
20521 number-of-threads="3"
20526 @subheading The @code{-thread-select} Command
20527 @findex -thread-select
20529 @subsubheading Synopsis
20532 -thread-select @var{threadnum}
20535 Make @var{threadnum} the current thread. It prints the number of the new
20536 current thread, and the topmost frame for that thread.
20538 This command is deprecated in favor of explicitly using the
20539 @samp{--thread} option to each command.
20541 @subsubheading @value{GDBN} Command
20543 The corresponding @value{GDBN} command is @samp{thread}.
20545 @subsubheading Example
20552 *stopped,reason="end-stepping-range",thread-id="2",line="187",
20553 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
20557 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
20558 number-of-threads="3"
20561 ^done,new-thread-id="3",
20562 frame=@{level="0",func="vprintf",
20563 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
20564 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
20568 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20569 @node GDB/MI Program Execution
20570 @section @sc{gdb/mi} Program Execution
20572 These are the asynchronous commands which generate the out-of-band
20573 record @samp{*stopped}. Currently @value{GDBN} only really executes
20574 asynchronously with remote targets and this interaction is mimicked in
20577 @subheading The @code{-exec-continue} Command
20578 @findex -exec-continue
20580 @subsubheading Synopsis
20583 -exec-continue [--all|--thread-group N]
20586 Resumes the execution of the inferior program until a breakpoint is
20587 encountered, or until the inferior exits. In all-stop mode
20588 (@pxref{All-Stop Mode}), may resume only one thread, or all threads,
20589 depending on the value of the @samp{scheduler-locking} variable. In
20590 non-stop mode (@pxref{Non-Stop Mode}), if the @samp{--all} is not
20591 specified, only the thread specified with the @samp{--thread} option
20592 (or current thread, if no @samp{--thread} is provided) is resumed. If
20593 @samp{--all} is specified, all threads will be resumed. The
20594 @samp{--all} option is ignored in all-stop mode. If the
20595 @samp{--thread-group} options is specified, then all threads in that
20596 thread group are resumed.
20598 @subsubheading @value{GDBN} Command
20600 The corresponding @value{GDBN} corresponding is @samp{continue}.
20602 @subsubheading Example
20609 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
20610 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
20616 @subheading The @code{-exec-finish} Command
20617 @findex -exec-finish
20619 @subsubheading Synopsis
20625 Resumes the execution of the inferior program until the current
20626 function is exited. Displays the results returned by the function.
20628 @subsubheading @value{GDBN} Command
20630 The corresponding @value{GDBN} command is @samp{finish}.
20632 @subsubheading Example
20634 Function returning @code{void}.
20641 *stopped,reason="function-finished",frame=@{func="main",args=[],
20642 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
20646 Function returning other than @code{void}. The name of the internal
20647 @value{GDBN} variable storing the result is printed, together with the
20654 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
20655 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
20656 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
20657 gdb-result-var="$1",return-value="0"
20662 @subheading The @code{-exec-interrupt} Command
20663 @findex -exec-interrupt
20665 @subsubheading Synopsis
20668 -exec-interrupt [--all|--thread-group N]
20671 Interrupts the background execution of the target. Note how the token
20672 associated with the stop message is the one for the execution command
20673 that has been interrupted. The token for the interrupt itself only
20674 appears in the @samp{^done} output. If the user is trying to
20675 interrupt a non-running program, an error message will be printed.
20677 Note that when asynchronous execution is enabled, this command is
20678 asynchronous just like other execution commands. That is, first the
20679 @samp{^done} response will be printed, and the target stop will be
20680 reported after that using the @samp{*stopped} notification.
20682 In non-stop mode, only the context thread is interrupted by default.
20683 All threads will be interrupted if the @samp{--all} option is
20684 specified. If the @samp{--thread-group} option is specified, all
20685 threads in that group will be interrupted.
20687 @subsubheading @value{GDBN} Command
20689 The corresponding @value{GDBN} command is @samp{interrupt}.
20691 @subsubheading Example
20702 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
20703 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
20704 fullname="/home/foo/bar/try.c",line="13"@}
20709 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
20714 @subheading The @code{-exec-next} Command
20717 @subsubheading Synopsis
20723 Resumes execution of the inferior program, stopping when the beginning
20724 of the next source line is reached.
20726 @subsubheading @value{GDBN} Command
20728 The corresponding @value{GDBN} command is @samp{next}.
20730 @subsubheading Example
20736 *stopped,reason="end-stepping-range",line="8",file="hello.c"
20741 @subheading The @code{-exec-next-instruction} Command
20742 @findex -exec-next-instruction
20744 @subsubheading Synopsis
20747 -exec-next-instruction
20750 Executes one machine instruction. If the instruction is a function
20751 call, continues until the function returns. If the program stops at an
20752 instruction in the middle of a source line, the address will be
20755 @subsubheading @value{GDBN} Command
20757 The corresponding @value{GDBN} command is @samp{nexti}.
20759 @subsubheading Example
20763 -exec-next-instruction
20767 *stopped,reason="end-stepping-range",
20768 addr="0x000100d4",line="5",file="hello.c"
20773 @subheading The @code{-exec-return} Command
20774 @findex -exec-return
20776 @subsubheading Synopsis
20782 Makes current function return immediately. Doesn't execute the inferior.
20783 Displays the new current frame.
20785 @subsubheading @value{GDBN} Command
20787 The corresponding @value{GDBN} command is @samp{return}.
20789 @subsubheading Example
20793 200-break-insert callee4
20794 200^done,bkpt=@{number="1",addr="0x00010734",
20795 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
20800 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
20801 frame=@{func="callee4",args=[],
20802 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20803 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
20809 111^done,frame=@{level="0",func="callee3",
20810 args=[@{name="strarg",
20811 value="0x11940 \"A string argument.\""@}],
20812 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
20813 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
20818 @subheading The @code{-exec-run} Command
20821 @subsubheading Synopsis
20827 Starts execution of the inferior from the beginning. The inferior
20828 executes until either a breakpoint is encountered or the program
20829 exits. In the latter case the output will include an exit code, if
20830 the program has exited exceptionally.
20832 @subsubheading @value{GDBN} Command
20834 The corresponding @value{GDBN} command is @samp{run}.
20836 @subsubheading Examples
20841 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
20846 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
20847 frame=@{func="main",args=[],file="recursive2.c",
20848 fullname="/home/foo/bar/recursive2.c",line="4"@}
20853 Program exited normally:
20861 *stopped,reason="exited-normally"
20866 Program exited exceptionally:
20874 *stopped,reason="exited",exit-code="01"
20878 Another way the program can terminate is if it receives a signal such as
20879 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
20883 *stopped,reason="exited-signalled",signal-name="SIGINT",
20884 signal-meaning="Interrupt"
20888 @c @subheading -exec-signal
20891 @subheading The @code{-exec-step} Command
20894 @subsubheading Synopsis
20900 Resumes execution of the inferior program, stopping when the beginning
20901 of the next source line is reached, if the next source line is not a
20902 function call. If it is, stop at the first instruction of the called
20905 @subsubheading @value{GDBN} Command
20907 The corresponding @value{GDBN} command is @samp{step}.
20909 @subsubheading Example
20911 Stepping into a function:
20917 *stopped,reason="end-stepping-range",
20918 frame=@{func="foo",args=[@{name="a",value="10"@},
20919 @{name="b",value="0"@}],file="recursive2.c",
20920 fullname="/home/foo/bar/recursive2.c",line="11"@}
20930 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
20935 @subheading The @code{-exec-step-instruction} Command
20936 @findex -exec-step-instruction
20938 @subsubheading Synopsis
20941 -exec-step-instruction
20944 Resumes the inferior which executes one machine instruction. The
20945 output, once @value{GDBN} has stopped, will vary depending on whether
20946 we have stopped in the middle of a source line or not. In the former
20947 case, the address at which the program stopped will be printed as
20950 @subsubheading @value{GDBN} Command
20952 The corresponding @value{GDBN} command is @samp{stepi}.
20954 @subsubheading Example
20958 -exec-step-instruction
20962 *stopped,reason="end-stepping-range",
20963 frame=@{func="foo",args=[],file="try.c",
20964 fullname="/home/foo/bar/try.c",line="10"@}
20966 -exec-step-instruction
20970 *stopped,reason="end-stepping-range",
20971 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
20972 fullname="/home/foo/bar/try.c",line="10"@}
20977 @subheading The @code{-exec-until} Command
20978 @findex -exec-until
20980 @subsubheading Synopsis
20983 -exec-until [ @var{location} ]
20986 Executes the inferior until the @var{location} specified in the
20987 argument is reached. If there is no argument, the inferior executes
20988 until a source line greater than the current one is reached. The
20989 reason for stopping in this case will be @samp{location-reached}.
20991 @subsubheading @value{GDBN} Command
20993 The corresponding @value{GDBN} command is @samp{until}.
20995 @subsubheading Example
20999 -exec-until recursive2.c:6
21003 *stopped,reason="location-reached",frame=@{func="main",args=[],
21004 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
21009 @subheading -file-clear
21010 Is this going away????
21013 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21014 @node GDB/MI Stack Manipulation
21015 @section @sc{gdb/mi} Stack Manipulation Commands
21018 @subheading The @code{-stack-info-frame} Command
21019 @findex -stack-info-frame
21021 @subsubheading Synopsis
21027 Get info on the selected frame.
21029 @subsubheading @value{GDBN} Command
21031 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
21032 (without arguments).
21034 @subsubheading Example
21039 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
21040 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21041 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
21045 @subheading The @code{-stack-info-depth} Command
21046 @findex -stack-info-depth
21048 @subsubheading Synopsis
21051 -stack-info-depth [ @var{max-depth} ]
21054 Return the depth of the stack. If the integer argument @var{max-depth}
21055 is specified, do not count beyond @var{max-depth} frames.
21057 @subsubheading @value{GDBN} Command
21059 There's no equivalent @value{GDBN} command.
21061 @subsubheading Example
21063 For a stack with frame levels 0 through 11:
21070 -stack-info-depth 4
21073 -stack-info-depth 12
21076 -stack-info-depth 11
21079 -stack-info-depth 13
21084 @subheading The @code{-stack-list-arguments} Command
21085 @findex -stack-list-arguments
21087 @subsubheading Synopsis
21090 -stack-list-arguments @var{show-values}
21091 [ @var{low-frame} @var{high-frame} ]
21094 Display a list of the arguments for the frames between @var{low-frame}
21095 and @var{high-frame} (inclusive). If @var{low-frame} and
21096 @var{high-frame} are not provided, list the arguments for the whole
21097 call stack. If the two arguments are equal, show the single frame
21098 at the corresponding level. It is an error if @var{low-frame} is
21099 larger than the actual number of frames. On the other hand,
21100 @var{high-frame} may be larger than the actual number of frames, in
21101 which case only existing frames will be returned.
21103 The @var{show-values} argument must have a value of 0 or 1. A value of
21104 0 means that only the names of the arguments are listed, a value of 1
21105 means that both names and values of the arguments are printed.
21107 @subsubheading @value{GDBN} Command
21109 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
21110 @samp{gdb_get_args} command which partially overlaps with the
21111 functionality of @samp{-stack-list-arguments}.
21113 @subsubheading Example
21120 frame=@{level="0",addr="0x00010734",func="callee4",
21121 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21122 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
21123 frame=@{level="1",addr="0x0001076c",func="callee3",
21124 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21125 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
21126 frame=@{level="2",addr="0x0001078c",func="callee2",
21127 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21128 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
21129 frame=@{level="3",addr="0x000107b4",func="callee1",
21130 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21131 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
21132 frame=@{level="4",addr="0x000107e0",func="main",
21133 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21134 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
21136 -stack-list-arguments 0
21139 frame=@{level="0",args=[]@},
21140 frame=@{level="1",args=[name="strarg"]@},
21141 frame=@{level="2",args=[name="intarg",name="strarg"]@},
21142 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
21143 frame=@{level="4",args=[]@}]
21145 -stack-list-arguments 1
21148 frame=@{level="0",args=[]@},
21150 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
21151 frame=@{level="2",args=[
21152 @{name="intarg",value="2"@},
21153 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
21154 @{frame=@{level="3",args=[
21155 @{name="intarg",value="2"@},
21156 @{name="strarg",value="0x11940 \"A string argument.\""@},
21157 @{name="fltarg",value="3.5"@}]@},
21158 frame=@{level="4",args=[]@}]
21160 -stack-list-arguments 0 2 2
21161 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
21163 -stack-list-arguments 1 2 2
21164 ^done,stack-args=[frame=@{level="2",
21165 args=[@{name="intarg",value="2"@},
21166 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
21170 @c @subheading -stack-list-exception-handlers
21173 @subheading The @code{-stack-list-frames} Command
21174 @findex -stack-list-frames
21176 @subsubheading Synopsis
21179 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
21182 List the frames currently on the stack. For each frame it displays the
21187 The frame number, 0 being the topmost frame, i.e., the innermost function.
21189 The @code{$pc} value for that frame.
21193 File name of the source file where the function lives.
21195 Line number corresponding to the @code{$pc}.
21198 If invoked without arguments, this command prints a backtrace for the
21199 whole stack. If given two integer arguments, it shows the frames whose
21200 levels are between the two arguments (inclusive). If the two arguments
21201 are equal, it shows the single frame at the corresponding level. It is
21202 an error if @var{low-frame} is larger than the actual number of
21203 frames. On the other hand, @var{high-frame} may be larger than the
21204 actual number of frames, in which case only existing frames will be returned.
21206 @subsubheading @value{GDBN} Command
21208 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
21210 @subsubheading Example
21212 Full stack backtrace:
21218 [frame=@{level="0",addr="0x0001076c",func="foo",
21219 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
21220 frame=@{level="1",addr="0x000107a4",func="foo",
21221 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21222 frame=@{level="2",addr="0x000107a4",func="foo",
21223 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21224 frame=@{level="3",addr="0x000107a4",func="foo",
21225 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21226 frame=@{level="4",addr="0x000107a4",func="foo",
21227 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21228 frame=@{level="5",addr="0x000107a4",func="foo",
21229 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21230 frame=@{level="6",addr="0x000107a4",func="foo",
21231 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21232 frame=@{level="7",addr="0x000107a4",func="foo",
21233 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21234 frame=@{level="8",addr="0x000107a4",func="foo",
21235 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21236 frame=@{level="9",addr="0x000107a4",func="foo",
21237 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21238 frame=@{level="10",addr="0x000107a4",func="foo",
21239 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21240 frame=@{level="11",addr="0x00010738",func="main",
21241 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
21245 Show frames between @var{low_frame} and @var{high_frame}:
21249 -stack-list-frames 3 5
21251 [frame=@{level="3",addr="0x000107a4",func="foo",
21252 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21253 frame=@{level="4",addr="0x000107a4",func="foo",
21254 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
21255 frame=@{level="5",addr="0x000107a4",func="foo",
21256 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
21260 Show a single frame:
21264 -stack-list-frames 3 3
21266 [frame=@{level="3",addr="0x000107a4",func="foo",
21267 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
21272 @subheading The @code{-stack-list-locals} Command
21273 @findex -stack-list-locals
21275 @subsubheading Synopsis
21278 -stack-list-locals @var{print-values}
21281 Display the local variable names for the selected frame. If
21282 @var{print-values} is 0 or @code{--no-values}, print only the names of
21283 the variables; if it is 1 or @code{--all-values}, print also their
21284 values; and if it is 2 or @code{--simple-values}, print the name,
21285 type and value for simple data types and the name and type for arrays,
21286 structures and unions. In this last case, a frontend can immediately
21287 display the value of simple data types and create variable objects for
21288 other data types when the user wishes to explore their values in
21291 @subsubheading @value{GDBN} Command
21293 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
21295 @subsubheading Example
21299 -stack-list-locals 0
21300 ^done,locals=[name="A",name="B",name="C"]
21302 -stack-list-locals --all-values
21303 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
21304 @{name="C",value="@{1, 2, 3@}"@}]
21305 -stack-list-locals --simple-values
21306 ^done,locals=[@{name="A",type="int",value="1"@},
21307 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
21312 @subheading The @code{-stack-select-frame} Command
21313 @findex -stack-select-frame
21315 @subsubheading Synopsis
21318 -stack-select-frame @var{framenum}
21321 Change the selected frame. Select a different frame @var{framenum} on
21324 This command in deprecated in favor of passing the @samp{--frame}
21325 option to every command.
21327 @subsubheading @value{GDBN} Command
21329 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
21330 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
21332 @subsubheading Example
21336 -stack-select-frame 2
21341 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21342 @node GDB/MI Variable Objects
21343 @section @sc{gdb/mi} Variable Objects
21347 @subheading Motivation for Variable Objects in @sc{gdb/mi}
21349 For the implementation of a variable debugger window (locals, watched
21350 expressions, etc.), we are proposing the adaptation of the existing code
21351 used by @code{Insight}.
21353 The two main reasons for that are:
21357 It has been proven in practice (it is already on its second generation).
21360 It will shorten development time (needless to say how important it is
21364 The original interface was designed to be used by Tcl code, so it was
21365 slightly changed so it could be used through @sc{gdb/mi}. This section
21366 describes the @sc{gdb/mi} operations that will be available and gives some
21367 hints about their use.
21369 @emph{Note}: In addition to the set of operations described here, we
21370 expect the @sc{gui} implementation of a variable window to require, at
21371 least, the following operations:
21374 @item @code{-gdb-show} @code{output-radix}
21375 @item @code{-stack-list-arguments}
21376 @item @code{-stack-list-locals}
21377 @item @code{-stack-select-frame}
21382 @subheading Introduction to Variable Objects
21384 @cindex variable objects in @sc{gdb/mi}
21386 Variable objects are "object-oriented" MI interface for examining and
21387 changing values of expressions. Unlike some other MI interfaces that
21388 work with expressions, variable objects are specifically designed for
21389 simple and efficient presentation in the frontend. A variable object
21390 is identified by string name. When a variable object is created, the
21391 frontend specifies the expression for that variable object. The
21392 expression can be a simple variable, or it can be an arbitrary complex
21393 expression, and can even involve CPU registers. After creating a
21394 variable object, the frontend can invoke other variable object
21395 operations---for example to obtain or change the value of a variable
21396 object, or to change display format.
21398 Variable objects have hierarchical tree structure. Any variable object
21399 that corresponds to a composite type, such as structure in C, has
21400 a number of child variable objects, for example corresponding to each
21401 element of a structure. A child variable object can itself have
21402 children, recursively. Recursion ends when we reach
21403 leaf variable objects, which always have built-in types. Child variable
21404 objects are created only by explicit request, so if a frontend
21405 is not interested in the children of a particular variable object, no
21406 child will be created.
21408 For a leaf variable object it is possible to obtain its value as a
21409 string, or set the value from a string. String value can be also
21410 obtained for a non-leaf variable object, but it's generally a string
21411 that only indicates the type of the object, and does not list its
21412 contents. Assignment to a non-leaf variable object is not allowed.
21414 A frontend does not need to read the values of all variable objects each time
21415 the program stops. Instead, MI provides an update command that lists all
21416 variable objects whose values has changed since the last update
21417 operation. This considerably reduces the amount of data that must
21418 be transferred to the frontend. As noted above, children variable
21419 objects are created on demand, and only leaf variable objects have a
21420 real value. As result, gdb will read target memory only for leaf
21421 variables that frontend has created.
21423 The automatic update is not always desirable. For example, a frontend
21424 might want to keep a value of some expression for future reference,
21425 and never update it. For another example, fetching memory is
21426 relatively slow for embedded targets, so a frontend might want
21427 to disable automatic update for the variables that are either not
21428 visible on the screen, or ``closed''. This is possible using so
21429 called ``frozen variable objects''. Such variable objects are never
21430 implicitly updated.
21432 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
21433 fixed variable object, the expression is parsed when the variable
21434 object is created, including associating identifiers to specific
21435 variables. The meaning of expression never changes. For a floating
21436 variable object the values of variables whose names appear in the
21437 expressions are re-evaluated every time in the context of the current
21438 frame. Consider this example:
21443 struct work_state state;
21450 If a fixed variable object for the @code{state} variable is created in
21451 this function, and we enter the recursive call, the the variable
21452 object will report the value of @code{state} in the top-level
21453 @code{do_work} invocation. On the other hand, a floating variable
21454 object will report the value of @code{state} in the current frame.
21456 If an expression specified when creating a fixed variable object
21457 refers to a local variable, the variable object becomes bound to the
21458 thread and frame in which the variable object is created. When such
21459 variable object is updated, @value{GDBN} makes sure that the
21460 thread/frame combination the variable object is bound to still exists,
21461 and re-evaluates the variable object in context of that thread/frame.
21463 The following is the complete set of @sc{gdb/mi} operations defined to
21464 access this functionality:
21466 @multitable @columnfractions .4 .6
21467 @item @strong{Operation}
21468 @tab @strong{Description}
21470 @item @code{-var-create}
21471 @tab create a variable object
21472 @item @code{-var-delete}
21473 @tab delete the variable object and/or its children
21474 @item @code{-var-set-format}
21475 @tab set the display format of this variable
21476 @item @code{-var-show-format}
21477 @tab show the display format of this variable
21478 @item @code{-var-info-num-children}
21479 @tab tells how many children this object has
21480 @item @code{-var-list-children}
21481 @tab return a list of the object's children
21482 @item @code{-var-info-type}
21483 @tab show the type of this variable object
21484 @item @code{-var-info-expression}
21485 @tab print parent-relative expression that this variable object represents
21486 @item @code{-var-info-path-expression}
21487 @tab print full expression that this variable object represents
21488 @item @code{-var-show-attributes}
21489 @tab is this variable editable? does it exist here?
21490 @item @code{-var-evaluate-expression}
21491 @tab get the value of this variable
21492 @item @code{-var-assign}
21493 @tab set the value of this variable
21494 @item @code{-var-update}
21495 @tab update the variable and its children
21496 @item @code{-var-set-frozen}
21497 @tab set frozeness attribute
21500 In the next subsection we describe each operation in detail and suggest
21501 how it can be used.
21503 @subheading Description And Use of Operations on Variable Objects
21505 @subheading The @code{-var-create} Command
21506 @findex -var-create
21508 @subsubheading Synopsis
21511 -var-create @{@var{name} | "-"@}
21512 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
21515 This operation creates a variable object, which allows the monitoring of
21516 a variable, the result of an expression, a memory cell or a CPU
21519 The @var{name} parameter is the string by which the object can be
21520 referenced. It must be unique. If @samp{-} is specified, the varobj
21521 system will generate a string ``varNNNNNN'' automatically. It will be
21522 unique provided that one does not specify @var{name} of that format.
21523 The command fails if a duplicate name is found.
21525 The frame under which the expression should be evaluated can be
21526 specified by @var{frame-addr}. A @samp{*} indicates that the current
21527 frame should be used. A @samp{@@} indicates that a floating variable
21528 object must be created.
21530 @var{expression} is any expression valid on the current language set (must not
21531 begin with a @samp{*}), or one of the following:
21535 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
21538 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
21541 @samp{$@var{regname}} --- a CPU register name
21544 @subsubheading Result
21546 This operation returns the name, number of children and the type of the
21547 object created. Type is returned as a string as the ones generated by
21548 the @value{GDBN} CLI. If a fixed variable object is bound to a
21549 specific thread, the thread is is also printed:
21552 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}"
21556 @subheading The @code{-var-delete} Command
21557 @findex -var-delete
21559 @subsubheading Synopsis
21562 -var-delete [ -c ] @var{name}
21565 Deletes a previously created variable object and all of its children.
21566 With the @samp{-c} option, just deletes the children.
21568 Returns an error if the object @var{name} is not found.
21571 @subheading The @code{-var-set-format} Command
21572 @findex -var-set-format
21574 @subsubheading Synopsis
21577 -var-set-format @var{name} @var{format-spec}
21580 Sets the output format for the value of the object @var{name} to be
21583 @anchor{-var-set-format}
21584 The syntax for the @var{format-spec} is as follows:
21587 @var{format-spec} @expansion{}
21588 @{binary | decimal | hexadecimal | octal | natural@}
21591 The natural format is the default format choosen automatically
21592 based on the variable type (like decimal for an @code{int}, hex
21593 for pointers, etc.).
21595 For a variable with children, the format is set only on the
21596 variable itself, and the children are not affected.
21598 @subheading The @code{-var-show-format} Command
21599 @findex -var-show-format
21601 @subsubheading Synopsis
21604 -var-show-format @var{name}
21607 Returns the format used to display the value of the object @var{name}.
21610 @var{format} @expansion{}
21615 @subheading The @code{-var-info-num-children} Command
21616 @findex -var-info-num-children
21618 @subsubheading Synopsis
21621 -var-info-num-children @var{name}
21624 Returns the number of children of a variable object @var{name}:
21631 @subheading The @code{-var-list-children} Command
21632 @findex -var-list-children
21634 @subsubheading Synopsis
21637 -var-list-children [@var{print-values}] @var{name}
21639 @anchor{-var-list-children}
21641 Return a list of the children of the specified variable object and
21642 create variable objects for them, if they do not already exist. With
21643 a single argument or if @var{print-values} has a value for of 0 or
21644 @code{--no-values}, print only the names of the variables; if
21645 @var{print-values} is 1 or @code{--all-values}, also print their
21646 values; and if it is 2 or @code{--simple-values} print the name and
21647 value for simple data types and just the name for arrays, structures
21650 @subsubheading Example
21654 -var-list-children n
21655 ^done,numchild=@var{n},children=[@{name=@var{name},
21656 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
21658 -var-list-children --all-values n
21659 ^done,numchild=@var{n},children=[@{name=@var{name},
21660 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
21664 @subheading The @code{-var-info-type} Command
21665 @findex -var-info-type
21667 @subsubheading Synopsis
21670 -var-info-type @var{name}
21673 Returns the type of the specified variable @var{name}. The type is
21674 returned as a string in the same format as it is output by the
21678 type=@var{typename}
21682 @subheading The @code{-var-info-expression} Command
21683 @findex -var-info-expression
21685 @subsubheading Synopsis
21688 -var-info-expression @var{name}
21691 Returns a string that is suitable for presenting this
21692 variable object in user interface. The string is generally
21693 not valid expression in the current language, and cannot be evaluated.
21695 For example, if @code{a} is an array, and variable object
21696 @code{A} was created for @code{a}, then we'll get this output:
21699 (gdb) -var-info-expression A.1
21700 ^done,lang="C",exp="1"
21704 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
21706 Note that the output of the @code{-var-list-children} command also
21707 includes those expressions, so the @code{-var-info-expression} command
21710 @subheading The @code{-var-info-path-expression} Command
21711 @findex -var-info-path-expression
21713 @subsubheading Synopsis
21716 -var-info-path-expression @var{name}
21719 Returns an expression that can be evaluated in the current
21720 context and will yield the same value that a variable object has.
21721 Compare this with the @code{-var-info-expression} command, which
21722 result can be used only for UI presentation. Typical use of
21723 the @code{-var-info-path-expression} command is creating a
21724 watchpoint from a variable object.
21726 For example, suppose @code{C} is a C@t{++} class, derived from class
21727 @code{Base}, and that the @code{Base} class has a member called
21728 @code{m_size}. Assume a variable @code{c} is has the type of
21729 @code{C} and a variable object @code{C} was created for variable
21730 @code{c}. Then, we'll get this output:
21732 (gdb) -var-info-path-expression C.Base.public.m_size
21733 ^done,path_expr=((Base)c).m_size)
21736 @subheading The @code{-var-show-attributes} Command
21737 @findex -var-show-attributes
21739 @subsubheading Synopsis
21742 -var-show-attributes @var{name}
21745 List attributes of the specified variable object @var{name}:
21748 status=@var{attr} [ ( ,@var{attr} )* ]
21752 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
21754 @subheading The @code{-var-evaluate-expression} Command
21755 @findex -var-evaluate-expression
21757 @subsubheading Synopsis
21760 -var-evaluate-expression [-f @var{format-spec}] @var{name}
21763 Evaluates the expression that is represented by the specified variable
21764 object and returns its value as a string. The format of the string
21765 can be specified with the @samp{-f} option. The possible values of
21766 this option are the same as for @code{-var-set-format}
21767 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
21768 the current display format will be used. The current display format
21769 can be changed using the @code{-var-set-format} command.
21775 Note that one must invoke @code{-var-list-children} for a variable
21776 before the value of a child variable can be evaluated.
21778 @subheading The @code{-var-assign} Command
21779 @findex -var-assign
21781 @subsubheading Synopsis
21784 -var-assign @var{name} @var{expression}
21787 Assigns the value of @var{expression} to the variable object specified
21788 by @var{name}. The object must be @samp{editable}. If the variable's
21789 value is altered by the assign, the variable will show up in any
21790 subsequent @code{-var-update} list.
21792 @subsubheading Example
21800 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
21804 @subheading The @code{-var-update} Command
21805 @findex -var-update
21807 @subsubheading Synopsis
21810 -var-update [@var{print-values}] @{@var{name} | "*"@}
21813 Reevaluate the expressions corresponding to the variable object
21814 @var{name} and all its direct and indirect children, and return the
21815 list of variable objects whose values have changed; @var{name} must
21816 be a root variable object. Here, ``changed'' means that the result of
21817 @code{-var-evaluate-expression} before and after the
21818 @code{-var-update} is different. If @samp{*} is used as the variable
21819 object names, all existing variable objects are updated, except
21820 for frozen ones (@pxref{-var-set-frozen}). The option
21821 @var{print-values} determines whether both names and values, or just
21822 names are printed. The possible values of this option are the same
21823 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
21824 recommended to use the @samp{--all-values} option, to reduce the
21825 number of MI commands needed on each program stop.
21827 With the @samp{*} parameter, if a variable object is bound to a
21828 currently running thread, it will not be updated, without any
21831 @subsubheading Example
21838 -var-update --all-values var1
21839 ^done,changelist=[@{name="var1",value="3",in_scope="true",
21840 type_changed="false"@}]
21844 @anchor{-var-update}
21845 The field in_scope may take three values:
21849 The variable object's current value is valid.
21852 The variable object does not currently hold a valid value but it may
21853 hold one in the future if its associated expression comes back into
21857 The variable object no longer holds a valid value.
21858 This can occur when the executable file being debugged has changed,
21859 either through recompilation or by using the @value{GDBN} @code{file}
21860 command. The front end should normally choose to delete these variable
21864 In the future new values may be added to this list so the front should
21865 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
21867 @subheading The @code{-var-set-frozen} Command
21868 @findex -var-set-frozen
21869 @anchor{-var-set-frozen}
21871 @subsubheading Synopsis
21874 -var-set-frozen @var{name} @var{flag}
21877 Set the frozenness flag on the variable object @var{name}. The
21878 @var{flag} parameter should be either @samp{1} to make the variable
21879 frozen or @samp{0} to make it unfrozen. If a variable object is
21880 frozen, then neither itself, nor any of its children, are
21881 implicitly updated by @code{-var-update} of
21882 a parent variable or by @code{-var-update *}. Only
21883 @code{-var-update} of the variable itself will update its value and
21884 values of its children. After a variable object is unfrozen, it is
21885 implicitly updated by all subsequent @code{-var-update} operations.
21886 Unfreezing a variable does not update it, only subsequent
21887 @code{-var-update} does.
21889 @subsubheading Example
21893 -var-set-frozen V 1
21899 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21900 @node GDB/MI Data Manipulation
21901 @section @sc{gdb/mi} Data Manipulation
21903 @cindex data manipulation, in @sc{gdb/mi}
21904 @cindex @sc{gdb/mi}, data manipulation
21905 This section describes the @sc{gdb/mi} commands that manipulate data:
21906 examine memory and registers, evaluate expressions, etc.
21908 @c REMOVED FROM THE INTERFACE.
21909 @c @subheading -data-assign
21910 @c Change the value of a program variable. Plenty of side effects.
21911 @c @subsubheading GDB Command
21913 @c @subsubheading Example
21916 @subheading The @code{-data-disassemble} Command
21917 @findex -data-disassemble
21919 @subsubheading Synopsis
21923 [ -s @var{start-addr} -e @var{end-addr} ]
21924 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
21932 @item @var{start-addr}
21933 is the beginning address (or @code{$pc})
21934 @item @var{end-addr}
21936 @item @var{filename}
21937 is the name of the file to disassemble
21938 @item @var{linenum}
21939 is the line number to disassemble around
21941 is the number of disassembly lines to be produced. If it is -1,
21942 the whole function will be disassembled, in case no @var{end-addr} is
21943 specified. If @var{end-addr} is specified as a non-zero value, and
21944 @var{lines} is lower than the number of disassembly lines between
21945 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
21946 displayed; if @var{lines} is higher than the number of lines between
21947 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
21950 is either 0 (meaning only disassembly) or 1 (meaning mixed source and
21954 @subsubheading Result
21956 The output for each instruction is composed of four fields:
21965 Note that whatever included in the instruction field, is not manipulated
21966 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
21968 @subsubheading @value{GDBN} Command
21970 There's no direct mapping from this command to the CLI.
21972 @subsubheading Example
21974 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
21978 -data-disassemble -s $pc -e "$pc + 20" -- 0
21981 @{address="0x000107c0",func-name="main",offset="4",
21982 inst="mov 2, %o0"@},
21983 @{address="0x000107c4",func-name="main",offset="8",
21984 inst="sethi %hi(0x11800), %o2"@},
21985 @{address="0x000107c8",func-name="main",offset="12",
21986 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
21987 @{address="0x000107cc",func-name="main",offset="16",
21988 inst="sethi %hi(0x11800), %o2"@},
21989 @{address="0x000107d0",func-name="main",offset="20",
21990 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
21994 Disassemble the whole @code{main} function. Line 32 is part of
21998 -data-disassemble -f basics.c -l 32 -- 0
22000 @{address="0x000107bc",func-name="main",offset="0",
22001 inst="save %sp, -112, %sp"@},
22002 @{address="0x000107c0",func-name="main",offset="4",
22003 inst="mov 2, %o0"@},
22004 @{address="0x000107c4",func-name="main",offset="8",
22005 inst="sethi %hi(0x11800), %o2"@},
22007 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
22008 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
22012 Disassemble 3 instructions from the start of @code{main}:
22016 -data-disassemble -f basics.c -l 32 -n 3 -- 0
22018 @{address="0x000107bc",func-name="main",offset="0",
22019 inst="save %sp, -112, %sp"@},
22020 @{address="0x000107c0",func-name="main",offset="4",
22021 inst="mov 2, %o0"@},
22022 @{address="0x000107c4",func-name="main",offset="8",
22023 inst="sethi %hi(0x11800), %o2"@}]
22027 Disassemble 3 instructions from the start of @code{main} in mixed mode:
22031 -data-disassemble -f basics.c -l 32 -n 3 -- 1
22033 src_and_asm_line=@{line="31",
22034 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
22035 testsuite/gdb.mi/basics.c",line_asm_insn=[
22036 @{address="0x000107bc",func-name="main",offset="0",
22037 inst="save %sp, -112, %sp"@}]@},
22038 src_and_asm_line=@{line="32",
22039 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
22040 testsuite/gdb.mi/basics.c",line_asm_insn=[
22041 @{address="0x000107c0",func-name="main",offset="4",
22042 inst="mov 2, %o0"@},
22043 @{address="0x000107c4",func-name="main",offset="8",
22044 inst="sethi %hi(0x11800), %o2"@}]@}]
22049 @subheading The @code{-data-evaluate-expression} Command
22050 @findex -data-evaluate-expression
22052 @subsubheading Synopsis
22055 -data-evaluate-expression @var{expr}
22058 Evaluate @var{expr} as an expression. The expression could contain an
22059 inferior function call. The function call will execute synchronously.
22060 If the expression contains spaces, it must be enclosed in double quotes.
22062 @subsubheading @value{GDBN} Command
22064 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
22065 @samp{call}. In @code{gdbtk} only, there's a corresponding
22066 @samp{gdb_eval} command.
22068 @subsubheading Example
22070 In the following example, the numbers that precede the commands are the
22071 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
22072 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
22076 211-data-evaluate-expression A
22079 311-data-evaluate-expression &A
22080 311^done,value="0xefffeb7c"
22082 411-data-evaluate-expression A+3
22085 511-data-evaluate-expression "A + 3"
22091 @subheading The @code{-data-list-changed-registers} Command
22092 @findex -data-list-changed-registers
22094 @subsubheading Synopsis
22097 -data-list-changed-registers
22100 Display a list of the registers that have changed.
22102 @subsubheading @value{GDBN} Command
22104 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
22105 has the corresponding command @samp{gdb_changed_register_list}.
22107 @subsubheading Example
22109 On a PPC MBX board:
22117 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
22118 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
22121 -data-list-changed-registers
22122 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
22123 "10","11","13","14","15","16","17","18","19","20","21","22","23",
22124 "24","25","26","27","28","30","31","64","65","66","67","69"]
22129 @subheading The @code{-data-list-register-names} Command
22130 @findex -data-list-register-names
22132 @subsubheading Synopsis
22135 -data-list-register-names [ ( @var{regno} )+ ]
22138 Show a list of register names for the current target. If no arguments
22139 are given, it shows a list of the names of all the registers. If
22140 integer numbers are given as arguments, it will print a list of the
22141 names of the registers corresponding to the arguments. To ensure
22142 consistency between a register name and its number, the output list may
22143 include empty register names.
22145 @subsubheading @value{GDBN} Command
22147 @value{GDBN} does not have a command which corresponds to
22148 @samp{-data-list-register-names}. In @code{gdbtk} there is a
22149 corresponding command @samp{gdb_regnames}.
22151 @subsubheading Example
22153 For the PPC MBX board:
22156 -data-list-register-names
22157 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
22158 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
22159 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
22160 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
22161 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
22162 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
22163 "", "pc","ps","cr","lr","ctr","xer"]
22165 -data-list-register-names 1 2 3
22166 ^done,register-names=["r1","r2","r3"]
22170 @subheading The @code{-data-list-register-values} Command
22171 @findex -data-list-register-values
22173 @subsubheading Synopsis
22176 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
22179 Display the registers' contents. @var{fmt} is the format according to
22180 which the registers' contents are to be returned, followed by an optional
22181 list of numbers specifying the registers to display. A missing list of
22182 numbers indicates that the contents of all the registers must be returned.
22184 Allowed formats for @var{fmt} are:
22201 @subsubheading @value{GDBN} Command
22203 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
22204 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
22206 @subsubheading Example
22208 For a PPC MBX board (note: line breaks are for readability only, they
22209 don't appear in the actual output):
22213 -data-list-register-values r 64 65
22214 ^done,register-values=[@{number="64",value="0xfe00a300"@},
22215 @{number="65",value="0x00029002"@}]
22217 -data-list-register-values x
22218 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
22219 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
22220 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
22221 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
22222 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
22223 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
22224 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
22225 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
22226 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
22227 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
22228 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
22229 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
22230 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
22231 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
22232 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
22233 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
22234 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
22235 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
22236 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
22237 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
22238 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
22239 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
22240 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
22241 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
22242 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
22243 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
22244 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
22245 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
22246 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
22247 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
22248 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
22249 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
22250 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
22251 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
22252 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
22253 @{number="69",value="0x20002b03"@}]
22258 @subheading The @code{-data-read-memory} Command
22259 @findex -data-read-memory
22261 @subsubheading Synopsis
22264 -data-read-memory [ -o @var{byte-offset} ]
22265 @var{address} @var{word-format} @var{word-size}
22266 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
22273 @item @var{address}
22274 An expression specifying the address of the first memory word to be
22275 read. Complex expressions containing embedded white space should be
22276 quoted using the C convention.
22278 @item @var{word-format}
22279 The format to be used to print the memory words. The notation is the
22280 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
22283 @item @var{word-size}
22284 The size of each memory word in bytes.
22286 @item @var{nr-rows}
22287 The number of rows in the output table.
22289 @item @var{nr-cols}
22290 The number of columns in the output table.
22293 If present, indicates that each row should include an @sc{ascii} dump. The
22294 value of @var{aschar} is used as a padding character when a byte is not a
22295 member of the printable @sc{ascii} character set (printable @sc{ascii}
22296 characters are those whose code is between 32 and 126, inclusively).
22298 @item @var{byte-offset}
22299 An offset to add to the @var{address} before fetching memory.
22302 This command displays memory contents as a table of @var{nr-rows} by
22303 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
22304 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
22305 (returned as @samp{total-bytes}). Should less than the requested number
22306 of bytes be returned by the target, the missing words are identified
22307 using @samp{N/A}. The number of bytes read from the target is returned
22308 in @samp{nr-bytes} and the starting address used to read memory in
22311 The address of the next/previous row or page is available in
22312 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
22315 @subsubheading @value{GDBN} Command
22317 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
22318 @samp{gdb_get_mem} memory read command.
22320 @subsubheading Example
22322 Read six bytes of memory starting at @code{bytes+6} but then offset by
22323 @code{-6} bytes. Format as three rows of two columns. One byte per
22324 word. Display each word in hex.
22328 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
22329 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
22330 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
22331 prev-page="0x0000138a",memory=[
22332 @{addr="0x00001390",data=["0x00","0x01"]@},
22333 @{addr="0x00001392",data=["0x02","0x03"]@},
22334 @{addr="0x00001394",data=["0x04","0x05"]@}]
22338 Read two bytes of memory starting at address @code{shorts + 64} and
22339 display as a single word formatted in decimal.
22343 5-data-read-memory shorts+64 d 2 1 1
22344 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
22345 next-row="0x00001512",prev-row="0x0000150e",
22346 next-page="0x00001512",prev-page="0x0000150e",memory=[
22347 @{addr="0x00001510",data=["128"]@}]
22351 Read thirty two bytes of memory starting at @code{bytes+16} and format
22352 as eight rows of four columns. Include a string encoding with @samp{x}
22353 used as the non-printable character.
22357 4-data-read-memory bytes+16 x 1 8 4 x
22358 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
22359 next-row="0x000013c0",prev-row="0x0000139c",
22360 next-page="0x000013c0",prev-page="0x00001380",memory=[
22361 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
22362 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
22363 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
22364 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
22365 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
22366 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
22367 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
22368 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
22372 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22373 @node GDB/MI Tracepoint Commands
22374 @section @sc{gdb/mi} Tracepoint Commands
22376 The tracepoint commands are not yet implemented.
22378 @c @subheading -trace-actions
22380 @c @subheading -trace-delete
22382 @c @subheading -trace-disable
22384 @c @subheading -trace-dump
22386 @c @subheading -trace-enable
22388 @c @subheading -trace-exists
22390 @c @subheading -trace-find
22392 @c @subheading -trace-frame-number
22394 @c @subheading -trace-info
22396 @c @subheading -trace-insert
22398 @c @subheading -trace-list
22400 @c @subheading -trace-pass-count
22402 @c @subheading -trace-save
22404 @c @subheading -trace-start
22406 @c @subheading -trace-stop
22409 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22410 @node GDB/MI Symbol Query
22411 @section @sc{gdb/mi} Symbol Query Commands
22414 @subheading The @code{-symbol-info-address} Command
22415 @findex -symbol-info-address
22417 @subsubheading Synopsis
22420 -symbol-info-address @var{symbol}
22423 Describe where @var{symbol} is stored.
22425 @subsubheading @value{GDBN} Command
22427 The corresponding @value{GDBN} command is @samp{info address}.
22429 @subsubheading Example
22433 @subheading The @code{-symbol-info-file} Command
22434 @findex -symbol-info-file
22436 @subsubheading Synopsis
22442 Show the file for the symbol.
22444 @subsubheading @value{GDBN} Command
22446 There's no equivalent @value{GDBN} command. @code{gdbtk} has
22447 @samp{gdb_find_file}.
22449 @subsubheading Example
22453 @subheading The @code{-symbol-info-function} Command
22454 @findex -symbol-info-function
22456 @subsubheading Synopsis
22459 -symbol-info-function
22462 Show which function the symbol lives in.
22464 @subsubheading @value{GDBN} Command
22466 @samp{gdb_get_function} in @code{gdbtk}.
22468 @subsubheading Example
22472 @subheading The @code{-symbol-info-line} Command
22473 @findex -symbol-info-line
22475 @subsubheading Synopsis
22481 Show the core addresses of the code for a source line.
22483 @subsubheading @value{GDBN} Command
22485 The corresponding @value{GDBN} command is @samp{info line}.
22486 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
22488 @subsubheading Example
22492 @subheading The @code{-symbol-info-symbol} Command
22493 @findex -symbol-info-symbol
22495 @subsubheading Synopsis
22498 -symbol-info-symbol @var{addr}
22501 Describe what symbol is at location @var{addr}.
22503 @subsubheading @value{GDBN} Command
22505 The corresponding @value{GDBN} command is @samp{info symbol}.
22507 @subsubheading Example
22511 @subheading The @code{-symbol-list-functions} Command
22512 @findex -symbol-list-functions
22514 @subsubheading Synopsis
22517 -symbol-list-functions
22520 List the functions in the executable.
22522 @subsubheading @value{GDBN} Command
22524 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
22525 @samp{gdb_search} in @code{gdbtk}.
22527 @subsubheading Example
22531 @subheading The @code{-symbol-list-lines} Command
22532 @findex -symbol-list-lines
22534 @subsubheading Synopsis
22537 -symbol-list-lines @var{filename}
22540 Print the list of lines that contain code and their associated program
22541 addresses for the given source filename. The entries are sorted in
22542 ascending PC order.
22544 @subsubheading @value{GDBN} Command
22546 There is no corresponding @value{GDBN} command.
22548 @subsubheading Example
22551 -symbol-list-lines basics.c
22552 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
22557 @subheading The @code{-symbol-list-types} Command
22558 @findex -symbol-list-types
22560 @subsubheading Synopsis
22566 List all the type names.
22568 @subsubheading @value{GDBN} Command
22570 The corresponding commands are @samp{info types} in @value{GDBN},
22571 @samp{gdb_search} in @code{gdbtk}.
22573 @subsubheading Example
22577 @subheading The @code{-symbol-list-variables} Command
22578 @findex -symbol-list-variables
22580 @subsubheading Synopsis
22583 -symbol-list-variables
22586 List all the global and static variable names.
22588 @subsubheading @value{GDBN} Command
22590 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
22592 @subsubheading Example
22596 @subheading The @code{-symbol-locate} Command
22597 @findex -symbol-locate
22599 @subsubheading Synopsis
22605 @subsubheading @value{GDBN} Command
22607 @samp{gdb_loc} in @code{gdbtk}.
22609 @subsubheading Example
22613 @subheading The @code{-symbol-type} Command
22614 @findex -symbol-type
22616 @subsubheading Synopsis
22619 -symbol-type @var{variable}
22622 Show type of @var{variable}.
22624 @subsubheading @value{GDBN} Command
22626 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
22627 @samp{gdb_obj_variable}.
22629 @subsubheading Example
22633 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22634 @node GDB/MI File Commands
22635 @section @sc{gdb/mi} File Commands
22637 This section describes the GDB/MI commands to specify executable file names
22638 and to read in and obtain symbol table information.
22640 @subheading The @code{-file-exec-and-symbols} Command
22641 @findex -file-exec-and-symbols
22643 @subsubheading Synopsis
22646 -file-exec-and-symbols @var{file}
22649 Specify the executable file to be debugged. This file is the one from
22650 which the symbol table is also read. If no file is specified, the
22651 command clears the executable and symbol information. If breakpoints
22652 are set when using this command with no arguments, @value{GDBN} will produce
22653 error messages. Otherwise, no output is produced, except a completion
22656 @subsubheading @value{GDBN} Command
22658 The corresponding @value{GDBN} command is @samp{file}.
22660 @subsubheading Example
22664 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
22670 @subheading The @code{-file-exec-file} Command
22671 @findex -file-exec-file
22673 @subsubheading Synopsis
22676 -file-exec-file @var{file}
22679 Specify the executable file to be debugged. Unlike
22680 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
22681 from this file. If used without argument, @value{GDBN} clears the information
22682 about the executable file. No output is produced, except a completion
22685 @subsubheading @value{GDBN} Command
22687 The corresponding @value{GDBN} command is @samp{exec-file}.
22689 @subsubheading Example
22693 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
22699 @subheading The @code{-file-list-exec-sections} Command
22700 @findex -file-list-exec-sections
22702 @subsubheading Synopsis
22705 -file-list-exec-sections
22708 List the sections of the current executable file.
22710 @subsubheading @value{GDBN} Command
22712 The @value{GDBN} command @samp{info file} shows, among the rest, the same
22713 information as this command. @code{gdbtk} has a corresponding command
22714 @samp{gdb_load_info}.
22716 @subsubheading Example
22720 @subheading The @code{-file-list-exec-source-file} Command
22721 @findex -file-list-exec-source-file
22723 @subsubheading Synopsis
22726 -file-list-exec-source-file
22729 List the line number, the current source file, and the absolute path
22730 to the current source file for the current executable. The macro
22731 information field has a value of @samp{1} or @samp{0} depending on
22732 whether or not the file includes preprocessor macro information.
22734 @subsubheading @value{GDBN} Command
22736 The @value{GDBN} equivalent is @samp{info source}
22738 @subsubheading Example
22742 123-file-list-exec-source-file
22743 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
22748 @subheading The @code{-file-list-exec-source-files} Command
22749 @findex -file-list-exec-source-files
22751 @subsubheading Synopsis
22754 -file-list-exec-source-files
22757 List the source files for the current executable.
22759 It will always output the filename, but only when @value{GDBN} can find
22760 the absolute file name of a source file, will it output the fullname.
22762 @subsubheading @value{GDBN} Command
22764 The @value{GDBN} equivalent is @samp{info sources}.
22765 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
22767 @subsubheading Example
22770 -file-list-exec-source-files
22772 @{file=foo.c,fullname=/home/foo.c@},
22773 @{file=/home/bar.c,fullname=/home/bar.c@},
22774 @{file=gdb_could_not_find_fullpath.c@}]
22778 @subheading The @code{-file-list-shared-libraries} Command
22779 @findex -file-list-shared-libraries
22781 @subsubheading Synopsis
22784 -file-list-shared-libraries
22787 List the shared libraries in the program.
22789 @subsubheading @value{GDBN} Command
22791 The corresponding @value{GDBN} command is @samp{info shared}.
22793 @subsubheading Example
22797 @subheading The @code{-file-list-symbol-files} Command
22798 @findex -file-list-symbol-files
22800 @subsubheading Synopsis
22803 -file-list-symbol-files
22808 @subsubheading @value{GDBN} Command
22810 The corresponding @value{GDBN} command is @samp{info file} (part of it).
22812 @subsubheading Example
22816 @subheading The @code{-file-symbol-file} Command
22817 @findex -file-symbol-file
22819 @subsubheading Synopsis
22822 -file-symbol-file @var{file}
22825 Read symbol table info from the specified @var{file} argument. When
22826 used without arguments, clears @value{GDBN}'s symbol table info. No output is
22827 produced, except for a completion notification.
22829 @subsubheading @value{GDBN} Command
22831 The corresponding @value{GDBN} command is @samp{symbol-file}.
22833 @subsubheading Example
22837 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
22843 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22844 @node GDB/MI Memory Overlay Commands
22845 @section @sc{gdb/mi} Memory Overlay Commands
22847 The memory overlay commands are not implemented.
22849 @c @subheading -overlay-auto
22851 @c @subheading -overlay-list-mapping-state
22853 @c @subheading -overlay-list-overlays
22855 @c @subheading -overlay-map
22857 @c @subheading -overlay-off
22859 @c @subheading -overlay-on
22861 @c @subheading -overlay-unmap
22863 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22864 @node GDB/MI Signal Handling Commands
22865 @section @sc{gdb/mi} Signal Handling Commands
22867 Signal handling commands are not implemented.
22869 @c @subheading -signal-handle
22871 @c @subheading -signal-list-handle-actions
22873 @c @subheading -signal-list-signal-types
22877 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22878 @node GDB/MI Target Manipulation
22879 @section @sc{gdb/mi} Target Manipulation Commands
22882 @subheading The @code{-target-attach} Command
22883 @findex -target-attach
22885 @subsubheading Synopsis
22888 -target-attach @var{pid} | @var{gid} | @var{file}
22891 Attach to a process @var{pid} or a file @var{file} outside of
22892 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
22893 group, the id previously returned by
22894 @samp{-list-thread-groups --available} must be used.
22896 @subsubheading @value{GDBN} Command
22898 The corresponding @value{GDBN} command is @samp{attach}.
22900 @subsubheading Example
22904 =thread-created,id="1"
22905 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
22910 @subheading The @code{-target-compare-sections} Command
22911 @findex -target-compare-sections
22913 @subsubheading Synopsis
22916 -target-compare-sections [ @var{section} ]
22919 Compare data of section @var{section} on target to the exec file.
22920 Without the argument, all sections are compared.
22922 @subsubheading @value{GDBN} Command
22924 The @value{GDBN} equivalent is @samp{compare-sections}.
22926 @subsubheading Example
22930 @subheading The @code{-target-detach} Command
22931 @findex -target-detach
22933 @subsubheading Synopsis
22936 -target-detach [ @var{pid} | @var{gid} ]
22939 Detach from the remote target which normally resumes its execution.
22940 If either @var{pid} or @var{gid} is specified, detaches from either
22941 the specified process, or specified thread group. There's no output.
22943 @subsubheading @value{GDBN} Command
22945 The corresponding @value{GDBN} command is @samp{detach}.
22947 @subsubheading Example
22957 @subheading The @code{-target-disconnect} Command
22958 @findex -target-disconnect
22960 @subsubheading Synopsis
22966 Disconnect from the remote target. There's no output and the target is
22967 generally not resumed.
22969 @subsubheading @value{GDBN} Command
22971 The corresponding @value{GDBN} command is @samp{disconnect}.
22973 @subsubheading Example
22983 @subheading The @code{-target-download} Command
22984 @findex -target-download
22986 @subsubheading Synopsis
22992 Loads the executable onto the remote target.
22993 It prints out an update message every half second, which includes the fields:
22997 The name of the section.
22999 The size of what has been sent so far for that section.
23001 The size of the section.
23003 The total size of what was sent so far (the current and the previous sections).
23005 The size of the overall executable to download.
23009 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
23010 @sc{gdb/mi} Output Syntax}).
23012 In addition, it prints the name and size of the sections, as they are
23013 downloaded. These messages include the following fields:
23017 The name of the section.
23019 The size of the section.
23021 The size of the overall executable to download.
23025 At the end, a summary is printed.
23027 @subsubheading @value{GDBN} Command
23029 The corresponding @value{GDBN} command is @samp{load}.
23031 @subsubheading Example
23033 Note: each status message appears on a single line. Here the messages
23034 have been broken down so that they can fit onto a page.
23039 +download,@{section=".text",section-size="6668",total-size="9880"@}
23040 +download,@{section=".text",section-sent="512",section-size="6668",
23041 total-sent="512",total-size="9880"@}
23042 +download,@{section=".text",section-sent="1024",section-size="6668",
23043 total-sent="1024",total-size="9880"@}
23044 +download,@{section=".text",section-sent="1536",section-size="6668",
23045 total-sent="1536",total-size="9880"@}
23046 +download,@{section=".text",section-sent="2048",section-size="6668",
23047 total-sent="2048",total-size="9880"@}
23048 +download,@{section=".text",section-sent="2560",section-size="6668",
23049 total-sent="2560",total-size="9880"@}
23050 +download,@{section=".text",section-sent="3072",section-size="6668",
23051 total-sent="3072",total-size="9880"@}
23052 +download,@{section=".text",section-sent="3584",section-size="6668",
23053 total-sent="3584",total-size="9880"@}
23054 +download,@{section=".text",section-sent="4096",section-size="6668",
23055 total-sent="4096",total-size="9880"@}
23056 +download,@{section=".text",section-sent="4608",section-size="6668",
23057 total-sent="4608",total-size="9880"@}
23058 +download,@{section=".text",section-sent="5120",section-size="6668",
23059 total-sent="5120",total-size="9880"@}
23060 +download,@{section=".text",section-sent="5632",section-size="6668",
23061 total-sent="5632",total-size="9880"@}
23062 +download,@{section=".text",section-sent="6144",section-size="6668",
23063 total-sent="6144",total-size="9880"@}
23064 +download,@{section=".text",section-sent="6656",section-size="6668",
23065 total-sent="6656",total-size="9880"@}
23066 +download,@{section=".init",section-size="28",total-size="9880"@}
23067 +download,@{section=".fini",section-size="28",total-size="9880"@}
23068 +download,@{section=".data",section-size="3156",total-size="9880"@}
23069 +download,@{section=".data",section-sent="512",section-size="3156",
23070 total-sent="7236",total-size="9880"@}
23071 +download,@{section=".data",section-sent="1024",section-size="3156",
23072 total-sent="7748",total-size="9880"@}
23073 +download,@{section=".data",section-sent="1536",section-size="3156",
23074 total-sent="8260",total-size="9880"@}
23075 +download,@{section=".data",section-sent="2048",section-size="3156",
23076 total-sent="8772",total-size="9880"@}
23077 +download,@{section=".data",section-sent="2560",section-size="3156",
23078 total-sent="9284",total-size="9880"@}
23079 +download,@{section=".data",section-sent="3072",section-size="3156",
23080 total-sent="9796",total-size="9880"@}
23081 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
23087 @subheading The @code{-target-exec-status} Command
23088 @findex -target-exec-status
23090 @subsubheading Synopsis
23093 -target-exec-status
23096 Provide information on the state of the target (whether it is running or
23097 not, for instance).
23099 @subsubheading @value{GDBN} Command
23101 There's no equivalent @value{GDBN} command.
23103 @subsubheading Example
23107 @subheading The @code{-target-list-available-targets} Command
23108 @findex -target-list-available-targets
23110 @subsubheading Synopsis
23113 -target-list-available-targets
23116 List the possible targets to connect to.
23118 @subsubheading @value{GDBN} Command
23120 The corresponding @value{GDBN} command is @samp{help target}.
23122 @subsubheading Example
23126 @subheading The @code{-target-list-current-targets} Command
23127 @findex -target-list-current-targets
23129 @subsubheading Synopsis
23132 -target-list-current-targets
23135 Describe the current target.
23137 @subsubheading @value{GDBN} Command
23139 The corresponding information is printed by @samp{info file} (among
23142 @subsubheading Example
23146 @subheading The @code{-target-list-parameters} Command
23147 @findex -target-list-parameters
23149 @subsubheading Synopsis
23152 -target-list-parameters
23157 @subsubheading @value{GDBN} Command
23161 @subsubheading Example
23165 @subheading The @code{-target-select} Command
23166 @findex -target-select
23168 @subsubheading Synopsis
23171 -target-select @var{type} @var{parameters @dots{}}
23174 Connect @value{GDBN} to the remote target. This command takes two args:
23178 The type of target, for instance @samp{remote}, etc.
23179 @item @var{parameters}
23180 Device names, host names and the like. @xref{Target Commands, ,
23181 Commands for Managing Targets}, for more details.
23184 The output is a connection notification, followed by the address at
23185 which the target program is, in the following form:
23188 ^connected,addr="@var{address}",func="@var{function name}",
23189 args=[@var{arg list}]
23192 @subsubheading @value{GDBN} Command
23194 The corresponding @value{GDBN} command is @samp{target}.
23196 @subsubheading Example
23200 -target-select remote /dev/ttya
23201 ^connected,addr="0xfe00a300",func="??",args=[]
23205 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23206 @node GDB/MI File Transfer Commands
23207 @section @sc{gdb/mi} File Transfer Commands
23210 @subheading The @code{-target-file-put} Command
23211 @findex -target-file-put
23213 @subsubheading Synopsis
23216 -target-file-put @var{hostfile} @var{targetfile}
23219 Copy file @var{hostfile} from the host system (the machine running
23220 @value{GDBN}) to @var{targetfile} on the target system.
23222 @subsubheading @value{GDBN} Command
23224 The corresponding @value{GDBN} command is @samp{remote put}.
23226 @subsubheading Example
23230 -target-file-put localfile remotefile
23236 @subheading The @code{-target-file-get} Command
23237 @findex -target-file-get
23239 @subsubheading Synopsis
23242 -target-file-get @var{targetfile} @var{hostfile}
23245 Copy file @var{targetfile} from the target system to @var{hostfile}
23246 on the host system.
23248 @subsubheading @value{GDBN} Command
23250 The corresponding @value{GDBN} command is @samp{remote get}.
23252 @subsubheading Example
23256 -target-file-get remotefile localfile
23262 @subheading The @code{-target-file-delete} Command
23263 @findex -target-file-delete
23265 @subsubheading Synopsis
23268 -target-file-delete @var{targetfile}
23271 Delete @var{targetfile} from the target system.
23273 @subsubheading @value{GDBN} Command
23275 The corresponding @value{GDBN} command is @samp{remote delete}.
23277 @subsubheading Example
23281 -target-file-delete remotefile
23287 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23288 @node GDB/MI Miscellaneous Commands
23289 @section Miscellaneous @sc{gdb/mi} Commands
23291 @c @subheading -gdb-complete
23293 @subheading The @code{-gdb-exit} Command
23296 @subsubheading Synopsis
23302 Exit @value{GDBN} immediately.
23304 @subsubheading @value{GDBN} Command
23306 Approximately corresponds to @samp{quit}.
23308 @subsubheading Example
23317 @subheading The @code{-exec-abort} Command
23318 @findex -exec-abort
23320 @subsubheading Synopsis
23326 Kill the inferior running program.
23328 @subsubheading @value{GDBN} Command
23330 The corresponding @value{GDBN} command is @samp{kill}.
23332 @subsubheading Example
23336 @subheading The @code{-gdb-set} Command
23339 @subsubheading Synopsis
23345 Set an internal @value{GDBN} variable.
23346 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
23348 @subsubheading @value{GDBN} Command
23350 The corresponding @value{GDBN} command is @samp{set}.
23352 @subsubheading Example
23362 @subheading The @code{-gdb-show} Command
23365 @subsubheading Synopsis
23371 Show the current value of a @value{GDBN} variable.
23373 @subsubheading @value{GDBN} Command
23375 The corresponding @value{GDBN} command is @samp{show}.
23377 @subsubheading Example
23386 @c @subheading -gdb-source
23389 @subheading The @code{-gdb-version} Command
23390 @findex -gdb-version
23392 @subsubheading Synopsis
23398 Show version information for @value{GDBN}. Used mostly in testing.
23400 @subsubheading @value{GDBN} Command
23402 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
23403 default shows this information when you start an interactive session.
23405 @subsubheading Example
23407 @c This example modifies the actual output from GDB to avoid overfull
23413 ~Copyright 2000 Free Software Foundation, Inc.
23414 ~GDB is free software, covered by the GNU General Public License, and
23415 ~you are welcome to change it and/or distribute copies of it under
23416 ~ certain conditions.
23417 ~Type "show copying" to see the conditions.
23418 ~There is absolutely no warranty for GDB. Type "show warranty" for
23420 ~This GDB was configured as
23421 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
23426 @subheading The @code{-list-features} Command
23427 @findex -list-features
23429 Returns a list of particular features of the MI protocol that
23430 this version of gdb implements. A feature can be a command,
23431 or a new field in an output of some command, or even an
23432 important bugfix. While a frontend can sometimes detect presence
23433 of a feature at runtime, it is easier to perform detection at debugger
23436 The command returns a list of strings, with each string naming an
23437 available feature. Each returned string is just a name, it does not
23438 have any internal structure. The list of possible feature names
23444 (gdb) -list-features
23445 ^done,result=["feature1","feature2"]
23448 The current list of features is:
23451 @item frozen-varobjs
23452 Indicates presence of the @code{-var-set-frozen} command, as well
23453 as possible presense of the @code{frozen} field in the output
23454 of @code{-varobj-create}.
23455 @item pending-breakpoints
23456 Indicates presence of the @option{-f} option to the @code{-break-insert} command.
23458 Indicates presence of the @code{-thread-info} command.
23462 @subheading The @code{-list-target-features} Command
23463 @findex -list-target-features
23465 Returns a list of particular features that are supported by the
23466 target. Those features affect the permitted MI commands, but
23467 unlike the features reported by the @code{-list-features} command, the
23468 features depend on which target GDB is using at the moment. Whenever
23469 a target can change, due to commands such as @code{-target-select},
23470 @code{-target-attach} or @code{-exec-run}, the list of target features
23471 may change, and the frontend should obtain it again.
23475 (gdb) -list-features
23476 ^done,result=["async"]
23479 The current list of features is:
23483 Indicates that the target is capable of asynchronous command
23484 execution, which means that @value{GDBN} will accept further commands
23485 while the target is running.
23489 @subheading The @code{-list-thread-groups} Command
23490 @findex -list-thread-groups
23492 @subheading Synopsis
23495 -list-thread-groups [ --available ] [ @var{group} ]
23498 When used without the @var{group} parameter, lists top-level thread
23499 groups that are being debugged. When used with the @var{group}
23500 parameter, the children of the specified group are listed. The
23501 children can be either threads, or other groups. At present,
23502 @value{GDBN} will not report both threads and groups as children at
23503 the same time, but it may change in future.
23505 With the @samp{--available} option, instead of reporting groups that
23506 are been debugged, GDB will report all thread groups available on the
23507 target. Using the @samp{--available} option together with @var{group}
23510 @subheading Example
23514 -list-thread-groups
23515 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
23516 -list-thread-groups 17
23517 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
23518 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
23519 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
23520 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
23521 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
23524 @subheading The @code{-interpreter-exec} Command
23525 @findex -interpreter-exec
23527 @subheading Synopsis
23530 -interpreter-exec @var{interpreter} @var{command}
23532 @anchor{-interpreter-exec}
23534 Execute the specified @var{command} in the given @var{interpreter}.
23536 @subheading @value{GDBN} Command
23538 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
23540 @subheading Example
23544 -interpreter-exec console "break main"
23545 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
23546 &"During symbol reading, bad structure-type format.\n"
23547 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
23552 @subheading The @code{-inferior-tty-set} Command
23553 @findex -inferior-tty-set
23555 @subheading Synopsis
23558 -inferior-tty-set /dev/pts/1
23561 Set terminal for future runs of the program being debugged.
23563 @subheading @value{GDBN} Command
23565 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
23567 @subheading Example
23571 -inferior-tty-set /dev/pts/1
23576 @subheading The @code{-inferior-tty-show} Command
23577 @findex -inferior-tty-show
23579 @subheading Synopsis
23585 Show terminal for future runs of program being debugged.
23587 @subheading @value{GDBN} Command
23589 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
23591 @subheading Example
23595 -inferior-tty-set /dev/pts/1
23599 ^done,inferior_tty_terminal="/dev/pts/1"
23603 @subheading The @code{-enable-timings} Command
23604 @findex -enable-timings
23606 @subheading Synopsis
23609 -enable-timings [yes | no]
23612 Toggle the printing of the wallclock, user and system times for an MI
23613 command as a field in its output. This command is to help frontend
23614 developers optimize the performance of their code. No argument is
23615 equivalent to @samp{yes}.
23617 @subheading @value{GDBN} Command
23621 @subheading Example
23629 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
23630 addr="0x080484ed",func="main",file="myprog.c",
23631 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
23632 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
23640 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
23641 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
23642 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
23643 fullname="/home/nickrob/myprog.c",line="73"@}
23648 @chapter @value{GDBN} Annotations
23650 This chapter describes annotations in @value{GDBN}. Annotations were
23651 designed to interface @value{GDBN} to graphical user interfaces or other
23652 similar programs which want to interact with @value{GDBN} at a
23653 relatively high level.
23655 The annotation mechanism has largely been superseded by @sc{gdb/mi}
23659 This is Edition @value{EDITION}, @value{DATE}.
23663 * Annotations Overview:: What annotations are; the general syntax.
23664 * Server Prefix:: Issuing a command without affecting user state.
23665 * Prompting:: Annotations marking @value{GDBN}'s need for input.
23666 * Errors:: Annotations for error messages.
23667 * Invalidation:: Some annotations describe things now invalid.
23668 * Annotations for Running::
23669 Whether the program is running, how it stopped, etc.
23670 * Source Annotations:: Annotations describing source code.
23673 @node Annotations Overview
23674 @section What is an Annotation?
23675 @cindex annotations
23677 Annotations start with a newline character, two @samp{control-z}
23678 characters, and the name of the annotation. If there is no additional
23679 information associated with this annotation, the name of the annotation
23680 is followed immediately by a newline. If there is additional
23681 information, the name of the annotation is followed by a space, the
23682 additional information, and a newline. The additional information
23683 cannot contain newline characters.
23685 Any output not beginning with a newline and two @samp{control-z}
23686 characters denotes literal output from @value{GDBN}. Currently there is
23687 no need for @value{GDBN} to output a newline followed by two
23688 @samp{control-z} characters, but if there was such a need, the
23689 annotations could be extended with an @samp{escape} annotation which
23690 means those three characters as output.
23692 The annotation @var{level}, which is specified using the
23693 @option{--annotate} command line option (@pxref{Mode Options}), controls
23694 how much information @value{GDBN} prints together with its prompt,
23695 values of expressions, source lines, and other types of output. Level 0
23696 is for no annotations, level 1 is for use when @value{GDBN} is run as a
23697 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
23698 for programs that control @value{GDBN}, and level 2 annotations have
23699 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
23700 Interface, annotate, GDB's Obsolete Annotations}).
23703 @kindex set annotate
23704 @item set annotate @var{level}
23705 The @value{GDBN} command @code{set annotate} sets the level of
23706 annotations to the specified @var{level}.
23708 @item show annotate
23709 @kindex show annotate
23710 Show the current annotation level.
23713 This chapter describes level 3 annotations.
23715 A simple example of starting up @value{GDBN} with annotations is:
23718 $ @kbd{gdb --annotate=3}
23720 Copyright 2003 Free Software Foundation, Inc.
23721 GDB is free software, covered by the GNU General Public License,
23722 and you are welcome to change it and/or distribute copies of it
23723 under certain conditions.
23724 Type "show copying" to see the conditions.
23725 There is absolutely no warranty for GDB. Type "show warranty"
23727 This GDB was configured as "i386-pc-linux-gnu"
23738 Here @samp{quit} is input to @value{GDBN}; the rest is output from
23739 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
23740 denotes a @samp{control-z} character) are annotations; the rest is
23741 output from @value{GDBN}.
23743 @node Server Prefix
23744 @section The Server Prefix
23745 @cindex server prefix
23747 If you prefix a command with @samp{server } then it will not affect
23748 the command history, nor will it affect @value{GDBN}'s notion of which
23749 command to repeat if @key{RET} is pressed on a line by itself. This
23750 means that commands can be run behind a user's back by a front-end in
23751 a transparent manner.
23753 The server prefix does not affect the recording of values into the value
23754 history; to print a value without recording it into the value history,
23755 use the @code{output} command instead of the @code{print} command.
23758 @section Annotation for @value{GDBN} Input
23760 @cindex annotations for prompts
23761 When @value{GDBN} prompts for input, it annotates this fact so it is possible
23762 to know when to send output, when the output from a given command is
23765 Different kinds of input each have a different @dfn{input type}. Each
23766 input type has three annotations: a @code{pre-} annotation, which
23767 denotes the beginning of any prompt which is being output, a plain
23768 annotation, which denotes the end of the prompt, and then a @code{post-}
23769 annotation which denotes the end of any echo which may (or may not) be
23770 associated with the input. For example, the @code{prompt} input type
23771 features the following annotations:
23779 The input types are
23782 @findex pre-prompt annotation
23783 @findex prompt annotation
23784 @findex post-prompt annotation
23786 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
23788 @findex pre-commands annotation
23789 @findex commands annotation
23790 @findex post-commands annotation
23792 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
23793 command. The annotations are repeated for each command which is input.
23795 @findex pre-overload-choice annotation
23796 @findex overload-choice annotation
23797 @findex post-overload-choice annotation
23798 @item overload-choice
23799 When @value{GDBN} wants the user to select between various overloaded functions.
23801 @findex pre-query annotation
23802 @findex query annotation
23803 @findex post-query annotation
23805 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
23807 @findex pre-prompt-for-continue annotation
23808 @findex prompt-for-continue annotation
23809 @findex post-prompt-for-continue annotation
23810 @item prompt-for-continue
23811 When @value{GDBN} is asking the user to press return to continue. Note: Don't
23812 expect this to work well; instead use @code{set height 0} to disable
23813 prompting. This is because the counting of lines is buggy in the
23814 presence of annotations.
23819 @cindex annotations for errors, warnings and interrupts
23821 @findex quit annotation
23826 This annotation occurs right before @value{GDBN} responds to an interrupt.
23828 @findex error annotation
23833 This annotation occurs right before @value{GDBN} responds to an error.
23835 Quit and error annotations indicate that any annotations which @value{GDBN} was
23836 in the middle of may end abruptly. For example, if a
23837 @code{value-history-begin} annotation is followed by a @code{error}, one
23838 cannot expect to receive the matching @code{value-history-end}. One
23839 cannot expect not to receive it either, however; an error annotation
23840 does not necessarily mean that @value{GDBN} is immediately returning all the way
23843 @findex error-begin annotation
23844 A quit or error annotation may be preceded by
23850 Any output between that and the quit or error annotation is the error
23853 Warning messages are not yet annotated.
23854 @c If we want to change that, need to fix warning(), type_error(),
23855 @c range_error(), and possibly other places.
23858 @section Invalidation Notices
23860 @cindex annotations for invalidation messages
23861 The following annotations say that certain pieces of state may have
23865 @findex frames-invalid annotation
23866 @item ^Z^Zframes-invalid
23868 The frames (for example, output from the @code{backtrace} command) may
23871 @findex breakpoints-invalid annotation
23872 @item ^Z^Zbreakpoints-invalid
23874 The breakpoints may have changed. For example, the user just added or
23875 deleted a breakpoint.
23878 @node Annotations for Running
23879 @section Running the Program
23880 @cindex annotations for running programs
23882 @findex starting annotation
23883 @findex stopping annotation
23884 When the program starts executing due to a @value{GDBN} command such as
23885 @code{step} or @code{continue},
23891 is output. When the program stops,
23897 is output. Before the @code{stopped} annotation, a variety of
23898 annotations describe how the program stopped.
23901 @findex exited annotation
23902 @item ^Z^Zexited @var{exit-status}
23903 The program exited, and @var{exit-status} is the exit status (zero for
23904 successful exit, otherwise nonzero).
23906 @findex signalled annotation
23907 @findex signal-name annotation
23908 @findex signal-name-end annotation
23909 @findex signal-string annotation
23910 @findex signal-string-end annotation
23911 @item ^Z^Zsignalled
23912 The program exited with a signal. After the @code{^Z^Zsignalled}, the
23913 annotation continues:
23919 ^Z^Zsignal-name-end
23923 ^Z^Zsignal-string-end
23928 where @var{name} is the name of the signal, such as @code{SIGILL} or
23929 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
23930 as @code{Illegal Instruction} or @code{Segmentation fault}.
23931 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
23932 user's benefit and have no particular format.
23934 @findex signal annotation
23936 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
23937 just saying that the program received the signal, not that it was
23938 terminated with it.
23940 @findex breakpoint annotation
23941 @item ^Z^Zbreakpoint @var{number}
23942 The program hit breakpoint number @var{number}.
23944 @findex watchpoint annotation
23945 @item ^Z^Zwatchpoint @var{number}
23946 The program hit watchpoint number @var{number}.
23949 @node Source Annotations
23950 @section Displaying Source
23951 @cindex annotations for source display
23953 @findex source annotation
23954 The following annotation is used instead of displaying source code:
23957 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
23960 where @var{filename} is an absolute file name indicating which source
23961 file, @var{line} is the line number within that file (where 1 is the
23962 first line in the file), @var{character} is the character position
23963 within the file (where 0 is the first character in the file) (for most
23964 debug formats this will necessarily point to the beginning of a line),
23965 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
23966 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
23967 @var{addr} is the address in the target program associated with the
23968 source which is being displayed. @var{addr} is in the form @samp{0x}
23969 followed by one or more lowercase hex digits (note that this does not
23970 depend on the language).
23973 @chapter Reporting Bugs in @value{GDBN}
23974 @cindex bugs in @value{GDBN}
23975 @cindex reporting bugs in @value{GDBN}
23977 Your bug reports play an essential role in making @value{GDBN} reliable.
23979 Reporting a bug may help you by bringing a solution to your problem, or it
23980 may not. But in any case the principal function of a bug report is to help
23981 the entire community by making the next version of @value{GDBN} work better. Bug
23982 reports are your contribution to the maintenance of @value{GDBN}.
23984 In order for a bug report to serve its purpose, you must include the
23985 information that enables us to fix the bug.
23988 * Bug Criteria:: Have you found a bug?
23989 * Bug Reporting:: How to report bugs
23993 @section Have You Found a Bug?
23994 @cindex bug criteria
23996 If you are not sure whether you have found a bug, here are some guidelines:
23999 @cindex fatal signal
24000 @cindex debugger crash
24001 @cindex crash of debugger
24003 If the debugger gets a fatal signal, for any input whatever, that is a
24004 @value{GDBN} bug. Reliable debuggers never crash.
24006 @cindex error on valid input
24008 If @value{GDBN} produces an error message for valid input, that is a
24009 bug. (Note that if you're cross debugging, the problem may also be
24010 somewhere in the connection to the target.)
24012 @cindex invalid input
24014 If @value{GDBN} does not produce an error message for invalid input,
24015 that is a bug. However, you should note that your idea of
24016 ``invalid input'' might be our idea of ``an extension'' or ``support
24017 for traditional practice''.
24020 If you are an experienced user of debugging tools, your suggestions
24021 for improvement of @value{GDBN} are welcome in any case.
24024 @node Bug Reporting
24025 @section How to Report Bugs
24026 @cindex bug reports
24027 @cindex @value{GDBN} bugs, reporting
24029 A number of companies and individuals offer support for @sc{gnu} products.
24030 If you obtained @value{GDBN} from a support organization, we recommend you
24031 contact that organization first.
24033 You can find contact information for many support companies and
24034 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
24036 @c should add a web page ref...
24039 @ifset BUGURL_DEFAULT
24040 In any event, we also recommend that you submit bug reports for
24041 @value{GDBN}. The preferred method is to submit them directly using
24042 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
24043 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
24046 @strong{Do not send bug reports to @samp{info-gdb}, or to
24047 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
24048 not want to receive bug reports. Those that do have arranged to receive
24051 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
24052 serves as a repeater. The mailing list and the newsgroup carry exactly
24053 the same messages. Often people think of posting bug reports to the
24054 newsgroup instead of mailing them. This appears to work, but it has one
24055 problem which can be crucial: a newsgroup posting often lacks a mail
24056 path back to the sender. Thus, if we need to ask for more information,
24057 we may be unable to reach you. For this reason, it is better to send
24058 bug reports to the mailing list.
24060 @ifclear BUGURL_DEFAULT
24061 In any event, we also recommend that you submit bug reports for
24062 @value{GDBN} to @value{BUGURL}.
24066 The fundamental principle of reporting bugs usefully is this:
24067 @strong{report all the facts}. If you are not sure whether to state a
24068 fact or leave it out, state it!
24070 Often people omit facts because they think they know what causes the
24071 problem and assume that some details do not matter. Thus, you might
24072 assume that the name of the variable you use in an example does not matter.
24073 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
24074 stray memory reference which happens to fetch from the location where that
24075 name is stored in memory; perhaps, if the name were different, the contents
24076 of that location would fool the debugger into doing the right thing despite
24077 the bug. Play it safe and give a specific, complete example. That is the
24078 easiest thing for you to do, and the most helpful.
24080 Keep in mind that the purpose of a bug report is to enable us to fix the
24081 bug. It may be that the bug has been reported previously, but neither
24082 you nor we can know that unless your bug report is complete and
24085 Sometimes people give a few sketchy facts and ask, ``Does this ring a
24086 bell?'' Those bug reports are useless, and we urge everyone to
24087 @emph{refuse to respond to them} except to chide the sender to report
24090 To enable us to fix the bug, you should include all these things:
24094 The version of @value{GDBN}. @value{GDBN} announces it if you start
24095 with no arguments; you can also print it at any time using @code{show
24098 Without this, we will not know whether there is any point in looking for
24099 the bug in the current version of @value{GDBN}.
24102 The type of machine you are using, and the operating system name and
24106 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
24107 ``@value{GCC}--2.8.1''.
24110 What compiler (and its version) was used to compile the program you are
24111 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
24112 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
24113 to get this information; for other compilers, see the documentation for
24117 The command arguments you gave the compiler to compile your example and
24118 observe the bug. For example, did you use @samp{-O}? To guarantee
24119 you will not omit something important, list them all. A copy of the
24120 Makefile (or the output from make) is sufficient.
24122 If we were to try to guess the arguments, we would probably guess wrong
24123 and then we might not encounter the bug.
24126 A complete input script, and all necessary source files, that will
24130 A description of what behavior you observe that you believe is
24131 incorrect. For example, ``It gets a fatal signal.''
24133 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
24134 will certainly notice it. But if the bug is incorrect output, we might
24135 not notice unless it is glaringly wrong. You might as well not give us
24136 a chance to make a mistake.
24138 Even if the problem you experience is a fatal signal, you should still
24139 say so explicitly. Suppose something strange is going on, such as, your
24140 copy of @value{GDBN} is out of synch, or you have encountered a bug in
24141 the C library on your system. (This has happened!) Your copy might
24142 crash and ours would not. If you told us to expect a crash, then when
24143 ours fails to crash, we would know that the bug was not happening for
24144 us. If you had not told us to expect a crash, then we would not be able
24145 to draw any conclusion from our observations.
24148 @cindex recording a session script
24149 To collect all this information, you can use a session recording program
24150 such as @command{script}, which is available on many Unix systems.
24151 Just run your @value{GDBN} session inside @command{script} and then
24152 include the @file{typescript} file with your bug report.
24154 Another way to record a @value{GDBN} session is to run @value{GDBN}
24155 inside Emacs and then save the entire buffer to a file.
24158 If you wish to suggest changes to the @value{GDBN} source, send us context
24159 diffs. If you even discuss something in the @value{GDBN} source, refer to
24160 it by context, not by line number.
24162 The line numbers in our development sources will not match those in your
24163 sources. Your line numbers would convey no useful information to us.
24167 Here are some things that are not necessary:
24171 A description of the envelope of the bug.
24173 Often people who encounter a bug spend a lot of time investigating
24174 which changes to the input file will make the bug go away and which
24175 changes will not affect it.
24177 This is often time consuming and not very useful, because the way we
24178 will find the bug is by running a single example under the debugger
24179 with breakpoints, not by pure deduction from a series of examples.
24180 We recommend that you save your time for something else.
24182 Of course, if you can find a simpler example to report @emph{instead}
24183 of the original one, that is a convenience for us. Errors in the
24184 output will be easier to spot, running under the debugger will take
24185 less time, and so on.
24187 However, simplification is not vital; if you do not want to do this,
24188 report the bug anyway and send us the entire test case you used.
24191 A patch for the bug.
24193 A patch for the bug does help us if it is a good one. But do not omit
24194 the necessary information, such as the test case, on the assumption that
24195 a patch is all we need. We might see problems with your patch and decide
24196 to fix the problem another way, or we might not understand it at all.
24198 Sometimes with a program as complicated as @value{GDBN} it is very hard to
24199 construct an example that will make the program follow a certain path
24200 through the code. If you do not send us the example, we will not be able
24201 to construct one, so we will not be able to verify that the bug is fixed.
24203 And if we cannot understand what bug you are trying to fix, or why your
24204 patch should be an improvement, we will not install it. A test case will
24205 help us to understand.
24208 A guess about what the bug is or what it depends on.
24210 Such guesses are usually wrong. Even we cannot guess right about such
24211 things without first using the debugger to find the facts.
24214 @c The readline documentation is distributed with the readline code
24215 @c and consists of the two following files:
24217 @c inc-hist.texinfo
24218 @c Use -I with makeinfo to point to the appropriate directory,
24219 @c environment var TEXINPUTS with TeX.
24220 @include rluser.texi
24221 @include inc-hist.texinfo
24224 @node Formatting Documentation
24225 @appendix Formatting Documentation
24227 @cindex @value{GDBN} reference card
24228 @cindex reference card
24229 The @value{GDBN} 4 release includes an already-formatted reference card, ready
24230 for printing with PostScript or Ghostscript, in the @file{gdb}
24231 subdirectory of the main source directory@footnote{In
24232 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
24233 release.}. If you can use PostScript or Ghostscript with your printer,
24234 you can print the reference card immediately with @file{refcard.ps}.
24236 The release also includes the source for the reference card. You
24237 can format it, using @TeX{}, by typing:
24243 The @value{GDBN} reference card is designed to print in @dfn{landscape}
24244 mode on US ``letter'' size paper;
24245 that is, on a sheet 11 inches wide by 8.5 inches
24246 high. You will need to specify this form of printing as an option to
24247 your @sc{dvi} output program.
24249 @cindex documentation
24251 All the documentation for @value{GDBN} comes as part of the machine-readable
24252 distribution. The documentation is written in Texinfo format, which is
24253 a documentation system that uses a single source file to produce both
24254 on-line information and a printed manual. You can use one of the Info
24255 formatting commands to create the on-line version of the documentation
24256 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
24258 @value{GDBN} includes an already formatted copy of the on-line Info
24259 version of this manual in the @file{gdb} subdirectory. The main Info
24260 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
24261 subordinate files matching @samp{gdb.info*} in the same directory. If
24262 necessary, you can print out these files, or read them with any editor;
24263 but they are easier to read using the @code{info} subsystem in @sc{gnu}
24264 Emacs or the standalone @code{info} program, available as part of the
24265 @sc{gnu} Texinfo distribution.
24267 If you want to format these Info files yourself, you need one of the
24268 Info formatting programs, such as @code{texinfo-format-buffer} or
24271 If you have @code{makeinfo} installed, and are in the top level
24272 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
24273 version @value{GDBVN}), you can make the Info file by typing:
24280 If you want to typeset and print copies of this manual, you need @TeX{},
24281 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
24282 Texinfo definitions file.
24284 @TeX{} is a typesetting program; it does not print files directly, but
24285 produces output files called @sc{dvi} files. To print a typeset
24286 document, you need a program to print @sc{dvi} files. If your system
24287 has @TeX{} installed, chances are it has such a program. The precise
24288 command to use depends on your system; @kbd{lpr -d} is common; another
24289 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
24290 require a file name without any extension or a @samp{.dvi} extension.
24292 @TeX{} also requires a macro definitions file called
24293 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
24294 written in Texinfo format. On its own, @TeX{} cannot either read or
24295 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
24296 and is located in the @file{gdb-@var{version-number}/texinfo}
24299 If you have @TeX{} and a @sc{dvi} printer program installed, you can
24300 typeset and print this manual. First switch to the @file{gdb}
24301 subdirectory of the main source directory (for example, to
24302 @file{gdb-@value{GDBVN}/gdb}) and type:
24308 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
24310 @node Installing GDB
24311 @appendix Installing @value{GDBN}
24312 @cindex installation
24315 * Requirements:: Requirements for building @value{GDBN}
24316 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
24317 * Separate Objdir:: Compiling @value{GDBN} in another directory
24318 * Config Names:: Specifying names for hosts and targets
24319 * Configure Options:: Summary of options for configure
24323 @section Requirements for Building @value{GDBN}
24324 @cindex building @value{GDBN}, requirements for
24326 Building @value{GDBN} requires various tools and packages to be available.
24327 Other packages will be used only if they are found.
24329 @heading Tools/Packages Necessary for Building @value{GDBN}
24331 @item ISO C90 compiler
24332 @value{GDBN} is written in ISO C90. It should be buildable with any
24333 working C90 compiler, e.g.@: GCC.
24337 @heading Tools/Packages Optional for Building @value{GDBN}
24341 @value{GDBN} can use the Expat XML parsing library. This library may be
24342 included with your operating system distribution; if it is not, you
24343 can get the latest version from @url{http://expat.sourceforge.net}.
24344 The @file{configure} script will search for this library in several
24345 standard locations; if it is installed in an unusual path, you can
24346 use the @option{--with-libexpat-prefix} option to specify its location.
24352 Remote protocol memory maps (@pxref{Memory Map Format})
24354 Target descriptions (@pxref{Target Descriptions})
24356 Remote shared library lists (@pxref{Library List Format})
24358 MS-Windows shared libraries (@pxref{Shared Libraries})
24362 @cindex compressed debug sections
24363 @value{GDBN} will use the @samp{zlib} library, if available, to read
24364 compressed debug sections. Some linkers, such as GNU gold, are capable
24365 of producing binaries with compressed debug sections. If @value{GDBN}
24366 is compiled with @samp{zlib}, it will be able to read the debug
24367 information in such binaries.
24369 The @samp{zlib} library is likely included with your operating system
24370 distribution; if it is not, you can get the latest version from
24371 @url{http://zlib.net}.
24375 @node Running Configure
24376 @section Invoking the @value{GDBN} @file{configure} Script
24377 @cindex configuring @value{GDBN}
24378 @value{GDBN} comes with a @file{configure} script that automates the process
24379 of preparing @value{GDBN} for installation; you can then use @code{make} to
24380 build the @code{gdb} program.
24382 @c irrelevant in info file; it's as current as the code it lives with.
24383 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
24384 look at the @file{README} file in the sources; we may have improved the
24385 installation procedures since publishing this manual.}
24388 The @value{GDBN} distribution includes all the source code you need for
24389 @value{GDBN} in a single directory, whose name is usually composed by
24390 appending the version number to @samp{gdb}.
24392 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
24393 @file{gdb-@value{GDBVN}} directory. That directory contains:
24396 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
24397 script for configuring @value{GDBN} and all its supporting libraries
24399 @item gdb-@value{GDBVN}/gdb
24400 the source specific to @value{GDBN} itself
24402 @item gdb-@value{GDBVN}/bfd
24403 source for the Binary File Descriptor library
24405 @item gdb-@value{GDBVN}/include
24406 @sc{gnu} include files
24408 @item gdb-@value{GDBVN}/libiberty
24409 source for the @samp{-liberty} free software library
24411 @item gdb-@value{GDBVN}/opcodes
24412 source for the library of opcode tables and disassemblers
24414 @item gdb-@value{GDBVN}/readline
24415 source for the @sc{gnu} command-line interface
24417 @item gdb-@value{GDBVN}/glob
24418 source for the @sc{gnu} filename pattern-matching subroutine
24420 @item gdb-@value{GDBVN}/mmalloc
24421 source for the @sc{gnu} memory-mapped malloc package
24424 The simplest way to configure and build @value{GDBN} is to run @file{configure}
24425 from the @file{gdb-@var{version-number}} source directory, which in
24426 this example is the @file{gdb-@value{GDBVN}} directory.
24428 First switch to the @file{gdb-@var{version-number}} source directory
24429 if you are not already in it; then run @file{configure}. Pass the
24430 identifier for the platform on which @value{GDBN} will run as an
24436 cd gdb-@value{GDBVN}
24437 ./configure @var{host}
24442 where @var{host} is an identifier such as @samp{sun4} or
24443 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
24444 (You can often leave off @var{host}; @file{configure} tries to guess the
24445 correct value by examining your system.)
24447 Running @samp{configure @var{host}} and then running @code{make} builds the
24448 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
24449 libraries, then @code{gdb} itself. The configured source files, and the
24450 binaries, are left in the corresponding source directories.
24453 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
24454 system does not recognize this automatically when you run a different
24455 shell, you may need to run @code{sh} on it explicitly:
24458 sh configure @var{host}
24461 If you run @file{configure} from a directory that contains source
24462 directories for multiple libraries or programs, such as the
24463 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
24465 creates configuration files for every directory level underneath (unless
24466 you tell it not to, with the @samp{--norecursion} option).
24468 You should run the @file{configure} script from the top directory in the
24469 source tree, the @file{gdb-@var{version-number}} directory. If you run
24470 @file{configure} from one of the subdirectories, you will configure only
24471 that subdirectory. That is usually not what you want. In particular,
24472 if you run the first @file{configure} from the @file{gdb} subdirectory
24473 of the @file{gdb-@var{version-number}} directory, you will omit the
24474 configuration of @file{bfd}, @file{readline}, and other sibling
24475 directories of the @file{gdb} subdirectory. This leads to build errors
24476 about missing include files such as @file{bfd/bfd.h}.
24478 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
24479 However, you should make sure that the shell on your path (named by
24480 the @samp{SHELL} environment variable) is publicly readable. Remember
24481 that @value{GDBN} uses the shell to start your program---some systems refuse to
24482 let @value{GDBN} debug child processes whose programs are not readable.
24484 @node Separate Objdir
24485 @section Compiling @value{GDBN} in Another Directory
24487 If you want to run @value{GDBN} versions for several host or target machines,
24488 you need a different @code{gdb} compiled for each combination of
24489 host and target. @file{configure} is designed to make this easy by
24490 allowing you to generate each configuration in a separate subdirectory,
24491 rather than in the source directory. If your @code{make} program
24492 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
24493 @code{make} in each of these directories builds the @code{gdb}
24494 program specified there.
24496 To build @code{gdb} in a separate directory, run @file{configure}
24497 with the @samp{--srcdir} option to specify where to find the source.
24498 (You also need to specify a path to find @file{configure}
24499 itself from your working directory. If the path to @file{configure}
24500 would be the same as the argument to @samp{--srcdir}, you can leave out
24501 the @samp{--srcdir} option; it is assumed.)
24503 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
24504 separate directory for a Sun 4 like this:
24508 cd gdb-@value{GDBVN}
24511 ../gdb-@value{GDBVN}/configure sun4
24516 When @file{configure} builds a configuration using a remote source
24517 directory, it creates a tree for the binaries with the same structure
24518 (and using the same names) as the tree under the source directory. In
24519 the example, you'd find the Sun 4 library @file{libiberty.a} in the
24520 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
24521 @file{gdb-sun4/gdb}.
24523 Make sure that your path to the @file{configure} script has just one
24524 instance of @file{gdb} in it. If your path to @file{configure} looks
24525 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
24526 one subdirectory of @value{GDBN}, not the whole package. This leads to
24527 build errors about missing include files such as @file{bfd/bfd.h}.
24529 One popular reason to build several @value{GDBN} configurations in separate
24530 directories is to configure @value{GDBN} for cross-compiling (where
24531 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
24532 programs that run on another machine---the @dfn{target}).
24533 You specify a cross-debugging target by
24534 giving the @samp{--target=@var{target}} option to @file{configure}.
24536 When you run @code{make} to build a program or library, you must run
24537 it in a configured directory---whatever directory you were in when you
24538 called @file{configure} (or one of its subdirectories).
24540 The @code{Makefile} that @file{configure} generates in each source
24541 directory also runs recursively. If you type @code{make} in a source
24542 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
24543 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
24544 will build all the required libraries, and then build GDB.
24546 When you have multiple hosts or targets configured in separate
24547 directories, you can run @code{make} on them in parallel (for example,
24548 if they are NFS-mounted on each of the hosts); they will not interfere
24552 @section Specifying Names for Hosts and Targets
24554 The specifications used for hosts and targets in the @file{configure}
24555 script are based on a three-part naming scheme, but some short predefined
24556 aliases are also supported. The full naming scheme encodes three pieces
24557 of information in the following pattern:
24560 @var{architecture}-@var{vendor}-@var{os}
24563 For example, you can use the alias @code{sun4} as a @var{host} argument,
24564 or as the value for @var{target} in a @code{--target=@var{target}}
24565 option. The equivalent full name is @samp{sparc-sun-sunos4}.
24567 The @file{configure} script accompanying @value{GDBN} does not provide
24568 any query facility to list all supported host and target names or
24569 aliases. @file{configure} calls the Bourne shell script
24570 @code{config.sub} to map abbreviations to full names; you can read the
24571 script, if you wish, or you can use it to test your guesses on
24572 abbreviations---for example:
24575 % sh config.sub i386-linux
24577 % sh config.sub alpha-linux
24578 alpha-unknown-linux-gnu
24579 % sh config.sub hp9k700
24581 % sh config.sub sun4
24582 sparc-sun-sunos4.1.1
24583 % sh config.sub sun3
24584 m68k-sun-sunos4.1.1
24585 % sh config.sub i986v
24586 Invalid configuration `i986v': machine `i986v' not recognized
24590 @code{config.sub} is also distributed in the @value{GDBN} source
24591 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
24593 @node Configure Options
24594 @section @file{configure} Options
24596 Here is a summary of the @file{configure} options and arguments that
24597 are most often useful for building @value{GDBN}. @file{configure} also has
24598 several other options not listed here. @inforef{What Configure
24599 Does,,configure.info}, for a full explanation of @file{configure}.
24602 configure @r{[}--help@r{]}
24603 @r{[}--prefix=@var{dir}@r{]}
24604 @r{[}--exec-prefix=@var{dir}@r{]}
24605 @r{[}--srcdir=@var{dirname}@r{]}
24606 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
24607 @r{[}--target=@var{target}@r{]}
24612 You may introduce options with a single @samp{-} rather than
24613 @samp{--} if you prefer; but you may abbreviate option names if you use
24618 Display a quick summary of how to invoke @file{configure}.
24620 @item --prefix=@var{dir}
24621 Configure the source to install programs and files under directory
24624 @item --exec-prefix=@var{dir}
24625 Configure the source to install programs under directory
24628 @c avoid splitting the warning from the explanation:
24630 @item --srcdir=@var{dirname}
24631 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
24632 @code{make} that implements the @code{VPATH} feature.}@*
24633 Use this option to make configurations in directories separate from the
24634 @value{GDBN} source directories. Among other things, you can use this to
24635 build (or maintain) several configurations simultaneously, in separate
24636 directories. @file{configure} writes configuration-specific files in
24637 the current directory, but arranges for them to use the source in the
24638 directory @var{dirname}. @file{configure} creates directories under
24639 the working directory in parallel to the source directories below
24642 @item --norecursion
24643 Configure only the directory level where @file{configure} is executed; do not
24644 propagate configuration to subdirectories.
24646 @item --target=@var{target}
24647 Configure @value{GDBN} for cross-debugging programs running on the specified
24648 @var{target}. Without this option, @value{GDBN} is configured to debug
24649 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
24651 There is no convenient way to generate a list of all available targets.
24653 @item @var{host} @dots{}
24654 Configure @value{GDBN} to run on the specified @var{host}.
24656 There is no convenient way to generate a list of all available hosts.
24659 There are many other options available as well, but they are generally
24660 needed for special purposes only.
24662 @node Maintenance Commands
24663 @appendix Maintenance Commands
24664 @cindex maintenance commands
24665 @cindex internal commands
24667 In addition to commands intended for @value{GDBN} users, @value{GDBN}
24668 includes a number of commands intended for @value{GDBN} developers,
24669 that are not documented elsewhere in this manual. These commands are
24670 provided here for reference. (For commands that turn on debugging
24671 messages, see @ref{Debugging Output}.)
24674 @kindex maint agent
24675 @item maint agent @var{expression}
24676 Translate the given @var{expression} into remote agent bytecodes.
24677 This command is useful for debugging the Agent Expression mechanism
24678 (@pxref{Agent Expressions}).
24680 @kindex maint info breakpoints
24681 @item @anchor{maint info breakpoints}maint info breakpoints
24682 Using the same format as @samp{info breakpoints}, display both the
24683 breakpoints you've set explicitly, and those @value{GDBN} is using for
24684 internal purposes. Internal breakpoints are shown with negative
24685 breakpoint numbers. The type column identifies what kind of breakpoint
24690 Normal, explicitly set breakpoint.
24693 Normal, explicitly set watchpoint.
24696 Internal breakpoint, used to handle correctly stepping through
24697 @code{longjmp} calls.
24699 @item longjmp resume
24700 Internal breakpoint at the target of a @code{longjmp}.
24703 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
24706 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
24709 Shared library events.
24713 @kindex set displaced-stepping
24714 @kindex show displaced-stepping
24715 @cindex displaced stepping support
24716 @cindex out-of-line single-stepping
24717 @item set displaced-stepping
24718 @itemx show displaced-stepping
24719 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
24720 if the target supports it. Displaced stepping is a way to single-step
24721 over breakpoints without removing them from the inferior, by executing
24722 an out-of-line copy of the instruction that was originally at the
24723 breakpoint location. It is also known as out-of-line single-stepping.
24726 @item set displaced-stepping on
24727 If the target architecture supports it, @value{GDBN} will use
24728 displaced stepping to step over breakpoints.
24730 @item set displaced-stepping off
24731 @value{GDBN} will not use displaced stepping to step over breakpoints,
24732 even if such is supported by the target architecture.
24734 @cindex non-stop mode, and @samp{set displaced-stepping}
24735 @item set displaced-stepping auto
24736 This is the default mode. @value{GDBN} will use displaced stepping
24737 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
24738 architecture supports displaced stepping.
24741 @kindex maint check-symtabs
24742 @item maint check-symtabs
24743 Check the consistency of psymtabs and symtabs.
24745 @kindex maint cplus first_component
24746 @item maint cplus first_component @var{name}
24747 Print the first C@t{++} class/namespace component of @var{name}.
24749 @kindex maint cplus namespace
24750 @item maint cplus namespace
24751 Print the list of possible C@t{++} namespaces.
24753 @kindex maint demangle
24754 @item maint demangle @var{name}
24755 Demangle a C@t{++} or Objective-C mangled @var{name}.
24757 @kindex maint deprecate
24758 @kindex maint undeprecate
24759 @cindex deprecated commands
24760 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
24761 @itemx maint undeprecate @var{command}
24762 Deprecate or undeprecate the named @var{command}. Deprecated commands
24763 cause @value{GDBN} to issue a warning when you use them. The optional
24764 argument @var{replacement} says which newer command should be used in
24765 favor of the deprecated one; if it is given, @value{GDBN} will mention
24766 the replacement as part of the warning.
24768 @kindex maint dump-me
24769 @item maint dump-me
24770 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
24771 Cause a fatal signal in the debugger and force it to dump its core.
24772 This is supported only on systems which support aborting a program
24773 with the @code{SIGQUIT} signal.
24775 @kindex maint internal-error
24776 @kindex maint internal-warning
24777 @item maint internal-error @r{[}@var{message-text}@r{]}
24778 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
24779 Cause @value{GDBN} to call the internal function @code{internal_error}
24780 or @code{internal_warning} and hence behave as though an internal error
24781 or internal warning has been detected. In addition to reporting the
24782 internal problem, these functions give the user the opportunity to
24783 either quit @value{GDBN} or create a core file of the current
24784 @value{GDBN} session.
24786 These commands take an optional parameter @var{message-text} that is
24787 used as the text of the error or warning message.
24789 Here's an example of using @code{internal-error}:
24792 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
24793 @dots{}/maint.c:121: internal-error: testing, 1, 2
24794 A problem internal to GDB has been detected. Further
24795 debugging may prove unreliable.
24796 Quit this debugging session? (y or n) @kbd{n}
24797 Create a core file? (y or n) @kbd{n}
24801 @kindex maint packet
24802 @item maint packet @var{text}
24803 If @value{GDBN} is talking to an inferior via the serial protocol,
24804 then this command sends the string @var{text} to the inferior, and
24805 displays the response packet. @value{GDBN} supplies the initial
24806 @samp{$} character, the terminating @samp{#} character, and the
24809 @kindex maint print architecture
24810 @item maint print architecture @r{[}@var{file}@r{]}
24811 Print the entire architecture configuration. The optional argument
24812 @var{file} names the file where the output goes.
24814 @kindex maint print c-tdesc
24815 @item maint print c-tdesc
24816 Print the current target description (@pxref{Target Descriptions}) as
24817 a C source file. The created source file can be used in @value{GDBN}
24818 when an XML parser is not available to parse the description.
24820 @kindex maint print dummy-frames
24821 @item maint print dummy-frames
24822 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
24825 (@value{GDBP}) @kbd{b add}
24827 (@value{GDBP}) @kbd{print add(2,3)}
24828 Breakpoint 2, add (a=2, b=3) at @dots{}
24830 The program being debugged stopped while in a function called from GDB.
24832 (@value{GDBP}) @kbd{maint print dummy-frames}
24833 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
24834 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
24835 call_lo=0x01014000 call_hi=0x01014001
24839 Takes an optional file parameter.
24841 @kindex maint print registers
24842 @kindex maint print raw-registers
24843 @kindex maint print cooked-registers
24844 @kindex maint print register-groups
24845 @item maint print registers @r{[}@var{file}@r{]}
24846 @itemx maint print raw-registers @r{[}@var{file}@r{]}
24847 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
24848 @itemx maint print register-groups @r{[}@var{file}@r{]}
24849 Print @value{GDBN}'s internal register data structures.
24851 The command @code{maint print raw-registers} includes the contents of
24852 the raw register cache; the command @code{maint print cooked-registers}
24853 includes the (cooked) value of all registers; and the command
24854 @code{maint print register-groups} includes the groups that each
24855 register is a member of. @xref{Registers,, Registers, gdbint,
24856 @value{GDBN} Internals}.
24858 These commands take an optional parameter, a file name to which to
24859 write the information.
24861 @kindex maint print reggroups
24862 @item maint print reggroups @r{[}@var{file}@r{]}
24863 Print @value{GDBN}'s internal register group data structures. The
24864 optional argument @var{file} tells to what file to write the
24867 The register groups info looks like this:
24870 (@value{GDBP}) @kbd{maint print reggroups}
24883 This command forces @value{GDBN} to flush its internal register cache.
24885 @kindex maint print objfiles
24886 @cindex info for known object files
24887 @item maint print objfiles
24888 Print a dump of all known object files. For each object file, this
24889 command prints its name, address in memory, and all of its psymtabs
24892 @kindex maint print statistics
24893 @cindex bcache statistics
24894 @item maint print statistics
24895 This command prints, for each object file in the program, various data
24896 about that object file followed by the byte cache (@dfn{bcache})
24897 statistics for the object file. The objfile data includes the number
24898 of minimal, partial, full, and stabs symbols, the number of types
24899 defined by the objfile, the number of as yet unexpanded psym tables,
24900 the number of line tables and string tables, and the amount of memory
24901 used by the various tables. The bcache statistics include the counts,
24902 sizes, and counts of duplicates of all and unique objects, max,
24903 average, and median entry size, total memory used and its overhead and
24904 savings, and various measures of the hash table size and chain
24907 @kindex maint print target-stack
24908 @cindex target stack description
24909 @item maint print target-stack
24910 A @dfn{target} is an interface between the debugger and a particular
24911 kind of file or process. Targets can be stacked in @dfn{strata},
24912 so that more than one target can potentially respond to a request.
24913 In particular, memory accesses will walk down the stack of targets
24914 until they find a target that is interested in handling that particular
24917 This command prints a short description of each layer that was pushed on
24918 the @dfn{target stack}, starting from the top layer down to the bottom one.
24920 @kindex maint print type
24921 @cindex type chain of a data type
24922 @item maint print type @var{expr}
24923 Print the type chain for a type specified by @var{expr}. The argument
24924 can be either a type name or a symbol. If it is a symbol, the type of
24925 that symbol is described. The type chain produced by this command is
24926 a recursive definition of the data type as stored in @value{GDBN}'s
24927 data structures, including its flags and contained types.
24929 @kindex maint set dwarf2 max-cache-age
24930 @kindex maint show dwarf2 max-cache-age
24931 @item maint set dwarf2 max-cache-age
24932 @itemx maint show dwarf2 max-cache-age
24933 Control the DWARF 2 compilation unit cache.
24935 @cindex DWARF 2 compilation units cache
24936 In object files with inter-compilation-unit references, such as those
24937 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
24938 reader needs to frequently refer to previously read compilation units.
24939 This setting controls how long a compilation unit will remain in the
24940 cache if it is not referenced. A higher limit means that cached
24941 compilation units will be stored in memory longer, and more total
24942 memory will be used. Setting it to zero disables caching, which will
24943 slow down @value{GDBN} startup, but reduce memory consumption.
24945 @kindex maint set profile
24946 @kindex maint show profile
24947 @cindex profiling GDB
24948 @item maint set profile
24949 @itemx maint show profile
24950 Control profiling of @value{GDBN}.
24952 Profiling will be disabled until you use the @samp{maint set profile}
24953 command to enable it. When you enable profiling, the system will begin
24954 collecting timing and execution count data; when you disable profiling or
24955 exit @value{GDBN}, the results will be written to a log file. Remember that
24956 if you use profiling, @value{GDBN} will overwrite the profiling log file
24957 (often called @file{gmon.out}). If you have a record of important profiling
24958 data in a @file{gmon.out} file, be sure to move it to a safe location.
24960 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
24961 compiled with the @samp{-pg} compiler option.
24963 @kindex maint set linux-async
24964 @kindex maint show linux-async
24965 @cindex asynchronous support
24966 @item maint set linux-async
24967 @itemx maint show linux-async
24968 Control the GNU/Linux native asynchronous support
24969 (@pxref{Background Execution}) of @value{GDBN}.
24971 GNU/Linux native asynchronous support will be disabled until you use
24972 the @samp{maint set linux-async} command to enable it.
24974 @kindex maint set remote-async
24975 @kindex maint show remote-async
24976 @cindex asynchronous support
24977 @item maint set remote-async
24978 @itemx maint show remote-async
24979 Control the remote asynchronous support
24980 (@pxref{Background Execution}) of @value{GDBN}.
24982 Remote asynchronous support will be disabled until you use
24983 the @samp{maint set remote-async} command to enable it.
24985 @kindex maint show-debug-regs
24986 @cindex x86 hardware debug registers
24987 @item maint show-debug-regs
24988 Control whether to show variables that mirror the x86 hardware debug
24989 registers. Use @code{ON} to enable, @code{OFF} to disable. If
24990 enabled, the debug registers values are shown when @value{GDBN} inserts or
24991 removes a hardware breakpoint or watchpoint, and when the inferior
24992 triggers a hardware-assisted breakpoint or watchpoint.
24994 @kindex maint space
24995 @cindex memory used by commands
24997 Control whether to display memory usage for each command. If set to a
24998 nonzero value, @value{GDBN} will display how much memory each command
24999 took, following the command's own output. This can also be requested
25000 by invoking @value{GDBN} with the @option{--statistics} command-line
25001 switch (@pxref{Mode Options}).
25004 @cindex time of command execution
25006 Control whether to display the execution time for each command. If
25007 set to a nonzero value, @value{GDBN} will display how much time it
25008 took to execute each command, following the command's own output.
25009 The time is not printed for the commands that run the target, since
25010 there's no mechanism currently to compute how much time was spend
25011 by @value{GDBN} and how much time was spend by the program been debugged.
25012 it's not possibly currently
25013 This can also be requested by invoking @value{GDBN} with the
25014 @option{--statistics} command-line switch (@pxref{Mode Options}).
25016 @kindex maint translate-address
25017 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
25018 Find the symbol stored at the location specified by the address
25019 @var{addr} and an optional section name @var{section}. If found,
25020 @value{GDBN} prints the name of the closest symbol and an offset from
25021 the symbol's location to the specified address. This is similar to
25022 the @code{info address} command (@pxref{Symbols}), except that this
25023 command also allows to find symbols in other sections.
25025 If section was not specified, the section in which the symbol was found
25026 is also printed. For dynamically linked executables, the name of
25027 executable or shared library containing the symbol is printed as well.
25031 The following command is useful for non-interactive invocations of
25032 @value{GDBN}, such as in the test suite.
25035 @item set watchdog @var{nsec}
25036 @kindex set watchdog
25037 @cindex watchdog timer
25038 @cindex timeout for commands
25039 Set the maximum number of seconds @value{GDBN} will wait for the
25040 target operation to finish. If this time expires, @value{GDBN}
25041 reports and error and the command is aborted.
25043 @item show watchdog
25044 Show the current setting of the target wait timeout.
25047 @node Remote Protocol
25048 @appendix @value{GDBN} Remote Serial Protocol
25053 * Stop Reply Packets::
25054 * General Query Packets::
25055 * Register Packet Format::
25056 * Tracepoint Packets::
25057 * Host I/O Packets::
25059 * Notification Packets::
25060 * Remote Non-Stop::
25061 * Packet Acknowledgment::
25063 * File-I/O Remote Protocol Extension::
25064 * Library List Format::
25065 * Memory Map Format::
25071 There may be occasions when you need to know something about the
25072 protocol---for example, if there is only one serial port to your target
25073 machine, you might want your program to do something special if it
25074 recognizes a packet meant for @value{GDBN}.
25076 In the examples below, @samp{->} and @samp{<-} are used to indicate
25077 transmitted and received data, respectively.
25079 @cindex protocol, @value{GDBN} remote serial
25080 @cindex serial protocol, @value{GDBN} remote
25081 @cindex remote serial protocol
25082 All @value{GDBN} commands and responses (other than acknowledgments
25083 and notifications, see @ref{Notification Packets}) are sent as a
25084 @var{packet}. A @var{packet} is introduced with the character
25085 @samp{$}, the actual @var{packet-data}, and the terminating character
25086 @samp{#} followed by a two-digit @var{checksum}:
25089 @code{$}@var{packet-data}@code{#}@var{checksum}
25093 @cindex checksum, for @value{GDBN} remote
25095 The two-digit @var{checksum} is computed as the modulo 256 sum of all
25096 characters between the leading @samp{$} and the trailing @samp{#} (an
25097 eight bit unsigned checksum).
25099 Implementors should note that prior to @value{GDBN} 5.0 the protocol
25100 specification also included an optional two-digit @var{sequence-id}:
25103 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
25106 @cindex sequence-id, for @value{GDBN} remote
25108 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
25109 has never output @var{sequence-id}s. Stubs that handle packets added
25110 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
25112 When either the host or the target machine receives a packet, the first
25113 response expected is an acknowledgment: either @samp{+} (to indicate
25114 the package was received correctly) or @samp{-} (to request
25118 -> @code{$}@var{packet-data}@code{#}@var{checksum}
25123 The @samp{+}/@samp{-} acknowledgments can be disabled
25124 once a connection is established.
25125 @xref{Packet Acknowledgment}, for details.
25127 The host (@value{GDBN}) sends @var{command}s, and the target (the
25128 debugging stub incorporated in your program) sends a @var{response}. In
25129 the case of step and continue @var{command}s, the response is only sent
25130 when the operation has completed, and the target has again stopped all
25131 threads in all attached processes. This is the default all-stop mode
25132 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
25133 execution mode; see @ref{Remote Non-Stop}, for details.
25135 @var{packet-data} consists of a sequence of characters with the
25136 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
25139 @cindex remote protocol, field separator
25140 Fields within the packet should be separated using @samp{,} @samp{;} or
25141 @samp{:}. Except where otherwise noted all numbers are represented in
25142 @sc{hex} with leading zeros suppressed.
25144 Implementors should note that prior to @value{GDBN} 5.0, the character
25145 @samp{:} could not appear as the third character in a packet (as it
25146 would potentially conflict with the @var{sequence-id}).
25148 @cindex remote protocol, binary data
25149 @anchor{Binary Data}
25150 Binary data in most packets is encoded either as two hexadecimal
25151 digits per byte of binary data. This allowed the traditional remote
25152 protocol to work over connections which were only seven-bit clean.
25153 Some packets designed more recently assume an eight-bit clean
25154 connection, and use a more efficient encoding to send and receive
25157 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
25158 as an escape character. Any escaped byte is transmitted as the escape
25159 character followed by the original character XORed with @code{0x20}.
25160 For example, the byte @code{0x7d} would be transmitted as the two
25161 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
25162 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
25163 @samp{@}}) must always be escaped. Responses sent by the stub
25164 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
25165 is not interpreted as the start of a run-length encoded sequence
25168 Response @var{data} can be run-length encoded to save space.
25169 Run-length encoding replaces runs of identical characters with one
25170 instance of the repeated character, followed by a @samp{*} and a
25171 repeat count. The repeat count is itself sent encoded, to avoid
25172 binary characters in @var{data}: a value of @var{n} is sent as
25173 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
25174 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
25175 code 32) for a repeat count of 3. (This is because run-length
25176 encoding starts to win for counts 3 or more.) Thus, for example,
25177 @samp{0* } is a run-length encoding of ``0000'': the space character
25178 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
25181 The printable characters @samp{#} and @samp{$} or with a numeric value
25182 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
25183 seven repeats (@samp{$}) can be expanded using a repeat count of only
25184 five (@samp{"}). For example, @samp{00000000} can be encoded as
25187 The error response returned for some packets includes a two character
25188 error number. That number is not well defined.
25190 @cindex empty response, for unsupported packets
25191 For any @var{command} not supported by the stub, an empty response
25192 (@samp{$#00}) should be returned. That way it is possible to extend the
25193 protocol. A newer @value{GDBN} can tell if a packet is supported based
25196 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
25197 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
25203 The following table provides a complete list of all currently defined
25204 @var{command}s and their corresponding response @var{data}.
25205 @xref{File-I/O Remote Protocol Extension}, for details about the File
25206 I/O extension of the remote protocol.
25208 Each packet's description has a template showing the packet's overall
25209 syntax, followed by an explanation of the packet's meaning. We
25210 include spaces in some of the templates for clarity; these are not
25211 part of the packet's syntax. No @value{GDBN} packet uses spaces to
25212 separate its components. For example, a template like @samp{foo
25213 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
25214 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
25215 @var{baz}. @value{GDBN} does not transmit a space character between the
25216 @samp{foo} and the @var{bar}, or between the @var{bar} and the
25219 @cindex @var{thread-id}, in remote protocol
25220 @anchor{thread-id syntax}
25221 Several packets and replies include a @var{thread-id} field to identify
25222 a thread. Normally these are positive numbers with a target-specific
25223 interpretation, formatted as big-endian hex strings. A @var{thread-id}
25224 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
25227 In addition, the remote protocol supports a multiprocess feature in
25228 which the @var{thread-id} syntax is extended to optionally include both
25229 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
25230 The @var{pid} (process) and @var{tid} (thread) components each have the
25231 format described above: a positive number with target-specific
25232 interpretation formatted as a big-endian hex string, literal @samp{-1}
25233 to indicate all processes or threads (respectively), or @samp{0} to
25234 indicate an arbitrary process or thread. Specifying just a process, as
25235 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
25236 error to specify all processes but a specific thread, such as
25237 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
25238 for those packets and replies explicitly documented to include a process
25239 ID, rather than a @var{thread-id}.
25241 The multiprocess @var{thread-id} syntax extensions are only used if both
25242 @value{GDBN} and the stub report support for the @samp{multiprocess}
25243 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
25246 Note that all packet forms beginning with an upper- or lower-case
25247 letter, other than those described here, are reserved for future use.
25249 Here are the packet descriptions.
25254 @cindex @samp{!} packet
25255 @anchor{extended mode}
25256 Enable extended mode. In extended mode, the remote server is made
25257 persistent. The @samp{R} packet is used to restart the program being
25263 The remote target both supports and has enabled extended mode.
25267 @cindex @samp{?} packet
25268 Indicate the reason the target halted. The reply is the same as for
25269 step and continue. This packet has a special interpretation when the
25270 target is in non-stop mode; see @ref{Remote Non-Stop}.
25273 @xref{Stop Reply Packets}, for the reply specifications.
25275 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
25276 @cindex @samp{A} packet
25277 Initialized @code{argv[]} array passed into program. @var{arglen}
25278 specifies the number of bytes in the hex encoded byte stream
25279 @var{arg}. See @code{gdbserver} for more details.
25284 The arguments were set.
25290 @cindex @samp{b} packet
25291 (Don't use this packet; its behavior is not well-defined.)
25292 Change the serial line speed to @var{baud}.
25294 JTC: @emph{When does the transport layer state change? When it's
25295 received, or after the ACK is transmitted. In either case, there are
25296 problems if the command or the acknowledgment packet is dropped.}
25298 Stan: @emph{If people really wanted to add something like this, and get
25299 it working for the first time, they ought to modify ser-unix.c to send
25300 some kind of out-of-band message to a specially-setup stub and have the
25301 switch happen "in between" packets, so that from remote protocol's point
25302 of view, nothing actually happened.}
25304 @item B @var{addr},@var{mode}
25305 @cindex @samp{B} packet
25306 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
25307 breakpoint at @var{addr}.
25309 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
25310 (@pxref{insert breakpoint or watchpoint packet}).
25313 @cindex @samp{bc} packet
25314 Backward continue. Execute the target system in reverse. No parameter.
25315 @xref{Reverse Execution}, for more information.
25318 @xref{Stop Reply Packets}, for the reply specifications.
25321 @cindex @samp{bs} packet
25322 Backward single step. Execute one instruction in reverse. No parameter.
25323 @xref{Reverse Execution}, for more information.
25326 @xref{Stop Reply Packets}, for the reply specifications.
25328 @item c @r{[}@var{addr}@r{]}
25329 @cindex @samp{c} packet
25330 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
25331 resume at current address.
25334 @xref{Stop Reply Packets}, for the reply specifications.
25336 @item C @var{sig}@r{[};@var{addr}@r{]}
25337 @cindex @samp{C} packet
25338 Continue with signal @var{sig} (hex signal number). If
25339 @samp{;@var{addr}} is omitted, resume at same address.
25342 @xref{Stop Reply Packets}, for the reply specifications.
25345 @cindex @samp{d} packet
25348 Don't use this packet; instead, define a general set packet
25349 (@pxref{General Query Packets}).
25353 @cindex @samp{D} packet
25354 The first form of the packet is used to detach @value{GDBN} from the
25355 remote system. It is sent to the remote target
25356 before @value{GDBN} disconnects via the @code{detach} command.
25358 The second form, including a process ID, is used when multiprocess
25359 protocol extensions are enabled (@pxref{multiprocess extensions}), to
25360 detach only a specific process. The @var{pid} is specified as a
25361 big-endian hex string.
25371 @item F @var{RC},@var{EE},@var{CF};@var{XX}
25372 @cindex @samp{F} packet
25373 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
25374 This is part of the File-I/O protocol extension. @xref{File-I/O
25375 Remote Protocol Extension}, for the specification.
25378 @anchor{read registers packet}
25379 @cindex @samp{g} packet
25380 Read general registers.
25384 @item @var{XX@dots{}}
25385 Each byte of register data is described by two hex digits. The bytes
25386 with the register are transmitted in target byte order. The size of
25387 each register and their position within the @samp{g} packet are
25388 determined by the @value{GDBN} internal gdbarch functions
25389 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
25390 specification of several standard @samp{g} packets is specified below.
25395 @item G @var{XX@dots{}}
25396 @cindex @samp{G} packet
25397 Write general registers. @xref{read registers packet}, for a
25398 description of the @var{XX@dots{}} data.
25408 @item H @var{c} @var{thread-id}
25409 @cindex @samp{H} packet
25410 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
25411 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
25412 should be @samp{c} for step and continue operations, @samp{g} for other
25413 operations. The thread designator @var{thread-id} has the format and
25414 interpretation described in @ref{thread-id syntax}.
25425 @c 'H': How restrictive (or permissive) is the thread model. If a
25426 @c thread is selected and stopped, are other threads allowed
25427 @c to continue to execute? As I mentioned above, I think the
25428 @c semantics of each command when a thread is selected must be
25429 @c described. For example:
25431 @c 'g': If the stub supports threads and a specific thread is
25432 @c selected, returns the register block from that thread;
25433 @c otherwise returns current registers.
25435 @c 'G' If the stub supports threads and a specific thread is
25436 @c selected, sets the registers of the register block of
25437 @c that thread; otherwise sets current registers.
25439 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
25440 @anchor{cycle step packet}
25441 @cindex @samp{i} packet
25442 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
25443 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
25444 step starting at that address.
25447 @cindex @samp{I} packet
25448 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
25452 @cindex @samp{k} packet
25455 FIXME: @emph{There is no description of how to operate when a specific
25456 thread context has been selected (i.e.@: does 'k' kill only that
25459 @item m @var{addr},@var{length}
25460 @cindex @samp{m} packet
25461 Read @var{length} bytes of memory starting at address @var{addr}.
25462 Note that @var{addr} may not be aligned to any particular boundary.
25464 The stub need not use any particular size or alignment when gathering
25465 data from memory for the response; even if @var{addr} is word-aligned
25466 and @var{length} is a multiple of the word size, the stub is free to
25467 use byte accesses, or not. For this reason, this packet may not be
25468 suitable for accessing memory-mapped I/O devices.
25469 @cindex alignment of remote memory accesses
25470 @cindex size of remote memory accesses
25471 @cindex memory, alignment and size of remote accesses
25475 @item @var{XX@dots{}}
25476 Memory contents; each byte is transmitted as a two-digit hexadecimal
25477 number. The reply may contain fewer bytes than requested if the
25478 server was able to read only part of the region of memory.
25483 @item M @var{addr},@var{length}:@var{XX@dots{}}
25484 @cindex @samp{M} packet
25485 Write @var{length} bytes of memory starting at address @var{addr}.
25486 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
25487 hexadecimal number.
25494 for an error (this includes the case where only part of the data was
25499 @cindex @samp{p} packet
25500 Read the value of register @var{n}; @var{n} is in hex.
25501 @xref{read registers packet}, for a description of how the returned
25502 register value is encoded.
25506 @item @var{XX@dots{}}
25507 the register's value
25511 Indicating an unrecognized @var{query}.
25514 @item P @var{n@dots{}}=@var{r@dots{}}
25515 @anchor{write register packet}
25516 @cindex @samp{P} packet
25517 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
25518 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
25519 digits for each byte in the register (target byte order).
25529 @item q @var{name} @var{params}@dots{}
25530 @itemx Q @var{name} @var{params}@dots{}
25531 @cindex @samp{q} packet
25532 @cindex @samp{Q} packet
25533 General query (@samp{q}) and set (@samp{Q}). These packets are
25534 described fully in @ref{General Query Packets}.
25537 @cindex @samp{r} packet
25538 Reset the entire system.
25540 Don't use this packet; use the @samp{R} packet instead.
25543 @cindex @samp{R} packet
25544 Restart the program being debugged. @var{XX}, while needed, is ignored.
25545 This packet is only available in extended mode (@pxref{extended mode}).
25547 The @samp{R} packet has no reply.
25549 @item s @r{[}@var{addr}@r{]}
25550 @cindex @samp{s} packet
25551 Single step. @var{addr} is the address at which to resume. If
25552 @var{addr} is omitted, resume at same address.
25555 @xref{Stop Reply Packets}, for the reply specifications.
25557 @item S @var{sig}@r{[};@var{addr}@r{]}
25558 @anchor{step with signal packet}
25559 @cindex @samp{S} packet
25560 Step with signal. This is analogous to the @samp{C} packet, but
25561 requests a single-step, rather than a normal resumption of execution.
25564 @xref{Stop Reply Packets}, for the reply specifications.
25566 @item t @var{addr}:@var{PP},@var{MM}
25567 @cindex @samp{t} packet
25568 Search backwards starting at address @var{addr} for a match with pattern
25569 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
25570 @var{addr} must be at least 3 digits.
25572 @item T @var{thread-id}
25573 @cindex @samp{T} packet
25574 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
25579 thread is still alive
25585 Packets starting with @samp{v} are identified by a multi-letter name,
25586 up to the first @samp{;} or @samp{?} (or the end of the packet).
25588 @item vAttach;@var{pid}
25589 @cindex @samp{vAttach} packet
25590 Attach to a new process with the specified process ID @var{pid}.
25591 The process ID is a
25592 hexadecimal integer identifying the process. In all-stop mode, all
25593 threads in the attached process are stopped; in non-stop mode, it may be
25594 attached without being stopped if that is supported by the target.
25596 @c In non-stop mode, on a successful vAttach, the stub should set the
25597 @c current thread to a thread of the newly-attached process. After
25598 @c attaching, GDB queries for the attached process's thread ID with qC.
25599 @c Also note that, from a user perspective, whether or not the
25600 @c target is stopped on attach in non-stop mode depends on whether you
25601 @c use the foreground or background version of the attach command, not
25602 @c on what vAttach does; GDB does the right thing with respect to either
25603 @c stopping or restarting threads.
25605 This packet is only available in extended mode (@pxref{extended mode}).
25611 @item @r{Any stop packet}
25612 for success in all-stop mode (@pxref{Stop Reply Packets})
25614 for success in non-stop mode (@pxref{Remote Non-Stop})
25617 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
25618 @cindex @samp{vCont} packet
25619 Resume the inferior, specifying different actions for each thread.
25620 If an action is specified with no @var{thread-id}, then it is applied to any
25621 threads that don't have a specific action specified; if no default action is
25622 specified then other threads should remain stopped in all-stop mode and
25623 in their current state in non-stop mode.
25624 Specifying multiple
25625 default actions is an error; specifying no actions is also an error.
25626 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
25628 Currently supported actions are:
25634 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
25638 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
25642 Stop with signal @var{sig}. The signal @var{sig} should be two hex digits.
25645 The optional argument @var{addr} normally associated with the
25646 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
25647 not supported in @samp{vCont}.
25649 The @samp{t} and @samp{T} actions are only relevant in non-stop mode
25650 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
25651 A stop reply should be generated for any affected thread not already stopped.
25652 When a thread is stopped by means of a @samp{t} action,
25653 the corresponding stop reply should indicate that the thread has stopped with
25654 signal @samp{0}, regardless of whether the target uses some other signal
25655 as an implementation detail.
25658 @xref{Stop Reply Packets}, for the reply specifications.
25661 @cindex @samp{vCont?} packet
25662 Request a list of actions supported by the @samp{vCont} packet.
25666 @item vCont@r{[};@var{action}@dots{}@r{]}
25667 The @samp{vCont} packet is supported. Each @var{action} is a supported
25668 command in the @samp{vCont} packet.
25670 The @samp{vCont} packet is not supported.
25673 @item vFile:@var{operation}:@var{parameter}@dots{}
25674 @cindex @samp{vFile} packet
25675 Perform a file operation on the target system. For details,
25676 see @ref{Host I/O Packets}.
25678 @item vFlashErase:@var{addr},@var{length}
25679 @cindex @samp{vFlashErase} packet
25680 Direct the stub to erase @var{length} bytes of flash starting at
25681 @var{addr}. The region may enclose any number of flash blocks, but
25682 its start and end must fall on block boundaries, as indicated by the
25683 flash block size appearing in the memory map (@pxref{Memory Map
25684 Format}). @value{GDBN} groups flash memory programming operations
25685 together, and sends a @samp{vFlashDone} request after each group; the
25686 stub is allowed to delay erase operation until the @samp{vFlashDone}
25687 packet is received.
25689 The stub must support @samp{vCont} if it reports support for
25690 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
25691 this case @samp{vCont} actions can be specified to apply to all threads
25692 in a process by using the @samp{p@var{pid}.-1} form of the
25703 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
25704 @cindex @samp{vFlashWrite} packet
25705 Direct the stub to write data to flash address @var{addr}. The data
25706 is passed in binary form using the same encoding as for the @samp{X}
25707 packet (@pxref{Binary Data}). The memory ranges specified by
25708 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
25709 not overlap, and must appear in order of increasing addresses
25710 (although @samp{vFlashErase} packets for higher addresses may already
25711 have been received; the ordering is guaranteed only between
25712 @samp{vFlashWrite} packets). If a packet writes to an address that was
25713 neither erased by a preceding @samp{vFlashErase} packet nor by some other
25714 target-specific method, the results are unpredictable.
25722 for vFlashWrite addressing non-flash memory
25728 @cindex @samp{vFlashDone} packet
25729 Indicate to the stub that flash programming operation is finished.
25730 The stub is permitted to delay or batch the effects of a group of
25731 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
25732 @samp{vFlashDone} packet is received. The contents of the affected
25733 regions of flash memory are unpredictable until the @samp{vFlashDone}
25734 request is completed.
25736 @item vKill;@var{pid}
25737 @cindex @samp{vKill} packet
25738 Kill the process with the specified process ID. @var{pid} is a
25739 hexadecimal integer identifying the process. This packet is used in
25740 preference to @samp{k} when multiprocess protocol extensions are
25741 supported; see @ref{multiprocess extensions}.
25751 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
25752 @cindex @samp{vRun} packet
25753 Run the program @var{filename}, passing it each @var{argument} on its
25754 command line. The file and arguments are hex-encoded strings. If
25755 @var{filename} is an empty string, the stub may use a default program
25756 (e.g.@: the last program run). The program is created in the stopped
25759 @c FIXME: What about non-stop mode?
25761 This packet is only available in extended mode (@pxref{extended mode}).
25767 @item @r{Any stop packet}
25768 for success (@pxref{Stop Reply Packets})
25772 @anchor{vStopped packet}
25773 @cindex @samp{vStopped} packet
25775 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
25776 reply and prompt for the stub to report another one.
25780 @item @r{Any stop packet}
25781 if there is another unreported stop event (@pxref{Stop Reply Packets})
25783 if there are no unreported stop events
25786 @item X @var{addr},@var{length}:@var{XX@dots{}}
25788 @cindex @samp{X} packet
25789 Write data to memory, where the data is transmitted in binary.
25790 @var{addr} is address, @var{length} is number of bytes,
25791 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
25801 @item z @var{type},@var{addr},@var{length}
25802 @itemx Z @var{type},@var{addr},@var{length}
25803 @anchor{insert breakpoint or watchpoint packet}
25804 @cindex @samp{z} packet
25805 @cindex @samp{Z} packets
25806 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
25807 watchpoint starting at address @var{address} and covering the next
25808 @var{length} bytes.
25810 Each breakpoint and watchpoint packet @var{type} is documented
25813 @emph{Implementation notes: A remote target shall return an empty string
25814 for an unrecognized breakpoint or watchpoint packet @var{type}. A
25815 remote target shall support either both or neither of a given
25816 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
25817 avoid potential problems with duplicate packets, the operations should
25818 be implemented in an idempotent way.}
25820 @item z0,@var{addr},@var{length}
25821 @itemx Z0,@var{addr},@var{length}
25822 @cindex @samp{z0} packet
25823 @cindex @samp{Z0} packet
25824 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
25825 @var{addr} of size @var{length}.
25827 A memory breakpoint is implemented by replacing the instruction at
25828 @var{addr} with a software breakpoint or trap instruction. The
25829 @var{length} is used by targets that indicates the size of the
25830 breakpoint (in bytes) that should be inserted (e.g., the @sc{arm} and
25831 @sc{mips} can insert either a 2 or 4 byte breakpoint).
25833 @emph{Implementation note: It is possible for a target to copy or move
25834 code that contains memory breakpoints (e.g., when implementing
25835 overlays). The behavior of this packet, in the presence of such a
25836 target, is not defined.}
25848 @item z1,@var{addr},@var{length}
25849 @itemx Z1,@var{addr},@var{length}
25850 @cindex @samp{z1} packet
25851 @cindex @samp{Z1} packet
25852 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
25853 address @var{addr} of size @var{length}.
25855 A hardware breakpoint is implemented using a mechanism that is not
25856 dependant on being able to modify the target's memory.
25858 @emph{Implementation note: A hardware breakpoint is not affected by code
25871 @item z2,@var{addr},@var{length}
25872 @itemx Z2,@var{addr},@var{length}
25873 @cindex @samp{z2} packet
25874 @cindex @samp{Z2} packet
25875 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint.
25887 @item z3,@var{addr},@var{length}
25888 @itemx Z3,@var{addr},@var{length}
25889 @cindex @samp{z3} packet
25890 @cindex @samp{Z3} packet
25891 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint.
25903 @item z4,@var{addr},@var{length}
25904 @itemx Z4,@var{addr},@var{length}
25905 @cindex @samp{z4} packet
25906 @cindex @samp{Z4} packet
25907 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint.
25921 @node Stop Reply Packets
25922 @section Stop Reply Packets
25923 @cindex stop reply packets
25925 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
25926 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
25927 receive any of the below as a reply. Except for @samp{?}
25928 and @samp{vStopped}, that reply is only returned
25929 when the target halts. In the below the exact meaning of @dfn{signal
25930 number} is defined by the header @file{include/gdb/signals.h} in the
25931 @value{GDBN} source code.
25933 As in the description of request packets, we include spaces in the
25934 reply templates for clarity; these are not part of the reply packet's
25935 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
25941 The program received signal number @var{AA} (a two-digit hexadecimal
25942 number). This is equivalent to a @samp{T} response with no
25943 @var{n}:@var{r} pairs.
25945 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
25946 @cindex @samp{T} packet reply
25947 The program received signal number @var{AA} (a two-digit hexadecimal
25948 number). This is equivalent to an @samp{S} response, except that the
25949 @samp{@var{n}:@var{r}} pairs can carry values of important registers
25950 and other information directly in the stop reply packet, reducing
25951 round-trip latency. Single-step and breakpoint traps are reported
25952 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
25956 If @var{n} is a hexadecimal number, it is a register number, and the
25957 corresponding @var{r} gives that register's value. @var{r} is a
25958 series of bytes in target byte order, with each byte given by a
25959 two-digit hex number.
25962 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
25963 the stopped thread, as specified in @ref{thread-id syntax}.
25966 If @var{n} is a recognized @dfn{stop reason}, it describes a more
25967 specific event that stopped the target. The currently defined stop
25968 reasons are listed below. @var{aa} should be @samp{05}, the trap
25969 signal. At most one stop reason should be present.
25972 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
25973 and go on to the next; this allows us to extend the protocol in the
25977 The currently defined stop reasons are:
25983 The packet indicates a watchpoint hit, and @var{r} is the data address, in
25986 @cindex shared library events, remote reply
25988 The packet indicates that the loaded libraries have changed.
25989 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
25990 list of loaded libraries. @var{r} is ignored.
25992 @cindex replay log events, remote reply
25994 The packet indicates that the target cannot continue replaying
25995 logged execution events, because it has reached the end (or the
25996 beginning when executing backward) of the log. The value of @var{r}
25997 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
25998 for more information.
26004 @itemx W @var{AA} ; process:@var{pid}
26005 The process exited, and @var{AA} is the exit status. This is only
26006 applicable to certain targets.
26008 The second form of the response, including the process ID of the exited
26009 process, can be used only when @value{GDBN} has reported support for
26010 multiprocess protocol extensions; see @ref{multiprocess extensions}.
26011 The @var{pid} is formatted as a big-endian hex string.
26014 @itemx X @var{AA} ; process:@var{pid}
26015 The process terminated with signal @var{AA}.
26017 The second form of the response, including the process ID of the
26018 terminated process, can be used only when @value{GDBN} has reported
26019 support for multiprocess protocol extensions; see @ref{multiprocess
26020 extensions}. The @var{pid} is formatted as a big-endian hex string.
26022 @item O @var{XX}@dots{}
26023 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
26024 written as the program's console output. This can happen at any time
26025 while the program is running and the debugger should continue to wait
26026 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
26028 @item F @var{call-id},@var{parameter}@dots{}
26029 @var{call-id} is the identifier which says which host system call should
26030 be called. This is just the name of the function. Translation into the
26031 correct system call is only applicable as it's defined in @value{GDBN}.
26032 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
26035 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
26036 this very system call.
26038 The target replies with this packet when it expects @value{GDBN} to
26039 call a host system call on behalf of the target. @value{GDBN} replies
26040 with an appropriate @samp{F} packet and keeps up waiting for the next
26041 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
26042 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
26043 Protocol Extension}, for more details.
26047 @node General Query Packets
26048 @section General Query Packets
26049 @cindex remote query requests
26051 Packets starting with @samp{q} are @dfn{general query packets};
26052 packets starting with @samp{Q} are @dfn{general set packets}. General
26053 query and set packets are a semi-unified form for retrieving and
26054 sending information to and from the stub.
26056 The initial letter of a query or set packet is followed by a name
26057 indicating what sort of thing the packet applies to. For example,
26058 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
26059 definitions with the stub. These packet names follow some
26064 The name must not contain commas, colons or semicolons.
26066 Most @value{GDBN} query and set packets have a leading upper case
26069 The names of custom vendor packets should use a company prefix, in
26070 lower case, followed by a period. For example, packets designed at
26071 the Acme Corporation might begin with @samp{qacme.foo} (for querying
26072 foos) or @samp{Qacme.bar} (for setting bars).
26075 The name of a query or set packet should be separated from any
26076 parameters by a @samp{:}; the parameters themselves should be
26077 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
26078 full packet name, and check for a separator or the end of the packet,
26079 in case two packet names share a common prefix. New packets should not begin
26080 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
26081 packets predate these conventions, and have arguments without any terminator
26082 for the packet name; we suspect they are in widespread use in places that
26083 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
26084 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
26087 Like the descriptions of the other packets, each description here
26088 has a template showing the packet's overall syntax, followed by an
26089 explanation of the packet's meaning. We include spaces in some of the
26090 templates for clarity; these are not part of the packet's syntax. No
26091 @value{GDBN} packet uses spaces to separate its components.
26093 Here are the currently defined query and set packets:
26098 @cindex current thread, remote request
26099 @cindex @samp{qC} packet
26100 Return the current thread ID.
26104 @item QC @var{thread-id}
26105 Where @var{thread-id} is a thread ID as documented in
26106 @ref{thread-id syntax}.
26107 @item @r{(anything else)}
26108 Any other reply implies the old thread ID.
26111 @item qCRC:@var{addr},@var{length}
26112 @cindex CRC of memory block, remote request
26113 @cindex @samp{qCRC} packet
26114 Compute the CRC checksum of a block of memory.
26118 An error (such as memory fault)
26119 @item C @var{crc32}
26120 The specified memory region's checksum is @var{crc32}.
26124 @itemx qsThreadInfo
26125 @cindex list active threads, remote request
26126 @cindex @samp{qfThreadInfo} packet
26127 @cindex @samp{qsThreadInfo} packet
26128 Obtain a list of all active thread IDs from the target (OS). Since there
26129 may be too many active threads to fit into one reply packet, this query
26130 works iteratively: it may require more than one query/reply sequence to
26131 obtain the entire list of threads. The first query of the sequence will
26132 be the @samp{qfThreadInfo} query; subsequent queries in the
26133 sequence will be the @samp{qsThreadInfo} query.
26135 NOTE: This packet replaces the @samp{qL} query (see below).
26139 @item m @var{thread-id}
26141 @item m @var{thread-id},@var{thread-id}@dots{}
26142 a comma-separated list of thread IDs
26144 (lower case letter @samp{L}) denotes end of list.
26147 In response to each query, the target will reply with a list of one or
26148 more thread IDs, separated by commas.
26149 @value{GDBN} will respond to each reply with a request for more thread
26150 ids (using the @samp{qs} form of the query), until the target responds
26151 with @samp{l} (lower-case el, for @dfn{last}).
26152 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
26155 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
26156 @cindex get thread-local storage address, remote request
26157 @cindex @samp{qGetTLSAddr} packet
26158 Fetch the address associated with thread local storage specified
26159 by @var{thread-id}, @var{offset}, and @var{lm}.
26161 @var{thread-id} is the thread ID associated with the
26162 thread for which to fetch the TLS address. @xref{thread-id syntax}.
26164 @var{offset} is the (big endian, hex encoded) offset associated with the
26165 thread local variable. (This offset is obtained from the debug
26166 information associated with the variable.)
26168 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
26169 the load module associated with the thread local storage. For example,
26170 a @sc{gnu}/Linux system will pass the link map address of the shared
26171 object associated with the thread local storage under consideration.
26172 Other operating environments may choose to represent the load module
26173 differently, so the precise meaning of this parameter will vary.
26177 @item @var{XX}@dots{}
26178 Hex encoded (big endian) bytes representing the address of the thread
26179 local storage requested.
26182 An error occurred. @var{nn} are hex digits.
26185 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
26188 @item qL @var{startflag} @var{threadcount} @var{nextthread}
26189 Obtain thread information from RTOS. Where: @var{startflag} (one hex
26190 digit) is one to indicate the first query and zero to indicate a
26191 subsequent query; @var{threadcount} (two hex digits) is the maximum
26192 number of threads the response packet can contain; and @var{nextthread}
26193 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
26194 returned in the response as @var{argthread}.
26196 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
26200 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
26201 Where: @var{count} (two hex digits) is the number of threads being
26202 returned; @var{done} (one hex digit) is zero to indicate more threads
26203 and one indicates no further threads; @var{argthreadid} (eight hex
26204 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
26205 is a sequence of thread IDs from the target. @var{threadid} (eight hex
26206 digits). See @code{remote.c:parse_threadlist_response()}.
26210 @cindex section offsets, remote request
26211 @cindex @samp{qOffsets} packet
26212 Get section offsets that the target used when relocating the downloaded
26217 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
26218 Relocate the @code{Text} section by @var{xxx} from its original address.
26219 Relocate the @code{Data} section by @var{yyy} from its original address.
26220 If the object file format provides segment information (e.g.@: @sc{elf}
26221 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
26222 segments by the supplied offsets.
26224 @emph{Note: while a @code{Bss} offset may be included in the response,
26225 @value{GDBN} ignores this and instead applies the @code{Data} offset
26226 to the @code{Bss} section.}
26228 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
26229 Relocate the first segment of the object file, which conventionally
26230 contains program code, to a starting address of @var{xxx}. If
26231 @samp{DataSeg} is specified, relocate the second segment, which
26232 conventionally contains modifiable data, to a starting address of
26233 @var{yyy}. @value{GDBN} will report an error if the object file
26234 does not contain segment information, or does not contain at least
26235 as many segments as mentioned in the reply. Extra segments are
26236 kept at fixed offsets relative to the last relocated segment.
26239 @item qP @var{mode} @var{thread-id}
26240 @cindex thread information, remote request
26241 @cindex @samp{qP} packet
26242 Returns information on @var{thread-id}. Where: @var{mode} is a hex
26243 encoded 32 bit mode; @var{thread-id} is a thread ID
26244 (@pxref{thread-id syntax}).
26246 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
26249 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
26253 @cindex non-stop mode, remote request
26254 @cindex @samp{QNonStop} packet
26256 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
26257 @xref{Remote Non-Stop}, for more information.
26262 The request succeeded.
26265 An error occurred. @var{nn} are hex digits.
26268 An empty reply indicates that @samp{QNonStop} is not supported by
26272 This packet is not probed by default; the remote stub must request it,
26273 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
26274 Use of this packet is controlled by the @code{set non-stop} command;
26275 @pxref{Non-Stop Mode}.
26277 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
26278 @cindex pass signals to inferior, remote request
26279 @cindex @samp{QPassSignals} packet
26280 @anchor{QPassSignals}
26281 Each listed @var{signal} should be passed directly to the inferior process.
26282 Signals are numbered identically to continue packets and stop replies
26283 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
26284 strictly greater than the previous item. These signals do not need to stop
26285 the inferior, or be reported to @value{GDBN}. All other signals should be
26286 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
26287 combine; any earlier @samp{QPassSignals} list is completely replaced by the
26288 new list. This packet improves performance when using @samp{handle
26289 @var{signal} nostop noprint pass}.
26294 The request succeeded.
26297 An error occurred. @var{nn} are hex digits.
26300 An empty reply indicates that @samp{QPassSignals} is not supported by
26304 Use of this packet is controlled by the @code{set remote pass-signals}
26305 command (@pxref{Remote Configuration, set remote pass-signals}).
26306 This packet is not probed by default; the remote stub must request it,
26307 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
26309 @item qRcmd,@var{command}
26310 @cindex execute remote command, remote request
26311 @cindex @samp{qRcmd} packet
26312 @var{command} (hex encoded) is passed to the local interpreter for
26313 execution. Invalid commands should be reported using the output
26314 string. Before the final result packet, the target may also respond
26315 with a number of intermediate @samp{O@var{output}} console output
26316 packets. @emph{Implementors should note that providing access to a
26317 stubs's interpreter may have security implications}.
26322 A command response with no output.
26324 A command response with the hex encoded output string @var{OUTPUT}.
26326 Indicate a badly formed request.
26328 An empty reply indicates that @samp{qRcmd} is not recognized.
26331 (Note that the @code{qRcmd} packet's name is separated from the
26332 command by a @samp{,}, not a @samp{:}, contrary to the naming
26333 conventions above. Please don't use this packet as a model for new
26336 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
26337 @cindex searching memory, in remote debugging
26338 @cindex @samp{qSearch:memory} packet
26339 @anchor{qSearch memory}
26340 Search @var{length} bytes at @var{address} for @var{search-pattern}.
26341 @var{address} and @var{length} are encoded in hex.
26342 @var{search-pattern} is a sequence of bytes, hex encoded.
26347 The pattern was not found.
26349 The pattern was found at @var{address}.
26351 A badly formed request or an error was encountered while searching memory.
26353 An empty reply indicates that @samp{qSearch:memory} is not recognized.
26356 @item QStartNoAckMode
26357 @cindex @samp{QStartNoAckMode} packet
26358 @anchor{QStartNoAckMode}
26359 Request that the remote stub disable the normal @samp{+}/@samp{-}
26360 protocol acknowledgments (@pxref{Packet Acknowledgment}).
26365 The stub has switched to no-acknowledgment mode.
26366 @value{GDBN} acknowledges this reponse,
26367 but neither the stub nor @value{GDBN} shall send or expect further
26368 @samp{+}/@samp{-} acknowledgments in the current connection.
26370 An empty reply indicates that the stub does not support no-acknowledgment mode.
26373 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
26374 @cindex supported packets, remote query
26375 @cindex features of the remote protocol
26376 @cindex @samp{qSupported} packet
26377 @anchor{qSupported}
26378 Tell the remote stub about features supported by @value{GDBN}, and
26379 query the stub for features it supports. This packet allows
26380 @value{GDBN} and the remote stub to take advantage of each others'
26381 features. @samp{qSupported} also consolidates multiple feature probes
26382 at startup, to improve @value{GDBN} performance---a single larger
26383 packet performs better than multiple smaller probe packets on
26384 high-latency links. Some features may enable behavior which must not
26385 be on by default, e.g.@: because it would confuse older clients or
26386 stubs. Other features may describe packets which could be
26387 automatically probed for, but are not. These features must be
26388 reported before @value{GDBN} will use them. This ``default
26389 unsupported'' behavior is not appropriate for all packets, but it
26390 helps to keep the initial connection time under control with new
26391 versions of @value{GDBN} which support increasing numbers of packets.
26395 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
26396 The stub supports or does not support each returned @var{stubfeature},
26397 depending on the form of each @var{stubfeature} (see below for the
26400 An empty reply indicates that @samp{qSupported} is not recognized,
26401 or that no features needed to be reported to @value{GDBN}.
26404 The allowed forms for each feature (either a @var{gdbfeature} in the
26405 @samp{qSupported} packet, or a @var{stubfeature} in the response)
26409 @item @var{name}=@var{value}
26410 The remote protocol feature @var{name} is supported, and associated
26411 with the specified @var{value}. The format of @var{value} depends
26412 on the feature, but it must not include a semicolon.
26414 The remote protocol feature @var{name} is supported, and does not
26415 need an associated value.
26417 The remote protocol feature @var{name} is not supported.
26419 The remote protocol feature @var{name} may be supported, and
26420 @value{GDBN} should auto-detect support in some other way when it is
26421 needed. This form will not be used for @var{gdbfeature} notifications,
26422 but may be used for @var{stubfeature} responses.
26425 Whenever the stub receives a @samp{qSupported} request, the
26426 supplied set of @value{GDBN} features should override any previous
26427 request. This allows @value{GDBN} to put the stub in a known
26428 state, even if the stub had previously been communicating with
26429 a different version of @value{GDBN}.
26431 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
26436 This feature indicates whether @value{GDBN} supports multiprocess
26437 extensions to the remote protocol. @value{GDBN} does not use such
26438 extensions unless the stub also reports that it supports them by
26439 including @samp{multiprocess+} in its @samp{qSupported} reply.
26440 @xref{multiprocess extensions}, for details.
26443 Stubs should ignore any unknown values for
26444 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
26445 packet supports receiving packets of unlimited length (earlier
26446 versions of @value{GDBN} may reject overly long responses). Additional values
26447 for @var{gdbfeature} may be defined in the future to let the stub take
26448 advantage of new features in @value{GDBN}, e.g.@: incompatible
26449 improvements in the remote protocol---the @samp{multiprocess} feature is
26450 an example of such a feature. The stub's reply should be independent
26451 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
26452 describes all the features it supports, and then the stub replies with
26453 all the features it supports.
26455 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
26456 responses, as long as each response uses one of the standard forms.
26458 Some features are flags. A stub which supports a flag feature
26459 should respond with a @samp{+} form response. Other features
26460 require values, and the stub should respond with an @samp{=}
26463 Each feature has a default value, which @value{GDBN} will use if
26464 @samp{qSupported} is not available or if the feature is not mentioned
26465 in the @samp{qSupported} response. The default values are fixed; a
26466 stub is free to omit any feature responses that match the defaults.
26468 Not all features can be probed, but for those which can, the probing
26469 mechanism is useful: in some cases, a stub's internal
26470 architecture may not allow the protocol layer to know some information
26471 about the underlying target in advance. This is especially common in
26472 stubs which may be configured for multiple targets.
26474 These are the currently defined stub features and their properties:
26476 @multitable @columnfractions 0.35 0.2 0.12 0.2
26477 @c NOTE: The first row should be @headitem, but we do not yet require
26478 @c a new enough version of Texinfo (4.7) to use @headitem.
26480 @tab Value Required
26484 @item @samp{PacketSize}
26489 @item @samp{qXfer:auxv:read}
26494 @item @samp{qXfer:features:read}
26499 @item @samp{qXfer:libraries:read}
26504 @item @samp{qXfer:memory-map:read}
26509 @item @samp{qXfer:spu:read}
26514 @item @samp{qXfer:spu:write}
26519 @item @samp{QNonStop}
26524 @item @samp{QPassSignals}
26529 @item @samp{QStartNoAckMode}
26534 @item @samp{multiprocess}
26541 These are the currently defined stub features, in more detail:
26544 @cindex packet size, remote protocol
26545 @item PacketSize=@var{bytes}
26546 The remote stub can accept packets up to at least @var{bytes} in
26547 length. @value{GDBN} will send packets up to this size for bulk
26548 transfers, and will never send larger packets. This is a limit on the
26549 data characters in the packet, including the frame and checksum.
26550 There is no trailing NUL byte in a remote protocol packet; if the stub
26551 stores packets in a NUL-terminated format, it should allow an extra
26552 byte in its buffer for the NUL. If this stub feature is not supported,
26553 @value{GDBN} guesses based on the size of the @samp{g} packet response.
26555 @item qXfer:auxv:read
26556 The remote stub understands the @samp{qXfer:auxv:read} packet
26557 (@pxref{qXfer auxiliary vector read}).
26559 @item qXfer:features:read
26560 The remote stub understands the @samp{qXfer:features:read} packet
26561 (@pxref{qXfer target description read}).
26563 @item qXfer:libraries:read
26564 The remote stub understands the @samp{qXfer:libraries:read} packet
26565 (@pxref{qXfer library list read}).
26567 @item qXfer:memory-map:read
26568 The remote stub understands the @samp{qXfer:memory-map:read} packet
26569 (@pxref{qXfer memory map read}).
26571 @item qXfer:spu:read
26572 The remote stub understands the @samp{qXfer:spu:read} packet
26573 (@pxref{qXfer spu read}).
26575 @item qXfer:spu:write
26576 The remote stub understands the @samp{qXfer:spu:write} packet
26577 (@pxref{qXfer spu write}).
26580 The remote stub understands the @samp{QNonStop} packet
26581 (@pxref{QNonStop}).
26584 The remote stub understands the @samp{QPassSignals} packet
26585 (@pxref{QPassSignals}).
26587 @item QStartNoAckMode
26588 The remote stub understands the @samp{QStartNoAckMode} packet and
26589 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
26592 @anchor{multiprocess extensions}
26593 @cindex multiprocess extensions, in remote protocol
26594 The remote stub understands the multiprocess extensions to the remote
26595 protocol syntax. The multiprocess extensions affect the syntax of
26596 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
26597 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
26598 replies. Note that reporting this feature indicates support for the
26599 syntactic extensions only, not that the stub necessarily supports
26600 debugging of more than one process at a time. The stub must not use
26601 multiprocess extensions in packet replies unless @value{GDBN} has also
26602 indicated it supports them in its @samp{qSupported} request.
26604 @item qXfer:osdata:read
26605 The remote stub understands the @samp{qXfer:osdata:read} packet
26606 ((@pxref{qXfer osdata read}).
26611 @cindex symbol lookup, remote request
26612 @cindex @samp{qSymbol} packet
26613 Notify the target that @value{GDBN} is prepared to serve symbol lookup
26614 requests. Accept requests from the target for the values of symbols.
26619 The target does not need to look up any (more) symbols.
26620 @item qSymbol:@var{sym_name}
26621 The target requests the value of symbol @var{sym_name} (hex encoded).
26622 @value{GDBN} may provide the value by using the
26623 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
26627 @item qSymbol:@var{sym_value}:@var{sym_name}
26628 Set the value of @var{sym_name} to @var{sym_value}.
26630 @var{sym_name} (hex encoded) is the name of a symbol whose value the
26631 target has previously requested.
26633 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
26634 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
26640 The target does not need to look up any (more) symbols.
26641 @item qSymbol:@var{sym_name}
26642 The target requests the value of a new symbol @var{sym_name} (hex
26643 encoded). @value{GDBN} will continue to supply the values of symbols
26644 (if available), until the target ceases to request them.
26649 @xref{Tracepoint Packets}.
26651 @item qThreadExtraInfo,@var{thread-id}
26652 @cindex thread attributes info, remote request
26653 @cindex @samp{qThreadExtraInfo} packet
26654 Obtain a printable string description of a thread's attributes from
26655 the target OS. @var{thread-id} is a thread ID;
26656 see @ref{thread-id syntax}. This
26657 string may contain anything that the target OS thinks is interesting
26658 for @value{GDBN} to tell the user about the thread. The string is
26659 displayed in @value{GDBN}'s @code{info threads} display. Some
26660 examples of possible thread extra info strings are @samp{Runnable}, or
26661 @samp{Blocked on Mutex}.
26665 @item @var{XX}@dots{}
26666 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
26667 comprising the printable string containing the extra information about
26668 the thread's attributes.
26671 (Note that the @code{qThreadExtraInfo} packet's name is separated from
26672 the command by a @samp{,}, not a @samp{:}, contrary to the naming
26673 conventions above. Please don't use this packet as a model for new
26681 @xref{Tracepoint Packets}.
26683 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
26684 @cindex read special object, remote request
26685 @cindex @samp{qXfer} packet
26686 @anchor{qXfer read}
26687 Read uninterpreted bytes from the target's special data area
26688 identified by the keyword @var{object}. Request @var{length} bytes
26689 starting at @var{offset} bytes into the data. The content and
26690 encoding of @var{annex} is specific to @var{object}; it can supply
26691 additional details about what data to access.
26693 Here are the specific requests of this form defined so far. All
26694 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
26695 formats, listed below.
26698 @item qXfer:auxv:read::@var{offset},@var{length}
26699 @anchor{qXfer auxiliary vector read}
26700 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
26701 auxiliary vector}. Note @var{annex} must be empty.
26703 This packet is not probed by default; the remote stub must request it,
26704 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
26706 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
26707 @anchor{qXfer target description read}
26708 Access the @dfn{target description}. @xref{Target Descriptions}. The
26709 annex specifies which XML document to access. The main description is
26710 always loaded from the @samp{target.xml} annex.
26712 This packet is not probed by default; the remote stub must request it,
26713 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
26715 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
26716 @anchor{qXfer library list read}
26717 Access the target's list of loaded libraries. @xref{Library List Format}.
26718 The annex part of the generic @samp{qXfer} packet must be empty
26719 (@pxref{qXfer read}).
26721 Targets which maintain a list of libraries in the program's memory do
26722 not need to implement this packet; it is designed for platforms where
26723 the operating system manages the list of loaded libraries.
26725 This packet is not probed by default; the remote stub must request it,
26726 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
26728 @item qXfer:memory-map:read::@var{offset},@var{length}
26729 @anchor{qXfer memory map read}
26730 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
26731 annex part of the generic @samp{qXfer} packet must be empty
26732 (@pxref{qXfer read}).
26734 This packet is not probed by default; the remote stub must request it,
26735 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
26737 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
26738 @anchor{qXfer spu read}
26739 Read contents of an @code{spufs} file on the target system. The
26740 annex specifies which file to read; it must be of the form
26741 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
26742 in the target process, and @var{name} identifes the @code{spufs} file
26743 in that context to be accessed.
26745 This packet is not probed by default; the remote stub must request it,
26746 by supplying an appropriate @samp{qSupported} response
26747 (@pxref{qSupported}).
26749 @item qXfer:osdata:read::@var{offset},@var{length}
26750 @anchor{qXfer osdata read}
26751 Access the target's @dfn{operating system information}.
26752 @xref{Operating System Information}.
26759 Data @var{data} (@pxref{Binary Data}) has been read from the
26760 target. There may be more data at a higher address (although
26761 it is permitted to return @samp{m} even for the last valid
26762 block of data, as long as at least one byte of data was read).
26763 @var{data} may have fewer bytes than the @var{length} in the
26767 Data @var{data} (@pxref{Binary Data}) has been read from the target.
26768 There is no more data to be read. @var{data} may have fewer bytes
26769 than the @var{length} in the request.
26772 The @var{offset} in the request is at the end of the data.
26773 There is no more data to be read.
26776 The request was malformed, or @var{annex} was invalid.
26779 The offset was invalid, or there was an error encountered reading the data.
26780 @var{nn} is a hex-encoded @code{errno} value.
26783 An empty reply indicates the @var{object} string was not recognized by
26784 the stub, or that the object does not support reading.
26787 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
26788 @cindex write data into object, remote request
26789 Write uninterpreted bytes into the target's special data area
26790 identified by the keyword @var{object}, starting at @var{offset} bytes
26791 into the data. @var{data}@dots{} is the binary-encoded data
26792 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
26793 is specific to @var{object}; it can supply additional details about what data
26796 Here are the specific requests of this form defined so far. All
26797 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
26798 formats, listed below.
26801 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
26802 @anchor{qXfer spu write}
26803 Write @var{data} to an @code{spufs} file on the target system. The
26804 annex specifies which file to write; it must be of the form
26805 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
26806 in the target process, and @var{name} identifes the @code{spufs} file
26807 in that context to be accessed.
26809 This packet is not probed by default; the remote stub must request it,
26810 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
26816 @var{nn} (hex encoded) is the number of bytes written.
26817 This may be fewer bytes than supplied in the request.
26820 The request was malformed, or @var{annex} was invalid.
26823 The offset was invalid, or there was an error encountered writing the data.
26824 @var{nn} is a hex-encoded @code{errno} value.
26827 An empty reply indicates the @var{object} string was not
26828 recognized by the stub, or that the object does not support writing.
26831 @item qXfer:@var{object}:@var{operation}:@dots{}
26832 Requests of this form may be added in the future. When a stub does
26833 not recognize the @var{object} keyword, or its support for
26834 @var{object} does not recognize the @var{operation} keyword, the stub
26835 must respond with an empty packet.
26839 @node Register Packet Format
26840 @section Register Packet Format
26842 The following @code{g}/@code{G} packets have previously been defined.
26843 In the below, some thirty-two bit registers are transferred as
26844 sixty-four bits. Those registers should be zero/sign extended (which?)
26845 to fill the space allocated. Register bytes are transferred in target
26846 byte order. The two nibbles within a register byte are transferred
26847 most-significant - least-significant.
26853 All registers are transferred as thirty-two bit quantities in the order:
26854 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
26855 registers; fsr; fir; fp.
26859 All registers are transferred as sixty-four bit quantities (including
26860 thirty-two bit registers such as @code{sr}). The ordering is the same
26865 @node Tracepoint Packets
26866 @section Tracepoint Packets
26867 @cindex tracepoint packets
26868 @cindex packets, tracepoint
26870 Here we describe the packets @value{GDBN} uses to implement
26871 tracepoints (@pxref{Tracepoints}).
26875 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}@r{[}-@r{]}
26876 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
26877 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
26878 the tracepoint is disabled. @var{step} is the tracepoint's step
26879 count, and @var{pass} is its pass count. If the trailing @samp{-} is
26880 present, further @samp{QTDP} packets will follow to specify this
26881 tracepoint's actions.
26886 The packet was understood and carried out.
26888 The packet was not recognized.
26891 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
26892 Define actions to be taken when a tracepoint is hit. @var{n} and
26893 @var{addr} must be the same as in the initial @samp{QTDP} packet for
26894 this tracepoint. This packet may only be sent immediately after
26895 another @samp{QTDP} packet that ended with a @samp{-}. If the
26896 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
26897 specifying more actions for this tracepoint.
26899 In the series of action packets for a given tracepoint, at most one
26900 can have an @samp{S} before its first @var{action}. If such a packet
26901 is sent, it and the following packets define ``while-stepping''
26902 actions. Any prior packets define ordinary actions --- that is, those
26903 taken when the tracepoint is first hit. If no action packet has an
26904 @samp{S}, then all the packets in the series specify ordinary
26905 tracepoint actions.
26907 The @samp{@var{action}@dots{}} portion of the packet is a series of
26908 actions, concatenated without separators. Each action has one of the
26914 Collect the registers whose bits are set in @var{mask}. @var{mask} is
26915 a hexadecimal number whose @var{i}'th bit is set if register number
26916 @var{i} should be collected. (The least significant bit is numbered
26917 zero.) Note that @var{mask} may be any number of digits long; it may
26918 not fit in a 32-bit word.
26920 @item M @var{basereg},@var{offset},@var{len}
26921 Collect @var{len} bytes of memory starting at the address in register
26922 number @var{basereg}, plus @var{offset}. If @var{basereg} is
26923 @samp{-1}, then the range has a fixed address: @var{offset} is the
26924 address of the lowest byte to collect. The @var{basereg},
26925 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
26926 values (the @samp{-1} value for @var{basereg} is a special case).
26928 @item X @var{len},@var{expr}
26929 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
26930 it directs. @var{expr} is an agent expression, as described in
26931 @ref{Agent Expressions}. Each byte of the expression is encoded as a
26932 two-digit hex number in the packet; @var{len} is the number of bytes
26933 in the expression (and thus one-half the number of hex digits in the
26938 Any number of actions may be packed together in a single @samp{QTDP}
26939 packet, as long as the packet does not exceed the maximum packet
26940 length (400 bytes, for many stubs). There may be only one @samp{R}
26941 action per tracepoint, and it must precede any @samp{M} or @samp{X}
26942 actions. Any registers referred to by @samp{M} and @samp{X} actions
26943 must be collected by a preceding @samp{R} action. (The
26944 ``while-stepping'' actions are treated as if they were attached to a
26945 separate tracepoint, as far as these restrictions are concerned.)
26950 The packet was understood and carried out.
26952 The packet was not recognized.
26955 @item QTFrame:@var{n}
26956 Select the @var{n}'th tracepoint frame from the buffer, and use the
26957 register and memory contents recorded there to answer subsequent
26958 request packets from @value{GDBN}.
26960 A successful reply from the stub indicates that the stub has found the
26961 requested frame. The response is a series of parts, concatenated
26962 without separators, describing the frame we selected. Each part has
26963 one of the following forms:
26967 The selected frame is number @var{n} in the trace frame buffer;
26968 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
26969 was no frame matching the criteria in the request packet.
26972 The selected trace frame records a hit of tracepoint number @var{t};
26973 @var{t} is a hexadecimal number.
26977 @item QTFrame:pc:@var{addr}
26978 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
26979 currently selected frame whose PC is @var{addr};
26980 @var{addr} is a hexadecimal number.
26982 @item QTFrame:tdp:@var{t}
26983 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
26984 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
26985 is a hexadecimal number.
26987 @item QTFrame:range:@var{start}:@var{end}
26988 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
26989 currently selected frame whose PC is between @var{start} (inclusive)
26990 and @var{end} (exclusive); @var{start} and @var{end} are hexadecimal
26993 @item QTFrame:outside:@var{start}:@var{end}
26994 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
26995 frame @emph{outside} the given range of addresses.
26998 Begin the tracepoint experiment. Begin collecting data from tracepoint
26999 hits in the trace frame buffer.
27002 End the tracepoint experiment. Stop collecting trace frames.
27005 Clear the table of tracepoints, and empty the trace frame buffer.
27007 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
27008 Establish the given ranges of memory as ``transparent''. The stub
27009 will answer requests for these ranges from memory's current contents,
27010 if they were not collected as part of the tracepoint hit.
27012 @value{GDBN} uses this to mark read-only regions of memory, like those
27013 containing program code. Since these areas never change, they should
27014 still have the same contents they did when the tracepoint was hit, so
27015 there's no reason for the stub to refuse to provide their contents.
27018 Ask the stub if there is a trace experiment running right now.
27023 There is no trace experiment running.
27025 There is a trace experiment running.
27031 @node Host I/O Packets
27032 @section Host I/O Packets
27033 @cindex Host I/O, remote protocol
27034 @cindex file transfer, remote protocol
27036 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
27037 operations on the far side of a remote link. For example, Host I/O is
27038 used to upload and download files to a remote target with its own
27039 filesystem. Host I/O uses the same constant values and data structure
27040 layout as the target-initiated File-I/O protocol. However, the
27041 Host I/O packets are structured differently. The target-initiated
27042 protocol relies on target memory to store parameters and buffers.
27043 Host I/O requests are initiated by @value{GDBN}, and the
27044 target's memory is not involved. @xref{File-I/O Remote Protocol
27045 Extension}, for more details on the target-initiated protocol.
27047 The Host I/O request packets all encode a single operation along with
27048 its arguments. They have this format:
27052 @item vFile:@var{operation}: @var{parameter}@dots{}
27053 @var{operation} is the name of the particular request; the target
27054 should compare the entire packet name up to the second colon when checking
27055 for a supported operation. The format of @var{parameter} depends on
27056 the operation. Numbers are always passed in hexadecimal. Negative
27057 numbers have an explicit minus sign (i.e.@: two's complement is not
27058 used). Strings (e.g.@: filenames) are encoded as a series of
27059 hexadecimal bytes. The last argument to a system call may be a
27060 buffer of escaped binary data (@pxref{Binary Data}).
27064 The valid responses to Host I/O packets are:
27068 @item F @var{result} [, @var{errno}] [; @var{attachment}]
27069 @var{result} is the integer value returned by this operation, usually
27070 non-negative for success and -1 for errors. If an error has occured,
27071 @var{errno} will be included in the result. @var{errno} will have a
27072 value defined by the File-I/O protocol (@pxref{Errno Values}). For
27073 operations which return data, @var{attachment} supplies the data as a
27074 binary buffer. Binary buffers in response packets are escaped in the
27075 normal way (@pxref{Binary Data}). See the individual packet
27076 documentation for the interpretation of @var{result} and
27080 An empty response indicates that this operation is not recognized.
27084 These are the supported Host I/O operations:
27087 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
27088 Open a file at @var{pathname} and return a file descriptor for it, or
27089 return -1 if an error occurs. @var{pathname} is a string,
27090 @var{flags} is an integer indicating a mask of open flags
27091 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
27092 of mode bits to use if the file is created (@pxref{mode_t Values}).
27093 @xref{open}, for details of the open flags and mode values.
27095 @item vFile:close: @var{fd}
27096 Close the open file corresponding to @var{fd} and return 0, or
27097 -1 if an error occurs.
27099 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
27100 Read data from the open file corresponding to @var{fd}. Up to
27101 @var{count} bytes will be read from the file, starting at @var{offset}
27102 relative to the start of the file. The target may read fewer bytes;
27103 common reasons include packet size limits and an end-of-file
27104 condition. The number of bytes read is returned. Zero should only be
27105 returned for a successful read at the end of the file, or if
27106 @var{count} was zero.
27108 The data read should be returned as a binary attachment on success.
27109 If zero bytes were read, the response should include an empty binary
27110 attachment (i.e.@: a trailing semicolon). The return value is the
27111 number of target bytes read; the binary attachment may be longer if
27112 some characters were escaped.
27114 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
27115 Write @var{data} (a binary buffer) to the open file corresponding
27116 to @var{fd}. Start the write at @var{offset} from the start of the
27117 file. Unlike many @code{write} system calls, there is no
27118 separate @var{count} argument; the length of @var{data} in the
27119 packet is used. @samp{vFile:write} returns the number of bytes written,
27120 which may be shorter than the length of @var{data}, or -1 if an
27123 @item vFile:unlink: @var{pathname}
27124 Delete the file at @var{pathname} on the target. Return 0,
27125 or -1 if an error occurs. @var{pathname} is a string.
27130 @section Interrupts
27131 @cindex interrupts (remote protocol)
27133 When a program on the remote target is running, @value{GDBN} may
27134 attempt to interrupt it by sending a @samp{Ctrl-C} or a @code{BREAK},
27135 control of which is specified via @value{GDBN}'s @samp{remotebreak}
27136 setting (@pxref{set remotebreak}).
27138 The precise meaning of @code{BREAK} is defined by the transport
27139 mechanism and may, in fact, be undefined. @value{GDBN} does not
27140 currently define a @code{BREAK} mechanism for any of the network
27141 interfaces except for TCP, in which case @value{GDBN} sends the
27142 @code{telnet} BREAK sequence.
27144 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
27145 transport mechanisms. It is represented by sending the single byte
27146 @code{0x03} without any of the usual packet overhead described in
27147 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
27148 transmitted as part of a packet, it is considered to be packet data
27149 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
27150 (@pxref{X packet}), used for binary downloads, may include an unescaped
27151 @code{0x03} as part of its packet.
27153 Stubs are not required to recognize these interrupt mechanisms and the
27154 precise meaning associated with receipt of the interrupt is
27155 implementation defined. If the target supports debugging of multiple
27156 threads and/or processes, it should attempt to interrupt all
27157 currently-executing threads and processes.
27158 If the stub is successful at interrupting the
27159 running program, it should send one of the stop
27160 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
27161 of successfully stopping the program in all-stop mode, and a stop reply
27162 for each stopped thread in non-stop mode.
27163 Interrupts received while the
27164 program is stopped are discarded.
27166 @node Notification Packets
27167 @section Notification Packets
27168 @cindex notification packets
27169 @cindex packets, notification
27171 The @value{GDBN} remote serial protocol includes @dfn{notifications},
27172 packets that require no acknowledgment. Both the GDB and the stub
27173 may send notifications (although the only notifications defined at
27174 present are sent by the stub). Notifications carry information
27175 without incurring the round-trip latency of an acknowledgment, and so
27176 are useful for low-impact communications where occasional packet loss
27179 A notification packet has the form @samp{% @var{data} #
27180 @var{checksum}}, where @var{data} is the content of the notification,
27181 and @var{checksum} is a checksum of @var{data}, computed and formatted
27182 as for ordinary @value{GDBN} packets. A notification's @var{data}
27183 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
27184 receiving a notification, the recipient sends no @samp{+} or @samp{-}
27185 to acknowledge the notification's receipt or to report its corruption.
27187 Every notification's @var{data} begins with a name, which contains no
27188 colon characters, followed by a colon character.
27190 Recipients should silently ignore corrupted notifications and
27191 notifications they do not understand. Recipients should restart
27192 timeout periods on receipt of a well-formed notification, whether or
27193 not they understand it.
27195 Senders should only send the notifications described here when this
27196 protocol description specifies that they are permitted. In the
27197 future, we may extend the protocol to permit existing notifications in
27198 new contexts; this rule helps older senders avoid confusing newer
27201 (Older versions of @value{GDBN} ignore bytes received until they see
27202 the @samp{$} byte that begins an ordinary packet, so new stubs may
27203 transmit notifications without fear of confusing older clients. There
27204 are no notifications defined for @value{GDBN} to send at the moment, but we
27205 assume that most older stubs would ignore them, as well.)
27207 The following notification packets from the stub to @value{GDBN} are
27211 @item Stop: @var{reply}
27212 Report an asynchronous stop event in non-stop mode.
27213 The @var{reply} has the form of a stop reply, as
27214 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
27215 for information on how these notifications are acknowledged by
27219 @node Remote Non-Stop
27220 @section Remote Protocol Support for Non-Stop Mode
27222 @value{GDBN}'s remote protocol supports non-stop debugging of
27223 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
27224 supports non-stop mode, it should report that to @value{GDBN} by including
27225 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
27227 @value{GDBN} typically sends a @samp{QNonStop} packet only when
27228 establishing a new connection with the stub. Entering non-stop mode
27229 does not alter the state of any currently-running threads, but targets
27230 must stop all threads in any already-attached processes when entering
27231 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
27232 probe the target state after a mode change.
27234 In non-stop mode, when an attached process encounters an event that
27235 would otherwise be reported with a stop reply, it uses the
27236 asynchronous notification mechanism (@pxref{Notification Packets}) to
27237 inform @value{GDBN}. In contrast to all-stop mode, where all threads
27238 in all processes are stopped when a stop reply is sent, in non-stop
27239 mode only the thread reporting the stop event is stopped. That is,
27240 when reporting a @samp{S} or @samp{T} response to indicate completion
27241 of a step operation, hitting a breakpoint, or a fault, only the
27242 affected thread is stopped; any other still-running threads continue
27243 to run. When reporting a @samp{W} or @samp{X} response, all running
27244 threads belonging to other attached processes continue to run.
27246 Only one stop reply notification at a time may be pending; if
27247 additional stop events occur before @value{GDBN} has acknowledged the
27248 previous notification, they must be queued by the stub for later
27249 synchronous transmission in response to @samp{vStopped} packets from
27250 @value{GDBN}. Because the notification mechanism is unreliable,
27251 the stub is permitted to resend a stop reply notification
27252 if it believes @value{GDBN} may not have received it. @value{GDBN}
27253 ignores additional stop reply notifications received before it has
27254 finished processing a previous notification and the stub has completed
27255 sending any queued stop events.
27257 Otherwise, @value{GDBN} must be prepared to receive a stop reply
27258 notification at any time. Specifically, they may appear when
27259 @value{GDBN} is not otherwise reading input from the stub, or when
27260 @value{GDBN} is expecting to read a normal synchronous response or a
27261 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
27262 Notification packets are distinct from any other communication from
27263 the stub so there is no ambiguity.
27265 After receiving a stop reply notification, @value{GDBN} shall
27266 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
27267 as a regular, synchronous request to the stub. Such acknowledgment
27268 is not required to happen immediately, as @value{GDBN} is permitted to
27269 send other, unrelated packets to the stub first, which the stub should
27272 Upon receiving a @samp{vStopped} packet, if the stub has other queued
27273 stop events to report to @value{GDBN}, it shall respond by sending a
27274 normal stop reply response. @value{GDBN} shall then send another
27275 @samp{vStopped} packet to solicit further responses; again, it is
27276 permitted to send other, unrelated packets as well which the stub
27277 should process normally.
27279 If the stub receives a @samp{vStopped} packet and there are no
27280 additional stop events to report, the stub shall return an @samp{OK}
27281 response. At this point, if further stop events occur, the stub shall
27282 send a new stop reply notification, @value{GDBN} shall accept the
27283 notification, and the process shall be repeated.
27285 In non-stop mode, the target shall respond to the @samp{?} packet as
27286 follows. First, any incomplete stop reply notification/@samp{vStopped}
27287 sequence in progress is abandoned. The target must begin a new
27288 sequence reporting stop events for all stopped threads, whether or not
27289 it has previously reported those events to @value{GDBN}. The first
27290 stop reply is sent as a synchronous reply to the @samp{?} packet, and
27291 subsequent stop replies are sent as responses to @samp{vStopped} packets
27292 using the mechanism described above. The target must not send
27293 asynchronous stop reply notifications until the sequence is complete.
27294 If all threads are running when the target receives the @samp{?} packet,
27295 or if the target is not attached to any process, it shall respond
27298 @node Packet Acknowledgment
27299 @section Packet Acknowledgment
27301 @cindex acknowledgment, for @value{GDBN} remote
27302 @cindex packet acknowledgment, for @value{GDBN} remote
27303 By default, when either the host or the target machine receives a packet,
27304 the first response expected is an acknowledgment: either @samp{+} (to indicate
27305 the package was received correctly) or @samp{-} (to request retransmission).
27306 This mechanism allows the @value{GDBN} remote protocol to operate over
27307 unreliable transport mechanisms, such as a serial line.
27309 In cases where the transport mechanism is itself reliable (such as a pipe or
27310 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
27311 It may be desirable to disable them in that case to reduce communication
27312 overhead, or for other reasons. This can be accomplished by means of the
27313 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
27315 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
27316 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
27317 and response format still includes the normal checksum, as described in
27318 @ref{Overview}, but the checksum may be ignored by the receiver.
27320 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
27321 no-acknowledgment mode, it should report that to @value{GDBN}
27322 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
27323 @pxref{qSupported}.
27324 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
27325 disabled via the @code{set remote noack-packet off} command
27326 (@pxref{Remote Configuration}),
27327 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
27328 Only then may the stub actually turn off packet acknowledgments.
27329 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
27330 response, which can be safely ignored by the stub.
27332 Note that @code{set remote noack-packet} command only affects negotiation
27333 between @value{GDBN} and the stub when subsequent connections are made;
27334 it does not affect the protocol acknowledgment state for any current
27336 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
27337 new connection is established,
27338 there is also no protocol request to re-enable the acknowledgments
27339 for the current connection, once disabled.
27344 Example sequence of a target being re-started. Notice how the restart
27345 does not get any direct output:
27350 @emph{target restarts}
27353 <- @code{T001:1234123412341234}
27357 Example sequence of a target being stepped by a single instruction:
27360 -> @code{G1445@dots{}}
27365 <- @code{T001:1234123412341234}
27369 <- @code{1455@dots{}}
27373 @node File-I/O Remote Protocol Extension
27374 @section File-I/O Remote Protocol Extension
27375 @cindex File-I/O remote protocol extension
27378 * File-I/O Overview::
27379 * Protocol Basics::
27380 * The F Request Packet::
27381 * The F Reply Packet::
27382 * The Ctrl-C Message::
27384 * List of Supported Calls::
27385 * Protocol-specific Representation of Datatypes::
27387 * File-I/O Examples::
27390 @node File-I/O Overview
27391 @subsection File-I/O Overview
27392 @cindex file-i/o overview
27394 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
27395 target to use the host's file system and console I/O to perform various
27396 system calls. System calls on the target system are translated into a
27397 remote protocol packet to the host system, which then performs the needed
27398 actions and returns a response packet to the target system.
27399 This simulates file system operations even on targets that lack file systems.
27401 The protocol is defined to be independent of both the host and target systems.
27402 It uses its own internal representation of datatypes and values. Both
27403 @value{GDBN} and the target's @value{GDBN} stub are responsible for
27404 translating the system-dependent value representations into the internal
27405 protocol representations when data is transmitted.
27407 The communication is synchronous. A system call is possible only when
27408 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
27409 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
27410 the target is stopped to allow deterministic access to the target's
27411 memory. Therefore File-I/O is not interruptible by target signals. On
27412 the other hand, it is possible to interrupt File-I/O by a user interrupt
27413 (@samp{Ctrl-C}) within @value{GDBN}.
27415 The target's request to perform a host system call does not finish
27416 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
27417 after finishing the system call, the target returns to continuing the
27418 previous activity (continue, step). No additional continue or step
27419 request from @value{GDBN} is required.
27422 (@value{GDBP}) continue
27423 <- target requests 'system call X'
27424 target is stopped, @value{GDBN} executes system call
27425 -> @value{GDBN} returns result
27426 ... target continues, @value{GDBN} returns to wait for the target
27427 <- target hits breakpoint and sends a Txx packet
27430 The protocol only supports I/O on the console and to regular files on
27431 the host file system. Character or block special devices, pipes,
27432 named pipes, sockets or any other communication method on the host
27433 system are not supported by this protocol.
27435 File I/O is not supported in non-stop mode.
27437 @node Protocol Basics
27438 @subsection Protocol Basics
27439 @cindex protocol basics, file-i/o
27441 The File-I/O protocol uses the @code{F} packet as the request as well
27442 as reply packet. Since a File-I/O system call can only occur when
27443 @value{GDBN} is waiting for a response from the continuing or stepping target,
27444 the File-I/O request is a reply that @value{GDBN} has to expect as a result
27445 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
27446 This @code{F} packet contains all information needed to allow @value{GDBN}
27447 to call the appropriate host system call:
27451 A unique identifier for the requested system call.
27454 All parameters to the system call. Pointers are given as addresses
27455 in the target memory address space. Pointers to strings are given as
27456 pointer/length pair. Numerical values are given as they are.
27457 Numerical control flags are given in a protocol-specific representation.
27461 At this point, @value{GDBN} has to perform the following actions.
27465 If the parameters include pointer values to data needed as input to a
27466 system call, @value{GDBN} requests this data from the target with a
27467 standard @code{m} packet request. This additional communication has to be
27468 expected by the target implementation and is handled as any other @code{m}
27472 @value{GDBN} translates all value from protocol representation to host
27473 representation as needed. Datatypes are coerced into the host types.
27476 @value{GDBN} calls the system call.
27479 It then coerces datatypes back to protocol representation.
27482 If the system call is expected to return data in buffer space specified
27483 by pointer parameters to the call, the data is transmitted to the
27484 target using a @code{M} or @code{X} packet. This packet has to be expected
27485 by the target implementation and is handled as any other @code{M} or @code{X}
27490 Eventually @value{GDBN} replies with another @code{F} packet which contains all
27491 necessary information for the target to continue. This at least contains
27498 @code{errno}, if has been changed by the system call.
27505 After having done the needed type and value coercion, the target continues
27506 the latest continue or step action.
27508 @node The F Request Packet
27509 @subsection The @code{F} Request Packet
27510 @cindex file-i/o request packet
27511 @cindex @code{F} request packet
27513 The @code{F} request packet has the following format:
27516 @item F@var{call-id},@var{parameter@dots{}}
27518 @var{call-id} is the identifier to indicate the host system call to be called.
27519 This is just the name of the function.
27521 @var{parameter@dots{}} are the parameters to the system call.
27522 Parameters are hexadecimal integer values, either the actual values in case
27523 of scalar datatypes, pointers to target buffer space in case of compound
27524 datatypes and unspecified memory areas, or pointer/length pairs in case
27525 of string parameters. These are appended to the @var{call-id} as a
27526 comma-delimited list. All values are transmitted in ASCII
27527 string representation, pointer/length pairs separated by a slash.
27533 @node The F Reply Packet
27534 @subsection The @code{F} Reply Packet
27535 @cindex file-i/o reply packet
27536 @cindex @code{F} reply packet
27538 The @code{F} reply packet has the following format:
27542 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
27544 @var{retcode} is the return code of the system call as hexadecimal value.
27546 @var{errno} is the @code{errno} set by the call, in protocol-specific
27548 This parameter can be omitted if the call was successful.
27550 @var{Ctrl-C flag} is only sent if the user requested a break. In this
27551 case, @var{errno} must be sent as well, even if the call was successful.
27552 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
27559 or, if the call was interrupted before the host call has been performed:
27566 assuming 4 is the protocol-specific representation of @code{EINTR}.
27571 @node The Ctrl-C Message
27572 @subsection The @samp{Ctrl-C} Message
27573 @cindex ctrl-c message, in file-i/o protocol
27575 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
27576 reply packet (@pxref{The F Reply Packet}),
27577 the target should behave as if it had
27578 gotten a break message. The meaning for the target is ``system call
27579 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
27580 (as with a break message) and return to @value{GDBN} with a @code{T02}
27583 It's important for the target to know in which
27584 state the system call was interrupted. There are two possible cases:
27588 The system call hasn't been performed on the host yet.
27591 The system call on the host has been finished.
27595 These two states can be distinguished by the target by the value of the
27596 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
27597 call hasn't been performed. This is equivalent to the @code{EINTR} handling
27598 on POSIX systems. In any other case, the target may presume that the
27599 system call has been finished --- successfully or not --- and should behave
27600 as if the break message arrived right after the system call.
27602 @value{GDBN} must behave reliably. If the system call has not been called
27603 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
27604 @code{errno} in the packet. If the system call on the host has been finished
27605 before the user requests a break, the full action must be finished by
27606 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
27607 The @code{F} packet may only be sent when either nothing has happened
27608 or the full action has been completed.
27611 @subsection Console I/O
27612 @cindex console i/o as part of file-i/o
27614 By default and if not explicitly closed by the target system, the file
27615 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
27616 on the @value{GDBN} console is handled as any other file output operation
27617 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
27618 by @value{GDBN} so that after the target read request from file descriptor
27619 0 all following typing is buffered until either one of the following
27624 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
27626 system call is treated as finished.
27629 The user presses @key{RET}. This is treated as end of input with a trailing
27633 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
27634 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
27638 If the user has typed more characters than fit in the buffer given to
27639 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
27640 either another @code{read(0, @dots{})} is requested by the target, or debugging
27641 is stopped at the user's request.
27644 @node List of Supported Calls
27645 @subsection List of Supported Calls
27646 @cindex list of supported file-i/o calls
27663 @unnumberedsubsubsec open
27664 @cindex open, file-i/o system call
27669 int open(const char *pathname, int flags);
27670 int open(const char *pathname, int flags, mode_t mode);
27674 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
27677 @var{flags} is the bitwise @code{OR} of the following values:
27681 If the file does not exist it will be created. The host
27682 rules apply as far as file ownership and time stamps
27686 When used with @code{O_CREAT}, if the file already exists it is
27687 an error and open() fails.
27690 If the file already exists and the open mode allows
27691 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
27692 truncated to zero length.
27695 The file is opened in append mode.
27698 The file is opened for reading only.
27701 The file is opened for writing only.
27704 The file is opened for reading and writing.
27708 Other bits are silently ignored.
27712 @var{mode} is the bitwise @code{OR} of the following values:
27716 User has read permission.
27719 User has write permission.
27722 Group has read permission.
27725 Group has write permission.
27728 Others have read permission.
27731 Others have write permission.
27735 Other bits are silently ignored.
27738 @item Return value:
27739 @code{open} returns the new file descriptor or -1 if an error
27746 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
27749 @var{pathname} refers to a directory.
27752 The requested access is not allowed.
27755 @var{pathname} was too long.
27758 A directory component in @var{pathname} does not exist.
27761 @var{pathname} refers to a device, pipe, named pipe or socket.
27764 @var{pathname} refers to a file on a read-only filesystem and
27765 write access was requested.
27768 @var{pathname} is an invalid pointer value.
27771 No space on device to create the file.
27774 The process already has the maximum number of files open.
27777 The limit on the total number of files open on the system
27781 The call was interrupted by the user.
27787 @unnumberedsubsubsec close
27788 @cindex close, file-i/o system call
27797 @samp{Fclose,@var{fd}}
27799 @item Return value:
27800 @code{close} returns zero on success, or -1 if an error occurred.
27806 @var{fd} isn't a valid open file descriptor.
27809 The call was interrupted by the user.
27815 @unnumberedsubsubsec read
27816 @cindex read, file-i/o system call
27821 int read(int fd, void *buf, unsigned int count);
27825 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
27827 @item Return value:
27828 On success, the number of bytes read is returned.
27829 Zero indicates end of file. If count is zero, read
27830 returns zero as well. On error, -1 is returned.
27836 @var{fd} is not a valid file descriptor or is not open for
27840 @var{bufptr} is an invalid pointer value.
27843 The call was interrupted by the user.
27849 @unnumberedsubsubsec write
27850 @cindex write, file-i/o system call
27855 int write(int fd, const void *buf, unsigned int count);
27859 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
27861 @item Return value:
27862 On success, the number of bytes written are returned.
27863 Zero indicates nothing was written. On error, -1
27870 @var{fd} is not a valid file descriptor or is not open for
27874 @var{bufptr} is an invalid pointer value.
27877 An attempt was made to write a file that exceeds the
27878 host-specific maximum file size allowed.
27881 No space on device to write the data.
27884 The call was interrupted by the user.
27890 @unnumberedsubsubsec lseek
27891 @cindex lseek, file-i/o system call
27896 long lseek (int fd, long offset, int flag);
27900 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
27902 @var{flag} is one of:
27906 The offset is set to @var{offset} bytes.
27909 The offset is set to its current location plus @var{offset}
27913 The offset is set to the size of the file plus @var{offset}
27917 @item Return value:
27918 On success, the resulting unsigned offset in bytes from
27919 the beginning of the file is returned. Otherwise, a
27920 value of -1 is returned.
27926 @var{fd} is not a valid open file descriptor.
27929 @var{fd} is associated with the @value{GDBN} console.
27932 @var{flag} is not a proper value.
27935 The call was interrupted by the user.
27941 @unnumberedsubsubsec rename
27942 @cindex rename, file-i/o system call
27947 int rename(const char *oldpath, const char *newpath);
27951 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
27953 @item Return value:
27954 On success, zero is returned. On error, -1 is returned.
27960 @var{newpath} is an existing directory, but @var{oldpath} is not a
27964 @var{newpath} is a non-empty directory.
27967 @var{oldpath} or @var{newpath} is a directory that is in use by some
27971 An attempt was made to make a directory a subdirectory
27975 A component used as a directory in @var{oldpath} or new
27976 path is not a directory. Or @var{oldpath} is a directory
27977 and @var{newpath} exists but is not a directory.
27980 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
27983 No access to the file or the path of the file.
27987 @var{oldpath} or @var{newpath} was too long.
27990 A directory component in @var{oldpath} or @var{newpath} does not exist.
27993 The file is on a read-only filesystem.
27996 The device containing the file has no room for the new
28000 The call was interrupted by the user.
28006 @unnumberedsubsubsec unlink
28007 @cindex unlink, file-i/o system call
28012 int unlink(const char *pathname);
28016 @samp{Funlink,@var{pathnameptr}/@var{len}}
28018 @item Return value:
28019 On success, zero is returned. On error, -1 is returned.
28025 No access to the file or the path of the file.
28028 The system does not allow unlinking of directories.
28031 The file @var{pathname} cannot be unlinked because it's
28032 being used by another process.
28035 @var{pathnameptr} is an invalid pointer value.
28038 @var{pathname} was too long.
28041 A directory component in @var{pathname} does not exist.
28044 A component of the path is not a directory.
28047 The file is on a read-only filesystem.
28050 The call was interrupted by the user.
28056 @unnumberedsubsubsec stat/fstat
28057 @cindex fstat, file-i/o system call
28058 @cindex stat, file-i/o system call
28063 int stat(const char *pathname, struct stat *buf);
28064 int fstat(int fd, struct stat *buf);
28068 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
28069 @samp{Ffstat,@var{fd},@var{bufptr}}
28071 @item Return value:
28072 On success, zero is returned. On error, -1 is returned.
28078 @var{fd} is not a valid open file.
28081 A directory component in @var{pathname} does not exist or the
28082 path is an empty string.
28085 A component of the path is not a directory.
28088 @var{pathnameptr} is an invalid pointer value.
28091 No access to the file or the path of the file.
28094 @var{pathname} was too long.
28097 The call was interrupted by the user.
28103 @unnumberedsubsubsec gettimeofday
28104 @cindex gettimeofday, file-i/o system call
28109 int gettimeofday(struct timeval *tv, void *tz);
28113 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
28115 @item Return value:
28116 On success, 0 is returned, -1 otherwise.
28122 @var{tz} is a non-NULL pointer.
28125 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
28131 @unnumberedsubsubsec isatty
28132 @cindex isatty, file-i/o system call
28137 int isatty(int fd);
28141 @samp{Fisatty,@var{fd}}
28143 @item Return value:
28144 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
28150 The call was interrupted by the user.
28155 Note that the @code{isatty} call is treated as a special case: it returns
28156 1 to the target if the file descriptor is attached
28157 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
28158 would require implementing @code{ioctl} and would be more complex than
28163 @unnumberedsubsubsec system
28164 @cindex system, file-i/o system call
28169 int system(const char *command);
28173 @samp{Fsystem,@var{commandptr}/@var{len}}
28175 @item Return value:
28176 If @var{len} is zero, the return value indicates whether a shell is
28177 available. A zero return value indicates a shell is not available.
28178 For non-zero @var{len}, the value returned is -1 on error and the
28179 return status of the command otherwise. Only the exit status of the
28180 command is returned, which is extracted from the host's @code{system}
28181 return value by calling @code{WEXITSTATUS(retval)}. In case
28182 @file{/bin/sh} could not be executed, 127 is returned.
28188 The call was interrupted by the user.
28193 @value{GDBN} takes over the full task of calling the necessary host calls
28194 to perform the @code{system} call. The return value of @code{system} on
28195 the host is simplified before it's returned
28196 to the target. Any termination signal information from the child process
28197 is discarded, and the return value consists
28198 entirely of the exit status of the called command.
28200 Due to security concerns, the @code{system} call is by default refused
28201 by @value{GDBN}. The user has to allow this call explicitly with the
28202 @code{set remote system-call-allowed 1} command.
28205 @item set remote system-call-allowed
28206 @kindex set remote system-call-allowed
28207 Control whether to allow the @code{system} calls in the File I/O
28208 protocol for the remote target. The default is zero (disabled).
28210 @item show remote system-call-allowed
28211 @kindex show remote system-call-allowed
28212 Show whether the @code{system} calls are allowed in the File I/O
28216 @node Protocol-specific Representation of Datatypes
28217 @subsection Protocol-specific Representation of Datatypes
28218 @cindex protocol-specific representation of datatypes, in file-i/o protocol
28221 * Integral Datatypes::
28223 * Memory Transfer::
28228 @node Integral Datatypes
28229 @unnumberedsubsubsec Integral Datatypes
28230 @cindex integral datatypes, in file-i/o protocol
28232 The integral datatypes used in the system calls are @code{int},
28233 @code{unsigned int}, @code{long}, @code{unsigned long},
28234 @code{mode_t}, and @code{time_t}.
28236 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
28237 implemented as 32 bit values in this protocol.
28239 @code{long} and @code{unsigned long} are implemented as 64 bit types.
28241 @xref{Limits}, for corresponding MIN and MAX values (similar to those
28242 in @file{limits.h}) to allow range checking on host and target.
28244 @code{time_t} datatypes are defined as seconds since the Epoch.
28246 All integral datatypes transferred as part of a memory read or write of a
28247 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
28250 @node Pointer Values
28251 @unnumberedsubsubsec Pointer Values
28252 @cindex pointer values, in file-i/o protocol
28254 Pointers to target data are transmitted as they are. An exception
28255 is made for pointers to buffers for which the length isn't
28256 transmitted as part of the function call, namely strings. Strings
28257 are transmitted as a pointer/length pair, both as hex values, e.g.@:
28264 which is a pointer to data of length 18 bytes at position 0x1aaf.
28265 The length is defined as the full string length in bytes, including
28266 the trailing null byte. For example, the string @code{"hello world"}
28267 at address 0x123456 is transmitted as
28273 @node Memory Transfer
28274 @unnumberedsubsubsec Memory Transfer
28275 @cindex memory transfer, in file-i/o protocol
28277 Structured data which is transferred using a memory read or write (for
28278 example, a @code{struct stat}) is expected to be in a protocol-specific format
28279 with all scalar multibyte datatypes being big endian. Translation to
28280 this representation needs to be done both by the target before the @code{F}
28281 packet is sent, and by @value{GDBN} before
28282 it transfers memory to the target. Transferred pointers to structured
28283 data should point to the already-coerced data at any time.
28287 @unnumberedsubsubsec struct stat
28288 @cindex struct stat, in file-i/o protocol
28290 The buffer of type @code{struct stat} used by the target and @value{GDBN}
28291 is defined as follows:
28295 unsigned int st_dev; /* device */
28296 unsigned int st_ino; /* inode */
28297 mode_t st_mode; /* protection */
28298 unsigned int st_nlink; /* number of hard links */
28299 unsigned int st_uid; /* user ID of owner */
28300 unsigned int st_gid; /* group ID of owner */
28301 unsigned int st_rdev; /* device type (if inode device) */
28302 unsigned long st_size; /* total size, in bytes */
28303 unsigned long st_blksize; /* blocksize for filesystem I/O */
28304 unsigned long st_blocks; /* number of blocks allocated */
28305 time_t st_atime; /* time of last access */
28306 time_t st_mtime; /* time of last modification */
28307 time_t st_ctime; /* time of last change */
28311 The integral datatypes conform to the definitions given in the
28312 appropriate section (see @ref{Integral Datatypes}, for details) so this
28313 structure is of size 64 bytes.
28315 The values of several fields have a restricted meaning and/or
28321 A value of 0 represents a file, 1 the console.
28324 No valid meaning for the target. Transmitted unchanged.
28327 Valid mode bits are described in @ref{Constants}. Any other
28328 bits have currently no meaning for the target.
28333 No valid meaning for the target. Transmitted unchanged.
28338 These values have a host and file system dependent
28339 accuracy. Especially on Windows hosts, the file system may not
28340 support exact timing values.
28343 The target gets a @code{struct stat} of the above representation and is
28344 responsible for coercing it to the target representation before
28347 Note that due to size differences between the host, target, and protocol
28348 representations of @code{struct stat} members, these members could eventually
28349 get truncated on the target.
28351 @node struct timeval
28352 @unnumberedsubsubsec struct timeval
28353 @cindex struct timeval, in file-i/o protocol
28355 The buffer of type @code{struct timeval} used by the File-I/O protocol
28356 is defined as follows:
28360 time_t tv_sec; /* second */
28361 long tv_usec; /* microsecond */
28365 The integral datatypes conform to the definitions given in the
28366 appropriate section (see @ref{Integral Datatypes}, for details) so this
28367 structure is of size 8 bytes.
28370 @subsection Constants
28371 @cindex constants, in file-i/o protocol
28373 The following values are used for the constants inside of the
28374 protocol. @value{GDBN} and target are responsible for translating these
28375 values before and after the call as needed.
28386 @unnumberedsubsubsec Open Flags
28387 @cindex open flags, in file-i/o protocol
28389 All values are given in hexadecimal representation.
28401 @node mode_t Values
28402 @unnumberedsubsubsec mode_t Values
28403 @cindex mode_t values, in file-i/o protocol
28405 All values are given in octal representation.
28422 @unnumberedsubsubsec Errno Values
28423 @cindex errno values, in file-i/o protocol
28425 All values are given in decimal representation.
28450 @code{EUNKNOWN} is used as a fallback error value if a host system returns
28451 any error value not in the list of supported error numbers.
28454 @unnumberedsubsubsec Lseek Flags
28455 @cindex lseek flags, in file-i/o protocol
28464 @unnumberedsubsubsec Limits
28465 @cindex limits, in file-i/o protocol
28467 All values are given in decimal representation.
28470 INT_MIN -2147483648
28472 UINT_MAX 4294967295
28473 LONG_MIN -9223372036854775808
28474 LONG_MAX 9223372036854775807
28475 ULONG_MAX 18446744073709551615
28478 @node File-I/O Examples
28479 @subsection File-I/O Examples
28480 @cindex file-i/o examples
28482 Example sequence of a write call, file descriptor 3, buffer is at target
28483 address 0x1234, 6 bytes should be written:
28486 <- @code{Fwrite,3,1234,6}
28487 @emph{request memory read from target}
28490 @emph{return "6 bytes written"}
28494 Example sequence of a read call, file descriptor 3, buffer is at target
28495 address 0x1234, 6 bytes should be read:
28498 <- @code{Fread,3,1234,6}
28499 @emph{request memory write to target}
28500 -> @code{X1234,6:XXXXXX}
28501 @emph{return "6 bytes read"}
28505 Example sequence of a read call, call fails on the host due to invalid
28506 file descriptor (@code{EBADF}):
28509 <- @code{Fread,3,1234,6}
28513 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
28517 <- @code{Fread,3,1234,6}
28522 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
28526 <- @code{Fread,3,1234,6}
28527 -> @code{X1234,6:XXXXXX}
28531 @node Library List Format
28532 @section Library List Format
28533 @cindex library list format, remote protocol
28535 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
28536 same process as your application to manage libraries. In this case,
28537 @value{GDBN} can use the loader's symbol table and normal memory
28538 operations to maintain a list of shared libraries. On other
28539 platforms, the operating system manages loaded libraries.
28540 @value{GDBN} can not retrieve the list of currently loaded libraries
28541 through memory operations, so it uses the @samp{qXfer:libraries:read}
28542 packet (@pxref{qXfer library list read}) instead. The remote stub
28543 queries the target's operating system and reports which libraries
28546 The @samp{qXfer:libraries:read} packet returns an XML document which
28547 lists loaded libraries and their offsets. Each library has an
28548 associated name and one or more segment or section base addresses,
28549 which report where the library was loaded in memory.
28551 For the common case of libraries that are fully linked binaries, the
28552 library should have a list of segments. If the target supports
28553 dynamic linking of a relocatable object file, its library XML element
28554 should instead include a list of allocated sections. The segment or
28555 section bases are start addresses, not relocation offsets; they do not
28556 depend on the library's link-time base addresses.
28558 @value{GDBN} must be linked with the Expat library to support XML
28559 library lists. @xref{Expat}.
28561 A simple memory map, with one loaded library relocated by a single
28562 offset, looks like this:
28566 <library name="/lib/libc.so.6">
28567 <segment address="0x10000000"/>
28572 Another simple memory map, with one loaded library with three
28573 allocated sections (.text, .data, .bss), looks like this:
28577 <library name="sharedlib.o">
28578 <section address="0x10000000"/>
28579 <section address="0x20000000"/>
28580 <section address="0x30000000"/>
28585 The format of a library list is described by this DTD:
28588 <!-- library-list: Root element with versioning -->
28589 <!ELEMENT library-list (library)*>
28590 <!ATTLIST library-list version CDATA #FIXED "1.0">
28591 <!ELEMENT library (segment*, section*)>
28592 <!ATTLIST library name CDATA #REQUIRED>
28593 <!ELEMENT segment EMPTY>
28594 <!ATTLIST segment address CDATA #REQUIRED>
28595 <!ELEMENT section EMPTY>
28596 <!ATTLIST section address CDATA #REQUIRED>
28599 In addition, segments and section descriptors cannot be mixed within a
28600 single library element, and you must supply at least one segment or
28601 section for each library.
28603 @node Memory Map Format
28604 @section Memory Map Format
28605 @cindex memory map format
28607 To be able to write into flash memory, @value{GDBN} needs to obtain a
28608 memory map from the target. This section describes the format of the
28611 The memory map is obtained using the @samp{qXfer:memory-map:read}
28612 (@pxref{qXfer memory map read}) packet and is an XML document that
28613 lists memory regions.
28615 @value{GDBN} must be linked with the Expat library to support XML
28616 memory maps. @xref{Expat}.
28618 The top-level structure of the document is shown below:
28621 <?xml version="1.0"?>
28622 <!DOCTYPE memory-map
28623 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
28624 "http://sourceware.org/gdb/gdb-memory-map.dtd">
28630 Each region can be either:
28635 A region of RAM starting at @var{addr} and extending for @var{length}
28639 <memory type="ram" start="@var{addr}" length="@var{length}"/>
28644 A region of read-only memory:
28647 <memory type="rom" start="@var{addr}" length="@var{length}"/>
28652 A region of flash memory, with erasure blocks @var{blocksize}
28656 <memory type="flash" start="@var{addr}" length="@var{length}">
28657 <property name="blocksize">@var{blocksize}</property>
28663 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
28664 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
28665 packets to write to addresses in such ranges.
28667 The formal DTD for memory map format is given below:
28670 <!-- ................................................... -->
28671 <!-- Memory Map XML DTD ................................ -->
28672 <!-- File: memory-map.dtd .............................. -->
28673 <!-- .................................... .............. -->
28674 <!-- memory-map.dtd -->
28675 <!-- memory-map: Root element with versioning -->
28676 <!ELEMENT memory-map (memory | property)>
28677 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
28678 <!ELEMENT memory (property)>
28679 <!-- memory: Specifies a memory region,
28680 and its type, or device. -->
28681 <!ATTLIST memory type CDATA #REQUIRED
28682 start CDATA #REQUIRED
28683 length CDATA #REQUIRED
28684 device CDATA #IMPLIED>
28685 <!-- property: Generic attribute tag -->
28686 <!ELEMENT property (#PCDATA | property)*>
28687 <!ATTLIST property name CDATA #REQUIRED>
28690 @include agentexpr.texi
28692 @node Target Descriptions
28693 @appendix Target Descriptions
28694 @cindex target descriptions
28696 @strong{Warning:} target descriptions are still under active development,
28697 and the contents and format may change between @value{GDBN} releases.
28698 The format is expected to stabilize in the future.
28700 One of the challenges of using @value{GDBN} to debug embedded systems
28701 is that there are so many minor variants of each processor
28702 architecture in use. It is common practice for vendors to start with
28703 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
28704 and then make changes to adapt it to a particular market niche. Some
28705 architectures have hundreds of variants, available from dozens of
28706 vendors. This leads to a number of problems:
28710 With so many different customized processors, it is difficult for
28711 the @value{GDBN} maintainers to keep up with the changes.
28713 Since individual variants may have short lifetimes or limited
28714 audiences, it may not be worthwhile to carry information about every
28715 variant in the @value{GDBN} source tree.
28717 When @value{GDBN} does support the architecture of the embedded system
28718 at hand, the task of finding the correct architecture name to give the
28719 @command{set architecture} command can be error-prone.
28722 To address these problems, the @value{GDBN} remote protocol allows a
28723 target system to not only identify itself to @value{GDBN}, but to
28724 actually describe its own features. This lets @value{GDBN} support
28725 processor variants it has never seen before --- to the extent that the
28726 descriptions are accurate, and that @value{GDBN} understands them.
28728 @value{GDBN} must be linked with the Expat library to support XML
28729 target descriptions. @xref{Expat}.
28732 * Retrieving Descriptions:: How descriptions are fetched from a target.
28733 * Target Description Format:: The contents of a target description.
28734 * Predefined Target Types:: Standard types available for target
28736 * Standard Target Features:: Features @value{GDBN} knows about.
28739 @node Retrieving Descriptions
28740 @section Retrieving Descriptions
28742 Target descriptions can be read from the target automatically, or
28743 specified by the user manually. The default behavior is to read the
28744 description from the target. @value{GDBN} retrieves it via the remote
28745 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
28746 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
28747 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
28748 XML document, of the form described in @ref{Target Description
28751 Alternatively, you can specify a file to read for the target description.
28752 If a file is set, the target will not be queried. The commands to
28753 specify a file are:
28756 @cindex set tdesc filename
28757 @item set tdesc filename @var{path}
28758 Read the target description from @var{path}.
28760 @cindex unset tdesc filename
28761 @item unset tdesc filename
28762 Do not read the XML target description from a file. @value{GDBN}
28763 will use the description supplied by the current target.
28765 @cindex show tdesc filename
28766 @item show tdesc filename
28767 Show the filename to read for a target description, if any.
28771 @node Target Description Format
28772 @section Target Description Format
28773 @cindex target descriptions, XML format
28775 A target description annex is an @uref{http://www.w3.org/XML/, XML}
28776 document which complies with the Document Type Definition provided in
28777 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
28778 means you can use generally available tools like @command{xmllint} to
28779 check that your feature descriptions are well-formed and valid.
28780 However, to help people unfamiliar with XML write descriptions for
28781 their targets, we also describe the grammar here.
28783 Target descriptions can identify the architecture of the remote target
28784 and (for some architectures) provide information about custom register
28785 sets. @value{GDBN} can use this information to autoconfigure for your
28786 target, or to warn you if you connect to an unsupported target.
28788 Here is a simple target description:
28791 <target version="1.0">
28792 <architecture>i386:x86-64</architecture>
28797 This minimal description only says that the target uses
28798 the x86-64 architecture.
28800 A target description has the following overall form, with [ ] marking
28801 optional elements and @dots{} marking repeatable elements. The elements
28802 are explained further below.
28805 <?xml version="1.0"?>
28806 <!DOCTYPE target SYSTEM "gdb-target.dtd">
28807 <target version="1.0">
28808 @r{[}@var{architecture}@r{]}
28809 @r{[}@var{feature}@dots{}@r{]}
28814 The description is generally insensitive to whitespace and line
28815 breaks, under the usual common-sense rules. The XML version
28816 declaration and document type declaration can generally be omitted
28817 (@value{GDBN} does not require them), but specifying them may be
28818 useful for XML validation tools. The @samp{version} attribute for
28819 @samp{<target>} may also be omitted, but we recommend
28820 including it; if future versions of @value{GDBN} use an incompatible
28821 revision of @file{gdb-target.dtd}, they will detect and report
28822 the version mismatch.
28824 @subsection Inclusion
28825 @cindex target descriptions, inclusion
28828 @cindex <xi:include>
28831 It can sometimes be valuable to split a target description up into
28832 several different annexes, either for organizational purposes, or to
28833 share files between different possible target descriptions. You can
28834 divide a description into multiple files by replacing any element of
28835 the target description with an inclusion directive of the form:
28838 <xi:include href="@var{document}"/>
28842 When @value{GDBN} encounters an element of this form, it will retrieve
28843 the named XML @var{document}, and replace the inclusion directive with
28844 the contents of that document. If the current description was read
28845 using @samp{qXfer}, then so will be the included document;
28846 @var{document} will be interpreted as the name of an annex. If the
28847 current description was read from a file, @value{GDBN} will look for
28848 @var{document} as a file in the same directory where it found the
28849 original description.
28851 @subsection Architecture
28852 @cindex <architecture>
28854 An @samp{<architecture>} element has this form:
28857 <architecture>@var{arch}</architecture>
28860 @var{arch} is an architecture name from the same selection
28861 accepted by @code{set architecture} (@pxref{Targets, ,Specifying a
28862 Debugging Target}).
28864 @subsection Features
28867 Each @samp{<feature>} describes some logical portion of the target
28868 system. Features are currently used to describe available CPU
28869 registers and the types of their contents. A @samp{<feature>} element
28873 <feature name="@var{name}">
28874 @r{[}@var{type}@dots{}@r{]}
28880 Each feature's name should be unique within the description. The name
28881 of a feature does not matter unless @value{GDBN} has some special
28882 knowledge of the contents of that feature; if it does, the feature
28883 should have its standard name. @xref{Standard Target Features}.
28887 Any register's value is a collection of bits which @value{GDBN} must
28888 interpret. The default interpretation is a two's complement integer,
28889 but other types can be requested by name in the register description.
28890 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
28891 Target Types}), and the description can define additional composite types.
28893 Each type element must have an @samp{id} attribute, which gives
28894 a unique (within the containing @samp{<feature>}) name to the type.
28895 Types must be defined before they are used.
28898 Some targets offer vector registers, which can be treated as arrays
28899 of scalar elements. These types are written as @samp{<vector>} elements,
28900 specifying the array element type, @var{type}, and the number of elements,
28904 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
28908 If a register's value is usefully viewed in multiple ways, define it
28909 with a union type containing the useful representations. The
28910 @samp{<union>} element contains one or more @samp{<field>} elements,
28911 each of which has a @var{name} and a @var{type}:
28914 <union id="@var{id}">
28915 <field name="@var{name}" type="@var{type}"/>
28920 @subsection Registers
28923 Each register is represented as an element with this form:
28926 <reg name="@var{name}"
28927 bitsize="@var{size}"
28928 @r{[}regnum="@var{num}"@r{]}
28929 @r{[}save-restore="@var{save-restore}"@r{]}
28930 @r{[}type="@var{type}"@r{]}
28931 @r{[}group="@var{group}"@r{]}/>
28935 The components are as follows:
28940 The register's name; it must be unique within the target description.
28943 The register's size, in bits.
28946 The register's number. If omitted, a register's number is one greater
28947 than that of the previous register (either in the current feature or in
28948 a preceeding feature); the first register in the target description
28949 defaults to zero. This register number is used to read or write
28950 the register; e.g.@: it is used in the remote @code{p} and @code{P}
28951 packets, and registers appear in the @code{g} and @code{G} packets
28952 in order of increasing register number.
28955 Whether the register should be preserved across inferior function
28956 calls; this must be either @code{yes} or @code{no}. The default is
28957 @code{yes}, which is appropriate for most registers except for
28958 some system control registers; this is not related to the target's
28962 The type of the register. @var{type} may be a predefined type, a type
28963 defined in the current feature, or one of the special types @code{int}
28964 and @code{float}. @code{int} is an integer type of the correct size
28965 for @var{bitsize}, and @code{float} is a floating point type (in the
28966 architecture's normal floating point format) of the correct size for
28967 @var{bitsize}. The default is @code{int}.
28970 The register group to which this register belongs. @var{group} must
28971 be either @code{general}, @code{float}, or @code{vector}. If no
28972 @var{group} is specified, @value{GDBN} will not display the register
28973 in @code{info registers}.
28977 @node Predefined Target Types
28978 @section Predefined Target Types
28979 @cindex target descriptions, predefined types
28981 Type definitions in the self-description can build up composite types
28982 from basic building blocks, but can not define fundamental types. Instead,
28983 standard identifiers are provided by @value{GDBN} for the fundamental
28984 types. The currently supported types are:
28993 Signed integer types holding the specified number of bits.
29000 Unsigned integer types holding the specified number of bits.
29004 Pointers to unspecified code and data. The program counter and
29005 any dedicated return address register may be marked as code
29006 pointers; printing a code pointer converts it into a symbolic
29007 address. The stack pointer and any dedicated address registers
29008 may be marked as data pointers.
29011 Single precision IEEE floating point.
29014 Double precision IEEE floating point.
29017 The 12-byte extended precision format used by ARM FPA registers.
29021 @node Standard Target Features
29022 @section Standard Target Features
29023 @cindex target descriptions, standard features
29025 A target description must contain either no registers or all the
29026 target's registers. If the description contains no registers, then
29027 @value{GDBN} will assume a default register layout, selected based on
29028 the architecture. If the description contains any registers, the
29029 default layout will not be used; the standard registers must be
29030 described in the target description, in such a way that @value{GDBN}
29031 can recognize them.
29033 This is accomplished by giving specific names to feature elements
29034 which contain standard registers. @value{GDBN} will look for features
29035 with those names and verify that they contain the expected registers;
29036 if any known feature is missing required registers, or if any required
29037 feature is missing, @value{GDBN} will reject the target
29038 description. You can add additional registers to any of the
29039 standard features --- @value{GDBN} will display them just as if
29040 they were added to an unrecognized feature.
29042 This section lists the known features and their expected contents.
29043 Sample XML documents for these features are included in the
29044 @value{GDBN} source tree, in the directory @file{gdb/features}.
29046 Names recognized by @value{GDBN} should include the name of the
29047 company or organization which selected the name, and the overall
29048 architecture to which the feature applies; so e.g.@: the feature
29049 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
29051 The names of registers are not case sensitive for the purpose
29052 of recognizing standard features, but @value{GDBN} will only display
29053 registers using the capitalization used in the description.
29059 * PowerPC Features::
29064 @subsection ARM Features
29065 @cindex target descriptions, ARM features
29067 The @samp{org.gnu.gdb.arm.core} feature is required for ARM targets.
29068 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
29069 @samp{lr}, @samp{pc}, and @samp{cpsr}.
29071 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
29072 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
29074 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
29075 it should contain at least registers @samp{wR0} through @samp{wR15} and
29076 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
29077 @samp{wCSSF}, and @samp{wCASF} registers are optional.
29079 @node MIPS Features
29080 @subsection MIPS Features
29081 @cindex target descriptions, MIPS features
29083 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
29084 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
29085 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
29088 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
29089 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
29090 registers. They may be 32-bit or 64-bit depending on the target.
29092 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
29093 it may be optional in a future version of @value{GDBN}. It should
29094 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
29095 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
29097 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
29098 contain a single register, @samp{restart}, which is used by the
29099 Linux kernel to control restartable syscalls.
29101 @node M68K Features
29102 @subsection M68K Features
29103 @cindex target descriptions, M68K features
29106 @item @samp{org.gnu.gdb.m68k.core}
29107 @itemx @samp{org.gnu.gdb.coldfire.core}
29108 @itemx @samp{org.gnu.gdb.fido.core}
29109 One of those features must be always present.
29110 The feature that is present determines which flavor of m68k is
29111 used. The feature that is present should contain registers
29112 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
29113 @samp{sp}, @samp{ps} and @samp{pc}.
29115 @item @samp{org.gnu.gdb.coldfire.fp}
29116 This feature is optional. If present, it should contain registers
29117 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
29121 @node PowerPC Features
29122 @subsection PowerPC Features
29123 @cindex target descriptions, PowerPC features
29125 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
29126 targets. It should contain registers @samp{r0} through @samp{r31},
29127 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
29128 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
29130 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
29131 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
29133 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
29134 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
29137 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
29138 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
29139 will combine these registers with the floating point registers
29140 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
29141 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
29142 through @samp{vs63}, the set of vector registers for POWER7.
29144 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
29145 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
29146 @samp{spefscr}. SPE targets should provide 32-bit registers in
29147 @samp{org.gnu.gdb.power.core} and provide the upper halves in
29148 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
29149 these to present registers @samp{ev0} through @samp{ev31} to the
29152 @node Operating System Information
29153 @appendix Operating System Information
29154 @cindex operating system information
29160 Users of @value{GDBN} often wish to obtain information about the state of
29161 the operating system running on the target---for example the list of
29162 processes, or the list of open files. This section describes the
29163 mechanism that makes it possible. This mechanism is similar to the
29164 target features mechanism (@pxref{Target Descriptions}), but focuses
29165 on a different aspect of target.
29167 Operating system information is retrived from the target via the
29168 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
29169 read}). The object name in the request should be @samp{osdata}, and
29170 the @var{annex} identifies the data to be fetched.
29173 @appendixsection Process list
29174 @cindex operating system information, process list
29176 When requesting the process list, the @var{annex} field in the
29177 @samp{qXfer} request should be @samp{processes}. The returned data is
29178 an XML document. The formal syntax of this document is defined in
29179 @file{gdb/features/osdata.dtd}.
29181 An example document is:
29184 <?xml version="1.0"?>
29185 <!DOCTYPE target SYSTEM "osdata.dtd">
29186 <osdata type="processes">
29188 <column name="pid">1</column>
29189 <column name="user">root</column>
29190 <column name="command">/sbin/init</column>
29195 Each item should include a column whose name is @samp{pid}. The value
29196 of that column should identify the process on the target. The
29197 @samp{user} and @samp{command} columns are optional, and will be
29198 displayed by @value{GDBN}. Target may provide additional columns,
29199 which @value{GDBN} currently ignores.
29213 % I think something like @colophon should be in texinfo. In the
29215 \long\def\colophon{\hbox to0pt{}\vfill
29216 \centerline{The body of this manual is set in}
29217 \centerline{\fontname\tenrm,}
29218 \centerline{with headings in {\bf\fontname\tenbf}}
29219 \centerline{and examples in {\tt\fontname\tentt}.}
29220 \centerline{{\it\fontname\tenit\/},}
29221 \centerline{{\bf\fontname\tenbf}, and}
29222 \centerline{{\sl\fontname\tensl\/}}
29223 \centerline{are used for emphasis.}\vfill}
29225 % Blame: doc@cygnus.com, 1991.