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, 2010
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
26 @c readline appendices use @vindex, @findex and @ftable,
27 @c annotate.texi and gdbmi use @findex.
31 @c !!set GDB manual's edition---not the same as GDB version!
32 @c This is updated by GNU Press.
35 @c !!set GDB edit command default editor
38 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
40 @c This is a dir.info fragment to support semi-automated addition of
41 @c manuals to an info tree.
42 @dircategory Software development
44 * Gdb: (gdb). The GNU debugger.
48 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996,
49 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
50 Free Software Foundation, Inc.
52 Permission is granted to copy, distribute and/or modify this document
53 under the terms of the GNU Free Documentation License, Version 1.1 or
54 any later version published by the Free Software Foundation; with the
55 Invariant Sections being ``Free Software'' and ``Free Software Needs
56 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
57 and with the Back-Cover Texts as in (a) below.
59 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
60 this GNU Manual. Buying copies from GNU Press supports the FSF in
61 developing GNU and promoting software freedom.''
65 This file documents the @sc{gnu} debugger @value{GDBN}.
67 This is the @value{EDITION} Edition, of @cite{Debugging with
68 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
69 @ifset VERSION_PACKAGE
70 @value{VERSION_PACKAGE}
72 Version @value{GDBVN}.
78 @title Debugging with @value{GDBN}
79 @subtitle The @sc{gnu} Source-Level Debugger
81 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
82 @ifset VERSION_PACKAGE
84 @subtitle @value{VERSION_PACKAGE}
86 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
90 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
91 \hfill {\it Debugging with @value{GDBN}}\par
92 \hfill \TeX{}info \texinfoversion\par
96 @vskip 0pt plus 1filll
97 Published by the Free Software Foundation @*
98 51 Franklin Street, Fifth Floor,
99 Boston, MA 02110-1301, USA@*
100 ISBN 1-882114-77-9 @*
104 This edition of the GDB manual is dedicated to the memory of Fred
105 Fish. Fred was a long-standing contributor to GDB and to Free
106 software in general. We will miss him.
111 @node Top, Summary, (dir), (dir)
113 @top Debugging with @value{GDBN}
115 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
117 This is the @value{EDITION} Edition, for @value{GDBN}
118 @ifset VERSION_PACKAGE
119 @value{VERSION_PACKAGE}
121 Version @value{GDBVN}.
123 Copyright (C) 1988-2010 Free Software Foundation, Inc.
125 This edition of the GDB manual is dedicated to the memory of Fred
126 Fish. Fred was a long-standing contributor to GDB and to Free
127 software in general. We will miss him.
130 * Summary:: Summary of @value{GDBN}
131 * Sample Session:: A sample @value{GDBN} session
133 * Invocation:: Getting in and out of @value{GDBN}
134 * Commands:: @value{GDBN} commands
135 * Running:: Running programs under @value{GDBN}
136 * Stopping:: Stopping and continuing
137 * Reverse Execution:: Running programs backward
138 * Process Record and Replay:: Recording inferior's execution and replaying it
139 * Stack:: Examining the stack
140 * Source:: Examining source files
141 * Data:: Examining data
142 * Optimized Code:: Debugging optimized code
143 * Macros:: Preprocessor Macros
144 * Tracepoints:: Debugging remote targets non-intrusively
145 * Overlays:: Debugging programs that use overlays
147 * Languages:: Using @value{GDBN} with different languages
149 * Symbols:: Examining the symbol table
150 * Altering:: Altering execution
151 * GDB Files:: @value{GDBN} files
152 * Targets:: Specifying a debugging target
153 * Remote Debugging:: Debugging remote programs
154 * Configurations:: Configuration-specific information
155 * Controlling GDB:: Controlling @value{GDBN}
156 * Extending GDB:: Extending @value{GDBN}
157 * Interpreters:: Command Interpreters
158 * TUI:: @value{GDBN} Text User Interface
159 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
160 * GDB/MI:: @value{GDBN}'s Machine Interface.
161 * Annotations:: @value{GDBN}'s annotation interface.
162 * JIT Interface:: Using the JIT debugging interface.
164 * GDB Bugs:: Reporting bugs in @value{GDBN}
166 * Command Line Editing:: Command Line Editing
167 * Using History Interactively:: Using History Interactively
168 * Formatting Documentation:: How to format and print @value{GDBN} documentation
169 * Installing GDB:: Installing GDB
170 * Maintenance Commands:: Maintenance Commands
171 * Remote Protocol:: GDB Remote Serial Protocol
172 * Agent Expressions:: The GDB Agent Expression Mechanism
173 * Target Descriptions:: How targets can describe themselves to
175 * Operating System Information:: Getting additional information from
177 * Trace File Format:: GDB trace file format
178 * Copying:: GNU General Public License says
179 how you can copy and share GDB
180 * GNU Free Documentation License:: The license for this documentation
189 @unnumbered Summary of @value{GDBN}
191 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
192 going on ``inside'' another program while it executes---or what another
193 program was doing at the moment it crashed.
195 @value{GDBN} can do four main kinds of things (plus other things in support of
196 these) to help you catch bugs in the act:
200 Start your program, specifying anything that might affect its behavior.
203 Make your program stop on specified conditions.
206 Examine what has happened, when your program has stopped.
209 Change things in your program, so you can experiment with correcting the
210 effects of one bug and go on to learn about another.
213 You can use @value{GDBN} to debug programs written in C and C@t{++}.
214 For more information, see @ref{Supported Languages,,Supported Languages}.
215 For more information, see @ref{C,,C and C++}.
218 Support for Modula-2 is partial. For information on Modula-2, see
219 @ref{Modula-2,,Modula-2}.
222 Debugging Pascal programs which use sets, subranges, file variables, or
223 nested functions does not currently work. @value{GDBN} does not support
224 entering expressions, printing values, or similar features using Pascal
228 @value{GDBN} can be used to debug programs written in Fortran, although
229 it may be necessary to refer to some variables with a trailing
232 @value{GDBN} can be used to debug programs written in Objective-C,
233 using either the Apple/NeXT or the GNU Objective-C runtime.
236 * Free Software:: Freely redistributable software
237 * Contributors:: Contributors to GDB
241 @unnumberedsec Free Software
243 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
244 General Public License
245 (GPL). The GPL gives you the freedom to copy or adapt a licensed
246 program---but every person getting a copy also gets with it the
247 freedom to modify that copy (which means that they must get access to
248 the source code), and the freedom to distribute further copies.
249 Typical software companies use copyrights to limit your freedoms; the
250 Free Software Foundation uses the GPL to preserve these freedoms.
252 Fundamentally, the General Public License is a license which says that
253 you have these freedoms and that you cannot take these freedoms away
256 @unnumberedsec Free Software Needs Free Documentation
258 The biggest deficiency in the free software community today is not in
259 the software---it is the lack of good free documentation that we can
260 include with the free software. Many of our most important
261 programs do not come with free reference manuals and free introductory
262 texts. Documentation is an essential part of any software package;
263 when an important free software package does not come with a free
264 manual and a free tutorial, that is a major gap. We have many such
267 Consider Perl, for instance. The tutorial manuals that people
268 normally use are non-free. How did this come about? Because the
269 authors of those manuals published them with restrictive terms---no
270 copying, no modification, source files not available---which exclude
271 them from the free software world.
273 That wasn't the first time this sort of thing happened, and it was far
274 from the last. Many times we have heard a GNU user eagerly describe a
275 manual that he is writing, his intended contribution to the community,
276 only to learn that he had ruined everything by signing a publication
277 contract to make it non-free.
279 Free documentation, like free software, is a matter of freedom, not
280 price. The problem with the non-free manual is not that publishers
281 charge a price for printed copies---that in itself is fine. (The Free
282 Software Foundation sells printed copies of manuals, too.) The
283 problem is the restrictions on the use of the manual. Free manuals
284 are available in source code form, and give you permission to copy and
285 modify. Non-free manuals do not allow this.
287 The criteria of freedom for a free manual are roughly the same as for
288 free software. Redistribution (including the normal kinds of
289 commercial redistribution) must be permitted, so that the manual can
290 accompany every copy of the program, both on-line and on paper.
292 Permission for modification of the technical content is crucial too.
293 When people modify the software, adding or changing features, if they
294 are conscientious they will change the manual too---so they can
295 provide accurate and clear documentation for the modified program. A
296 manual that leaves you no choice but to write a new manual to document
297 a changed version of the program is not really available to our
300 Some kinds of limits on the way modification is handled are
301 acceptable. For example, requirements to preserve the original
302 author's copyright notice, the distribution terms, or the list of
303 authors, are ok. It is also no problem to require modified versions
304 to include notice that they were modified. Even entire sections that
305 may not be deleted or changed are acceptable, as long as they deal
306 with nontechnical topics (like this one). These kinds of restrictions
307 are acceptable because they don't obstruct the community's normal use
310 However, it must be possible to modify all the @emph{technical}
311 content of the manual, and then distribute the result in all the usual
312 media, through all the usual channels. Otherwise, the restrictions
313 obstruct the use of the manual, it is not free, and we need another
314 manual to replace it.
316 Please spread the word about this issue. Our community continues to
317 lose manuals to proprietary publishing. If we spread the word that
318 free software needs free reference manuals and free tutorials, perhaps
319 the next person who wants to contribute by writing documentation will
320 realize, before it is too late, that only free manuals contribute to
321 the free software community.
323 If you are writing documentation, please insist on publishing it under
324 the GNU Free Documentation License or another free documentation
325 license. Remember that this decision requires your approval---you
326 don't have to let the publisher decide. Some commercial publishers
327 will use a free license if you insist, but they will not propose the
328 option; it is up to you to raise the issue and say firmly that this is
329 what you want. If the publisher you are dealing with refuses, please
330 try other publishers. If you're not sure whether a proposed license
331 is free, write to @email{licensing@@gnu.org}.
333 You can encourage commercial publishers to sell more free, copylefted
334 manuals and tutorials by buying them, and particularly by buying
335 copies from the publishers that paid for their writing or for major
336 improvements. Meanwhile, try to avoid buying non-free documentation
337 at all. Check the distribution terms of a manual before you buy it,
338 and insist that whoever seeks your business must respect your freedom.
339 Check the history of the book, and try to reward the publishers that
340 have paid or pay the authors to work on it.
342 The Free Software Foundation maintains a list of free documentation
343 published by other publishers, at
344 @url{http://www.fsf.org/doc/other-free-books.html}.
347 @unnumberedsec Contributors to @value{GDBN}
349 Richard Stallman was the original author of @value{GDBN}, and of many
350 other @sc{gnu} programs. Many others have contributed to its
351 development. This section attempts to credit major contributors. One
352 of the virtues of free software is that everyone is free to contribute
353 to it; with regret, we cannot actually acknowledge everyone here. The
354 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
355 blow-by-blow account.
357 Changes much prior to version 2.0 are lost in the mists of time.
360 @emph{Plea:} Additions to this section are particularly welcome. If you
361 or your friends (or enemies, to be evenhanded) have been unfairly
362 omitted from this list, we would like to add your names!
365 So that they may not regard their many labors as thankless, we
366 particularly thank those who shepherded @value{GDBN} through major
368 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
369 Jim Blandy (release 4.18);
370 Jason Molenda (release 4.17);
371 Stan Shebs (release 4.14);
372 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
373 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
374 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
375 Jim Kingdon (releases 3.5, 3.4, and 3.3);
376 and Randy Smith (releases 3.2, 3.1, and 3.0).
378 Richard Stallman, assisted at various times by Peter TerMaat, Chris
379 Hanson, and Richard Mlynarik, handled releases through 2.8.
381 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
382 in @value{GDBN}, with significant additional contributions from Per
383 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
384 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
385 much general update work leading to release 3.0).
387 @value{GDBN} uses the BFD subroutine library to examine multiple
388 object-file formats; BFD was a joint project of David V.
389 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
391 David Johnson wrote the original COFF support; Pace Willison did
392 the original support for encapsulated COFF.
394 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
396 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
397 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
399 Jean-Daniel Fekete contributed Sun 386i support.
400 Chris Hanson improved the HP9000 support.
401 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
402 David Johnson contributed Encore Umax support.
403 Jyrki Kuoppala contributed Altos 3068 support.
404 Jeff Law contributed HP PA and SOM support.
405 Keith Packard contributed NS32K support.
406 Doug Rabson contributed Acorn Risc Machine support.
407 Bob Rusk contributed Harris Nighthawk CX-UX support.
408 Chris Smith contributed Convex support (and Fortran debugging).
409 Jonathan Stone contributed Pyramid support.
410 Michael Tiemann contributed SPARC support.
411 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
412 Pace Willison contributed Intel 386 support.
413 Jay Vosburgh contributed Symmetry support.
414 Marko Mlinar contributed OpenRISC 1000 support.
416 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
418 Rich Schaefer and Peter Schauer helped with support of SunOS shared
421 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
422 about several machine instruction sets.
424 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
425 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
426 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
427 and RDI targets, respectively.
429 Brian Fox is the author of the readline libraries providing
430 command-line editing and command history.
432 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
433 Modula-2 support, and contributed the Languages chapter of this manual.
435 Fred Fish wrote most of the support for Unix System Vr4.
436 He also enhanced the command-completion support to cover C@t{++} overloaded
439 Hitachi America (now Renesas America), Ltd. sponsored the support for
440 H8/300, H8/500, and Super-H processors.
442 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
444 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
447 Toshiba sponsored the support for the TX39 Mips processor.
449 Matsushita sponsored the support for the MN10200 and MN10300 processors.
451 Fujitsu sponsored the support for SPARClite and FR30 processors.
453 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
456 Michael Snyder added support for tracepoints.
458 Stu Grossman wrote gdbserver.
460 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
461 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
463 The following people at the Hewlett-Packard Company contributed
464 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
465 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
466 compiler, and the Text User Interface (nee Terminal User Interface):
467 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
468 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
469 provided HP-specific information in this manual.
471 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
472 Robert Hoehne made significant contributions to the DJGPP port.
474 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
475 development since 1991. Cygnus engineers who have worked on @value{GDBN}
476 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
477 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
478 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
479 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
480 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
481 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
482 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
483 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
484 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
485 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
486 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
487 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
488 Zuhn have made contributions both large and small.
490 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
491 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
493 Jim Blandy added support for preprocessor macros, while working for Red
496 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
497 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
498 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
499 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
500 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
501 with the migration of old architectures to this new framework.
503 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
504 unwinder framework, this consisting of a fresh new design featuring
505 frame IDs, independent frame sniffers, and the sentinel frame. Mark
506 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
507 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
508 trad unwinders. The architecture-specific changes, each involving a
509 complete rewrite of the architecture's frame code, were carried out by
510 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
511 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
512 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
513 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
516 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
517 Tensilica, Inc.@: contributed support for Xtensa processors. Others
518 who have worked on the Xtensa port of @value{GDBN} in the past include
519 Steve Tjiang, John Newlin, and Scott Foehner.
521 Michael Eager and staff of Xilinx, Inc., contributed support for the
522 Xilinx MicroBlaze architecture.
525 @chapter A Sample @value{GDBN} Session
527 You can use this manual at your leisure to read all about @value{GDBN}.
528 However, a handful of commands are enough to get started using the
529 debugger. This chapter illustrates those commands.
532 In this sample session, we emphasize user input like this: @b{input},
533 to make it easier to pick out from the surrounding output.
536 @c FIXME: this example may not be appropriate for some configs, where
537 @c FIXME...primary interest is in remote use.
539 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
540 processor) exhibits the following bug: sometimes, when we change its
541 quote strings from the default, the commands used to capture one macro
542 definition within another stop working. In the following short @code{m4}
543 session, we define a macro @code{foo} which expands to @code{0000}; we
544 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
545 same thing. However, when we change the open quote string to
546 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
547 procedure fails to define a new synonym @code{baz}:
556 @b{define(bar,defn(`foo'))}
560 @b{changequote(<QUOTE>,<UNQUOTE>)}
562 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
565 m4: End of input: 0: fatal error: EOF in string
569 Let us use @value{GDBN} to try to see what is going on.
572 $ @b{@value{GDBP} m4}
573 @c FIXME: this falsifies the exact text played out, to permit smallbook
574 @c FIXME... format to come out better.
575 @value{GDBN} is free software and you are welcome to distribute copies
576 of it under certain conditions; type "show copying" to see
578 There is absolutely no warranty for @value{GDBN}; type "show warranty"
581 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
586 @value{GDBN} reads only enough symbol data to know where to find the
587 rest when needed; as a result, the first prompt comes up very quickly.
588 We now tell @value{GDBN} to use a narrower display width than usual, so
589 that examples fit in this manual.
592 (@value{GDBP}) @b{set width 70}
596 We need to see how the @code{m4} built-in @code{changequote} works.
597 Having looked at the source, we know the relevant subroutine is
598 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
599 @code{break} command.
602 (@value{GDBP}) @b{break m4_changequote}
603 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
607 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
608 control; as long as control does not reach the @code{m4_changequote}
609 subroutine, the program runs as usual:
612 (@value{GDBP}) @b{run}
613 Starting program: /work/Editorial/gdb/gnu/m4/m4
621 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
622 suspends execution of @code{m4}, displaying information about the
623 context where it stops.
626 @b{changequote(<QUOTE>,<UNQUOTE>)}
628 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
630 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
634 Now we use the command @code{n} (@code{next}) to advance execution to
635 the next line of the current function.
639 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
644 @code{set_quotes} looks like a promising subroutine. We can go into it
645 by using the command @code{s} (@code{step}) instead of @code{next}.
646 @code{step} goes to the next line to be executed in @emph{any}
647 subroutine, so it steps into @code{set_quotes}.
651 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
653 530 if (lquote != def_lquote)
657 The display that shows the subroutine where @code{m4} is now
658 suspended (and its arguments) is called a stack frame display. It
659 shows a summary of the stack. We can use the @code{backtrace}
660 command (which can also be spelled @code{bt}), to see where we are
661 in the stack as a whole: the @code{backtrace} command displays a
662 stack frame for each active subroutine.
665 (@value{GDBP}) @b{bt}
666 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
668 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
670 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
671 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
673 #4 0x79dc in expand_input () at macro.c:40
674 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
678 We step through a few more lines to see what happens. The first two
679 times, we can use @samp{s}; the next two times we use @code{n} to avoid
680 falling into the @code{xstrdup} subroutine.
684 0x3b5c 532 if (rquote != def_rquote)
686 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
687 def_lquote : xstrdup(lq);
689 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
692 538 len_lquote = strlen(rquote);
696 The last line displayed looks a little odd; we can examine the variables
697 @code{lquote} and @code{rquote} to see if they are in fact the new left
698 and right quotes we specified. We use the command @code{p}
699 (@code{print}) to see their values.
702 (@value{GDBP}) @b{p lquote}
703 $1 = 0x35d40 "<QUOTE>"
704 (@value{GDBP}) @b{p rquote}
705 $2 = 0x35d50 "<UNQUOTE>"
709 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
710 To look at some context, we can display ten lines of source
711 surrounding the current line with the @code{l} (@code{list}) command.
717 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
719 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
722 538 len_lquote = strlen(rquote);
723 539 len_rquote = strlen(lquote);
730 Let us step past the two lines that set @code{len_lquote} and
731 @code{len_rquote}, and then examine the values of those variables.
735 539 len_rquote = strlen(lquote);
738 (@value{GDBP}) @b{p len_lquote}
740 (@value{GDBP}) @b{p len_rquote}
745 That certainly looks wrong, assuming @code{len_lquote} and
746 @code{len_rquote} are meant to be the lengths of @code{lquote} and
747 @code{rquote} respectively. We can set them to better values using
748 the @code{p} command, since it can print the value of
749 any expression---and that expression can include subroutine calls and
753 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
755 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
760 Is that enough to fix the problem of using the new quotes with the
761 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
762 executing with the @code{c} (@code{continue}) command, and then try the
763 example that caused trouble initially:
769 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
776 Success! The new quotes now work just as well as the default ones. The
777 problem seems to have been just the two typos defining the wrong
778 lengths. We allow @code{m4} exit by giving it an EOF as input:
782 Program exited normally.
786 The message @samp{Program exited normally.} is from @value{GDBN}; it
787 indicates @code{m4} has finished executing. We can end our @value{GDBN}
788 session with the @value{GDBN} @code{quit} command.
791 (@value{GDBP}) @b{quit}
795 @chapter Getting In and Out of @value{GDBN}
797 This chapter discusses how to start @value{GDBN}, and how to get out of it.
801 type @samp{@value{GDBP}} to start @value{GDBN}.
803 type @kbd{quit} or @kbd{Ctrl-d} to exit.
807 * Invoking GDB:: How to start @value{GDBN}
808 * Quitting GDB:: How to quit @value{GDBN}
809 * Shell Commands:: How to use shell commands inside @value{GDBN}
810 * Logging Output:: How to log @value{GDBN}'s output to a file
814 @section Invoking @value{GDBN}
816 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
817 @value{GDBN} reads commands from the terminal until you tell it to exit.
819 You can also run @code{@value{GDBP}} with a variety of arguments and options,
820 to specify more of your debugging environment at the outset.
822 The command-line options described here are designed
823 to cover a variety of situations; in some environments, some of these
824 options may effectively be unavailable.
826 The most usual way to start @value{GDBN} is with one argument,
827 specifying an executable program:
830 @value{GDBP} @var{program}
834 You can also start with both an executable program and a core file
838 @value{GDBP} @var{program} @var{core}
841 You can, instead, specify a process ID as a second argument, if you want
842 to debug a running process:
845 @value{GDBP} @var{program} 1234
849 would attach @value{GDBN} to process @code{1234} (unless you also have a file
850 named @file{1234}; @value{GDBN} does check for a core file first).
852 Taking advantage of the second command-line argument requires a fairly
853 complete operating system; when you use @value{GDBN} as a remote
854 debugger attached to a bare board, there may not be any notion of
855 ``process'', and there is often no way to get a core dump. @value{GDBN}
856 will warn you if it is unable to attach or to read core dumps.
858 You can optionally have @code{@value{GDBP}} pass any arguments after the
859 executable file to the inferior using @code{--args}. This option stops
862 @value{GDBP} --args gcc -O2 -c foo.c
864 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
865 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
867 You can run @code{@value{GDBP}} without printing the front material, which describes
868 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
875 You can further control how @value{GDBN} starts up by using command-line
876 options. @value{GDBN} itself can remind you of the options available.
886 to display all available options and briefly describe their use
887 (@samp{@value{GDBP} -h} is a shorter equivalent).
889 All options and command line arguments you give are processed
890 in sequential order. The order makes a difference when the
891 @samp{-x} option is used.
895 * File Options:: Choosing files
896 * Mode Options:: Choosing modes
897 * Startup:: What @value{GDBN} does during startup
901 @subsection Choosing Files
903 When @value{GDBN} starts, it reads any arguments other than options as
904 specifying an executable file and core file (or process ID). This is
905 the same as if the arguments were specified by the @samp{-se} and
906 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
907 first argument that does not have an associated option flag as
908 equivalent to the @samp{-se} option followed by that argument; and the
909 second argument that does not have an associated option flag, if any, as
910 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
911 If the second argument begins with a decimal digit, @value{GDBN} will
912 first attempt to attach to it as a process, and if that fails, attempt
913 to open it as a corefile. If you have a corefile whose name begins with
914 a digit, you can prevent @value{GDBN} from treating it as a pid by
915 prefixing it with @file{./}, e.g.@: @file{./12345}.
917 If @value{GDBN} has not been configured to included core file support,
918 such as for most embedded targets, then it will complain about a second
919 argument and ignore it.
921 Many options have both long and short forms; both are shown in the
922 following list. @value{GDBN} also recognizes the long forms if you truncate
923 them, so long as enough of the option is present to be unambiguous.
924 (If you prefer, you can flag option arguments with @samp{--} rather
925 than @samp{-}, though we illustrate the more usual convention.)
927 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
928 @c way, both those who look for -foo and --foo in the index, will find
932 @item -symbols @var{file}
934 @cindex @code{--symbols}
936 Read symbol table from file @var{file}.
938 @item -exec @var{file}
940 @cindex @code{--exec}
942 Use file @var{file} as the executable file to execute when appropriate,
943 and for examining pure data in conjunction with a core dump.
947 Read symbol table from file @var{file} and use it as the executable
950 @item -core @var{file}
952 @cindex @code{--core}
954 Use file @var{file} as a core dump to examine.
956 @item -pid @var{number}
957 @itemx -p @var{number}
960 Connect to process ID @var{number}, as with the @code{attach} command.
962 @item -command @var{file}
964 @cindex @code{--command}
966 Execute commands from file @var{file}. The contents of this file is
967 evaluated exactly as the @code{source} command would.
968 @xref{Command Files,, Command files}.
970 @item -eval-command @var{command}
971 @itemx -ex @var{command}
972 @cindex @code{--eval-command}
974 Execute a single @value{GDBN} command.
976 This option may be used multiple times to call multiple commands. It may
977 also be interleaved with @samp{-command} as required.
980 @value{GDBP} -ex 'target sim' -ex 'load' \
981 -x setbreakpoints -ex 'run' a.out
984 @item -directory @var{directory}
985 @itemx -d @var{directory}
986 @cindex @code{--directory}
988 Add @var{directory} to the path to search for source and script files.
992 @cindex @code{--readnow}
994 Read each symbol file's entire symbol table immediately, rather than
995 the default, which is to read it incrementally as it is needed.
996 This makes startup slower, but makes future operations faster.
1001 @subsection Choosing Modes
1003 You can run @value{GDBN} in various alternative modes---for example, in
1004 batch mode or quiet mode.
1011 Do not execute commands found in any initialization files. Normally,
1012 @value{GDBN} executes the commands in these files after all the command
1013 options and arguments have been processed. @xref{Command Files,,Command
1019 @cindex @code{--quiet}
1020 @cindex @code{--silent}
1022 ``Quiet''. Do not print the introductory and copyright messages. These
1023 messages are also suppressed in batch mode.
1026 @cindex @code{--batch}
1027 Run in batch mode. Exit with status @code{0} after processing all the
1028 command files specified with @samp{-x} (and all commands from
1029 initialization files, if not inhibited with @samp{-n}). Exit with
1030 nonzero status if an error occurs in executing the @value{GDBN} commands
1031 in the command files. Batch mode also disables pagination;
1032 @pxref{Screen Size} and acts as if @kbd{set confirm off} were in
1033 effect (@pxref{Messages/Warnings}).
1035 Batch mode may be useful for running @value{GDBN} as a filter, for
1036 example to download and run a program on another computer; in order to
1037 make this more useful, the message
1040 Program exited normally.
1044 (which is ordinarily issued whenever a program running under
1045 @value{GDBN} control terminates) is not issued when running in batch
1049 @cindex @code{--batch-silent}
1050 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1051 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1052 unaffected). This is much quieter than @samp{-silent} and would be useless
1053 for an interactive session.
1055 This is particularly useful when using targets that give @samp{Loading section}
1056 messages, for example.
1058 Note that targets that give their output via @value{GDBN}, as opposed to
1059 writing directly to @code{stdout}, will also be made silent.
1061 @item -return-child-result
1062 @cindex @code{--return-child-result}
1063 The return code from @value{GDBN} will be the return code from the child
1064 process (the process being debugged), with the following exceptions:
1068 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1069 internal error. In this case the exit code is the same as it would have been
1070 without @samp{-return-child-result}.
1072 The user quits with an explicit value. E.g., @samp{quit 1}.
1074 The child process never runs, or is not allowed to terminate, in which case
1075 the exit code will be -1.
1078 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1079 when @value{GDBN} is being used as a remote program loader or simulator
1084 @cindex @code{--nowindows}
1086 ``No windows''. If @value{GDBN} comes with a graphical user interface
1087 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1088 interface. If no GUI is available, this option has no effect.
1092 @cindex @code{--windows}
1094 If @value{GDBN} includes a GUI, then this option requires it to be
1097 @item -cd @var{directory}
1099 Run @value{GDBN} using @var{directory} as its working directory,
1100 instead of the current directory.
1104 @cindex @code{--fullname}
1106 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1107 subprocess. It tells @value{GDBN} to output the full file name and line
1108 number in a standard, recognizable fashion each time a stack frame is
1109 displayed (which includes each time your program stops). This
1110 recognizable format looks like two @samp{\032} characters, followed by
1111 the file name, line number and character position separated by colons,
1112 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1113 @samp{\032} characters as a signal to display the source code for the
1117 @cindex @code{--epoch}
1118 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1119 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1120 routines so as to allow Epoch to display values of expressions in a
1123 @item -annotate @var{level}
1124 @cindex @code{--annotate}
1125 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1126 effect is identical to using @samp{set annotate @var{level}}
1127 (@pxref{Annotations}). The annotation @var{level} controls how much
1128 information @value{GDBN} prints together with its prompt, values of
1129 expressions, source lines, and other types of output. Level 0 is the
1130 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1131 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1132 that control @value{GDBN}, and level 2 has been deprecated.
1134 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1138 @cindex @code{--args}
1139 Change interpretation of command line so that arguments following the
1140 executable file are passed as command line arguments to the inferior.
1141 This option stops option processing.
1143 @item -baud @var{bps}
1145 @cindex @code{--baud}
1147 Set the line speed (baud rate or bits per second) of any serial
1148 interface used by @value{GDBN} for remote debugging.
1150 @item -l @var{timeout}
1152 Set the timeout (in seconds) of any communication used by @value{GDBN}
1153 for remote debugging.
1155 @item -tty @var{device}
1156 @itemx -t @var{device}
1157 @cindex @code{--tty}
1159 Run using @var{device} for your program's standard input and output.
1160 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1162 @c resolve the situation of these eventually
1164 @cindex @code{--tui}
1165 Activate the @dfn{Text User Interface} when starting. The Text User
1166 Interface manages several text windows on the terminal, showing
1167 source, assembly, registers and @value{GDBN} command outputs
1168 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1169 Text User Interface can be enabled by invoking the program
1170 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1171 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1174 @c @cindex @code{--xdb}
1175 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1176 @c For information, see the file @file{xdb_trans.html}, which is usually
1177 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1180 @item -interpreter @var{interp}
1181 @cindex @code{--interpreter}
1182 Use the interpreter @var{interp} for interface with the controlling
1183 program or device. This option is meant to be set by programs which
1184 communicate with @value{GDBN} using it as a back end.
1185 @xref{Interpreters, , Command Interpreters}.
1187 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1188 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1189 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1190 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1191 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1192 @sc{gdb/mi} interfaces are no longer supported.
1195 @cindex @code{--write}
1196 Open the executable and core files for both reading and writing. This
1197 is equivalent to the @samp{set write on} command inside @value{GDBN}
1201 @cindex @code{--statistics}
1202 This option causes @value{GDBN} to print statistics about time and
1203 memory usage after it completes each command and returns to the prompt.
1206 @cindex @code{--version}
1207 This option causes @value{GDBN} to print its version number and
1208 no-warranty blurb, and exit.
1213 @subsection What @value{GDBN} Does During Startup
1214 @cindex @value{GDBN} startup
1216 Here's the description of what @value{GDBN} does during session startup:
1220 Sets up the command interpreter as specified by the command line
1221 (@pxref{Mode Options, interpreter}).
1225 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1226 used when building @value{GDBN}; @pxref{System-wide configuration,
1227 ,System-wide configuration and settings}) and executes all the commands in
1231 Reads the init file (if any) in your home directory@footnote{On
1232 DOS/Windows systems, the home directory is the one pointed to by the
1233 @code{HOME} environment variable.} and executes all the commands in
1237 Processes command line options and operands.
1240 Reads and executes the commands from init file (if any) in the current
1241 working directory. This is only done if the current directory is
1242 different from your home directory. Thus, you can have more than one
1243 init file, one generic in your home directory, and another, specific
1244 to the program you are debugging, in the directory where you invoke
1248 Reads command files specified by the @samp{-x} option. @xref{Command
1249 Files}, for more details about @value{GDBN} command files.
1252 Reads the command history recorded in the @dfn{history file}.
1253 @xref{Command History}, for more details about the command history and the
1254 files where @value{GDBN} records it.
1257 Init files use the same syntax as @dfn{command files} (@pxref{Command
1258 Files}) and are processed by @value{GDBN} in the same way. The init
1259 file in your home directory can set options (such as @samp{set
1260 complaints}) that affect subsequent processing of command line options
1261 and operands. Init files are not executed if you use the @samp{-nx}
1262 option (@pxref{Mode Options, ,Choosing Modes}).
1264 To display the list of init files loaded by gdb at startup, you
1265 can use @kbd{gdb --help}.
1267 @cindex init file name
1268 @cindex @file{.gdbinit}
1269 @cindex @file{gdb.ini}
1270 The @value{GDBN} init files are normally called @file{.gdbinit}.
1271 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1272 the limitations of file names imposed by DOS filesystems. The Windows
1273 ports of @value{GDBN} use the standard name, but if they find a
1274 @file{gdb.ini} file, they warn you about that and suggest to rename
1275 the file to the standard name.
1279 @section Quitting @value{GDBN}
1280 @cindex exiting @value{GDBN}
1281 @cindex leaving @value{GDBN}
1284 @kindex quit @r{[}@var{expression}@r{]}
1285 @kindex q @r{(@code{quit})}
1286 @item quit @r{[}@var{expression}@r{]}
1288 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1289 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1290 do not supply @var{expression}, @value{GDBN} will terminate normally;
1291 otherwise it will terminate using the result of @var{expression} as the
1296 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1297 terminates the action of any @value{GDBN} command that is in progress and
1298 returns to @value{GDBN} command level. It is safe to type the interrupt
1299 character at any time because @value{GDBN} does not allow it to take effect
1300 until a time when it is safe.
1302 If you have been using @value{GDBN} to control an attached process or
1303 device, you can release it with the @code{detach} command
1304 (@pxref{Attach, ,Debugging an Already-running Process}).
1306 @node Shell Commands
1307 @section Shell Commands
1309 If you need to execute occasional shell commands during your
1310 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1311 just use the @code{shell} command.
1315 @cindex shell escape
1316 @item shell @var{command string}
1317 Invoke a standard shell to execute @var{command string}.
1318 If it exists, the environment variable @code{SHELL} determines which
1319 shell to run. Otherwise @value{GDBN} uses the default shell
1320 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1323 The utility @code{make} is often needed in development environments.
1324 You do not have to use the @code{shell} command for this purpose in
1329 @cindex calling make
1330 @item make @var{make-args}
1331 Execute the @code{make} program with the specified
1332 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1335 @node Logging Output
1336 @section Logging Output
1337 @cindex logging @value{GDBN} output
1338 @cindex save @value{GDBN} output to a file
1340 You may want to save the output of @value{GDBN} commands to a file.
1341 There are several commands to control @value{GDBN}'s logging.
1345 @item set logging on
1347 @item set logging off
1349 @cindex logging file name
1350 @item set logging file @var{file}
1351 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1352 @item set logging overwrite [on|off]
1353 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1354 you want @code{set logging on} to overwrite the logfile instead.
1355 @item set logging redirect [on|off]
1356 By default, @value{GDBN} output will go to both the terminal and the logfile.
1357 Set @code{redirect} if you want output to go only to the log file.
1358 @kindex show logging
1360 Show the current values of the logging settings.
1364 @chapter @value{GDBN} Commands
1366 You can abbreviate a @value{GDBN} command to the first few letters of the command
1367 name, if that abbreviation is unambiguous; and you can repeat certain
1368 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1369 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1370 show you the alternatives available, if there is more than one possibility).
1373 * Command Syntax:: How to give commands to @value{GDBN}
1374 * Completion:: Command completion
1375 * Help:: How to ask @value{GDBN} for help
1378 @node Command Syntax
1379 @section Command Syntax
1381 A @value{GDBN} command is a single line of input. There is no limit on
1382 how long it can be. It starts with a command name, which is followed by
1383 arguments whose meaning depends on the command name. For example, the
1384 command @code{step} accepts an argument which is the number of times to
1385 step, as in @samp{step 5}. You can also use the @code{step} command
1386 with no arguments. Some commands do not allow any arguments.
1388 @cindex abbreviation
1389 @value{GDBN} command names may always be truncated if that abbreviation is
1390 unambiguous. Other possible command abbreviations are listed in the
1391 documentation for individual commands. In some cases, even ambiguous
1392 abbreviations are allowed; for example, @code{s} is specially defined as
1393 equivalent to @code{step} even though there are other commands whose
1394 names start with @code{s}. You can test abbreviations by using them as
1395 arguments to the @code{help} command.
1397 @cindex repeating commands
1398 @kindex RET @r{(repeat last command)}
1399 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1400 repeat the previous command. Certain commands (for example, @code{run})
1401 will not repeat this way; these are commands whose unintentional
1402 repetition might cause trouble and which you are unlikely to want to
1403 repeat. User-defined commands can disable this feature; see
1404 @ref{Define, dont-repeat}.
1406 The @code{list} and @code{x} commands, when you repeat them with
1407 @key{RET}, construct new arguments rather than repeating
1408 exactly as typed. This permits easy scanning of source or memory.
1410 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1411 output, in a way similar to the common utility @code{more}
1412 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1413 @key{RET} too many in this situation, @value{GDBN} disables command
1414 repetition after any command that generates this sort of display.
1416 @kindex # @r{(a comment)}
1418 Any text from a @kbd{#} to the end of the line is a comment; it does
1419 nothing. This is useful mainly in command files (@pxref{Command
1420 Files,,Command Files}).
1422 @cindex repeating command sequences
1423 @kindex Ctrl-o @r{(operate-and-get-next)}
1424 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1425 commands. This command accepts the current line, like @key{RET}, and
1426 then fetches the next line relative to the current line from the history
1430 @section Command Completion
1433 @cindex word completion
1434 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1435 only one possibility; it can also show you what the valid possibilities
1436 are for the next word in a command, at any time. This works for @value{GDBN}
1437 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1439 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1440 of a word. If there is only one possibility, @value{GDBN} fills in the
1441 word, and waits for you to finish the command (or press @key{RET} to
1442 enter it). For example, if you type
1444 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1445 @c complete accuracy in these examples; space introduced for clarity.
1446 @c If texinfo enhancements make it unnecessary, it would be nice to
1447 @c replace " @key" by "@key" in the following...
1449 (@value{GDBP}) info bre @key{TAB}
1453 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1454 the only @code{info} subcommand beginning with @samp{bre}:
1457 (@value{GDBP}) info breakpoints
1461 You can either press @key{RET} at this point, to run the @code{info
1462 breakpoints} command, or backspace and enter something else, if
1463 @samp{breakpoints} does not look like the command you expected. (If you
1464 were sure you wanted @code{info breakpoints} in the first place, you
1465 might as well just type @key{RET} immediately after @samp{info bre},
1466 to exploit command abbreviations rather than command completion).
1468 If there is more than one possibility for the next word when you press
1469 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1470 characters and try again, or just press @key{TAB} a second time;
1471 @value{GDBN} displays all the possible completions for that word. For
1472 example, you might want to set a breakpoint on a subroutine whose name
1473 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1474 just sounds the bell. Typing @key{TAB} again displays all the
1475 function names in your program that begin with those characters, for
1479 (@value{GDBP}) b make_ @key{TAB}
1480 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1481 make_a_section_from_file make_environ
1482 make_abs_section make_function_type
1483 make_blockvector make_pointer_type
1484 make_cleanup make_reference_type
1485 make_command make_symbol_completion_list
1486 (@value{GDBP}) b make_
1490 After displaying the available possibilities, @value{GDBN} copies your
1491 partial input (@samp{b make_} in the example) so you can finish the
1494 If you just want to see the list of alternatives in the first place, you
1495 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1496 means @kbd{@key{META} ?}. You can type this either by holding down a
1497 key designated as the @key{META} shift on your keyboard (if there is
1498 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1500 @cindex quotes in commands
1501 @cindex completion of quoted strings
1502 Sometimes the string you need, while logically a ``word'', may contain
1503 parentheses or other characters that @value{GDBN} normally excludes from
1504 its notion of a word. To permit word completion to work in this
1505 situation, you may enclose words in @code{'} (single quote marks) in
1506 @value{GDBN} commands.
1508 The most likely situation where you might need this is in typing the
1509 name of a C@t{++} function. This is because C@t{++} allows function
1510 overloading (multiple definitions of the same function, distinguished
1511 by argument type). For example, when you want to set a breakpoint you
1512 may need to distinguish whether you mean the version of @code{name}
1513 that takes an @code{int} parameter, @code{name(int)}, or the version
1514 that takes a @code{float} parameter, @code{name(float)}. To use the
1515 word-completion facilities in this situation, type a single quote
1516 @code{'} at the beginning of the function name. This alerts
1517 @value{GDBN} that it may need to consider more information than usual
1518 when you press @key{TAB} or @kbd{M-?} to request word completion:
1521 (@value{GDBP}) b 'bubble( @kbd{M-?}
1522 bubble(double,double) bubble(int,int)
1523 (@value{GDBP}) b 'bubble(
1526 In some cases, @value{GDBN} can tell that completing a name requires using
1527 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1528 completing as much as it can) if you do not type the quote in the first
1532 (@value{GDBP}) b bub @key{TAB}
1533 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1534 (@value{GDBP}) b 'bubble(
1538 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1539 you have not yet started typing the argument list when you ask for
1540 completion on an overloaded symbol.
1542 For more information about overloaded functions, see @ref{C Plus Plus
1543 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1544 overload-resolution off} to disable overload resolution;
1545 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1547 @cindex completion of structure field names
1548 @cindex structure field name completion
1549 @cindex completion of union field names
1550 @cindex union field name completion
1551 When completing in an expression which looks up a field in a
1552 structure, @value{GDBN} also tries@footnote{The completer can be
1553 confused by certain kinds of invalid expressions. Also, it only
1554 examines the static type of the expression, not the dynamic type.} to
1555 limit completions to the field names available in the type of the
1559 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1560 magic to_delete to_fputs to_put to_rewind
1561 to_data to_flush to_isatty to_read to_write
1565 This is because the @code{gdb_stdout} is a variable of the type
1566 @code{struct ui_file} that is defined in @value{GDBN} sources as
1573 ui_file_flush_ftype *to_flush;
1574 ui_file_write_ftype *to_write;
1575 ui_file_fputs_ftype *to_fputs;
1576 ui_file_read_ftype *to_read;
1577 ui_file_delete_ftype *to_delete;
1578 ui_file_isatty_ftype *to_isatty;
1579 ui_file_rewind_ftype *to_rewind;
1580 ui_file_put_ftype *to_put;
1587 @section Getting Help
1588 @cindex online documentation
1591 You can always ask @value{GDBN} itself for information on its commands,
1592 using the command @code{help}.
1595 @kindex h @r{(@code{help})}
1598 You can use @code{help} (abbreviated @code{h}) with no arguments to
1599 display a short list of named classes of commands:
1603 List of classes of commands:
1605 aliases -- Aliases of other commands
1606 breakpoints -- Making program stop at certain points
1607 data -- Examining data
1608 files -- Specifying and examining files
1609 internals -- Maintenance commands
1610 obscure -- Obscure features
1611 running -- Running the program
1612 stack -- Examining the stack
1613 status -- Status inquiries
1614 support -- Support facilities
1615 tracepoints -- Tracing of program execution without
1616 stopping the program
1617 user-defined -- User-defined commands
1619 Type "help" followed by a class name for a list of
1620 commands in that class.
1621 Type "help" followed by command name for full
1623 Command name abbreviations are allowed if unambiguous.
1626 @c the above line break eliminates huge line overfull...
1628 @item help @var{class}
1629 Using one of the general help classes as an argument, you can get a
1630 list of the individual commands in that class. For example, here is the
1631 help display for the class @code{status}:
1634 (@value{GDBP}) help status
1639 @c Line break in "show" line falsifies real output, but needed
1640 @c to fit in smallbook page size.
1641 info -- Generic command for showing things
1642 about the program being debugged
1643 show -- Generic command for showing things
1646 Type "help" followed by command name for full
1648 Command name abbreviations are allowed if unambiguous.
1652 @item help @var{command}
1653 With a command name as @code{help} argument, @value{GDBN} displays a
1654 short paragraph on how to use that command.
1657 @item apropos @var{args}
1658 The @code{apropos} command searches through all of the @value{GDBN}
1659 commands, and their documentation, for the regular expression specified in
1660 @var{args}. It prints out all matches found. For example:
1671 set symbol-reloading -- Set dynamic symbol table reloading
1672 multiple times in one run
1673 show symbol-reloading -- Show dynamic symbol table reloading
1674 multiple times in one run
1679 @item complete @var{args}
1680 The @code{complete @var{args}} command lists all the possible completions
1681 for the beginning of a command. Use @var{args} to specify the beginning of the
1682 command you want completed. For example:
1688 @noindent results in:
1699 @noindent This is intended for use by @sc{gnu} Emacs.
1702 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1703 and @code{show} to inquire about the state of your program, or the state
1704 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1705 manual introduces each of them in the appropriate context. The listings
1706 under @code{info} and under @code{show} in the Index point to
1707 all the sub-commands. @xref{Index}.
1712 @kindex i @r{(@code{info})}
1714 This command (abbreviated @code{i}) is for describing the state of your
1715 program. For example, you can show the arguments passed to a function
1716 with @code{info args}, list the registers currently in use with @code{info
1717 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1718 You can get a complete list of the @code{info} sub-commands with
1719 @w{@code{help info}}.
1723 You can assign the result of an expression to an environment variable with
1724 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1725 @code{set prompt $}.
1729 In contrast to @code{info}, @code{show} is for describing the state of
1730 @value{GDBN} itself.
1731 You can change most of the things you can @code{show}, by using the
1732 related command @code{set}; for example, you can control what number
1733 system is used for displays with @code{set radix}, or simply inquire
1734 which is currently in use with @code{show radix}.
1737 To display all the settable parameters and their current
1738 values, you can use @code{show} with no arguments; you may also use
1739 @code{info set}. Both commands produce the same display.
1740 @c FIXME: "info set" violates the rule that "info" is for state of
1741 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1742 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1746 Here are three miscellaneous @code{show} subcommands, all of which are
1747 exceptional in lacking corresponding @code{set} commands:
1750 @kindex show version
1751 @cindex @value{GDBN} version number
1753 Show what version of @value{GDBN} is running. You should include this
1754 information in @value{GDBN} bug-reports. If multiple versions of
1755 @value{GDBN} are in use at your site, you may need to determine which
1756 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1757 commands are introduced, and old ones may wither away. Also, many
1758 system vendors ship variant versions of @value{GDBN}, and there are
1759 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1760 The version number is the same as the one announced when you start
1763 @kindex show copying
1764 @kindex info copying
1765 @cindex display @value{GDBN} copyright
1768 Display information about permission for copying @value{GDBN}.
1770 @kindex show warranty
1771 @kindex info warranty
1773 @itemx info warranty
1774 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1775 if your version of @value{GDBN} comes with one.
1780 @chapter Running Programs Under @value{GDBN}
1782 When you run a program under @value{GDBN}, you must first generate
1783 debugging information when you compile it.
1785 You may start @value{GDBN} with its arguments, if any, in an environment
1786 of your choice. If you are doing native debugging, you may redirect
1787 your program's input and output, debug an already running process, or
1788 kill a child process.
1791 * Compilation:: Compiling for debugging
1792 * Starting:: Starting your program
1793 * Arguments:: Your program's arguments
1794 * Environment:: Your program's environment
1796 * Working Directory:: Your program's working directory
1797 * Input/Output:: Your program's input and output
1798 * Attach:: Debugging an already-running process
1799 * Kill Process:: Killing the child process
1801 * Inferiors and Programs:: Debugging multiple inferiors and programs
1802 * Threads:: Debugging programs with multiple threads
1803 * Forks:: Debugging forks
1804 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1808 @section Compiling for Debugging
1810 In order to debug a program effectively, you need to generate
1811 debugging information when you compile it. This debugging information
1812 is stored in the object file; it describes the data type of each
1813 variable or function and the correspondence between source line numbers
1814 and addresses in the executable code.
1816 To request debugging information, specify the @samp{-g} option when you run
1819 Programs that are to be shipped to your customers are compiled with
1820 optimizations, using the @samp{-O} compiler option. However, some
1821 compilers are unable to handle the @samp{-g} and @samp{-O} options
1822 together. Using those compilers, you cannot generate optimized
1823 executables containing debugging information.
1825 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1826 without @samp{-O}, making it possible to debug optimized code. We
1827 recommend that you @emph{always} use @samp{-g} whenever you compile a
1828 program. You may think your program is correct, but there is no sense
1829 in pushing your luck. For more information, see @ref{Optimized Code}.
1831 Older versions of the @sc{gnu} C compiler permitted a variant option
1832 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1833 format; if your @sc{gnu} C compiler has this option, do not use it.
1835 @value{GDBN} knows about preprocessor macros and can show you their
1836 expansion (@pxref{Macros}). Most compilers do not include information
1837 about preprocessor macros in the debugging information if you specify
1838 the @option{-g} flag alone, because this information is rather large.
1839 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1840 provides macro information if you specify the options
1841 @option{-gdwarf-2} and @option{-g3}; the former option requests
1842 debugging information in the Dwarf 2 format, and the latter requests
1843 ``extra information''. In the future, we hope to find more compact
1844 ways to represent macro information, so that it can be included with
1849 @section Starting your Program
1855 @kindex r @r{(@code{run})}
1858 Use the @code{run} command to start your program under @value{GDBN}.
1859 You must first specify the program name (except on VxWorks) with an
1860 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1861 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1862 (@pxref{Files, ,Commands to Specify Files}).
1866 If you are running your program in an execution environment that
1867 supports processes, @code{run} creates an inferior process and makes
1868 that process run your program. In some environments without processes,
1869 @code{run} jumps to the start of your program. Other targets,
1870 like @samp{remote}, are always running. If you get an error
1871 message like this one:
1874 The "remote" target does not support "run".
1875 Try "help target" or "continue".
1879 then use @code{continue} to run your program. You may need @code{load}
1880 first (@pxref{load}).
1882 The execution of a program is affected by certain information it
1883 receives from its superior. @value{GDBN} provides ways to specify this
1884 information, which you must do @emph{before} starting your program. (You
1885 can change it after starting your program, but such changes only affect
1886 your program the next time you start it.) This information may be
1887 divided into four categories:
1890 @item The @emph{arguments.}
1891 Specify the arguments to give your program as the arguments of the
1892 @code{run} command. If a shell is available on your target, the shell
1893 is used to pass the arguments, so that you may use normal conventions
1894 (such as wildcard expansion or variable substitution) in describing
1896 In Unix systems, you can control which shell is used with the
1897 @code{SHELL} environment variable.
1898 @xref{Arguments, ,Your Program's Arguments}.
1900 @item The @emph{environment.}
1901 Your program normally inherits its environment from @value{GDBN}, but you can
1902 use the @value{GDBN} commands @code{set environment} and @code{unset
1903 environment} to change parts of the environment that affect
1904 your program. @xref{Environment, ,Your Program's Environment}.
1906 @item The @emph{working directory.}
1907 Your program inherits its working directory from @value{GDBN}. You can set
1908 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1909 @xref{Working Directory, ,Your Program's Working Directory}.
1911 @item The @emph{standard input and output.}
1912 Your program normally uses the same device for standard input and
1913 standard output as @value{GDBN} is using. You can redirect input and output
1914 in the @code{run} command line, or you can use the @code{tty} command to
1915 set a different device for your program.
1916 @xref{Input/Output, ,Your Program's Input and Output}.
1919 @emph{Warning:} While input and output redirection work, you cannot use
1920 pipes to pass the output of the program you are debugging to another
1921 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1925 When you issue the @code{run} command, your program begins to execute
1926 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1927 of how to arrange for your program to stop. Once your program has
1928 stopped, you may call functions in your program, using the @code{print}
1929 or @code{call} commands. @xref{Data, ,Examining Data}.
1931 If the modification time of your symbol file has changed since the last
1932 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1933 table, and reads it again. When it does this, @value{GDBN} tries to retain
1934 your current breakpoints.
1939 @cindex run to main procedure
1940 The name of the main procedure can vary from language to language.
1941 With C or C@t{++}, the main procedure name is always @code{main}, but
1942 other languages such as Ada do not require a specific name for their
1943 main procedure. The debugger provides a convenient way to start the
1944 execution of the program and to stop at the beginning of the main
1945 procedure, depending on the language used.
1947 The @samp{start} command does the equivalent of setting a temporary
1948 breakpoint at the beginning of the main procedure and then invoking
1949 the @samp{run} command.
1951 @cindex elaboration phase
1952 Some programs contain an @dfn{elaboration} phase where some startup code is
1953 executed before the main procedure is called. This depends on the
1954 languages used to write your program. In C@t{++}, for instance,
1955 constructors for static and global objects are executed before
1956 @code{main} is called. It is therefore possible that the debugger stops
1957 before reaching the main procedure. However, the temporary breakpoint
1958 will remain to halt execution.
1960 Specify the arguments to give to your program as arguments to the
1961 @samp{start} command. These arguments will be given verbatim to the
1962 underlying @samp{run} command. Note that the same arguments will be
1963 reused if no argument is provided during subsequent calls to
1964 @samp{start} or @samp{run}.
1966 It is sometimes necessary to debug the program during elaboration. In
1967 these cases, using the @code{start} command would stop the execution of
1968 your program too late, as the program would have already completed the
1969 elaboration phase. Under these circumstances, insert breakpoints in your
1970 elaboration code before running your program.
1972 @kindex set exec-wrapper
1973 @item set exec-wrapper @var{wrapper}
1974 @itemx show exec-wrapper
1975 @itemx unset exec-wrapper
1976 When @samp{exec-wrapper} is set, the specified wrapper is used to
1977 launch programs for debugging. @value{GDBN} starts your program
1978 with a shell command of the form @kbd{exec @var{wrapper}
1979 @var{program}}. Quoting is added to @var{program} and its
1980 arguments, but not to @var{wrapper}, so you should add quotes if
1981 appropriate for your shell. The wrapper runs until it executes
1982 your program, and then @value{GDBN} takes control.
1984 You can use any program that eventually calls @code{execve} with
1985 its arguments as a wrapper. Several standard Unix utilities do
1986 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
1987 with @code{exec "$@@"} will also work.
1989 For example, you can use @code{env} to pass an environment variable to
1990 the debugged program, without setting the variable in your shell's
1994 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
1998 This command is available when debugging locally on most targets, excluding
1999 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2001 @kindex set disable-randomization
2002 @item set disable-randomization
2003 @itemx set disable-randomization on
2004 This option (enabled by default in @value{GDBN}) will turn off the native
2005 randomization of the virtual address space of the started program. This option
2006 is useful for multiple debugging sessions to make the execution better
2007 reproducible and memory addresses reusable across debugging sessions.
2009 This feature is implemented only on @sc{gnu}/Linux. You can get the same
2013 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2016 @item set disable-randomization off
2017 Leave the behavior of the started executable unchanged. Some bugs rear their
2018 ugly heads only when the program is loaded at certain addresses. If your bug
2019 disappears when you run the program under @value{GDBN}, that might be because
2020 @value{GDBN} by default disables the address randomization on platforms, such
2021 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2022 disable-randomization off} to try to reproduce such elusive bugs.
2024 The virtual address space randomization is implemented only on @sc{gnu}/Linux.
2025 It protects the programs against some kinds of security attacks. In these
2026 cases the attacker needs to know the exact location of a concrete executable
2027 code. Randomizing its location makes it impossible to inject jumps misusing
2028 a code at its expected addresses.
2030 Prelinking shared libraries provides a startup performance advantage but it
2031 makes addresses in these libraries predictable for privileged processes by
2032 having just unprivileged access at the target system. Reading the shared
2033 library binary gives enough information for assembling the malicious code
2034 misusing it. Still even a prelinked shared library can get loaded at a new
2035 random address just requiring the regular relocation process during the
2036 startup. Shared libraries not already prelinked are always loaded at
2037 a randomly chosen address.
2039 Position independent executables (PIE) contain position independent code
2040 similar to the shared libraries and therefore such executables get loaded at
2041 a randomly chosen address upon startup. PIE executables always load even
2042 already prelinked shared libraries at a random address. You can build such
2043 executable using @command{gcc -fPIE -pie}.
2045 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2046 (as long as the randomization is enabled).
2048 @item show disable-randomization
2049 Show the current setting of the explicit disable of the native randomization of
2050 the virtual address space of the started program.
2055 @section Your Program's Arguments
2057 @cindex arguments (to your program)
2058 The arguments to your program can be specified by the arguments of the
2060 They are passed to a shell, which expands wildcard characters and
2061 performs redirection of I/O, and thence to your program. Your
2062 @code{SHELL} environment variable (if it exists) specifies what shell
2063 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2064 the default shell (@file{/bin/sh} on Unix).
2066 On non-Unix systems, the program is usually invoked directly by
2067 @value{GDBN}, which emulates I/O redirection via the appropriate system
2068 calls, and the wildcard characters are expanded by the startup code of
2069 the program, not by the shell.
2071 @code{run} with no arguments uses the same arguments used by the previous
2072 @code{run}, or those set by the @code{set args} command.
2077 Specify the arguments to be used the next time your program is run. If
2078 @code{set args} has no arguments, @code{run} executes your program
2079 with no arguments. Once you have run your program with arguments,
2080 using @code{set args} before the next @code{run} is the only way to run
2081 it again without arguments.
2085 Show the arguments to give your program when it is started.
2089 @section Your Program's Environment
2091 @cindex environment (of your program)
2092 The @dfn{environment} consists of a set of environment variables and
2093 their values. Environment variables conventionally record such things as
2094 your user name, your home directory, your terminal type, and your search
2095 path for programs to run. Usually you set up environment variables with
2096 the shell and they are inherited by all the other programs you run. When
2097 debugging, it can be useful to try running your program with a modified
2098 environment without having to start @value{GDBN} over again.
2102 @item path @var{directory}
2103 Add @var{directory} to the front of the @code{PATH} environment variable
2104 (the search path for executables) that will be passed to your program.
2105 The value of @code{PATH} used by @value{GDBN} does not change.
2106 You may specify several directory names, separated by whitespace or by a
2107 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2108 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2109 is moved to the front, so it is searched sooner.
2111 You can use the string @samp{$cwd} to refer to whatever is the current
2112 working directory at the time @value{GDBN} searches the path. If you
2113 use @samp{.} instead, it refers to the directory where you executed the
2114 @code{path} command. @value{GDBN} replaces @samp{.} in the
2115 @var{directory} argument (with the current path) before adding
2116 @var{directory} to the search path.
2117 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2118 @c document that, since repeating it would be a no-op.
2122 Display the list of search paths for executables (the @code{PATH}
2123 environment variable).
2125 @kindex show environment
2126 @item show environment @r{[}@var{varname}@r{]}
2127 Print the value of environment variable @var{varname} to be given to
2128 your program when it starts. If you do not supply @var{varname},
2129 print the names and values of all environment variables to be given to
2130 your program. You can abbreviate @code{environment} as @code{env}.
2132 @kindex set environment
2133 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2134 Set environment variable @var{varname} to @var{value}. The value
2135 changes for your program only, not for @value{GDBN} itself. @var{value} may
2136 be any string; the values of environment variables are just strings, and
2137 any interpretation is supplied by your program itself. The @var{value}
2138 parameter is optional; if it is eliminated, the variable is set to a
2140 @c "any string" here does not include leading, trailing
2141 @c blanks. Gnu asks: does anyone care?
2143 For example, this command:
2150 tells the debugged program, when subsequently run, that its user is named
2151 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2152 are not actually required.)
2154 @kindex unset environment
2155 @item unset environment @var{varname}
2156 Remove variable @var{varname} from the environment to be passed to your
2157 program. This is different from @samp{set env @var{varname} =};
2158 @code{unset environment} removes the variable from the environment,
2159 rather than assigning it an empty value.
2162 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2164 by your @code{SHELL} environment variable if it exists (or
2165 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2166 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2167 @file{.bashrc} for BASH---any variables you set in that file affect
2168 your program. You may wish to move setting of environment variables to
2169 files that are only run when you sign on, such as @file{.login} or
2172 @node Working Directory
2173 @section Your Program's Working Directory
2175 @cindex working directory (of your program)
2176 Each time you start your program with @code{run}, it inherits its
2177 working directory from the current working directory of @value{GDBN}.
2178 The @value{GDBN} working directory is initially whatever it inherited
2179 from its parent process (typically the shell), but you can specify a new
2180 working directory in @value{GDBN} with the @code{cd} command.
2182 The @value{GDBN} working directory also serves as a default for the commands
2183 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2188 @cindex change working directory
2189 @item cd @var{directory}
2190 Set the @value{GDBN} working directory to @var{directory}.
2194 Print the @value{GDBN} working directory.
2197 It is generally impossible to find the current working directory of
2198 the process being debugged (since a program can change its directory
2199 during its run). If you work on a system where @value{GDBN} is
2200 configured with the @file{/proc} support, you can use the @code{info
2201 proc} command (@pxref{SVR4 Process Information}) to find out the
2202 current working directory of the debuggee.
2205 @section Your Program's Input and Output
2210 By default, the program you run under @value{GDBN} does input and output to
2211 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2212 to its own terminal modes to interact with you, but it records the terminal
2213 modes your program was using and switches back to them when you continue
2214 running your program.
2217 @kindex info terminal
2219 Displays information recorded by @value{GDBN} about the terminal modes your
2223 You can redirect your program's input and/or output using shell
2224 redirection with the @code{run} command. For example,
2231 starts your program, diverting its output to the file @file{outfile}.
2234 @cindex controlling terminal
2235 Another way to specify where your program should do input and output is
2236 with the @code{tty} command. This command accepts a file name as
2237 argument, and causes this file to be the default for future @code{run}
2238 commands. It also resets the controlling terminal for the child
2239 process, for future @code{run} commands. For example,
2246 directs that processes started with subsequent @code{run} commands
2247 default to do input and output on the terminal @file{/dev/ttyb} and have
2248 that as their controlling terminal.
2250 An explicit redirection in @code{run} overrides the @code{tty} command's
2251 effect on the input/output device, but not its effect on the controlling
2254 When you use the @code{tty} command or redirect input in the @code{run}
2255 command, only the input @emph{for your program} is affected. The input
2256 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2257 for @code{set inferior-tty}.
2259 @cindex inferior tty
2260 @cindex set inferior controlling terminal
2261 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2262 display the name of the terminal that will be used for future runs of your
2266 @item set inferior-tty /dev/ttyb
2267 @kindex set inferior-tty
2268 Set the tty for the program being debugged to /dev/ttyb.
2270 @item show inferior-tty
2271 @kindex show inferior-tty
2272 Show the current tty for the program being debugged.
2276 @section Debugging an Already-running Process
2281 @item attach @var{process-id}
2282 This command attaches to a running process---one that was started
2283 outside @value{GDBN}. (@code{info files} shows your active
2284 targets.) The command takes as argument a process ID. The usual way to
2285 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2286 or with the @samp{jobs -l} shell command.
2288 @code{attach} does not repeat if you press @key{RET} a second time after
2289 executing the command.
2292 To use @code{attach}, your program must be running in an environment
2293 which supports processes; for example, @code{attach} does not work for
2294 programs on bare-board targets that lack an operating system. You must
2295 also have permission to send the process a signal.
2297 When you use @code{attach}, the debugger finds the program running in
2298 the process first by looking in the current working directory, then (if
2299 the program is not found) by using the source file search path
2300 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2301 the @code{file} command to load the program. @xref{Files, ,Commands to
2304 The first thing @value{GDBN} does after arranging to debug the specified
2305 process is to stop it. You can examine and modify an attached process
2306 with all the @value{GDBN} commands that are ordinarily available when
2307 you start processes with @code{run}. You can insert breakpoints; you
2308 can step and continue; you can modify storage. If you would rather the
2309 process continue running, you may use the @code{continue} command after
2310 attaching @value{GDBN} to the process.
2315 When you have finished debugging the attached process, you can use the
2316 @code{detach} command to release it from @value{GDBN} control. Detaching
2317 the process continues its execution. After the @code{detach} command,
2318 that process and @value{GDBN} become completely independent once more, and you
2319 are ready to @code{attach} another process or start one with @code{run}.
2320 @code{detach} does not repeat if you press @key{RET} again after
2321 executing the command.
2324 If you exit @value{GDBN} while you have an attached process, you detach
2325 that process. If you use the @code{run} command, you kill that process.
2326 By default, @value{GDBN} asks for confirmation if you try to do either of these
2327 things; you can control whether or not you need to confirm by using the
2328 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2332 @section Killing the Child Process
2337 Kill the child process in which your program is running under @value{GDBN}.
2340 This command is useful if you wish to debug a core dump instead of a
2341 running process. @value{GDBN} ignores any core dump file while your program
2344 On some operating systems, a program cannot be executed outside @value{GDBN}
2345 while you have breakpoints set on it inside @value{GDBN}. You can use the
2346 @code{kill} command in this situation to permit running your program
2347 outside the debugger.
2349 The @code{kill} command is also useful if you wish to recompile and
2350 relink your program, since on many systems it is impossible to modify an
2351 executable file while it is running in a process. In this case, when you
2352 next type @code{run}, @value{GDBN} notices that the file has changed, and
2353 reads the symbol table again (while trying to preserve your current
2354 breakpoint settings).
2356 @node Inferiors and Programs
2357 @section Debugging Multiple Inferiors and Programs
2359 @value{GDBN} lets you run and debug multiple programs in a single
2360 session. In addition, @value{GDBN} on some systems may let you run
2361 several programs simultaneously (otherwise you have to exit from one
2362 before starting another). In the most general case, you can have
2363 multiple threads of execution in each of multiple processes, launched
2364 from multiple executables.
2367 @value{GDBN} represents the state of each program execution with an
2368 object called an @dfn{inferior}. An inferior typically corresponds to
2369 a process, but is more general and applies also to targets that do not
2370 have processes. Inferiors may be created before a process runs, and
2371 may be retained after a process exits. Inferiors have unique
2372 identifiers that are different from process ids. Usually each
2373 inferior will also have its own distinct address space, although some
2374 embedded targets may have several inferiors running in different parts
2375 of a single address space. Each inferior may in turn have multiple
2376 threads running in it.
2378 To find out what inferiors exist at any moment, use @w{@code{info
2382 @kindex info inferiors
2383 @item info inferiors
2384 Print a list of all inferiors currently being managed by @value{GDBN}.
2386 @value{GDBN} displays for each inferior (in this order):
2390 the inferior number assigned by @value{GDBN}
2393 the target system's inferior identifier
2396 the name of the executable the inferior is running.
2401 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2402 indicates the current inferior.
2406 @c end table here to get a little more width for example
2409 (@value{GDBP}) info inferiors
2410 Num Description Executable
2411 2 process 2307 hello
2412 * 1 process 3401 goodbye
2415 To switch focus between inferiors, use the @code{inferior} command:
2418 @kindex inferior @var{infno}
2419 @item inferior @var{infno}
2420 Make inferior number @var{infno} the current inferior. The argument
2421 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2422 in the first field of the @samp{info inferiors} display.
2426 You can get multiple executables into a debugging session via the
2427 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2428 systems @value{GDBN} can add inferiors to the debug session
2429 automatically by following calls to @code{fork} and @code{exec}. To
2430 remove inferiors from the debugging session use the
2431 @w{@code{remove-inferior}} command.
2434 @kindex add-inferior
2435 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2436 Adds @var{n} inferiors to be run using @var{executable} as the
2437 executable. @var{n} defaults to 1. If no executable is specified,
2438 the inferiors begins empty, with no program. You can still assign or
2439 change the program assigned to the inferior at any time by using the
2440 @code{file} command with the executable name as its argument.
2442 @kindex clone-inferior
2443 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2444 Adds @var{n} inferiors ready to execute the same program as inferior
2445 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2446 number of the current inferior. This is a convenient command when you
2447 want to run another instance of the inferior you are debugging.
2450 (@value{GDBP}) info inferiors
2451 Num Description Executable
2452 * 1 process 29964 helloworld
2453 (@value{GDBP}) clone-inferior
2456 (@value{GDBP}) info inferiors
2457 Num Description Executable
2459 * 1 process 29964 helloworld
2462 You can now simply switch focus to inferior 2 and run it.
2464 @kindex remove-inferior
2465 @item remove-inferior @var{infno}
2466 Removes the inferior @var{infno}. It is not possible to remove an
2467 inferior that is running with this command. For those, use the
2468 @code{kill} or @code{detach} command first.
2472 To quit debugging one of the running inferiors that is not the current
2473 inferior, you can either detach from it by using the @w{@code{detach
2474 inferior}} command (allowing it to run independently), or kill it
2475 using the @w{@code{kill inferior}} command:
2478 @kindex detach inferior @var{infno}
2479 @item detach inferior @var{infno}
2480 Detach from the inferior identified by @value{GDBN} inferior number
2481 @var{infno}, and remove it from the inferior list.
2483 @kindex kill inferior @var{infno}
2484 @item kill inferior @var{infno}
2485 Kill the inferior identified by @value{GDBN} inferior number
2486 @var{infno}, and remove it from the inferior list.
2489 After the successful completion of a command such as @code{detach},
2490 @code{detach inferior}, @code{kill} or @code{kill inferior}, or after
2491 a normal process exit, the inferior is still valid and listed with
2492 @code{info inferiors}, ready to be restarted.
2495 To be notified when inferiors are started or exit under @value{GDBN}'s
2496 control use @w{@code{set print inferior-events}}:
2499 @kindex set print inferior-events
2500 @cindex print messages on inferior start and exit
2501 @item set print inferior-events
2502 @itemx set print inferior-events on
2503 @itemx set print inferior-events off
2504 The @code{set print inferior-events} command allows you to enable or
2505 disable printing of messages when @value{GDBN} notices that new
2506 inferiors have started or that inferiors have exited or have been
2507 detached. By default, these messages will not be printed.
2509 @kindex show print inferior-events
2510 @item show print inferior-events
2511 Show whether messages will be printed when @value{GDBN} detects that
2512 inferiors have started, exited or have been detached.
2515 Many commands will work the same with multiple programs as with a
2516 single program: e.g., @code{print myglobal} will simply display the
2517 value of @code{myglobal} in the current inferior.
2520 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2521 get more info about the relationship of inferiors, programs, address
2522 spaces in a debug session. You can do that with the @w{@code{maint
2523 info program-spaces}} command.
2526 @kindex maint info program-spaces
2527 @item maint info program-spaces
2528 Print a list of all program spaces currently being managed by
2531 @value{GDBN} displays for each program space (in this order):
2535 the program space number assigned by @value{GDBN}
2538 the name of the executable loaded into the program space, with e.g.,
2539 the @code{file} command.
2544 An asterisk @samp{*} preceding the @value{GDBN} program space number
2545 indicates the current program space.
2547 In addition, below each program space line, @value{GDBN} prints extra
2548 information that isn't suitable to display in tabular form. For
2549 example, the list of inferiors bound to the program space.
2552 (@value{GDBP}) maint info program-spaces
2555 Bound inferiors: ID 1 (process 21561)
2559 Here we can see that no inferior is running the program @code{hello},
2560 while @code{process 21561} is running the program @code{goodbye}. On
2561 some targets, it is possible that multiple inferiors are bound to the
2562 same program space. The most common example is that of debugging both
2563 the parent and child processes of a @code{vfork} call. For example,
2566 (@value{GDBP}) maint info program-spaces
2569 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2572 Here, both inferior 2 and inferior 1 are running in the same program
2573 space as a result of inferior 1 having executed a @code{vfork} call.
2577 @section Debugging Programs with Multiple Threads
2579 @cindex threads of execution
2580 @cindex multiple threads
2581 @cindex switching threads
2582 In some operating systems, such as HP-UX and Solaris, a single program
2583 may have more than one @dfn{thread} of execution. The precise semantics
2584 of threads differ from one operating system to another, but in general
2585 the threads of a single program are akin to multiple processes---except
2586 that they share one address space (that is, they can all examine and
2587 modify the same variables). On the other hand, each thread has its own
2588 registers and execution stack, and perhaps private memory.
2590 @value{GDBN} provides these facilities for debugging multi-thread
2594 @item automatic notification of new threads
2595 @item @samp{thread @var{threadno}}, a command to switch among threads
2596 @item @samp{info threads}, a command to inquire about existing threads
2597 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2598 a command to apply a command to a list of threads
2599 @item thread-specific breakpoints
2600 @item @samp{set print thread-events}, which controls printing of
2601 messages on thread start and exit.
2602 @item @samp{set libthread-db-search-path @var{path}}, which lets
2603 the user specify which @code{libthread_db} to use if the default choice
2604 isn't compatible with the program.
2608 @emph{Warning:} These facilities are not yet available on every
2609 @value{GDBN} configuration where the operating system supports threads.
2610 If your @value{GDBN} does not support threads, these commands have no
2611 effect. For example, a system without thread support shows no output
2612 from @samp{info threads}, and always rejects the @code{thread} command,
2616 (@value{GDBP}) info threads
2617 (@value{GDBP}) thread 1
2618 Thread ID 1 not known. Use the "info threads" command to
2619 see the IDs of currently known threads.
2621 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2622 @c doesn't support threads"?
2625 @cindex focus of debugging
2626 @cindex current thread
2627 The @value{GDBN} thread debugging facility allows you to observe all
2628 threads while your program runs---but whenever @value{GDBN} takes
2629 control, one thread in particular is always the focus of debugging.
2630 This thread is called the @dfn{current thread}. Debugging commands show
2631 program information from the perspective of the current thread.
2633 @cindex @code{New} @var{systag} message
2634 @cindex thread identifier (system)
2635 @c FIXME-implementors!! It would be more helpful if the [New...] message
2636 @c included GDB's numeric thread handle, so you could just go to that
2637 @c thread without first checking `info threads'.
2638 Whenever @value{GDBN} detects a new thread in your program, it displays
2639 the target system's identification for the thread with a message in the
2640 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2641 whose form varies depending on the particular system. For example, on
2642 @sc{gnu}/Linux, you might see
2645 [New Thread 46912507313328 (LWP 25582)]
2649 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2650 the @var{systag} is simply something like @samp{process 368}, with no
2653 @c FIXME!! (1) Does the [New...] message appear even for the very first
2654 @c thread of a program, or does it only appear for the
2655 @c second---i.e.@: when it becomes obvious we have a multithread
2657 @c (2) *Is* there necessarily a first thread always? Or do some
2658 @c multithread systems permit starting a program with multiple
2659 @c threads ab initio?
2661 @cindex thread number
2662 @cindex thread identifier (GDB)
2663 For debugging purposes, @value{GDBN} associates its own thread
2664 number---always a single integer---with each thread in your program.
2667 @kindex info threads
2669 Display a summary of all threads currently in your
2670 program. @value{GDBN} displays for each thread (in this order):
2674 the thread number assigned by @value{GDBN}
2677 the target system's thread identifier (@var{systag})
2680 the current stack frame summary for that thread
2684 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2685 indicates the current thread.
2689 @c end table here to get a little more width for example
2692 (@value{GDBP}) info threads
2693 3 process 35 thread 27 0x34e5 in sigpause ()
2694 2 process 35 thread 23 0x34e5 in sigpause ()
2695 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2701 @cindex debugging multithreaded programs (on HP-UX)
2702 @cindex thread identifier (GDB), on HP-UX
2703 For debugging purposes, @value{GDBN} associates its own thread
2704 number---a small integer assigned in thread-creation order---with each
2705 thread in your program.
2707 @cindex @code{New} @var{systag} message, on HP-UX
2708 @cindex thread identifier (system), on HP-UX
2709 @c FIXME-implementors!! It would be more helpful if the [New...] message
2710 @c included GDB's numeric thread handle, so you could just go to that
2711 @c thread without first checking `info threads'.
2712 Whenever @value{GDBN} detects a new thread in your program, it displays
2713 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2714 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2715 whose form varies depending on the particular system. For example, on
2719 [New thread 2 (system thread 26594)]
2723 when @value{GDBN} notices a new thread.
2726 @kindex info threads (HP-UX)
2728 Display a summary of all threads currently in your
2729 program. @value{GDBN} displays for each thread (in this order):
2732 @item the thread number assigned by @value{GDBN}
2734 @item the target system's thread identifier (@var{systag})
2736 @item the current stack frame summary for that thread
2740 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2741 indicates the current thread.
2745 @c end table here to get a little more width for example
2748 (@value{GDBP}) info threads
2749 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2751 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2752 from /usr/lib/libc.2
2753 1 system thread 27905 0x7b003498 in _brk () \@*
2754 from /usr/lib/libc.2
2757 On Solaris, you can display more information about user threads with a
2758 Solaris-specific command:
2761 @item maint info sol-threads
2762 @kindex maint info sol-threads
2763 @cindex thread info (Solaris)
2764 Display info on Solaris user threads.
2768 @kindex thread @var{threadno}
2769 @item thread @var{threadno}
2770 Make thread number @var{threadno} the current thread. The command
2771 argument @var{threadno} is the internal @value{GDBN} thread number, as
2772 shown in the first field of the @samp{info threads} display.
2773 @value{GDBN} responds by displaying the system identifier of the thread
2774 you selected, and its current stack frame summary:
2777 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2778 (@value{GDBP}) thread 2
2779 [Switching to process 35 thread 23]
2780 0x34e5 in sigpause ()
2784 As with the @samp{[New @dots{}]} message, the form of the text after
2785 @samp{Switching to} depends on your system's conventions for identifying
2788 @kindex thread apply
2789 @cindex apply command to several threads
2790 @item thread apply [@var{threadno}] [@var{all}] @var{command}
2791 The @code{thread apply} command allows you to apply the named
2792 @var{command} to one or more threads. Specify the numbers of the
2793 threads that you want affected with the command argument
2794 @var{threadno}. It can be a single thread number, one of the numbers
2795 shown in the first field of the @samp{info threads} display; or it
2796 could be a range of thread numbers, as in @code{2-4}. To apply a
2797 command to all threads, type @kbd{thread apply all @var{command}}.
2799 @kindex set print thread-events
2800 @cindex print messages on thread start and exit
2801 @item set print thread-events
2802 @itemx set print thread-events on
2803 @itemx set print thread-events off
2804 The @code{set print thread-events} command allows you to enable or
2805 disable printing of messages when @value{GDBN} notices that new threads have
2806 started or that threads have exited. By default, these messages will
2807 be printed if detection of these events is supported by the target.
2808 Note that these messages cannot be disabled on all targets.
2810 @kindex show print thread-events
2811 @item show print thread-events
2812 Show whether messages will be printed when @value{GDBN} detects that threads
2813 have started and exited.
2816 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2817 more information about how @value{GDBN} behaves when you stop and start
2818 programs with multiple threads.
2820 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2821 watchpoints in programs with multiple threads.
2824 @kindex set libthread-db-search-path
2825 @cindex search path for @code{libthread_db}
2826 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2827 If this variable is set, @var{path} is a colon-separated list of
2828 directories @value{GDBN} will use to search for @code{libthread_db}.
2829 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2832 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2833 @code{libthread_db} library to obtain information about threads in the
2834 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2835 to find @code{libthread_db}. If that fails, @value{GDBN} will continue
2836 with default system shared library directories, and finally the directory
2837 from which @code{libpthread} was loaded in the inferior process.
2839 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2840 @value{GDBN} attempts to initialize it with the current inferior process.
2841 If this initialization fails (which could happen because of a version
2842 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2843 will unload @code{libthread_db}, and continue with the next directory.
2844 If none of @code{libthread_db} libraries initialize successfully,
2845 @value{GDBN} will issue a warning and thread debugging will be disabled.
2847 Setting @code{libthread-db-search-path} is currently implemented
2848 only on some platforms.
2850 @kindex show libthread-db-search-path
2851 @item show libthread-db-search-path
2852 Display current libthread_db search path.
2856 @section Debugging Forks
2858 @cindex fork, debugging programs which call
2859 @cindex multiple processes
2860 @cindex processes, multiple
2861 On most systems, @value{GDBN} has no special support for debugging
2862 programs which create additional processes using the @code{fork}
2863 function. When a program forks, @value{GDBN} will continue to debug the
2864 parent process and the child process will run unimpeded. If you have
2865 set a breakpoint in any code which the child then executes, the child
2866 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2867 will cause it to terminate.
2869 However, if you want to debug the child process there is a workaround
2870 which isn't too painful. Put a call to @code{sleep} in the code which
2871 the child process executes after the fork. It may be useful to sleep
2872 only if a certain environment variable is set, or a certain file exists,
2873 so that the delay need not occur when you don't want to run @value{GDBN}
2874 on the child. While the child is sleeping, use the @code{ps} program to
2875 get its process ID. Then tell @value{GDBN} (a new invocation of
2876 @value{GDBN} if you are also debugging the parent process) to attach to
2877 the child process (@pxref{Attach}). From that point on you can debug
2878 the child process just like any other process which you attached to.
2880 On some systems, @value{GDBN} provides support for debugging programs that
2881 create additional processes using the @code{fork} or @code{vfork} functions.
2882 Currently, the only platforms with this feature are HP-UX (11.x and later
2883 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2885 By default, when a program forks, @value{GDBN} will continue to debug
2886 the parent process and the child process will run unimpeded.
2888 If you want to follow the child process instead of the parent process,
2889 use the command @w{@code{set follow-fork-mode}}.
2892 @kindex set follow-fork-mode
2893 @item set follow-fork-mode @var{mode}
2894 Set the debugger response to a program call of @code{fork} or
2895 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2896 process. The @var{mode} argument can be:
2900 The original process is debugged after a fork. The child process runs
2901 unimpeded. This is the default.
2904 The new process is debugged after a fork. The parent process runs
2909 @kindex show follow-fork-mode
2910 @item show follow-fork-mode
2911 Display the current debugger response to a @code{fork} or @code{vfork} call.
2914 @cindex debugging multiple processes
2915 On Linux, if you want to debug both the parent and child processes, use the
2916 command @w{@code{set detach-on-fork}}.
2919 @kindex set detach-on-fork
2920 @item set detach-on-fork @var{mode}
2921 Tells gdb whether to detach one of the processes after a fork, or
2922 retain debugger control over them both.
2926 The child process (or parent process, depending on the value of
2927 @code{follow-fork-mode}) will be detached and allowed to run
2928 independently. This is the default.
2931 Both processes will be held under the control of @value{GDBN}.
2932 One process (child or parent, depending on the value of
2933 @code{follow-fork-mode}) is debugged as usual, while the other
2938 @kindex show detach-on-fork
2939 @item show detach-on-fork
2940 Show whether detach-on-fork mode is on/off.
2943 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
2944 will retain control of all forked processes (including nested forks).
2945 You can list the forked processes under the control of @value{GDBN} by
2946 using the @w{@code{info inferiors}} command, and switch from one fork
2947 to another by using the @code{inferior} command (@pxref{Inferiors and
2948 Programs, ,Debugging Multiple Inferiors and Programs}).
2950 To quit debugging one of the forked processes, you can either detach
2951 from it by using the @w{@code{detach inferior}} command (allowing it
2952 to run independently), or kill it using the @w{@code{kill inferior}}
2953 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
2956 If you ask to debug a child process and a @code{vfork} is followed by an
2957 @code{exec}, @value{GDBN} executes the new target up to the first
2958 breakpoint in the new target. If you have a breakpoint set on
2959 @code{main} in your original program, the breakpoint will also be set on
2960 the child process's @code{main}.
2962 On some systems, when a child process is spawned by @code{vfork}, you
2963 cannot debug the child or parent until an @code{exec} call completes.
2965 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2966 call executes, the new target restarts. To restart the parent
2967 process, use the @code{file} command with the parent executable name
2968 as its argument. By default, after an @code{exec} call executes,
2969 @value{GDBN} discards the symbols of the previous executable image.
2970 You can change this behaviour with the @w{@code{set follow-exec-mode}}
2974 @kindex set follow-exec-mode
2975 @item set follow-exec-mode @var{mode}
2977 Set debugger response to a program call of @code{exec}. An
2978 @code{exec} call replaces the program image of a process.
2980 @code{follow-exec-mode} can be:
2984 @value{GDBN} creates a new inferior and rebinds the process to this
2985 new inferior. The program the process was running before the
2986 @code{exec} call can be restarted afterwards by restarting the
2992 (@value{GDBP}) info inferiors
2994 Id Description Executable
2997 process 12020 is executing new program: prog2
2998 Program exited normally.
2999 (@value{GDBP}) info inferiors
3000 Id Description Executable
3006 @value{GDBN} keeps the process bound to the same inferior. The new
3007 executable image replaces the previous executable loaded in the
3008 inferior. Restarting the inferior after the @code{exec} call, with
3009 e.g., the @code{run} command, restarts the executable the process was
3010 running after the @code{exec} call. This is the default mode.
3015 (@value{GDBP}) info inferiors
3016 Id Description Executable
3019 process 12020 is executing new program: prog2
3020 Program exited normally.
3021 (@value{GDBP}) info inferiors
3022 Id Description Executable
3029 You can use the @code{catch} command to make @value{GDBN} stop whenever
3030 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3031 Catchpoints, ,Setting Catchpoints}.
3033 @node Checkpoint/Restart
3034 @section Setting a @emph{Bookmark} to Return to Later
3039 @cindex snapshot of a process
3040 @cindex rewind program state
3042 On certain operating systems@footnote{Currently, only
3043 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3044 program's state, called a @dfn{checkpoint}, and come back to it
3047 Returning to a checkpoint effectively undoes everything that has
3048 happened in the program since the @code{checkpoint} was saved. This
3049 includes changes in memory, registers, and even (within some limits)
3050 system state. Effectively, it is like going back in time to the
3051 moment when the checkpoint was saved.
3053 Thus, if you're stepping thru a program and you think you're
3054 getting close to the point where things go wrong, you can save
3055 a checkpoint. Then, if you accidentally go too far and miss
3056 the critical statement, instead of having to restart your program
3057 from the beginning, you can just go back to the checkpoint and
3058 start again from there.
3060 This can be especially useful if it takes a lot of time or
3061 steps to reach the point where you think the bug occurs.
3063 To use the @code{checkpoint}/@code{restart} method of debugging:
3068 Save a snapshot of the debugged program's current execution state.
3069 The @code{checkpoint} command takes no arguments, but each checkpoint
3070 is assigned a small integer id, similar to a breakpoint id.
3072 @kindex info checkpoints
3073 @item info checkpoints
3074 List the checkpoints that have been saved in the current debugging
3075 session. For each checkpoint, the following information will be
3082 @item Source line, or label
3085 @kindex restart @var{checkpoint-id}
3086 @item restart @var{checkpoint-id}
3087 Restore the program state that was saved as checkpoint number
3088 @var{checkpoint-id}. All program variables, registers, stack frames
3089 etc.@: will be returned to the values that they had when the checkpoint
3090 was saved. In essence, gdb will ``wind back the clock'' to the point
3091 in time when the checkpoint was saved.
3093 Note that breakpoints, @value{GDBN} variables, command history etc.
3094 are not affected by restoring a checkpoint. In general, a checkpoint
3095 only restores things that reside in the program being debugged, not in
3098 @kindex delete checkpoint @var{checkpoint-id}
3099 @item delete checkpoint @var{checkpoint-id}
3100 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3104 Returning to a previously saved checkpoint will restore the user state
3105 of the program being debugged, plus a significant subset of the system
3106 (OS) state, including file pointers. It won't ``un-write'' data from
3107 a file, but it will rewind the file pointer to the previous location,
3108 so that the previously written data can be overwritten. For files
3109 opened in read mode, the pointer will also be restored so that the
3110 previously read data can be read again.
3112 Of course, characters that have been sent to a printer (or other
3113 external device) cannot be ``snatched back'', and characters received
3114 from eg.@: a serial device can be removed from internal program buffers,
3115 but they cannot be ``pushed back'' into the serial pipeline, ready to
3116 be received again. Similarly, the actual contents of files that have
3117 been changed cannot be restored (at this time).
3119 However, within those constraints, you actually can ``rewind'' your
3120 program to a previously saved point in time, and begin debugging it
3121 again --- and you can change the course of events so as to debug a
3122 different execution path this time.
3124 @cindex checkpoints and process id
3125 Finally, there is one bit of internal program state that will be
3126 different when you return to a checkpoint --- the program's process
3127 id. Each checkpoint will have a unique process id (or @var{pid}),
3128 and each will be different from the program's original @var{pid}.
3129 If your program has saved a local copy of its process id, this could
3130 potentially pose a problem.
3132 @subsection A Non-obvious Benefit of Using Checkpoints
3134 On some systems such as @sc{gnu}/Linux, address space randomization
3135 is performed on new processes for security reasons. This makes it
3136 difficult or impossible to set a breakpoint, or watchpoint, on an
3137 absolute address if you have to restart the program, since the
3138 absolute location of a symbol will change from one execution to the
3141 A checkpoint, however, is an @emph{identical} copy of a process.
3142 Therefore if you create a checkpoint at (eg.@:) the start of main,
3143 and simply return to that checkpoint instead of restarting the
3144 process, you can avoid the effects of address randomization and
3145 your symbols will all stay in the same place.
3148 @chapter Stopping and Continuing
3150 The principal purposes of using a debugger are so that you can stop your
3151 program before it terminates; or so that, if your program runs into
3152 trouble, you can investigate and find out why.
3154 Inside @value{GDBN}, your program may stop for any of several reasons,
3155 such as a signal, a breakpoint, or reaching a new line after a
3156 @value{GDBN} command such as @code{step}. You may then examine and
3157 change variables, set new breakpoints or remove old ones, and then
3158 continue execution. Usually, the messages shown by @value{GDBN} provide
3159 ample explanation of the status of your program---but you can also
3160 explicitly request this information at any time.
3163 @kindex info program
3165 Display information about the status of your program: whether it is
3166 running or not, what process it is, and why it stopped.
3170 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3171 * Continuing and Stepping:: Resuming execution
3173 * Thread Stops:: Stopping and starting multi-thread programs
3177 @section Breakpoints, Watchpoints, and Catchpoints
3180 A @dfn{breakpoint} makes your program stop whenever a certain point in
3181 the program is reached. For each breakpoint, you can add conditions to
3182 control in finer detail whether your program stops. You can set
3183 breakpoints with the @code{break} command and its variants (@pxref{Set
3184 Breaks, ,Setting Breakpoints}), to specify the place where your program
3185 should stop by line number, function name or exact address in the
3188 On some systems, you can set breakpoints in shared libraries before
3189 the executable is run. There is a minor limitation on HP-UX systems:
3190 you must wait until the executable is run in order to set breakpoints
3191 in shared library routines that are not called directly by the program
3192 (for example, routines that are arguments in a @code{pthread_create}
3196 @cindex data breakpoints
3197 @cindex memory tracing
3198 @cindex breakpoint on memory address
3199 @cindex breakpoint on variable modification
3200 A @dfn{watchpoint} is a special breakpoint that stops your program
3201 when the value of an expression changes. The expression may be a value
3202 of a variable, or it could involve values of one or more variables
3203 combined by operators, such as @samp{a + b}. This is sometimes called
3204 @dfn{data breakpoints}. You must use a different command to set
3205 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3206 from that, you can manage a watchpoint like any other breakpoint: you
3207 enable, disable, and delete both breakpoints and watchpoints using the
3210 You can arrange to have values from your program displayed automatically
3211 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3215 @cindex breakpoint on events
3216 A @dfn{catchpoint} is another special breakpoint that stops your program
3217 when a certain kind of event occurs, such as the throwing of a C@t{++}
3218 exception or the loading of a library. As with watchpoints, you use a
3219 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3220 Catchpoints}), but aside from that, you can manage a catchpoint like any
3221 other breakpoint. (To stop when your program receives a signal, use the
3222 @code{handle} command; see @ref{Signals, ,Signals}.)
3224 @cindex breakpoint numbers
3225 @cindex numbers for breakpoints
3226 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3227 catchpoint when you create it; these numbers are successive integers
3228 starting with one. In many of the commands for controlling various
3229 features of breakpoints you use the breakpoint number to say which
3230 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3231 @dfn{disabled}; if disabled, it has no effect on your program until you
3234 @cindex breakpoint ranges
3235 @cindex ranges of breakpoints
3236 Some @value{GDBN} commands accept a range of breakpoints on which to
3237 operate. A breakpoint range is either a single breakpoint number, like
3238 @samp{5}, or two such numbers, in increasing order, separated by a
3239 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3240 all breakpoints in that range are operated on.
3243 * Set Breaks:: Setting breakpoints
3244 * Set Watchpoints:: Setting watchpoints
3245 * Set Catchpoints:: Setting catchpoints
3246 * Delete Breaks:: Deleting breakpoints
3247 * Disabling:: Disabling breakpoints
3248 * Conditions:: Break conditions
3249 * Break Commands:: Breakpoint command lists
3250 * Save Breakpoints:: How to save breakpoints in a file
3251 * Error in Breakpoints:: ``Cannot insert breakpoints''
3252 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3256 @subsection Setting Breakpoints
3258 @c FIXME LMB what does GDB do if no code on line of breakpt?
3259 @c consider in particular declaration with/without initialization.
3261 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3264 @kindex b @r{(@code{break})}
3265 @vindex $bpnum@r{, convenience variable}
3266 @cindex latest breakpoint
3267 Breakpoints are set with the @code{break} command (abbreviated
3268 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3269 number of the breakpoint you've set most recently; see @ref{Convenience
3270 Vars,, Convenience Variables}, for a discussion of what you can do with
3271 convenience variables.
3274 @item break @var{location}
3275 Set a breakpoint at the given @var{location}, which can specify a
3276 function name, a line number, or an address of an instruction.
3277 (@xref{Specify Location}, for a list of all the possible ways to
3278 specify a @var{location}.) The breakpoint will stop your program just
3279 before it executes any of the code in the specified @var{location}.
3281 When using source languages that permit overloading of symbols, such as
3282 C@t{++}, a function name may refer to more than one possible place to break.
3283 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3286 It is also possible to insert a breakpoint that will stop the program
3287 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3288 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3291 When called without any arguments, @code{break} sets a breakpoint at
3292 the next instruction to be executed in the selected stack frame
3293 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3294 innermost, this makes your program stop as soon as control
3295 returns to that frame. This is similar to the effect of a
3296 @code{finish} command in the frame inside the selected frame---except
3297 that @code{finish} does not leave an active breakpoint. If you use
3298 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3299 the next time it reaches the current location; this may be useful
3302 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3303 least one instruction has been executed. If it did not do this, you
3304 would be unable to proceed past a breakpoint without first disabling the
3305 breakpoint. This rule applies whether or not the breakpoint already
3306 existed when your program stopped.
3308 @item break @dots{} if @var{cond}
3309 Set a breakpoint with condition @var{cond}; evaluate the expression
3310 @var{cond} each time the breakpoint is reached, and stop only if the
3311 value is nonzero---that is, if @var{cond} evaluates as true.
3312 @samp{@dots{}} stands for one of the possible arguments described
3313 above (or no argument) specifying where to break. @xref{Conditions,
3314 ,Break Conditions}, for more information on breakpoint conditions.
3317 @item tbreak @var{args}
3318 Set a breakpoint enabled only for one stop. @var{args} are the
3319 same as for the @code{break} command, and the breakpoint is set in the same
3320 way, but the breakpoint is automatically deleted after the first time your
3321 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3324 @cindex hardware breakpoints
3325 @item hbreak @var{args}
3326 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3327 @code{break} command and the breakpoint is set in the same way, but the
3328 breakpoint requires hardware support and some target hardware may not
3329 have this support. The main purpose of this is EPROM/ROM code
3330 debugging, so you can set a breakpoint at an instruction without
3331 changing the instruction. This can be used with the new trap-generation
3332 provided by SPARClite DSU and most x86-based targets. These targets
3333 will generate traps when a program accesses some data or instruction
3334 address that is assigned to the debug registers. However the hardware
3335 breakpoint registers can take a limited number of breakpoints. For
3336 example, on the DSU, only two data breakpoints can be set at a time, and
3337 @value{GDBN} will reject this command if more than two are used. Delete
3338 or disable unused hardware breakpoints before setting new ones
3339 (@pxref{Disabling, ,Disabling Breakpoints}).
3340 @xref{Conditions, ,Break Conditions}.
3341 For remote targets, you can restrict the number of hardware
3342 breakpoints @value{GDBN} will use, see @ref{set remote
3343 hardware-breakpoint-limit}.
3346 @item thbreak @var{args}
3347 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3348 are the same as for the @code{hbreak} command and the breakpoint is set in
3349 the same way. However, like the @code{tbreak} command,
3350 the breakpoint is automatically deleted after the
3351 first time your program stops there. Also, like the @code{hbreak}
3352 command, the breakpoint requires hardware support and some target hardware
3353 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3354 See also @ref{Conditions, ,Break Conditions}.
3357 @cindex regular expression
3358 @cindex breakpoints in functions matching a regexp
3359 @cindex set breakpoints in many functions
3360 @item rbreak @var{regex}
3361 Set breakpoints on all functions matching the regular expression
3362 @var{regex}. This command sets an unconditional breakpoint on all
3363 matches, printing a list of all breakpoints it set. Once these
3364 breakpoints are set, they are treated just like the breakpoints set with
3365 the @code{break} command. You can delete them, disable them, or make
3366 them conditional the same way as any other breakpoint.
3368 The syntax of the regular expression is the standard one used with tools
3369 like @file{grep}. Note that this is different from the syntax used by
3370 shells, so for instance @code{foo*} matches all functions that include
3371 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3372 @code{.*} leading and trailing the regular expression you supply, so to
3373 match only functions that begin with @code{foo}, use @code{^foo}.
3375 @cindex non-member C@t{++} functions, set breakpoint in
3376 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3377 breakpoints on overloaded functions that are not members of any special
3380 @cindex set breakpoints on all functions
3381 The @code{rbreak} command can be used to set breakpoints in
3382 @strong{all} the functions in a program, like this:
3385 (@value{GDBP}) rbreak .
3388 @kindex info breakpoints
3389 @cindex @code{$_} and @code{info breakpoints}
3390 @item info breakpoints @r{[}@var{n}@r{]}
3391 @itemx info break @r{[}@var{n}@r{]}
3392 Print a table of all breakpoints, watchpoints, and catchpoints set and
3393 not deleted. Optional argument @var{n} means print information only
3394 about the specified breakpoint (or watchpoint or catchpoint). For
3395 each breakpoint, following columns are printed:
3398 @item Breakpoint Numbers
3400 Breakpoint, watchpoint, or catchpoint.
3402 Whether the breakpoint is marked to be disabled or deleted when hit.
3403 @item Enabled or Disabled
3404 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3405 that are not enabled.
3407 Where the breakpoint is in your program, as a memory address. For a
3408 pending breakpoint whose address is not yet known, this field will
3409 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3410 library that has the symbol or line referred by breakpoint is loaded.
3411 See below for details. A breakpoint with several locations will
3412 have @samp{<MULTIPLE>} in this field---see below for details.
3414 Where the breakpoint is in the source for your program, as a file and
3415 line number. For a pending breakpoint, the original string passed to
3416 the breakpoint command will be listed as it cannot be resolved until
3417 the appropriate shared library is loaded in the future.
3421 If a breakpoint is conditional, @code{info break} shows the condition on
3422 the line following the affected breakpoint; breakpoint commands, if any,
3423 are listed after that. A pending breakpoint is allowed to have a condition
3424 specified for it. The condition is not parsed for validity until a shared
3425 library is loaded that allows the pending breakpoint to resolve to a
3429 @code{info break} with a breakpoint
3430 number @var{n} as argument lists only that breakpoint. The
3431 convenience variable @code{$_} and the default examining-address for
3432 the @code{x} command are set to the address of the last breakpoint
3433 listed (@pxref{Memory, ,Examining Memory}).
3436 @code{info break} displays a count of the number of times the breakpoint
3437 has been hit. This is especially useful in conjunction with the
3438 @code{ignore} command. You can ignore a large number of breakpoint
3439 hits, look at the breakpoint info to see how many times the breakpoint
3440 was hit, and then run again, ignoring one less than that number. This
3441 will get you quickly to the last hit of that breakpoint.
3444 @value{GDBN} allows you to set any number of breakpoints at the same place in
3445 your program. There is nothing silly or meaningless about this. When
3446 the breakpoints are conditional, this is even useful
3447 (@pxref{Conditions, ,Break Conditions}).
3449 @cindex multiple locations, breakpoints
3450 @cindex breakpoints, multiple locations
3451 It is possible that a breakpoint corresponds to several locations
3452 in your program. Examples of this situation are:
3456 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3457 instances of the function body, used in different cases.
3460 For a C@t{++} template function, a given line in the function can
3461 correspond to any number of instantiations.
3464 For an inlined function, a given source line can correspond to
3465 several places where that function is inlined.
3468 In all those cases, @value{GDBN} will insert a breakpoint at all
3469 the relevant locations@footnote{
3470 As of this writing, multiple-location breakpoints work only if there's
3471 line number information for all the locations. This means that they
3472 will generally not work in system libraries, unless you have debug
3473 info with line numbers for them.}.
3475 A breakpoint with multiple locations is displayed in the breakpoint
3476 table using several rows---one header row, followed by one row for
3477 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3478 address column. The rows for individual locations contain the actual
3479 addresses for locations, and show the functions to which those
3480 locations belong. The number column for a location is of the form
3481 @var{breakpoint-number}.@var{location-number}.
3486 Num Type Disp Enb Address What
3487 1 breakpoint keep y <MULTIPLE>
3489 breakpoint already hit 1 time
3490 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3491 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3494 Each location can be individually enabled or disabled by passing
3495 @var{breakpoint-number}.@var{location-number} as argument to the
3496 @code{enable} and @code{disable} commands. Note that you cannot
3497 delete the individual locations from the list, you can only delete the
3498 entire list of locations that belong to their parent breakpoint (with
3499 the @kbd{delete @var{num}} command, where @var{num} is the number of
3500 the parent breakpoint, 1 in the above example). Disabling or enabling
3501 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3502 that belong to that breakpoint.
3504 @cindex pending breakpoints
3505 It's quite common to have a breakpoint inside a shared library.
3506 Shared libraries can be loaded and unloaded explicitly,
3507 and possibly repeatedly, as the program is executed. To support
3508 this use case, @value{GDBN} updates breakpoint locations whenever
3509 any shared library is loaded or unloaded. Typically, you would
3510 set a breakpoint in a shared library at the beginning of your
3511 debugging session, when the library is not loaded, and when the
3512 symbols from the library are not available. When you try to set
3513 breakpoint, @value{GDBN} will ask you if you want to set
3514 a so called @dfn{pending breakpoint}---breakpoint whose address
3515 is not yet resolved.
3517 After the program is run, whenever a new shared library is loaded,
3518 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3519 shared library contains the symbol or line referred to by some
3520 pending breakpoint, that breakpoint is resolved and becomes an
3521 ordinary breakpoint. When a library is unloaded, all breakpoints
3522 that refer to its symbols or source lines become pending again.
3524 This logic works for breakpoints with multiple locations, too. For
3525 example, if you have a breakpoint in a C@t{++} template function, and
3526 a newly loaded shared library has an instantiation of that template,
3527 a new location is added to the list of locations for the breakpoint.
3529 Except for having unresolved address, pending breakpoints do not
3530 differ from regular breakpoints. You can set conditions or commands,
3531 enable and disable them and perform other breakpoint operations.
3533 @value{GDBN} provides some additional commands for controlling what
3534 happens when the @samp{break} command cannot resolve breakpoint
3535 address specification to an address:
3537 @kindex set breakpoint pending
3538 @kindex show breakpoint pending
3540 @item set breakpoint pending auto
3541 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3542 location, it queries you whether a pending breakpoint should be created.
3544 @item set breakpoint pending on
3545 This indicates that an unrecognized breakpoint location should automatically
3546 result in a pending breakpoint being created.
3548 @item set breakpoint pending off
3549 This indicates that pending breakpoints are not to be created. Any
3550 unrecognized breakpoint location results in an error. This setting does
3551 not affect any pending breakpoints previously created.
3553 @item show breakpoint pending
3554 Show the current behavior setting for creating pending breakpoints.
3557 The settings above only affect the @code{break} command and its
3558 variants. Once breakpoint is set, it will be automatically updated
3559 as shared libraries are loaded and unloaded.
3561 @cindex automatic hardware breakpoints
3562 For some targets, @value{GDBN} can automatically decide if hardware or
3563 software breakpoints should be used, depending on whether the
3564 breakpoint address is read-only or read-write. This applies to
3565 breakpoints set with the @code{break} command as well as to internal
3566 breakpoints set by commands like @code{next} and @code{finish}. For
3567 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3570 You can control this automatic behaviour with the following commands::
3572 @kindex set breakpoint auto-hw
3573 @kindex show breakpoint auto-hw
3575 @item set breakpoint auto-hw on
3576 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3577 will try to use the target memory map to decide if software or hardware
3578 breakpoint must be used.
3580 @item set breakpoint auto-hw off
3581 This indicates @value{GDBN} should not automatically select breakpoint
3582 type. If the target provides a memory map, @value{GDBN} will warn when
3583 trying to set software breakpoint at a read-only address.
3586 @value{GDBN} normally implements breakpoints by replacing the program code
3587 at the breakpoint address with a special instruction, which, when
3588 executed, given control to the debugger. By default, the program
3589 code is so modified only when the program is resumed. As soon as
3590 the program stops, @value{GDBN} restores the original instructions. This
3591 behaviour guards against leaving breakpoints inserted in the
3592 target should gdb abrubptly disconnect. However, with slow remote
3593 targets, inserting and removing breakpoint can reduce the performance.
3594 This behavior can be controlled with the following commands::
3596 @kindex set breakpoint always-inserted
3597 @kindex show breakpoint always-inserted
3599 @item set breakpoint always-inserted off
3600 All breakpoints, including newly added by the user, are inserted in
3601 the target only when the target is resumed. All breakpoints are
3602 removed from the target when it stops.
3604 @item set breakpoint always-inserted on
3605 Causes all breakpoints to be inserted in the target at all times. If
3606 the user adds a new breakpoint, or changes an existing breakpoint, the
3607 breakpoints in the target are updated immediately. A breakpoint is
3608 removed from the target only when breakpoint itself is removed.
3610 @cindex non-stop mode, and @code{breakpoint always-inserted}
3611 @item set breakpoint always-inserted auto
3612 This is the default mode. If @value{GDBN} is controlling the inferior
3613 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3614 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3615 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3616 @code{breakpoint always-inserted} mode is off.
3619 @cindex negative breakpoint numbers
3620 @cindex internal @value{GDBN} breakpoints
3621 @value{GDBN} itself sometimes sets breakpoints in your program for
3622 special purposes, such as proper handling of @code{longjmp} (in C
3623 programs). These internal breakpoints are assigned negative numbers,
3624 starting with @code{-1}; @samp{info breakpoints} does not display them.
3625 You can see these breakpoints with the @value{GDBN} maintenance command
3626 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3629 @node Set Watchpoints
3630 @subsection Setting Watchpoints
3632 @cindex setting watchpoints
3633 You can use a watchpoint to stop execution whenever the value of an
3634 expression changes, without having to predict a particular place where
3635 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3636 The expression may be as simple as the value of a single variable, or
3637 as complex as many variables combined by operators. Examples include:
3641 A reference to the value of a single variable.
3644 An address cast to an appropriate data type. For example,
3645 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3646 address (assuming an @code{int} occupies 4 bytes).
3649 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3650 expression can use any operators valid in the program's native
3651 language (@pxref{Languages}).
3654 You can set a watchpoint on an expression even if the expression can
3655 not be evaluated yet. For instance, you can set a watchpoint on
3656 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3657 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3658 the expression produces a valid value. If the expression becomes
3659 valid in some other way than changing a variable (e.g.@: if the memory
3660 pointed to by @samp{*global_ptr} becomes readable as the result of a
3661 @code{malloc} call), @value{GDBN} may not stop until the next time
3662 the expression changes.
3664 @cindex software watchpoints
3665 @cindex hardware watchpoints
3666 Depending on your system, watchpoints may be implemented in software or
3667 hardware. @value{GDBN} does software watchpointing by single-stepping your
3668 program and testing the variable's value each time, which is hundreds of
3669 times slower than normal execution. (But this may still be worth it, to
3670 catch errors where you have no clue what part of your program is the
3673 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3674 x86-based targets, @value{GDBN} includes support for hardware
3675 watchpoints, which do not slow down the running of your program.
3679 @item watch @var{expr} @r{[}thread @var{threadnum}@r{]}
3680 Set a watchpoint for an expression. @value{GDBN} will break when the
3681 expression @var{expr} is written into by the program and its value
3682 changes. The simplest (and the most popular) use of this command is
3683 to watch the value of a single variable:
3686 (@value{GDBP}) watch foo
3689 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3690 clause, @value{GDBN} breaks only when the thread identified by
3691 @var{threadnum} changes the value of @var{expr}. If any other threads
3692 change the value of @var{expr}, @value{GDBN} will not break. Note
3693 that watchpoints restricted to a single thread in this way only work
3694 with Hardware Watchpoints.
3697 @item rwatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3698 Set a watchpoint that will break when the value of @var{expr} is read
3702 @item awatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3703 Set a watchpoint that will break when @var{expr} is either read from
3704 or written into by the program.
3706 @kindex info watchpoints @r{[}@var{n}@r{]}
3707 @item info watchpoints
3708 This command prints a list of watchpoints, using the same format as
3709 @code{info break} (@pxref{Set Breaks}).
3712 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3713 watchpoints execute very quickly, and the debugger reports a change in
3714 value at the exact instruction where the change occurs. If @value{GDBN}
3715 cannot set a hardware watchpoint, it sets a software watchpoint, which
3716 executes more slowly and reports the change in value at the next
3717 @emph{statement}, not the instruction, after the change occurs.
3719 @cindex use only software watchpoints
3720 You can force @value{GDBN} to use only software watchpoints with the
3721 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3722 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3723 the underlying system supports them. (Note that hardware-assisted
3724 watchpoints that were set @emph{before} setting
3725 @code{can-use-hw-watchpoints} to zero will still use the hardware
3726 mechanism of watching expression values.)
3729 @item set can-use-hw-watchpoints
3730 @kindex set can-use-hw-watchpoints
3731 Set whether or not to use hardware watchpoints.
3733 @item show can-use-hw-watchpoints
3734 @kindex show can-use-hw-watchpoints
3735 Show the current mode of using hardware watchpoints.
3738 For remote targets, you can restrict the number of hardware
3739 watchpoints @value{GDBN} will use, see @ref{set remote
3740 hardware-breakpoint-limit}.
3742 When you issue the @code{watch} command, @value{GDBN} reports
3745 Hardware watchpoint @var{num}: @var{expr}
3749 if it was able to set a hardware watchpoint.
3751 Currently, the @code{awatch} and @code{rwatch} commands can only set
3752 hardware watchpoints, because accesses to data that don't change the
3753 value of the watched expression cannot be detected without examining
3754 every instruction as it is being executed, and @value{GDBN} does not do
3755 that currently. If @value{GDBN} finds that it is unable to set a
3756 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3757 will print a message like this:
3760 Expression cannot be implemented with read/access watchpoint.
3763 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3764 data type of the watched expression is wider than what a hardware
3765 watchpoint on the target machine can handle. For example, some systems
3766 can only watch regions that are up to 4 bytes wide; on such systems you
3767 cannot set hardware watchpoints for an expression that yields a
3768 double-precision floating-point number (which is typically 8 bytes
3769 wide). As a work-around, it might be possible to break the large region
3770 into a series of smaller ones and watch them with separate watchpoints.
3772 If you set too many hardware watchpoints, @value{GDBN} might be unable
3773 to insert all of them when you resume the execution of your program.
3774 Since the precise number of active watchpoints is unknown until such
3775 time as the program is about to be resumed, @value{GDBN} might not be
3776 able to warn you about this when you set the watchpoints, and the
3777 warning will be printed only when the program is resumed:
3780 Hardware watchpoint @var{num}: Could not insert watchpoint
3784 If this happens, delete or disable some of the watchpoints.
3786 Watching complex expressions that reference many variables can also
3787 exhaust the resources available for hardware-assisted watchpoints.
3788 That's because @value{GDBN} needs to watch every variable in the
3789 expression with separately allocated resources.
3791 If you call a function interactively using @code{print} or @code{call},
3792 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3793 kind of breakpoint or the call completes.
3795 @value{GDBN} automatically deletes watchpoints that watch local
3796 (automatic) variables, or expressions that involve such variables, when
3797 they go out of scope, that is, when the execution leaves the block in
3798 which these variables were defined. In particular, when the program
3799 being debugged terminates, @emph{all} local variables go out of scope,
3800 and so only watchpoints that watch global variables remain set. If you
3801 rerun the program, you will need to set all such watchpoints again. One
3802 way of doing that would be to set a code breakpoint at the entry to the
3803 @code{main} function and when it breaks, set all the watchpoints.
3805 @cindex watchpoints and threads
3806 @cindex threads and watchpoints
3807 In multi-threaded programs, watchpoints will detect changes to the
3808 watched expression from every thread.
3811 @emph{Warning:} In multi-threaded programs, software watchpoints
3812 have only limited usefulness. If @value{GDBN} creates a software
3813 watchpoint, it can only watch the value of an expression @emph{in a
3814 single thread}. If you are confident that the expression can only
3815 change due to the current thread's activity (and if you are also
3816 confident that no other thread can become current), then you can use
3817 software watchpoints as usual. However, @value{GDBN} may not notice
3818 when a non-current thread's activity changes the expression. (Hardware
3819 watchpoints, in contrast, watch an expression in all threads.)
3822 @xref{set remote hardware-watchpoint-limit}.
3824 @node Set Catchpoints
3825 @subsection Setting Catchpoints
3826 @cindex catchpoints, setting
3827 @cindex exception handlers
3828 @cindex event handling
3830 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3831 kinds of program events, such as C@t{++} exceptions or the loading of a
3832 shared library. Use the @code{catch} command to set a catchpoint.
3836 @item catch @var{event}
3837 Stop when @var{event} occurs. @var{event} can be any of the following:
3840 @cindex stop on C@t{++} exceptions
3841 The throwing of a C@t{++} exception.
3844 The catching of a C@t{++} exception.
3847 @cindex Ada exception catching
3848 @cindex catch Ada exceptions
3849 An Ada exception being raised. If an exception name is specified
3850 at the end of the command (eg @code{catch exception Program_Error}),
3851 the debugger will stop only when this specific exception is raised.
3852 Otherwise, the debugger stops execution when any Ada exception is raised.
3854 When inserting an exception catchpoint on a user-defined exception whose
3855 name is identical to one of the exceptions defined by the language, the
3856 fully qualified name must be used as the exception name. Otherwise,
3857 @value{GDBN} will assume that it should stop on the pre-defined exception
3858 rather than the user-defined one. For instance, assuming an exception
3859 called @code{Constraint_Error} is defined in package @code{Pck}, then
3860 the command to use to catch such exceptions is @kbd{catch exception
3861 Pck.Constraint_Error}.
3863 @item exception unhandled
3864 An exception that was raised but is not handled by the program.
3867 A failed Ada assertion.
3870 @cindex break on fork/exec
3871 A call to @code{exec}. This is currently only available for HP-UX
3875 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
3876 @cindex break on a system call.
3877 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
3878 syscall is a mechanism for application programs to request a service
3879 from the operating system (OS) or one of the OS system services.
3880 @value{GDBN} can catch some or all of the syscalls issued by the
3881 debuggee, and show the related information for each syscall. If no
3882 argument is specified, calls to and returns from all system calls
3885 @var{name} can be any system call name that is valid for the
3886 underlying OS. Just what syscalls are valid depends on the OS. On
3887 GNU and Unix systems, you can find the full list of valid syscall
3888 names on @file{/usr/include/asm/unistd.h}.
3890 @c For MS-Windows, the syscall names and the corresponding numbers
3891 @c can be found, e.g., on this URL:
3892 @c http://www.metasploit.com/users/opcode/syscalls.html
3893 @c but we don't support Windows syscalls yet.
3895 Normally, @value{GDBN} knows in advance which syscalls are valid for
3896 each OS, so you can use the @value{GDBN} command-line completion
3897 facilities (@pxref{Completion,, command completion}) to list the
3900 You may also specify the system call numerically. A syscall's
3901 number is the value passed to the OS's syscall dispatcher to
3902 identify the requested service. When you specify the syscall by its
3903 name, @value{GDBN} uses its database of syscalls to convert the name
3904 into the corresponding numeric code, but using the number directly
3905 may be useful if @value{GDBN}'s database does not have the complete
3906 list of syscalls on your system (e.g., because @value{GDBN} lags
3907 behind the OS upgrades).
3909 The example below illustrates how this command works if you don't provide
3913 (@value{GDBP}) catch syscall
3914 Catchpoint 1 (syscall)
3916 Starting program: /tmp/catch-syscall
3918 Catchpoint 1 (call to syscall 'close'), \
3919 0xffffe424 in __kernel_vsyscall ()
3923 Catchpoint 1 (returned from syscall 'close'), \
3924 0xffffe424 in __kernel_vsyscall ()
3928 Here is an example of catching a system call by name:
3931 (@value{GDBP}) catch syscall chroot
3932 Catchpoint 1 (syscall 'chroot' [61])
3934 Starting program: /tmp/catch-syscall
3936 Catchpoint 1 (call to syscall 'chroot'), \
3937 0xffffe424 in __kernel_vsyscall ()
3941 Catchpoint 1 (returned from syscall 'chroot'), \
3942 0xffffe424 in __kernel_vsyscall ()
3946 An example of specifying a system call numerically. In the case
3947 below, the syscall number has a corresponding entry in the XML
3948 file, so @value{GDBN} finds its name and prints it:
3951 (@value{GDBP}) catch syscall 252
3952 Catchpoint 1 (syscall(s) 'exit_group')
3954 Starting program: /tmp/catch-syscall
3956 Catchpoint 1 (call to syscall 'exit_group'), \
3957 0xffffe424 in __kernel_vsyscall ()
3961 Program exited normally.
3965 However, there can be situations when there is no corresponding name
3966 in XML file for that syscall number. In this case, @value{GDBN} prints
3967 a warning message saying that it was not able to find the syscall name,
3968 but the catchpoint will be set anyway. See the example below:
3971 (@value{GDBP}) catch syscall 764
3972 warning: The number '764' does not represent a known syscall.
3973 Catchpoint 2 (syscall 764)
3977 If you configure @value{GDBN} using the @samp{--without-expat} option,
3978 it will not be able to display syscall names. Also, if your
3979 architecture does not have an XML file describing its system calls,
3980 you will not be able to see the syscall names. It is important to
3981 notice that these two features are used for accessing the syscall
3982 name database. In either case, you will see a warning like this:
3985 (@value{GDBP}) catch syscall
3986 warning: Could not open "syscalls/i386-linux.xml"
3987 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
3988 GDB will not be able to display syscall names.
3989 Catchpoint 1 (syscall)
3993 Of course, the file name will change depending on your architecture and system.
3995 Still using the example above, you can also try to catch a syscall by its
3996 number. In this case, you would see something like:
3999 (@value{GDBP}) catch syscall 252
4000 Catchpoint 1 (syscall(s) 252)
4003 Again, in this case @value{GDBN} would not be able to display syscall's names.
4006 A call to @code{fork}. This is currently only available for HP-UX
4010 A call to @code{vfork}. This is currently only available for HP-UX
4015 @item tcatch @var{event}
4016 Set a catchpoint that is enabled only for one stop. The catchpoint is
4017 automatically deleted after the first time the event is caught.
4021 Use the @code{info break} command to list the current catchpoints.
4023 There are currently some limitations to C@t{++} exception handling
4024 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
4028 If you call a function interactively, @value{GDBN} normally returns
4029 control to you when the function has finished executing. If the call
4030 raises an exception, however, the call may bypass the mechanism that
4031 returns control to you and cause your program either to abort or to
4032 simply continue running until it hits a breakpoint, catches a signal
4033 that @value{GDBN} is listening for, or exits. This is the case even if
4034 you set a catchpoint for the exception; catchpoints on exceptions are
4035 disabled within interactive calls.
4038 You cannot raise an exception interactively.
4041 You cannot install an exception handler interactively.
4044 @cindex raise exceptions
4045 Sometimes @code{catch} is not the best way to debug exception handling:
4046 if you need to know exactly where an exception is raised, it is better to
4047 stop @emph{before} the exception handler is called, since that way you
4048 can see the stack before any unwinding takes place. If you set a
4049 breakpoint in an exception handler instead, it may not be easy to find
4050 out where the exception was raised.
4052 To stop just before an exception handler is called, you need some
4053 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
4054 raised by calling a library function named @code{__raise_exception}
4055 which has the following ANSI C interface:
4058 /* @var{addr} is where the exception identifier is stored.
4059 @var{id} is the exception identifier. */
4060 void __raise_exception (void **addr, void *id);
4064 To make the debugger catch all exceptions before any stack
4065 unwinding takes place, set a breakpoint on @code{__raise_exception}
4066 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
4068 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
4069 that depends on the value of @var{id}, you can stop your program when
4070 a specific exception is raised. You can use multiple conditional
4071 breakpoints to stop your program when any of a number of exceptions are
4076 @subsection Deleting Breakpoints
4078 @cindex clearing breakpoints, watchpoints, catchpoints
4079 @cindex deleting breakpoints, watchpoints, catchpoints
4080 It is often necessary to eliminate a breakpoint, watchpoint, or
4081 catchpoint once it has done its job and you no longer want your program
4082 to stop there. This is called @dfn{deleting} the breakpoint. A
4083 breakpoint that has been deleted no longer exists; it is forgotten.
4085 With the @code{clear} command you can delete breakpoints according to
4086 where they are in your program. With the @code{delete} command you can
4087 delete individual breakpoints, watchpoints, or catchpoints by specifying
4088 their breakpoint numbers.
4090 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4091 automatically ignores breakpoints on the first instruction to be executed
4092 when you continue execution without changing the execution address.
4097 Delete any breakpoints at the next instruction to be executed in the
4098 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4099 the innermost frame is selected, this is a good way to delete a
4100 breakpoint where your program just stopped.
4102 @item clear @var{location}
4103 Delete any breakpoints set at the specified @var{location}.
4104 @xref{Specify Location}, for the various forms of @var{location}; the
4105 most useful ones are listed below:
4108 @item clear @var{function}
4109 @itemx clear @var{filename}:@var{function}
4110 Delete any breakpoints set at entry to the named @var{function}.
4112 @item clear @var{linenum}
4113 @itemx clear @var{filename}:@var{linenum}
4114 Delete any breakpoints set at or within the code of the specified
4115 @var{linenum} of the specified @var{filename}.
4118 @cindex delete breakpoints
4120 @kindex d @r{(@code{delete})}
4121 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4122 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4123 ranges specified as arguments. If no argument is specified, delete all
4124 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4125 confirm off}). You can abbreviate this command as @code{d}.
4129 @subsection Disabling Breakpoints
4131 @cindex enable/disable a breakpoint
4132 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4133 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4134 it had been deleted, but remembers the information on the breakpoint so
4135 that you can @dfn{enable} it again later.
4137 You disable and enable breakpoints, watchpoints, and catchpoints with
4138 the @code{enable} and @code{disable} commands, optionally specifying
4139 one or more breakpoint numbers as arguments. Use @code{info break} to
4140 print a list of all breakpoints, watchpoints, and catchpoints if you
4141 do not know which numbers to use.
4143 Disabling and enabling a breakpoint that has multiple locations
4144 affects all of its locations.
4146 A breakpoint, watchpoint, or catchpoint can have any of four different
4147 states of enablement:
4151 Enabled. The breakpoint stops your program. A breakpoint set
4152 with the @code{break} command starts out in this state.
4154 Disabled. The breakpoint has no effect on your program.
4156 Enabled once. The breakpoint stops your program, but then becomes
4159 Enabled for deletion. The breakpoint stops your program, but
4160 immediately after it does so it is deleted permanently. A breakpoint
4161 set with the @code{tbreak} command starts out in this state.
4164 You can use the following commands to enable or disable breakpoints,
4165 watchpoints, and catchpoints:
4169 @kindex dis @r{(@code{disable})}
4170 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4171 Disable the specified breakpoints---or all breakpoints, if none are
4172 listed. A disabled breakpoint has no effect but is not forgotten. All
4173 options such as ignore-counts, conditions and commands are remembered in
4174 case the breakpoint is enabled again later. You may abbreviate
4175 @code{disable} as @code{dis}.
4178 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4179 Enable the specified breakpoints (or all defined breakpoints). They
4180 become effective once again in stopping your program.
4182 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4183 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4184 of these breakpoints immediately after stopping your program.
4186 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4187 Enable the specified breakpoints to work once, then die. @value{GDBN}
4188 deletes any of these breakpoints as soon as your program stops there.
4189 Breakpoints set by the @code{tbreak} command start out in this state.
4192 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4193 @c confusing: tbreak is also initially enabled.
4194 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4195 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4196 subsequently, they become disabled or enabled only when you use one of
4197 the commands above. (The command @code{until} can set and delete a
4198 breakpoint of its own, but it does not change the state of your other
4199 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4203 @subsection Break Conditions
4204 @cindex conditional breakpoints
4205 @cindex breakpoint conditions
4207 @c FIXME what is scope of break condition expr? Context where wanted?
4208 @c in particular for a watchpoint?
4209 The simplest sort of breakpoint breaks every time your program reaches a
4210 specified place. You can also specify a @dfn{condition} for a
4211 breakpoint. A condition is just a Boolean expression in your
4212 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4213 a condition evaluates the expression each time your program reaches it,
4214 and your program stops only if the condition is @emph{true}.
4216 This is the converse of using assertions for program validation; in that
4217 situation, you want to stop when the assertion is violated---that is,
4218 when the condition is false. In C, if you want to test an assertion expressed
4219 by the condition @var{assert}, you should set the condition
4220 @samp{! @var{assert}} on the appropriate breakpoint.
4222 Conditions are also accepted for watchpoints; you may not need them,
4223 since a watchpoint is inspecting the value of an expression anyhow---but
4224 it might be simpler, say, to just set a watchpoint on a variable name,
4225 and specify a condition that tests whether the new value is an interesting
4228 Break conditions can have side effects, and may even call functions in
4229 your program. This can be useful, for example, to activate functions
4230 that log program progress, or to use your own print functions to
4231 format special data structures. The effects are completely predictable
4232 unless there is another enabled breakpoint at the same address. (In
4233 that case, @value{GDBN} might see the other breakpoint first and stop your
4234 program without checking the condition of this one.) Note that
4235 breakpoint commands are usually more convenient and flexible than break
4237 purpose of performing side effects when a breakpoint is reached
4238 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4240 Break conditions can be specified when a breakpoint is set, by using
4241 @samp{if} in the arguments to the @code{break} command. @xref{Set
4242 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4243 with the @code{condition} command.
4245 You can also use the @code{if} keyword with the @code{watch} command.
4246 The @code{catch} command does not recognize the @code{if} keyword;
4247 @code{condition} is the only way to impose a further condition on a
4252 @item condition @var{bnum} @var{expression}
4253 Specify @var{expression} as the break condition for breakpoint,
4254 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4255 breakpoint @var{bnum} stops your program only if the value of
4256 @var{expression} is true (nonzero, in C). When you use
4257 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4258 syntactic correctness, and to determine whether symbols in it have
4259 referents in the context of your breakpoint. If @var{expression} uses
4260 symbols not referenced in the context of the breakpoint, @value{GDBN}
4261 prints an error message:
4264 No symbol "foo" in current context.
4269 not actually evaluate @var{expression} at the time the @code{condition}
4270 command (or a command that sets a breakpoint with a condition, like
4271 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4273 @item condition @var{bnum}
4274 Remove the condition from breakpoint number @var{bnum}. It becomes
4275 an ordinary unconditional breakpoint.
4278 @cindex ignore count (of breakpoint)
4279 A special case of a breakpoint condition is to stop only when the
4280 breakpoint has been reached a certain number of times. This is so
4281 useful that there is a special way to do it, using the @dfn{ignore
4282 count} of the breakpoint. Every breakpoint has an ignore count, which
4283 is an integer. Most of the time, the ignore count is zero, and
4284 therefore has no effect. But if your program reaches a breakpoint whose
4285 ignore count is positive, then instead of stopping, it just decrements
4286 the ignore count by one and continues. As a result, if the ignore count
4287 value is @var{n}, the breakpoint does not stop the next @var{n} times
4288 your program reaches it.
4292 @item ignore @var{bnum} @var{count}
4293 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4294 The next @var{count} times the breakpoint is reached, your program's
4295 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4298 To make the breakpoint stop the next time it is reached, specify
4301 When you use @code{continue} to resume execution of your program from a
4302 breakpoint, you can specify an ignore count directly as an argument to
4303 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4304 Stepping,,Continuing and Stepping}.
4306 If a breakpoint has a positive ignore count and a condition, the
4307 condition is not checked. Once the ignore count reaches zero,
4308 @value{GDBN} resumes checking the condition.
4310 You could achieve the effect of the ignore count with a condition such
4311 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4312 is decremented each time. @xref{Convenience Vars, ,Convenience
4316 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4319 @node Break Commands
4320 @subsection Breakpoint Command Lists
4322 @cindex breakpoint commands
4323 You can give any breakpoint (or watchpoint or catchpoint) a series of
4324 commands to execute when your program stops due to that breakpoint. For
4325 example, you might want to print the values of certain expressions, or
4326 enable other breakpoints.
4330 @kindex end@r{ (breakpoint commands)}
4331 @item commands @r{[}@var{range}@dots{}@r{]}
4332 @itemx @dots{} @var{command-list} @dots{}
4334 Specify a list of commands for the given breakpoints. The commands
4335 themselves appear on the following lines. Type a line containing just
4336 @code{end} to terminate the commands.
4338 To remove all commands from a breakpoint, type @code{commands} and
4339 follow it immediately with @code{end}; that is, give no commands.
4341 With no argument, @code{commands} refers to the last breakpoint,
4342 watchpoint, or catchpoint set (not to the breakpoint most recently
4343 encountered). If the most recent breakpoints were set with a single
4344 command, then the @code{commands} will apply to all the breakpoints
4345 set by that command. This applies to breakpoints set by
4346 @code{rbreak}, and also applies when a single @code{break} command
4347 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4351 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4352 disabled within a @var{command-list}.
4354 You can use breakpoint commands to start your program up again. Simply
4355 use the @code{continue} command, or @code{step}, or any other command
4356 that resumes execution.
4358 Any other commands in the command list, after a command that resumes
4359 execution, are ignored. This is because any time you resume execution
4360 (even with a simple @code{next} or @code{step}), you may encounter
4361 another breakpoint---which could have its own command list, leading to
4362 ambiguities about which list to execute.
4365 If the first command you specify in a command list is @code{silent}, the
4366 usual message about stopping at a breakpoint is not printed. This may
4367 be desirable for breakpoints that are to print a specific message and
4368 then continue. If none of the remaining commands print anything, you
4369 see no sign that the breakpoint was reached. @code{silent} is
4370 meaningful only at the beginning of a breakpoint command list.
4372 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4373 print precisely controlled output, and are often useful in silent
4374 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4376 For example, here is how you could use breakpoint commands to print the
4377 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4383 printf "x is %d\n",x
4388 One application for breakpoint commands is to compensate for one bug so
4389 you can test for another. Put a breakpoint just after the erroneous line
4390 of code, give it a condition to detect the case in which something
4391 erroneous has been done, and give it commands to assign correct values
4392 to any variables that need them. End with the @code{continue} command
4393 so that your program does not stop, and start with the @code{silent}
4394 command so that no output is produced. Here is an example:
4405 @node Save Breakpoints
4406 @subsection How to save breakpoints to a file
4408 To save breakpoint definitions to a file use the @w{@code{save
4409 breakpoints}} command.
4412 @kindex save breakpoints
4413 @cindex save breakpoints to a file for future sessions
4414 @item save breakpoints [@var{filename}]
4415 This command saves all current breakpoint definitions together with
4416 their commands and ignore counts, into a file @file{@var{filename}}
4417 suitable for use in a later debugging session. This includes all
4418 types of breakpoints (breakpoints, watchpoints, catchpoints,
4419 tracepoints). To read the saved breakpoint definitions, use the
4420 @code{source} command (@pxref{Command Files}). Note that watchpoints
4421 with expressions involving local variables may fail to be recreated
4422 because it may not be possible to access the context where the
4423 watchpoint is valid anymore. Because the saved breakpoint definitions
4424 are simply a sequence of @value{GDBN} commands that recreate the
4425 breakpoints, you can edit the file in your favorite editing program,
4426 and remove the breakpoint definitions you're not interested in, or
4427 that can no longer be recreated.
4430 @c @ifclear BARETARGET
4431 @node Error in Breakpoints
4432 @subsection ``Cannot insert breakpoints''
4434 If you request too many active hardware-assisted breakpoints and
4435 watchpoints, you will see this error message:
4437 @c FIXME: the precise wording of this message may change; the relevant
4438 @c source change is not committed yet (Sep 3, 1999).
4440 Stopped; cannot insert breakpoints.
4441 You may have requested too many hardware breakpoints and watchpoints.
4445 This message is printed when you attempt to resume the program, since
4446 only then @value{GDBN} knows exactly how many hardware breakpoints and
4447 watchpoints it needs to insert.
4449 When this message is printed, you need to disable or remove some of the
4450 hardware-assisted breakpoints and watchpoints, and then continue.
4452 @node Breakpoint-related Warnings
4453 @subsection ``Breakpoint address adjusted...''
4454 @cindex breakpoint address adjusted
4456 Some processor architectures place constraints on the addresses at
4457 which breakpoints may be placed. For architectures thus constrained,
4458 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4459 with the constraints dictated by the architecture.
4461 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4462 a VLIW architecture in which a number of RISC-like instructions may be
4463 bundled together for parallel execution. The FR-V architecture
4464 constrains the location of a breakpoint instruction within such a
4465 bundle to the instruction with the lowest address. @value{GDBN}
4466 honors this constraint by adjusting a breakpoint's address to the
4467 first in the bundle.
4469 It is not uncommon for optimized code to have bundles which contain
4470 instructions from different source statements, thus it may happen that
4471 a breakpoint's address will be adjusted from one source statement to
4472 another. Since this adjustment may significantly alter @value{GDBN}'s
4473 breakpoint related behavior from what the user expects, a warning is
4474 printed when the breakpoint is first set and also when the breakpoint
4477 A warning like the one below is printed when setting a breakpoint
4478 that's been subject to address adjustment:
4481 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4484 Such warnings are printed both for user settable and @value{GDBN}'s
4485 internal breakpoints. If you see one of these warnings, you should
4486 verify that a breakpoint set at the adjusted address will have the
4487 desired affect. If not, the breakpoint in question may be removed and
4488 other breakpoints may be set which will have the desired behavior.
4489 E.g., it may be sufficient to place the breakpoint at a later
4490 instruction. A conditional breakpoint may also be useful in some
4491 cases to prevent the breakpoint from triggering too often.
4493 @value{GDBN} will also issue a warning when stopping at one of these
4494 adjusted breakpoints:
4497 warning: Breakpoint 1 address previously adjusted from 0x00010414
4501 When this warning is encountered, it may be too late to take remedial
4502 action except in cases where the breakpoint is hit earlier or more
4503 frequently than expected.
4505 @node Continuing and Stepping
4506 @section Continuing and Stepping
4510 @cindex resuming execution
4511 @dfn{Continuing} means resuming program execution until your program
4512 completes normally. In contrast, @dfn{stepping} means executing just
4513 one more ``step'' of your program, where ``step'' may mean either one
4514 line of source code, or one machine instruction (depending on what
4515 particular command you use). Either when continuing or when stepping,
4516 your program may stop even sooner, due to a breakpoint or a signal. (If
4517 it stops due to a signal, you may want to use @code{handle}, or use
4518 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4522 @kindex c @r{(@code{continue})}
4523 @kindex fg @r{(resume foreground execution)}
4524 @item continue @r{[}@var{ignore-count}@r{]}
4525 @itemx c @r{[}@var{ignore-count}@r{]}
4526 @itemx fg @r{[}@var{ignore-count}@r{]}
4527 Resume program execution, at the address where your program last stopped;
4528 any breakpoints set at that address are bypassed. The optional argument
4529 @var{ignore-count} allows you to specify a further number of times to
4530 ignore a breakpoint at this location; its effect is like that of
4531 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4533 The argument @var{ignore-count} is meaningful only when your program
4534 stopped due to a breakpoint. At other times, the argument to
4535 @code{continue} is ignored.
4537 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4538 debugged program is deemed to be the foreground program) are provided
4539 purely for convenience, and have exactly the same behavior as
4543 To resume execution at a different place, you can use @code{return}
4544 (@pxref{Returning, ,Returning from a Function}) to go back to the
4545 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4546 Different Address}) to go to an arbitrary location in your program.
4548 A typical technique for using stepping is to set a breakpoint
4549 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4550 beginning of the function or the section of your program where a problem
4551 is believed to lie, run your program until it stops at that breakpoint,
4552 and then step through the suspect area, examining the variables that are
4553 interesting, until you see the problem happen.
4557 @kindex s @r{(@code{step})}
4559 Continue running your program until control reaches a different source
4560 line, then stop it and return control to @value{GDBN}. This command is
4561 abbreviated @code{s}.
4564 @c "without debugging information" is imprecise; actually "without line
4565 @c numbers in the debugging information". (gcc -g1 has debugging info but
4566 @c not line numbers). But it seems complex to try to make that
4567 @c distinction here.
4568 @emph{Warning:} If you use the @code{step} command while control is
4569 within a function that was compiled without debugging information,
4570 execution proceeds until control reaches a function that does have
4571 debugging information. Likewise, it will not step into a function which
4572 is compiled without debugging information. To step through functions
4573 without debugging information, use the @code{stepi} command, described
4577 The @code{step} command only stops at the first instruction of a source
4578 line. This prevents the multiple stops that could otherwise occur in
4579 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4580 to stop if a function that has debugging information is called within
4581 the line. In other words, @code{step} @emph{steps inside} any functions
4582 called within the line.
4584 Also, the @code{step} command only enters a function if there is line
4585 number information for the function. Otherwise it acts like the
4586 @code{next} command. This avoids problems when using @code{cc -gl}
4587 on MIPS machines. Previously, @code{step} entered subroutines if there
4588 was any debugging information about the routine.
4590 @item step @var{count}
4591 Continue running as in @code{step}, but do so @var{count} times. If a
4592 breakpoint is reached, or a signal not related to stepping occurs before
4593 @var{count} steps, stepping stops right away.
4596 @kindex n @r{(@code{next})}
4597 @item next @r{[}@var{count}@r{]}
4598 Continue to the next source line in the current (innermost) stack frame.
4599 This is similar to @code{step}, but function calls that appear within
4600 the line of code are executed without stopping. Execution stops when
4601 control reaches a different line of code at the original stack level
4602 that was executing when you gave the @code{next} command. This command
4603 is abbreviated @code{n}.
4605 An argument @var{count} is a repeat count, as for @code{step}.
4608 @c FIX ME!! Do we delete this, or is there a way it fits in with
4609 @c the following paragraph? --- Vctoria
4611 @c @code{next} within a function that lacks debugging information acts like
4612 @c @code{step}, but any function calls appearing within the code of the
4613 @c function are executed without stopping.
4615 The @code{next} command only stops at the first instruction of a
4616 source line. This prevents multiple stops that could otherwise occur in
4617 @code{switch} statements, @code{for} loops, etc.
4619 @kindex set step-mode
4621 @cindex functions without line info, and stepping
4622 @cindex stepping into functions with no line info
4623 @itemx set step-mode on
4624 The @code{set step-mode on} command causes the @code{step} command to
4625 stop at the first instruction of a function which contains no debug line
4626 information rather than stepping over it.
4628 This is useful in cases where you may be interested in inspecting the
4629 machine instructions of a function which has no symbolic info and do not
4630 want @value{GDBN} to automatically skip over this function.
4632 @item set step-mode off
4633 Causes the @code{step} command to step over any functions which contains no
4634 debug information. This is the default.
4636 @item show step-mode
4637 Show whether @value{GDBN} will stop in or step over functions without
4638 source line debug information.
4641 @kindex fin @r{(@code{finish})}
4643 Continue running until just after function in the selected stack frame
4644 returns. Print the returned value (if any). This command can be
4645 abbreviated as @code{fin}.
4647 Contrast this with the @code{return} command (@pxref{Returning,
4648 ,Returning from a Function}).
4651 @kindex u @r{(@code{until})}
4652 @cindex run until specified location
4655 Continue running until a source line past the current line, in the
4656 current stack frame, is reached. This command is used to avoid single
4657 stepping through a loop more than once. It is like the @code{next}
4658 command, except that when @code{until} encounters a jump, it
4659 automatically continues execution until the program counter is greater
4660 than the address of the jump.
4662 This means that when you reach the end of a loop after single stepping
4663 though it, @code{until} makes your program continue execution until it
4664 exits the loop. In contrast, a @code{next} command at the end of a loop
4665 simply steps back to the beginning of the loop, which forces you to step
4666 through the next iteration.
4668 @code{until} always stops your program if it attempts to exit the current
4671 @code{until} may produce somewhat counterintuitive results if the order
4672 of machine code does not match the order of the source lines. For
4673 example, in the following excerpt from a debugging session, the @code{f}
4674 (@code{frame}) command shows that execution is stopped at line
4675 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4679 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4681 (@value{GDBP}) until
4682 195 for ( ; argc > 0; NEXTARG) @{
4685 This happened because, for execution efficiency, the compiler had
4686 generated code for the loop closure test at the end, rather than the
4687 start, of the loop---even though the test in a C @code{for}-loop is
4688 written before the body of the loop. The @code{until} command appeared
4689 to step back to the beginning of the loop when it advanced to this
4690 expression; however, it has not really gone to an earlier
4691 statement---not in terms of the actual machine code.
4693 @code{until} with no argument works by means of single
4694 instruction stepping, and hence is slower than @code{until} with an
4697 @item until @var{location}
4698 @itemx u @var{location}
4699 Continue running your program until either the specified location is
4700 reached, or the current stack frame returns. @var{location} is any of
4701 the forms described in @ref{Specify Location}.
4702 This form of the command uses temporary breakpoints, and
4703 hence is quicker than @code{until} without an argument. The specified
4704 location is actually reached only if it is in the current frame. This
4705 implies that @code{until} can be used to skip over recursive function
4706 invocations. For instance in the code below, if the current location is
4707 line @code{96}, issuing @code{until 99} will execute the program up to
4708 line @code{99} in the same invocation of factorial, i.e., after the inner
4709 invocations have returned.
4712 94 int factorial (int value)
4714 96 if (value > 1) @{
4715 97 value *= factorial (value - 1);
4722 @kindex advance @var{location}
4723 @itemx advance @var{location}
4724 Continue running the program up to the given @var{location}. An argument is
4725 required, which should be of one of the forms described in
4726 @ref{Specify Location}.
4727 Execution will also stop upon exit from the current stack
4728 frame. This command is similar to @code{until}, but @code{advance} will
4729 not skip over recursive function calls, and the target location doesn't
4730 have to be in the same frame as the current one.
4734 @kindex si @r{(@code{stepi})}
4736 @itemx stepi @var{arg}
4738 Execute one machine instruction, then stop and return to the debugger.
4740 It is often useful to do @samp{display/i $pc} when stepping by machine
4741 instructions. This makes @value{GDBN} automatically display the next
4742 instruction to be executed, each time your program stops. @xref{Auto
4743 Display,, Automatic Display}.
4745 An argument is a repeat count, as in @code{step}.
4749 @kindex ni @r{(@code{nexti})}
4751 @itemx nexti @var{arg}
4753 Execute one machine instruction, but if it is a function call,
4754 proceed until the function returns.
4756 An argument is a repeat count, as in @code{next}.
4763 A signal is an asynchronous event that can happen in a program. The
4764 operating system defines the possible kinds of signals, and gives each
4765 kind a name and a number. For example, in Unix @code{SIGINT} is the
4766 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4767 @code{SIGSEGV} is the signal a program gets from referencing a place in
4768 memory far away from all the areas in use; @code{SIGALRM} occurs when
4769 the alarm clock timer goes off (which happens only if your program has
4770 requested an alarm).
4772 @cindex fatal signals
4773 Some signals, including @code{SIGALRM}, are a normal part of the
4774 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4775 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4776 program has not specified in advance some other way to handle the signal.
4777 @code{SIGINT} does not indicate an error in your program, but it is normally
4778 fatal so it can carry out the purpose of the interrupt: to kill the program.
4780 @value{GDBN} has the ability to detect any occurrence of a signal in your
4781 program. You can tell @value{GDBN} in advance what to do for each kind of
4784 @cindex handling signals
4785 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4786 @code{SIGALRM} be silently passed to your program
4787 (so as not to interfere with their role in the program's functioning)
4788 but to stop your program immediately whenever an error signal happens.
4789 You can change these settings with the @code{handle} command.
4792 @kindex info signals
4796 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4797 handle each one. You can use this to see the signal numbers of all
4798 the defined types of signals.
4800 @item info signals @var{sig}
4801 Similar, but print information only about the specified signal number.
4803 @code{info handle} is an alias for @code{info signals}.
4806 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4807 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4808 can be the number of a signal or its name (with or without the
4809 @samp{SIG} at the beginning); a list of signal numbers of the form
4810 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4811 known signals. Optional arguments @var{keywords}, described below,
4812 say what change to make.
4816 The keywords allowed by the @code{handle} command can be abbreviated.
4817 Their full names are:
4821 @value{GDBN} should not stop your program when this signal happens. It may
4822 still print a message telling you that the signal has come in.
4825 @value{GDBN} should stop your program when this signal happens. This implies
4826 the @code{print} keyword as well.
4829 @value{GDBN} should print a message when this signal happens.
4832 @value{GDBN} should not mention the occurrence of the signal at all. This
4833 implies the @code{nostop} keyword as well.
4837 @value{GDBN} should allow your program to see this signal; your program
4838 can handle the signal, or else it may terminate if the signal is fatal
4839 and not handled. @code{pass} and @code{noignore} are synonyms.
4843 @value{GDBN} should not allow your program to see this signal.
4844 @code{nopass} and @code{ignore} are synonyms.
4848 When a signal stops your program, the signal is not visible to the
4850 continue. Your program sees the signal then, if @code{pass} is in
4851 effect for the signal in question @emph{at that time}. In other words,
4852 after @value{GDBN} reports a signal, you can use the @code{handle}
4853 command with @code{pass} or @code{nopass} to control whether your
4854 program sees that signal when you continue.
4856 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4857 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4858 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4861 You can also use the @code{signal} command to prevent your program from
4862 seeing a signal, or cause it to see a signal it normally would not see,
4863 or to give it any signal at any time. For example, if your program stopped
4864 due to some sort of memory reference error, you might store correct
4865 values into the erroneous variables and continue, hoping to see more
4866 execution; but your program would probably terminate immediately as
4867 a result of the fatal signal once it saw the signal. To prevent this,
4868 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4871 @cindex extra signal information
4872 @anchor{extra signal information}
4874 On some targets, @value{GDBN} can inspect extra signal information
4875 associated with the intercepted signal, before it is actually
4876 delivered to the program being debugged. This information is exported
4877 by the convenience variable @code{$_siginfo}, and consists of data
4878 that is passed by the kernel to the signal handler at the time of the
4879 receipt of a signal. The data type of the information itself is
4880 target dependent. You can see the data type using the @code{ptype
4881 $_siginfo} command. On Unix systems, it typically corresponds to the
4882 standard @code{siginfo_t} type, as defined in the @file{signal.h}
4885 Here's an example, on a @sc{gnu}/Linux system, printing the stray
4886 referenced address that raised a segmentation fault.
4890 (@value{GDBP}) continue
4891 Program received signal SIGSEGV, Segmentation fault.
4892 0x0000000000400766 in main ()
4894 (@value{GDBP}) ptype $_siginfo
4901 struct @{...@} _kill;
4902 struct @{...@} _timer;
4904 struct @{...@} _sigchld;
4905 struct @{...@} _sigfault;
4906 struct @{...@} _sigpoll;
4909 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
4913 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
4914 $1 = (void *) 0x7ffff7ff7000
4918 Depending on target support, @code{$_siginfo} may also be writable.
4921 @section Stopping and Starting Multi-thread Programs
4923 @cindex stopped threads
4924 @cindex threads, stopped
4926 @cindex continuing threads
4927 @cindex threads, continuing
4929 @value{GDBN} supports debugging programs with multiple threads
4930 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
4931 are two modes of controlling execution of your program within the
4932 debugger. In the default mode, referred to as @dfn{all-stop mode},
4933 when any thread in your program stops (for example, at a breakpoint
4934 or while being stepped), all other threads in the program are also stopped by
4935 @value{GDBN}. On some targets, @value{GDBN} also supports
4936 @dfn{non-stop mode}, in which other threads can continue to run freely while
4937 you examine the stopped thread in the debugger.
4940 * All-Stop Mode:: All threads stop when GDB takes control
4941 * Non-Stop Mode:: Other threads continue to execute
4942 * Background Execution:: Running your program asynchronously
4943 * Thread-Specific Breakpoints:: Controlling breakpoints
4944 * Interrupted System Calls:: GDB may interfere with system calls
4948 @subsection All-Stop Mode
4950 @cindex all-stop mode
4952 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
4953 @emph{all} threads of execution stop, not just the current thread. This
4954 allows you to examine the overall state of the program, including
4955 switching between threads, without worrying that things may change
4958 Conversely, whenever you restart the program, @emph{all} threads start
4959 executing. @emph{This is true even when single-stepping} with commands
4960 like @code{step} or @code{next}.
4962 In particular, @value{GDBN} cannot single-step all threads in lockstep.
4963 Since thread scheduling is up to your debugging target's operating
4964 system (not controlled by @value{GDBN}), other threads may
4965 execute more than one statement while the current thread completes a
4966 single step. Moreover, in general other threads stop in the middle of a
4967 statement, rather than at a clean statement boundary, when the program
4970 You might even find your program stopped in another thread after
4971 continuing or even single-stepping. This happens whenever some other
4972 thread runs into a breakpoint, a signal, or an exception before the
4973 first thread completes whatever you requested.
4975 @cindex automatic thread selection
4976 @cindex switching threads automatically
4977 @cindex threads, automatic switching
4978 Whenever @value{GDBN} stops your program, due to a breakpoint or a
4979 signal, it automatically selects the thread where that breakpoint or
4980 signal happened. @value{GDBN} alerts you to the context switch with a
4981 message such as @samp{[Switching to Thread @var{n}]} to identify the
4984 On some OSes, you can modify @value{GDBN}'s default behavior by
4985 locking the OS scheduler to allow only a single thread to run.
4988 @item set scheduler-locking @var{mode}
4989 @cindex scheduler locking mode
4990 @cindex lock scheduler
4991 Set the scheduler locking mode. If it is @code{off}, then there is no
4992 locking and any thread may run at any time. If @code{on}, then only the
4993 current thread may run when the inferior is resumed. The @code{step}
4994 mode optimizes for single-stepping; it prevents other threads
4995 from preempting the current thread while you are stepping, so that
4996 the focus of debugging does not change unexpectedly.
4997 Other threads only rarely (or never) get a chance to run
4998 when you step. They are more likely to run when you @samp{next} over a
4999 function call, and they are completely free to run when you use commands
5000 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
5001 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
5002 the current thread away from the thread that you are debugging.
5004 @item show scheduler-locking
5005 Display the current scheduler locking mode.
5008 @cindex resume threads of multiple processes simultaneously
5009 By default, when you issue one of the execution commands such as
5010 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5011 threads of the current inferior to run. For example, if @value{GDBN}
5012 is attached to two inferiors, each with two threads, the
5013 @code{continue} command resumes only the two threads of the current
5014 inferior. This is useful, for example, when you debug a program that
5015 forks and you want to hold the parent stopped (so that, for instance,
5016 it doesn't run to exit), while you debug the child. In other
5017 situations, you may not be interested in inspecting the current state
5018 of any of the processes @value{GDBN} is attached to, and you may want
5019 to resume them all until some breakpoint is hit. In the latter case,
5020 you can instruct @value{GDBN} to allow all threads of all the
5021 inferiors to run with the @w{@code{set schedule-multiple}} command.
5024 @kindex set schedule-multiple
5025 @item set schedule-multiple
5026 Set the mode for allowing threads of multiple processes to be resumed
5027 when an execution command is issued. When @code{on}, all threads of
5028 all processes are allowed to run. When @code{off}, only the threads
5029 of the current process are resumed. The default is @code{off}. The
5030 @code{scheduler-locking} mode takes precedence when set to @code{on},
5031 or while you are stepping and set to @code{step}.
5033 @item show schedule-multiple
5034 Display the current mode for resuming the execution of threads of
5039 @subsection Non-Stop Mode
5041 @cindex non-stop mode
5043 @c This section is really only a place-holder, and needs to be expanded
5044 @c with more details.
5046 For some multi-threaded targets, @value{GDBN} supports an optional
5047 mode of operation in which you can examine stopped program threads in
5048 the debugger while other threads continue to execute freely. This
5049 minimizes intrusion when debugging live systems, such as programs
5050 where some threads have real-time constraints or must continue to
5051 respond to external events. This is referred to as @dfn{non-stop} mode.
5053 In non-stop mode, when a thread stops to report a debugging event,
5054 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5055 threads as well, in contrast to the all-stop mode behavior. Additionally,
5056 execution commands such as @code{continue} and @code{step} apply by default
5057 only to the current thread in non-stop mode, rather than all threads as
5058 in all-stop mode. This allows you to control threads explicitly in
5059 ways that are not possible in all-stop mode --- for example, stepping
5060 one thread while allowing others to run freely, stepping
5061 one thread while holding all others stopped, or stepping several threads
5062 independently and simultaneously.
5064 To enter non-stop mode, use this sequence of commands before you run
5065 or attach to your program:
5068 # Enable the async interface.
5071 # If using the CLI, pagination breaks non-stop.
5074 # Finally, turn it on!
5078 You can use these commands to manipulate the non-stop mode setting:
5081 @kindex set non-stop
5082 @item set non-stop on
5083 Enable selection of non-stop mode.
5084 @item set non-stop off
5085 Disable selection of non-stop mode.
5086 @kindex show non-stop
5088 Show the current non-stop enablement setting.
5091 Note these commands only reflect whether non-stop mode is enabled,
5092 not whether the currently-executing program is being run in non-stop mode.
5093 In particular, the @code{set non-stop} preference is only consulted when
5094 @value{GDBN} starts or connects to the target program, and it is generally
5095 not possible to switch modes once debugging has started. Furthermore,
5096 since not all targets support non-stop mode, even when you have enabled
5097 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5100 In non-stop mode, all execution commands apply only to the current thread
5101 by default. That is, @code{continue} only continues one thread.
5102 To continue all threads, issue @code{continue -a} or @code{c -a}.
5104 You can use @value{GDBN}'s background execution commands
5105 (@pxref{Background Execution}) to run some threads in the background
5106 while you continue to examine or step others from @value{GDBN}.
5107 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5108 always executed asynchronously in non-stop mode.
5110 Suspending execution is done with the @code{interrupt} command when
5111 running in the background, or @kbd{Ctrl-c} during foreground execution.
5112 In all-stop mode, this stops the whole process;
5113 but in non-stop mode the interrupt applies only to the current thread.
5114 To stop the whole program, use @code{interrupt -a}.
5116 Other execution commands do not currently support the @code{-a} option.
5118 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5119 that thread current, as it does in all-stop mode. This is because the
5120 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5121 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5122 changed to a different thread just as you entered a command to operate on the
5123 previously current thread.
5125 @node Background Execution
5126 @subsection Background Execution
5128 @cindex foreground execution
5129 @cindex background execution
5130 @cindex asynchronous execution
5131 @cindex execution, foreground, background and asynchronous
5133 @value{GDBN}'s execution commands have two variants: the normal
5134 foreground (synchronous) behavior, and a background
5135 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5136 the program to report that some thread has stopped before prompting for
5137 another command. In background execution, @value{GDBN} immediately gives
5138 a command prompt so that you can issue other commands while your program runs.
5140 You need to explicitly enable asynchronous mode before you can use
5141 background execution commands. You can use these commands to
5142 manipulate the asynchronous mode setting:
5145 @kindex set target-async
5146 @item set target-async on
5147 Enable asynchronous mode.
5148 @item set target-async off
5149 Disable asynchronous mode.
5150 @kindex show target-async
5151 @item show target-async
5152 Show the current target-async setting.
5155 If the target doesn't support async mode, @value{GDBN} issues an error
5156 message if you attempt to use the background execution commands.
5158 To specify background execution, add a @code{&} to the command. For example,
5159 the background form of the @code{continue} command is @code{continue&}, or
5160 just @code{c&}. The execution commands that accept background execution
5166 @xref{Starting, , Starting your Program}.
5170 @xref{Attach, , Debugging an Already-running Process}.
5174 @xref{Continuing and Stepping, step}.
5178 @xref{Continuing and Stepping, stepi}.
5182 @xref{Continuing and Stepping, next}.
5186 @xref{Continuing and Stepping, nexti}.
5190 @xref{Continuing and Stepping, continue}.
5194 @xref{Continuing and Stepping, finish}.
5198 @xref{Continuing and Stepping, until}.
5202 Background execution is especially useful in conjunction with non-stop
5203 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5204 However, you can also use these commands in the normal all-stop mode with
5205 the restriction that you cannot issue another execution command until the
5206 previous one finishes. Examples of commands that are valid in all-stop
5207 mode while the program is running include @code{help} and @code{info break}.
5209 You can interrupt your program while it is running in the background by
5210 using the @code{interrupt} command.
5217 Suspend execution of the running program. In all-stop mode,
5218 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5219 only the current thread. To stop the whole program in non-stop mode,
5220 use @code{interrupt -a}.
5223 @node Thread-Specific Breakpoints
5224 @subsection Thread-Specific Breakpoints
5226 When your program has multiple threads (@pxref{Threads,, Debugging
5227 Programs with Multiple Threads}), you can choose whether to set
5228 breakpoints on all threads, or on a particular thread.
5231 @cindex breakpoints and threads
5232 @cindex thread breakpoints
5233 @kindex break @dots{} thread @var{threadno}
5234 @item break @var{linespec} thread @var{threadno}
5235 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5236 @var{linespec} specifies source lines; there are several ways of
5237 writing them (@pxref{Specify Location}), but the effect is always to
5238 specify some source line.
5240 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5241 to specify that you only want @value{GDBN} to stop the program when a
5242 particular thread reaches this breakpoint. @var{threadno} is one of the
5243 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5244 column of the @samp{info threads} display.
5246 If you do not specify @samp{thread @var{threadno}} when you set a
5247 breakpoint, the breakpoint applies to @emph{all} threads of your
5250 You can use the @code{thread} qualifier on conditional breakpoints as
5251 well; in this case, place @samp{thread @var{threadno}} before or
5252 after the breakpoint condition, like this:
5255 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5260 @node Interrupted System Calls
5261 @subsection Interrupted System Calls
5263 @cindex thread breakpoints and system calls
5264 @cindex system calls and thread breakpoints
5265 @cindex premature return from system calls
5266 There is an unfortunate side effect when using @value{GDBN} to debug
5267 multi-threaded programs. If one thread stops for a
5268 breakpoint, or for some other reason, and another thread is blocked in a
5269 system call, then the system call may return prematurely. This is a
5270 consequence of the interaction between multiple threads and the signals
5271 that @value{GDBN} uses to implement breakpoints and other events that
5274 To handle this problem, your program should check the return value of
5275 each system call and react appropriately. This is good programming
5278 For example, do not write code like this:
5284 The call to @code{sleep} will return early if a different thread stops
5285 at a breakpoint or for some other reason.
5287 Instead, write this:
5292 unslept = sleep (unslept);
5295 A system call is allowed to return early, so the system is still
5296 conforming to its specification. But @value{GDBN} does cause your
5297 multi-threaded program to behave differently than it would without
5300 Also, @value{GDBN} uses internal breakpoints in the thread library to
5301 monitor certain events such as thread creation and thread destruction.
5302 When such an event happens, a system call in another thread may return
5303 prematurely, even though your program does not appear to stop.
5306 @node Reverse Execution
5307 @chapter Running programs backward
5308 @cindex reverse execution
5309 @cindex running programs backward
5311 When you are debugging a program, it is not unusual to realize that
5312 you have gone too far, and some event of interest has already happened.
5313 If the target environment supports it, @value{GDBN} can allow you to
5314 ``rewind'' the program by running it backward.
5316 A target environment that supports reverse execution should be able
5317 to ``undo'' the changes in machine state that have taken place as the
5318 program was executing normally. Variables, registers etc.@: should
5319 revert to their previous values. Obviously this requires a great
5320 deal of sophistication on the part of the target environment; not
5321 all target environments can support reverse execution.
5323 When a program is executed in reverse, the instructions that
5324 have most recently been executed are ``un-executed'', in reverse
5325 order. The program counter runs backward, following the previous
5326 thread of execution in reverse. As each instruction is ``un-executed'',
5327 the values of memory and/or registers that were changed by that
5328 instruction are reverted to their previous states. After executing
5329 a piece of source code in reverse, all side effects of that code
5330 should be ``undone'', and all variables should be returned to their
5331 prior values@footnote{
5332 Note that some side effects are easier to undo than others. For instance,
5333 memory and registers are relatively easy, but device I/O is hard. Some
5334 targets may be able undo things like device I/O, and some may not.
5336 The contract between @value{GDBN} and the reverse executing target
5337 requires only that the target do something reasonable when
5338 @value{GDBN} tells it to execute backwards, and then report the
5339 results back to @value{GDBN}. Whatever the target reports back to
5340 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
5341 assumes that the memory and registers that the target reports are in a
5342 consistant state, but @value{GDBN} accepts whatever it is given.
5345 If you are debugging in a target environment that supports
5346 reverse execution, @value{GDBN} provides the following commands.
5349 @kindex reverse-continue
5350 @kindex rc @r{(@code{reverse-continue})}
5351 @item reverse-continue @r{[}@var{ignore-count}@r{]}
5352 @itemx rc @r{[}@var{ignore-count}@r{]}
5353 Beginning at the point where your program last stopped, start executing
5354 in reverse. Reverse execution will stop for breakpoints and synchronous
5355 exceptions (signals), just like normal execution. Behavior of
5356 asynchronous signals depends on the target environment.
5358 @kindex reverse-step
5359 @kindex rs @r{(@code{step})}
5360 @item reverse-step @r{[}@var{count}@r{]}
5361 Run the program backward until control reaches the start of a
5362 different source line; then stop it, and return control to @value{GDBN}.
5364 Like the @code{step} command, @code{reverse-step} will only stop
5365 at the beginning of a source line. It ``un-executes'' the previously
5366 executed source line. If the previous source line included calls to
5367 debuggable functions, @code{reverse-step} will step (backward) into
5368 the called function, stopping at the beginning of the @emph{last}
5369 statement in the called function (typically a return statement).
5371 Also, as with the @code{step} command, if non-debuggable functions are
5372 called, @code{reverse-step} will run thru them backward without stopping.
5374 @kindex reverse-stepi
5375 @kindex rsi @r{(@code{reverse-stepi})}
5376 @item reverse-stepi @r{[}@var{count}@r{]}
5377 Reverse-execute one machine instruction. Note that the instruction
5378 to be reverse-executed is @emph{not} the one pointed to by the program
5379 counter, but the instruction executed prior to that one. For instance,
5380 if the last instruction was a jump, @code{reverse-stepi} will take you
5381 back from the destination of the jump to the jump instruction itself.
5383 @kindex reverse-next
5384 @kindex rn @r{(@code{reverse-next})}
5385 @item reverse-next @r{[}@var{count}@r{]}
5386 Run backward to the beginning of the previous line executed in
5387 the current (innermost) stack frame. If the line contains function
5388 calls, they will be ``un-executed'' without stopping. Starting from
5389 the first line of a function, @code{reverse-next} will take you back
5390 to the caller of that function, @emph{before} the function was called,
5391 just as the normal @code{next} command would take you from the last
5392 line of a function back to its return to its caller
5393 @footnote{Unless the code is too heavily optimized.}.
5395 @kindex reverse-nexti
5396 @kindex rni @r{(@code{reverse-nexti})}
5397 @item reverse-nexti @r{[}@var{count}@r{]}
5398 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
5399 in reverse, except that called functions are ``un-executed'' atomically.
5400 That is, if the previously executed instruction was a return from
5401 another function, @code{reverse-nexti} will continue to execute
5402 in reverse until the call to that function (from the current stack
5405 @kindex reverse-finish
5406 @item reverse-finish
5407 Just as the @code{finish} command takes you to the point where the
5408 current function returns, @code{reverse-finish} takes you to the point
5409 where it was called. Instead of ending up at the end of the current
5410 function invocation, you end up at the beginning.
5412 @kindex set exec-direction
5413 @item set exec-direction
5414 Set the direction of target execution.
5415 @itemx set exec-direction reverse
5416 @cindex execute forward or backward in time
5417 @value{GDBN} will perform all execution commands in reverse, until the
5418 exec-direction mode is changed to ``forward''. Affected commands include
5419 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
5420 command cannot be used in reverse mode.
5421 @item set exec-direction forward
5422 @value{GDBN} will perform all execution commands in the normal fashion.
5423 This is the default.
5427 @node Process Record and Replay
5428 @chapter Recording Inferior's Execution and Replaying It
5429 @cindex process record and replay
5430 @cindex recording inferior's execution and replaying it
5432 On some platforms, @value{GDBN} provides a special @dfn{process record
5433 and replay} target that can record a log of the process execution, and
5434 replay it later with both forward and reverse execution commands.
5437 When this target is in use, if the execution log includes the record
5438 for the next instruction, @value{GDBN} will debug in @dfn{replay
5439 mode}. In the replay mode, the inferior does not really execute code
5440 instructions. Instead, all the events that normally happen during
5441 code execution are taken from the execution log. While code is not
5442 really executed in replay mode, the values of registers (including the
5443 program counter register) and the memory of the inferior are still
5444 changed as they normally would. Their contents are taken from the
5448 If the record for the next instruction is not in the execution log,
5449 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
5450 inferior executes normally, and @value{GDBN} records the execution log
5453 The process record and replay target supports reverse execution
5454 (@pxref{Reverse Execution}), even if the platform on which the
5455 inferior runs does not. However, the reverse execution is limited in
5456 this case by the range of the instructions recorded in the execution
5457 log. In other words, reverse execution on platforms that don't
5458 support it directly can only be done in the replay mode.
5460 When debugging in the reverse direction, @value{GDBN} will work in
5461 replay mode as long as the execution log includes the record for the
5462 previous instruction; otherwise, it will work in record mode, if the
5463 platform supports reverse execution, or stop if not.
5465 For architecture environments that support process record and replay,
5466 @value{GDBN} provides the following commands:
5469 @kindex target record
5473 This command starts the process record and replay target. The process
5474 record and replay target can only debug a process that is already
5475 running. Therefore, you need first to start the process with the
5476 @kbd{run} or @kbd{start} commands, and then start the recording with
5477 the @kbd{target record} command.
5479 Both @code{record} and @code{rec} are aliases of @code{target record}.
5481 @cindex displaced stepping, and process record and replay
5482 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
5483 will be automatically disabled when process record and replay target
5484 is started. That's because the process record and replay target
5485 doesn't support displaced stepping.
5487 @cindex non-stop mode, and process record and replay
5488 @cindex asynchronous execution, and process record and replay
5489 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
5490 the asynchronous execution mode (@pxref{Background Execution}), the
5491 process record and replay target cannot be started because it doesn't
5492 support these two modes.
5497 Stop the process record and replay target. When process record and
5498 replay target stops, the entire execution log will be deleted and the
5499 inferior will either be terminated, or will remain in its final state.
5501 When you stop the process record and replay target in record mode (at
5502 the end of the execution log), the inferior will be stopped at the
5503 next instruction that would have been recorded. In other words, if
5504 you record for a while and then stop recording, the inferior process
5505 will be left in the same state as if the recording never happened.
5507 On the other hand, if the process record and replay target is stopped
5508 while in replay mode (that is, not at the end of the execution log,
5509 but at some earlier point), the inferior process will become ``live''
5510 at that earlier state, and it will then be possible to continue the
5511 usual ``live'' debugging of the process from that state.
5513 When the inferior process exits, or @value{GDBN} detaches from it,
5514 process record and replay target will automatically stop itself.
5516 @kindex set record insn-number-max
5517 @item set record insn-number-max @var{limit}
5518 Set the limit of instructions to be recorded. Default value is 200000.
5520 If @var{limit} is a positive number, then @value{GDBN} will start
5521 deleting instructions from the log once the number of the record
5522 instructions becomes greater than @var{limit}. For every new recorded
5523 instruction, @value{GDBN} will delete the earliest recorded
5524 instruction to keep the number of recorded instructions at the limit.
5525 (Since deleting recorded instructions loses information, @value{GDBN}
5526 lets you control what happens when the limit is reached, by means of
5527 the @code{stop-at-limit} option, described below.)
5529 If @var{limit} is zero, @value{GDBN} will never delete recorded
5530 instructions from the execution log. The number of recorded
5531 instructions is unlimited in this case.
5533 @kindex show record insn-number-max
5534 @item show record insn-number-max
5535 Show the limit of instructions to be recorded.
5537 @kindex set record stop-at-limit
5538 @item set record stop-at-limit
5539 Control the behavior when the number of recorded instructions reaches
5540 the limit. If ON (the default), @value{GDBN} will stop when the limit
5541 is reached for the first time and ask you whether you want to stop the
5542 inferior or continue running it and recording the execution log. If
5543 you decide to continue recording, each new recorded instruction will
5544 cause the oldest one to be deleted.
5546 If this option is OFF, @value{GDBN} will automatically delete the
5547 oldest record to make room for each new one, without asking.
5549 @kindex show record stop-at-limit
5550 @item show record stop-at-limit
5551 Show the current setting of @code{stop-at-limit}.
5555 Show various statistics about the state of process record and its
5556 in-memory execution log buffer, including:
5560 Whether in record mode or replay mode.
5562 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
5564 Highest recorded instruction number.
5566 Current instruction about to be replayed (if in replay mode).
5568 Number of instructions contained in the execution log.
5570 Maximum number of instructions that may be contained in the execution log.
5573 @kindex record delete
5576 When record target runs in replay mode (``in the past''), delete the
5577 subsequent execution log and begin to record a new execution log starting
5578 from the current address. This means you will abandon the previously
5579 recorded ``future'' and begin recording a new ``future''.
5584 @chapter Examining the Stack
5586 When your program has stopped, the first thing you need to know is where it
5587 stopped and how it got there.
5590 Each time your program performs a function call, information about the call
5592 That information includes the location of the call in your program,
5593 the arguments of the call,
5594 and the local variables of the function being called.
5595 The information is saved in a block of data called a @dfn{stack frame}.
5596 The stack frames are allocated in a region of memory called the @dfn{call
5599 When your program stops, the @value{GDBN} commands for examining the
5600 stack allow you to see all of this information.
5602 @cindex selected frame
5603 One of the stack frames is @dfn{selected} by @value{GDBN} and many
5604 @value{GDBN} commands refer implicitly to the selected frame. In
5605 particular, whenever you ask @value{GDBN} for the value of a variable in
5606 your program, the value is found in the selected frame. There are
5607 special @value{GDBN} commands to select whichever frame you are
5608 interested in. @xref{Selection, ,Selecting a Frame}.
5610 When your program stops, @value{GDBN} automatically selects the
5611 currently executing frame and describes it briefly, similar to the
5612 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
5615 * Frames:: Stack frames
5616 * Backtrace:: Backtraces
5617 * Selection:: Selecting a frame
5618 * Frame Info:: Information on a frame
5623 @section Stack Frames
5625 @cindex frame, definition
5627 The call stack is divided up into contiguous pieces called @dfn{stack
5628 frames}, or @dfn{frames} for short; each frame is the data associated
5629 with one call to one function. The frame contains the arguments given
5630 to the function, the function's local variables, and the address at
5631 which the function is executing.
5633 @cindex initial frame
5634 @cindex outermost frame
5635 @cindex innermost frame
5636 When your program is started, the stack has only one frame, that of the
5637 function @code{main}. This is called the @dfn{initial} frame or the
5638 @dfn{outermost} frame. Each time a function is called, a new frame is
5639 made. Each time a function returns, the frame for that function invocation
5640 is eliminated. If a function is recursive, there can be many frames for
5641 the same function. The frame for the function in which execution is
5642 actually occurring is called the @dfn{innermost} frame. This is the most
5643 recently created of all the stack frames that still exist.
5645 @cindex frame pointer
5646 Inside your program, stack frames are identified by their addresses. A
5647 stack frame consists of many bytes, each of which has its own address; each
5648 kind of computer has a convention for choosing one byte whose
5649 address serves as the address of the frame. Usually this address is kept
5650 in a register called the @dfn{frame pointer register}
5651 (@pxref{Registers, $fp}) while execution is going on in that frame.
5653 @cindex frame number
5654 @value{GDBN} assigns numbers to all existing stack frames, starting with
5655 zero for the innermost frame, one for the frame that called it,
5656 and so on upward. These numbers do not really exist in your program;
5657 they are assigned by @value{GDBN} to give you a way of designating stack
5658 frames in @value{GDBN} commands.
5660 @c The -fomit-frame-pointer below perennially causes hbox overflow
5661 @c underflow problems.
5662 @cindex frameless execution
5663 Some compilers provide a way to compile functions so that they operate
5664 without stack frames. (For example, the @value{NGCC} option
5666 @samp{-fomit-frame-pointer}
5668 generates functions without a frame.)
5669 This is occasionally done with heavily used library functions to save
5670 the frame setup time. @value{GDBN} has limited facilities for dealing
5671 with these function invocations. If the innermost function invocation
5672 has no stack frame, @value{GDBN} nevertheless regards it as though
5673 it had a separate frame, which is numbered zero as usual, allowing
5674 correct tracing of the function call chain. However, @value{GDBN} has
5675 no provision for frameless functions elsewhere in the stack.
5678 @kindex frame@r{, command}
5679 @cindex current stack frame
5680 @item frame @var{args}
5681 The @code{frame} command allows you to move from one stack frame to another,
5682 and to print the stack frame you select. @var{args} may be either the
5683 address of the frame or the stack frame number. Without an argument,
5684 @code{frame} prints the current stack frame.
5686 @kindex select-frame
5687 @cindex selecting frame silently
5689 The @code{select-frame} command allows you to move from one stack frame
5690 to another without printing the frame. This is the silent version of
5698 @cindex call stack traces
5699 A backtrace is a summary of how your program got where it is. It shows one
5700 line per frame, for many frames, starting with the currently executing
5701 frame (frame zero), followed by its caller (frame one), and on up the
5706 @kindex bt @r{(@code{backtrace})}
5709 Print a backtrace of the entire stack: one line per frame for all
5710 frames in the stack.
5712 You can stop the backtrace at any time by typing the system interrupt
5713 character, normally @kbd{Ctrl-c}.
5715 @item backtrace @var{n}
5717 Similar, but print only the innermost @var{n} frames.
5719 @item backtrace -@var{n}
5721 Similar, but print only the outermost @var{n} frames.
5723 @item backtrace full
5725 @itemx bt full @var{n}
5726 @itemx bt full -@var{n}
5727 Print the values of the local variables also. @var{n} specifies the
5728 number of frames to print, as described above.
5733 The names @code{where} and @code{info stack} (abbreviated @code{info s})
5734 are additional aliases for @code{backtrace}.
5736 @cindex multiple threads, backtrace
5737 In a multi-threaded program, @value{GDBN} by default shows the
5738 backtrace only for the current thread. To display the backtrace for
5739 several or all of the threads, use the command @code{thread apply}
5740 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
5741 apply all backtrace}, @value{GDBN} will display the backtrace for all
5742 the threads; this is handy when you debug a core dump of a
5743 multi-threaded program.
5745 Each line in the backtrace shows the frame number and the function name.
5746 The program counter value is also shown---unless you use @code{set
5747 print address off}. The backtrace also shows the source file name and
5748 line number, as well as the arguments to the function. The program
5749 counter value is omitted if it is at the beginning of the code for that
5752 Here is an example of a backtrace. It was made with the command
5753 @samp{bt 3}, so it shows the innermost three frames.
5757 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5759 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
5760 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
5762 (More stack frames follow...)
5767 The display for frame zero does not begin with a program counter
5768 value, indicating that your program has stopped at the beginning of the
5769 code for line @code{993} of @code{builtin.c}.
5772 The value of parameter @code{data} in frame 1 has been replaced by
5773 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
5774 only if it is a scalar (integer, pointer, enumeration, etc). See command
5775 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
5776 on how to configure the way function parameter values are printed.
5778 @cindex value optimized out, in backtrace
5779 @cindex function call arguments, optimized out
5780 If your program was compiled with optimizations, some compilers will
5781 optimize away arguments passed to functions if those arguments are
5782 never used after the call. Such optimizations generate code that
5783 passes arguments through registers, but doesn't store those arguments
5784 in the stack frame. @value{GDBN} has no way of displaying such
5785 arguments in stack frames other than the innermost one. Here's what
5786 such a backtrace might look like:
5790 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5792 #1 0x6e38 in expand_macro (sym=<value optimized out>) at macro.c:242
5793 #2 0x6840 in expand_token (obs=0x0, t=<value optimized out>, td=0xf7fffb08)
5795 (More stack frames follow...)
5800 The values of arguments that were not saved in their stack frames are
5801 shown as @samp{<value optimized out>}.
5803 If you need to display the values of such optimized-out arguments,
5804 either deduce that from other variables whose values depend on the one
5805 you are interested in, or recompile without optimizations.
5807 @cindex backtrace beyond @code{main} function
5808 @cindex program entry point
5809 @cindex startup code, and backtrace
5810 Most programs have a standard user entry point---a place where system
5811 libraries and startup code transition into user code. For C this is
5812 @code{main}@footnote{
5813 Note that embedded programs (the so-called ``free-standing''
5814 environment) are not required to have a @code{main} function as the
5815 entry point. They could even have multiple entry points.}.
5816 When @value{GDBN} finds the entry function in a backtrace
5817 it will terminate the backtrace, to avoid tracing into highly
5818 system-specific (and generally uninteresting) code.
5820 If you need to examine the startup code, or limit the number of levels
5821 in a backtrace, you can change this behavior:
5824 @item set backtrace past-main
5825 @itemx set backtrace past-main on
5826 @kindex set backtrace
5827 Backtraces will continue past the user entry point.
5829 @item set backtrace past-main off
5830 Backtraces will stop when they encounter the user entry point. This is the
5833 @item show backtrace past-main
5834 @kindex show backtrace
5835 Display the current user entry point backtrace policy.
5837 @item set backtrace past-entry
5838 @itemx set backtrace past-entry on
5839 Backtraces will continue past the internal entry point of an application.
5840 This entry point is encoded by the linker when the application is built,
5841 and is likely before the user entry point @code{main} (or equivalent) is called.
5843 @item set backtrace past-entry off
5844 Backtraces will stop when they encounter the internal entry point of an
5845 application. This is the default.
5847 @item show backtrace past-entry
5848 Display the current internal entry point backtrace policy.
5850 @item set backtrace limit @var{n}
5851 @itemx set backtrace limit 0
5852 @cindex backtrace limit
5853 Limit the backtrace to @var{n} levels. A value of zero means
5856 @item show backtrace limit
5857 Display the current limit on backtrace levels.
5861 @section Selecting a Frame
5863 Most commands for examining the stack and other data in your program work on
5864 whichever stack frame is selected at the moment. Here are the commands for
5865 selecting a stack frame; all of them finish by printing a brief description
5866 of the stack frame just selected.
5869 @kindex frame@r{, selecting}
5870 @kindex f @r{(@code{frame})}
5873 Select frame number @var{n}. Recall that frame zero is the innermost
5874 (currently executing) frame, frame one is the frame that called the
5875 innermost one, and so on. The highest-numbered frame is the one for
5878 @item frame @var{addr}
5880 Select the frame at address @var{addr}. This is useful mainly if the
5881 chaining of stack frames has been damaged by a bug, making it
5882 impossible for @value{GDBN} to assign numbers properly to all frames. In
5883 addition, this can be useful when your program has multiple stacks and
5884 switches between them.
5886 On the SPARC architecture, @code{frame} needs two addresses to
5887 select an arbitrary frame: a frame pointer and a stack pointer.
5889 On the MIPS and Alpha architecture, it needs two addresses: a stack
5890 pointer and a program counter.
5892 On the 29k architecture, it needs three addresses: a register stack
5893 pointer, a program counter, and a memory stack pointer.
5897 Move @var{n} frames up the stack. For positive numbers @var{n}, this
5898 advances toward the outermost frame, to higher frame numbers, to frames
5899 that have existed longer. @var{n} defaults to one.
5902 @kindex do @r{(@code{down})}
5904 Move @var{n} frames down the stack. For positive numbers @var{n}, this
5905 advances toward the innermost frame, to lower frame numbers, to frames
5906 that were created more recently. @var{n} defaults to one. You may
5907 abbreviate @code{down} as @code{do}.
5910 All of these commands end by printing two lines of output describing the
5911 frame. The first line shows the frame number, the function name, the
5912 arguments, and the source file and line number of execution in that
5913 frame. The second line shows the text of that source line.
5921 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
5923 10 read_input_file (argv[i]);
5927 After such a printout, the @code{list} command with no arguments
5928 prints ten lines centered on the point of execution in the frame.
5929 You can also edit the program at the point of execution with your favorite
5930 editing program by typing @code{edit}.
5931 @xref{List, ,Printing Source Lines},
5935 @kindex down-silently
5937 @item up-silently @var{n}
5938 @itemx down-silently @var{n}
5939 These two commands are variants of @code{up} and @code{down},
5940 respectively; they differ in that they do their work silently, without
5941 causing display of the new frame. They are intended primarily for use
5942 in @value{GDBN} command scripts, where the output might be unnecessary and
5947 @section Information About a Frame
5949 There are several other commands to print information about the selected
5955 When used without any argument, this command does not change which
5956 frame is selected, but prints a brief description of the currently
5957 selected stack frame. It can be abbreviated @code{f}. With an
5958 argument, this command is used to select a stack frame.
5959 @xref{Selection, ,Selecting a Frame}.
5962 @kindex info f @r{(@code{info frame})}
5965 This command prints a verbose description of the selected stack frame,
5970 the address of the frame
5972 the address of the next frame down (called by this frame)
5974 the address of the next frame up (caller of this frame)
5976 the language in which the source code corresponding to this frame is written
5978 the address of the frame's arguments
5980 the address of the frame's local variables
5982 the program counter saved in it (the address of execution in the caller frame)
5984 which registers were saved in the frame
5987 @noindent The verbose description is useful when
5988 something has gone wrong that has made the stack format fail to fit
5989 the usual conventions.
5991 @item info frame @var{addr}
5992 @itemx info f @var{addr}
5993 Print a verbose description of the frame at address @var{addr}, without
5994 selecting that frame. The selected frame remains unchanged by this
5995 command. This requires the same kind of address (more than one for some
5996 architectures) that you specify in the @code{frame} command.
5997 @xref{Selection, ,Selecting a Frame}.
6001 Print the arguments of the selected frame, each on a separate line.
6005 Print the local variables of the selected frame, each on a separate
6006 line. These are all variables (declared either static or automatic)
6007 accessible at the point of execution of the selected frame.
6010 @cindex catch exceptions, list active handlers
6011 @cindex exception handlers, how to list
6013 Print a list of all the exception handlers that are active in the
6014 current stack frame at the current point of execution. To see other
6015 exception handlers, visit the associated frame (using the @code{up},
6016 @code{down}, or @code{frame} commands); then type @code{info catch}.
6017 @xref{Set Catchpoints, , Setting Catchpoints}.
6023 @chapter Examining Source Files
6025 @value{GDBN} can print parts of your program's source, since the debugging
6026 information recorded in the program tells @value{GDBN} what source files were
6027 used to build it. When your program stops, @value{GDBN} spontaneously prints
6028 the line where it stopped. Likewise, when you select a stack frame
6029 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
6030 execution in that frame has stopped. You can print other portions of
6031 source files by explicit command.
6033 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
6034 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
6035 @value{GDBN} under @sc{gnu} Emacs}.
6038 * List:: Printing source lines
6039 * Specify Location:: How to specify code locations
6040 * Edit:: Editing source files
6041 * Search:: Searching source files
6042 * Source Path:: Specifying source directories
6043 * Machine Code:: Source and machine code
6047 @section Printing Source Lines
6050 @kindex l @r{(@code{list})}
6051 To print lines from a source file, use the @code{list} command
6052 (abbreviated @code{l}). By default, ten lines are printed.
6053 There are several ways to specify what part of the file you want to
6054 print; see @ref{Specify Location}, for the full list.
6056 Here are the forms of the @code{list} command most commonly used:
6059 @item list @var{linenum}
6060 Print lines centered around line number @var{linenum} in the
6061 current source file.
6063 @item list @var{function}
6064 Print lines centered around the beginning of function
6068 Print more lines. If the last lines printed were printed with a
6069 @code{list} command, this prints lines following the last lines
6070 printed; however, if the last line printed was a solitary line printed
6071 as part of displaying a stack frame (@pxref{Stack, ,Examining the
6072 Stack}), this prints lines centered around that line.
6075 Print lines just before the lines last printed.
6078 @cindex @code{list}, how many lines to display
6079 By default, @value{GDBN} prints ten source lines with any of these forms of
6080 the @code{list} command. You can change this using @code{set listsize}:
6083 @kindex set listsize
6084 @item set listsize @var{count}
6085 Make the @code{list} command display @var{count} source lines (unless
6086 the @code{list} argument explicitly specifies some other number).
6088 @kindex show listsize
6090 Display the number of lines that @code{list} prints.
6093 Repeating a @code{list} command with @key{RET} discards the argument,
6094 so it is equivalent to typing just @code{list}. This is more useful
6095 than listing the same lines again. An exception is made for an
6096 argument of @samp{-}; that argument is preserved in repetition so that
6097 each repetition moves up in the source file.
6099 In general, the @code{list} command expects you to supply zero, one or two
6100 @dfn{linespecs}. Linespecs specify source lines; there are several ways
6101 of writing them (@pxref{Specify Location}), but the effect is always
6102 to specify some source line.
6104 Here is a complete description of the possible arguments for @code{list}:
6107 @item list @var{linespec}
6108 Print lines centered around the line specified by @var{linespec}.
6110 @item list @var{first},@var{last}
6111 Print lines from @var{first} to @var{last}. Both arguments are
6112 linespecs. When a @code{list} command has two linespecs, and the
6113 source file of the second linespec is omitted, this refers to
6114 the same source file as the first linespec.
6116 @item list ,@var{last}
6117 Print lines ending with @var{last}.
6119 @item list @var{first},
6120 Print lines starting with @var{first}.
6123 Print lines just after the lines last printed.
6126 Print lines just before the lines last printed.
6129 As described in the preceding table.
6132 @node Specify Location
6133 @section Specifying a Location
6134 @cindex specifying location
6137 Several @value{GDBN} commands accept arguments that specify a location
6138 of your program's code. Since @value{GDBN} is a source-level
6139 debugger, a location usually specifies some line in the source code;
6140 for that reason, locations are also known as @dfn{linespecs}.
6142 Here are all the different ways of specifying a code location that
6143 @value{GDBN} understands:
6147 Specifies the line number @var{linenum} of the current source file.
6150 @itemx +@var{offset}
6151 Specifies the line @var{offset} lines before or after the @dfn{current
6152 line}. For the @code{list} command, the current line is the last one
6153 printed; for the breakpoint commands, this is the line at which
6154 execution stopped in the currently selected @dfn{stack frame}
6155 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
6156 used as the second of the two linespecs in a @code{list} command,
6157 this specifies the line @var{offset} lines up or down from the first
6160 @item @var{filename}:@var{linenum}
6161 Specifies the line @var{linenum} in the source file @var{filename}.
6163 @item @var{function}
6164 Specifies the line that begins the body of the function @var{function}.
6165 For example, in C, this is the line with the open brace.
6167 @item @var{filename}:@var{function}
6168 Specifies the line that begins the body of the function @var{function}
6169 in the file @var{filename}. You only need the file name with a
6170 function name to avoid ambiguity when there are identically named
6171 functions in different source files.
6173 @item *@var{address}
6174 Specifies the program address @var{address}. For line-oriented
6175 commands, such as @code{list} and @code{edit}, this specifies a source
6176 line that contains @var{address}. For @code{break} and other
6177 breakpoint oriented commands, this can be used to set breakpoints in
6178 parts of your program which do not have debugging information or
6181 Here @var{address} may be any expression valid in the current working
6182 language (@pxref{Languages, working language}) that specifies a code
6183 address. In addition, as a convenience, @value{GDBN} extends the
6184 semantics of expressions used in locations to cover the situations
6185 that frequently happen during debugging. Here are the various forms
6189 @item @var{expression}
6190 Any expression valid in the current working language.
6192 @item @var{funcaddr}
6193 An address of a function or procedure derived from its name. In C,
6194 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
6195 simply the function's name @var{function} (and actually a special case
6196 of a valid expression). In Pascal and Modula-2, this is
6197 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
6198 (although the Pascal form also works).
6200 This form specifies the address of the function's first instruction,
6201 before the stack frame and arguments have been set up.
6203 @item '@var{filename}'::@var{funcaddr}
6204 Like @var{funcaddr} above, but also specifies the name of the source
6205 file explicitly. This is useful if the name of the function does not
6206 specify the function unambiguously, e.g., if there are several
6207 functions with identical names in different source files.
6214 @section Editing Source Files
6215 @cindex editing source files
6218 @kindex e @r{(@code{edit})}
6219 To edit the lines in a source file, use the @code{edit} command.
6220 The editing program of your choice
6221 is invoked with the current line set to
6222 the active line in the program.
6223 Alternatively, there are several ways to specify what part of the file you
6224 want to print if you want to see other parts of the program:
6227 @item edit @var{location}
6228 Edit the source file specified by @code{location}. Editing starts at
6229 that @var{location}, e.g., at the specified source line of the
6230 specified file. @xref{Specify Location}, for all the possible forms
6231 of the @var{location} argument; here are the forms of the @code{edit}
6232 command most commonly used:
6235 @item edit @var{number}
6236 Edit the current source file with @var{number} as the active line number.
6238 @item edit @var{function}
6239 Edit the file containing @var{function} at the beginning of its definition.
6244 @subsection Choosing your Editor
6245 You can customize @value{GDBN} to use any editor you want
6247 The only restriction is that your editor (say @code{ex}), recognizes the
6248 following command-line syntax:
6250 ex +@var{number} file
6252 The optional numeric value +@var{number} specifies the number of the line in
6253 the file where to start editing.}.
6254 By default, it is @file{@value{EDITOR}}, but you can change this
6255 by setting the environment variable @code{EDITOR} before using
6256 @value{GDBN}. For example, to configure @value{GDBN} to use the
6257 @code{vi} editor, you could use these commands with the @code{sh} shell:
6263 or in the @code{csh} shell,
6265 setenv EDITOR /usr/bin/vi
6270 @section Searching Source Files
6271 @cindex searching source files
6273 There are two commands for searching through the current source file for a
6278 @kindex forward-search
6279 @item forward-search @var{regexp}
6280 @itemx search @var{regexp}
6281 The command @samp{forward-search @var{regexp}} checks each line,
6282 starting with the one following the last line listed, for a match for
6283 @var{regexp}. It lists the line that is found. You can use the
6284 synonym @samp{search @var{regexp}} or abbreviate the command name as
6287 @kindex reverse-search
6288 @item reverse-search @var{regexp}
6289 The command @samp{reverse-search @var{regexp}} checks each line, starting
6290 with the one before the last line listed and going backward, for a match
6291 for @var{regexp}. It lists the line that is found. You can abbreviate
6292 this command as @code{rev}.
6296 @section Specifying Source Directories
6299 @cindex directories for source files
6300 Executable programs sometimes do not record the directories of the source
6301 files from which they were compiled, just the names. Even when they do,
6302 the directories could be moved between the compilation and your debugging
6303 session. @value{GDBN} has a list of directories to search for source files;
6304 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
6305 it tries all the directories in the list, in the order they are present
6306 in the list, until it finds a file with the desired name.
6308 For example, suppose an executable references the file
6309 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
6310 @file{/mnt/cross}. The file is first looked up literally; if this
6311 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
6312 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
6313 message is printed. @value{GDBN} does not look up the parts of the
6314 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
6315 Likewise, the subdirectories of the source path are not searched: if
6316 the source path is @file{/mnt/cross}, and the binary refers to
6317 @file{foo.c}, @value{GDBN} would not find it under
6318 @file{/mnt/cross/usr/src/foo-1.0/lib}.
6320 Plain file names, relative file names with leading directories, file
6321 names containing dots, etc.@: are all treated as described above; for
6322 instance, if the source path is @file{/mnt/cross}, and the source file
6323 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
6324 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
6325 that---@file{/mnt/cross/foo.c}.
6327 Note that the executable search path is @emph{not} used to locate the
6330 Whenever you reset or rearrange the source path, @value{GDBN} clears out
6331 any information it has cached about where source files are found and where
6332 each line is in the file.
6336 When you start @value{GDBN}, its source path includes only @samp{cdir}
6337 and @samp{cwd}, in that order.
6338 To add other directories, use the @code{directory} command.
6340 The search path is used to find both program source files and @value{GDBN}
6341 script files (read using the @samp{-command} option and @samp{source} command).
6343 In addition to the source path, @value{GDBN} provides a set of commands
6344 that manage a list of source path substitution rules. A @dfn{substitution
6345 rule} specifies how to rewrite source directories stored in the program's
6346 debug information in case the sources were moved to a different
6347 directory between compilation and debugging. A rule is made of
6348 two strings, the first specifying what needs to be rewritten in
6349 the path, and the second specifying how it should be rewritten.
6350 In @ref{set substitute-path}, we name these two parts @var{from} and
6351 @var{to} respectively. @value{GDBN} does a simple string replacement
6352 of @var{from} with @var{to} at the start of the directory part of the
6353 source file name, and uses that result instead of the original file
6354 name to look up the sources.
6356 Using the previous example, suppose the @file{foo-1.0} tree has been
6357 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
6358 @value{GDBN} to replace @file{/usr/src} in all source path names with
6359 @file{/mnt/cross}. The first lookup will then be
6360 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
6361 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
6362 substitution rule, use the @code{set substitute-path} command
6363 (@pxref{set substitute-path}).
6365 To avoid unexpected substitution results, a rule is applied only if the
6366 @var{from} part of the directory name ends at a directory separator.
6367 For instance, a rule substituting @file{/usr/source} into
6368 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
6369 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
6370 is applied only at the beginning of the directory name, this rule will
6371 not be applied to @file{/root/usr/source/baz.c} either.
6373 In many cases, you can achieve the same result using the @code{directory}
6374 command. However, @code{set substitute-path} can be more efficient in
6375 the case where the sources are organized in a complex tree with multiple
6376 subdirectories. With the @code{directory} command, you need to add each
6377 subdirectory of your project. If you moved the entire tree while
6378 preserving its internal organization, then @code{set substitute-path}
6379 allows you to direct the debugger to all the sources with one single
6382 @code{set substitute-path} is also more than just a shortcut command.
6383 The source path is only used if the file at the original location no
6384 longer exists. On the other hand, @code{set substitute-path} modifies
6385 the debugger behavior to look at the rewritten location instead. So, if
6386 for any reason a source file that is not relevant to your executable is
6387 located at the original location, a substitution rule is the only
6388 method available to point @value{GDBN} at the new location.
6390 @cindex @samp{--with-relocated-sources}
6391 @cindex default source path substitution
6392 You can configure a default source path substitution rule by
6393 configuring @value{GDBN} with the
6394 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
6395 should be the name of a directory under @value{GDBN}'s configured
6396 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
6397 directory names in debug information under @var{dir} will be adjusted
6398 automatically if the installed @value{GDBN} is moved to a new
6399 location. This is useful if @value{GDBN}, libraries or executables
6400 with debug information and corresponding source code are being moved
6404 @item directory @var{dirname} @dots{}
6405 @item dir @var{dirname} @dots{}
6406 Add directory @var{dirname} to the front of the source path. Several
6407 directory names may be given to this command, separated by @samp{:}
6408 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
6409 part of absolute file names) or
6410 whitespace. You may specify a directory that is already in the source
6411 path; this moves it forward, so @value{GDBN} searches it sooner.
6415 @vindex $cdir@r{, convenience variable}
6416 @vindex $cwd@r{, convenience variable}
6417 @cindex compilation directory
6418 @cindex current directory
6419 @cindex working directory
6420 @cindex directory, current
6421 @cindex directory, compilation
6422 You can use the string @samp{$cdir} to refer to the compilation
6423 directory (if one is recorded), and @samp{$cwd} to refer to the current
6424 working directory. @samp{$cwd} is not the same as @samp{.}---the former
6425 tracks the current working directory as it changes during your @value{GDBN}
6426 session, while the latter is immediately expanded to the current
6427 directory at the time you add an entry to the source path.
6430 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
6432 @c RET-repeat for @code{directory} is explicitly disabled, but since
6433 @c repeating it would be a no-op we do not say that. (thanks to RMS)
6435 @item show directories
6436 @kindex show directories
6437 Print the source path: show which directories it contains.
6439 @anchor{set substitute-path}
6440 @item set substitute-path @var{from} @var{to}
6441 @kindex set substitute-path
6442 Define a source path substitution rule, and add it at the end of the
6443 current list of existing substitution rules. If a rule with the same
6444 @var{from} was already defined, then the old rule is also deleted.
6446 For example, if the file @file{/foo/bar/baz.c} was moved to
6447 @file{/mnt/cross/baz.c}, then the command
6450 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
6454 will tell @value{GDBN} to replace @samp{/usr/src} with
6455 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
6456 @file{baz.c} even though it was moved.
6458 In the case when more than one substitution rule have been defined,
6459 the rules are evaluated one by one in the order where they have been
6460 defined. The first one matching, if any, is selected to perform
6463 For instance, if we had entered the following commands:
6466 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
6467 (@value{GDBP}) set substitute-path /usr/src /mnt/src
6471 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
6472 @file{/mnt/include/defs.h} by using the first rule. However, it would
6473 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
6474 @file{/mnt/src/lib/foo.c}.
6477 @item unset substitute-path [path]
6478 @kindex unset substitute-path
6479 If a path is specified, search the current list of substitution rules
6480 for a rule that would rewrite that path. Delete that rule if found.
6481 A warning is emitted by the debugger if no rule could be found.
6483 If no path is specified, then all substitution rules are deleted.
6485 @item show substitute-path [path]
6486 @kindex show substitute-path
6487 If a path is specified, then print the source path substitution rule
6488 which would rewrite that path, if any.
6490 If no path is specified, then print all existing source path substitution
6495 If your source path is cluttered with directories that are no longer of
6496 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
6497 versions of source. You can correct the situation as follows:
6501 Use @code{directory} with no argument to reset the source path to its default value.
6504 Use @code{directory} with suitable arguments to reinstall the
6505 directories you want in the source path. You can add all the
6506 directories in one command.
6510 @section Source and Machine Code
6511 @cindex source line and its code address
6513 You can use the command @code{info line} to map source lines to program
6514 addresses (and vice versa), and the command @code{disassemble} to display
6515 a range of addresses as machine instructions. You can use the command
6516 @code{set disassemble-next-line} to set whether to disassemble next
6517 source line when execution stops. When run under @sc{gnu} Emacs
6518 mode, the @code{info line} command causes the arrow to point to the
6519 line specified. Also, @code{info line} prints addresses in symbolic form as
6524 @item info line @var{linespec}
6525 Print the starting and ending addresses of the compiled code for
6526 source line @var{linespec}. You can specify source lines in any of
6527 the ways documented in @ref{Specify Location}.
6530 For example, we can use @code{info line} to discover the location of
6531 the object code for the first line of function
6532 @code{m4_changequote}:
6534 @c FIXME: I think this example should also show the addresses in
6535 @c symbolic form, as they usually would be displayed.
6537 (@value{GDBP}) info line m4_changequote
6538 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
6542 @cindex code address and its source line
6543 We can also inquire (using @code{*@var{addr}} as the form for
6544 @var{linespec}) what source line covers a particular address:
6546 (@value{GDBP}) info line *0x63ff
6547 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
6550 @cindex @code{$_} and @code{info line}
6551 @cindex @code{x} command, default address
6552 @kindex x@r{(examine), and} info line
6553 After @code{info line}, the default address for the @code{x} command
6554 is changed to the starting address of the line, so that @samp{x/i} is
6555 sufficient to begin examining the machine code (@pxref{Memory,
6556 ,Examining Memory}). Also, this address is saved as the value of the
6557 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
6562 @cindex assembly instructions
6563 @cindex instructions, assembly
6564 @cindex machine instructions
6565 @cindex listing machine instructions
6567 @itemx disassemble /m
6568 @itemx disassemble /r
6569 This specialized command dumps a range of memory as machine
6570 instructions. It can also print mixed source+disassembly by specifying
6571 the @code{/m} modifier and print the raw instructions in hex as well as
6572 in symbolic form by specifying the @code{/r}.
6573 The default memory range is the function surrounding the
6574 program counter of the selected frame. A single argument to this
6575 command is a program counter value; @value{GDBN} dumps the function
6576 surrounding this value. When two arguments are given, they should
6577 be separated by a comma, possibly surrounded by whitespace. The
6578 arguments specify a range of addresses (first inclusive, second exclusive)
6579 to dump. In that case, the name of the function is also printed (since
6580 there could be several functions in the given range).
6582 The argument(s) can be any expression yielding a numeric value, such as
6583 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
6585 If the range of memory being disassembled contains current program counter,
6586 the instruction at that location is shown with a @code{=>} marker.
6589 The following example shows the disassembly of a range of addresses of
6590 HP PA-RISC 2.0 code:
6593 (@value{GDBP}) disas 0x32c4, 0x32e4
6594 Dump of assembler code from 0x32c4 to 0x32e4:
6595 0x32c4 <main+204>: addil 0,dp
6596 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
6597 0x32cc <main+212>: ldil 0x3000,r31
6598 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
6599 0x32d4 <main+220>: ldo 0(r31),rp
6600 0x32d8 <main+224>: addil -0x800,dp
6601 0x32dc <main+228>: ldo 0x588(r1),r26
6602 0x32e0 <main+232>: ldil 0x3000,r31
6603 End of assembler dump.
6606 Here is an example showing mixed source+assembly for Intel x86, when the
6607 program is stopped just after function prologue:
6610 (@value{GDBP}) disas /m main
6611 Dump of assembler code for function main:
6613 0x08048330 <+0>: push %ebp
6614 0x08048331 <+1>: mov %esp,%ebp
6615 0x08048333 <+3>: sub $0x8,%esp
6616 0x08048336 <+6>: and $0xfffffff0,%esp
6617 0x08048339 <+9>: sub $0x10,%esp
6619 6 printf ("Hello.\n");
6620 => 0x0804833c <+12>: movl $0x8048440,(%esp)
6621 0x08048343 <+19>: call 0x8048284 <puts@@plt>
6625 0x08048348 <+24>: mov $0x0,%eax
6626 0x0804834d <+29>: leave
6627 0x0804834e <+30>: ret
6629 End of assembler dump.
6632 Some architectures have more than one commonly-used set of instruction
6633 mnemonics or other syntax.
6635 For programs that were dynamically linked and use shared libraries,
6636 instructions that call functions or branch to locations in the shared
6637 libraries might show a seemingly bogus location---it's actually a
6638 location of the relocation table. On some architectures, @value{GDBN}
6639 might be able to resolve these to actual function names.
6642 @kindex set disassembly-flavor
6643 @cindex Intel disassembly flavor
6644 @cindex AT&T disassembly flavor
6645 @item set disassembly-flavor @var{instruction-set}
6646 Select the instruction set to use when disassembling the
6647 program via the @code{disassemble} or @code{x/i} commands.
6649 Currently this command is only defined for the Intel x86 family. You
6650 can set @var{instruction-set} to either @code{intel} or @code{att}.
6651 The default is @code{att}, the AT&T flavor used by default by Unix
6652 assemblers for x86-based targets.
6654 @kindex show disassembly-flavor
6655 @item show disassembly-flavor
6656 Show the current setting of the disassembly flavor.
6660 @kindex set disassemble-next-line
6661 @kindex show disassemble-next-line
6662 @item set disassemble-next-line
6663 @itemx show disassemble-next-line
6664 Control whether or not @value{GDBN} will disassemble the next source
6665 line or instruction when execution stops. If ON, @value{GDBN} will
6666 display disassembly of the next source line when execution of the
6667 program being debugged stops. This is @emph{in addition} to
6668 displaying the source line itself, which @value{GDBN} always does if
6669 possible. If the next source line cannot be displayed for some reason
6670 (e.g., if @value{GDBN} cannot find the source file, or there's no line
6671 info in the debug info), @value{GDBN} will display disassembly of the
6672 next @emph{instruction} instead of showing the next source line. If
6673 AUTO, @value{GDBN} will display disassembly of next instruction only
6674 if the source line cannot be displayed. This setting causes
6675 @value{GDBN} to display some feedback when you step through a function
6676 with no line info or whose source file is unavailable. The default is
6677 OFF, which means never display the disassembly of the next line or
6683 @chapter Examining Data
6685 @cindex printing data
6686 @cindex examining data
6689 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
6690 @c document because it is nonstandard... Under Epoch it displays in a
6691 @c different window or something like that.
6692 The usual way to examine data in your program is with the @code{print}
6693 command (abbreviated @code{p}), or its synonym @code{inspect}. It
6694 evaluates and prints the value of an expression of the language your
6695 program is written in (@pxref{Languages, ,Using @value{GDBN} with
6696 Different Languages}). It may also print the expression using a
6697 Python-based pretty-printer (@pxref{Pretty Printing}).
6700 @item print @var{expr}
6701 @itemx print /@var{f} @var{expr}
6702 @var{expr} is an expression (in the source language). By default the
6703 value of @var{expr} is printed in a format appropriate to its data type;
6704 you can choose a different format by specifying @samp{/@var{f}}, where
6705 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
6709 @itemx print /@var{f}
6710 @cindex reprint the last value
6711 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
6712 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
6713 conveniently inspect the same value in an alternative format.
6716 A more low-level way of examining data is with the @code{x} command.
6717 It examines data in memory at a specified address and prints it in a
6718 specified format. @xref{Memory, ,Examining Memory}.
6720 If you are interested in information about types, or about how the
6721 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
6722 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
6726 * Expressions:: Expressions
6727 * Ambiguous Expressions:: Ambiguous Expressions
6728 * Variables:: Program variables
6729 * Arrays:: Artificial arrays
6730 * Output Formats:: Output formats
6731 * Memory:: Examining memory
6732 * Auto Display:: Automatic display
6733 * Print Settings:: Print settings
6734 * Value History:: Value history
6735 * Convenience Vars:: Convenience variables
6736 * Registers:: Registers
6737 * Floating Point Hardware:: Floating point hardware
6738 * Vector Unit:: Vector Unit
6739 * OS Information:: Auxiliary data provided by operating system
6740 * Memory Region Attributes:: Memory region attributes
6741 * Dump/Restore Files:: Copy between memory and a file
6742 * Core File Generation:: Cause a program dump its core
6743 * Character Sets:: Debugging programs that use a different
6744 character set than GDB does
6745 * Caching Remote Data:: Data caching for remote targets
6746 * Searching Memory:: Searching memory for a sequence of bytes
6750 @section Expressions
6753 @code{print} and many other @value{GDBN} commands accept an expression and
6754 compute its value. Any kind of constant, variable or operator defined
6755 by the programming language you are using is valid in an expression in
6756 @value{GDBN}. This includes conditional expressions, function calls,
6757 casts, and string constants. It also includes preprocessor macros, if
6758 you compiled your program to include this information; see
6761 @cindex arrays in expressions
6762 @value{GDBN} supports array constants in expressions input by
6763 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
6764 you can use the command @code{print @{1, 2, 3@}} to create an array
6765 of three integers. If you pass an array to a function or assign it
6766 to a program variable, @value{GDBN} copies the array to memory that
6767 is @code{malloc}ed in the target program.
6769 Because C is so widespread, most of the expressions shown in examples in
6770 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
6771 Languages}, for information on how to use expressions in other
6774 In this section, we discuss operators that you can use in @value{GDBN}
6775 expressions regardless of your programming language.
6777 @cindex casts, in expressions
6778 Casts are supported in all languages, not just in C, because it is so
6779 useful to cast a number into a pointer in order to examine a structure
6780 at that address in memory.
6781 @c FIXME: casts supported---Mod2 true?
6783 @value{GDBN} supports these operators, in addition to those common
6784 to programming languages:
6788 @samp{@@} is a binary operator for treating parts of memory as arrays.
6789 @xref{Arrays, ,Artificial Arrays}, for more information.
6792 @samp{::} allows you to specify a variable in terms of the file or
6793 function where it is defined. @xref{Variables, ,Program Variables}.
6795 @cindex @{@var{type}@}
6796 @cindex type casting memory
6797 @cindex memory, viewing as typed object
6798 @cindex casts, to view memory
6799 @item @{@var{type}@} @var{addr}
6800 Refers to an object of type @var{type} stored at address @var{addr} in
6801 memory. @var{addr} may be any expression whose value is an integer or
6802 pointer (but parentheses are required around binary operators, just as in
6803 a cast). This construct is allowed regardless of what kind of data is
6804 normally supposed to reside at @var{addr}.
6807 @node Ambiguous Expressions
6808 @section Ambiguous Expressions
6809 @cindex ambiguous expressions
6811 Expressions can sometimes contain some ambiguous elements. For instance,
6812 some programming languages (notably Ada, C@t{++} and Objective-C) permit
6813 a single function name to be defined several times, for application in
6814 different contexts. This is called @dfn{overloading}. Another example
6815 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
6816 templates and is typically instantiated several times, resulting in
6817 the same function name being defined in different contexts.
6819 In some cases and depending on the language, it is possible to adjust
6820 the expression to remove the ambiguity. For instance in C@t{++}, you
6821 can specify the signature of the function you want to break on, as in
6822 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
6823 qualified name of your function often makes the expression unambiguous
6826 When an ambiguity that needs to be resolved is detected, the debugger
6827 has the capability to display a menu of numbered choices for each
6828 possibility, and then waits for the selection with the prompt @samp{>}.
6829 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
6830 aborts the current command. If the command in which the expression was
6831 used allows more than one choice to be selected, the next option in the
6832 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
6835 For example, the following session excerpt shows an attempt to set a
6836 breakpoint at the overloaded symbol @code{String::after}.
6837 We choose three particular definitions of that function name:
6839 @c FIXME! This is likely to change to show arg type lists, at least
6842 (@value{GDBP}) b String::after
6845 [2] file:String.cc; line number:867
6846 [3] file:String.cc; line number:860
6847 [4] file:String.cc; line number:875
6848 [5] file:String.cc; line number:853
6849 [6] file:String.cc; line number:846
6850 [7] file:String.cc; line number:735
6852 Breakpoint 1 at 0xb26c: file String.cc, line 867.
6853 Breakpoint 2 at 0xb344: file String.cc, line 875.
6854 Breakpoint 3 at 0xafcc: file String.cc, line 846.
6855 Multiple breakpoints were set.
6856 Use the "delete" command to delete unwanted
6863 @kindex set multiple-symbols
6864 @item set multiple-symbols @var{mode}
6865 @cindex multiple-symbols menu
6867 This option allows you to adjust the debugger behavior when an expression
6870 By default, @var{mode} is set to @code{all}. If the command with which
6871 the expression is used allows more than one choice, then @value{GDBN}
6872 automatically selects all possible choices. For instance, inserting
6873 a breakpoint on a function using an ambiguous name results in a breakpoint
6874 inserted on each possible match. However, if a unique choice must be made,
6875 then @value{GDBN} uses the menu to help you disambiguate the expression.
6876 For instance, printing the address of an overloaded function will result
6877 in the use of the menu.
6879 When @var{mode} is set to @code{ask}, the debugger always uses the menu
6880 when an ambiguity is detected.
6882 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
6883 an error due to the ambiguity and the command is aborted.
6885 @kindex show multiple-symbols
6886 @item show multiple-symbols
6887 Show the current value of the @code{multiple-symbols} setting.
6891 @section Program Variables
6893 The most common kind of expression to use is the name of a variable
6896 Variables in expressions are understood in the selected stack frame
6897 (@pxref{Selection, ,Selecting a Frame}); they must be either:
6901 global (or file-static)
6908 visible according to the scope rules of the
6909 programming language from the point of execution in that frame
6912 @noindent This means that in the function
6927 you can examine and use the variable @code{a} whenever your program is
6928 executing within the function @code{foo}, but you can only use or
6929 examine the variable @code{b} while your program is executing inside
6930 the block where @code{b} is declared.
6932 @cindex variable name conflict
6933 There is an exception: you can refer to a variable or function whose
6934 scope is a single source file even if the current execution point is not
6935 in this file. But it is possible to have more than one such variable or
6936 function with the same name (in different source files). If that
6937 happens, referring to that name has unpredictable effects. If you wish,
6938 you can specify a static variable in a particular function or file,
6939 using the colon-colon (@code{::}) notation:
6941 @cindex colon-colon, context for variables/functions
6943 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
6944 @cindex @code{::}, context for variables/functions
6947 @var{file}::@var{variable}
6948 @var{function}::@var{variable}
6952 Here @var{file} or @var{function} is the name of the context for the
6953 static @var{variable}. In the case of file names, you can use quotes to
6954 make sure @value{GDBN} parses the file name as a single word---for example,
6955 to print a global value of @code{x} defined in @file{f2.c}:
6958 (@value{GDBP}) p 'f2.c'::x
6961 @cindex C@t{++} scope resolution
6962 This use of @samp{::} is very rarely in conflict with the very similar
6963 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
6964 scope resolution operator in @value{GDBN} expressions.
6965 @c FIXME: Um, so what happens in one of those rare cases where it's in
6968 @cindex wrong values
6969 @cindex variable values, wrong
6970 @cindex function entry/exit, wrong values of variables
6971 @cindex optimized code, wrong values of variables
6973 @emph{Warning:} Occasionally, a local variable may appear to have the
6974 wrong value at certain points in a function---just after entry to a new
6975 scope, and just before exit.
6977 You may see this problem when you are stepping by machine instructions.
6978 This is because, on most machines, it takes more than one instruction to
6979 set up a stack frame (including local variable definitions); if you are
6980 stepping by machine instructions, variables may appear to have the wrong
6981 values until the stack frame is completely built. On exit, it usually
6982 also takes more than one machine instruction to destroy a stack frame;
6983 after you begin stepping through that group of instructions, local
6984 variable definitions may be gone.
6986 This may also happen when the compiler does significant optimizations.
6987 To be sure of always seeing accurate values, turn off all optimization
6990 @cindex ``No symbol "foo" in current context''
6991 Another possible effect of compiler optimizations is to optimize
6992 unused variables out of existence, or assign variables to registers (as
6993 opposed to memory addresses). Depending on the support for such cases
6994 offered by the debug info format used by the compiler, @value{GDBN}
6995 might not be able to display values for such local variables. If that
6996 happens, @value{GDBN} will print a message like this:
6999 No symbol "foo" in current context.
7002 To solve such problems, either recompile without optimizations, or use a
7003 different debug info format, if the compiler supports several such
7004 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
7005 usually supports the @option{-gstabs+} option. @option{-gstabs+}
7006 produces debug info in a format that is superior to formats such as
7007 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
7008 an effective form for debug info. @xref{Debugging Options,,Options
7009 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
7010 Compiler Collection (GCC)}.
7011 @xref{C, ,C and C@t{++}}, for more information about debug info formats
7012 that are best suited to C@t{++} programs.
7014 If you ask to print an object whose contents are unknown to
7015 @value{GDBN}, e.g., because its data type is not completely specified
7016 by the debug information, @value{GDBN} will say @samp{<incomplete
7017 type>}. @xref{Symbols, incomplete type}, for more about this.
7019 Strings are identified as arrays of @code{char} values without specified
7020 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
7021 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
7022 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
7023 defines literal string type @code{"char"} as @code{char} without a sign.
7028 signed char var1[] = "A";
7031 You get during debugging
7036 $2 = @{65 'A', 0 '\0'@}
7040 @section Artificial Arrays
7042 @cindex artificial array
7044 @kindex @@@r{, referencing memory as an array}
7045 It is often useful to print out several successive objects of the
7046 same type in memory; a section of an array, or an array of
7047 dynamically determined size for which only a pointer exists in the
7050 You can do this by referring to a contiguous span of memory as an
7051 @dfn{artificial array}, using the binary operator @samp{@@}. The left
7052 operand of @samp{@@} should be the first element of the desired array
7053 and be an individual object. The right operand should be the desired length
7054 of the array. The result is an array value whose elements are all of
7055 the type of the left argument. The first element is actually the left
7056 argument; the second element comes from bytes of memory immediately
7057 following those that hold the first element, and so on. Here is an
7058 example. If a program says
7061 int *array = (int *) malloc (len * sizeof (int));
7065 you can print the contents of @code{array} with
7071 The left operand of @samp{@@} must reside in memory. Array values made
7072 with @samp{@@} in this way behave just like other arrays in terms of
7073 subscripting, and are coerced to pointers when used in expressions.
7074 Artificial arrays most often appear in expressions via the value history
7075 (@pxref{Value History, ,Value History}), after printing one out.
7077 Another way to create an artificial array is to use a cast.
7078 This re-interprets a value as if it were an array.
7079 The value need not be in memory:
7081 (@value{GDBP}) p/x (short[2])0x12345678
7082 $1 = @{0x1234, 0x5678@}
7085 As a convenience, if you leave the array length out (as in
7086 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
7087 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
7089 (@value{GDBP}) p/x (short[])0x12345678
7090 $2 = @{0x1234, 0x5678@}
7093 Sometimes the artificial array mechanism is not quite enough; in
7094 moderately complex data structures, the elements of interest may not
7095 actually be adjacent---for example, if you are interested in the values
7096 of pointers in an array. One useful work-around in this situation is
7097 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
7098 Variables}) as a counter in an expression that prints the first
7099 interesting value, and then repeat that expression via @key{RET}. For
7100 instance, suppose you have an array @code{dtab} of pointers to
7101 structures, and you are interested in the values of a field @code{fv}
7102 in each structure. Here is an example of what you might type:
7112 @node Output Formats
7113 @section Output Formats
7115 @cindex formatted output
7116 @cindex output formats
7117 By default, @value{GDBN} prints a value according to its data type. Sometimes
7118 this is not what you want. For example, you might want to print a number
7119 in hex, or a pointer in decimal. Or you might want to view data in memory
7120 at a certain address as a character string or as an instruction. To do
7121 these things, specify an @dfn{output format} when you print a value.
7123 The simplest use of output formats is to say how to print a value
7124 already computed. This is done by starting the arguments of the
7125 @code{print} command with a slash and a format letter. The format
7126 letters supported are:
7130 Regard the bits of the value as an integer, and print the integer in
7134 Print as integer in signed decimal.
7137 Print as integer in unsigned decimal.
7140 Print as integer in octal.
7143 Print as integer in binary. The letter @samp{t} stands for ``two''.
7144 @footnote{@samp{b} cannot be used because these format letters are also
7145 used with the @code{x} command, where @samp{b} stands for ``byte'';
7146 see @ref{Memory,,Examining Memory}.}
7149 @cindex unknown address, locating
7150 @cindex locate address
7151 Print as an address, both absolute in hexadecimal and as an offset from
7152 the nearest preceding symbol. You can use this format used to discover
7153 where (in what function) an unknown address is located:
7156 (@value{GDBP}) p/a 0x54320
7157 $3 = 0x54320 <_initialize_vx+396>
7161 The command @code{info symbol 0x54320} yields similar results.
7162 @xref{Symbols, info symbol}.
7165 Regard as an integer and print it as a character constant. This
7166 prints both the numerical value and its character representation. The
7167 character representation is replaced with the octal escape @samp{\nnn}
7168 for characters outside the 7-bit @sc{ascii} range.
7170 Without this format, @value{GDBN} displays @code{char},
7171 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
7172 constants. Single-byte members of vectors are displayed as integer
7176 Regard the bits of the value as a floating point number and print
7177 using typical floating point syntax.
7180 @cindex printing strings
7181 @cindex printing byte arrays
7182 Regard as a string, if possible. With this format, pointers to single-byte
7183 data are displayed as null-terminated strings and arrays of single-byte data
7184 are displayed as fixed-length strings. Other values are displayed in their
7187 Without this format, @value{GDBN} displays pointers to and arrays of
7188 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
7189 strings. Single-byte members of a vector are displayed as an integer
7193 @cindex raw printing
7194 Print using the @samp{raw} formatting. By default, @value{GDBN} will
7195 use a Python-based pretty-printer, if one is available (@pxref{Pretty
7196 Printing}). This typically results in a higher-level display of the
7197 value's contents. The @samp{r} format bypasses any Python
7198 pretty-printer which might exist.
7201 For example, to print the program counter in hex (@pxref{Registers}), type
7208 Note that no space is required before the slash; this is because command
7209 names in @value{GDBN} cannot contain a slash.
7211 To reprint the last value in the value history with a different format,
7212 you can use the @code{print} command with just a format and no
7213 expression. For example, @samp{p/x} reprints the last value in hex.
7216 @section Examining Memory
7218 You can use the command @code{x} (for ``examine'') to examine memory in
7219 any of several formats, independently of your program's data types.
7221 @cindex examining memory
7223 @kindex x @r{(examine memory)}
7224 @item x/@var{nfu} @var{addr}
7227 Use the @code{x} command to examine memory.
7230 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
7231 much memory to display and how to format it; @var{addr} is an
7232 expression giving the address where you want to start displaying memory.
7233 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
7234 Several commands set convenient defaults for @var{addr}.
7237 @item @var{n}, the repeat count
7238 The repeat count is a decimal integer; the default is 1. It specifies
7239 how much memory (counting by units @var{u}) to display.
7240 @c This really is **decimal**; unaffected by 'set radix' as of GDB
7243 @item @var{f}, the display format
7244 The display format is one of the formats used by @code{print}
7245 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
7246 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
7247 The default is @samp{x} (hexadecimal) initially. The default changes
7248 each time you use either @code{x} or @code{print}.
7250 @item @var{u}, the unit size
7251 The unit size is any of
7257 Halfwords (two bytes).
7259 Words (four bytes). This is the initial default.
7261 Giant words (eight bytes).
7264 Each time you specify a unit size with @code{x}, that size becomes the
7265 default unit the next time you use @code{x}. (For the @samp{s} and
7266 @samp{i} formats, the unit size is ignored and is normally not written.)
7268 @item @var{addr}, starting display address
7269 @var{addr} is the address where you want @value{GDBN} to begin displaying
7270 memory. The expression need not have a pointer value (though it may);
7271 it is always interpreted as an integer address of a byte of memory.
7272 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
7273 @var{addr} is usually just after the last address examined---but several
7274 other commands also set the default address: @code{info breakpoints} (to
7275 the address of the last breakpoint listed), @code{info line} (to the
7276 starting address of a line), and @code{print} (if you use it to display
7277 a value from memory).
7280 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
7281 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
7282 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
7283 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
7284 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
7286 Since the letters indicating unit sizes are all distinct from the
7287 letters specifying output formats, you do not have to remember whether
7288 unit size or format comes first; either order works. The output
7289 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
7290 (However, the count @var{n} must come first; @samp{wx4} does not work.)
7292 Even though the unit size @var{u} is ignored for the formats @samp{s}
7293 and @samp{i}, you might still want to use a count @var{n}; for example,
7294 @samp{3i} specifies that you want to see three machine instructions,
7295 including any operands. For convenience, especially when used with
7296 the @code{display} command, the @samp{i} format also prints branch delay
7297 slot instructions, if any, beyond the count specified, which immediately
7298 follow the last instruction that is within the count. The command
7299 @code{disassemble} gives an alternative way of inspecting machine
7300 instructions; see @ref{Machine Code,,Source and Machine Code}.
7302 All the defaults for the arguments to @code{x} are designed to make it
7303 easy to continue scanning memory with minimal specifications each time
7304 you use @code{x}. For example, after you have inspected three machine
7305 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
7306 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
7307 the repeat count @var{n} is used again; the other arguments default as
7308 for successive uses of @code{x}.
7310 When examining machine instructions, the instruction at current program
7311 counter is shown with a @code{=>} marker. For example:
7314 (@value{GDBP}) x/5i $pc-6
7315 0x804837f <main+11>: mov %esp,%ebp
7316 0x8048381 <main+13>: push %ecx
7317 0x8048382 <main+14>: sub $0x4,%esp
7318 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
7319 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
7322 @cindex @code{$_}, @code{$__}, and value history
7323 The addresses and contents printed by the @code{x} command are not saved
7324 in the value history because there is often too much of them and they
7325 would get in the way. Instead, @value{GDBN} makes these values available for
7326 subsequent use in expressions as values of the convenience variables
7327 @code{$_} and @code{$__}. After an @code{x} command, the last address
7328 examined is available for use in expressions in the convenience variable
7329 @code{$_}. The contents of that address, as examined, are available in
7330 the convenience variable @code{$__}.
7332 If the @code{x} command has a repeat count, the address and contents saved
7333 are from the last memory unit printed; this is not the same as the last
7334 address printed if several units were printed on the last line of output.
7336 @cindex remote memory comparison
7337 @cindex verify remote memory image
7338 When you are debugging a program running on a remote target machine
7339 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
7340 remote machine's memory against the executable file you downloaded to
7341 the target. The @code{compare-sections} command is provided for such
7345 @kindex compare-sections
7346 @item compare-sections @r{[}@var{section-name}@r{]}
7347 Compare the data of a loadable section @var{section-name} in the
7348 executable file of the program being debugged with the same section in
7349 the remote machine's memory, and report any mismatches. With no
7350 arguments, compares all loadable sections. This command's
7351 availability depends on the target's support for the @code{"qCRC"}
7356 @section Automatic Display
7357 @cindex automatic display
7358 @cindex display of expressions
7360 If you find that you want to print the value of an expression frequently
7361 (to see how it changes), you might want to add it to the @dfn{automatic
7362 display list} so that @value{GDBN} prints its value each time your program stops.
7363 Each expression added to the list is given a number to identify it;
7364 to remove an expression from the list, you specify that number.
7365 The automatic display looks like this:
7369 3: bar[5] = (struct hack *) 0x3804
7373 This display shows item numbers, expressions and their current values. As with
7374 displays you request manually using @code{x} or @code{print}, you can
7375 specify the output format you prefer; in fact, @code{display} decides
7376 whether to use @code{print} or @code{x} depending your format
7377 specification---it uses @code{x} if you specify either the @samp{i}
7378 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
7382 @item display @var{expr}
7383 Add the expression @var{expr} to the list of expressions to display
7384 each time your program stops. @xref{Expressions, ,Expressions}.
7386 @code{display} does not repeat if you press @key{RET} again after using it.
7388 @item display/@var{fmt} @var{expr}
7389 For @var{fmt} specifying only a display format and not a size or
7390 count, add the expression @var{expr} to the auto-display list but
7391 arrange to display it each time in the specified format @var{fmt}.
7392 @xref{Output Formats,,Output Formats}.
7394 @item display/@var{fmt} @var{addr}
7395 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
7396 number of units, add the expression @var{addr} as a memory address to
7397 be examined each time your program stops. Examining means in effect
7398 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
7401 For example, @samp{display/i $pc} can be helpful, to see the machine
7402 instruction about to be executed each time execution stops (@samp{$pc}
7403 is a common name for the program counter; @pxref{Registers, ,Registers}).
7406 @kindex delete display
7408 @item undisplay @var{dnums}@dots{}
7409 @itemx delete display @var{dnums}@dots{}
7410 Remove item numbers @var{dnums} from the list of expressions to display.
7412 @code{undisplay} does not repeat if you press @key{RET} after using it.
7413 (Otherwise you would just get the error @samp{No display number @dots{}}.)
7415 @kindex disable display
7416 @item disable display @var{dnums}@dots{}
7417 Disable the display of item numbers @var{dnums}. A disabled display
7418 item is not printed automatically, but is not forgotten. It may be
7419 enabled again later.
7421 @kindex enable display
7422 @item enable display @var{dnums}@dots{}
7423 Enable display of item numbers @var{dnums}. It becomes effective once
7424 again in auto display of its expression, until you specify otherwise.
7427 Display the current values of the expressions on the list, just as is
7428 done when your program stops.
7430 @kindex info display
7432 Print the list of expressions previously set up to display
7433 automatically, each one with its item number, but without showing the
7434 values. This includes disabled expressions, which are marked as such.
7435 It also includes expressions which would not be displayed right now
7436 because they refer to automatic variables not currently available.
7439 @cindex display disabled out of scope
7440 If a display expression refers to local variables, then it does not make
7441 sense outside the lexical context for which it was set up. Such an
7442 expression is disabled when execution enters a context where one of its
7443 variables is not defined. For example, if you give the command
7444 @code{display last_char} while inside a function with an argument
7445 @code{last_char}, @value{GDBN} displays this argument while your program
7446 continues to stop inside that function. When it stops elsewhere---where
7447 there is no variable @code{last_char}---the display is disabled
7448 automatically. The next time your program stops where @code{last_char}
7449 is meaningful, you can enable the display expression once again.
7451 @node Print Settings
7452 @section Print Settings
7454 @cindex format options
7455 @cindex print settings
7456 @value{GDBN} provides the following ways to control how arrays, structures,
7457 and symbols are printed.
7460 These settings are useful for debugging programs in any language:
7464 @item set print address
7465 @itemx set print address on
7466 @cindex print/don't print memory addresses
7467 @value{GDBN} prints memory addresses showing the location of stack
7468 traces, structure values, pointer values, breakpoints, and so forth,
7469 even when it also displays the contents of those addresses. The default
7470 is @code{on}. For example, this is what a stack frame display looks like with
7471 @code{set print address on}:
7476 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
7478 530 if (lquote != def_lquote)
7482 @item set print address off
7483 Do not print addresses when displaying their contents. For example,
7484 this is the same stack frame displayed with @code{set print address off}:
7488 (@value{GDBP}) set print addr off
7490 #0 set_quotes (lq="<<", rq=">>") at input.c:530
7491 530 if (lquote != def_lquote)
7495 You can use @samp{set print address off} to eliminate all machine
7496 dependent displays from the @value{GDBN} interface. For example, with
7497 @code{print address off}, you should get the same text for backtraces on
7498 all machines---whether or not they involve pointer arguments.
7501 @item show print address
7502 Show whether or not addresses are to be printed.
7505 When @value{GDBN} prints a symbolic address, it normally prints the
7506 closest earlier symbol plus an offset. If that symbol does not uniquely
7507 identify the address (for example, it is a name whose scope is a single
7508 source file), you may need to clarify. One way to do this is with
7509 @code{info line}, for example @samp{info line *0x4537}. Alternately,
7510 you can set @value{GDBN} to print the source file and line number when
7511 it prints a symbolic address:
7514 @item set print symbol-filename on
7515 @cindex source file and line of a symbol
7516 @cindex symbol, source file and line
7517 Tell @value{GDBN} to print the source file name and line number of a
7518 symbol in the symbolic form of an address.
7520 @item set print symbol-filename off
7521 Do not print source file name and line number of a symbol. This is the
7524 @item show print symbol-filename
7525 Show whether or not @value{GDBN} will print the source file name and
7526 line number of a symbol in the symbolic form of an address.
7529 Another situation where it is helpful to show symbol filenames and line
7530 numbers is when disassembling code; @value{GDBN} shows you the line
7531 number and source file that corresponds to each instruction.
7533 Also, you may wish to see the symbolic form only if the address being
7534 printed is reasonably close to the closest earlier symbol:
7537 @item set print max-symbolic-offset @var{max-offset}
7538 @cindex maximum value for offset of closest symbol
7539 Tell @value{GDBN} to only display the symbolic form of an address if the
7540 offset between the closest earlier symbol and the address is less than
7541 @var{max-offset}. The default is 0, which tells @value{GDBN}
7542 to always print the symbolic form of an address if any symbol precedes it.
7544 @item show print max-symbolic-offset
7545 Ask how large the maximum offset is that @value{GDBN} prints in a
7549 @cindex wild pointer, interpreting
7550 @cindex pointer, finding referent
7551 If you have a pointer and you are not sure where it points, try
7552 @samp{set print symbol-filename on}. Then you can determine the name
7553 and source file location of the variable where it points, using
7554 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
7555 For example, here @value{GDBN} shows that a variable @code{ptt} points
7556 at another variable @code{t}, defined in @file{hi2.c}:
7559 (@value{GDBP}) set print symbol-filename on
7560 (@value{GDBP}) p/a ptt
7561 $4 = 0xe008 <t in hi2.c>
7565 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
7566 does not show the symbol name and filename of the referent, even with
7567 the appropriate @code{set print} options turned on.
7570 Other settings control how different kinds of objects are printed:
7573 @item set print array
7574 @itemx set print array on
7575 @cindex pretty print arrays
7576 Pretty print arrays. This format is more convenient to read,
7577 but uses more space. The default is off.
7579 @item set print array off
7580 Return to compressed format for arrays.
7582 @item show print array
7583 Show whether compressed or pretty format is selected for displaying
7586 @cindex print array indexes
7587 @item set print array-indexes
7588 @itemx set print array-indexes on
7589 Print the index of each element when displaying arrays. May be more
7590 convenient to locate a given element in the array or quickly find the
7591 index of a given element in that printed array. The default is off.
7593 @item set print array-indexes off
7594 Stop printing element indexes when displaying arrays.
7596 @item show print array-indexes
7597 Show whether the index of each element is printed when displaying
7600 @item set print elements @var{number-of-elements}
7601 @cindex number of array elements to print
7602 @cindex limit on number of printed array elements
7603 Set a limit on how many elements of an array @value{GDBN} will print.
7604 If @value{GDBN} is printing a large array, it stops printing after it has
7605 printed the number of elements set by the @code{set print elements} command.
7606 This limit also applies to the display of strings.
7607 When @value{GDBN} starts, this limit is set to 200.
7608 Setting @var{number-of-elements} to zero means that the printing is unlimited.
7610 @item show print elements
7611 Display the number of elements of a large array that @value{GDBN} will print.
7612 If the number is 0, then the printing is unlimited.
7614 @item set print frame-arguments @var{value}
7615 @kindex set print frame-arguments
7616 @cindex printing frame argument values
7617 @cindex print all frame argument values
7618 @cindex print frame argument values for scalars only
7619 @cindex do not print frame argument values
7620 This command allows to control how the values of arguments are printed
7621 when the debugger prints a frame (@pxref{Frames}). The possible
7626 The values of all arguments are printed.
7629 Print the value of an argument only if it is a scalar. The value of more
7630 complex arguments such as arrays, structures, unions, etc, is replaced
7631 by @code{@dots{}}. This is the default. Here is an example where
7632 only scalar arguments are shown:
7635 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
7640 None of the argument values are printed. Instead, the value of each argument
7641 is replaced by @code{@dots{}}. In this case, the example above now becomes:
7644 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
7649 By default, only scalar arguments are printed. This command can be used
7650 to configure the debugger to print the value of all arguments, regardless
7651 of their type. However, it is often advantageous to not print the value
7652 of more complex parameters. For instance, it reduces the amount of
7653 information printed in each frame, making the backtrace more readable.
7654 Also, it improves performance when displaying Ada frames, because
7655 the computation of large arguments can sometimes be CPU-intensive,
7656 especially in large applications. Setting @code{print frame-arguments}
7657 to @code{scalars} (the default) or @code{none} avoids this computation,
7658 thus speeding up the display of each Ada frame.
7660 @item show print frame-arguments
7661 Show how the value of arguments should be displayed when printing a frame.
7663 @item set print repeats
7664 @cindex repeated array elements
7665 Set the threshold for suppressing display of repeated array
7666 elements. When the number of consecutive identical elements of an
7667 array exceeds the threshold, @value{GDBN} prints the string
7668 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
7669 identical repetitions, instead of displaying the identical elements
7670 themselves. Setting the threshold to zero will cause all elements to
7671 be individually printed. The default threshold is 10.
7673 @item show print repeats
7674 Display the current threshold for printing repeated identical
7677 @item set print null-stop
7678 @cindex @sc{null} elements in arrays
7679 Cause @value{GDBN} to stop printing the characters of an array when the first
7680 @sc{null} is encountered. This is useful when large arrays actually
7681 contain only short strings.
7684 @item show print null-stop
7685 Show whether @value{GDBN} stops printing an array on the first
7686 @sc{null} character.
7688 @item set print pretty on
7689 @cindex print structures in indented form
7690 @cindex indentation in structure display
7691 Cause @value{GDBN} to print structures in an indented format with one member
7692 per line, like this:
7707 @item set print pretty off
7708 Cause @value{GDBN} to print structures in a compact format, like this:
7712 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
7713 meat = 0x54 "Pork"@}
7718 This is the default format.
7720 @item show print pretty
7721 Show which format @value{GDBN} is using to print structures.
7723 @item set print sevenbit-strings on
7724 @cindex eight-bit characters in strings
7725 @cindex octal escapes in strings
7726 Print using only seven-bit characters; if this option is set,
7727 @value{GDBN} displays any eight-bit characters (in strings or
7728 character values) using the notation @code{\}@var{nnn}. This setting is
7729 best if you are working in English (@sc{ascii}) and you use the
7730 high-order bit of characters as a marker or ``meta'' bit.
7732 @item set print sevenbit-strings off
7733 Print full eight-bit characters. This allows the use of more
7734 international character sets, and is the default.
7736 @item show print sevenbit-strings
7737 Show whether or not @value{GDBN} is printing only seven-bit characters.
7739 @item set print union on
7740 @cindex unions in structures, printing
7741 Tell @value{GDBN} to print unions which are contained in structures
7742 and other unions. This is the default setting.
7744 @item set print union off
7745 Tell @value{GDBN} not to print unions which are contained in
7746 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
7749 @item show print union
7750 Ask @value{GDBN} whether or not it will print unions which are contained in
7751 structures and other unions.
7753 For example, given the declarations
7756 typedef enum @{Tree, Bug@} Species;
7757 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
7758 typedef enum @{Caterpillar, Cocoon, Butterfly@}
7769 struct thing foo = @{Tree, @{Acorn@}@};
7773 with @code{set print union on} in effect @samp{p foo} would print
7776 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
7780 and with @code{set print union off} in effect it would print
7783 $1 = @{it = Tree, form = @{...@}@}
7787 @code{set print union} affects programs written in C-like languages
7793 These settings are of interest when debugging C@t{++} programs:
7796 @cindex demangling C@t{++} names
7797 @item set print demangle
7798 @itemx set print demangle on
7799 Print C@t{++} names in their source form rather than in the encoded
7800 (``mangled'') form passed to the assembler and linker for type-safe
7801 linkage. The default is on.
7803 @item show print demangle
7804 Show whether C@t{++} names are printed in mangled or demangled form.
7806 @item set print asm-demangle
7807 @itemx set print asm-demangle on
7808 Print C@t{++} names in their source form rather than their mangled form, even
7809 in assembler code printouts such as instruction disassemblies.
7812 @item show print asm-demangle
7813 Show whether C@t{++} names in assembly listings are printed in mangled
7816 @cindex C@t{++} symbol decoding style
7817 @cindex symbol decoding style, C@t{++}
7818 @kindex set demangle-style
7819 @item set demangle-style @var{style}
7820 Choose among several encoding schemes used by different compilers to
7821 represent C@t{++} names. The choices for @var{style} are currently:
7825 Allow @value{GDBN} to choose a decoding style by inspecting your program.
7828 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
7829 This is the default.
7832 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
7835 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
7838 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
7839 @strong{Warning:} this setting alone is not sufficient to allow
7840 debugging @code{cfront}-generated executables. @value{GDBN} would
7841 require further enhancement to permit that.
7844 If you omit @var{style}, you will see a list of possible formats.
7846 @item show demangle-style
7847 Display the encoding style currently in use for decoding C@t{++} symbols.
7849 @item set print object
7850 @itemx set print object on
7851 @cindex derived type of an object, printing
7852 @cindex display derived types
7853 When displaying a pointer to an object, identify the @emph{actual}
7854 (derived) type of the object rather than the @emph{declared} type, using
7855 the virtual function table.
7857 @item set print object off
7858 Display only the declared type of objects, without reference to the
7859 virtual function table. This is the default setting.
7861 @item show print object
7862 Show whether actual, or declared, object types are displayed.
7864 @item set print static-members
7865 @itemx set print static-members on
7866 @cindex static members of C@t{++} objects
7867 Print static members when displaying a C@t{++} object. The default is on.
7869 @item set print static-members off
7870 Do not print static members when displaying a C@t{++} object.
7872 @item show print static-members
7873 Show whether C@t{++} static members are printed or not.
7875 @item set print pascal_static-members
7876 @itemx set print pascal_static-members on
7877 @cindex static members of Pascal objects
7878 @cindex Pascal objects, static members display
7879 Print static members when displaying a Pascal object. The default is on.
7881 @item set print pascal_static-members off
7882 Do not print static members when displaying a Pascal object.
7884 @item show print pascal_static-members
7885 Show whether Pascal static members are printed or not.
7887 @c These don't work with HP ANSI C++ yet.
7888 @item set print vtbl
7889 @itemx set print vtbl on
7890 @cindex pretty print C@t{++} virtual function tables
7891 @cindex virtual functions (C@t{++}) display
7892 @cindex VTBL display
7893 Pretty print C@t{++} virtual function tables. The default is off.
7894 (The @code{vtbl} commands do not work on programs compiled with the HP
7895 ANSI C@t{++} compiler (@code{aCC}).)
7897 @item set print vtbl off
7898 Do not pretty print C@t{++} virtual function tables.
7900 @item show print vtbl
7901 Show whether C@t{++} virtual function tables are pretty printed, or not.
7905 @section Value History
7907 @cindex value history
7908 @cindex history of values printed by @value{GDBN}
7909 Values printed by the @code{print} command are saved in the @value{GDBN}
7910 @dfn{value history}. This allows you to refer to them in other expressions.
7911 Values are kept until the symbol table is re-read or discarded
7912 (for example with the @code{file} or @code{symbol-file} commands).
7913 When the symbol table changes, the value history is discarded,
7914 since the values may contain pointers back to the types defined in the
7919 @cindex history number
7920 The values printed are given @dfn{history numbers} by which you can
7921 refer to them. These are successive integers starting with one.
7922 @code{print} shows you the history number assigned to a value by
7923 printing @samp{$@var{num} = } before the value; here @var{num} is the
7926 To refer to any previous value, use @samp{$} followed by the value's
7927 history number. The way @code{print} labels its output is designed to
7928 remind you of this. Just @code{$} refers to the most recent value in
7929 the history, and @code{$$} refers to the value before that.
7930 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
7931 is the value just prior to @code{$$}, @code{$$1} is equivalent to
7932 @code{$$}, and @code{$$0} is equivalent to @code{$}.
7934 For example, suppose you have just printed a pointer to a structure and
7935 want to see the contents of the structure. It suffices to type
7941 If you have a chain of structures where the component @code{next} points
7942 to the next one, you can print the contents of the next one with this:
7949 You can print successive links in the chain by repeating this
7950 command---which you can do by just typing @key{RET}.
7952 Note that the history records values, not expressions. If the value of
7953 @code{x} is 4 and you type these commands:
7961 then the value recorded in the value history by the @code{print} command
7962 remains 4 even though the value of @code{x} has changed.
7967 Print the last ten values in the value history, with their item numbers.
7968 This is like @samp{p@ $$9} repeated ten times, except that @code{show
7969 values} does not change the history.
7971 @item show values @var{n}
7972 Print ten history values centered on history item number @var{n}.
7975 Print ten history values just after the values last printed. If no more
7976 values are available, @code{show values +} produces no display.
7979 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
7980 same effect as @samp{show values +}.
7982 @node Convenience Vars
7983 @section Convenience Variables
7985 @cindex convenience variables
7986 @cindex user-defined variables
7987 @value{GDBN} provides @dfn{convenience variables} that you can use within
7988 @value{GDBN} to hold on to a value and refer to it later. These variables
7989 exist entirely within @value{GDBN}; they are not part of your program, and
7990 setting a convenience variable has no direct effect on further execution
7991 of your program. That is why you can use them freely.
7993 Convenience variables are prefixed with @samp{$}. Any name preceded by
7994 @samp{$} can be used for a convenience variable, unless it is one of
7995 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
7996 (Value history references, in contrast, are @emph{numbers} preceded
7997 by @samp{$}. @xref{Value History, ,Value History}.)
7999 You can save a value in a convenience variable with an assignment
8000 expression, just as you would set a variable in your program.
8004 set $foo = *object_ptr
8008 would save in @code{$foo} the value contained in the object pointed to by
8011 Using a convenience variable for the first time creates it, but its
8012 value is @code{void} until you assign a new value. You can alter the
8013 value with another assignment at any time.
8015 Convenience variables have no fixed types. You can assign a convenience
8016 variable any type of value, including structures and arrays, even if
8017 that variable already has a value of a different type. The convenience
8018 variable, when used as an expression, has the type of its current value.
8021 @kindex show convenience
8022 @cindex show all user variables
8023 @item show convenience
8024 Print a list of convenience variables used so far, and their values.
8025 Abbreviated @code{show conv}.
8027 @kindex init-if-undefined
8028 @cindex convenience variables, initializing
8029 @item init-if-undefined $@var{variable} = @var{expression}
8030 Set a convenience variable if it has not already been set. This is useful
8031 for user-defined commands that keep some state. It is similar, in concept,
8032 to using local static variables with initializers in C (except that
8033 convenience variables are global). It can also be used to allow users to
8034 override default values used in a command script.
8036 If the variable is already defined then the expression is not evaluated so
8037 any side-effects do not occur.
8040 One of the ways to use a convenience variable is as a counter to be
8041 incremented or a pointer to be advanced. For example, to print
8042 a field from successive elements of an array of structures:
8046 print bar[$i++]->contents
8050 Repeat that command by typing @key{RET}.
8052 Some convenience variables are created automatically by @value{GDBN} and given
8053 values likely to be useful.
8056 @vindex $_@r{, convenience variable}
8058 The variable @code{$_} is automatically set by the @code{x} command to
8059 the last address examined (@pxref{Memory, ,Examining Memory}). Other
8060 commands which provide a default address for @code{x} to examine also
8061 set @code{$_} to that address; these commands include @code{info line}
8062 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
8063 except when set by the @code{x} command, in which case it is a pointer
8064 to the type of @code{$__}.
8066 @vindex $__@r{, convenience variable}
8068 The variable @code{$__} is automatically set by the @code{x} command
8069 to the value found in the last address examined. Its type is chosen
8070 to match the format in which the data was printed.
8073 @vindex $_exitcode@r{, convenience variable}
8074 The variable @code{$_exitcode} is automatically set to the exit code when
8075 the program being debugged terminates.
8078 @vindex $_siginfo@r{, convenience variable}
8079 The variable @code{$_siginfo} contains extra signal information
8080 (@pxref{extra signal information}). Note that @code{$_siginfo}
8081 could be empty, if the application has not yet received any signals.
8082 For example, it will be empty before you execute the @code{run} command.
8085 @vindex $_tlb@r{, convenience variable}
8086 The variable @code{$_tlb} is automatically set when debugging
8087 applications running on MS-Windows in native mode or connected to
8088 gdbserver that supports the @code{qGetTIBAddr} request.
8089 @xref{General Query Packets}.
8090 This variable contains the address of the thread information block.
8094 On HP-UX systems, if you refer to a function or variable name that
8095 begins with a dollar sign, @value{GDBN} searches for a user or system
8096 name first, before it searches for a convenience variable.
8098 @cindex convenience functions
8099 @value{GDBN} also supplies some @dfn{convenience functions}. These
8100 have a syntax similar to convenience variables. A convenience
8101 function can be used in an expression just like an ordinary function;
8102 however, a convenience function is implemented internally to
8107 @kindex help function
8108 @cindex show all convenience functions
8109 Print a list of all convenience functions.
8116 You can refer to machine register contents, in expressions, as variables
8117 with names starting with @samp{$}. The names of registers are different
8118 for each machine; use @code{info registers} to see the names used on
8122 @kindex info registers
8123 @item info registers
8124 Print the names and values of all registers except floating-point
8125 and vector registers (in the selected stack frame).
8127 @kindex info all-registers
8128 @cindex floating point registers
8129 @item info all-registers
8130 Print the names and values of all registers, including floating-point
8131 and vector registers (in the selected stack frame).
8133 @item info registers @var{regname} @dots{}
8134 Print the @dfn{relativized} value of each specified register @var{regname}.
8135 As discussed in detail below, register values are normally relative to
8136 the selected stack frame. @var{regname} may be any register name valid on
8137 the machine you are using, with or without the initial @samp{$}.
8140 @cindex stack pointer register
8141 @cindex program counter register
8142 @cindex process status register
8143 @cindex frame pointer register
8144 @cindex standard registers
8145 @value{GDBN} has four ``standard'' register names that are available (in
8146 expressions) on most machines---whenever they do not conflict with an
8147 architecture's canonical mnemonics for registers. The register names
8148 @code{$pc} and @code{$sp} are used for the program counter register and
8149 the stack pointer. @code{$fp} is used for a register that contains a
8150 pointer to the current stack frame, and @code{$ps} is used for a
8151 register that contains the processor status. For example,
8152 you could print the program counter in hex with
8159 or print the instruction to be executed next with
8166 or add four to the stack pointer@footnote{This is a way of removing
8167 one word from the stack, on machines where stacks grow downward in
8168 memory (most machines, nowadays). This assumes that the innermost
8169 stack frame is selected; setting @code{$sp} is not allowed when other
8170 stack frames are selected. To pop entire frames off the stack,
8171 regardless of machine architecture, use @code{return};
8172 see @ref{Returning, ,Returning from a Function}.} with
8178 Whenever possible, these four standard register names are available on
8179 your machine even though the machine has different canonical mnemonics,
8180 so long as there is no conflict. The @code{info registers} command
8181 shows the canonical names. For example, on the SPARC, @code{info
8182 registers} displays the processor status register as @code{$psr} but you
8183 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
8184 is an alias for the @sc{eflags} register.
8186 @value{GDBN} always considers the contents of an ordinary register as an
8187 integer when the register is examined in this way. Some machines have
8188 special registers which can hold nothing but floating point; these
8189 registers are considered to have floating point values. There is no way
8190 to refer to the contents of an ordinary register as floating point value
8191 (although you can @emph{print} it as a floating point value with
8192 @samp{print/f $@var{regname}}).
8194 Some registers have distinct ``raw'' and ``virtual'' data formats. This
8195 means that the data format in which the register contents are saved by
8196 the operating system is not the same one that your program normally
8197 sees. For example, the registers of the 68881 floating point
8198 coprocessor are always saved in ``extended'' (raw) format, but all C
8199 programs expect to work with ``double'' (virtual) format. In such
8200 cases, @value{GDBN} normally works with the virtual format only (the format
8201 that makes sense for your program), but the @code{info registers} command
8202 prints the data in both formats.
8204 @cindex SSE registers (x86)
8205 @cindex MMX registers (x86)
8206 Some machines have special registers whose contents can be interpreted
8207 in several different ways. For example, modern x86-based machines
8208 have SSE and MMX registers that can hold several values packed
8209 together in several different formats. @value{GDBN} refers to such
8210 registers in @code{struct} notation:
8213 (@value{GDBP}) print $xmm1
8215 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
8216 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
8217 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
8218 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
8219 v4_int32 = @{0, 20657912, 11, 13@},
8220 v2_int64 = @{88725056443645952, 55834574859@},
8221 uint128 = 0x0000000d0000000b013b36f800000000
8226 To set values of such registers, you need to tell @value{GDBN} which
8227 view of the register you wish to change, as if you were assigning
8228 value to a @code{struct} member:
8231 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
8234 Normally, register values are relative to the selected stack frame
8235 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
8236 value that the register would contain if all stack frames farther in
8237 were exited and their saved registers restored. In order to see the
8238 true contents of hardware registers, you must select the innermost
8239 frame (with @samp{frame 0}).
8241 However, @value{GDBN} must deduce where registers are saved, from the machine
8242 code generated by your compiler. If some registers are not saved, or if
8243 @value{GDBN} is unable to locate the saved registers, the selected stack
8244 frame makes no difference.
8246 @node Floating Point Hardware
8247 @section Floating Point Hardware
8248 @cindex floating point
8250 Depending on the configuration, @value{GDBN} may be able to give
8251 you more information about the status of the floating point hardware.
8256 Display hardware-dependent information about the floating
8257 point unit. The exact contents and layout vary depending on the
8258 floating point chip. Currently, @samp{info float} is supported on
8259 the ARM and x86 machines.
8263 @section Vector Unit
8266 Depending on the configuration, @value{GDBN} may be able to give you
8267 more information about the status of the vector unit.
8272 Display information about the vector unit. The exact contents and
8273 layout vary depending on the hardware.
8276 @node OS Information
8277 @section Operating System Auxiliary Information
8278 @cindex OS information
8280 @value{GDBN} provides interfaces to useful OS facilities that can help
8281 you debug your program.
8283 @cindex @code{ptrace} system call
8284 @cindex @code{struct user} contents
8285 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
8286 machines), it interfaces with the inferior via the @code{ptrace}
8287 system call. The operating system creates a special sata structure,
8288 called @code{struct user}, for this interface. You can use the
8289 command @code{info udot} to display the contents of this data
8295 Display the contents of the @code{struct user} maintained by the OS
8296 kernel for the program being debugged. @value{GDBN} displays the
8297 contents of @code{struct user} as a list of hex numbers, similar to
8298 the @code{examine} command.
8301 @cindex auxiliary vector
8302 @cindex vector, auxiliary
8303 Some operating systems supply an @dfn{auxiliary vector} to programs at
8304 startup. This is akin to the arguments and environment that you
8305 specify for a program, but contains a system-dependent variety of
8306 binary values that tell system libraries important details about the
8307 hardware, operating system, and process. Each value's purpose is
8308 identified by an integer tag; the meanings are well-known but system-specific.
8309 Depending on the configuration and operating system facilities,
8310 @value{GDBN} may be able to show you this information. For remote
8311 targets, this functionality may further depend on the remote stub's
8312 support of the @samp{qXfer:auxv:read} packet, see
8313 @ref{qXfer auxiliary vector read}.
8318 Display the auxiliary vector of the inferior, which can be either a
8319 live process or a core dump file. @value{GDBN} prints each tag value
8320 numerically, and also shows names and text descriptions for recognized
8321 tags. Some values in the vector are numbers, some bit masks, and some
8322 pointers to strings or other data. @value{GDBN} displays each value in the
8323 most appropriate form for a recognized tag, and in hexadecimal for
8324 an unrecognized tag.
8327 On some targets, @value{GDBN} can access operating-system-specific information
8328 and display it to user, without interpretation. For remote targets,
8329 this functionality depends on the remote stub's support of the
8330 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
8333 @kindex info os processes
8334 @item info os processes
8335 Display the list of processes on the target. For each process,
8336 @value{GDBN} prints the process identifier, the name of the user, and
8337 the command corresponding to the process.
8340 @node Memory Region Attributes
8341 @section Memory Region Attributes
8342 @cindex memory region attributes
8344 @dfn{Memory region attributes} allow you to describe special handling
8345 required by regions of your target's memory. @value{GDBN} uses
8346 attributes to determine whether to allow certain types of memory
8347 accesses; whether to use specific width accesses; and whether to cache
8348 target memory. By default the description of memory regions is
8349 fetched from the target (if the current target supports this), but the
8350 user can override the fetched regions.
8352 Defined memory regions can be individually enabled and disabled. When a
8353 memory region is disabled, @value{GDBN} uses the default attributes when
8354 accessing memory in that region. Similarly, if no memory regions have
8355 been defined, @value{GDBN} uses the default attributes when accessing
8358 When a memory region is defined, it is given a number to identify it;
8359 to enable, disable, or remove a memory region, you specify that number.
8363 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
8364 Define a memory region bounded by @var{lower} and @var{upper} with
8365 attributes @var{attributes}@dots{}, and add it to the list of regions
8366 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
8367 case: it is treated as the target's maximum memory address.
8368 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
8371 Discard any user changes to the memory regions and use target-supplied
8372 regions, if available, or no regions if the target does not support.
8375 @item delete mem @var{nums}@dots{}
8376 Remove memory regions @var{nums}@dots{} from the list of regions
8377 monitored by @value{GDBN}.
8380 @item disable mem @var{nums}@dots{}
8381 Disable monitoring of memory regions @var{nums}@dots{}.
8382 A disabled memory region is not forgotten.
8383 It may be enabled again later.
8386 @item enable mem @var{nums}@dots{}
8387 Enable monitoring of memory regions @var{nums}@dots{}.
8391 Print a table of all defined memory regions, with the following columns
8395 @item Memory Region Number
8396 @item Enabled or Disabled.
8397 Enabled memory regions are marked with @samp{y}.
8398 Disabled memory regions are marked with @samp{n}.
8401 The address defining the inclusive lower bound of the memory region.
8404 The address defining the exclusive upper bound of the memory region.
8407 The list of attributes set for this memory region.
8412 @subsection Attributes
8414 @subsubsection Memory Access Mode
8415 The access mode attributes set whether @value{GDBN} may make read or
8416 write accesses to a memory region.
8418 While these attributes prevent @value{GDBN} from performing invalid
8419 memory accesses, they do nothing to prevent the target system, I/O DMA,
8420 etc.@: from accessing memory.
8424 Memory is read only.
8426 Memory is write only.
8428 Memory is read/write. This is the default.
8431 @subsubsection Memory Access Size
8432 The access size attribute tells @value{GDBN} to use specific sized
8433 accesses in the memory region. Often memory mapped device registers
8434 require specific sized accesses. If no access size attribute is
8435 specified, @value{GDBN} may use accesses of any size.
8439 Use 8 bit memory accesses.
8441 Use 16 bit memory accesses.
8443 Use 32 bit memory accesses.
8445 Use 64 bit memory accesses.
8448 @c @subsubsection Hardware/Software Breakpoints
8449 @c The hardware/software breakpoint attributes set whether @value{GDBN}
8450 @c will use hardware or software breakpoints for the internal breakpoints
8451 @c used by the step, next, finish, until, etc. commands.
8455 @c Always use hardware breakpoints
8456 @c @item swbreak (default)
8459 @subsubsection Data Cache
8460 The data cache attributes set whether @value{GDBN} will cache target
8461 memory. While this generally improves performance by reducing debug
8462 protocol overhead, it can lead to incorrect results because @value{GDBN}
8463 does not know about volatile variables or memory mapped device
8468 Enable @value{GDBN} to cache target memory.
8470 Disable @value{GDBN} from caching target memory. This is the default.
8473 @subsection Memory Access Checking
8474 @value{GDBN} can be instructed to refuse accesses to memory that is
8475 not explicitly described. This can be useful if accessing such
8476 regions has undesired effects for a specific target, or to provide
8477 better error checking. The following commands control this behaviour.
8480 @kindex set mem inaccessible-by-default
8481 @item set mem inaccessible-by-default [on|off]
8482 If @code{on} is specified, make @value{GDBN} treat memory not
8483 explicitly described by the memory ranges as non-existent and refuse accesses
8484 to such memory. The checks are only performed if there's at least one
8485 memory range defined. If @code{off} is specified, make @value{GDBN}
8486 treat the memory not explicitly described by the memory ranges as RAM.
8487 The default value is @code{on}.
8488 @kindex show mem inaccessible-by-default
8489 @item show mem inaccessible-by-default
8490 Show the current handling of accesses to unknown memory.
8494 @c @subsubsection Memory Write Verification
8495 @c The memory write verification attributes set whether @value{GDBN}
8496 @c will re-reads data after each write to verify the write was successful.
8500 @c @item noverify (default)
8503 @node Dump/Restore Files
8504 @section Copy Between Memory and a File
8505 @cindex dump/restore files
8506 @cindex append data to a file
8507 @cindex dump data to a file
8508 @cindex restore data from a file
8510 You can use the commands @code{dump}, @code{append}, and
8511 @code{restore} to copy data between target memory and a file. The
8512 @code{dump} and @code{append} commands write data to a file, and the
8513 @code{restore} command reads data from a file back into the inferior's
8514 memory. Files may be in binary, Motorola S-record, Intel hex, or
8515 Tektronix Hex format; however, @value{GDBN} can only append to binary
8521 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8522 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
8523 Dump the contents of memory from @var{start_addr} to @var{end_addr},
8524 or the value of @var{expr}, to @var{filename} in the given format.
8526 The @var{format} parameter may be any one of:
8533 Motorola S-record format.
8535 Tektronix Hex format.
8538 @value{GDBN} uses the same definitions of these formats as the
8539 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
8540 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
8544 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8545 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
8546 Append the contents of memory from @var{start_addr} to @var{end_addr},
8547 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
8548 (@value{GDBN} can only append data to files in raw binary form.)
8551 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
8552 Restore the contents of file @var{filename} into memory. The
8553 @code{restore} command can automatically recognize any known @sc{bfd}
8554 file format, except for raw binary. To restore a raw binary file you
8555 must specify the optional keyword @code{binary} after the filename.
8557 If @var{bias} is non-zero, its value will be added to the addresses
8558 contained in the file. Binary files always start at address zero, so
8559 they will be restored at address @var{bias}. Other bfd files have
8560 a built-in location; they will be restored at offset @var{bias}
8563 If @var{start} and/or @var{end} are non-zero, then only data between
8564 file offset @var{start} and file offset @var{end} will be restored.
8565 These offsets are relative to the addresses in the file, before
8566 the @var{bias} argument is applied.
8570 @node Core File Generation
8571 @section How to Produce a Core File from Your Program
8572 @cindex dump core from inferior
8574 A @dfn{core file} or @dfn{core dump} is a file that records the memory
8575 image of a running process and its process status (register values
8576 etc.). Its primary use is post-mortem debugging of a program that
8577 crashed while it ran outside a debugger. A program that crashes
8578 automatically produces a core file, unless this feature is disabled by
8579 the user. @xref{Files}, for information on invoking @value{GDBN} in
8580 the post-mortem debugging mode.
8582 Occasionally, you may wish to produce a core file of the program you
8583 are debugging in order to preserve a snapshot of its state.
8584 @value{GDBN} has a special command for that.
8588 @kindex generate-core-file
8589 @item generate-core-file [@var{file}]
8590 @itemx gcore [@var{file}]
8591 Produce a core dump of the inferior process. The optional argument
8592 @var{file} specifies the file name where to put the core dump. If not
8593 specified, the file name defaults to @file{core.@var{pid}}, where
8594 @var{pid} is the inferior process ID.
8596 Note that this command is implemented only for some systems (as of
8597 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
8600 @node Character Sets
8601 @section Character Sets
8602 @cindex character sets
8604 @cindex translating between character sets
8605 @cindex host character set
8606 @cindex target character set
8608 If the program you are debugging uses a different character set to
8609 represent characters and strings than the one @value{GDBN} uses itself,
8610 @value{GDBN} can automatically translate between the character sets for
8611 you. The character set @value{GDBN} uses we call the @dfn{host
8612 character set}; the one the inferior program uses we call the
8613 @dfn{target character set}.
8615 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
8616 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
8617 remote protocol (@pxref{Remote Debugging}) to debug a program
8618 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
8619 then the host character set is Latin-1, and the target character set is
8620 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
8621 target-charset EBCDIC-US}, then @value{GDBN} translates between
8622 @sc{ebcdic} and Latin 1 as you print character or string values, or use
8623 character and string literals in expressions.
8625 @value{GDBN} has no way to automatically recognize which character set
8626 the inferior program uses; you must tell it, using the @code{set
8627 target-charset} command, described below.
8629 Here are the commands for controlling @value{GDBN}'s character set
8633 @item set target-charset @var{charset}
8634 @kindex set target-charset
8635 Set the current target character set to @var{charset}. To display the
8636 list of supported target character sets, type
8637 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
8639 @item set host-charset @var{charset}
8640 @kindex set host-charset
8641 Set the current host character set to @var{charset}.
8643 By default, @value{GDBN} uses a host character set appropriate to the
8644 system it is running on; you can override that default using the
8645 @code{set host-charset} command. On some systems, @value{GDBN} cannot
8646 automatically determine the appropriate host character set. In this
8647 case, @value{GDBN} uses @samp{UTF-8}.
8649 @value{GDBN} can only use certain character sets as its host character
8650 set. If you type @kbd{@w{set target-charset @key{TAB}@key{TAB}}},
8651 @value{GDBN} will list the host character sets it supports.
8653 @item set charset @var{charset}
8655 Set the current host and target character sets to @var{charset}. As
8656 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
8657 @value{GDBN} will list the names of the character sets that can be used
8658 for both host and target.
8661 @kindex show charset
8662 Show the names of the current host and target character sets.
8664 @item show host-charset
8665 @kindex show host-charset
8666 Show the name of the current host character set.
8668 @item show target-charset
8669 @kindex show target-charset
8670 Show the name of the current target character set.
8672 @item set target-wide-charset @var{charset}
8673 @kindex set target-wide-charset
8674 Set the current target's wide character set to @var{charset}. This is
8675 the character set used by the target's @code{wchar_t} type. To
8676 display the list of supported wide character sets, type
8677 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
8679 @item show target-wide-charset
8680 @kindex show target-wide-charset
8681 Show the name of the current target's wide character set.
8684 Here is an example of @value{GDBN}'s character set support in action.
8685 Assume that the following source code has been placed in the file
8686 @file{charset-test.c}:
8692 = @{72, 101, 108, 108, 111, 44, 32, 119,
8693 111, 114, 108, 100, 33, 10, 0@};
8694 char ibm1047_hello[]
8695 = @{200, 133, 147, 147, 150, 107, 64, 166,
8696 150, 153, 147, 132, 90, 37, 0@};
8700 printf ("Hello, world!\n");
8704 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
8705 containing the string @samp{Hello, world!} followed by a newline,
8706 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
8708 We compile the program, and invoke the debugger on it:
8711 $ gcc -g charset-test.c -o charset-test
8712 $ gdb -nw charset-test
8713 GNU gdb 2001-12-19-cvs
8714 Copyright 2001 Free Software Foundation, Inc.
8719 We can use the @code{show charset} command to see what character sets
8720 @value{GDBN} is currently using to interpret and display characters and
8724 (@value{GDBP}) show charset
8725 The current host and target character set is `ISO-8859-1'.
8729 For the sake of printing this manual, let's use @sc{ascii} as our
8730 initial character set:
8732 (@value{GDBP}) set charset ASCII
8733 (@value{GDBP}) show charset
8734 The current host and target character set is `ASCII'.
8738 Let's assume that @sc{ascii} is indeed the correct character set for our
8739 host system --- in other words, let's assume that if @value{GDBN} prints
8740 characters using the @sc{ascii} character set, our terminal will display
8741 them properly. Since our current target character set is also
8742 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
8745 (@value{GDBP}) print ascii_hello
8746 $1 = 0x401698 "Hello, world!\n"
8747 (@value{GDBP}) print ascii_hello[0]
8752 @value{GDBN} uses the target character set for character and string
8753 literals you use in expressions:
8756 (@value{GDBP}) print '+'
8761 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
8764 @value{GDBN} relies on the user to tell it which character set the
8765 target program uses. If we print @code{ibm1047_hello} while our target
8766 character set is still @sc{ascii}, we get jibberish:
8769 (@value{GDBP}) print ibm1047_hello
8770 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
8771 (@value{GDBP}) print ibm1047_hello[0]
8776 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
8777 @value{GDBN} tells us the character sets it supports:
8780 (@value{GDBP}) set target-charset
8781 ASCII EBCDIC-US IBM1047 ISO-8859-1
8782 (@value{GDBP}) set target-charset
8785 We can select @sc{ibm1047} as our target character set, and examine the
8786 program's strings again. Now the @sc{ascii} string is wrong, but
8787 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
8788 target character set, @sc{ibm1047}, to the host character set,
8789 @sc{ascii}, and they display correctly:
8792 (@value{GDBP}) set target-charset IBM1047
8793 (@value{GDBP}) show charset
8794 The current host character set is `ASCII'.
8795 The current target character set is `IBM1047'.
8796 (@value{GDBP}) print ascii_hello
8797 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
8798 (@value{GDBP}) print ascii_hello[0]
8800 (@value{GDBP}) print ibm1047_hello
8801 $8 = 0x4016a8 "Hello, world!\n"
8802 (@value{GDBP}) print ibm1047_hello[0]
8807 As above, @value{GDBN} uses the target character set for character and
8808 string literals you use in expressions:
8811 (@value{GDBP}) print '+'
8816 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
8819 @node Caching Remote Data
8820 @section Caching Data of Remote Targets
8821 @cindex caching data of remote targets
8823 @value{GDBN} caches data exchanged between the debugger and a
8824 remote target (@pxref{Remote Debugging}). Such caching generally improves
8825 performance, because it reduces the overhead of the remote protocol by
8826 bundling memory reads and writes into large chunks. Unfortunately, simply
8827 caching everything would lead to incorrect results, since @value{GDBN}
8828 does not necessarily know anything about volatile values, memory-mapped I/O
8829 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
8830 memory can be changed @emph{while} a gdb command is executing.
8831 Therefore, by default, @value{GDBN} only caches data
8832 known to be on the stack@footnote{In non-stop mode, it is moderately
8833 rare for a running thread to modify the stack of a stopped thread
8834 in a way that would interfere with a backtrace, and caching of
8835 stack reads provides a significant speed up of remote backtraces.}.
8836 Other regions of memory can be explicitly marked as
8837 cacheable; see @pxref{Memory Region Attributes}.
8840 @kindex set remotecache
8841 @item set remotecache on
8842 @itemx set remotecache off
8843 This option no longer does anything; it exists for compatibility
8846 @kindex show remotecache
8847 @item show remotecache
8848 Show the current state of the obsolete remotecache flag.
8850 @kindex set stack-cache
8851 @item set stack-cache on
8852 @itemx set stack-cache off
8853 Enable or disable caching of stack accesses. When @code{ON}, use
8854 caching. By default, this option is @code{ON}.
8856 @kindex show stack-cache
8857 @item show stack-cache
8858 Show the current state of data caching for memory accesses.
8861 @item info dcache @r{[}line@r{]}
8862 Print the information about the data cache performance. The
8863 information displayed includes the dcache width and depth, and for
8864 each cache line, its number, address, and how many times it was
8865 referenced. This command is useful for debugging the data cache
8868 If a line number is specified, the contents of that line will be
8872 @node Searching Memory
8873 @section Search Memory
8874 @cindex searching memory
8876 Memory can be searched for a particular sequence of bytes with the
8877 @code{find} command.
8881 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8882 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8883 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
8884 etc. The search begins at address @var{start_addr} and continues for either
8885 @var{len} bytes or through to @var{end_addr} inclusive.
8888 @var{s} and @var{n} are optional parameters.
8889 They may be specified in either order, apart or together.
8892 @item @var{s}, search query size
8893 The size of each search query value.
8899 halfwords (two bytes)
8903 giant words (eight bytes)
8906 All values are interpreted in the current language.
8907 This means, for example, that if the current source language is C/C@t{++}
8908 then searching for the string ``hello'' includes the trailing '\0'.
8910 If the value size is not specified, it is taken from the
8911 value's type in the current language.
8912 This is useful when one wants to specify the search
8913 pattern as a mixture of types.
8914 Note that this means, for example, that in the case of C-like languages
8915 a search for an untyped 0x42 will search for @samp{(int) 0x42}
8916 which is typically four bytes.
8918 @item @var{n}, maximum number of finds
8919 The maximum number of matches to print. The default is to print all finds.
8922 You can use strings as search values. Quote them with double-quotes
8924 The string value is copied into the search pattern byte by byte,
8925 regardless of the endianness of the target and the size specification.
8927 The address of each match found is printed as well as a count of the
8928 number of matches found.
8930 The address of the last value found is stored in convenience variable
8932 A count of the number of matches is stored in @samp{$numfound}.
8934 For example, if stopped at the @code{printf} in this function:
8940 static char hello[] = "hello-hello";
8941 static struct @{ char c; short s; int i; @}
8942 __attribute__ ((packed)) mixed
8943 = @{ 'c', 0x1234, 0x87654321 @};
8944 printf ("%s\n", hello);
8949 you get during debugging:
8952 (gdb) find &hello[0], +sizeof(hello), "hello"
8953 0x804956d <hello.1620+6>
8955 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
8956 0x8049567 <hello.1620>
8957 0x804956d <hello.1620+6>
8959 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
8960 0x8049567 <hello.1620>
8962 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
8963 0x8049560 <mixed.1625>
8965 (gdb) print $numfound
8968 $2 = (void *) 0x8049560
8971 @node Optimized Code
8972 @chapter Debugging Optimized Code
8973 @cindex optimized code, debugging
8974 @cindex debugging optimized code
8976 Almost all compilers support optimization. With optimization
8977 disabled, the compiler generates assembly code that corresponds
8978 directly to your source code, in a simplistic way. As the compiler
8979 applies more powerful optimizations, the generated assembly code
8980 diverges from your original source code. With help from debugging
8981 information generated by the compiler, @value{GDBN} can map from
8982 the running program back to constructs from your original source.
8984 @value{GDBN} is more accurate with optimization disabled. If you
8985 can recompile without optimization, it is easier to follow the
8986 progress of your program during debugging. But, there are many cases
8987 where you may need to debug an optimized version.
8989 When you debug a program compiled with @samp{-g -O}, remember that the
8990 optimizer has rearranged your code; the debugger shows you what is
8991 really there. Do not be too surprised when the execution path does not
8992 exactly match your source file! An extreme example: if you define a
8993 variable, but never use it, @value{GDBN} never sees that
8994 variable---because the compiler optimizes it out of existence.
8996 Some things do not work as well with @samp{-g -O} as with just
8997 @samp{-g}, particularly on machines with instruction scheduling. If in
8998 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
8999 please report it to us as a bug (including a test case!).
9000 @xref{Variables}, for more information about debugging optimized code.
9003 * Inline Functions:: How @value{GDBN} presents inlining
9006 @node Inline Functions
9007 @section Inline Functions
9008 @cindex inline functions, debugging
9010 @dfn{Inlining} is an optimization that inserts a copy of the function
9011 body directly at each call site, instead of jumping to a shared
9012 routine. @value{GDBN} displays inlined functions just like
9013 non-inlined functions. They appear in backtraces. You can view their
9014 arguments and local variables, step into them with @code{step}, skip
9015 them with @code{next}, and escape from them with @code{finish}.
9016 You can check whether a function was inlined by using the
9017 @code{info frame} command.
9019 For @value{GDBN} to support inlined functions, the compiler must
9020 record information about inlining in the debug information ---
9021 @value{NGCC} using the @sc{dwarf 2} format does this, and several
9022 other compilers do also. @value{GDBN} only supports inlined functions
9023 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
9024 do not emit two required attributes (@samp{DW_AT_call_file} and
9025 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
9026 function calls with earlier versions of @value{NGCC}. It instead
9027 displays the arguments and local variables of inlined functions as
9028 local variables in the caller.
9030 The body of an inlined function is directly included at its call site;
9031 unlike a non-inlined function, there are no instructions devoted to
9032 the call. @value{GDBN} still pretends that the call site and the
9033 start of the inlined function are different instructions. Stepping to
9034 the call site shows the call site, and then stepping again shows
9035 the first line of the inlined function, even though no additional
9036 instructions are executed.
9038 This makes source-level debugging much clearer; you can see both the
9039 context of the call and then the effect of the call. Only stepping by
9040 a single instruction using @code{stepi} or @code{nexti} does not do
9041 this; single instruction steps always show the inlined body.
9043 There are some ways that @value{GDBN} does not pretend that inlined
9044 function calls are the same as normal calls:
9048 You cannot set breakpoints on inlined functions. @value{GDBN}
9049 either reports that there is no symbol with that name, or else sets the
9050 breakpoint only on non-inlined copies of the function. This limitation
9051 will be removed in a future version of @value{GDBN}; until then,
9052 set a breakpoint by line number on the first line of the inlined
9056 Setting breakpoints at the call site of an inlined function may not
9057 work, because the call site does not contain any code. @value{GDBN}
9058 may incorrectly move the breakpoint to the next line of the enclosing
9059 function, after the call. This limitation will be removed in a future
9060 version of @value{GDBN}; until then, set a breakpoint on an earlier line
9061 or inside the inlined function instead.
9064 @value{GDBN} cannot locate the return value of inlined calls after
9065 using the @code{finish} command. This is a limitation of compiler-generated
9066 debugging information; after @code{finish}, you can step to the next line
9067 and print a variable where your program stored the return value.
9073 @chapter C Preprocessor Macros
9075 Some languages, such as C and C@t{++}, provide a way to define and invoke
9076 ``preprocessor macros'' which expand into strings of tokens.
9077 @value{GDBN} can evaluate expressions containing macro invocations, show
9078 the result of macro expansion, and show a macro's definition, including
9079 where it was defined.
9081 You may need to compile your program specially to provide @value{GDBN}
9082 with information about preprocessor macros. Most compilers do not
9083 include macros in their debugging information, even when you compile
9084 with the @option{-g} flag. @xref{Compilation}.
9086 A program may define a macro at one point, remove that definition later,
9087 and then provide a different definition after that. Thus, at different
9088 points in the program, a macro may have different definitions, or have
9089 no definition at all. If there is a current stack frame, @value{GDBN}
9090 uses the macros in scope at that frame's source code line. Otherwise,
9091 @value{GDBN} uses the macros in scope at the current listing location;
9094 Whenever @value{GDBN} evaluates an expression, it always expands any
9095 macro invocations present in the expression. @value{GDBN} also provides
9096 the following commands for working with macros explicitly.
9100 @kindex macro expand
9101 @cindex macro expansion, showing the results of preprocessor
9102 @cindex preprocessor macro expansion, showing the results of
9103 @cindex expanding preprocessor macros
9104 @item macro expand @var{expression}
9105 @itemx macro exp @var{expression}
9106 Show the results of expanding all preprocessor macro invocations in
9107 @var{expression}. Since @value{GDBN} simply expands macros, but does
9108 not parse the result, @var{expression} need not be a valid expression;
9109 it can be any string of tokens.
9112 @item macro expand-once @var{expression}
9113 @itemx macro exp1 @var{expression}
9114 @cindex expand macro once
9115 @i{(This command is not yet implemented.)} Show the results of
9116 expanding those preprocessor macro invocations that appear explicitly in
9117 @var{expression}. Macro invocations appearing in that expansion are
9118 left unchanged. This command allows you to see the effect of a
9119 particular macro more clearly, without being confused by further
9120 expansions. Since @value{GDBN} simply expands macros, but does not
9121 parse the result, @var{expression} need not be a valid expression; it
9122 can be any string of tokens.
9125 @cindex macro definition, showing
9126 @cindex definition, showing a macro's
9127 @item info macro @var{macro}
9128 Show the definition of the macro named @var{macro}, and describe the
9129 source location or compiler command-line where that definition was established.
9131 @kindex macro define
9132 @cindex user-defined macros
9133 @cindex defining macros interactively
9134 @cindex macros, user-defined
9135 @item macro define @var{macro} @var{replacement-list}
9136 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
9137 Introduce a definition for a preprocessor macro named @var{macro},
9138 invocations of which are replaced by the tokens given in
9139 @var{replacement-list}. The first form of this command defines an
9140 ``object-like'' macro, which takes no arguments; the second form
9141 defines a ``function-like'' macro, which takes the arguments given in
9144 A definition introduced by this command is in scope in every
9145 expression evaluated in @value{GDBN}, until it is removed with the
9146 @code{macro undef} command, described below. The definition overrides
9147 all definitions for @var{macro} present in the program being debugged,
9148 as well as any previous user-supplied definition.
9151 @item macro undef @var{macro}
9152 Remove any user-supplied definition for the macro named @var{macro}.
9153 This command only affects definitions provided with the @code{macro
9154 define} command, described above; it cannot remove definitions present
9155 in the program being debugged.
9159 List all the macros defined using the @code{macro define} command.
9162 @cindex macros, example of debugging with
9163 Here is a transcript showing the above commands in action. First, we
9164 show our source files:
9172 #define ADD(x) (M + x)
9177 printf ("Hello, world!\n");
9179 printf ("We're so creative.\n");
9181 printf ("Goodbye, world!\n");
9188 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
9189 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
9190 compiler includes information about preprocessor macros in the debugging
9194 $ gcc -gdwarf-2 -g3 sample.c -o sample
9198 Now, we start @value{GDBN} on our sample program:
9202 GNU gdb 2002-05-06-cvs
9203 Copyright 2002 Free Software Foundation, Inc.
9204 GDB is free software, @dots{}
9208 We can expand macros and examine their definitions, even when the
9209 program is not running. @value{GDBN} uses the current listing position
9210 to decide which macro definitions are in scope:
9213 (@value{GDBP}) list main
9216 5 #define ADD(x) (M + x)
9221 10 printf ("Hello, world!\n");
9223 12 printf ("We're so creative.\n");
9224 (@value{GDBP}) info macro ADD
9225 Defined at /home/jimb/gdb/macros/play/sample.c:5
9226 #define ADD(x) (M + x)
9227 (@value{GDBP}) info macro Q
9228 Defined at /home/jimb/gdb/macros/play/sample.h:1
9229 included at /home/jimb/gdb/macros/play/sample.c:2
9231 (@value{GDBP}) macro expand ADD(1)
9232 expands to: (42 + 1)
9233 (@value{GDBP}) macro expand-once ADD(1)
9234 expands to: once (M + 1)
9238 In the example above, note that @code{macro expand-once} expands only
9239 the macro invocation explicit in the original text --- the invocation of
9240 @code{ADD} --- but does not expand the invocation of the macro @code{M},
9241 which was introduced by @code{ADD}.
9243 Once the program is running, @value{GDBN} uses the macro definitions in
9244 force at the source line of the current stack frame:
9247 (@value{GDBP}) break main
9248 Breakpoint 1 at 0x8048370: file sample.c, line 10.
9250 Starting program: /home/jimb/gdb/macros/play/sample
9252 Breakpoint 1, main () at sample.c:10
9253 10 printf ("Hello, world!\n");
9257 At line 10, the definition of the macro @code{N} at line 9 is in force:
9260 (@value{GDBP}) info macro N
9261 Defined at /home/jimb/gdb/macros/play/sample.c:9
9263 (@value{GDBP}) macro expand N Q M
9265 (@value{GDBP}) print N Q M
9270 As we step over directives that remove @code{N}'s definition, and then
9271 give it a new definition, @value{GDBN} finds the definition (or lack
9272 thereof) in force at each point:
9277 12 printf ("We're so creative.\n");
9278 (@value{GDBP}) info macro N
9279 The symbol `N' has no definition as a C/C++ preprocessor macro
9280 at /home/jimb/gdb/macros/play/sample.c:12
9283 14 printf ("Goodbye, world!\n");
9284 (@value{GDBP}) info macro N
9285 Defined at /home/jimb/gdb/macros/play/sample.c:13
9287 (@value{GDBP}) macro expand N Q M
9288 expands to: 1729 < 42
9289 (@value{GDBP}) print N Q M
9294 In addition to source files, macros can be defined on the compilation command
9295 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
9296 such a way, @value{GDBN} displays the location of their definition as line zero
9297 of the source file submitted to the compiler.
9300 (@value{GDBP}) info macro __STDC__
9301 Defined at /home/jimb/gdb/macros/play/sample.c:0
9308 @chapter Tracepoints
9309 @c This chapter is based on the documentation written by Michael
9310 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
9313 In some applications, it is not feasible for the debugger to interrupt
9314 the program's execution long enough for the developer to learn
9315 anything helpful about its behavior. If the program's correctness
9316 depends on its real-time behavior, delays introduced by a debugger
9317 might cause the program to change its behavior drastically, or perhaps
9318 fail, even when the code itself is correct. It is useful to be able
9319 to observe the program's behavior without interrupting it.
9321 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
9322 specify locations in the program, called @dfn{tracepoints}, and
9323 arbitrary expressions to evaluate when those tracepoints are reached.
9324 Later, using the @code{tfind} command, you can examine the values
9325 those expressions had when the program hit the tracepoints. The
9326 expressions may also denote objects in memory---structures or arrays,
9327 for example---whose values @value{GDBN} should record; while visiting
9328 a particular tracepoint, you may inspect those objects as if they were
9329 in memory at that moment. However, because @value{GDBN} records these
9330 values without interacting with you, it can do so quickly and
9331 unobtrusively, hopefully not disturbing the program's behavior.
9333 The tracepoint facility is currently available only for remote
9334 targets. @xref{Targets}. In addition, your remote target must know
9335 how to collect trace data. This functionality is implemented in the
9336 remote stub; however, none of the stubs distributed with @value{GDBN}
9337 support tracepoints as of this writing. The format of the remote
9338 packets used to implement tracepoints are described in @ref{Tracepoint
9341 It is also possible to get trace data from a file, in a manner reminiscent
9342 of corefiles; you specify the filename, and use @code{tfind} to search
9343 through the file. @xref{Trace Files}, for more details.
9345 This chapter describes the tracepoint commands and features.
9349 * Analyze Collected Data::
9350 * Tracepoint Variables::
9354 @node Set Tracepoints
9355 @section Commands to Set Tracepoints
9357 Before running such a @dfn{trace experiment}, an arbitrary number of
9358 tracepoints can be set. A tracepoint is actually a special type of
9359 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
9360 standard breakpoint commands. For instance, as with breakpoints,
9361 tracepoint numbers are successive integers starting from one, and many
9362 of the commands associated with tracepoints take the tracepoint number
9363 as their argument, to identify which tracepoint to work on.
9365 For each tracepoint, you can specify, in advance, some arbitrary set
9366 of data that you want the target to collect in the trace buffer when
9367 it hits that tracepoint. The collected data can include registers,
9368 local variables, or global data. Later, you can use @value{GDBN}
9369 commands to examine the values these data had at the time the
9372 Tracepoints do not support every breakpoint feature. Ignore counts on
9373 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
9374 commands when they are hit. Tracepoints may not be thread-specific
9377 @cindex fast tracepoints
9378 Some targets may support @dfn{fast tracepoints}, which are inserted in
9379 a different way (such as with a jump instead of a trap), that is
9380 faster but possibly restricted in where they may be installed.
9382 This section describes commands to set tracepoints and associated
9383 conditions and actions.
9386 * Create and Delete Tracepoints::
9387 * Enable and Disable Tracepoints::
9388 * Tracepoint Passcounts::
9389 * Tracepoint Conditions::
9390 * Trace State Variables::
9391 * Tracepoint Actions::
9392 * Listing Tracepoints::
9393 * Starting and Stopping Trace Experiments::
9394 * Tracepoint Restrictions::
9397 @node Create and Delete Tracepoints
9398 @subsection Create and Delete Tracepoints
9401 @cindex set tracepoint
9403 @item trace @var{location}
9404 The @code{trace} command is very similar to the @code{break} command.
9405 Its argument @var{location} can be a source line, a function name, or
9406 an address in the target program. @xref{Specify Location}. The
9407 @code{trace} command defines a tracepoint, which is a point in the
9408 target program where the debugger will briefly stop, collect some
9409 data, and then allow the program to continue. Setting a tracepoint or
9410 changing its actions doesn't take effect until the next @code{tstart}
9411 command, and once a trace experiment is running, further changes will
9412 not have any effect until the next trace experiment starts.
9414 Here are some examples of using the @code{trace} command:
9417 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
9419 (@value{GDBP}) @b{trace +2} // 2 lines forward
9421 (@value{GDBP}) @b{trace my_function} // first source line of function
9423 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
9425 (@value{GDBP}) @b{trace *0x2117c4} // an address
9429 You can abbreviate @code{trace} as @code{tr}.
9431 @item trace @var{location} if @var{cond}
9432 Set a tracepoint with condition @var{cond}; evaluate the expression
9433 @var{cond} each time the tracepoint is reached, and collect data only
9434 if the value is nonzero---that is, if @var{cond} evaluates as true.
9435 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
9436 information on tracepoint conditions.
9438 @item ftrace @var{location} [ if @var{cond} ]
9439 @cindex set fast tracepoint
9441 The @code{ftrace} command sets a fast tracepoint. For targets that
9442 support them, fast tracepoints will use a more efficient but possibly
9443 less general technique to trigger data collection, such as a jump
9444 instruction instead of a trap, or some sort of hardware support. It
9445 may not be possible to create a fast tracepoint at the desired
9446 location, in which case the command will exit with an explanatory
9449 @value{GDBN} handles arguments to @code{ftrace} exactly as for
9453 @cindex last tracepoint number
9454 @cindex recent tracepoint number
9455 @cindex tracepoint number
9456 The convenience variable @code{$tpnum} records the tracepoint number
9457 of the most recently set tracepoint.
9459 @kindex delete tracepoint
9460 @cindex tracepoint deletion
9461 @item delete tracepoint @r{[}@var{num}@r{]}
9462 Permanently delete one or more tracepoints. With no argument, the
9463 default is to delete all tracepoints. Note that the regular
9464 @code{delete} command can remove tracepoints also.
9469 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
9471 (@value{GDBP}) @b{delete trace} // remove all tracepoints
9475 You can abbreviate this command as @code{del tr}.
9478 @node Enable and Disable Tracepoints
9479 @subsection Enable and Disable Tracepoints
9481 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
9484 @kindex disable tracepoint
9485 @item disable tracepoint @r{[}@var{num}@r{]}
9486 Disable tracepoint @var{num}, or all tracepoints if no argument
9487 @var{num} is given. A disabled tracepoint will have no effect during
9488 the next trace experiment, but it is not forgotten. You can re-enable
9489 a disabled tracepoint using the @code{enable tracepoint} command.
9491 @kindex enable tracepoint
9492 @item enable tracepoint @r{[}@var{num}@r{]}
9493 Enable tracepoint @var{num}, or all tracepoints. The enabled
9494 tracepoints will become effective the next time a trace experiment is
9498 @node Tracepoint Passcounts
9499 @subsection Tracepoint Passcounts
9503 @cindex tracepoint pass count
9504 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
9505 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
9506 automatically stop a trace experiment. If a tracepoint's passcount is
9507 @var{n}, then the trace experiment will be automatically stopped on
9508 the @var{n}'th time that tracepoint is hit. If the tracepoint number
9509 @var{num} is not specified, the @code{passcount} command sets the
9510 passcount of the most recently defined tracepoint. If no passcount is
9511 given, the trace experiment will run until stopped explicitly by the
9517 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
9518 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
9520 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
9521 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
9522 (@value{GDBP}) @b{trace foo}
9523 (@value{GDBP}) @b{pass 3}
9524 (@value{GDBP}) @b{trace bar}
9525 (@value{GDBP}) @b{pass 2}
9526 (@value{GDBP}) @b{trace baz}
9527 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
9528 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
9529 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
9530 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
9534 @node Tracepoint Conditions
9535 @subsection Tracepoint Conditions
9536 @cindex conditional tracepoints
9537 @cindex tracepoint conditions
9539 The simplest sort of tracepoint collects data every time your program
9540 reaches a specified place. You can also specify a @dfn{condition} for
9541 a tracepoint. A condition is just a Boolean expression in your
9542 programming language (@pxref{Expressions, ,Expressions}). A
9543 tracepoint with a condition evaluates the expression each time your
9544 program reaches it, and data collection happens only if the condition
9547 Tracepoint conditions can be specified when a tracepoint is set, by
9548 using @samp{if} in the arguments to the @code{trace} command.
9549 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
9550 also be set or changed at any time with the @code{condition} command,
9551 just as with breakpoints.
9553 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
9554 the conditional expression itself. Instead, @value{GDBN} encodes the
9555 expression into an agent expression (@pxref{Agent Expressions}
9556 suitable for execution on the target, independently of @value{GDBN}.
9557 Global variables become raw memory locations, locals become stack
9558 accesses, and so forth.
9560 For instance, suppose you have a function that is usually called
9561 frequently, but should not be called after an error has occurred. You
9562 could use the following tracepoint command to collect data about calls
9563 of that function that happen while the error code is propagating
9564 through the program; an unconditional tracepoint could end up
9565 collecting thousands of useless trace frames that you would have to
9569 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
9572 @node Trace State Variables
9573 @subsection Trace State Variables
9574 @cindex trace state variables
9576 A @dfn{trace state variable} is a special type of variable that is
9577 created and managed by target-side code. The syntax is the same as
9578 that for GDB's convenience variables (a string prefixed with ``$''),
9579 but they are stored on the target. They must be created explicitly,
9580 using a @code{tvariable} command. They are always 64-bit signed
9583 Trace state variables are remembered by @value{GDBN}, and downloaded
9584 to the target along with tracepoint information when the trace
9585 experiment starts. There are no intrinsic limits on the number of
9586 trace state variables, beyond memory limitations of the target.
9588 @cindex convenience variables, and trace state variables
9589 Although trace state variables are managed by the target, you can use
9590 them in print commands and expressions as if they were convenience
9591 variables; @value{GDBN} will get the current value from the target
9592 while the trace experiment is running. Trace state variables share
9593 the same namespace as other ``$'' variables, which means that you
9594 cannot have trace state variables with names like @code{$23} or
9595 @code{$pc}, nor can you have a trace state variable and a convenience
9596 variable with the same name.
9600 @item tvariable $@var{name} [ = @var{expression} ]
9602 The @code{tvariable} command creates a new trace state variable named
9603 @code{$@var{name}}, and optionally gives it an initial value of
9604 @var{expression}. @var{expression} is evaluated when this command is
9605 entered; the result will be converted to an integer if possible,
9606 otherwise @value{GDBN} will report an error. A subsequent
9607 @code{tvariable} command specifying the same name does not create a
9608 variable, but instead assigns the supplied initial value to the
9609 existing variable of that name, overwriting any previous initial
9610 value. The default initial value is 0.
9612 @item info tvariables
9613 @kindex info tvariables
9614 List all the trace state variables along with their initial values.
9615 Their current values may also be displayed, if the trace experiment is
9618 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
9619 @kindex delete tvariable
9620 Delete the given trace state variables, or all of them if no arguments
9625 @node Tracepoint Actions
9626 @subsection Tracepoint Action Lists
9630 @cindex tracepoint actions
9631 @item actions @r{[}@var{num}@r{]}
9632 This command will prompt for a list of actions to be taken when the
9633 tracepoint is hit. If the tracepoint number @var{num} is not
9634 specified, this command sets the actions for the one that was most
9635 recently defined (so that you can define a tracepoint and then say
9636 @code{actions} without bothering about its number). You specify the
9637 actions themselves on the following lines, one action at a time, and
9638 terminate the actions list with a line containing just @code{end}. So
9639 far, the only defined actions are @code{collect}, @code{teval}, and
9640 @code{while-stepping}.
9642 @cindex remove actions from a tracepoint
9643 To remove all actions from a tracepoint, type @samp{actions @var{num}}
9644 and follow it immediately with @samp{end}.
9647 (@value{GDBP}) @b{collect @var{data}} // collect some data
9649 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
9651 (@value{GDBP}) @b{end} // signals the end of actions.
9654 In the following example, the action list begins with @code{collect}
9655 commands indicating the things to be collected when the tracepoint is
9656 hit. Then, in order to single-step and collect additional data
9657 following the tracepoint, a @code{while-stepping} command is used,
9658 followed by the list of things to be collected after each step in a
9659 sequence of single steps. The @code{while-stepping} command is
9660 terminated by its own separate @code{end} command. Lastly, the action
9661 list is terminated by an @code{end} command.
9664 (@value{GDBP}) @b{trace foo}
9665 (@value{GDBP}) @b{actions}
9666 Enter actions for tracepoint 1, one per line:
9675 @kindex collect @r{(tracepoints)}
9676 @item collect @var{expr1}, @var{expr2}, @dots{}
9677 Collect values of the given expressions when the tracepoint is hit.
9678 This command accepts a comma-separated list of any valid expressions.
9679 In addition to global, static, or local variables, the following
9680 special arguments are supported:
9684 collect all registers
9687 collect all function arguments
9690 collect all local variables.
9693 You can give several consecutive @code{collect} commands, each one
9694 with a single argument, or one @code{collect} command with several
9695 arguments separated by commas: the effect is the same.
9697 The command @code{info scope} (@pxref{Symbols, info scope}) is
9698 particularly useful for figuring out what data to collect.
9700 @kindex teval @r{(tracepoints)}
9701 @item teval @var{expr1}, @var{expr2}, @dots{}
9702 Evaluate the given expressions when the tracepoint is hit. This
9703 command accepts a comma-separated list of expressions. The results
9704 are discarded, so this is mainly useful for assigning values to trace
9705 state variables (@pxref{Trace State Variables}) without adding those
9706 values to the trace buffer, as would be the case if the @code{collect}
9709 @kindex while-stepping @r{(tracepoints)}
9710 @item while-stepping @var{n}
9711 Perform @var{n} single-step instruction traces after the tracepoint,
9712 collecting new data after each step. The @code{while-stepping}
9713 command is followed by the list of what to collect while stepping
9714 (followed by its own @code{end} command):
9718 > collect $regs, myglobal
9724 Note that @code{$pc} is not automatically collected by
9725 @code{while-stepping}; you need to explicitly collect that register if
9726 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
9729 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
9730 @kindex set default-collect
9731 @cindex default collection action
9732 This variable is a list of expressions to collect at each tracepoint
9733 hit. It is effectively an additional @code{collect} action prepended
9734 to every tracepoint action list. The expressions are parsed
9735 individually for each tracepoint, so for instance a variable named
9736 @code{xyz} may be interpreted as a global for one tracepoint, and a
9737 local for another, as appropriate to the tracepoint's location.
9739 @item show default-collect
9740 @kindex show default-collect
9741 Show the list of expressions that are collected by default at each
9746 @node Listing Tracepoints
9747 @subsection Listing Tracepoints
9750 @kindex info tracepoints
9752 @cindex information about tracepoints
9753 @item info tracepoints @r{[}@var{num}@r{]}
9754 Display information about the tracepoint @var{num}. If you don't
9755 specify a tracepoint number, displays information about all the
9756 tracepoints defined so far. The format is similar to that used for
9757 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
9758 command, simply restricting itself to tracepoints.
9760 A tracepoint's listing may include additional information specific to
9765 its passcount as given by the @code{passcount @var{n}} command
9767 its step count as given by the @code{while-stepping @var{n}} command
9769 its action list as given by the @code{actions} command. The actions
9770 are prefixed with an @samp{A} so as to distinguish them from commands.
9774 (@value{GDBP}) @b{info trace}
9775 Num Type Disp Enb Address What
9776 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
9780 A collect globfoo, $regs
9788 This command can be abbreviated @code{info tp}.
9791 @node Starting and Stopping Trace Experiments
9792 @subsection Starting and Stopping Trace Experiments
9796 @cindex start a new trace experiment
9797 @cindex collected data discarded
9799 This command takes no arguments. It starts the trace experiment, and
9800 begins collecting data. This has the side effect of discarding all
9801 the data collected in the trace buffer during the previous trace
9805 @cindex stop a running trace experiment
9807 This command takes no arguments. It ends the trace experiment, and
9808 stops collecting data.
9810 @strong{Note}: a trace experiment and data collection may stop
9811 automatically if any tracepoint's passcount is reached
9812 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
9815 @cindex status of trace data collection
9816 @cindex trace experiment, status of
9818 This command displays the status of the current trace data
9822 Here is an example of the commands we described so far:
9825 (@value{GDBP}) @b{trace gdb_c_test}
9826 (@value{GDBP}) @b{actions}
9827 Enter actions for tracepoint #1, one per line.
9828 > collect $regs,$locals,$args
9833 (@value{GDBP}) @b{tstart}
9834 [time passes @dots{}]
9835 (@value{GDBP}) @b{tstop}
9838 @cindex disconnected tracing
9839 You can choose to continue running the trace experiment even if
9840 @value{GDBN} disconnects from the target, voluntarily or
9841 involuntarily. For commands such as @code{detach}, the debugger will
9842 ask what you want to do with the trace. But for unexpected
9843 terminations (@value{GDBN} crash, network outage), it would be
9844 unfortunate to lose hard-won trace data, so the variable
9845 @code{disconnected-tracing} lets you decide whether the trace should
9846 continue running without @value{GDBN}.
9849 @item set disconnected-tracing on
9850 @itemx set disconnected-tracing off
9851 @kindex set disconnected-tracing
9852 Choose whether a tracing run should continue to run if @value{GDBN}
9853 has disconnected from the target. Note that @code{detach} or
9854 @code{quit} will ask you directly what to do about a running trace no
9855 matter what this variable's setting, so the variable is mainly useful
9856 for handling unexpected situations, such as loss of the network.
9858 @item show disconnected-tracing
9859 @kindex show disconnected-tracing
9860 Show the current choice for disconnected tracing.
9864 When you reconnect to the target, the trace experiment may or may not
9865 still be running; it might have filled the trace buffer in the
9866 meantime, or stopped for one of the other reasons. If it is running,
9867 it will continue after reconnection.
9869 Upon reconnection, the target will upload information about the
9870 tracepoints in effect. @value{GDBN} will then compare that
9871 information to the set of tracepoints currently defined, and attempt
9872 to match them up, allowing for the possibility that the numbers may
9873 have changed due to creation and deletion in the meantime. If one of
9874 the target's tracepoints does not match any in @value{GDBN}, the
9875 debugger will create a new tracepoint, so that you have a number with
9876 which to specify that tracepoint. This matching-up process is
9877 necessarily heuristic, and it may result in useless tracepoints being
9878 created; you may simply delete them if they are of no use.
9880 @cindex circular trace buffer
9881 If your target agent supports a @dfn{circular trace buffer}, then you
9882 can run a trace experiment indefinitely without filling the trace
9883 buffer; when space runs out, the agent deletes already-collected trace
9884 frames, oldest first, until there is enough room to continue
9885 collecting. This is especially useful if your tracepoints are being
9886 hit too often, and your trace gets terminated prematurely because the
9887 buffer is full. To ask for a circular trace buffer, simply set
9888 @samp{circular_trace_buffer} to on. You can set this at any time,
9889 including during tracing; if the agent can do it, it will change
9890 buffer handling on the fly, otherwise it will not take effect until
9894 @item set circular-trace-buffer on
9895 @itemx set circular-trace-buffer off
9896 @kindex set circular-trace-buffer
9897 Choose whether a tracing run should use a linear or circular buffer
9898 for trace data. A linear buffer will not lose any trace data, but may
9899 fill up prematurely, while a circular buffer will discard old trace
9900 data, but it will have always room for the latest tracepoint hits.
9902 @item show circular-trace-buffer
9903 @kindex show circular-trace-buffer
9904 Show the current choice for the trace buffer. Note that this may not
9905 match the agent's current buffer handling, nor is it guaranteed to
9906 match the setting that might have been in effect during a past run,
9907 for instance if you are looking at frames from a trace file.
9911 @node Tracepoint Restrictions
9912 @subsection Tracepoint Restrictions
9914 @cindex tracepoint restrictions
9915 There are a number of restrictions on the use of tracepoints. As
9916 described above, tracepoint data gathering occurs on the target
9917 without interaction from @value{GDBN}. Thus the full capabilities of
9918 the debugger are not available during data gathering, and then at data
9919 examination time, you will be limited by only having what was
9920 collected. The following items describe some common problems, but it
9921 is not exhaustive, and you may run into additional difficulties not
9927 Tracepoint expressions are intended to gather objects (lvalues). Thus
9928 the full flexibility of GDB's expression evaluator is not available.
9929 You cannot call functions, cast objects to aggregate types, access
9930 convenience variables or modify values (except by assignment to trace
9931 state variables). Some language features may implicitly call
9932 functions (for instance Objective-C fields with accessors), and therefore
9933 cannot be collected either.
9936 Collection of local variables, either individually or in bulk with
9937 @code{$locals} or @code{$args}, during @code{while-stepping} may
9938 behave erratically. The stepping action may enter a new scope (for
9939 instance by stepping into a function), or the location of the variable
9940 may change (for instance it is loaded into a register). The
9941 tracepoint data recorded uses the location information for the
9942 variables that is correct for the tracepoint location. When the
9943 tracepoint is created, it is not possible, in general, to determine
9944 where the steps of a @code{while-stepping} sequence will advance the
9945 program---particularly if a conditional branch is stepped.
9948 Collection of an incompletely-initialized or partially-destroyed object
9949 may result in something that @value{GDBN} cannot display, or displays
9950 in a misleading way.
9953 When @value{GDBN} displays a pointer to character it automatically
9954 dereferences the pointer to also display characters of the string
9955 being pointed to. However, collecting the pointer during tracing does
9956 not automatically collect the string. You need to explicitly
9957 dereference the pointer and provide size information if you want to
9958 collect not only the pointer, but the memory pointed to. For example,
9959 @code{*ptr@@50} can be used to collect the 50 element array pointed to
9963 It is not possible to collect a complete stack backtrace at a
9964 tracepoint. Instead, you may collect the registers and a few hundred
9965 bytes from the stack pointer with something like @code{*$esp@@300}
9966 (adjust to use the name of the actual stack pointer register on your
9967 target architecture, and the amount of stack you wish to capture).
9968 Then the @code{backtrace} command will show a partial backtrace when
9969 using a trace frame. The number of stack frames that can be examined
9970 depends on the sizes of the frames in the collected stack. Note that
9971 if you ask for a block so large that it goes past the bottom of the
9972 stack, the target agent may report an error trying to read from an
9976 If you do not collect registers at a tracepoint, @value{GDBN} can
9977 infer that the value of @code{$pc} must be the same as the address of
9978 the tracepoint and use that when you are looking at a trace frame
9979 for that tracepoint. However, this cannot work if the tracepoint has
9980 multiple locations (for instance if it was set in a function that was
9981 inlined), or if it has a @code{while-stepping} loop. In those cases
9982 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
9987 @node Analyze Collected Data
9988 @section Using the Collected Data
9990 After the tracepoint experiment ends, you use @value{GDBN} commands
9991 for examining the trace data. The basic idea is that each tracepoint
9992 collects a trace @dfn{snapshot} every time it is hit and another
9993 snapshot every time it single-steps. All these snapshots are
9994 consecutively numbered from zero and go into a buffer, and you can
9995 examine them later. The way you examine them is to @dfn{focus} on a
9996 specific trace snapshot. When the remote stub is focused on a trace
9997 snapshot, it will respond to all @value{GDBN} requests for memory and
9998 registers by reading from the buffer which belongs to that snapshot,
9999 rather than from @emph{real} memory or registers of the program being
10000 debugged. This means that @strong{all} @value{GDBN} commands
10001 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
10002 behave as if we were currently debugging the program state as it was
10003 when the tracepoint occurred. Any requests for data that are not in
10004 the buffer will fail.
10007 * tfind:: How to select a trace snapshot
10008 * tdump:: How to display all data for a snapshot
10009 * save tracepoints:: How to save tracepoints for a future run
10013 @subsection @code{tfind @var{n}}
10016 @cindex select trace snapshot
10017 @cindex find trace snapshot
10018 The basic command for selecting a trace snapshot from the buffer is
10019 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
10020 counting from zero. If no argument @var{n} is given, the next
10021 snapshot is selected.
10023 Here are the various forms of using the @code{tfind} command.
10027 Find the first snapshot in the buffer. This is a synonym for
10028 @code{tfind 0} (since 0 is the number of the first snapshot).
10031 Stop debugging trace snapshots, resume @emph{live} debugging.
10034 Same as @samp{tfind none}.
10037 No argument means find the next trace snapshot.
10040 Find the previous trace snapshot before the current one. This permits
10041 retracing earlier steps.
10043 @item tfind tracepoint @var{num}
10044 Find the next snapshot associated with tracepoint @var{num}. Search
10045 proceeds forward from the last examined trace snapshot. If no
10046 argument @var{num} is given, it means find the next snapshot collected
10047 for the same tracepoint as the current snapshot.
10049 @item tfind pc @var{addr}
10050 Find the next snapshot associated with the value @var{addr} of the
10051 program counter. Search proceeds forward from the last examined trace
10052 snapshot. If no argument @var{addr} is given, it means find the next
10053 snapshot with the same value of PC as the current snapshot.
10055 @item tfind outside @var{addr1}, @var{addr2}
10056 Find the next snapshot whose PC is outside the given range of
10057 addresses (exclusive).
10059 @item tfind range @var{addr1}, @var{addr2}
10060 Find the next snapshot whose PC is between @var{addr1} and
10061 @var{addr2} (inclusive).
10063 @item tfind line @r{[}@var{file}:@r{]}@var{n}
10064 Find the next snapshot associated with the source line @var{n}. If
10065 the optional argument @var{file} is given, refer to line @var{n} in
10066 that source file. Search proceeds forward from the last examined
10067 trace snapshot. If no argument @var{n} is given, it means find the
10068 next line other than the one currently being examined; thus saying
10069 @code{tfind line} repeatedly can appear to have the same effect as
10070 stepping from line to line in a @emph{live} debugging session.
10073 The default arguments for the @code{tfind} commands are specifically
10074 designed to make it easy to scan through the trace buffer. For
10075 instance, @code{tfind} with no argument selects the next trace
10076 snapshot, and @code{tfind -} with no argument selects the previous
10077 trace snapshot. So, by giving one @code{tfind} command, and then
10078 simply hitting @key{RET} repeatedly you can examine all the trace
10079 snapshots in order. Or, by saying @code{tfind -} and then hitting
10080 @key{RET} repeatedly you can examine the snapshots in reverse order.
10081 The @code{tfind line} command with no argument selects the snapshot
10082 for the next source line executed. The @code{tfind pc} command with
10083 no argument selects the next snapshot with the same program counter
10084 (PC) as the current frame. The @code{tfind tracepoint} command with
10085 no argument selects the next trace snapshot collected by the same
10086 tracepoint as the current one.
10088 In addition to letting you scan through the trace buffer manually,
10089 these commands make it easy to construct @value{GDBN} scripts that
10090 scan through the trace buffer and print out whatever collected data
10091 you are interested in. Thus, if we want to examine the PC, FP, and SP
10092 registers from each trace frame in the buffer, we can say this:
10095 (@value{GDBP}) @b{tfind start}
10096 (@value{GDBP}) @b{while ($trace_frame != -1)}
10097 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
10098 $trace_frame, $pc, $sp, $fp
10102 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
10103 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
10104 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
10105 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
10106 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
10107 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
10108 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
10109 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
10110 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
10111 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
10112 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
10115 Or, if we want to examine the variable @code{X} at each source line in
10119 (@value{GDBP}) @b{tfind start}
10120 (@value{GDBP}) @b{while ($trace_frame != -1)}
10121 > printf "Frame %d, X == %d\n", $trace_frame, X
10131 @subsection @code{tdump}
10133 @cindex dump all data collected at tracepoint
10134 @cindex tracepoint data, display
10136 This command takes no arguments. It prints all the data collected at
10137 the current trace snapshot.
10140 (@value{GDBP}) @b{trace 444}
10141 (@value{GDBP}) @b{actions}
10142 Enter actions for tracepoint #2, one per line:
10143 > collect $regs, $locals, $args, gdb_long_test
10146 (@value{GDBP}) @b{tstart}
10148 (@value{GDBP}) @b{tfind line 444}
10149 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
10151 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
10153 (@value{GDBP}) @b{tdump}
10154 Data collected at tracepoint 2, trace frame 1:
10155 d0 0xc4aa0085 -995491707
10159 d4 0x71aea3d 119204413
10162 d7 0x380035 3670069
10163 a0 0x19e24a 1696330
10164 a1 0x3000668 50333288
10166 a3 0x322000 3284992
10167 a4 0x3000698 50333336
10168 a5 0x1ad3cc 1758156
10169 fp 0x30bf3c 0x30bf3c
10170 sp 0x30bf34 0x30bf34
10172 pc 0x20b2c8 0x20b2c8
10176 p = 0x20e5b4 "gdb-test"
10183 gdb_long_test = 17 '\021'
10188 @code{tdump} works by scanning the tracepoint's current collection
10189 actions and printing the value of each expression listed. So
10190 @code{tdump} can fail, if after a run, you change the tracepoint's
10191 actions to mention variables that were not collected during the run.
10193 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
10194 uses the collected value of @code{$pc} to distinguish between trace
10195 frames that were collected at the tracepoint hit, and frames that were
10196 collected while stepping. This allows it to correctly choose whether
10197 to display the basic list of collections, or the collections from the
10198 body of the while-stepping loop. However, if @code{$pc} was not collected,
10199 then @code{tdump} will always attempt to dump using the basic collection
10200 list, and may fail if a while-stepping frame does not include all the
10201 same data that is collected at the tracepoint hit.
10202 @c This is getting pretty arcane, example would be good.
10204 @node save tracepoints
10205 @subsection @code{save tracepoints @var{filename}}
10206 @kindex save tracepoints
10207 @kindex save-tracepoints
10208 @cindex save tracepoints for future sessions
10210 This command saves all current tracepoint definitions together with
10211 their actions and passcounts, into a file @file{@var{filename}}
10212 suitable for use in a later debugging session. To read the saved
10213 tracepoint definitions, use the @code{source} command (@pxref{Command
10214 Files}). The @w{@code{save-tracepoints}} command is a deprecated
10215 alias for @w{@code{save tracepoints}}
10217 @node Tracepoint Variables
10218 @section Convenience Variables for Tracepoints
10219 @cindex tracepoint variables
10220 @cindex convenience variables for tracepoints
10223 @vindex $trace_frame
10224 @item (int) $trace_frame
10225 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
10226 snapshot is selected.
10228 @vindex $tracepoint
10229 @item (int) $tracepoint
10230 The tracepoint for the current trace snapshot.
10232 @vindex $trace_line
10233 @item (int) $trace_line
10234 The line number for the current trace snapshot.
10236 @vindex $trace_file
10237 @item (char []) $trace_file
10238 The source file for the current trace snapshot.
10240 @vindex $trace_func
10241 @item (char []) $trace_func
10242 The name of the function containing @code{$tracepoint}.
10245 Note: @code{$trace_file} is not suitable for use in @code{printf},
10246 use @code{output} instead.
10248 Here's a simple example of using these convenience variables for
10249 stepping through all the trace snapshots and printing some of their
10250 data. Note that these are not the same as trace state variables,
10251 which are managed by the target.
10254 (@value{GDBP}) @b{tfind start}
10256 (@value{GDBP}) @b{while $trace_frame != -1}
10257 > output $trace_file
10258 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
10264 @section Using Trace Files
10265 @cindex trace files
10267 In some situations, the target running a trace experiment may no
10268 longer be available; perhaps it crashed, or the hardware was needed
10269 for a different activity. To handle these cases, you can arrange to
10270 dump the trace data into a file, and later use that file as a source
10271 of trace data, via the @code{target tfile} command.
10276 @item tsave [ -r ] @var{filename}
10277 Save the trace data to @var{filename}. By default, this command
10278 assumes that @var{filename} refers to the host filesystem, so if
10279 necessary @value{GDBN} will copy raw trace data up from the target and
10280 then save it. If the target supports it, you can also supply the
10281 optional argument @code{-r} (``remote'') to direct the target to save
10282 the data directly into @var{filename} in its own filesystem, which may be
10283 more efficient if the trace buffer is very large. (Note, however, that
10284 @code{target tfile} can only read from files accessible to the host.)
10286 @kindex target tfile
10288 @item target tfile @var{filename}
10289 Use the file named @var{filename} as a source of trace data. Commands
10290 that examine data work as they do with a live target, but it is not
10291 possible to run any new trace experiments. @code{tstatus} will report
10292 the state of the trace run at the moment the data was saved, as well
10293 as the current trace frame you are examining. @var{filename} must be
10294 on a filesystem accessible to the host.
10299 @chapter Debugging Programs That Use Overlays
10302 If your program is too large to fit completely in your target system's
10303 memory, you can sometimes use @dfn{overlays} to work around this
10304 problem. @value{GDBN} provides some support for debugging programs that
10308 * How Overlays Work:: A general explanation of overlays.
10309 * Overlay Commands:: Managing overlays in @value{GDBN}.
10310 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
10311 mapped by asking the inferior.
10312 * Overlay Sample Program:: A sample program using overlays.
10315 @node How Overlays Work
10316 @section How Overlays Work
10317 @cindex mapped overlays
10318 @cindex unmapped overlays
10319 @cindex load address, overlay's
10320 @cindex mapped address
10321 @cindex overlay area
10323 Suppose you have a computer whose instruction address space is only 64
10324 kilobytes long, but which has much more memory which can be accessed by
10325 other means: special instructions, segment registers, or memory
10326 management hardware, for example. Suppose further that you want to
10327 adapt a program which is larger than 64 kilobytes to run on this system.
10329 One solution is to identify modules of your program which are relatively
10330 independent, and need not call each other directly; call these modules
10331 @dfn{overlays}. Separate the overlays from the main program, and place
10332 their machine code in the larger memory. Place your main program in
10333 instruction memory, but leave at least enough space there to hold the
10334 largest overlay as well.
10336 Now, to call a function located in an overlay, you must first copy that
10337 overlay's machine code from the large memory into the space set aside
10338 for it in the instruction memory, and then jump to its entry point
10341 @c NB: In the below the mapped area's size is greater or equal to the
10342 @c size of all overlays. This is intentional to remind the developer
10343 @c that overlays don't necessarily need to be the same size.
10347 Data Instruction Larger
10348 Address Space Address Space Address Space
10349 +-----------+ +-----------+ +-----------+
10351 +-----------+ +-----------+ +-----------+<-- overlay 1
10352 | program | | main | .----| overlay 1 | load address
10353 | variables | | program | | +-----------+
10354 | and heap | | | | | |
10355 +-----------+ | | | +-----------+<-- overlay 2
10356 | | +-----------+ | | | load address
10357 +-----------+ | | | .-| overlay 2 |
10359 mapped --->+-----------+ | | +-----------+
10360 address | | | | | |
10361 | overlay | <-' | | |
10362 | area | <---' +-----------+<-- overlay 3
10363 | | <---. | | load address
10364 +-----------+ `--| overlay 3 |
10371 @anchor{A code overlay}A code overlay
10375 The diagram (@pxref{A code overlay}) shows a system with separate data
10376 and instruction address spaces. To map an overlay, the program copies
10377 its code from the larger address space to the instruction address space.
10378 Since the overlays shown here all use the same mapped address, only one
10379 may be mapped at a time. For a system with a single address space for
10380 data and instructions, the diagram would be similar, except that the
10381 program variables and heap would share an address space with the main
10382 program and the overlay area.
10384 An overlay loaded into instruction memory and ready for use is called a
10385 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
10386 instruction memory. An overlay not present (or only partially present)
10387 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
10388 is its address in the larger memory. The mapped address is also called
10389 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
10390 called the @dfn{load memory address}, or @dfn{LMA}.
10392 Unfortunately, overlays are not a completely transparent way to adapt a
10393 program to limited instruction memory. They introduce a new set of
10394 global constraints you must keep in mind as you design your program:
10399 Before calling or returning to a function in an overlay, your program
10400 must make sure that overlay is actually mapped. Otherwise, the call or
10401 return will transfer control to the right address, but in the wrong
10402 overlay, and your program will probably crash.
10405 If the process of mapping an overlay is expensive on your system, you
10406 will need to choose your overlays carefully to minimize their effect on
10407 your program's performance.
10410 The executable file you load onto your system must contain each
10411 overlay's instructions, appearing at the overlay's load address, not its
10412 mapped address. However, each overlay's instructions must be relocated
10413 and its symbols defined as if the overlay were at its mapped address.
10414 You can use GNU linker scripts to specify different load and relocation
10415 addresses for pieces of your program; see @ref{Overlay Description,,,
10416 ld.info, Using ld: the GNU linker}.
10419 The procedure for loading executable files onto your system must be able
10420 to load their contents into the larger address space as well as the
10421 instruction and data spaces.
10425 The overlay system described above is rather simple, and could be
10426 improved in many ways:
10431 If your system has suitable bank switch registers or memory management
10432 hardware, you could use those facilities to make an overlay's load area
10433 contents simply appear at their mapped address in instruction space.
10434 This would probably be faster than copying the overlay to its mapped
10435 area in the usual way.
10438 If your overlays are small enough, you could set aside more than one
10439 overlay area, and have more than one overlay mapped at a time.
10442 You can use overlays to manage data, as well as instructions. In
10443 general, data overlays are even less transparent to your design than
10444 code overlays: whereas code overlays only require care when you call or
10445 return to functions, data overlays require care every time you access
10446 the data. Also, if you change the contents of a data overlay, you
10447 must copy its contents back out to its load address before you can copy a
10448 different data overlay into the same mapped area.
10453 @node Overlay Commands
10454 @section Overlay Commands
10456 To use @value{GDBN}'s overlay support, each overlay in your program must
10457 correspond to a separate section of the executable file. The section's
10458 virtual memory address and load memory address must be the overlay's
10459 mapped and load addresses. Identifying overlays with sections allows
10460 @value{GDBN} to determine the appropriate address of a function or
10461 variable, depending on whether the overlay is mapped or not.
10463 @value{GDBN}'s overlay commands all start with the word @code{overlay};
10464 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
10469 Disable @value{GDBN}'s overlay support. When overlay support is
10470 disabled, @value{GDBN} assumes that all functions and variables are
10471 always present at their mapped addresses. By default, @value{GDBN}'s
10472 overlay support is disabled.
10474 @item overlay manual
10475 @cindex manual overlay debugging
10476 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
10477 relies on you to tell it which overlays are mapped, and which are not,
10478 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
10479 commands described below.
10481 @item overlay map-overlay @var{overlay}
10482 @itemx overlay map @var{overlay}
10483 @cindex map an overlay
10484 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
10485 be the name of the object file section containing the overlay. When an
10486 overlay is mapped, @value{GDBN} assumes it can find the overlay's
10487 functions and variables at their mapped addresses. @value{GDBN} assumes
10488 that any other overlays whose mapped ranges overlap that of
10489 @var{overlay} are now unmapped.
10491 @item overlay unmap-overlay @var{overlay}
10492 @itemx overlay unmap @var{overlay}
10493 @cindex unmap an overlay
10494 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
10495 must be the name of the object file section containing the overlay.
10496 When an overlay is unmapped, @value{GDBN} assumes it can find the
10497 overlay's functions and variables at their load addresses.
10500 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
10501 consults a data structure the overlay manager maintains in the inferior
10502 to see which overlays are mapped. For details, see @ref{Automatic
10503 Overlay Debugging}.
10505 @item overlay load-target
10506 @itemx overlay load
10507 @cindex reloading the overlay table
10508 Re-read the overlay table from the inferior. Normally, @value{GDBN}
10509 re-reads the table @value{GDBN} automatically each time the inferior
10510 stops, so this command should only be necessary if you have changed the
10511 overlay mapping yourself using @value{GDBN}. This command is only
10512 useful when using automatic overlay debugging.
10514 @item overlay list-overlays
10515 @itemx overlay list
10516 @cindex listing mapped overlays
10517 Display a list of the overlays currently mapped, along with their mapped
10518 addresses, load addresses, and sizes.
10522 Normally, when @value{GDBN} prints a code address, it includes the name
10523 of the function the address falls in:
10526 (@value{GDBP}) print main
10527 $3 = @{int ()@} 0x11a0 <main>
10530 When overlay debugging is enabled, @value{GDBN} recognizes code in
10531 unmapped overlays, and prints the names of unmapped functions with
10532 asterisks around them. For example, if @code{foo} is a function in an
10533 unmapped overlay, @value{GDBN} prints it this way:
10536 (@value{GDBP}) overlay list
10537 No sections are mapped.
10538 (@value{GDBP}) print foo
10539 $5 = @{int (int)@} 0x100000 <*foo*>
10542 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
10546 (@value{GDBP}) overlay list
10547 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
10548 mapped at 0x1016 - 0x104a
10549 (@value{GDBP}) print foo
10550 $6 = @{int (int)@} 0x1016 <foo>
10553 When overlay debugging is enabled, @value{GDBN} can find the correct
10554 address for functions and variables in an overlay, whether or not the
10555 overlay is mapped. This allows most @value{GDBN} commands, like
10556 @code{break} and @code{disassemble}, to work normally, even on unmapped
10557 code. However, @value{GDBN}'s breakpoint support has some limitations:
10561 @cindex breakpoints in overlays
10562 @cindex overlays, setting breakpoints in
10563 You can set breakpoints in functions in unmapped overlays, as long as
10564 @value{GDBN} can write to the overlay at its load address.
10566 @value{GDBN} can not set hardware or simulator-based breakpoints in
10567 unmapped overlays. However, if you set a breakpoint at the end of your
10568 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
10569 you are using manual overlay management), @value{GDBN} will re-set its
10570 breakpoints properly.
10574 @node Automatic Overlay Debugging
10575 @section Automatic Overlay Debugging
10576 @cindex automatic overlay debugging
10578 @value{GDBN} can automatically track which overlays are mapped and which
10579 are not, given some simple co-operation from the overlay manager in the
10580 inferior. If you enable automatic overlay debugging with the
10581 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
10582 looks in the inferior's memory for certain variables describing the
10583 current state of the overlays.
10585 Here are the variables your overlay manager must define to support
10586 @value{GDBN}'s automatic overlay debugging:
10590 @item @code{_ovly_table}:
10591 This variable must be an array of the following structures:
10596 /* The overlay's mapped address. */
10599 /* The size of the overlay, in bytes. */
10600 unsigned long size;
10602 /* The overlay's load address. */
10605 /* Non-zero if the overlay is currently mapped;
10607 unsigned long mapped;
10611 @item @code{_novlys}:
10612 This variable must be a four-byte signed integer, holding the total
10613 number of elements in @code{_ovly_table}.
10617 To decide whether a particular overlay is mapped or not, @value{GDBN}
10618 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
10619 @code{lma} members equal the VMA and LMA of the overlay's section in the
10620 executable file. When @value{GDBN} finds a matching entry, it consults
10621 the entry's @code{mapped} member to determine whether the overlay is
10624 In addition, your overlay manager may define a function called
10625 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
10626 will silently set a breakpoint there. If the overlay manager then
10627 calls this function whenever it has changed the overlay table, this
10628 will enable @value{GDBN} to accurately keep track of which overlays
10629 are in program memory, and update any breakpoints that may be set
10630 in overlays. This will allow breakpoints to work even if the
10631 overlays are kept in ROM or other non-writable memory while they
10632 are not being executed.
10634 @node Overlay Sample Program
10635 @section Overlay Sample Program
10636 @cindex overlay example program
10638 When linking a program which uses overlays, you must place the overlays
10639 at their load addresses, while relocating them to run at their mapped
10640 addresses. To do this, you must write a linker script (@pxref{Overlay
10641 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
10642 since linker scripts are specific to a particular host system, target
10643 architecture, and target memory layout, this manual cannot provide
10644 portable sample code demonstrating @value{GDBN}'s overlay support.
10646 However, the @value{GDBN} source distribution does contain an overlaid
10647 program, with linker scripts for a few systems, as part of its test
10648 suite. The program consists of the following files from
10649 @file{gdb/testsuite/gdb.base}:
10653 The main program file.
10655 A simple overlay manager, used by @file{overlays.c}.
10660 Overlay modules, loaded and used by @file{overlays.c}.
10663 Linker scripts for linking the test program on the @code{d10v-elf}
10664 and @code{m32r-elf} targets.
10667 You can build the test program using the @code{d10v-elf} GCC
10668 cross-compiler like this:
10671 $ d10v-elf-gcc -g -c overlays.c
10672 $ d10v-elf-gcc -g -c ovlymgr.c
10673 $ d10v-elf-gcc -g -c foo.c
10674 $ d10v-elf-gcc -g -c bar.c
10675 $ d10v-elf-gcc -g -c baz.c
10676 $ d10v-elf-gcc -g -c grbx.c
10677 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
10678 baz.o grbx.o -Wl,-Td10v.ld -o overlays
10681 The build process is identical for any other architecture, except that
10682 you must substitute the appropriate compiler and linker script for the
10683 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
10687 @chapter Using @value{GDBN} with Different Languages
10690 Although programming languages generally have common aspects, they are
10691 rarely expressed in the same manner. For instance, in ANSI C,
10692 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
10693 Modula-2, it is accomplished by @code{p^}. Values can also be
10694 represented (and displayed) differently. Hex numbers in C appear as
10695 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
10697 @cindex working language
10698 Language-specific information is built into @value{GDBN} for some languages,
10699 allowing you to express operations like the above in your program's
10700 native language, and allowing @value{GDBN} to output values in a manner
10701 consistent with the syntax of your program's native language. The
10702 language you use to build expressions is called the @dfn{working
10706 * Setting:: Switching between source languages
10707 * Show:: Displaying the language
10708 * Checks:: Type and range checks
10709 * Supported Languages:: Supported languages
10710 * Unsupported Languages:: Unsupported languages
10714 @section Switching Between Source Languages
10716 There are two ways to control the working language---either have @value{GDBN}
10717 set it automatically, or select it manually yourself. You can use the
10718 @code{set language} command for either purpose. On startup, @value{GDBN}
10719 defaults to setting the language automatically. The working language is
10720 used to determine how expressions you type are interpreted, how values
10723 In addition to the working language, every source file that
10724 @value{GDBN} knows about has its own working language. For some object
10725 file formats, the compiler might indicate which language a particular
10726 source file is in. However, most of the time @value{GDBN} infers the
10727 language from the name of the file. The language of a source file
10728 controls whether C@t{++} names are demangled---this way @code{backtrace} can
10729 show each frame appropriately for its own language. There is no way to
10730 set the language of a source file from within @value{GDBN}, but you can
10731 set the language associated with a filename extension. @xref{Show, ,
10732 Displaying the Language}.
10734 This is most commonly a problem when you use a program, such
10735 as @code{cfront} or @code{f2c}, that generates C but is written in
10736 another language. In that case, make the
10737 program use @code{#line} directives in its C output; that way
10738 @value{GDBN} will know the correct language of the source code of the original
10739 program, and will display that source code, not the generated C code.
10742 * Filenames:: Filename extensions and languages.
10743 * Manually:: Setting the working language manually
10744 * Automatically:: Having @value{GDBN} infer the source language
10748 @subsection List of Filename Extensions and Languages
10750 If a source file name ends in one of the following extensions, then
10751 @value{GDBN} infers that its language is the one indicated.
10769 C@t{++} source file
10772 Objective-C source file
10776 Fortran source file
10779 Modula-2 source file
10783 Assembler source file. This actually behaves almost like C, but
10784 @value{GDBN} does not skip over function prologues when stepping.
10787 In addition, you may set the language associated with a filename
10788 extension. @xref{Show, , Displaying the Language}.
10791 @subsection Setting the Working Language
10793 If you allow @value{GDBN} to set the language automatically,
10794 expressions are interpreted the same way in your debugging session and
10797 @kindex set language
10798 If you wish, you may set the language manually. To do this, issue the
10799 command @samp{set language @var{lang}}, where @var{lang} is the name of
10800 a language, such as
10801 @code{c} or @code{modula-2}.
10802 For a list of the supported languages, type @samp{set language}.
10804 Setting the language manually prevents @value{GDBN} from updating the working
10805 language automatically. This can lead to confusion if you try
10806 to debug a program when the working language is not the same as the
10807 source language, when an expression is acceptable to both
10808 languages---but means different things. For instance, if the current
10809 source file were written in C, and @value{GDBN} was parsing Modula-2, a
10817 might not have the effect you intended. In C, this means to add
10818 @code{b} and @code{c} and place the result in @code{a}. The result
10819 printed would be the value of @code{a}. In Modula-2, this means to compare
10820 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
10822 @node Automatically
10823 @subsection Having @value{GDBN} Infer the Source Language
10825 To have @value{GDBN} set the working language automatically, use
10826 @samp{set language local} or @samp{set language auto}. @value{GDBN}
10827 then infers the working language. That is, when your program stops in a
10828 frame (usually by encountering a breakpoint), @value{GDBN} sets the
10829 working language to the language recorded for the function in that
10830 frame. If the language for a frame is unknown (that is, if the function
10831 or block corresponding to the frame was defined in a source file that
10832 does not have a recognized extension), the current working language is
10833 not changed, and @value{GDBN} issues a warning.
10835 This may not seem necessary for most programs, which are written
10836 entirely in one source language. However, program modules and libraries
10837 written in one source language can be used by a main program written in
10838 a different source language. Using @samp{set language auto} in this
10839 case frees you from having to set the working language manually.
10842 @section Displaying the Language
10844 The following commands help you find out which language is the
10845 working language, and also what language source files were written in.
10848 @item show language
10849 @kindex show language
10850 Display the current working language. This is the
10851 language you can use with commands such as @code{print} to
10852 build and compute expressions that may involve variables in your program.
10855 @kindex info frame@r{, show the source language}
10856 Display the source language for this frame. This language becomes the
10857 working language if you use an identifier from this frame.
10858 @xref{Frame Info, ,Information about a Frame}, to identify the other
10859 information listed here.
10862 @kindex info source@r{, show the source language}
10863 Display the source language of this source file.
10864 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
10865 information listed here.
10868 In unusual circumstances, you may have source files with extensions
10869 not in the standard list. You can then set the extension associated
10870 with a language explicitly:
10873 @item set extension-language @var{ext} @var{language}
10874 @kindex set extension-language
10875 Tell @value{GDBN} that source files with extension @var{ext} are to be
10876 assumed as written in the source language @var{language}.
10878 @item info extensions
10879 @kindex info extensions
10880 List all the filename extensions and the associated languages.
10884 @section Type and Range Checking
10887 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
10888 checking are included, but they do not yet have any effect. This
10889 section documents the intended facilities.
10891 @c FIXME remove warning when type/range code added
10893 Some languages are designed to guard you against making seemingly common
10894 errors through a series of compile- and run-time checks. These include
10895 checking the type of arguments to functions and operators, and making
10896 sure mathematical overflows are caught at run time. Checks such as
10897 these help to ensure a program's correctness once it has been compiled
10898 by eliminating type mismatches, and providing active checks for range
10899 errors when your program is running.
10901 @value{GDBN} can check for conditions like the above if you wish.
10902 Although @value{GDBN} does not check the statements in your program,
10903 it can check expressions entered directly into @value{GDBN} for
10904 evaluation via the @code{print} command, for example. As with the
10905 working language, @value{GDBN} can also decide whether or not to check
10906 automatically based on your program's source language.
10907 @xref{Supported Languages, ,Supported Languages}, for the default
10908 settings of supported languages.
10911 * Type Checking:: An overview of type checking
10912 * Range Checking:: An overview of range checking
10915 @cindex type checking
10916 @cindex checks, type
10917 @node Type Checking
10918 @subsection An Overview of Type Checking
10920 Some languages, such as Modula-2, are strongly typed, meaning that the
10921 arguments to operators and functions have to be of the correct type,
10922 otherwise an error occurs. These checks prevent type mismatch
10923 errors from ever causing any run-time problems. For example,
10931 The second example fails because the @code{CARDINAL} 1 is not
10932 type-compatible with the @code{REAL} 2.3.
10934 For the expressions you use in @value{GDBN} commands, you can tell the
10935 @value{GDBN} type checker to skip checking;
10936 to treat any mismatches as errors and abandon the expression;
10937 or to only issue warnings when type mismatches occur,
10938 but evaluate the expression anyway. When you choose the last of
10939 these, @value{GDBN} evaluates expressions like the second example above, but
10940 also issues a warning.
10942 Even if you turn type checking off, there may be other reasons
10943 related to type that prevent @value{GDBN} from evaluating an expression.
10944 For instance, @value{GDBN} does not know how to add an @code{int} and
10945 a @code{struct foo}. These particular type errors have nothing to do
10946 with the language in use, and usually arise from expressions, such as
10947 the one described above, which make little sense to evaluate anyway.
10949 Each language defines to what degree it is strict about type. For
10950 instance, both Modula-2 and C require the arguments to arithmetical
10951 operators to be numbers. In C, enumerated types and pointers can be
10952 represented as numbers, so that they are valid arguments to mathematical
10953 operators. @xref{Supported Languages, ,Supported Languages}, for further
10954 details on specific languages.
10956 @value{GDBN} provides some additional commands for controlling the type checker:
10958 @kindex set check type
10959 @kindex show check type
10961 @item set check type auto
10962 Set type checking on or off based on the current working language.
10963 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10966 @item set check type on
10967 @itemx set check type off
10968 Set type checking on or off, overriding the default setting for the
10969 current working language. Issue a warning if the setting does not
10970 match the language default. If any type mismatches occur in
10971 evaluating an expression while type checking is on, @value{GDBN} prints a
10972 message and aborts evaluation of the expression.
10974 @item set check type warn
10975 Cause the type checker to issue warnings, but to always attempt to
10976 evaluate the expression. Evaluating the expression may still
10977 be impossible for other reasons. For example, @value{GDBN} cannot add
10978 numbers and structures.
10981 Show the current setting of the type checker, and whether or not @value{GDBN}
10982 is setting it automatically.
10985 @cindex range checking
10986 @cindex checks, range
10987 @node Range Checking
10988 @subsection An Overview of Range Checking
10990 In some languages (such as Modula-2), it is an error to exceed the
10991 bounds of a type; this is enforced with run-time checks. Such range
10992 checking is meant to ensure program correctness by making sure
10993 computations do not overflow, or indices on an array element access do
10994 not exceed the bounds of the array.
10996 For expressions you use in @value{GDBN} commands, you can tell
10997 @value{GDBN} to treat range errors in one of three ways: ignore them,
10998 always treat them as errors and abandon the expression, or issue
10999 warnings but evaluate the expression anyway.
11001 A range error can result from numerical overflow, from exceeding an
11002 array index bound, or when you type a constant that is not a member
11003 of any type. Some languages, however, do not treat overflows as an
11004 error. In many implementations of C, mathematical overflow causes the
11005 result to ``wrap around'' to lower values---for example, if @var{m} is
11006 the largest integer value, and @var{s} is the smallest, then
11009 @var{m} + 1 @result{} @var{s}
11012 This, too, is specific to individual languages, and in some cases
11013 specific to individual compilers or machines. @xref{Supported Languages, ,
11014 Supported Languages}, for further details on specific languages.
11016 @value{GDBN} provides some additional commands for controlling the range checker:
11018 @kindex set check range
11019 @kindex show check range
11021 @item set check range auto
11022 Set range checking on or off based on the current working language.
11023 @xref{Supported Languages, ,Supported Languages}, for the default settings for
11026 @item set check range on
11027 @itemx set check range off
11028 Set range checking on or off, overriding the default setting for the
11029 current working language. A warning is issued if the setting does not
11030 match the language default. If a range error occurs and range checking is on,
11031 then a message is printed and evaluation of the expression is aborted.
11033 @item set check range warn
11034 Output messages when the @value{GDBN} range checker detects a range error,
11035 but attempt to evaluate the expression anyway. Evaluating the
11036 expression may still be impossible for other reasons, such as accessing
11037 memory that the process does not own (a typical example from many Unix
11041 Show the current setting of the range checker, and whether or not it is
11042 being set automatically by @value{GDBN}.
11045 @node Supported Languages
11046 @section Supported Languages
11048 @value{GDBN} supports C, C@t{++}, Objective-C, Fortran, Java, Pascal,
11049 assembly, Modula-2, and Ada.
11050 @c This is false ...
11051 Some @value{GDBN} features may be used in expressions regardless of the
11052 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
11053 and the @samp{@{type@}addr} construct (@pxref{Expressions,
11054 ,Expressions}) can be used with the constructs of any supported
11057 The following sections detail to what degree each source language is
11058 supported by @value{GDBN}. These sections are not meant to be language
11059 tutorials or references, but serve only as a reference guide to what the
11060 @value{GDBN} expression parser accepts, and what input and output
11061 formats should look like for different languages. There are many good
11062 books written on each of these languages; please look to these for a
11063 language reference or tutorial.
11066 * C:: C and C@t{++}
11067 * Objective-C:: Objective-C
11068 * Fortran:: Fortran
11070 * Modula-2:: Modula-2
11075 @subsection C and C@t{++}
11077 @cindex C and C@t{++}
11078 @cindex expressions in C or C@t{++}
11080 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
11081 to both languages. Whenever this is the case, we discuss those languages
11085 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
11086 @cindex @sc{gnu} C@t{++}
11087 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
11088 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
11089 effectively, you must compile your C@t{++} programs with a supported
11090 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
11091 compiler (@code{aCC}).
11093 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
11094 format; if it doesn't work on your system, try the stabs+ debugging
11095 format. You can select those formats explicitly with the @code{g++}
11096 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
11097 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
11098 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
11101 * C Operators:: C and C@t{++} operators
11102 * C Constants:: C and C@t{++} constants
11103 * C Plus Plus Expressions:: C@t{++} expressions
11104 * C Defaults:: Default settings for C and C@t{++}
11105 * C Checks:: C and C@t{++} type and range checks
11106 * Debugging C:: @value{GDBN} and C
11107 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
11108 * Decimal Floating Point:: Numbers in Decimal Floating Point format
11112 @subsubsection C and C@t{++} Operators
11114 @cindex C and C@t{++} operators
11116 Operators must be defined on values of specific types. For instance,
11117 @code{+} is defined on numbers, but not on structures. Operators are
11118 often defined on groups of types.
11120 For the purposes of C and C@t{++}, the following definitions hold:
11125 @emph{Integral types} include @code{int} with any of its storage-class
11126 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
11129 @emph{Floating-point types} include @code{float}, @code{double}, and
11130 @code{long double} (if supported by the target platform).
11133 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
11136 @emph{Scalar types} include all of the above.
11141 The following operators are supported. They are listed here
11142 in order of increasing precedence:
11146 The comma or sequencing operator. Expressions in a comma-separated list
11147 are evaluated from left to right, with the result of the entire
11148 expression being the last expression evaluated.
11151 Assignment. The value of an assignment expression is the value
11152 assigned. Defined on scalar types.
11155 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
11156 and translated to @w{@code{@var{a} = @var{a op b}}}.
11157 @w{@code{@var{op}=}} and @code{=} have the same precedence.
11158 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
11159 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
11162 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
11163 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
11167 Logical @sc{or}. Defined on integral types.
11170 Logical @sc{and}. Defined on integral types.
11173 Bitwise @sc{or}. Defined on integral types.
11176 Bitwise exclusive-@sc{or}. Defined on integral types.
11179 Bitwise @sc{and}. Defined on integral types.
11182 Equality and inequality. Defined on scalar types. The value of these
11183 expressions is 0 for false and non-zero for true.
11185 @item <@r{, }>@r{, }<=@r{, }>=
11186 Less than, greater than, less than or equal, greater than or equal.
11187 Defined on scalar types. The value of these expressions is 0 for false
11188 and non-zero for true.
11191 left shift, and right shift. Defined on integral types.
11194 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11197 Addition and subtraction. Defined on integral types, floating-point types and
11200 @item *@r{, }/@r{, }%
11201 Multiplication, division, and modulus. Multiplication and division are
11202 defined on integral and floating-point types. Modulus is defined on
11206 Increment and decrement. When appearing before a variable, the
11207 operation is performed before the variable is used in an expression;
11208 when appearing after it, the variable's value is used before the
11209 operation takes place.
11212 Pointer dereferencing. Defined on pointer types. Same precedence as
11216 Address operator. Defined on variables. Same precedence as @code{++}.
11218 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
11219 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
11220 to examine the address
11221 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
11225 Negative. Defined on integral and floating-point types. Same
11226 precedence as @code{++}.
11229 Logical negation. Defined on integral types. Same precedence as
11233 Bitwise complement operator. Defined on integral types. Same precedence as
11238 Structure member, and pointer-to-structure member. For convenience,
11239 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
11240 pointer based on the stored type information.
11241 Defined on @code{struct} and @code{union} data.
11244 Dereferences of pointers to members.
11247 Array indexing. @code{@var{a}[@var{i}]} is defined as
11248 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
11251 Function parameter list. Same precedence as @code{->}.
11254 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
11255 and @code{class} types.
11258 Doubled colons also represent the @value{GDBN} scope operator
11259 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
11263 If an operator is redefined in the user code, @value{GDBN} usually
11264 attempts to invoke the redefined version instead of using the operator's
11265 predefined meaning.
11268 @subsubsection C and C@t{++} Constants
11270 @cindex C and C@t{++} constants
11272 @value{GDBN} allows you to express the constants of C and C@t{++} in the
11277 Integer constants are a sequence of digits. Octal constants are
11278 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
11279 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
11280 @samp{l}, specifying that the constant should be treated as a
11284 Floating point constants are a sequence of digits, followed by a decimal
11285 point, followed by a sequence of digits, and optionally followed by an
11286 exponent. An exponent is of the form:
11287 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
11288 sequence of digits. The @samp{+} is optional for positive exponents.
11289 A floating-point constant may also end with a letter @samp{f} or
11290 @samp{F}, specifying that the constant should be treated as being of
11291 the @code{float} (as opposed to the default @code{double}) type; or with
11292 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
11296 Enumerated constants consist of enumerated identifiers, or their
11297 integral equivalents.
11300 Character constants are a single character surrounded by single quotes
11301 (@code{'}), or a number---the ordinal value of the corresponding character
11302 (usually its @sc{ascii} value). Within quotes, the single character may
11303 be represented by a letter or by @dfn{escape sequences}, which are of
11304 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
11305 of the character's ordinal value; or of the form @samp{\@var{x}}, where
11306 @samp{@var{x}} is a predefined special character---for example,
11307 @samp{\n} for newline.
11310 String constants are a sequence of character constants surrounded by
11311 double quotes (@code{"}). Any valid character constant (as described
11312 above) may appear. Double quotes within the string must be preceded by
11313 a backslash, so for instance @samp{"a\"b'c"} is a string of five
11317 Pointer constants are an integral value. You can also write pointers
11318 to constants using the C operator @samp{&}.
11321 Array constants are comma-separated lists surrounded by braces @samp{@{}
11322 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
11323 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
11324 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
11327 @node C Plus Plus Expressions
11328 @subsubsection C@t{++} Expressions
11330 @cindex expressions in C@t{++}
11331 @value{GDBN} expression handling can interpret most C@t{++} expressions.
11333 @cindex debugging C@t{++} programs
11334 @cindex C@t{++} compilers
11335 @cindex debug formats and C@t{++}
11336 @cindex @value{NGCC} and C@t{++}
11338 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
11339 proper compiler and the proper debug format. Currently, @value{GDBN}
11340 works best when debugging C@t{++} code that is compiled with
11341 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
11342 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
11343 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
11344 stabs+ as their default debug format, so you usually don't need to
11345 specify a debug format explicitly. Other compilers and/or debug formats
11346 are likely to work badly or not at all when using @value{GDBN} to debug
11352 @cindex member functions
11354 Member function calls are allowed; you can use expressions like
11357 count = aml->GetOriginal(x, y)
11360 @vindex this@r{, inside C@t{++} member functions}
11361 @cindex namespace in C@t{++}
11363 While a member function is active (in the selected stack frame), your
11364 expressions have the same namespace available as the member function;
11365 that is, @value{GDBN} allows implicit references to the class instance
11366 pointer @code{this} following the same rules as C@t{++}.
11368 @cindex call overloaded functions
11369 @cindex overloaded functions, calling
11370 @cindex type conversions in C@t{++}
11372 You can call overloaded functions; @value{GDBN} resolves the function
11373 call to the right definition, with some restrictions. @value{GDBN} does not
11374 perform overload resolution involving user-defined type conversions,
11375 calls to constructors, or instantiations of templates that do not exist
11376 in the program. It also cannot handle ellipsis argument lists or
11379 It does perform integral conversions and promotions, floating-point
11380 promotions, arithmetic conversions, pointer conversions, conversions of
11381 class objects to base classes, and standard conversions such as those of
11382 functions or arrays to pointers; it requires an exact match on the
11383 number of function arguments.
11385 Overload resolution is always performed, unless you have specified
11386 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
11387 ,@value{GDBN} Features for C@t{++}}.
11389 You must specify @code{set overload-resolution off} in order to use an
11390 explicit function signature to call an overloaded function, as in
11392 p 'foo(char,int)'('x', 13)
11395 The @value{GDBN} command-completion facility can simplify this;
11396 see @ref{Completion, ,Command Completion}.
11398 @cindex reference declarations
11400 @value{GDBN} understands variables declared as C@t{++} references; you can use
11401 them in expressions just as you do in C@t{++} source---they are automatically
11404 In the parameter list shown when @value{GDBN} displays a frame, the values of
11405 reference variables are not displayed (unlike other variables); this
11406 avoids clutter, since references are often used for large structures.
11407 The @emph{address} of a reference variable is always shown, unless
11408 you have specified @samp{set print address off}.
11411 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
11412 expressions can use it just as expressions in your program do. Since
11413 one scope may be defined in another, you can use @code{::} repeatedly if
11414 necessary, for example in an expression like
11415 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
11416 resolving name scope by reference to source files, in both C and C@t{++}
11417 debugging (@pxref{Variables, ,Program Variables}).
11420 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
11421 calling virtual functions correctly, printing out virtual bases of
11422 objects, calling functions in a base subobject, casting objects, and
11423 invoking user-defined operators.
11426 @subsubsection C and C@t{++} Defaults
11428 @cindex C and C@t{++} defaults
11430 If you allow @value{GDBN} to set type and range checking automatically, they
11431 both default to @code{off} whenever the working language changes to
11432 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
11433 selects the working language.
11435 If you allow @value{GDBN} to set the language automatically, it
11436 recognizes source files whose names end with @file{.c}, @file{.C}, or
11437 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
11438 these files, it sets the working language to C or C@t{++}.
11439 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
11440 for further details.
11442 @c Type checking is (a) primarily motivated by Modula-2, and (b)
11443 @c unimplemented. If (b) changes, it might make sense to let this node
11444 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
11447 @subsubsection C and C@t{++} Type and Range Checks
11449 @cindex C and C@t{++} checks
11451 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
11452 is not used. However, if you turn type checking on, @value{GDBN}
11453 considers two variables type equivalent if:
11457 The two variables are structured and have the same structure, union, or
11461 The two variables have the same type name, or types that have been
11462 declared equivalent through @code{typedef}.
11465 @c leaving this out because neither J Gilmore nor R Pesch understand it.
11468 The two @code{struct}, @code{union}, or @code{enum} variables are
11469 declared in the same declaration. (Note: this may not be true for all C
11474 Range checking, if turned on, is done on mathematical operations. Array
11475 indices are not checked, since they are often used to index a pointer
11476 that is not itself an array.
11479 @subsubsection @value{GDBN} and C
11481 The @code{set print union} and @code{show print union} commands apply to
11482 the @code{union} type. When set to @samp{on}, any @code{union} that is
11483 inside a @code{struct} or @code{class} is also printed. Otherwise, it
11484 appears as @samp{@{...@}}.
11486 The @code{@@} operator aids in the debugging of dynamic arrays, formed
11487 with pointers and a memory allocation function. @xref{Expressions,
11490 @node Debugging C Plus Plus
11491 @subsubsection @value{GDBN} Features for C@t{++}
11493 @cindex commands for C@t{++}
11495 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
11496 designed specifically for use with C@t{++}. Here is a summary:
11499 @cindex break in overloaded functions
11500 @item @r{breakpoint menus}
11501 When you want a breakpoint in a function whose name is overloaded,
11502 @value{GDBN} has the capability to display a menu of possible breakpoint
11503 locations to help you specify which function definition you want.
11504 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
11506 @cindex overloading in C@t{++}
11507 @item rbreak @var{regex}
11508 Setting breakpoints using regular expressions is helpful for setting
11509 breakpoints on overloaded functions that are not members of any special
11511 @xref{Set Breaks, ,Setting Breakpoints}.
11513 @cindex C@t{++} exception handling
11516 Debug C@t{++} exception handling using these commands. @xref{Set
11517 Catchpoints, , Setting Catchpoints}.
11519 @cindex inheritance
11520 @item ptype @var{typename}
11521 Print inheritance relationships as well as other information for type
11523 @xref{Symbols, ,Examining the Symbol Table}.
11525 @cindex C@t{++} symbol display
11526 @item set print demangle
11527 @itemx show print demangle
11528 @itemx set print asm-demangle
11529 @itemx show print asm-demangle
11530 Control whether C@t{++} symbols display in their source form, both when
11531 displaying code as C@t{++} source and when displaying disassemblies.
11532 @xref{Print Settings, ,Print Settings}.
11534 @item set print object
11535 @itemx show print object
11536 Choose whether to print derived (actual) or declared types of objects.
11537 @xref{Print Settings, ,Print Settings}.
11539 @item set print vtbl
11540 @itemx show print vtbl
11541 Control the format for printing virtual function tables.
11542 @xref{Print Settings, ,Print Settings}.
11543 (The @code{vtbl} commands do not work on programs compiled with the HP
11544 ANSI C@t{++} compiler (@code{aCC}).)
11546 @kindex set overload-resolution
11547 @cindex overloaded functions, overload resolution
11548 @item set overload-resolution on
11549 Enable overload resolution for C@t{++} expression evaluation. The default
11550 is on. For overloaded functions, @value{GDBN} evaluates the arguments
11551 and searches for a function whose signature matches the argument types,
11552 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
11553 Expressions, ,C@t{++} Expressions}, for details).
11554 If it cannot find a match, it emits a message.
11556 @item set overload-resolution off
11557 Disable overload resolution for C@t{++} expression evaluation. For
11558 overloaded functions that are not class member functions, @value{GDBN}
11559 chooses the first function of the specified name that it finds in the
11560 symbol table, whether or not its arguments are of the correct type. For
11561 overloaded functions that are class member functions, @value{GDBN}
11562 searches for a function whose signature @emph{exactly} matches the
11565 @kindex show overload-resolution
11566 @item show overload-resolution
11567 Show the current setting of overload resolution.
11569 @item @r{Overloaded symbol names}
11570 You can specify a particular definition of an overloaded symbol, using
11571 the same notation that is used to declare such symbols in C@t{++}: type
11572 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
11573 also use the @value{GDBN} command-line word completion facilities to list the
11574 available choices, or to finish the type list for you.
11575 @xref{Completion,, Command Completion}, for details on how to do this.
11578 @node Decimal Floating Point
11579 @subsubsection Decimal Floating Point format
11580 @cindex decimal floating point format
11582 @value{GDBN} can examine, set and perform computations with numbers in
11583 decimal floating point format, which in the C language correspond to the
11584 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
11585 specified by the extension to support decimal floating-point arithmetic.
11587 There are two encodings in use, depending on the architecture: BID (Binary
11588 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
11589 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
11592 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
11593 to manipulate decimal floating point numbers, it is not possible to convert
11594 (using a cast, for example) integers wider than 32-bit to decimal float.
11596 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
11597 point computations, error checking in decimal float operations ignores
11598 underflow, overflow and divide by zero exceptions.
11600 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
11601 to inspect @code{_Decimal128} values stored in floating point registers.
11602 See @ref{PowerPC,,PowerPC} for more details.
11605 @subsection Objective-C
11607 @cindex Objective-C
11608 This section provides information about some commands and command
11609 options that are useful for debugging Objective-C code. See also
11610 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
11611 few more commands specific to Objective-C support.
11614 * Method Names in Commands::
11615 * The Print Command with Objective-C::
11618 @node Method Names in Commands
11619 @subsubsection Method Names in Commands
11621 The following commands have been extended to accept Objective-C method
11622 names as line specifications:
11624 @kindex clear@r{, and Objective-C}
11625 @kindex break@r{, and Objective-C}
11626 @kindex info line@r{, and Objective-C}
11627 @kindex jump@r{, and Objective-C}
11628 @kindex list@r{, and Objective-C}
11632 @item @code{info line}
11637 A fully qualified Objective-C method name is specified as
11640 -[@var{Class} @var{methodName}]
11643 where the minus sign is used to indicate an instance method and a
11644 plus sign (not shown) is used to indicate a class method. The class
11645 name @var{Class} and method name @var{methodName} are enclosed in
11646 brackets, similar to the way messages are specified in Objective-C
11647 source code. For example, to set a breakpoint at the @code{create}
11648 instance method of class @code{Fruit} in the program currently being
11652 break -[Fruit create]
11655 To list ten program lines around the @code{initialize} class method,
11659 list +[NSText initialize]
11662 In the current version of @value{GDBN}, the plus or minus sign is
11663 required. In future versions of @value{GDBN}, the plus or minus
11664 sign will be optional, but you can use it to narrow the search. It
11665 is also possible to specify just a method name:
11671 You must specify the complete method name, including any colons. If
11672 your program's source files contain more than one @code{create} method,
11673 you'll be presented with a numbered list of classes that implement that
11674 method. Indicate your choice by number, or type @samp{0} to exit if
11677 As another example, to clear a breakpoint established at the
11678 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
11681 clear -[NSWindow makeKeyAndOrderFront:]
11684 @node The Print Command with Objective-C
11685 @subsubsection The Print Command With Objective-C
11686 @cindex Objective-C, print objects
11687 @kindex print-object
11688 @kindex po @r{(@code{print-object})}
11690 The print command has also been extended to accept methods. For example:
11693 print -[@var{object} hash]
11696 @cindex print an Objective-C object description
11697 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
11699 will tell @value{GDBN} to send the @code{hash} message to @var{object}
11700 and print the result. Also, an additional command has been added,
11701 @code{print-object} or @code{po} for short, which is meant to print
11702 the description of an object. However, this command may only work
11703 with certain Objective-C libraries that have a particular hook
11704 function, @code{_NSPrintForDebugger}, defined.
11707 @subsection Fortran
11708 @cindex Fortran-specific support in @value{GDBN}
11710 @value{GDBN} can be used to debug programs written in Fortran, but it
11711 currently supports only the features of Fortran 77 language.
11713 @cindex trailing underscore, in Fortran symbols
11714 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
11715 among them) append an underscore to the names of variables and
11716 functions. When you debug programs compiled by those compilers, you
11717 will need to refer to variables and functions with a trailing
11721 * Fortran Operators:: Fortran operators and expressions
11722 * Fortran Defaults:: Default settings for Fortran
11723 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
11726 @node Fortran Operators
11727 @subsubsection Fortran Operators and Expressions
11729 @cindex Fortran operators and expressions
11731 Operators must be defined on values of specific types. For instance,
11732 @code{+} is defined on numbers, but not on characters or other non-
11733 arithmetic types. Operators are often defined on groups of types.
11737 The exponentiation operator. It raises the first operand to the power
11741 The range operator. Normally used in the form of array(low:high) to
11742 represent a section of array.
11745 The access component operator. Normally used to access elements in derived
11746 types. Also suitable for unions. As unions aren't part of regular Fortran,
11747 this can only happen when accessing a register that uses a gdbarch-defined
11751 @node Fortran Defaults
11752 @subsubsection Fortran Defaults
11754 @cindex Fortran Defaults
11756 Fortran symbols are usually case-insensitive, so @value{GDBN} by
11757 default uses case-insensitive matches for Fortran symbols. You can
11758 change that with the @samp{set case-insensitive} command, see
11759 @ref{Symbols}, for the details.
11761 @node Special Fortran Commands
11762 @subsubsection Special Fortran Commands
11764 @cindex Special Fortran commands
11766 @value{GDBN} has some commands to support Fortran-specific features,
11767 such as displaying common blocks.
11770 @cindex @code{COMMON} blocks, Fortran
11771 @kindex info common
11772 @item info common @r{[}@var{common-name}@r{]}
11773 This command prints the values contained in the Fortran @code{COMMON}
11774 block whose name is @var{common-name}. With no argument, the names of
11775 all @code{COMMON} blocks visible at the current program location are
11782 @cindex Pascal support in @value{GDBN}, limitations
11783 Debugging Pascal programs which use sets, subranges, file variables, or
11784 nested functions does not currently work. @value{GDBN} does not support
11785 entering expressions, printing values, or similar features using Pascal
11788 The Pascal-specific command @code{set print pascal_static-members}
11789 controls whether static members of Pascal objects are displayed.
11790 @xref{Print Settings, pascal_static-members}.
11793 @subsection Modula-2
11795 @cindex Modula-2, @value{GDBN} support
11797 The extensions made to @value{GDBN} to support Modula-2 only support
11798 output from the @sc{gnu} Modula-2 compiler (which is currently being
11799 developed). Other Modula-2 compilers are not currently supported, and
11800 attempting to debug executables produced by them is most likely
11801 to give an error as @value{GDBN} reads in the executable's symbol
11804 @cindex expressions in Modula-2
11806 * M2 Operators:: Built-in operators
11807 * Built-In Func/Proc:: Built-in functions and procedures
11808 * M2 Constants:: Modula-2 constants
11809 * M2 Types:: Modula-2 types
11810 * M2 Defaults:: Default settings for Modula-2
11811 * Deviations:: Deviations from standard Modula-2
11812 * M2 Checks:: Modula-2 type and range checks
11813 * M2 Scope:: The scope operators @code{::} and @code{.}
11814 * GDB/M2:: @value{GDBN} and Modula-2
11818 @subsubsection Operators
11819 @cindex Modula-2 operators
11821 Operators must be defined on values of specific types. For instance,
11822 @code{+} is defined on numbers, but not on structures. Operators are
11823 often defined on groups of types. For the purposes of Modula-2, the
11824 following definitions hold:
11829 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
11833 @emph{Character types} consist of @code{CHAR} and its subranges.
11836 @emph{Floating-point types} consist of @code{REAL}.
11839 @emph{Pointer types} consist of anything declared as @code{POINTER TO
11843 @emph{Scalar types} consist of all of the above.
11846 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
11849 @emph{Boolean types} consist of @code{BOOLEAN}.
11853 The following operators are supported, and appear in order of
11854 increasing precedence:
11858 Function argument or array index separator.
11861 Assignment. The value of @var{var} @code{:=} @var{value} is
11865 Less than, greater than on integral, floating-point, or enumerated
11869 Less than or equal to, greater than or equal to
11870 on integral, floating-point and enumerated types, or set inclusion on
11871 set types. Same precedence as @code{<}.
11873 @item =@r{, }<>@r{, }#
11874 Equality and two ways of expressing inequality, valid on scalar types.
11875 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
11876 available for inequality, since @code{#} conflicts with the script
11880 Set membership. Defined on set types and the types of their members.
11881 Same precedence as @code{<}.
11884 Boolean disjunction. Defined on boolean types.
11887 Boolean conjunction. Defined on boolean types.
11890 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11893 Addition and subtraction on integral and floating-point types, or union
11894 and difference on set types.
11897 Multiplication on integral and floating-point types, or set intersection
11901 Division on floating-point types, or symmetric set difference on set
11902 types. Same precedence as @code{*}.
11905 Integer division and remainder. Defined on integral types. Same
11906 precedence as @code{*}.
11909 Negative. Defined on @code{INTEGER} and @code{REAL} data.
11912 Pointer dereferencing. Defined on pointer types.
11915 Boolean negation. Defined on boolean types. Same precedence as
11919 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
11920 precedence as @code{^}.
11923 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
11926 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
11930 @value{GDBN} and Modula-2 scope operators.
11934 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
11935 treats the use of the operator @code{IN}, or the use of operators
11936 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
11937 @code{<=}, and @code{>=} on sets as an error.
11941 @node Built-In Func/Proc
11942 @subsubsection Built-in Functions and Procedures
11943 @cindex Modula-2 built-ins
11945 Modula-2 also makes available several built-in procedures and functions.
11946 In describing these, the following metavariables are used:
11951 represents an @code{ARRAY} variable.
11954 represents a @code{CHAR} constant or variable.
11957 represents a variable or constant of integral type.
11960 represents an identifier that belongs to a set. Generally used in the
11961 same function with the metavariable @var{s}. The type of @var{s} should
11962 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
11965 represents a variable or constant of integral or floating-point type.
11968 represents a variable or constant of floating-point type.
11974 represents a variable.
11977 represents a variable or constant of one of many types. See the
11978 explanation of the function for details.
11981 All Modula-2 built-in procedures also return a result, described below.
11985 Returns the absolute value of @var{n}.
11988 If @var{c} is a lower case letter, it returns its upper case
11989 equivalent, otherwise it returns its argument.
11992 Returns the character whose ordinal value is @var{i}.
11995 Decrements the value in the variable @var{v} by one. Returns the new value.
11997 @item DEC(@var{v},@var{i})
11998 Decrements the value in the variable @var{v} by @var{i}. Returns the
12001 @item EXCL(@var{m},@var{s})
12002 Removes the element @var{m} from the set @var{s}. Returns the new
12005 @item FLOAT(@var{i})
12006 Returns the floating point equivalent of the integer @var{i}.
12008 @item HIGH(@var{a})
12009 Returns the index of the last member of @var{a}.
12012 Increments the value in the variable @var{v} by one. Returns the new value.
12014 @item INC(@var{v},@var{i})
12015 Increments the value in the variable @var{v} by @var{i}. Returns the
12018 @item INCL(@var{m},@var{s})
12019 Adds the element @var{m} to the set @var{s} if it is not already
12020 there. Returns the new set.
12023 Returns the maximum value of the type @var{t}.
12026 Returns the minimum value of the type @var{t}.
12029 Returns boolean TRUE if @var{i} is an odd number.
12032 Returns the ordinal value of its argument. For example, the ordinal
12033 value of a character is its @sc{ascii} value (on machines supporting the
12034 @sc{ascii} character set). @var{x} must be of an ordered type, which include
12035 integral, character and enumerated types.
12037 @item SIZE(@var{x})
12038 Returns the size of its argument. @var{x} can be a variable or a type.
12040 @item TRUNC(@var{r})
12041 Returns the integral part of @var{r}.
12043 @item TSIZE(@var{x})
12044 Returns the size of its argument. @var{x} can be a variable or a type.
12046 @item VAL(@var{t},@var{i})
12047 Returns the member of the type @var{t} whose ordinal value is @var{i}.
12051 @emph{Warning:} Sets and their operations are not yet supported, so
12052 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
12056 @cindex Modula-2 constants
12058 @subsubsection Constants
12060 @value{GDBN} allows you to express the constants of Modula-2 in the following
12066 Integer constants are simply a sequence of digits. When used in an
12067 expression, a constant is interpreted to be type-compatible with the
12068 rest of the expression. Hexadecimal integers are specified by a
12069 trailing @samp{H}, and octal integers by a trailing @samp{B}.
12072 Floating point constants appear as a sequence of digits, followed by a
12073 decimal point and another sequence of digits. An optional exponent can
12074 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
12075 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
12076 digits of the floating point constant must be valid decimal (base 10)
12080 Character constants consist of a single character enclosed by a pair of
12081 like quotes, either single (@code{'}) or double (@code{"}). They may
12082 also be expressed by their ordinal value (their @sc{ascii} value, usually)
12083 followed by a @samp{C}.
12086 String constants consist of a sequence of characters enclosed by a
12087 pair of like quotes, either single (@code{'}) or double (@code{"}).
12088 Escape sequences in the style of C are also allowed. @xref{C
12089 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
12093 Enumerated constants consist of an enumerated identifier.
12096 Boolean constants consist of the identifiers @code{TRUE} and
12100 Pointer constants consist of integral values only.
12103 Set constants are not yet supported.
12107 @subsubsection Modula-2 Types
12108 @cindex Modula-2 types
12110 Currently @value{GDBN} can print the following data types in Modula-2
12111 syntax: array types, record types, set types, pointer types, procedure
12112 types, enumerated types, subrange types and base types. You can also
12113 print the contents of variables declared using these type.
12114 This section gives a number of simple source code examples together with
12115 sample @value{GDBN} sessions.
12117 The first example contains the following section of code:
12126 and you can request @value{GDBN} to interrogate the type and value of
12127 @code{r} and @code{s}.
12130 (@value{GDBP}) print s
12132 (@value{GDBP}) ptype s
12134 (@value{GDBP}) print r
12136 (@value{GDBP}) ptype r
12141 Likewise if your source code declares @code{s} as:
12145 s: SET ['A'..'Z'] ;
12149 then you may query the type of @code{s} by:
12152 (@value{GDBP}) ptype s
12153 type = SET ['A'..'Z']
12157 Note that at present you cannot interactively manipulate set
12158 expressions using the debugger.
12160 The following example shows how you might declare an array in Modula-2
12161 and how you can interact with @value{GDBN} to print its type and contents:
12165 s: ARRAY [-10..10] OF CHAR ;
12169 (@value{GDBP}) ptype s
12170 ARRAY [-10..10] OF CHAR
12173 Note that the array handling is not yet complete and although the type
12174 is printed correctly, expression handling still assumes that all
12175 arrays have a lower bound of zero and not @code{-10} as in the example
12178 Here are some more type related Modula-2 examples:
12182 colour = (blue, red, yellow, green) ;
12183 t = [blue..yellow] ;
12191 The @value{GDBN} interaction shows how you can query the data type
12192 and value of a variable.
12195 (@value{GDBP}) print s
12197 (@value{GDBP}) ptype t
12198 type = [blue..yellow]
12202 In this example a Modula-2 array is declared and its contents
12203 displayed. Observe that the contents are written in the same way as
12204 their @code{C} counterparts.
12208 s: ARRAY [1..5] OF CARDINAL ;
12214 (@value{GDBP}) print s
12215 $1 = @{1, 0, 0, 0, 0@}
12216 (@value{GDBP}) ptype s
12217 type = ARRAY [1..5] OF CARDINAL
12220 The Modula-2 language interface to @value{GDBN} also understands
12221 pointer types as shown in this example:
12225 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
12232 and you can request that @value{GDBN} describes the type of @code{s}.
12235 (@value{GDBP}) ptype s
12236 type = POINTER TO ARRAY [1..5] OF CARDINAL
12239 @value{GDBN} handles compound types as we can see in this example.
12240 Here we combine array types, record types, pointer types and subrange
12251 myarray = ARRAY myrange OF CARDINAL ;
12252 myrange = [-2..2] ;
12254 s: POINTER TO ARRAY myrange OF foo ;
12258 and you can ask @value{GDBN} to describe the type of @code{s} as shown
12262 (@value{GDBP}) ptype s
12263 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
12266 f3 : ARRAY [-2..2] OF CARDINAL;
12271 @subsubsection Modula-2 Defaults
12272 @cindex Modula-2 defaults
12274 If type and range checking are set automatically by @value{GDBN}, they
12275 both default to @code{on} whenever the working language changes to
12276 Modula-2. This happens regardless of whether you or @value{GDBN}
12277 selected the working language.
12279 If you allow @value{GDBN} to set the language automatically, then entering
12280 code compiled from a file whose name ends with @file{.mod} sets the
12281 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
12282 Infer the Source Language}, for further details.
12285 @subsubsection Deviations from Standard Modula-2
12286 @cindex Modula-2, deviations from
12288 A few changes have been made to make Modula-2 programs easier to debug.
12289 This is done primarily via loosening its type strictness:
12293 Unlike in standard Modula-2, pointer constants can be formed by
12294 integers. This allows you to modify pointer variables during
12295 debugging. (In standard Modula-2, the actual address contained in a
12296 pointer variable is hidden from you; it can only be modified
12297 through direct assignment to another pointer variable or expression that
12298 returned a pointer.)
12301 C escape sequences can be used in strings and characters to represent
12302 non-printable characters. @value{GDBN} prints out strings with these
12303 escape sequences embedded. Single non-printable characters are
12304 printed using the @samp{CHR(@var{nnn})} format.
12307 The assignment operator (@code{:=}) returns the value of its right-hand
12311 All built-in procedures both modify @emph{and} return their argument.
12315 @subsubsection Modula-2 Type and Range Checks
12316 @cindex Modula-2 checks
12319 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
12322 @c FIXME remove warning when type/range checks added
12324 @value{GDBN} considers two Modula-2 variables type equivalent if:
12328 They are of types that have been declared equivalent via a @code{TYPE
12329 @var{t1} = @var{t2}} statement
12332 They have been declared on the same line. (Note: This is true of the
12333 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
12336 As long as type checking is enabled, any attempt to combine variables
12337 whose types are not equivalent is an error.
12339 Range checking is done on all mathematical operations, assignment, array
12340 index bounds, and all built-in functions and procedures.
12343 @subsubsection The Scope Operators @code{::} and @code{.}
12345 @cindex @code{.}, Modula-2 scope operator
12346 @cindex colon, doubled as scope operator
12348 @vindex colon-colon@r{, in Modula-2}
12349 @c Info cannot handle :: but TeX can.
12352 @vindex ::@r{, in Modula-2}
12355 There are a few subtle differences between the Modula-2 scope operator
12356 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
12361 @var{module} . @var{id}
12362 @var{scope} :: @var{id}
12366 where @var{scope} is the name of a module or a procedure,
12367 @var{module} the name of a module, and @var{id} is any declared
12368 identifier within your program, except another module.
12370 Using the @code{::} operator makes @value{GDBN} search the scope
12371 specified by @var{scope} for the identifier @var{id}. If it is not
12372 found in the specified scope, then @value{GDBN} searches all scopes
12373 enclosing the one specified by @var{scope}.
12375 Using the @code{.} operator makes @value{GDBN} search the current scope for
12376 the identifier specified by @var{id} that was imported from the
12377 definition module specified by @var{module}. With this operator, it is
12378 an error if the identifier @var{id} was not imported from definition
12379 module @var{module}, or if @var{id} is not an identifier in
12383 @subsubsection @value{GDBN} and Modula-2
12385 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
12386 Five subcommands of @code{set print} and @code{show print} apply
12387 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
12388 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
12389 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
12390 analogue in Modula-2.
12392 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
12393 with any language, is not useful with Modula-2. Its
12394 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
12395 created in Modula-2 as they can in C or C@t{++}. However, because an
12396 address can be specified by an integral constant, the construct
12397 @samp{@{@var{type}@}@var{adrexp}} is still useful.
12399 @cindex @code{#} in Modula-2
12400 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
12401 interpreted as the beginning of a comment. Use @code{<>} instead.
12407 The extensions made to @value{GDBN} for Ada only support
12408 output from the @sc{gnu} Ada (GNAT) compiler.
12409 Other Ada compilers are not currently supported, and
12410 attempting to debug executables produced by them is most likely
12414 @cindex expressions in Ada
12416 * Ada Mode Intro:: General remarks on the Ada syntax
12417 and semantics supported by Ada mode
12419 * Omissions from Ada:: Restrictions on the Ada expression syntax.
12420 * Additions to Ada:: Extensions of the Ada expression syntax.
12421 * Stopping Before Main Program:: Debugging the program during elaboration.
12422 * Ada Tasks:: Listing and setting breakpoints in tasks.
12423 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
12424 * Ada Glitches:: Known peculiarities of Ada mode.
12427 @node Ada Mode Intro
12428 @subsubsection Introduction
12429 @cindex Ada mode, general
12431 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
12432 syntax, with some extensions.
12433 The philosophy behind the design of this subset is
12437 That @value{GDBN} should provide basic literals and access to operations for
12438 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
12439 leaving more sophisticated computations to subprograms written into the
12440 program (which therefore may be called from @value{GDBN}).
12443 That type safety and strict adherence to Ada language restrictions
12444 are not particularly important to the @value{GDBN} user.
12447 That brevity is important to the @value{GDBN} user.
12450 Thus, for brevity, the debugger acts as if all names declared in
12451 user-written packages are directly visible, even if they are not visible
12452 according to Ada rules, thus making it unnecessary to fully qualify most
12453 names with their packages, regardless of context. Where this causes
12454 ambiguity, @value{GDBN} asks the user's intent.
12456 The debugger will start in Ada mode if it detects an Ada main program.
12457 As for other languages, it will enter Ada mode when stopped in a program that
12458 was translated from an Ada source file.
12460 While in Ada mode, you may use `@t{--}' for comments. This is useful
12461 mostly for documenting command files. The standard @value{GDBN} comment
12462 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
12463 middle (to allow based literals).
12465 The debugger supports limited overloading. Given a subprogram call in which
12466 the function symbol has multiple definitions, it will use the number of
12467 actual parameters and some information about their types to attempt to narrow
12468 the set of definitions. It also makes very limited use of context, preferring
12469 procedures to functions in the context of the @code{call} command, and
12470 functions to procedures elsewhere.
12472 @node Omissions from Ada
12473 @subsubsection Omissions from Ada
12474 @cindex Ada, omissions from
12476 Here are the notable omissions from the subset:
12480 Only a subset of the attributes are supported:
12484 @t{'First}, @t{'Last}, and @t{'Length}
12485 on array objects (not on types and subtypes).
12488 @t{'Min} and @t{'Max}.
12491 @t{'Pos} and @t{'Val}.
12497 @t{'Range} on array objects (not subtypes), but only as the right
12498 operand of the membership (@code{in}) operator.
12501 @t{'Access}, @t{'Unchecked_Access}, and
12502 @t{'Unrestricted_Access} (a GNAT extension).
12510 @code{Characters.Latin_1} are not available and
12511 concatenation is not implemented. Thus, escape characters in strings are
12512 not currently available.
12515 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
12516 equality of representations. They will generally work correctly
12517 for strings and arrays whose elements have integer or enumeration types.
12518 They may not work correctly for arrays whose element
12519 types have user-defined equality, for arrays of real values
12520 (in particular, IEEE-conformant floating point, because of negative
12521 zeroes and NaNs), and for arrays whose elements contain unused bits with
12522 indeterminate values.
12525 The other component-by-component array operations (@code{and}, @code{or},
12526 @code{xor}, @code{not}, and relational tests other than equality)
12527 are not implemented.
12530 @cindex array aggregates (Ada)
12531 @cindex record aggregates (Ada)
12532 @cindex aggregates (Ada)
12533 There is limited support for array and record aggregates. They are
12534 permitted only on the right sides of assignments, as in these examples:
12537 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
12538 (@value{GDBP}) set An_Array := (1, others => 0)
12539 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
12540 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
12541 (@value{GDBP}) set A_Record := (1, "Peter", True);
12542 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
12546 discriminant's value by assigning an aggregate has an
12547 undefined effect if that discriminant is used within the record.
12548 However, you can first modify discriminants by directly assigning to
12549 them (which normally would not be allowed in Ada), and then performing an
12550 aggregate assignment. For example, given a variable @code{A_Rec}
12551 declared to have a type such as:
12554 type Rec (Len : Small_Integer := 0) is record
12556 Vals : IntArray (1 .. Len);
12560 you can assign a value with a different size of @code{Vals} with two
12564 (@value{GDBP}) set A_Rec.Len := 4
12565 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
12568 As this example also illustrates, @value{GDBN} is very loose about the usual
12569 rules concerning aggregates. You may leave out some of the
12570 components of an array or record aggregate (such as the @code{Len}
12571 component in the assignment to @code{A_Rec} above); they will retain their
12572 original values upon assignment. You may freely use dynamic values as
12573 indices in component associations. You may even use overlapping or
12574 redundant component associations, although which component values are
12575 assigned in such cases is not defined.
12578 Calls to dispatching subprograms are not implemented.
12581 The overloading algorithm is much more limited (i.e., less selective)
12582 than that of real Ada. It makes only limited use of the context in
12583 which a subexpression appears to resolve its meaning, and it is much
12584 looser in its rules for allowing type matches. As a result, some
12585 function calls will be ambiguous, and the user will be asked to choose
12586 the proper resolution.
12589 The @code{new} operator is not implemented.
12592 Entry calls are not implemented.
12595 Aside from printing, arithmetic operations on the native VAX floating-point
12596 formats are not supported.
12599 It is not possible to slice a packed array.
12602 The names @code{True} and @code{False}, when not part of a qualified name,
12603 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
12605 Should your program
12606 redefine these names in a package or procedure (at best a dubious practice),
12607 you will have to use fully qualified names to access their new definitions.
12610 @node Additions to Ada
12611 @subsubsection Additions to Ada
12612 @cindex Ada, deviations from
12614 As it does for other languages, @value{GDBN} makes certain generic
12615 extensions to Ada (@pxref{Expressions}):
12619 If the expression @var{E} is a variable residing in memory (typically
12620 a local variable or array element) and @var{N} is a positive integer,
12621 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
12622 @var{N}-1 adjacent variables following it in memory as an array. In
12623 Ada, this operator is generally not necessary, since its prime use is
12624 in displaying parts of an array, and slicing will usually do this in
12625 Ada. However, there are occasional uses when debugging programs in
12626 which certain debugging information has been optimized away.
12629 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
12630 appears in function or file @var{B}.'' When @var{B} is a file name,
12631 you must typically surround it in single quotes.
12634 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
12635 @var{type} that appears at address @var{addr}.''
12638 A name starting with @samp{$} is a convenience variable
12639 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
12642 In addition, @value{GDBN} provides a few other shortcuts and outright
12643 additions specific to Ada:
12647 The assignment statement is allowed as an expression, returning
12648 its right-hand operand as its value. Thus, you may enter
12651 (@value{GDBP}) set x := y + 3
12652 (@value{GDBP}) print A(tmp := y + 1)
12656 The semicolon is allowed as an ``operator,'' returning as its value
12657 the value of its right-hand operand.
12658 This allows, for example,
12659 complex conditional breaks:
12662 (@value{GDBP}) break f
12663 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
12667 Rather than use catenation and symbolic character names to introduce special
12668 characters into strings, one may instead use a special bracket notation,
12669 which is also used to print strings. A sequence of characters of the form
12670 @samp{["@var{XX}"]} within a string or character literal denotes the
12671 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
12672 sequence of characters @samp{["""]} also denotes a single quotation mark
12673 in strings. For example,
12675 "One line.["0a"]Next line.["0a"]"
12678 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
12682 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
12683 @t{'Max} is optional (and is ignored in any case). For example, it is valid
12687 (@value{GDBP}) print 'max(x, y)
12691 When printing arrays, @value{GDBN} uses positional notation when the
12692 array has a lower bound of 1, and uses a modified named notation otherwise.
12693 For example, a one-dimensional array of three integers with a lower bound
12694 of 3 might print as
12701 That is, in contrast to valid Ada, only the first component has a @code{=>}
12705 You may abbreviate attributes in expressions with any unique,
12706 multi-character subsequence of
12707 their names (an exact match gets preference).
12708 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
12709 in place of @t{a'length}.
12712 @cindex quoting Ada internal identifiers
12713 Since Ada is case-insensitive, the debugger normally maps identifiers you type
12714 to lower case. The GNAT compiler uses upper-case characters for
12715 some of its internal identifiers, which are normally of no interest to users.
12716 For the rare occasions when you actually have to look at them,
12717 enclose them in angle brackets to avoid the lower-case mapping.
12720 (@value{GDBP}) print <JMPBUF_SAVE>[0]
12724 Printing an object of class-wide type or dereferencing an
12725 access-to-class-wide value will display all the components of the object's
12726 specific type (as indicated by its run-time tag). Likewise, component
12727 selection on such a value will operate on the specific type of the
12732 @node Stopping Before Main Program
12733 @subsubsection Stopping at the Very Beginning
12735 @cindex breakpointing Ada elaboration code
12736 It is sometimes necessary to debug the program during elaboration, and
12737 before reaching the main procedure.
12738 As defined in the Ada Reference
12739 Manual, the elaboration code is invoked from a procedure called
12740 @code{adainit}. To run your program up to the beginning of
12741 elaboration, simply use the following two commands:
12742 @code{tbreak adainit} and @code{run}.
12745 @subsubsection Extensions for Ada Tasks
12746 @cindex Ada, tasking
12748 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
12749 @value{GDBN} provides the following task-related commands:
12754 This command shows a list of current Ada tasks, as in the following example:
12761 (@value{GDBP}) info tasks
12762 ID TID P-ID Pri State Name
12763 1 8088000 0 15 Child Activation Wait main_task
12764 2 80a4000 1 15 Accept Statement b
12765 3 809a800 1 15 Child Activation Wait a
12766 * 4 80ae800 3 15 Runnable c
12771 In this listing, the asterisk before the last task indicates it to be the
12772 task currently being inspected.
12776 Represents @value{GDBN}'s internal task number.
12782 The parent's task ID (@value{GDBN}'s internal task number).
12785 The base priority of the task.
12788 Current state of the task.
12792 The task has been created but has not been activated. It cannot be
12796 The task is not blocked for any reason known to Ada. (It may be waiting
12797 for a mutex, though.) It is conceptually "executing" in normal mode.
12800 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
12801 that were waiting on terminate alternatives have been awakened and have
12802 terminated themselves.
12804 @item Child Activation Wait
12805 The task is waiting for created tasks to complete activation.
12807 @item Accept Statement
12808 The task is waiting on an accept or selective wait statement.
12810 @item Waiting on entry call
12811 The task is waiting on an entry call.
12813 @item Async Select Wait
12814 The task is waiting to start the abortable part of an asynchronous
12818 The task is waiting on a select statement with only a delay
12821 @item Child Termination Wait
12822 The task is sleeping having completed a master within itself, and is
12823 waiting for the tasks dependent on that master to become terminated or
12824 waiting on a terminate Phase.
12826 @item Wait Child in Term Alt
12827 The task is sleeping waiting for tasks on terminate alternatives to
12828 finish terminating.
12830 @item Accepting RV with @var{taskno}
12831 The task is accepting a rendez-vous with the task @var{taskno}.
12835 Name of the task in the program.
12839 @kindex info task @var{taskno}
12840 @item info task @var{taskno}
12841 This command shows detailled informations on the specified task, as in
12842 the following example:
12847 (@value{GDBP}) info tasks
12848 ID TID P-ID Pri State Name
12849 1 8077880 0 15 Child Activation Wait main_task
12850 * 2 807c468 1 15 Runnable task_1
12851 (@value{GDBP}) info task 2
12852 Ada Task: 0x807c468
12855 Parent: 1 (main_task)
12861 @kindex task@r{ (Ada)}
12862 @cindex current Ada task ID
12863 This command prints the ID of the current task.
12869 (@value{GDBP}) info tasks
12870 ID TID P-ID Pri State Name
12871 1 8077870 0 15 Child Activation Wait main_task
12872 * 2 807c458 1 15 Runnable t
12873 (@value{GDBP}) task
12874 [Current task is 2]
12877 @item task @var{taskno}
12878 @cindex Ada task switching
12879 This command is like the @code{thread @var{threadno}}
12880 command (@pxref{Threads}). It switches the context of debugging
12881 from the current task to the given task.
12887 (@value{GDBP}) info tasks
12888 ID TID P-ID Pri State Name
12889 1 8077870 0 15 Child Activation Wait main_task
12890 * 2 807c458 1 15 Runnable t
12891 (@value{GDBP}) task 1
12892 [Switching to task 1]
12893 #0 0x8067726 in pthread_cond_wait ()
12895 #0 0x8067726 in pthread_cond_wait ()
12896 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
12897 #2 0x805cb63 in system.task_primitives.operations.sleep ()
12898 #3 0x806153e in system.tasking.stages.activate_tasks ()
12899 #4 0x804aacc in un () at un.adb:5
12902 @item break @var{linespec} task @var{taskno}
12903 @itemx break @var{linespec} task @var{taskno} if @dots{}
12904 @cindex breakpoints and tasks, in Ada
12905 @cindex task breakpoints, in Ada
12906 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
12907 These commands are like the @code{break @dots{} thread @dots{}}
12908 command (@pxref{Thread Stops}).
12909 @var{linespec} specifies source lines, as described
12910 in @ref{Specify Location}.
12912 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
12913 to specify that you only want @value{GDBN} to stop the program when a
12914 particular Ada task reaches this breakpoint. @var{taskno} is one of the
12915 numeric task identifiers assigned by @value{GDBN}, shown in the first
12916 column of the @samp{info tasks} display.
12918 If you do not specify @samp{task @var{taskno}} when you set a
12919 breakpoint, the breakpoint applies to @emph{all} tasks of your
12922 You can use the @code{task} qualifier on conditional breakpoints as
12923 well; in this case, place @samp{task @var{taskno}} before the
12924 breakpoint condition (before the @code{if}).
12932 (@value{GDBP}) info tasks
12933 ID TID P-ID Pri State Name
12934 1 140022020 0 15 Child Activation Wait main_task
12935 2 140045060 1 15 Accept/Select Wait t2
12936 3 140044840 1 15 Runnable t1
12937 * 4 140056040 1 15 Runnable t3
12938 (@value{GDBP}) b 15 task 2
12939 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
12940 (@value{GDBP}) cont
12945 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
12947 (@value{GDBP}) info tasks
12948 ID TID P-ID Pri State Name
12949 1 140022020 0 15 Child Activation Wait main_task
12950 * 2 140045060 1 15 Runnable t2
12951 3 140044840 1 15 Runnable t1
12952 4 140056040 1 15 Delay Sleep t3
12956 @node Ada Tasks and Core Files
12957 @subsubsection Tasking Support when Debugging Core Files
12958 @cindex Ada tasking and core file debugging
12960 When inspecting a core file, as opposed to debugging a live program,
12961 tasking support may be limited or even unavailable, depending on
12962 the platform being used.
12963 For instance, on x86-linux, the list of tasks is available, but task
12964 switching is not supported. On Tru64, however, task switching will work
12967 On certain platforms, including Tru64, the debugger needs to perform some
12968 memory writes in order to provide Ada tasking support. When inspecting
12969 a core file, this means that the core file must be opened with read-write
12970 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
12971 Under these circumstances, you should make a backup copy of the core
12972 file before inspecting it with @value{GDBN}.
12975 @subsubsection Known Peculiarities of Ada Mode
12976 @cindex Ada, problems
12978 Besides the omissions listed previously (@pxref{Omissions from Ada}),
12979 we know of several problems with and limitations of Ada mode in
12981 some of which will be fixed with planned future releases of the debugger
12982 and the GNU Ada compiler.
12986 Currently, the debugger
12987 has insufficient information to determine whether certain pointers represent
12988 pointers to objects or the objects themselves.
12989 Thus, the user may have to tack an extra @code{.all} after an expression
12990 to get it printed properly.
12993 Static constants that the compiler chooses not to materialize as objects in
12994 storage are invisible to the debugger.
12997 Named parameter associations in function argument lists are ignored (the
12998 argument lists are treated as positional).
13001 Many useful library packages are currently invisible to the debugger.
13004 Fixed-point arithmetic, conversions, input, and output is carried out using
13005 floating-point arithmetic, and may give results that only approximate those on
13009 The GNAT compiler never generates the prefix @code{Standard} for any of
13010 the standard symbols defined by the Ada language. @value{GDBN} knows about
13011 this: it will strip the prefix from names when you use it, and will never
13012 look for a name you have so qualified among local symbols, nor match against
13013 symbols in other packages or subprograms. If you have
13014 defined entities anywhere in your program other than parameters and
13015 local variables whose simple names match names in @code{Standard},
13016 GNAT's lack of qualification here can cause confusion. When this happens,
13017 you can usually resolve the confusion
13018 by qualifying the problematic names with package
13019 @code{Standard} explicitly.
13022 Older versions of the compiler sometimes generate erroneous debugging
13023 information, resulting in the debugger incorrectly printing the value
13024 of affected entities. In some cases, the debugger is able to work
13025 around an issue automatically. In other cases, the debugger is able
13026 to work around the issue, but the work-around has to be specifically
13029 @kindex set ada trust-PAD-over-XVS
13030 @kindex show ada trust-PAD-over-XVS
13033 @item set ada trust-PAD-over-XVS on
13034 Configure GDB to strictly follow the GNAT encoding when computing the
13035 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
13036 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
13037 a complete description of the encoding used by the GNAT compiler).
13038 This is the default.
13040 @item set ada trust-PAD-over-XVS off
13041 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
13042 sometimes prints the wrong value for certain entities, changing @code{ada
13043 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
13044 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
13045 @code{off}, but this incurs a slight performance penalty, so it is
13046 recommended to leave this setting to @code{on} unless necessary.
13050 @node Unsupported Languages
13051 @section Unsupported Languages
13053 @cindex unsupported languages
13054 @cindex minimal language
13055 In addition to the other fully-supported programming languages,
13056 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
13057 It does not represent a real programming language, but provides a set
13058 of capabilities close to what the C or assembly languages provide.
13059 This should allow most simple operations to be performed while debugging
13060 an application that uses a language currently not supported by @value{GDBN}.
13062 If the language is set to @code{auto}, @value{GDBN} will automatically
13063 select this language if the current frame corresponds to an unsupported
13067 @chapter Examining the Symbol Table
13069 The commands described in this chapter allow you to inquire about the
13070 symbols (names of variables, functions and types) defined in your
13071 program. This information is inherent in the text of your program and
13072 does not change as your program executes. @value{GDBN} finds it in your
13073 program's symbol table, in the file indicated when you started @value{GDBN}
13074 (@pxref{File Options, ,Choosing Files}), or by one of the
13075 file-management commands (@pxref{Files, ,Commands to Specify Files}).
13077 @cindex symbol names
13078 @cindex names of symbols
13079 @cindex quoting names
13080 Occasionally, you may need to refer to symbols that contain unusual
13081 characters, which @value{GDBN} ordinarily treats as word delimiters. The
13082 most frequent case is in referring to static variables in other
13083 source files (@pxref{Variables,,Program Variables}). File names
13084 are recorded in object files as debugging symbols, but @value{GDBN} would
13085 ordinarily parse a typical file name, like @file{foo.c}, as the three words
13086 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
13087 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
13094 looks up the value of @code{x} in the scope of the file @file{foo.c}.
13097 @cindex case-insensitive symbol names
13098 @cindex case sensitivity in symbol names
13099 @kindex set case-sensitive
13100 @item set case-sensitive on
13101 @itemx set case-sensitive off
13102 @itemx set case-sensitive auto
13103 Normally, when @value{GDBN} looks up symbols, it matches their names
13104 with case sensitivity determined by the current source language.
13105 Occasionally, you may wish to control that. The command @code{set
13106 case-sensitive} lets you do that by specifying @code{on} for
13107 case-sensitive matches or @code{off} for case-insensitive ones. If
13108 you specify @code{auto}, case sensitivity is reset to the default
13109 suitable for the source language. The default is case-sensitive
13110 matches for all languages except for Fortran, for which the default is
13111 case-insensitive matches.
13113 @kindex show case-sensitive
13114 @item show case-sensitive
13115 This command shows the current setting of case sensitivity for symbols
13118 @kindex info address
13119 @cindex address of a symbol
13120 @item info address @var{symbol}
13121 Describe where the data for @var{symbol} is stored. For a register
13122 variable, this says which register it is kept in. For a non-register
13123 local variable, this prints the stack-frame offset at which the variable
13126 Note the contrast with @samp{print &@var{symbol}}, which does not work
13127 at all for a register variable, and for a stack local variable prints
13128 the exact address of the current instantiation of the variable.
13130 @kindex info symbol
13131 @cindex symbol from address
13132 @cindex closest symbol and offset for an address
13133 @item info symbol @var{addr}
13134 Print the name of a symbol which is stored at the address @var{addr}.
13135 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
13136 nearest symbol and an offset from it:
13139 (@value{GDBP}) info symbol 0x54320
13140 _initialize_vx + 396 in section .text
13144 This is the opposite of the @code{info address} command. You can use
13145 it to find out the name of a variable or a function given its address.
13147 For dynamically linked executables, the name of executable or shared
13148 library containing the symbol is also printed:
13151 (@value{GDBP}) info symbol 0x400225
13152 _start + 5 in section .text of /tmp/a.out
13153 (@value{GDBP}) info symbol 0x2aaaac2811cf
13154 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
13158 @item whatis [@var{arg}]
13159 Print the data type of @var{arg}, which can be either an expression or
13160 a data type. With no argument, print the data type of @code{$}, the
13161 last value in the value history. If @var{arg} is an expression, it is
13162 not actually evaluated, and any side-effecting operations (such as
13163 assignments or function calls) inside it do not take place. If
13164 @var{arg} is a type name, it may be the name of a type or typedef, or
13165 for C code it may have the form @samp{class @var{class-name}},
13166 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
13167 @samp{enum @var{enum-tag}}.
13168 @xref{Expressions, ,Expressions}.
13171 @item ptype [@var{arg}]
13172 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
13173 detailed description of the type, instead of just the name of the type.
13174 @xref{Expressions, ,Expressions}.
13176 For example, for this variable declaration:
13179 struct complex @{double real; double imag;@} v;
13183 the two commands give this output:
13187 (@value{GDBP}) whatis v
13188 type = struct complex
13189 (@value{GDBP}) ptype v
13190 type = struct complex @{
13198 As with @code{whatis}, using @code{ptype} without an argument refers to
13199 the type of @code{$}, the last value in the value history.
13201 @cindex incomplete type
13202 Sometimes, programs use opaque data types or incomplete specifications
13203 of complex data structure. If the debug information included in the
13204 program does not allow @value{GDBN} to display a full declaration of
13205 the data type, it will say @samp{<incomplete type>}. For example,
13206 given these declarations:
13210 struct foo *fooptr;
13214 but no definition for @code{struct foo} itself, @value{GDBN} will say:
13217 (@value{GDBP}) ptype foo
13218 $1 = <incomplete type>
13222 ``Incomplete type'' is C terminology for data types that are not
13223 completely specified.
13226 @item info types @var{regexp}
13228 Print a brief description of all types whose names match the regular
13229 expression @var{regexp} (or all types in your program, if you supply
13230 no argument). Each complete typename is matched as though it were a
13231 complete line; thus, @samp{i type value} gives information on all
13232 types in your program whose names include the string @code{value}, but
13233 @samp{i type ^value$} gives information only on types whose complete
13234 name is @code{value}.
13236 This command differs from @code{ptype} in two ways: first, like
13237 @code{whatis}, it does not print a detailed description; second, it
13238 lists all source files where a type is defined.
13241 @cindex local variables
13242 @item info scope @var{location}
13243 List all the variables local to a particular scope. This command
13244 accepts a @var{location} argument---a function name, a source line, or
13245 an address preceded by a @samp{*}, and prints all the variables local
13246 to the scope defined by that location. (@xref{Specify Location}, for
13247 details about supported forms of @var{location}.) For example:
13250 (@value{GDBP}) @b{info scope command_line_handler}
13251 Scope for command_line_handler:
13252 Symbol rl is an argument at stack/frame offset 8, length 4.
13253 Symbol linebuffer is in static storage at address 0x150a18, length 4.
13254 Symbol linelength is in static storage at address 0x150a1c, length 4.
13255 Symbol p is a local variable in register $esi, length 4.
13256 Symbol p1 is a local variable in register $ebx, length 4.
13257 Symbol nline is a local variable in register $edx, length 4.
13258 Symbol repeat is a local variable at frame offset -8, length 4.
13262 This command is especially useful for determining what data to collect
13263 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
13266 @kindex info source
13268 Show information about the current source file---that is, the source file for
13269 the function containing the current point of execution:
13272 the name of the source file, and the directory containing it,
13274 the directory it was compiled in,
13276 its length, in lines,
13278 which programming language it is written in,
13280 whether the executable includes debugging information for that file, and
13281 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
13283 whether the debugging information includes information about
13284 preprocessor macros.
13288 @kindex info sources
13290 Print the names of all source files in your program for which there is
13291 debugging information, organized into two lists: files whose symbols
13292 have already been read, and files whose symbols will be read when needed.
13294 @kindex info functions
13295 @item info functions
13296 Print the names and data types of all defined functions.
13298 @item info functions @var{regexp}
13299 Print the names and data types of all defined functions
13300 whose names contain a match for regular expression @var{regexp}.
13301 Thus, @samp{info fun step} finds all functions whose names
13302 include @code{step}; @samp{info fun ^step} finds those whose names
13303 start with @code{step}. If a function name contains characters
13304 that conflict with the regular expression language (e.g.@:
13305 @samp{operator*()}), they may be quoted with a backslash.
13307 @kindex info variables
13308 @item info variables
13309 Print the names and data types of all variables that are defined
13310 outside of functions (i.e.@: excluding local variables).
13312 @item info variables @var{regexp}
13313 Print the names and data types of all variables (except for local
13314 variables) whose names contain a match for regular expression
13317 @kindex info classes
13318 @cindex Objective-C, classes and selectors
13320 @itemx info classes @var{regexp}
13321 Display all Objective-C classes in your program, or
13322 (with the @var{regexp} argument) all those matching a particular regular
13325 @kindex info selectors
13326 @item info selectors
13327 @itemx info selectors @var{regexp}
13328 Display all Objective-C selectors in your program, or
13329 (with the @var{regexp} argument) all those matching a particular regular
13333 This was never implemented.
13334 @kindex info methods
13336 @itemx info methods @var{regexp}
13337 The @code{info methods} command permits the user to examine all defined
13338 methods within C@t{++} program, or (with the @var{regexp} argument) a
13339 specific set of methods found in the various C@t{++} classes. Many
13340 C@t{++} classes provide a large number of methods. Thus, the output
13341 from the @code{ptype} command can be overwhelming and hard to use. The
13342 @code{info-methods} command filters the methods, printing only those
13343 which match the regular-expression @var{regexp}.
13346 @cindex reloading symbols
13347 Some systems allow individual object files that make up your program to
13348 be replaced without stopping and restarting your program. For example,
13349 in VxWorks you can simply recompile a defective object file and keep on
13350 running. If you are running on one of these systems, you can allow
13351 @value{GDBN} to reload the symbols for automatically relinked modules:
13354 @kindex set symbol-reloading
13355 @item set symbol-reloading on
13356 Replace symbol definitions for the corresponding source file when an
13357 object file with a particular name is seen again.
13359 @item set symbol-reloading off
13360 Do not replace symbol definitions when encountering object files of the
13361 same name more than once. This is the default state; if you are not
13362 running on a system that permits automatic relinking of modules, you
13363 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
13364 may discard symbols when linking large programs, that may contain
13365 several modules (from different directories or libraries) with the same
13368 @kindex show symbol-reloading
13369 @item show symbol-reloading
13370 Show the current @code{on} or @code{off} setting.
13373 @cindex opaque data types
13374 @kindex set opaque-type-resolution
13375 @item set opaque-type-resolution on
13376 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
13377 declared as a pointer to a @code{struct}, @code{class}, or
13378 @code{union}---for example, @code{struct MyType *}---that is used in one
13379 source file although the full declaration of @code{struct MyType} is in
13380 another source file. The default is on.
13382 A change in the setting of this subcommand will not take effect until
13383 the next time symbols for a file are loaded.
13385 @item set opaque-type-resolution off
13386 Tell @value{GDBN} not to resolve opaque types. In this case, the type
13387 is printed as follows:
13389 @{<no data fields>@}
13392 @kindex show opaque-type-resolution
13393 @item show opaque-type-resolution
13394 Show whether opaque types are resolved or not.
13396 @kindex maint print symbols
13397 @cindex symbol dump
13398 @kindex maint print psymbols
13399 @cindex partial symbol dump
13400 @item maint print symbols @var{filename}
13401 @itemx maint print psymbols @var{filename}
13402 @itemx maint print msymbols @var{filename}
13403 Write a dump of debugging symbol data into the file @var{filename}.
13404 These commands are used to debug the @value{GDBN} symbol-reading code. Only
13405 symbols with debugging data are included. If you use @samp{maint print
13406 symbols}, @value{GDBN} includes all the symbols for which it has already
13407 collected full details: that is, @var{filename} reflects symbols for
13408 only those files whose symbols @value{GDBN} has read. You can use the
13409 command @code{info sources} to find out which files these are. If you
13410 use @samp{maint print psymbols} instead, the dump shows information about
13411 symbols that @value{GDBN} only knows partially---that is, symbols defined in
13412 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
13413 @samp{maint print msymbols} dumps just the minimal symbol information
13414 required for each object file from which @value{GDBN} has read some symbols.
13415 @xref{Files, ,Commands to Specify Files}, for a discussion of how
13416 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
13418 @kindex maint info symtabs
13419 @kindex maint info psymtabs
13420 @cindex listing @value{GDBN}'s internal symbol tables
13421 @cindex symbol tables, listing @value{GDBN}'s internal
13422 @cindex full symbol tables, listing @value{GDBN}'s internal
13423 @cindex partial symbol tables, listing @value{GDBN}'s internal
13424 @item maint info symtabs @r{[} @var{regexp} @r{]}
13425 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
13427 List the @code{struct symtab} or @code{struct partial_symtab}
13428 structures whose names match @var{regexp}. If @var{regexp} is not
13429 given, list them all. The output includes expressions which you can
13430 copy into a @value{GDBN} debugging this one to examine a particular
13431 structure in more detail. For example:
13434 (@value{GDBP}) maint info psymtabs dwarf2read
13435 @{ objfile /home/gnu/build/gdb/gdb
13436 ((struct objfile *) 0x82e69d0)
13437 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
13438 ((struct partial_symtab *) 0x8474b10)
13441 text addresses 0x814d3c8 -- 0x8158074
13442 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
13443 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
13444 dependencies (none)
13447 (@value{GDBP}) maint info symtabs
13451 We see that there is one partial symbol table whose filename contains
13452 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
13453 and we see that @value{GDBN} has not read in any symtabs yet at all.
13454 If we set a breakpoint on a function, that will cause @value{GDBN} to
13455 read the symtab for the compilation unit containing that function:
13458 (@value{GDBP}) break dwarf2_psymtab_to_symtab
13459 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
13461 (@value{GDBP}) maint info symtabs
13462 @{ objfile /home/gnu/build/gdb/gdb
13463 ((struct objfile *) 0x82e69d0)
13464 @{ symtab /home/gnu/src/gdb/dwarf2read.c
13465 ((struct symtab *) 0x86c1f38)
13468 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
13469 linetable ((struct linetable *) 0x8370fa0)
13470 debugformat DWARF 2
13479 @chapter Altering Execution
13481 Once you think you have found an error in your program, you might want to
13482 find out for certain whether correcting the apparent error would lead to
13483 correct results in the rest of the run. You can find the answer by
13484 experiment, using the @value{GDBN} features for altering execution of the
13487 For example, you can store new values into variables or memory
13488 locations, give your program a signal, restart it at a different
13489 address, or even return prematurely from a function.
13492 * Assignment:: Assignment to variables
13493 * Jumping:: Continuing at a different address
13494 * Signaling:: Giving your program a signal
13495 * Returning:: Returning from a function
13496 * Calling:: Calling your program's functions
13497 * Patching:: Patching your program
13501 @section Assignment to Variables
13504 @cindex setting variables
13505 To alter the value of a variable, evaluate an assignment expression.
13506 @xref{Expressions, ,Expressions}. For example,
13513 stores the value 4 into the variable @code{x}, and then prints the
13514 value of the assignment expression (which is 4).
13515 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
13516 information on operators in supported languages.
13518 @kindex set variable
13519 @cindex variables, setting
13520 If you are not interested in seeing the value of the assignment, use the
13521 @code{set} command instead of the @code{print} command. @code{set} is
13522 really the same as @code{print} except that the expression's value is
13523 not printed and is not put in the value history (@pxref{Value History,
13524 ,Value History}). The expression is evaluated only for its effects.
13526 If the beginning of the argument string of the @code{set} command
13527 appears identical to a @code{set} subcommand, use the @code{set
13528 variable} command instead of just @code{set}. This command is identical
13529 to @code{set} except for its lack of subcommands. For example, if your
13530 program has a variable @code{width}, you get an error if you try to set
13531 a new value with just @samp{set width=13}, because @value{GDBN} has the
13532 command @code{set width}:
13535 (@value{GDBP}) whatis width
13537 (@value{GDBP}) p width
13539 (@value{GDBP}) set width=47
13540 Invalid syntax in expression.
13544 The invalid expression, of course, is @samp{=47}. In
13545 order to actually set the program's variable @code{width}, use
13548 (@value{GDBP}) set var width=47
13551 Because the @code{set} command has many subcommands that can conflict
13552 with the names of program variables, it is a good idea to use the
13553 @code{set variable} command instead of just @code{set}. For example, if
13554 your program has a variable @code{g}, you run into problems if you try
13555 to set a new value with just @samp{set g=4}, because @value{GDBN} has
13556 the command @code{set gnutarget}, abbreviated @code{set g}:
13560 (@value{GDBP}) whatis g
13564 (@value{GDBP}) set g=4
13568 The program being debugged has been started already.
13569 Start it from the beginning? (y or n) y
13570 Starting program: /home/smith/cc_progs/a.out
13571 "/home/smith/cc_progs/a.out": can't open to read symbols:
13572 Invalid bfd target.
13573 (@value{GDBP}) show g
13574 The current BFD target is "=4".
13579 The program variable @code{g} did not change, and you silently set the
13580 @code{gnutarget} to an invalid value. In order to set the variable
13584 (@value{GDBP}) set var g=4
13587 @value{GDBN} allows more implicit conversions in assignments than C; you can
13588 freely store an integer value into a pointer variable or vice versa,
13589 and you can convert any structure to any other structure that is the
13590 same length or shorter.
13591 @comment FIXME: how do structs align/pad in these conversions?
13592 @comment /doc@cygnus.com 18dec1990
13594 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
13595 construct to generate a value of specified type at a specified address
13596 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
13597 to memory location @code{0x83040} as an integer (which implies a certain size
13598 and representation in memory), and
13601 set @{int@}0x83040 = 4
13605 stores the value 4 into that memory location.
13608 @section Continuing at a Different Address
13610 Ordinarily, when you continue your program, you do so at the place where
13611 it stopped, with the @code{continue} command. You can instead continue at
13612 an address of your own choosing, with the following commands:
13616 @item jump @var{linespec}
13617 @itemx jump @var{location}
13618 Resume execution at line @var{linespec} or at address given by
13619 @var{location}. Execution stops again immediately if there is a
13620 breakpoint there. @xref{Specify Location}, for a description of the
13621 different forms of @var{linespec} and @var{location}. It is common
13622 practice to use the @code{tbreak} command in conjunction with
13623 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
13625 The @code{jump} command does not change the current stack frame, or
13626 the stack pointer, or the contents of any memory location or any
13627 register other than the program counter. If line @var{linespec} is in
13628 a different function from the one currently executing, the results may
13629 be bizarre if the two functions expect different patterns of arguments or
13630 of local variables. For this reason, the @code{jump} command requests
13631 confirmation if the specified line is not in the function currently
13632 executing. However, even bizarre results are predictable if you are
13633 well acquainted with the machine-language code of your program.
13636 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
13637 On many systems, you can get much the same effect as the @code{jump}
13638 command by storing a new value into the register @code{$pc}. The
13639 difference is that this does not start your program running; it only
13640 changes the address of where it @emph{will} run when you continue. For
13648 makes the next @code{continue} command or stepping command execute at
13649 address @code{0x485}, rather than at the address where your program stopped.
13650 @xref{Continuing and Stepping, ,Continuing and Stepping}.
13652 The most common occasion to use the @code{jump} command is to back
13653 up---perhaps with more breakpoints set---over a portion of a program
13654 that has already executed, in order to examine its execution in more
13659 @section Giving your Program a Signal
13660 @cindex deliver a signal to a program
13664 @item signal @var{signal}
13665 Resume execution where your program stopped, but immediately give it the
13666 signal @var{signal}. @var{signal} can be the name or the number of a
13667 signal. For example, on many systems @code{signal 2} and @code{signal
13668 SIGINT} are both ways of sending an interrupt signal.
13670 Alternatively, if @var{signal} is zero, continue execution without
13671 giving a signal. This is useful when your program stopped on account of
13672 a signal and would ordinary see the signal when resumed with the
13673 @code{continue} command; @samp{signal 0} causes it to resume without a
13676 @code{signal} does not repeat when you press @key{RET} a second time
13677 after executing the command.
13681 Invoking the @code{signal} command is not the same as invoking the
13682 @code{kill} utility from the shell. Sending a signal with @code{kill}
13683 causes @value{GDBN} to decide what to do with the signal depending on
13684 the signal handling tables (@pxref{Signals}). The @code{signal} command
13685 passes the signal directly to your program.
13689 @section Returning from a Function
13692 @cindex returning from a function
13695 @itemx return @var{expression}
13696 You can cancel execution of a function call with the @code{return}
13697 command. If you give an
13698 @var{expression} argument, its value is used as the function's return
13702 When you use @code{return}, @value{GDBN} discards the selected stack frame
13703 (and all frames within it). You can think of this as making the
13704 discarded frame return prematurely. If you wish to specify a value to
13705 be returned, give that value as the argument to @code{return}.
13707 This pops the selected stack frame (@pxref{Selection, ,Selecting a
13708 Frame}), and any other frames inside of it, leaving its caller as the
13709 innermost remaining frame. That frame becomes selected. The
13710 specified value is stored in the registers used for returning values
13713 The @code{return} command does not resume execution; it leaves the
13714 program stopped in the state that would exist if the function had just
13715 returned. In contrast, the @code{finish} command (@pxref{Continuing
13716 and Stepping, ,Continuing and Stepping}) resumes execution until the
13717 selected stack frame returns naturally.
13719 @value{GDBN} needs to know how the @var{expression} argument should be set for
13720 the inferior. The concrete registers assignment depends on the OS ABI and the
13721 type being returned by the selected stack frame. For example it is common for
13722 OS ABI to return floating point values in FPU registers while integer values in
13723 CPU registers. Still some ABIs return even floating point values in CPU
13724 registers. Larger integer widths (such as @code{long long int}) also have
13725 specific placement rules. @value{GDBN} already knows the OS ABI from its
13726 current target so it needs to find out also the type being returned to make the
13727 assignment into the right register(s).
13729 Normally, the selected stack frame has debug info. @value{GDBN} will always
13730 use the debug info instead of the implicit type of @var{expression} when the
13731 debug info is available. For example, if you type @kbd{return -1}, and the
13732 function in the current stack frame is declared to return a @code{long long
13733 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
13734 into a @code{long long int}:
13737 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
13739 (@value{GDBP}) return -1
13740 Make func return now? (y or n) y
13741 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
13742 43 printf ("result=%lld\n", func ());
13746 However, if the selected stack frame does not have a debug info, e.g., if the
13747 function was compiled without debug info, @value{GDBN} has to find out the type
13748 to return from user. Specifying a different type by mistake may set the value
13749 in different inferior registers than the caller code expects. For example,
13750 typing @kbd{return -1} with its implicit type @code{int} would set only a part
13751 of a @code{long long int} result for a debug info less function (on 32-bit
13752 architectures). Therefore the user is required to specify the return type by
13753 an appropriate cast explicitly:
13756 Breakpoint 2, 0x0040050b in func ()
13757 (@value{GDBP}) return -1
13758 Return value type not available for selected stack frame.
13759 Please use an explicit cast of the value to return.
13760 (@value{GDBP}) return (long long int) -1
13761 Make selected stack frame return now? (y or n) y
13762 #0 0x00400526 in main ()
13767 @section Calling Program Functions
13770 @cindex calling functions
13771 @cindex inferior functions, calling
13772 @item print @var{expr}
13773 Evaluate the expression @var{expr} and display the resulting value.
13774 @var{expr} may include calls to functions in the program being
13778 @item call @var{expr}
13779 Evaluate the expression @var{expr} without displaying @code{void}
13782 You can use this variant of the @code{print} command if you want to
13783 execute a function from your program that does not return anything
13784 (a.k.a.@: @dfn{a void function}), but without cluttering the output
13785 with @code{void} returned values that @value{GDBN} will otherwise
13786 print. If the result is not void, it is printed and saved in the
13790 It is possible for the function you call via the @code{print} or
13791 @code{call} command to generate a signal (e.g., if there's a bug in
13792 the function, or if you passed it incorrect arguments). What happens
13793 in that case is controlled by the @code{set unwindonsignal} command.
13795 Similarly, with a C@t{++} program it is possible for the function you
13796 call via the @code{print} or @code{call} command to generate an
13797 exception that is not handled due to the constraints of the dummy
13798 frame. In this case, any exception that is raised in the frame, but has
13799 an out-of-frame exception handler will not be found. GDB builds a
13800 dummy-frame for the inferior function call, and the unwinder cannot
13801 seek for exception handlers outside of this dummy-frame. What happens
13802 in that case is controlled by the
13803 @code{set unwind-on-terminating-exception} command.
13806 @item set unwindonsignal
13807 @kindex set unwindonsignal
13808 @cindex unwind stack in called functions
13809 @cindex call dummy stack unwinding
13810 Set unwinding of the stack if a signal is received while in a function
13811 that @value{GDBN} called in the program being debugged. If set to on,
13812 @value{GDBN} unwinds the stack it created for the call and restores
13813 the context to what it was before the call. If set to off (the
13814 default), @value{GDBN} stops in the frame where the signal was
13817 @item show unwindonsignal
13818 @kindex show unwindonsignal
13819 Show the current setting of stack unwinding in the functions called by
13822 @item set unwind-on-terminating-exception
13823 @kindex set unwind-on-terminating-exception
13824 @cindex unwind stack in called functions with unhandled exceptions
13825 @cindex call dummy stack unwinding on unhandled exception.
13826 Set unwinding of the stack if a C@t{++} exception is raised, but left
13827 unhandled while in a function that @value{GDBN} called in the program being
13828 debugged. If set to on (the default), @value{GDBN} unwinds the stack
13829 it created for the call and restores the context to what it was before
13830 the call. If set to off, @value{GDBN} the exception is delivered to
13831 the default C@t{++} exception handler and the inferior terminated.
13833 @item show unwind-on-terminating-exception
13834 @kindex show unwind-on-terminating-exception
13835 Show the current setting of stack unwinding in the functions called by
13840 @cindex weak alias functions
13841 Sometimes, a function you wish to call is actually a @dfn{weak alias}
13842 for another function. In such case, @value{GDBN} might not pick up
13843 the type information, including the types of the function arguments,
13844 which causes @value{GDBN} to call the inferior function incorrectly.
13845 As a result, the called function will function erroneously and may
13846 even crash. A solution to that is to use the name of the aliased
13850 @section Patching Programs
13852 @cindex patching binaries
13853 @cindex writing into executables
13854 @cindex writing into corefiles
13856 By default, @value{GDBN} opens the file containing your program's
13857 executable code (or the corefile) read-only. This prevents accidental
13858 alterations to machine code; but it also prevents you from intentionally
13859 patching your program's binary.
13861 If you'd like to be able to patch the binary, you can specify that
13862 explicitly with the @code{set write} command. For example, you might
13863 want to turn on internal debugging flags, or even to make emergency
13869 @itemx set write off
13870 If you specify @samp{set write on}, @value{GDBN} opens executable and
13871 core files for both reading and writing; if you specify @kbd{set write
13872 off} (the default), @value{GDBN} opens them read-only.
13874 If you have already loaded a file, you must load it again (using the
13875 @code{exec-file} or @code{core-file} command) after changing @code{set
13876 write}, for your new setting to take effect.
13880 Display whether executable files and core files are opened for writing
13881 as well as reading.
13885 @chapter @value{GDBN} Files
13887 @value{GDBN} needs to know the file name of the program to be debugged,
13888 both in order to read its symbol table and in order to start your
13889 program. To debug a core dump of a previous run, you must also tell
13890 @value{GDBN} the name of the core dump file.
13893 * Files:: Commands to specify files
13894 * Separate Debug Files:: Debugging information in separate files
13895 * Symbol Errors:: Errors reading symbol files
13896 * Data Files:: GDB data files
13900 @section Commands to Specify Files
13902 @cindex symbol table
13903 @cindex core dump file
13905 You may want to specify executable and core dump file names. The usual
13906 way to do this is at start-up time, using the arguments to
13907 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
13908 Out of @value{GDBN}}).
13910 Occasionally it is necessary to change to a different file during a
13911 @value{GDBN} session. Or you may run @value{GDBN} and forget to
13912 specify a file you want to use. Or you are debugging a remote target
13913 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
13914 Program}). In these situations the @value{GDBN} commands to specify
13915 new files are useful.
13918 @cindex executable file
13920 @item file @var{filename}
13921 Use @var{filename} as the program to be debugged. It is read for its
13922 symbols and for the contents of pure memory. It is also the program
13923 executed when you use the @code{run} command. If you do not specify a
13924 directory and the file is not found in the @value{GDBN} working directory,
13925 @value{GDBN} uses the environment variable @code{PATH} as a list of
13926 directories to search, just as the shell does when looking for a program
13927 to run. You can change the value of this variable, for both @value{GDBN}
13928 and your program, using the @code{path} command.
13930 @cindex unlinked object files
13931 @cindex patching object files
13932 You can load unlinked object @file{.o} files into @value{GDBN} using
13933 the @code{file} command. You will not be able to ``run'' an object
13934 file, but you can disassemble functions and inspect variables. Also,
13935 if the underlying BFD functionality supports it, you could use
13936 @kbd{gdb -write} to patch object files using this technique. Note
13937 that @value{GDBN} can neither interpret nor modify relocations in this
13938 case, so branches and some initialized variables will appear to go to
13939 the wrong place. But this feature is still handy from time to time.
13942 @code{file} with no argument makes @value{GDBN} discard any information it
13943 has on both executable file and the symbol table.
13946 @item exec-file @r{[} @var{filename} @r{]}
13947 Specify that the program to be run (but not the symbol table) is found
13948 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
13949 if necessary to locate your program. Omitting @var{filename} means to
13950 discard information on the executable file.
13952 @kindex symbol-file
13953 @item symbol-file @r{[} @var{filename} @r{]}
13954 Read symbol table information from file @var{filename}. @code{PATH} is
13955 searched when necessary. Use the @code{file} command to get both symbol
13956 table and program to run from the same file.
13958 @code{symbol-file} with no argument clears out @value{GDBN} information on your
13959 program's symbol table.
13961 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
13962 some breakpoints and auto-display expressions. This is because they may
13963 contain pointers to the internal data recording symbols and data types,
13964 which are part of the old symbol table data being discarded inside
13967 @code{symbol-file} does not repeat if you press @key{RET} again after
13970 When @value{GDBN} is configured for a particular environment, it
13971 understands debugging information in whatever format is the standard
13972 generated for that environment; you may use either a @sc{gnu} compiler, or
13973 other compilers that adhere to the local conventions.
13974 Best results are usually obtained from @sc{gnu} compilers; for example,
13975 using @code{@value{NGCC}} you can generate debugging information for
13978 For most kinds of object files, with the exception of old SVR3 systems
13979 using COFF, the @code{symbol-file} command does not normally read the
13980 symbol table in full right away. Instead, it scans the symbol table
13981 quickly to find which source files and which symbols are present. The
13982 details are read later, one source file at a time, as they are needed.
13984 The purpose of this two-stage reading strategy is to make @value{GDBN}
13985 start up faster. For the most part, it is invisible except for
13986 occasional pauses while the symbol table details for a particular source
13987 file are being read. (The @code{set verbose} command can turn these
13988 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
13989 Warnings and Messages}.)
13991 We have not implemented the two-stage strategy for COFF yet. When the
13992 symbol table is stored in COFF format, @code{symbol-file} reads the
13993 symbol table data in full right away. Note that ``stabs-in-COFF''
13994 still does the two-stage strategy, since the debug info is actually
13998 @cindex reading symbols immediately
13999 @cindex symbols, reading immediately
14000 @item symbol-file @r{[} -readnow @r{]} @var{filename}
14001 @itemx file @r{[} -readnow @r{]} @var{filename}
14002 You can override the @value{GDBN} two-stage strategy for reading symbol
14003 tables by using the @samp{-readnow} option with any of the commands that
14004 load symbol table information, if you want to be sure @value{GDBN} has the
14005 entire symbol table available.
14007 @c FIXME: for now no mention of directories, since this seems to be in
14008 @c flux. 13mar1992 status is that in theory GDB would look either in
14009 @c current dir or in same dir as myprog; but issues like competing
14010 @c GDB's, or clutter in system dirs, mean that in practice right now
14011 @c only current dir is used. FFish says maybe a special GDB hierarchy
14012 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
14016 @item core-file @r{[}@var{filename}@r{]}
14018 Specify the whereabouts of a core dump file to be used as the ``contents
14019 of memory''. Traditionally, core files contain only some parts of the
14020 address space of the process that generated them; @value{GDBN} can access the
14021 executable file itself for other parts.
14023 @code{core-file} with no argument specifies that no core file is
14026 Note that the core file is ignored when your program is actually running
14027 under @value{GDBN}. So, if you have been running your program and you
14028 wish to debug a core file instead, you must kill the subprocess in which
14029 the program is running. To do this, use the @code{kill} command
14030 (@pxref{Kill Process, ,Killing the Child Process}).
14032 @kindex add-symbol-file
14033 @cindex dynamic linking
14034 @item add-symbol-file @var{filename} @var{address}
14035 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
14036 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
14037 The @code{add-symbol-file} command reads additional symbol table
14038 information from the file @var{filename}. You would use this command
14039 when @var{filename} has been dynamically loaded (by some other means)
14040 into the program that is running. @var{address} should be the memory
14041 address at which the file has been loaded; @value{GDBN} cannot figure
14042 this out for itself. You can additionally specify an arbitrary number
14043 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
14044 section name and base address for that section. You can specify any
14045 @var{address} as an expression.
14047 The symbol table of the file @var{filename} is added to the symbol table
14048 originally read with the @code{symbol-file} command. You can use the
14049 @code{add-symbol-file} command any number of times; the new symbol data
14050 thus read keeps adding to the old. To discard all old symbol data
14051 instead, use the @code{symbol-file} command without any arguments.
14053 @cindex relocatable object files, reading symbols from
14054 @cindex object files, relocatable, reading symbols from
14055 @cindex reading symbols from relocatable object files
14056 @cindex symbols, reading from relocatable object files
14057 @cindex @file{.o} files, reading symbols from
14058 Although @var{filename} is typically a shared library file, an
14059 executable file, or some other object file which has been fully
14060 relocated for loading into a process, you can also load symbolic
14061 information from relocatable @file{.o} files, as long as:
14065 the file's symbolic information refers only to linker symbols defined in
14066 that file, not to symbols defined by other object files,
14068 every section the file's symbolic information refers to has actually
14069 been loaded into the inferior, as it appears in the file, and
14071 you can determine the address at which every section was loaded, and
14072 provide these to the @code{add-symbol-file} command.
14076 Some embedded operating systems, like Sun Chorus and VxWorks, can load
14077 relocatable files into an already running program; such systems
14078 typically make the requirements above easy to meet. However, it's
14079 important to recognize that many native systems use complex link
14080 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
14081 assembly, for example) that make the requirements difficult to meet. In
14082 general, one cannot assume that using @code{add-symbol-file} to read a
14083 relocatable object file's symbolic information will have the same effect
14084 as linking the relocatable object file into the program in the normal
14087 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
14089 @kindex add-symbol-file-from-memory
14090 @cindex @code{syscall DSO}
14091 @cindex load symbols from memory
14092 @item add-symbol-file-from-memory @var{address}
14093 Load symbols from the given @var{address} in a dynamically loaded
14094 object file whose image is mapped directly into the inferior's memory.
14095 For example, the Linux kernel maps a @code{syscall DSO} into each
14096 process's address space; this DSO provides kernel-specific code for
14097 some system calls. The argument can be any expression whose
14098 evaluation yields the address of the file's shared object file header.
14099 For this command to work, you must have used @code{symbol-file} or
14100 @code{exec-file} commands in advance.
14102 @kindex add-shared-symbol-files
14104 @item add-shared-symbol-files @var{library-file}
14105 @itemx assf @var{library-file}
14106 The @code{add-shared-symbol-files} command can currently be used only
14107 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
14108 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
14109 @value{GDBN} automatically looks for shared libraries, however if
14110 @value{GDBN} does not find yours, you can invoke
14111 @code{add-shared-symbol-files}. It takes one argument: the shared
14112 library's file name. @code{assf} is a shorthand alias for
14113 @code{add-shared-symbol-files}.
14116 @item section @var{section} @var{addr}
14117 The @code{section} command changes the base address of the named
14118 @var{section} of the exec file to @var{addr}. This can be used if the
14119 exec file does not contain section addresses, (such as in the
14120 @code{a.out} format), or when the addresses specified in the file
14121 itself are wrong. Each section must be changed separately. The
14122 @code{info files} command, described below, lists all the sections and
14126 @kindex info target
14129 @code{info files} and @code{info target} are synonymous; both print the
14130 current target (@pxref{Targets, ,Specifying a Debugging Target}),
14131 including the names of the executable and core dump files currently in
14132 use by @value{GDBN}, and the files from which symbols were loaded. The
14133 command @code{help target} lists all possible targets rather than
14136 @kindex maint info sections
14137 @item maint info sections
14138 Another command that can give you extra information about program sections
14139 is @code{maint info sections}. In addition to the section information
14140 displayed by @code{info files}, this command displays the flags and file
14141 offset of each section in the executable and core dump files. In addition,
14142 @code{maint info sections} provides the following command options (which
14143 may be arbitrarily combined):
14147 Display sections for all loaded object files, including shared libraries.
14148 @item @var{sections}
14149 Display info only for named @var{sections}.
14150 @item @var{section-flags}
14151 Display info only for sections for which @var{section-flags} are true.
14152 The section flags that @value{GDBN} currently knows about are:
14155 Section will have space allocated in the process when loaded.
14156 Set for all sections except those containing debug information.
14158 Section will be loaded from the file into the child process memory.
14159 Set for pre-initialized code and data, clear for @code{.bss} sections.
14161 Section needs to be relocated before loading.
14163 Section cannot be modified by the child process.
14165 Section contains executable code only.
14167 Section contains data only (no executable code).
14169 Section will reside in ROM.
14171 Section contains data for constructor/destructor lists.
14173 Section is not empty.
14175 An instruction to the linker to not output the section.
14176 @item COFF_SHARED_LIBRARY
14177 A notification to the linker that the section contains
14178 COFF shared library information.
14180 Section contains common symbols.
14183 @kindex set trust-readonly-sections
14184 @cindex read-only sections
14185 @item set trust-readonly-sections on
14186 Tell @value{GDBN} that readonly sections in your object file
14187 really are read-only (i.e.@: that their contents will not change).
14188 In that case, @value{GDBN} can fetch values from these sections
14189 out of the object file, rather than from the target program.
14190 For some targets (notably embedded ones), this can be a significant
14191 enhancement to debugging performance.
14193 The default is off.
14195 @item set trust-readonly-sections off
14196 Tell @value{GDBN} not to trust readonly sections. This means that
14197 the contents of the section might change while the program is running,
14198 and must therefore be fetched from the target when needed.
14200 @item show trust-readonly-sections
14201 Show the current setting of trusting readonly sections.
14204 All file-specifying commands allow both absolute and relative file names
14205 as arguments. @value{GDBN} always converts the file name to an absolute file
14206 name and remembers it that way.
14208 @cindex shared libraries
14209 @anchor{Shared Libraries}
14210 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
14211 and IBM RS/6000 AIX shared libraries.
14213 On MS-Windows @value{GDBN} must be linked with the Expat library to support
14214 shared libraries. @xref{Expat}.
14216 @value{GDBN} automatically loads symbol definitions from shared libraries
14217 when you use the @code{run} command, or when you examine a core file.
14218 (Before you issue the @code{run} command, @value{GDBN} does not understand
14219 references to a function in a shared library, however---unless you are
14220 debugging a core file).
14222 On HP-UX, if the program loads a library explicitly, @value{GDBN}
14223 automatically loads the symbols at the time of the @code{shl_load} call.
14225 @c FIXME: some @value{GDBN} release may permit some refs to undef
14226 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
14227 @c FIXME...lib; check this from time to time when updating manual
14229 There are times, however, when you may wish to not automatically load
14230 symbol definitions from shared libraries, such as when they are
14231 particularly large or there are many of them.
14233 To control the automatic loading of shared library symbols, use the
14237 @kindex set auto-solib-add
14238 @item set auto-solib-add @var{mode}
14239 If @var{mode} is @code{on}, symbols from all shared object libraries
14240 will be loaded automatically when the inferior begins execution, you
14241 attach to an independently started inferior, or when the dynamic linker
14242 informs @value{GDBN} that a new library has been loaded. If @var{mode}
14243 is @code{off}, symbols must be loaded manually, using the
14244 @code{sharedlibrary} command. The default value is @code{on}.
14246 @cindex memory used for symbol tables
14247 If your program uses lots of shared libraries with debug info that
14248 takes large amounts of memory, you can decrease the @value{GDBN}
14249 memory footprint by preventing it from automatically loading the
14250 symbols from shared libraries. To that end, type @kbd{set
14251 auto-solib-add off} before running the inferior, then load each
14252 library whose debug symbols you do need with @kbd{sharedlibrary
14253 @var{regexp}}, where @var{regexp} is a regular expression that matches
14254 the libraries whose symbols you want to be loaded.
14256 @kindex show auto-solib-add
14257 @item show auto-solib-add
14258 Display the current autoloading mode.
14261 @cindex load shared library
14262 To explicitly load shared library symbols, use the @code{sharedlibrary}
14266 @kindex info sharedlibrary
14268 @item info share @var{regex}
14269 @itemx info sharedlibrary @var{regex}
14270 Print the names of the shared libraries which are currently loaded
14271 that match @var{regex}. If @var{regex} is omitted then print
14272 all shared libraries that are loaded.
14274 @kindex sharedlibrary
14276 @item sharedlibrary @var{regex}
14277 @itemx share @var{regex}
14278 Load shared object library symbols for files matching a
14279 Unix regular expression.
14280 As with files loaded automatically, it only loads shared libraries
14281 required by your program for a core file or after typing @code{run}. If
14282 @var{regex} is omitted all shared libraries required by your program are
14285 @item nosharedlibrary
14286 @kindex nosharedlibrary
14287 @cindex unload symbols from shared libraries
14288 Unload all shared object library symbols. This discards all symbols
14289 that have been loaded from all shared libraries. Symbols from shared
14290 libraries that were loaded by explicit user requests are not
14294 Sometimes you may wish that @value{GDBN} stops and gives you control
14295 when any of shared library events happen. Use the @code{set
14296 stop-on-solib-events} command for this:
14299 @item set stop-on-solib-events
14300 @kindex set stop-on-solib-events
14301 This command controls whether @value{GDBN} should give you control
14302 when the dynamic linker notifies it about some shared library event.
14303 The most common event of interest is loading or unloading of a new
14306 @item show stop-on-solib-events
14307 @kindex show stop-on-solib-events
14308 Show whether @value{GDBN} stops and gives you control when shared
14309 library events happen.
14312 Shared libraries are also supported in many cross or remote debugging
14313 configurations. @value{GDBN} needs to have access to the target's libraries;
14314 this can be accomplished either by providing copies of the libraries
14315 on the host system, or by asking @value{GDBN} to automatically retrieve the
14316 libraries from the target. If copies of the target libraries are
14317 provided, they need to be the same as the target libraries, although the
14318 copies on the target can be stripped as long as the copies on the host are
14321 @cindex where to look for shared libraries
14322 For remote debugging, you need to tell @value{GDBN} where the target
14323 libraries are, so that it can load the correct copies---otherwise, it
14324 may try to load the host's libraries. @value{GDBN} has two variables
14325 to specify the search directories for target libraries.
14328 @cindex prefix for shared library file names
14329 @cindex system root, alternate
14330 @kindex set solib-absolute-prefix
14331 @kindex set sysroot
14332 @item set sysroot @var{path}
14333 Use @var{path} as the system root for the program being debugged. Any
14334 absolute shared library paths will be prefixed with @var{path}; many
14335 runtime loaders store the absolute paths to the shared library in the
14336 target program's memory. If you use @code{set sysroot} to find shared
14337 libraries, they need to be laid out in the same way that they are on
14338 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
14341 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
14342 retrieve the target libraries from the remote system. This is only
14343 supported when using a remote target that supports the @code{remote get}
14344 command (@pxref{File Transfer,,Sending files to a remote system}).
14345 The part of @var{path} following the initial @file{remote:}
14346 (if present) is used as system root prefix on the remote file system.
14347 @footnote{If you want to specify a local system root using a directory
14348 that happens to be named @file{remote:}, you need to use some equivalent
14349 variant of the name like @file{./remote:}.}
14351 The @code{set solib-absolute-prefix} command is an alias for @code{set
14354 @cindex default system root
14355 @cindex @samp{--with-sysroot}
14356 You can set the default system root by using the configure-time
14357 @samp{--with-sysroot} option. If the system root is inside
14358 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
14359 @samp{--exec-prefix}), then the default system root will be updated
14360 automatically if the installed @value{GDBN} is moved to a new
14363 @kindex show sysroot
14365 Display the current shared library prefix.
14367 @kindex set solib-search-path
14368 @item set solib-search-path @var{path}
14369 If this variable is set, @var{path} is a colon-separated list of
14370 directories to search for shared libraries. @samp{solib-search-path}
14371 is used after @samp{sysroot} fails to locate the library, or if the
14372 path to the library is relative instead of absolute. If you want to
14373 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
14374 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
14375 finding your host's libraries. @samp{sysroot} is preferred; setting
14376 it to a nonexistent directory may interfere with automatic loading
14377 of shared library symbols.
14379 @kindex show solib-search-path
14380 @item show solib-search-path
14381 Display the current shared library search path.
14385 @node Separate Debug Files
14386 @section Debugging Information in Separate Files
14387 @cindex separate debugging information files
14388 @cindex debugging information in separate files
14389 @cindex @file{.debug} subdirectories
14390 @cindex debugging information directory, global
14391 @cindex global debugging information directory
14392 @cindex build ID, and separate debugging files
14393 @cindex @file{.build-id} directory
14395 @value{GDBN} allows you to put a program's debugging information in a
14396 file separate from the executable itself, in a way that allows
14397 @value{GDBN} to find and load the debugging information automatically.
14398 Since debugging information can be very large---sometimes larger
14399 than the executable code itself---some systems distribute debugging
14400 information for their executables in separate files, which users can
14401 install only when they need to debug a problem.
14403 @value{GDBN} supports two ways of specifying the separate debug info
14408 The executable contains a @dfn{debug link} that specifies the name of
14409 the separate debug info file. The separate debug file's name is
14410 usually @file{@var{executable}.debug}, where @var{executable} is the
14411 name of the corresponding executable file without leading directories
14412 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
14413 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
14414 checksum for the debug file, which @value{GDBN} uses to validate that
14415 the executable and the debug file came from the same build.
14418 The executable contains a @dfn{build ID}, a unique bit string that is
14419 also present in the corresponding debug info file. (This is supported
14420 only on some operating systems, notably those which use the ELF format
14421 for binary files and the @sc{gnu} Binutils.) For more details about
14422 this feature, see the description of the @option{--build-id}
14423 command-line option in @ref{Options, , Command Line Options, ld.info,
14424 The GNU Linker}. The debug info file's name is not specified
14425 explicitly by the build ID, but can be computed from the build ID, see
14429 Depending on the way the debug info file is specified, @value{GDBN}
14430 uses two different methods of looking for the debug file:
14434 For the ``debug link'' method, @value{GDBN} looks up the named file in
14435 the directory of the executable file, then in a subdirectory of that
14436 directory named @file{.debug}, and finally under the global debug
14437 directory, in a subdirectory whose name is identical to the leading
14438 directories of the executable's absolute file name.
14441 For the ``build ID'' method, @value{GDBN} looks in the
14442 @file{.build-id} subdirectory of the global debug directory for a file
14443 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
14444 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
14445 are the rest of the bit string. (Real build ID strings are 32 or more
14446 hex characters, not 10.)
14449 So, for example, suppose you ask @value{GDBN} to debug
14450 @file{/usr/bin/ls}, which has a debug link that specifies the
14451 file @file{ls.debug}, and a build ID whose value in hex is
14452 @code{abcdef1234}. If the global debug directory is
14453 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
14454 debug information files, in the indicated order:
14458 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
14460 @file{/usr/bin/ls.debug}
14462 @file{/usr/bin/.debug/ls.debug}
14464 @file{/usr/lib/debug/usr/bin/ls.debug}.
14467 You can set the global debugging info directory's name, and view the
14468 name @value{GDBN} is currently using.
14472 @kindex set debug-file-directory
14473 @item set debug-file-directory @var{directories}
14474 Set the directories which @value{GDBN} searches for separate debugging
14475 information files to @var{directory}. Multiple directory components can be set
14476 concatenating them by a directory separator.
14478 @kindex show debug-file-directory
14479 @item show debug-file-directory
14480 Show the directories @value{GDBN} searches for separate debugging
14485 @cindex @code{.gnu_debuglink} sections
14486 @cindex debug link sections
14487 A debug link is a special section of the executable file named
14488 @code{.gnu_debuglink}. The section must contain:
14492 A filename, with any leading directory components removed, followed by
14495 zero to three bytes of padding, as needed to reach the next four-byte
14496 boundary within the section, and
14498 a four-byte CRC checksum, stored in the same endianness used for the
14499 executable file itself. The checksum is computed on the debugging
14500 information file's full contents by the function given below, passing
14501 zero as the @var{crc} argument.
14504 Any executable file format can carry a debug link, as long as it can
14505 contain a section named @code{.gnu_debuglink} with the contents
14508 @cindex @code{.note.gnu.build-id} sections
14509 @cindex build ID sections
14510 The build ID is a special section in the executable file (and in other
14511 ELF binary files that @value{GDBN} may consider). This section is
14512 often named @code{.note.gnu.build-id}, but that name is not mandatory.
14513 It contains unique identification for the built files---the ID remains
14514 the same across multiple builds of the same build tree. The default
14515 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
14516 content for the build ID string. The same section with an identical
14517 value is present in the original built binary with symbols, in its
14518 stripped variant, and in the separate debugging information file.
14520 The debugging information file itself should be an ordinary
14521 executable, containing a full set of linker symbols, sections, and
14522 debugging information. The sections of the debugging information file
14523 should have the same names, addresses, and sizes as the original file,
14524 but they need not contain any data---much like a @code{.bss} section
14525 in an ordinary executable.
14527 The @sc{gnu} binary utilities (Binutils) package includes the
14528 @samp{objcopy} utility that can produce
14529 the separated executable / debugging information file pairs using the
14530 following commands:
14533 @kbd{objcopy --only-keep-debug foo foo.debug}
14538 These commands remove the debugging
14539 information from the executable file @file{foo} and place it in the file
14540 @file{foo.debug}. You can use the first, second or both methods to link the
14545 The debug link method needs the following additional command to also leave
14546 behind a debug link in @file{foo}:
14549 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
14552 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
14553 a version of the @code{strip} command such that the command @kbd{strip foo -f
14554 foo.debug} has the same functionality as the two @code{objcopy} commands and
14555 the @code{ln -s} command above, together.
14558 Build ID gets embedded into the main executable using @code{ld --build-id} or
14559 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
14560 compatibility fixes for debug files separation are present in @sc{gnu} binary
14561 utilities (Binutils) package since version 2.18.
14566 @cindex CRC algorithm definition
14567 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
14568 IEEE 802.3 using the polynomial:
14570 @c TexInfo requires naked braces for multi-digit exponents for Tex
14571 @c output, but this causes HTML output to barf. HTML has to be set using
14572 @c raw commands. So we end up having to specify this equation in 2
14577 <em>x</em><sup>32</sup> + <em>x</em><sup>26</sup> + <em>x</em><sup>23</sup> + <em>x</em><sup>22</sup> + <em>x</em><sup>16</sup> + <em>x</em><sup>12</sup> + <em>x</em><sup>11</sup>
14578 + <em>x</em><sup>10</sup> + <em>x</em><sup>8</sup> + <em>x</em><sup>7</sup> + <em>x</em><sup>5</sup> + <em>x</em><sup>4</sup> + <em>x</em><sup>2</sup> + <em>x</em> + 1
14584 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
14585 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
14589 The function is computed byte at a time, taking the least
14590 significant bit of each byte first. The initial pattern
14591 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
14592 the final result is inverted to ensure trailing zeros also affect the
14595 @emph{Note:} This is the same CRC polynomial as used in handling the
14596 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
14597 , @value{GDBN} Remote Serial Protocol}). However in the
14598 case of the Remote Serial Protocol, the CRC is computed @emph{most}
14599 significant bit first, and the result is not inverted, so trailing
14600 zeros have no effect on the CRC value.
14602 To complete the description, we show below the code of the function
14603 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
14604 initially supplied @code{crc} argument means that an initial call to
14605 this function passing in zero will start computing the CRC using
14608 @kindex gnu_debuglink_crc32
14611 gnu_debuglink_crc32 (unsigned long crc,
14612 unsigned char *buf, size_t len)
14614 static const unsigned long crc32_table[256] =
14616 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
14617 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
14618 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
14619 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
14620 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
14621 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
14622 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
14623 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
14624 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
14625 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
14626 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
14627 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
14628 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
14629 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
14630 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
14631 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
14632 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
14633 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
14634 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
14635 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
14636 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
14637 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
14638 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
14639 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
14640 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
14641 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
14642 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
14643 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
14644 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
14645 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
14646 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
14647 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
14648 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
14649 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
14650 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
14651 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
14652 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
14653 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
14654 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
14655 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
14656 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
14657 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
14658 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
14659 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
14660 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
14661 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
14662 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
14663 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
14664 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
14665 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
14666 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
14669 unsigned char *end;
14671 crc = ~crc & 0xffffffff;
14672 for (end = buf + len; buf < end; ++buf)
14673 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
14674 return ~crc & 0xffffffff;
14679 This computation does not apply to the ``build ID'' method.
14682 @node Symbol Errors
14683 @section Errors Reading Symbol Files
14685 While reading a symbol file, @value{GDBN} occasionally encounters problems,
14686 such as symbol types it does not recognize, or known bugs in compiler
14687 output. By default, @value{GDBN} does not notify you of such problems, since
14688 they are relatively common and primarily of interest to people
14689 debugging compilers. If you are interested in seeing information
14690 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
14691 only one message about each such type of problem, no matter how many
14692 times the problem occurs; or you can ask @value{GDBN} to print more messages,
14693 to see how many times the problems occur, with the @code{set
14694 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
14697 The messages currently printed, and their meanings, include:
14700 @item inner block not inside outer block in @var{symbol}
14702 The symbol information shows where symbol scopes begin and end
14703 (such as at the start of a function or a block of statements). This
14704 error indicates that an inner scope block is not fully contained
14705 in its outer scope blocks.
14707 @value{GDBN} circumvents the problem by treating the inner block as if it had
14708 the same scope as the outer block. In the error message, @var{symbol}
14709 may be shown as ``@code{(don't know)}'' if the outer block is not a
14712 @item block at @var{address} out of order
14714 The symbol information for symbol scope blocks should occur in
14715 order of increasing addresses. This error indicates that it does not
14718 @value{GDBN} does not circumvent this problem, and has trouble
14719 locating symbols in the source file whose symbols it is reading. (You
14720 can often determine what source file is affected by specifying
14721 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
14724 @item bad block start address patched
14726 The symbol information for a symbol scope block has a start address
14727 smaller than the address of the preceding source line. This is known
14728 to occur in the SunOS 4.1.1 (and earlier) C compiler.
14730 @value{GDBN} circumvents the problem by treating the symbol scope block as
14731 starting on the previous source line.
14733 @item bad string table offset in symbol @var{n}
14736 Symbol number @var{n} contains a pointer into the string table which is
14737 larger than the size of the string table.
14739 @value{GDBN} circumvents the problem by considering the symbol to have the
14740 name @code{foo}, which may cause other problems if many symbols end up
14743 @item unknown symbol type @code{0x@var{nn}}
14745 The symbol information contains new data types that @value{GDBN} does
14746 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
14747 uncomprehended information, in hexadecimal.
14749 @value{GDBN} circumvents the error by ignoring this symbol information.
14750 This usually allows you to debug your program, though certain symbols
14751 are not accessible. If you encounter such a problem and feel like
14752 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
14753 on @code{complain}, then go up to the function @code{read_dbx_symtab}
14754 and examine @code{*bufp} to see the symbol.
14756 @item stub type has NULL name
14758 @value{GDBN} could not find the full definition for a struct or class.
14760 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
14761 The symbol information for a C@t{++} member function is missing some
14762 information that recent versions of the compiler should have output for
14765 @item info mismatch between compiler and debugger
14767 @value{GDBN} could not parse a type specification output by the compiler.
14772 @section GDB Data Files
14774 @cindex prefix for data files
14775 @value{GDBN} will sometimes read an auxiliary data file. These files
14776 are kept in a directory known as the @dfn{data directory}.
14778 You can set the data directory's name, and view the name @value{GDBN}
14779 is currently using.
14782 @kindex set data-directory
14783 @item set data-directory @var{directory}
14784 Set the directory which @value{GDBN} searches for auxiliary data files
14785 to @var{directory}.
14787 @kindex show data-directory
14788 @item show data-directory
14789 Show the directory @value{GDBN} searches for auxiliary data files.
14792 @cindex default data directory
14793 @cindex @samp{--with-gdb-datadir}
14794 You can set the default data directory by using the configure-time
14795 @samp{--with-gdb-datadir} option. If the data directory is inside
14796 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
14797 @samp{--exec-prefix}), then the default data directory will be updated
14798 automatically if the installed @value{GDBN} is moved to a new
14802 @chapter Specifying a Debugging Target
14804 @cindex debugging target
14805 A @dfn{target} is the execution environment occupied by your program.
14807 Often, @value{GDBN} runs in the same host environment as your program;
14808 in that case, the debugging target is specified as a side effect when
14809 you use the @code{file} or @code{core} commands. When you need more
14810 flexibility---for example, running @value{GDBN} on a physically separate
14811 host, or controlling a standalone system over a serial port or a
14812 realtime system over a TCP/IP connection---you can use the @code{target}
14813 command to specify one of the target types configured for @value{GDBN}
14814 (@pxref{Target Commands, ,Commands for Managing Targets}).
14816 @cindex target architecture
14817 It is possible to build @value{GDBN} for several different @dfn{target
14818 architectures}. When @value{GDBN} is built like that, you can choose
14819 one of the available architectures with the @kbd{set architecture}
14823 @kindex set architecture
14824 @kindex show architecture
14825 @item set architecture @var{arch}
14826 This command sets the current target architecture to @var{arch}. The
14827 value of @var{arch} can be @code{"auto"}, in addition to one of the
14828 supported architectures.
14830 @item show architecture
14831 Show the current target architecture.
14833 @item set processor
14835 @kindex set processor
14836 @kindex show processor
14837 These are alias commands for, respectively, @code{set architecture}
14838 and @code{show architecture}.
14842 * Active Targets:: Active targets
14843 * Target Commands:: Commands for managing targets
14844 * Byte Order:: Choosing target byte order
14847 @node Active Targets
14848 @section Active Targets
14850 @cindex stacking targets
14851 @cindex active targets
14852 @cindex multiple targets
14854 There are three classes of targets: processes, core files, and
14855 executable files. @value{GDBN} can work concurrently on up to three
14856 active targets, one in each class. This allows you to (for example)
14857 start a process and inspect its activity without abandoning your work on
14860 For example, if you execute @samp{gdb a.out}, then the executable file
14861 @code{a.out} is the only active target. If you designate a core file as
14862 well---presumably from a prior run that crashed and coredumped---then
14863 @value{GDBN} has two active targets and uses them in tandem, looking
14864 first in the corefile target, then in the executable file, to satisfy
14865 requests for memory addresses. (Typically, these two classes of target
14866 are complementary, since core files contain only a program's
14867 read-write memory---variables and so on---plus machine status, while
14868 executable files contain only the program text and initialized data.)
14870 When you type @code{run}, your executable file becomes an active process
14871 target as well. When a process target is active, all @value{GDBN}
14872 commands requesting memory addresses refer to that target; addresses in
14873 an active core file or executable file target are obscured while the
14874 process target is active.
14876 Use the @code{core-file} and @code{exec-file} commands to select a new
14877 core file or executable target (@pxref{Files, ,Commands to Specify
14878 Files}). To specify as a target a process that is already running, use
14879 the @code{attach} command (@pxref{Attach, ,Debugging an Already-running
14882 @node Target Commands
14883 @section Commands for Managing Targets
14886 @item target @var{type} @var{parameters}
14887 Connects the @value{GDBN} host environment to a target machine or
14888 process. A target is typically a protocol for talking to debugging
14889 facilities. You use the argument @var{type} to specify the type or
14890 protocol of the target machine.
14892 Further @var{parameters} are interpreted by the target protocol, but
14893 typically include things like device names or host names to connect
14894 with, process numbers, and baud rates.
14896 The @code{target} command does not repeat if you press @key{RET} again
14897 after executing the command.
14899 @kindex help target
14901 Displays the names of all targets available. To display targets
14902 currently selected, use either @code{info target} or @code{info files}
14903 (@pxref{Files, ,Commands to Specify Files}).
14905 @item help target @var{name}
14906 Describe a particular target, including any parameters necessary to
14909 @kindex set gnutarget
14910 @item set gnutarget @var{args}
14911 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
14912 knows whether it is reading an @dfn{executable},
14913 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
14914 with the @code{set gnutarget} command. Unlike most @code{target} commands,
14915 with @code{gnutarget} the @code{target} refers to a program, not a machine.
14918 @emph{Warning:} To specify a file format with @code{set gnutarget},
14919 you must know the actual BFD name.
14923 @xref{Files, , Commands to Specify Files}.
14925 @kindex show gnutarget
14926 @item show gnutarget
14927 Use the @code{show gnutarget} command to display what file format
14928 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
14929 @value{GDBN} will determine the file format for each file automatically,
14930 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
14933 @cindex common targets
14934 Here are some common targets (available, or not, depending on the GDB
14939 @item target exec @var{program}
14940 @cindex executable file target
14941 An executable file. @samp{target exec @var{program}} is the same as
14942 @samp{exec-file @var{program}}.
14944 @item target core @var{filename}
14945 @cindex core dump file target
14946 A core dump file. @samp{target core @var{filename}} is the same as
14947 @samp{core-file @var{filename}}.
14949 @item target remote @var{medium}
14950 @cindex remote target
14951 A remote system connected to @value{GDBN} via a serial line or network
14952 connection. This command tells @value{GDBN} to use its own remote
14953 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
14955 For example, if you have a board connected to @file{/dev/ttya} on the
14956 machine running @value{GDBN}, you could say:
14959 target remote /dev/ttya
14962 @code{target remote} supports the @code{load} command. This is only
14963 useful if you have some other way of getting the stub to the target
14964 system, and you can put it somewhere in memory where it won't get
14965 clobbered by the download.
14967 @item target sim @r{[}@var{simargs}@r{]} @dots{}
14968 @cindex built-in simulator target
14969 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
14977 works; however, you cannot assume that a specific memory map, device
14978 drivers, or even basic I/O is available, although some simulators do
14979 provide these. For info about any processor-specific simulator details,
14980 see the appropriate section in @ref{Embedded Processors, ,Embedded
14985 Some configurations may include these targets as well:
14989 @item target nrom @var{dev}
14990 @cindex NetROM ROM emulator target
14991 NetROM ROM emulator. This target only supports downloading.
14995 Different targets are available on different configurations of @value{GDBN};
14996 your configuration may have more or fewer targets.
14998 Many remote targets require you to download the executable's code once
14999 you've successfully established a connection. You may wish to control
15000 various aspects of this process.
15005 @kindex set hash@r{, for remote monitors}
15006 @cindex hash mark while downloading
15007 This command controls whether a hash mark @samp{#} is displayed while
15008 downloading a file to the remote monitor. If on, a hash mark is
15009 displayed after each S-record is successfully downloaded to the
15013 @kindex show hash@r{, for remote monitors}
15014 Show the current status of displaying the hash mark.
15016 @item set debug monitor
15017 @kindex set debug monitor
15018 @cindex display remote monitor communications
15019 Enable or disable display of communications messages between
15020 @value{GDBN} and the remote monitor.
15022 @item show debug monitor
15023 @kindex show debug monitor
15024 Show the current status of displaying communications between
15025 @value{GDBN} and the remote monitor.
15030 @kindex load @var{filename}
15031 @item load @var{filename}
15033 Depending on what remote debugging facilities are configured into
15034 @value{GDBN}, the @code{load} command may be available. Where it exists, it
15035 is meant to make @var{filename} (an executable) available for debugging
15036 on the remote system---by downloading, or dynamic linking, for example.
15037 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
15038 the @code{add-symbol-file} command.
15040 If your @value{GDBN} does not have a @code{load} command, attempting to
15041 execute it gets the error message ``@code{You can't do that when your
15042 target is @dots{}}''
15044 The file is loaded at whatever address is specified in the executable.
15045 For some object file formats, you can specify the load address when you
15046 link the program; for other formats, like a.out, the object file format
15047 specifies a fixed address.
15048 @c FIXME! This would be a good place for an xref to the GNU linker doc.
15050 Depending on the remote side capabilities, @value{GDBN} may be able to
15051 load programs into flash memory.
15053 @code{load} does not repeat if you press @key{RET} again after using it.
15057 @section Choosing Target Byte Order
15059 @cindex choosing target byte order
15060 @cindex target byte order
15062 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
15063 offer the ability to run either big-endian or little-endian byte
15064 orders. Usually the executable or symbol will include a bit to
15065 designate the endian-ness, and you will not need to worry about
15066 which to use. However, you may still find it useful to adjust
15067 @value{GDBN}'s idea of processor endian-ness manually.
15071 @item set endian big
15072 Instruct @value{GDBN} to assume the target is big-endian.
15074 @item set endian little
15075 Instruct @value{GDBN} to assume the target is little-endian.
15077 @item set endian auto
15078 Instruct @value{GDBN} to use the byte order associated with the
15082 Display @value{GDBN}'s current idea of the target byte order.
15086 Note that these commands merely adjust interpretation of symbolic
15087 data on the host, and that they have absolutely no effect on the
15091 @node Remote Debugging
15092 @chapter Debugging Remote Programs
15093 @cindex remote debugging
15095 If you are trying to debug a program running on a machine that cannot run
15096 @value{GDBN} in the usual way, it is often useful to use remote debugging.
15097 For example, you might use remote debugging on an operating system kernel,
15098 or on a small system which does not have a general purpose operating system
15099 powerful enough to run a full-featured debugger.
15101 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
15102 to make this work with particular debugging targets. In addition,
15103 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
15104 but not specific to any particular target system) which you can use if you
15105 write the remote stubs---the code that runs on the remote system to
15106 communicate with @value{GDBN}.
15108 Other remote targets may be available in your
15109 configuration of @value{GDBN}; use @code{help target} to list them.
15112 * Connecting:: Connecting to a remote target
15113 * File Transfer:: Sending files to a remote system
15114 * Server:: Using the gdbserver program
15115 * Remote Configuration:: Remote configuration
15116 * Remote Stub:: Implementing a remote stub
15120 @section Connecting to a Remote Target
15122 On the @value{GDBN} host machine, you will need an unstripped copy of
15123 your program, since @value{GDBN} needs symbol and debugging information.
15124 Start up @value{GDBN} as usual, using the name of the local copy of your
15125 program as the first argument.
15127 @cindex @code{target remote}
15128 @value{GDBN} can communicate with the target over a serial line, or
15129 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
15130 each case, @value{GDBN} uses the same protocol for debugging your
15131 program; only the medium carrying the debugging packets varies. The
15132 @code{target remote} command establishes a connection to the target.
15133 Its arguments indicate which medium to use:
15137 @item target remote @var{serial-device}
15138 @cindex serial line, @code{target remote}
15139 Use @var{serial-device} to communicate with the target. For example,
15140 to use a serial line connected to the device named @file{/dev/ttyb}:
15143 target remote /dev/ttyb
15146 If you're using a serial line, you may want to give @value{GDBN} the
15147 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
15148 (@pxref{Remote Configuration, set remotebaud}) before the
15149 @code{target} command.
15151 @item target remote @code{@var{host}:@var{port}}
15152 @itemx target remote @code{tcp:@var{host}:@var{port}}
15153 @cindex @acronym{TCP} port, @code{target remote}
15154 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
15155 The @var{host} may be either a host name or a numeric @acronym{IP}
15156 address; @var{port} must be a decimal number. The @var{host} could be
15157 the target machine itself, if it is directly connected to the net, or
15158 it might be a terminal server which in turn has a serial line to the
15161 For example, to connect to port 2828 on a terminal server named
15165 target remote manyfarms:2828
15168 If your remote target is actually running on the same machine as your
15169 debugger session (e.g.@: a simulator for your target running on the
15170 same host), you can omit the hostname. For example, to connect to
15171 port 1234 on your local machine:
15174 target remote :1234
15178 Note that the colon is still required here.
15180 @item target remote @code{udp:@var{host}:@var{port}}
15181 @cindex @acronym{UDP} port, @code{target remote}
15182 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
15183 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
15186 target remote udp:manyfarms:2828
15189 When using a @acronym{UDP} connection for remote debugging, you should
15190 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
15191 can silently drop packets on busy or unreliable networks, which will
15192 cause havoc with your debugging session.
15194 @item target remote | @var{command}
15195 @cindex pipe, @code{target remote} to
15196 Run @var{command} in the background and communicate with it using a
15197 pipe. The @var{command} is a shell command, to be parsed and expanded
15198 by the system's command shell, @code{/bin/sh}; it should expect remote
15199 protocol packets on its standard input, and send replies on its
15200 standard output. You could use this to run a stand-alone simulator
15201 that speaks the remote debugging protocol, to make net connections
15202 using programs like @code{ssh}, or for other similar tricks.
15204 If @var{command} closes its standard output (perhaps by exiting),
15205 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
15206 program has already exited, this will have no effect.)
15210 Once the connection has been established, you can use all the usual
15211 commands to examine and change data. The remote program is already
15212 running; you can use @kbd{step} and @kbd{continue}, and you do not
15213 need to use @kbd{run}.
15215 @cindex interrupting remote programs
15216 @cindex remote programs, interrupting
15217 Whenever @value{GDBN} is waiting for the remote program, if you type the
15218 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
15219 program. This may or may not succeed, depending in part on the hardware
15220 and the serial drivers the remote system uses. If you type the
15221 interrupt character once again, @value{GDBN} displays this prompt:
15224 Interrupted while waiting for the program.
15225 Give up (and stop debugging it)? (y or n)
15228 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
15229 (If you decide you want to try again later, you can use @samp{target
15230 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
15231 goes back to waiting.
15234 @kindex detach (remote)
15236 When you have finished debugging the remote program, you can use the
15237 @code{detach} command to release it from @value{GDBN} control.
15238 Detaching from the target normally resumes its execution, but the results
15239 will depend on your particular remote stub. After the @code{detach}
15240 command, @value{GDBN} is free to connect to another target.
15244 The @code{disconnect} command behaves like @code{detach}, except that
15245 the target is generally not resumed. It will wait for @value{GDBN}
15246 (this instance or another one) to connect and continue debugging. After
15247 the @code{disconnect} command, @value{GDBN} is again free to connect to
15250 @cindex send command to remote monitor
15251 @cindex extend @value{GDBN} for remote targets
15252 @cindex add new commands for external monitor
15254 @item monitor @var{cmd}
15255 This command allows you to send arbitrary commands directly to the
15256 remote monitor. Since @value{GDBN} doesn't care about the commands it
15257 sends like this, this command is the way to extend @value{GDBN}---you
15258 can add new commands that only the external monitor will understand
15262 @node File Transfer
15263 @section Sending files to a remote system
15264 @cindex remote target, file transfer
15265 @cindex file transfer
15266 @cindex sending files to remote systems
15268 Some remote targets offer the ability to transfer files over the same
15269 connection used to communicate with @value{GDBN}. This is convenient
15270 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
15271 running @code{gdbserver} over a network interface. For other targets,
15272 e.g.@: embedded devices with only a single serial port, this may be
15273 the only way to upload or download files.
15275 Not all remote targets support these commands.
15279 @item remote put @var{hostfile} @var{targetfile}
15280 Copy file @var{hostfile} from the host system (the machine running
15281 @value{GDBN}) to @var{targetfile} on the target system.
15284 @item remote get @var{targetfile} @var{hostfile}
15285 Copy file @var{targetfile} from the target system to @var{hostfile}
15286 on the host system.
15288 @kindex remote delete
15289 @item remote delete @var{targetfile}
15290 Delete @var{targetfile} from the target system.
15295 @section Using the @code{gdbserver} Program
15298 @cindex remote connection without stubs
15299 @code{gdbserver} is a control program for Unix-like systems, which
15300 allows you to connect your program with a remote @value{GDBN} via
15301 @code{target remote}---but without linking in the usual debugging stub.
15303 @code{gdbserver} is not a complete replacement for the debugging stubs,
15304 because it requires essentially the same operating-system facilities
15305 that @value{GDBN} itself does. In fact, a system that can run
15306 @code{gdbserver} to connect to a remote @value{GDBN} could also run
15307 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
15308 because it is a much smaller program than @value{GDBN} itself. It is
15309 also easier to port than all of @value{GDBN}, so you may be able to get
15310 started more quickly on a new system by using @code{gdbserver}.
15311 Finally, if you develop code for real-time systems, you may find that
15312 the tradeoffs involved in real-time operation make it more convenient to
15313 do as much development work as possible on another system, for example
15314 by cross-compiling. You can use @code{gdbserver} to make a similar
15315 choice for debugging.
15317 @value{GDBN} and @code{gdbserver} communicate via either a serial line
15318 or a TCP connection, using the standard @value{GDBN} remote serial
15322 @emph{Warning:} @code{gdbserver} does not have any built-in security.
15323 Do not run @code{gdbserver} connected to any public network; a
15324 @value{GDBN} connection to @code{gdbserver} provides access to the
15325 target system with the same privileges as the user running
15329 @subsection Running @code{gdbserver}
15330 @cindex arguments, to @code{gdbserver}
15332 Run @code{gdbserver} on the target system. You need a copy of the
15333 program you want to debug, including any libraries it requires.
15334 @code{gdbserver} does not need your program's symbol table, so you can
15335 strip the program if necessary to save space. @value{GDBN} on the host
15336 system does all the symbol handling.
15338 To use the server, you must tell it how to communicate with @value{GDBN};
15339 the name of your program; and the arguments for your program. The usual
15343 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
15346 @var{comm} is either a device name (to use a serial line) or a TCP
15347 hostname and portnumber. For example, to debug Emacs with the argument
15348 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
15352 target> gdbserver /dev/com1 emacs foo.txt
15355 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
15358 To use a TCP connection instead of a serial line:
15361 target> gdbserver host:2345 emacs foo.txt
15364 The only difference from the previous example is the first argument,
15365 specifying that you are communicating with the host @value{GDBN} via
15366 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
15367 expect a TCP connection from machine @samp{host} to local TCP port 2345.
15368 (Currently, the @samp{host} part is ignored.) You can choose any number
15369 you want for the port number as long as it does not conflict with any
15370 TCP ports already in use on the target system (for example, @code{23} is
15371 reserved for @code{telnet}).@footnote{If you choose a port number that
15372 conflicts with another service, @code{gdbserver} prints an error message
15373 and exits.} You must use the same port number with the host @value{GDBN}
15374 @code{target remote} command.
15376 @subsubsection Attaching to a Running Program
15378 On some targets, @code{gdbserver} can also attach to running programs.
15379 This is accomplished via the @code{--attach} argument. The syntax is:
15382 target> gdbserver --attach @var{comm} @var{pid}
15385 @var{pid} is the process ID of a currently running process. It isn't necessary
15386 to point @code{gdbserver} at a binary for the running process.
15389 @cindex attach to a program by name
15390 You can debug processes by name instead of process ID if your target has the
15391 @code{pidof} utility:
15394 target> gdbserver --attach @var{comm} `pidof @var{program}`
15397 In case more than one copy of @var{program} is running, or @var{program}
15398 has multiple threads, most versions of @code{pidof} support the
15399 @code{-s} option to only return the first process ID.
15401 @subsubsection Multi-Process Mode for @code{gdbserver}
15402 @cindex gdbserver, multiple processes
15403 @cindex multiple processes with gdbserver
15405 When you connect to @code{gdbserver} using @code{target remote},
15406 @code{gdbserver} debugs the specified program only once. When the
15407 program exits, or you detach from it, @value{GDBN} closes the connection
15408 and @code{gdbserver} exits.
15410 If you connect using @kbd{target extended-remote}, @code{gdbserver}
15411 enters multi-process mode. When the debugged program exits, or you
15412 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
15413 though no program is running. The @code{run} and @code{attach}
15414 commands instruct @code{gdbserver} to run or attach to a new program.
15415 The @code{run} command uses @code{set remote exec-file} (@pxref{set
15416 remote exec-file}) to select the program to run. Command line
15417 arguments are supported, except for wildcard expansion and I/O
15418 redirection (@pxref{Arguments}).
15420 To start @code{gdbserver} without supplying an initial command to run
15421 or process ID to attach, use the @option{--multi} command line option.
15422 Then you can connect using @kbd{target extended-remote} and start
15423 the program you want to debug.
15425 @code{gdbserver} does not automatically exit in multi-process mode.
15426 You can terminate it by using @code{monitor exit}
15427 (@pxref{Monitor Commands for gdbserver}).
15429 @subsubsection Other Command-Line Arguments for @code{gdbserver}
15431 The @option{--debug} option tells @code{gdbserver} to display extra
15432 status information about the debugging process. The
15433 @option{--remote-debug} option tells @code{gdbserver} to display
15434 remote protocol debug output. These options are intended for
15435 @code{gdbserver} development and for bug reports to the developers.
15437 The @option{--wrapper} option specifies a wrapper to launch programs
15438 for debugging. The option should be followed by the name of the
15439 wrapper, then any command-line arguments to pass to the wrapper, then
15440 @kbd{--} indicating the end of the wrapper arguments.
15442 @code{gdbserver} runs the specified wrapper program with a combined
15443 command line including the wrapper arguments, then the name of the
15444 program to debug, then any arguments to the program. The wrapper
15445 runs until it executes your program, and then @value{GDBN} gains control.
15447 You can use any program that eventually calls @code{execve} with
15448 its arguments as a wrapper. Several standard Unix utilities do
15449 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
15450 with @code{exec "$@@"} will also work.
15452 For example, you can use @code{env} to pass an environment variable to
15453 the debugged program, without setting the variable in @code{gdbserver}'s
15457 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
15460 @subsection Connecting to @code{gdbserver}
15462 Run @value{GDBN} on the host system.
15464 First make sure you have the necessary symbol files. Load symbols for
15465 your application using the @code{file} command before you connect. Use
15466 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
15467 was compiled with the correct sysroot using @code{--with-sysroot}).
15469 The symbol file and target libraries must exactly match the executable
15470 and libraries on the target, with one exception: the files on the host
15471 system should not be stripped, even if the files on the target system
15472 are. Mismatched or missing files will lead to confusing results
15473 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
15474 files may also prevent @code{gdbserver} from debugging multi-threaded
15477 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
15478 For TCP connections, you must start up @code{gdbserver} prior to using
15479 the @code{target remote} command. Otherwise you may get an error whose
15480 text depends on the host system, but which usually looks something like
15481 @samp{Connection refused}. Don't use the @code{load}
15482 command in @value{GDBN} when using @code{gdbserver}, since the program is
15483 already on the target.
15485 @subsection Monitor Commands for @code{gdbserver}
15486 @cindex monitor commands, for @code{gdbserver}
15487 @anchor{Monitor Commands for gdbserver}
15489 During a @value{GDBN} session using @code{gdbserver}, you can use the
15490 @code{monitor} command to send special requests to @code{gdbserver}.
15491 Here are the available commands.
15495 List the available monitor commands.
15497 @item monitor set debug 0
15498 @itemx monitor set debug 1
15499 Disable or enable general debugging messages.
15501 @item monitor set remote-debug 0
15502 @itemx monitor set remote-debug 1
15503 Disable or enable specific debugging messages associated with the remote
15504 protocol (@pxref{Remote Protocol}).
15506 @item monitor set libthread-db-search-path [PATH]
15507 @cindex gdbserver, search path for @code{libthread_db}
15508 When this command is issued, @var{path} is a colon-separated list of
15509 directories to search for @code{libthread_db} (@pxref{Threads,,set
15510 libthread-db-search-path}). If you omit @var{path},
15511 @samp{libthread-db-search-path} will be reset to an empty list.
15514 Tell gdbserver to exit immediately. This command should be followed by
15515 @code{disconnect} to close the debugging session. @code{gdbserver} will
15516 detach from any attached processes and kill any processes it created.
15517 Use @code{monitor exit} to terminate @code{gdbserver} at the end
15518 of a multi-process mode debug session.
15522 @node Remote Configuration
15523 @section Remote Configuration
15526 @kindex show remote
15527 This section documents the configuration options available when
15528 debugging remote programs. For the options related to the File I/O
15529 extensions of the remote protocol, see @ref{system,
15530 system-call-allowed}.
15533 @item set remoteaddresssize @var{bits}
15534 @cindex address size for remote targets
15535 @cindex bits in remote address
15536 Set the maximum size of address in a memory packet to the specified
15537 number of bits. @value{GDBN} will mask off the address bits above
15538 that number, when it passes addresses to the remote target. The
15539 default value is the number of bits in the target's address.
15541 @item show remoteaddresssize
15542 Show the current value of remote address size in bits.
15544 @item set remotebaud @var{n}
15545 @cindex baud rate for remote targets
15546 Set the baud rate for the remote serial I/O to @var{n} baud. The
15547 value is used to set the speed of the serial port used for debugging
15550 @item show remotebaud
15551 Show the current speed of the remote connection.
15553 @item set remotebreak
15554 @cindex interrupt remote programs
15555 @cindex BREAK signal instead of Ctrl-C
15556 @anchor{set remotebreak}
15557 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
15558 when you type @kbd{Ctrl-c} to interrupt the program running
15559 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
15560 character instead. The default is off, since most remote systems
15561 expect to see @samp{Ctrl-C} as the interrupt signal.
15563 @item show remotebreak
15564 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
15565 interrupt the remote program.
15567 @item set remoteflow on
15568 @itemx set remoteflow off
15569 @kindex set remoteflow
15570 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
15571 on the serial port used to communicate to the remote target.
15573 @item show remoteflow
15574 @kindex show remoteflow
15575 Show the current setting of hardware flow control.
15577 @item set remotelogbase @var{base}
15578 Set the base (a.k.a.@: radix) of logging serial protocol
15579 communications to @var{base}. Supported values of @var{base} are:
15580 @code{ascii}, @code{octal}, and @code{hex}. The default is
15583 @item show remotelogbase
15584 Show the current setting of the radix for logging remote serial
15587 @item set remotelogfile @var{file}
15588 @cindex record serial communications on file
15589 Record remote serial communications on the named @var{file}. The
15590 default is not to record at all.
15592 @item show remotelogfile.
15593 Show the current setting of the file name on which to record the
15594 serial communications.
15596 @item set remotetimeout @var{num}
15597 @cindex timeout for serial communications
15598 @cindex remote timeout
15599 Set the timeout limit to wait for the remote target to respond to
15600 @var{num} seconds. The default is 2 seconds.
15602 @item show remotetimeout
15603 Show the current number of seconds to wait for the remote target
15606 @cindex limit hardware breakpoints and watchpoints
15607 @cindex remote target, limit break- and watchpoints
15608 @anchor{set remote hardware-watchpoint-limit}
15609 @anchor{set remote hardware-breakpoint-limit}
15610 @item set remote hardware-watchpoint-limit @var{limit}
15611 @itemx set remote hardware-breakpoint-limit @var{limit}
15612 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
15613 watchpoints. A limit of -1, the default, is treated as unlimited.
15615 @item set remote exec-file @var{filename}
15616 @itemx show remote exec-file
15617 @anchor{set remote exec-file}
15618 @cindex executable file, for remote target
15619 Select the file used for @code{run} with @code{target
15620 extended-remote}. This should be set to a filename valid on the
15621 target system. If it is not set, the target will use a default
15622 filename (e.g.@: the last program run).
15624 @item set remote interrupt-sequence
15625 @cindex interrupt remote programs
15626 @cindex select Ctrl-C, BREAK or BREAK-g
15627 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
15628 @samp{BREAK-g} as the
15629 sequence to the remote target in order to interrupt the execution.
15630 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
15631 is high level of serial line for some certain time.
15632 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
15633 It is @code{BREAK} signal followed by character @code{g}.
15635 @item show interrupt-sequence
15636 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
15637 is sent by @value{GDBN} to interrupt the remote program.
15638 @code{BREAK-g} is BREAK signal followed by @code{g} and
15639 also known as Magic SysRq g.
15641 @item set remote interrupt-on-connect
15642 @cindex send interrupt-sequence on start
15643 Specify whether interrupt-sequence is sent to remote target when
15644 @value{GDBN} connects to it. This is mostly needed when you debug
15645 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
15646 which is known as Magic SysRq g in order to connect @value{GDBN}.
15648 @item show interrupt-on-connect
15649 Show whether interrupt-sequence is sent
15650 to remote target when @value{GDBN} connects to it.
15654 @item set tcp auto-retry on
15655 @cindex auto-retry, for remote TCP target
15656 Enable auto-retry for remote TCP connections. This is useful if the remote
15657 debugging agent is launched in parallel with @value{GDBN}; there is a race
15658 condition because the agent may not become ready to accept the connection
15659 before @value{GDBN} attempts to connect. When auto-retry is
15660 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
15661 to establish the connection using the timeout specified by
15662 @code{set tcp connect-timeout}.
15664 @item set tcp auto-retry off
15665 Do not auto-retry failed TCP connections.
15667 @item show tcp auto-retry
15668 Show the current auto-retry setting.
15670 @item set tcp connect-timeout @var{seconds}
15671 @cindex connection timeout, for remote TCP target
15672 @cindex timeout, for remote target connection
15673 Set the timeout for establishing a TCP connection to the remote target to
15674 @var{seconds}. The timeout affects both polling to retry failed connections
15675 (enabled by @code{set tcp auto-retry on}) and waiting for connections
15676 that are merely slow to complete, and represents an approximate cumulative
15679 @item show tcp connect-timeout
15680 Show the current connection timeout setting.
15683 @cindex remote packets, enabling and disabling
15684 The @value{GDBN} remote protocol autodetects the packets supported by
15685 your debugging stub. If you need to override the autodetection, you
15686 can use these commands to enable or disable individual packets. Each
15687 packet can be set to @samp{on} (the remote target supports this
15688 packet), @samp{off} (the remote target does not support this packet),
15689 or @samp{auto} (detect remote target support for this packet). They
15690 all default to @samp{auto}. For more information about each packet,
15691 see @ref{Remote Protocol}.
15693 During normal use, you should not have to use any of these commands.
15694 If you do, that may be a bug in your remote debugging stub, or a bug
15695 in @value{GDBN}. You may want to report the problem to the
15696 @value{GDBN} developers.
15698 For each packet @var{name}, the command to enable or disable the
15699 packet is @code{set remote @var{name}-packet}. The available settings
15702 @multitable @columnfractions 0.28 0.32 0.25
15705 @tab Related Features
15707 @item @code{fetch-register}
15709 @tab @code{info registers}
15711 @item @code{set-register}
15715 @item @code{binary-download}
15717 @tab @code{load}, @code{set}
15719 @item @code{read-aux-vector}
15720 @tab @code{qXfer:auxv:read}
15721 @tab @code{info auxv}
15723 @item @code{symbol-lookup}
15724 @tab @code{qSymbol}
15725 @tab Detecting multiple threads
15727 @item @code{attach}
15728 @tab @code{vAttach}
15731 @item @code{verbose-resume}
15733 @tab Stepping or resuming multiple threads
15739 @item @code{software-breakpoint}
15743 @item @code{hardware-breakpoint}
15747 @item @code{write-watchpoint}
15751 @item @code{read-watchpoint}
15755 @item @code{access-watchpoint}
15759 @item @code{target-features}
15760 @tab @code{qXfer:features:read}
15761 @tab @code{set architecture}
15763 @item @code{library-info}
15764 @tab @code{qXfer:libraries:read}
15765 @tab @code{info sharedlibrary}
15767 @item @code{memory-map}
15768 @tab @code{qXfer:memory-map:read}
15769 @tab @code{info mem}
15771 @item @code{read-spu-object}
15772 @tab @code{qXfer:spu:read}
15773 @tab @code{info spu}
15775 @item @code{write-spu-object}
15776 @tab @code{qXfer:spu:write}
15777 @tab @code{info spu}
15779 @item @code{read-siginfo-object}
15780 @tab @code{qXfer:siginfo:read}
15781 @tab @code{print $_siginfo}
15783 @item @code{write-siginfo-object}
15784 @tab @code{qXfer:siginfo:write}
15785 @tab @code{set $_siginfo}
15787 @item @code{threads}
15788 @tab @code{qXfer:threads:read}
15789 @tab @code{info threads}
15791 @item @code{get-thread-local-@*storage-address}
15792 @tab @code{qGetTLSAddr}
15793 @tab Displaying @code{__thread} variables
15795 @item @code{get-thread-information-block-address}
15796 @tab @code{qGetTIBAddr}
15797 @tab Display MS-Windows Thread Information Block.
15799 @item @code{search-memory}
15800 @tab @code{qSearch:memory}
15803 @item @code{supported-packets}
15804 @tab @code{qSupported}
15805 @tab Remote communications parameters
15807 @item @code{pass-signals}
15808 @tab @code{QPassSignals}
15809 @tab @code{handle @var{signal}}
15811 @item @code{hostio-close-packet}
15812 @tab @code{vFile:close}
15813 @tab @code{remote get}, @code{remote put}
15815 @item @code{hostio-open-packet}
15816 @tab @code{vFile:open}
15817 @tab @code{remote get}, @code{remote put}
15819 @item @code{hostio-pread-packet}
15820 @tab @code{vFile:pread}
15821 @tab @code{remote get}, @code{remote put}
15823 @item @code{hostio-pwrite-packet}
15824 @tab @code{vFile:pwrite}
15825 @tab @code{remote get}, @code{remote put}
15827 @item @code{hostio-unlink-packet}
15828 @tab @code{vFile:unlink}
15829 @tab @code{remote delete}
15831 @item @code{noack-packet}
15832 @tab @code{QStartNoAckMode}
15833 @tab Packet acknowledgment
15835 @item @code{osdata}
15836 @tab @code{qXfer:osdata:read}
15837 @tab @code{info os}
15839 @item @code{query-attached}
15840 @tab @code{qAttached}
15841 @tab Querying remote process attach state.
15845 @section Implementing a Remote Stub
15847 @cindex debugging stub, example
15848 @cindex remote stub, example
15849 @cindex stub example, remote debugging
15850 The stub files provided with @value{GDBN} implement the target side of the
15851 communication protocol, and the @value{GDBN} side is implemented in the
15852 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
15853 these subroutines to communicate, and ignore the details. (If you're
15854 implementing your own stub file, you can still ignore the details: start
15855 with one of the existing stub files. @file{sparc-stub.c} is the best
15856 organized, and therefore the easiest to read.)
15858 @cindex remote serial debugging, overview
15859 To debug a program running on another machine (the debugging
15860 @dfn{target} machine), you must first arrange for all the usual
15861 prerequisites for the program to run by itself. For example, for a C
15866 A startup routine to set up the C runtime environment; these usually
15867 have a name like @file{crt0}. The startup routine may be supplied by
15868 your hardware supplier, or you may have to write your own.
15871 A C subroutine library to support your program's
15872 subroutine calls, notably managing input and output.
15875 A way of getting your program to the other machine---for example, a
15876 download program. These are often supplied by the hardware
15877 manufacturer, but you may have to write your own from hardware
15881 The next step is to arrange for your program to use a serial port to
15882 communicate with the machine where @value{GDBN} is running (the @dfn{host}
15883 machine). In general terms, the scheme looks like this:
15887 @value{GDBN} already understands how to use this protocol; when everything
15888 else is set up, you can simply use the @samp{target remote} command
15889 (@pxref{Targets,,Specifying a Debugging Target}).
15891 @item On the target,
15892 you must link with your program a few special-purpose subroutines that
15893 implement the @value{GDBN} remote serial protocol. The file containing these
15894 subroutines is called a @dfn{debugging stub}.
15896 On certain remote targets, you can use an auxiliary program
15897 @code{gdbserver} instead of linking a stub into your program.
15898 @xref{Server,,Using the @code{gdbserver} Program}, for details.
15901 The debugging stub is specific to the architecture of the remote
15902 machine; for example, use @file{sparc-stub.c} to debug programs on
15905 @cindex remote serial stub list
15906 These working remote stubs are distributed with @value{GDBN}:
15911 @cindex @file{i386-stub.c}
15914 For Intel 386 and compatible architectures.
15917 @cindex @file{m68k-stub.c}
15918 @cindex Motorola 680x0
15920 For Motorola 680x0 architectures.
15923 @cindex @file{sh-stub.c}
15926 For Renesas SH architectures.
15929 @cindex @file{sparc-stub.c}
15931 For @sc{sparc} architectures.
15933 @item sparcl-stub.c
15934 @cindex @file{sparcl-stub.c}
15937 For Fujitsu @sc{sparclite} architectures.
15941 The @file{README} file in the @value{GDBN} distribution may list other
15942 recently added stubs.
15945 * Stub Contents:: What the stub can do for you
15946 * Bootstrapping:: What you must do for the stub
15947 * Debug Session:: Putting it all together
15950 @node Stub Contents
15951 @subsection What the Stub Can Do for You
15953 @cindex remote serial stub
15954 The debugging stub for your architecture supplies these three
15958 @item set_debug_traps
15959 @findex set_debug_traps
15960 @cindex remote serial stub, initialization
15961 This routine arranges for @code{handle_exception} to run when your
15962 program stops. You must call this subroutine explicitly near the
15963 beginning of your program.
15965 @item handle_exception
15966 @findex handle_exception
15967 @cindex remote serial stub, main routine
15968 This is the central workhorse, but your program never calls it
15969 explicitly---the setup code arranges for @code{handle_exception} to
15970 run when a trap is triggered.
15972 @code{handle_exception} takes control when your program stops during
15973 execution (for example, on a breakpoint), and mediates communications
15974 with @value{GDBN} on the host machine. This is where the communications
15975 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
15976 representative on the target machine. It begins by sending summary
15977 information on the state of your program, then continues to execute,
15978 retrieving and transmitting any information @value{GDBN} needs, until you
15979 execute a @value{GDBN} command that makes your program resume; at that point,
15980 @code{handle_exception} returns control to your own code on the target
15984 @cindex @code{breakpoint} subroutine, remote
15985 Use this auxiliary subroutine to make your program contain a
15986 breakpoint. Depending on the particular situation, this may be the only
15987 way for @value{GDBN} to get control. For instance, if your target
15988 machine has some sort of interrupt button, you won't need to call this;
15989 pressing the interrupt button transfers control to
15990 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
15991 simply receiving characters on the serial port may also trigger a trap;
15992 again, in that situation, you don't need to call @code{breakpoint} from
15993 your own program---simply running @samp{target remote} from the host
15994 @value{GDBN} session gets control.
15996 Call @code{breakpoint} if none of these is true, or if you simply want
15997 to make certain your program stops at a predetermined point for the
15998 start of your debugging session.
16001 @node Bootstrapping
16002 @subsection What You Must Do for the Stub
16004 @cindex remote stub, support routines
16005 The debugging stubs that come with @value{GDBN} are set up for a particular
16006 chip architecture, but they have no information about the rest of your
16007 debugging target machine.
16009 First of all you need to tell the stub how to communicate with the
16013 @item int getDebugChar()
16014 @findex getDebugChar
16015 Write this subroutine to read a single character from the serial port.
16016 It may be identical to @code{getchar} for your target system; a
16017 different name is used to allow you to distinguish the two if you wish.
16019 @item void putDebugChar(int)
16020 @findex putDebugChar
16021 Write this subroutine to write a single character to the serial port.
16022 It may be identical to @code{putchar} for your target system; a
16023 different name is used to allow you to distinguish the two if you wish.
16026 @cindex control C, and remote debugging
16027 @cindex interrupting remote targets
16028 If you want @value{GDBN} to be able to stop your program while it is
16029 running, you need to use an interrupt-driven serial driver, and arrange
16030 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
16031 character). That is the character which @value{GDBN} uses to tell the
16032 remote system to stop.
16034 Getting the debugging target to return the proper status to @value{GDBN}
16035 probably requires changes to the standard stub; one quick and dirty way
16036 is to just execute a breakpoint instruction (the ``dirty'' part is that
16037 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
16039 Other routines you need to supply are:
16042 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
16043 @findex exceptionHandler
16044 Write this function to install @var{exception_address} in the exception
16045 handling tables. You need to do this because the stub does not have any
16046 way of knowing what the exception handling tables on your target system
16047 are like (for example, the processor's table might be in @sc{rom},
16048 containing entries which point to a table in @sc{ram}).
16049 @var{exception_number} is the exception number which should be changed;
16050 its meaning is architecture-dependent (for example, different numbers
16051 might represent divide by zero, misaligned access, etc). When this
16052 exception occurs, control should be transferred directly to
16053 @var{exception_address}, and the processor state (stack, registers,
16054 and so on) should be just as it is when a processor exception occurs. So if
16055 you want to use a jump instruction to reach @var{exception_address}, it
16056 should be a simple jump, not a jump to subroutine.
16058 For the 386, @var{exception_address} should be installed as an interrupt
16059 gate so that interrupts are masked while the handler runs. The gate
16060 should be at privilege level 0 (the most privileged level). The
16061 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
16062 help from @code{exceptionHandler}.
16064 @item void flush_i_cache()
16065 @findex flush_i_cache
16066 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
16067 instruction cache, if any, on your target machine. If there is no
16068 instruction cache, this subroutine may be a no-op.
16070 On target machines that have instruction caches, @value{GDBN} requires this
16071 function to make certain that the state of your program is stable.
16075 You must also make sure this library routine is available:
16078 @item void *memset(void *, int, int)
16080 This is the standard library function @code{memset} that sets an area of
16081 memory to a known value. If you have one of the free versions of
16082 @code{libc.a}, @code{memset} can be found there; otherwise, you must
16083 either obtain it from your hardware manufacturer, or write your own.
16086 If you do not use the GNU C compiler, you may need other standard
16087 library subroutines as well; this varies from one stub to another,
16088 but in general the stubs are likely to use any of the common library
16089 subroutines which @code{@value{NGCC}} generates as inline code.
16092 @node Debug Session
16093 @subsection Putting it All Together
16095 @cindex remote serial debugging summary
16096 In summary, when your program is ready to debug, you must follow these
16101 Make sure you have defined the supporting low-level routines
16102 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
16104 @code{getDebugChar}, @code{putDebugChar},
16105 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
16109 Insert these lines near the top of your program:
16117 For the 680x0 stub only, you need to provide a variable called
16118 @code{exceptionHook}. Normally you just use:
16121 void (*exceptionHook)() = 0;
16125 but if before calling @code{set_debug_traps}, you set it to point to a
16126 function in your program, that function is called when
16127 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
16128 error). The function indicated by @code{exceptionHook} is called with
16129 one parameter: an @code{int} which is the exception number.
16132 Compile and link together: your program, the @value{GDBN} debugging stub for
16133 your target architecture, and the supporting subroutines.
16136 Make sure you have a serial connection between your target machine and
16137 the @value{GDBN} host, and identify the serial port on the host.
16140 @c The "remote" target now provides a `load' command, so we should
16141 @c document that. FIXME.
16142 Download your program to your target machine (or get it there by
16143 whatever means the manufacturer provides), and start it.
16146 Start @value{GDBN} on the host, and connect to the target
16147 (@pxref{Connecting,,Connecting to a Remote Target}).
16151 @node Configurations
16152 @chapter Configuration-Specific Information
16154 While nearly all @value{GDBN} commands are available for all native and
16155 cross versions of the debugger, there are some exceptions. This chapter
16156 describes things that are only available in certain configurations.
16158 There are three major categories of configurations: native
16159 configurations, where the host and target are the same, embedded
16160 operating system configurations, which are usually the same for several
16161 different processor architectures, and bare embedded processors, which
16162 are quite different from each other.
16167 * Embedded Processors::
16174 This section describes details specific to particular native
16179 * BSD libkvm Interface:: Debugging BSD kernel memory images
16180 * SVR4 Process Information:: SVR4 process information
16181 * DJGPP Native:: Features specific to the DJGPP port
16182 * Cygwin Native:: Features specific to the Cygwin port
16183 * Hurd Native:: Features specific to @sc{gnu} Hurd
16184 * Neutrino:: Features specific to QNX Neutrino
16185 * Darwin:: Features specific to Darwin
16191 On HP-UX systems, if you refer to a function or variable name that
16192 begins with a dollar sign, @value{GDBN} searches for a user or system
16193 name first, before it searches for a convenience variable.
16196 @node BSD libkvm Interface
16197 @subsection BSD libkvm Interface
16200 @cindex kernel memory image
16201 @cindex kernel crash dump
16203 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
16204 interface that provides a uniform interface for accessing kernel virtual
16205 memory images, including live systems and crash dumps. @value{GDBN}
16206 uses this interface to allow you to debug live kernels and kernel crash
16207 dumps on many native BSD configurations. This is implemented as a
16208 special @code{kvm} debugging target. For debugging a live system, load
16209 the currently running kernel into @value{GDBN} and connect to the
16213 (@value{GDBP}) @b{target kvm}
16216 For debugging crash dumps, provide the file name of the crash dump as an
16220 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
16223 Once connected to the @code{kvm} target, the following commands are
16229 Set current context from the @dfn{Process Control Block} (PCB) address.
16232 Set current context from proc address. This command isn't available on
16233 modern FreeBSD systems.
16236 @node SVR4 Process Information
16237 @subsection SVR4 Process Information
16239 @cindex examine process image
16240 @cindex process info via @file{/proc}
16242 Many versions of SVR4 and compatible systems provide a facility called
16243 @samp{/proc} that can be used to examine the image of a running
16244 process using file-system subroutines. If @value{GDBN} is configured
16245 for an operating system with this facility, the command @code{info
16246 proc} is available to report information about the process running
16247 your program, or about any process running on your system. @code{info
16248 proc} works only on SVR4 systems that include the @code{procfs} code.
16249 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
16250 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
16256 @itemx info proc @var{process-id}
16257 Summarize available information about any running process. If a
16258 process ID is specified by @var{process-id}, display information about
16259 that process; otherwise display information about the program being
16260 debugged. The summary includes the debugged process ID, the command
16261 line used to invoke it, its current working directory, and its
16262 executable file's absolute file name.
16264 On some systems, @var{process-id} can be of the form
16265 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
16266 within a process. If the optional @var{pid} part is missing, it means
16267 a thread from the process being debugged (the leading @samp{/} still
16268 needs to be present, or else @value{GDBN} will interpret the number as
16269 a process ID rather than a thread ID).
16271 @item info proc mappings
16272 @cindex memory address space mappings
16273 Report the memory address space ranges accessible in the program, with
16274 information on whether the process has read, write, or execute access
16275 rights to each range. On @sc{gnu}/Linux systems, each memory range
16276 includes the object file which is mapped to that range, instead of the
16277 memory access rights to that range.
16279 @item info proc stat
16280 @itemx info proc status
16281 @cindex process detailed status information
16282 These subcommands are specific to @sc{gnu}/Linux systems. They show
16283 the process-related information, including the user ID and group ID;
16284 how many threads are there in the process; its virtual memory usage;
16285 the signals that are pending, blocked, and ignored; its TTY; its
16286 consumption of system and user time; its stack size; its @samp{nice}
16287 value; etc. For more information, see the @samp{proc} man page
16288 (type @kbd{man 5 proc} from your shell prompt).
16290 @item info proc all
16291 Show all the information about the process described under all of the
16292 above @code{info proc} subcommands.
16295 @comment These sub-options of 'info proc' were not included when
16296 @comment procfs.c was re-written. Keep their descriptions around
16297 @comment against the day when someone finds the time to put them back in.
16298 @kindex info proc times
16299 @item info proc times
16300 Starting time, user CPU time, and system CPU time for your program and
16303 @kindex info proc id
16305 Report on the process IDs related to your program: its own process ID,
16306 the ID of its parent, the process group ID, and the session ID.
16309 @item set procfs-trace
16310 @kindex set procfs-trace
16311 @cindex @code{procfs} API calls
16312 This command enables and disables tracing of @code{procfs} API calls.
16314 @item show procfs-trace
16315 @kindex show procfs-trace
16316 Show the current state of @code{procfs} API call tracing.
16318 @item set procfs-file @var{file}
16319 @kindex set procfs-file
16320 Tell @value{GDBN} to write @code{procfs} API trace to the named
16321 @var{file}. @value{GDBN} appends the trace info to the previous
16322 contents of the file. The default is to display the trace on the
16325 @item show procfs-file
16326 @kindex show procfs-file
16327 Show the file to which @code{procfs} API trace is written.
16329 @item proc-trace-entry
16330 @itemx proc-trace-exit
16331 @itemx proc-untrace-entry
16332 @itemx proc-untrace-exit
16333 @kindex proc-trace-entry
16334 @kindex proc-trace-exit
16335 @kindex proc-untrace-entry
16336 @kindex proc-untrace-exit
16337 These commands enable and disable tracing of entries into and exits
16338 from the @code{syscall} interface.
16341 @kindex info pidlist
16342 @cindex process list, QNX Neutrino
16343 For QNX Neutrino only, this command displays the list of all the
16344 processes and all the threads within each process.
16347 @kindex info meminfo
16348 @cindex mapinfo list, QNX Neutrino
16349 For QNX Neutrino only, this command displays the list of all mapinfos.
16353 @subsection Features for Debugging @sc{djgpp} Programs
16354 @cindex @sc{djgpp} debugging
16355 @cindex native @sc{djgpp} debugging
16356 @cindex MS-DOS-specific commands
16359 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
16360 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
16361 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
16362 top of real-mode DOS systems and their emulations.
16364 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
16365 defines a few commands specific to the @sc{djgpp} port. This
16366 subsection describes those commands.
16371 This is a prefix of @sc{djgpp}-specific commands which print
16372 information about the target system and important OS structures.
16375 @cindex MS-DOS system info
16376 @cindex free memory information (MS-DOS)
16377 @item info dos sysinfo
16378 This command displays assorted information about the underlying
16379 platform: the CPU type and features, the OS version and flavor, the
16380 DPMI version, and the available conventional and DPMI memory.
16385 @cindex segment descriptor tables
16386 @cindex descriptor tables display
16388 @itemx info dos ldt
16389 @itemx info dos idt
16390 These 3 commands display entries from, respectively, Global, Local,
16391 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
16392 tables are data structures which store a descriptor for each segment
16393 that is currently in use. The segment's selector is an index into a
16394 descriptor table; the table entry for that index holds the
16395 descriptor's base address and limit, and its attributes and access
16398 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
16399 segment (used for both data and the stack), and a DOS segment (which
16400 allows access to DOS/BIOS data structures and absolute addresses in
16401 conventional memory). However, the DPMI host will usually define
16402 additional segments in order to support the DPMI environment.
16404 @cindex garbled pointers
16405 These commands allow to display entries from the descriptor tables.
16406 Without an argument, all entries from the specified table are
16407 displayed. An argument, which should be an integer expression, means
16408 display a single entry whose index is given by the argument. For
16409 example, here's a convenient way to display information about the
16410 debugged program's data segment:
16413 @exdent @code{(@value{GDBP}) info dos ldt $ds}
16414 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
16418 This comes in handy when you want to see whether a pointer is outside
16419 the data segment's limit (i.e.@: @dfn{garbled}).
16421 @cindex page tables display (MS-DOS)
16423 @itemx info dos pte
16424 These two commands display entries from, respectively, the Page
16425 Directory and the Page Tables. Page Directories and Page Tables are
16426 data structures which control how virtual memory addresses are mapped
16427 into physical addresses. A Page Table includes an entry for every
16428 page of memory that is mapped into the program's address space; there
16429 may be several Page Tables, each one holding up to 4096 entries. A
16430 Page Directory has up to 4096 entries, one each for every Page Table
16431 that is currently in use.
16433 Without an argument, @kbd{info dos pde} displays the entire Page
16434 Directory, and @kbd{info dos pte} displays all the entries in all of
16435 the Page Tables. An argument, an integer expression, given to the
16436 @kbd{info dos pde} command means display only that entry from the Page
16437 Directory table. An argument given to the @kbd{info dos pte} command
16438 means display entries from a single Page Table, the one pointed to by
16439 the specified entry in the Page Directory.
16441 @cindex direct memory access (DMA) on MS-DOS
16442 These commands are useful when your program uses @dfn{DMA} (Direct
16443 Memory Access), which needs physical addresses to program the DMA
16446 These commands are supported only with some DPMI servers.
16448 @cindex physical address from linear address
16449 @item info dos address-pte @var{addr}
16450 This command displays the Page Table entry for a specified linear
16451 address. The argument @var{addr} is a linear address which should
16452 already have the appropriate segment's base address added to it,
16453 because this command accepts addresses which may belong to @emph{any}
16454 segment. For example, here's how to display the Page Table entry for
16455 the page where a variable @code{i} is stored:
16458 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
16459 @exdent @code{Page Table entry for address 0x11a00d30:}
16460 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
16464 This says that @code{i} is stored at offset @code{0xd30} from the page
16465 whose physical base address is @code{0x02698000}, and shows all the
16466 attributes of that page.
16468 Note that you must cast the addresses of variables to a @code{char *},
16469 since otherwise the value of @code{__djgpp_base_address}, the base
16470 address of all variables and functions in a @sc{djgpp} program, will
16471 be added using the rules of C pointer arithmetics: if @code{i} is
16472 declared an @code{int}, @value{GDBN} will add 4 times the value of
16473 @code{__djgpp_base_address} to the address of @code{i}.
16475 Here's another example, it displays the Page Table entry for the
16479 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
16480 @exdent @code{Page Table entry for address 0x29110:}
16481 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
16485 (The @code{+ 3} offset is because the transfer buffer's address is the
16486 3rd member of the @code{_go32_info_block} structure.) The output
16487 clearly shows that this DPMI server maps the addresses in conventional
16488 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
16489 linear (@code{0x29110}) addresses are identical.
16491 This command is supported only with some DPMI servers.
16494 @cindex DOS serial data link, remote debugging
16495 In addition to native debugging, the DJGPP port supports remote
16496 debugging via a serial data link. The following commands are specific
16497 to remote serial debugging in the DJGPP port of @value{GDBN}.
16500 @kindex set com1base
16501 @kindex set com1irq
16502 @kindex set com2base
16503 @kindex set com2irq
16504 @kindex set com3base
16505 @kindex set com3irq
16506 @kindex set com4base
16507 @kindex set com4irq
16508 @item set com1base @var{addr}
16509 This command sets the base I/O port address of the @file{COM1} serial
16512 @item set com1irq @var{irq}
16513 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
16514 for the @file{COM1} serial port.
16516 There are similar commands @samp{set com2base}, @samp{set com3irq},
16517 etc.@: for setting the port address and the @code{IRQ} lines for the
16520 @kindex show com1base
16521 @kindex show com1irq
16522 @kindex show com2base
16523 @kindex show com2irq
16524 @kindex show com3base
16525 @kindex show com3irq
16526 @kindex show com4base
16527 @kindex show com4irq
16528 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
16529 display the current settings of the base address and the @code{IRQ}
16530 lines used by the COM ports.
16533 @kindex info serial
16534 @cindex DOS serial port status
16535 This command prints the status of the 4 DOS serial ports. For each
16536 port, it prints whether it's active or not, its I/O base address and
16537 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
16538 counts of various errors encountered so far.
16542 @node Cygwin Native
16543 @subsection Features for Debugging MS Windows PE Executables
16544 @cindex MS Windows debugging
16545 @cindex native Cygwin debugging
16546 @cindex Cygwin-specific commands
16548 @value{GDBN} supports native debugging of MS Windows programs, including
16549 DLLs with and without symbolic debugging information.
16551 @cindex Ctrl-BREAK, MS-Windows
16552 @cindex interrupt debuggee on MS-Windows
16553 MS-Windows programs that call @code{SetConsoleMode} to switch off the
16554 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
16555 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
16556 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
16557 sequence, which can be used to interrupt the debuggee even if it
16560 There are various additional Cygwin-specific commands, described in
16561 this section. Working with DLLs that have no debugging symbols is
16562 described in @ref{Non-debug DLL Symbols}.
16567 This is a prefix of MS Windows-specific commands which print
16568 information about the target system and important OS structures.
16570 @item info w32 selector
16571 This command displays information returned by
16572 the Win32 API @code{GetThreadSelectorEntry} function.
16573 It takes an optional argument that is evaluated to
16574 a long value to give the information about this given selector.
16575 Without argument, this command displays information
16576 about the six segment registers.
16578 @item info w32 thread-information-block
16579 This command displays thread specific information stored in the
16580 Thread Information Block (readable on the X86 CPU family using @code{$fs}
16581 selector for 32-bit programs and @code{$gs} for 64-bit programs).
16585 This is a Cygwin-specific alias of @code{info shared}.
16587 @kindex dll-symbols
16589 This command loads symbols from a dll similarly to
16590 add-sym command but without the need to specify a base address.
16592 @kindex set cygwin-exceptions
16593 @cindex debugging the Cygwin DLL
16594 @cindex Cygwin DLL, debugging
16595 @item set cygwin-exceptions @var{mode}
16596 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
16597 happen inside the Cygwin DLL. If @var{mode} is @code{off},
16598 @value{GDBN} will delay recognition of exceptions, and may ignore some
16599 exceptions which seem to be caused by internal Cygwin DLL
16600 ``bookkeeping''. This option is meant primarily for debugging the
16601 Cygwin DLL itself; the default value is @code{off} to avoid annoying
16602 @value{GDBN} users with false @code{SIGSEGV} signals.
16604 @kindex show cygwin-exceptions
16605 @item show cygwin-exceptions
16606 Displays whether @value{GDBN} will break on exceptions that happen
16607 inside the Cygwin DLL itself.
16609 @kindex set new-console
16610 @item set new-console @var{mode}
16611 If @var{mode} is @code{on} the debuggee will
16612 be started in a new console on next start.
16613 If @var{mode} is @code{off}, the debuggee will
16614 be started in the same console as the debugger.
16616 @kindex show new-console
16617 @item show new-console
16618 Displays whether a new console is used
16619 when the debuggee is started.
16621 @kindex set new-group
16622 @item set new-group @var{mode}
16623 This boolean value controls whether the debuggee should
16624 start a new group or stay in the same group as the debugger.
16625 This affects the way the Windows OS handles
16628 @kindex show new-group
16629 @item show new-group
16630 Displays current value of new-group boolean.
16632 @kindex set debugevents
16633 @item set debugevents
16634 This boolean value adds debug output concerning kernel events related
16635 to the debuggee seen by the debugger. This includes events that
16636 signal thread and process creation and exit, DLL loading and
16637 unloading, console interrupts, and debugging messages produced by the
16638 Windows @code{OutputDebugString} API call.
16640 @kindex set debugexec
16641 @item set debugexec
16642 This boolean value adds debug output concerning execute events
16643 (such as resume thread) seen by the debugger.
16645 @kindex set debugexceptions
16646 @item set debugexceptions
16647 This boolean value adds debug output concerning exceptions in the
16648 debuggee seen by the debugger.
16650 @kindex set debugmemory
16651 @item set debugmemory
16652 This boolean value adds debug output concerning debuggee memory reads
16653 and writes by the debugger.
16657 This boolean values specifies whether the debuggee is called
16658 via a shell or directly (default value is on).
16662 Displays if the debuggee will be started with a shell.
16667 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
16670 @node Non-debug DLL Symbols
16671 @subsubsection Support for DLLs without Debugging Symbols
16672 @cindex DLLs with no debugging symbols
16673 @cindex Minimal symbols and DLLs
16675 Very often on windows, some of the DLLs that your program relies on do
16676 not include symbolic debugging information (for example,
16677 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
16678 symbols in a DLL, it relies on the minimal amount of symbolic
16679 information contained in the DLL's export table. This section
16680 describes working with such symbols, known internally to @value{GDBN} as
16681 ``minimal symbols''.
16683 Note that before the debugged program has started execution, no DLLs
16684 will have been loaded. The easiest way around this problem is simply to
16685 start the program --- either by setting a breakpoint or letting the
16686 program run once to completion. It is also possible to force
16687 @value{GDBN} to load a particular DLL before starting the executable ---
16688 see the shared library information in @ref{Files}, or the
16689 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
16690 explicitly loading symbols from a DLL with no debugging information will
16691 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
16692 which may adversely affect symbol lookup performance.
16694 @subsubsection DLL Name Prefixes
16696 In keeping with the naming conventions used by the Microsoft debugging
16697 tools, DLL export symbols are made available with a prefix based on the
16698 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
16699 also entered into the symbol table, so @code{CreateFileA} is often
16700 sufficient. In some cases there will be name clashes within a program
16701 (particularly if the executable itself includes full debugging symbols)
16702 necessitating the use of the fully qualified name when referring to the
16703 contents of the DLL. Use single-quotes around the name to avoid the
16704 exclamation mark (``!'') being interpreted as a language operator.
16706 Note that the internal name of the DLL may be all upper-case, even
16707 though the file name of the DLL is lower-case, or vice-versa. Since
16708 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
16709 some confusion. If in doubt, try the @code{info functions} and
16710 @code{info variables} commands or even @code{maint print msymbols}
16711 (@pxref{Symbols}). Here's an example:
16714 (@value{GDBP}) info function CreateFileA
16715 All functions matching regular expression "CreateFileA":
16717 Non-debugging symbols:
16718 0x77e885f4 CreateFileA
16719 0x77e885f4 KERNEL32!CreateFileA
16723 (@value{GDBP}) info function !
16724 All functions matching regular expression "!":
16726 Non-debugging symbols:
16727 0x6100114c cygwin1!__assert
16728 0x61004034 cygwin1!_dll_crt0@@0
16729 0x61004240 cygwin1!dll_crt0(per_process *)
16733 @subsubsection Working with Minimal Symbols
16735 Symbols extracted from a DLL's export table do not contain very much
16736 type information. All that @value{GDBN} can do is guess whether a symbol
16737 refers to a function or variable depending on the linker section that
16738 contains the symbol. Also note that the actual contents of the memory
16739 contained in a DLL are not available unless the program is running. This
16740 means that you cannot examine the contents of a variable or disassemble
16741 a function within a DLL without a running program.
16743 Variables are generally treated as pointers and dereferenced
16744 automatically. For this reason, it is often necessary to prefix a
16745 variable name with the address-of operator (``&'') and provide explicit
16746 type information in the command. Here's an example of the type of
16750 (@value{GDBP}) print 'cygwin1!__argv'
16755 (@value{GDBP}) x 'cygwin1!__argv'
16756 0x10021610: "\230y\""
16759 And two possible solutions:
16762 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
16763 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
16767 (@value{GDBP}) x/2x &'cygwin1!__argv'
16768 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
16769 (@value{GDBP}) x/x 0x10021608
16770 0x10021608: 0x0022fd98
16771 (@value{GDBP}) x/s 0x0022fd98
16772 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
16775 Setting a break point within a DLL is possible even before the program
16776 starts execution. However, under these circumstances, @value{GDBN} can't
16777 examine the initial instructions of the function in order to skip the
16778 function's frame set-up code. You can work around this by using ``*&''
16779 to set the breakpoint at a raw memory address:
16782 (@value{GDBP}) break *&'python22!PyOS_Readline'
16783 Breakpoint 1 at 0x1e04eff0
16786 The author of these extensions is not entirely convinced that setting a
16787 break point within a shared DLL like @file{kernel32.dll} is completely
16791 @subsection Commands Specific to @sc{gnu} Hurd Systems
16792 @cindex @sc{gnu} Hurd debugging
16794 This subsection describes @value{GDBN} commands specific to the
16795 @sc{gnu} Hurd native debugging.
16800 @kindex set signals@r{, Hurd command}
16801 @kindex set sigs@r{, Hurd command}
16802 This command toggles the state of inferior signal interception by
16803 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
16804 affected by this command. @code{sigs} is a shorthand alias for
16809 @kindex show signals@r{, Hurd command}
16810 @kindex show sigs@r{, Hurd command}
16811 Show the current state of intercepting inferior's signals.
16813 @item set signal-thread
16814 @itemx set sigthread
16815 @kindex set signal-thread
16816 @kindex set sigthread
16817 This command tells @value{GDBN} which thread is the @code{libc} signal
16818 thread. That thread is run when a signal is delivered to a running
16819 process. @code{set sigthread} is the shorthand alias of @code{set
16822 @item show signal-thread
16823 @itemx show sigthread
16824 @kindex show signal-thread
16825 @kindex show sigthread
16826 These two commands show which thread will run when the inferior is
16827 delivered a signal.
16830 @kindex set stopped@r{, Hurd command}
16831 This commands tells @value{GDBN} that the inferior process is stopped,
16832 as with the @code{SIGSTOP} signal. The stopped process can be
16833 continued by delivering a signal to it.
16836 @kindex show stopped@r{, Hurd command}
16837 This command shows whether @value{GDBN} thinks the debuggee is
16840 @item set exceptions
16841 @kindex set exceptions@r{, Hurd command}
16842 Use this command to turn off trapping of exceptions in the inferior.
16843 When exception trapping is off, neither breakpoints nor
16844 single-stepping will work. To restore the default, set exception
16847 @item show exceptions
16848 @kindex show exceptions@r{, Hurd command}
16849 Show the current state of trapping exceptions in the inferior.
16851 @item set task pause
16852 @kindex set task@r{, Hurd commands}
16853 @cindex task attributes (@sc{gnu} Hurd)
16854 @cindex pause current task (@sc{gnu} Hurd)
16855 This command toggles task suspension when @value{GDBN} has control.
16856 Setting it to on takes effect immediately, and the task is suspended
16857 whenever @value{GDBN} gets control. Setting it to off will take
16858 effect the next time the inferior is continued. If this option is set
16859 to off, you can use @code{set thread default pause on} or @code{set
16860 thread pause on} (see below) to pause individual threads.
16862 @item show task pause
16863 @kindex show task@r{, Hurd commands}
16864 Show the current state of task suspension.
16866 @item set task detach-suspend-count
16867 @cindex task suspend count
16868 @cindex detach from task, @sc{gnu} Hurd
16869 This command sets the suspend count the task will be left with when
16870 @value{GDBN} detaches from it.
16872 @item show task detach-suspend-count
16873 Show the suspend count the task will be left with when detaching.
16875 @item set task exception-port
16876 @itemx set task excp
16877 @cindex task exception port, @sc{gnu} Hurd
16878 This command sets the task exception port to which @value{GDBN} will
16879 forward exceptions. The argument should be the value of the @dfn{send
16880 rights} of the task. @code{set task excp} is a shorthand alias.
16882 @item set noninvasive
16883 @cindex noninvasive task options
16884 This command switches @value{GDBN} to a mode that is the least
16885 invasive as far as interfering with the inferior is concerned. This
16886 is the same as using @code{set task pause}, @code{set exceptions}, and
16887 @code{set signals} to values opposite to the defaults.
16889 @item info send-rights
16890 @itemx info receive-rights
16891 @itemx info port-rights
16892 @itemx info port-sets
16893 @itemx info dead-names
16896 @cindex send rights, @sc{gnu} Hurd
16897 @cindex receive rights, @sc{gnu} Hurd
16898 @cindex port rights, @sc{gnu} Hurd
16899 @cindex port sets, @sc{gnu} Hurd
16900 @cindex dead names, @sc{gnu} Hurd
16901 These commands display information about, respectively, send rights,
16902 receive rights, port rights, port sets, and dead names of a task.
16903 There are also shorthand aliases: @code{info ports} for @code{info
16904 port-rights} and @code{info psets} for @code{info port-sets}.
16906 @item set thread pause
16907 @kindex set thread@r{, Hurd command}
16908 @cindex thread properties, @sc{gnu} Hurd
16909 @cindex pause current thread (@sc{gnu} Hurd)
16910 This command toggles current thread suspension when @value{GDBN} has
16911 control. Setting it to on takes effect immediately, and the current
16912 thread is suspended whenever @value{GDBN} gets control. Setting it to
16913 off will take effect the next time the inferior is continued.
16914 Normally, this command has no effect, since when @value{GDBN} has
16915 control, the whole task is suspended. However, if you used @code{set
16916 task pause off} (see above), this command comes in handy to suspend
16917 only the current thread.
16919 @item show thread pause
16920 @kindex show thread@r{, Hurd command}
16921 This command shows the state of current thread suspension.
16923 @item set thread run
16924 This command sets whether the current thread is allowed to run.
16926 @item show thread run
16927 Show whether the current thread is allowed to run.
16929 @item set thread detach-suspend-count
16930 @cindex thread suspend count, @sc{gnu} Hurd
16931 @cindex detach from thread, @sc{gnu} Hurd
16932 This command sets the suspend count @value{GDBN} will leave on a
16933 thread when detaching. This number is relative to the suspend count
16934 found by @value{GDBN} when it notices the thread; use @code{set thread
16935 takeover-suspend-count} to force it to an absolute value.
16937 @item show thread detach-suspend-count
16938 Show the suspend count @value{GDBN} will leave on the thread when
16941 @item set thread exception-port
16942 @itemx set thread excp
16943 Set the thread exception port to which to forward exceptions. This
16944 overrides the port set by @code{set task exception-port} (see above).
16945 @code{set thread excp} is the shorthand alias.
16947 @item set thread takeover-suspend-count
16948 Normally, @value{GDBN}'s thread suspend counts are relative to the
16949 value @value{GDBN} finds when it notices each thread. This command
16950 changes the suspend counts to be absolute instead.
16952 @item set thread default
16953 @itemx show thread default
16954 @cindex thread default settings, @sc{gnu} Hurd
16955 Each of the above @code{set thread} commands has a @code{set thread
16956 default} counterpart (e.g., @code{set thread default pause}, @code{set
16957 thread default exception-port}, etc.). The @code{thread default}
16958 variety of commands sets the default thread properties for all
16959 threads; you can then change the properties of individual threads with
16960 the non-default commands.
16965 @subsection QNX Neutrino
16966 @cindex QNX Neutrino
16968 @value{GDBN} provides the following commands specific to the QNX
16972 @item set debug nto-debug
16973 @kindex set debug nto-debug
16974 When set to on, enables debugging messages specific to the QNX
16977 @item show debug nto-debug
16978 @kindex show debug nto-debug
16979 Show the current state of QNX Neutrino messages.
16986 @value{GDBN} provides the following commands specific to the Darwin target:
16989 @item set debug darwin @var{num}
16990 @kindex set debug darwin
16991 When set to a non zero value, enables debugging messages specific to
16992 the Darwin support. Higher values produce more verbose output.
16994 @item show debug darwin
16995 @kindex show debug darwin
16996 Show the current state of Darwin messages.
16998 @item set debug mach-o @var{num}
16999 @kindex set debug mach-o
17000 When set to a non zero value, enables debugging messages while
17001 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
17002 file format used on Darwin for object and executable files.) Higher
17003 values produce more verbose output. This is a command to diagnose
17004 problems internal to @value{GDBN} and should not be needed in normal
17007 @item show debug mach-o
17008 @kindex show debug mach-o
17009 Show the current state of Mach-O file messages.
17011 @item set mach-exceptions on
17012 @itemx set mach-exceptions off
17013 @kindex set mach-exceptions
17014 On Darwin, faults are first reported as a Mach exception and are then
17015 mapped to a Posix signal. Use this command to turn on trapping of
17016 Mach exceptions in the inferior. This might be sometimes useful to
17017 better understand the cause of a fault. The default is off.
17019 @item show mach-exceptions
17020 @kindex show mach-exceptions
17021 Show the current state of exceptions trapping.
17026 @section Embedded Operating Systems
17028 This section describes configurations involving the debugging of
17029 embedded operating systems that are available for several different
17033 * VxWorks:: Using @value{GDBN} with VxWorks
17036 @value{GDBN} includes the ability to debug programs running on
17037 various real-time operating systems.
17040 @subsection Using @value{GDBN} with VxWorks
17046 @kindex target vxworks
17047 @item target vxworks @var{machinename}
17048 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
17049 is the target system's machine name or IP address.
17053 On VxWorks, @code{load} links @var{filename} dynamically on the
17054 current target system as well as adding its symbols in @value{GDBN}.
17056 @value{GDBN} enables developers to spawn and debug tasks running on networked
17057 VxWorks targets from a Unix host. Already-running tasks spawned from
17058 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
17059 both the Unix host and on the VxWorks target. The program
17060 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
17061 installed with the name @code{vxgdb}, to distinguish it from a
17062 @value{GDBN} for debugging programs on the host itself.)
17065 @item VxWorks-timeout @var{args}
17066 @kindex vxworks-timeout
17067 All VxWorks-based targets now support the option @code{vxworks-timeout}.
17068 This option is set by the user, and @var{args} represents the number of
17069 seconds @value{GDBN} waits for responses to rpc's. You might use this if
17070 your VxWorks target is a slow software simulator or is on the far side
17071 of a thin network line.
17074 The following information on connecting to VxWorks was current when
17075 this manual was produced; newer releases of VxWorks may use revised
17078 @findex INCLUDE_RDB
17079 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
17080 to include the remote debugging interface routines in the VxWorks
17081 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
17082 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
17083 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
17084 source debugging task @code{tRdbTask} when VxWorks is booted. For more
17085 information on configuring and remaking VxWorks, see the manufacturer's
17087 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
17089 Once you have included @file{rdb.a} in your VxWorks system image and set
17090 your Unix execution search path to find @value{GDBN}, you are ready to
17091 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
17092 @code{vxgdb}, depending on your installation).
17094 @value{GDBN} comes up showing the prompt:
17101 * VxWorks Connection:: Connecting to VxWorks
17102 * VxWorks Download:: VxWorks download
17103 * VxWorks Attach:: Running tasks
17106 @node VxWorks Connection
17107 @subsubsection Connecting to VxWorks
17109 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
17110 network. To connect to a target whose host name is ``@code{tt}'', type:
17113 (vxgdb) target vxworks tt
17117 @value{GDBN} displays messages like these:
17120 Attaching remote machine across net...
17125 @value{GDBN} then attempts to read the symbol tables of any object modules
17126 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
17127 these files by searching the directories listed in the command search
17128 path (@pxref{Environment, ,Your Program's Environment}); if it fails
17129 to find an object file, it displays a message such as:
17132 prog.o: No such file or directory.
17135 When this happens, add the appropriate directory to the search path with
17136 the @value{GDBN} command @code{path}, and execute the @code{target}
17139 @node VxWorks Download
17140 @subsubsection VxWorks Download
17142 @cindex download to VxWorks
17143 If you have connected to the VxWorks target and you want to debug an
17144 object that has not yet been loaded, you can use the @value{GDBN}
17145 @code{load} command to download a file from Unix to VxWorks
17146 incrementally. The object file given as an argument to the @code{load}
17147 command is actually opened twice: first by the VxWorks target in order
17148 to download the code, then by @value{GDBN} in order to read the symbol
17149 table. This can lead to problems if the current working directories on
17150 the two systems differ. If both systems have NFS mounted the same
17151 filesystems, you can avoid these problems by using absolute paths.
17152 Otherwise, it is simplest to set the working directory on both systems
17153 to the directory in which the object file resides, and then to reference
17154 the file by its name, without any path. For instance, a program
17155 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
17156 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
17157 program, type this on VxWorks:
17160 -> cd "@var{vxpath}/vw/demo/rdb"
17164 Then, in @value{GDBN}, type:
17167 (vxgdb) cd @var{hostpath}/vw/demo/rdb
17168 (vxgdb) load prog.o
17171 @value{GDBN} displays a response similar to this:
17174 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
17177 You can also use the @code{load} command to reload an object module
17178 after editing and recompiling the corresponding source file. Note that
17179 this makes @value{GDBN} delete all currently-defined breakpoints,
17180 auto-displays, and convenience variables, and to clear the value
17181 history. (This is necessary in order to preserve the integrity of
17182 debugger's data structures that reference the target system's symbol
17185 @node VxWorks Attach
17186 @subsubsection Running Tasks
17188 @cindex running VxWorks tasks
17189 You can also attach to an existing task using the @code{attach} command as
17193 (vxgdb) attach @var{task}
17197 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
17198 or suspended when you attach to it. Running tasks are suspended at
17199 the time of attachment.
17201 @node Embedded Processors
17202 @section Embedded Processors
17204 This section goes into details specific to particular embedded
17207 @cindex send command to simulator
17208 Whenever a specific embedded processor has a simulator, @value{GDBN}
17209 allows to send an arbitrary command to the simulator.
17212 @item sim @var{command}
17213 @kindex sim@r{, a command}
17214 Send an arbitrary @var{command} string to the simulator. Consult the
17215 documentation for the specific simulator in use for information about
17216 acceptable commands.
17222 * M32R/D:: Renesas M32R/D
17223 * M68K:: Motorola M68K
17224 * MicroBlaze:: Xilinx MicroBlaze
17225 * MIPS Embedded:: MIPS Embedded
17226 * OpenRISC 1000:: OpenRisc 1000
17227 * PA:: HP PA Embedded
17228 * PowerPC Embedded:: PowerPC Embedded
17229 * Sparclet:: Tsqware Sparclet
17230 * Sparclite:: Fujitsu Sparclite
17231 * Z8000:: Zilog Z8000
17234 * Super-H:: Renesas Super-H
17243 @item target rdi @var{dev}
17244 ARM Angel monitor, via RDI library interface to ADP protocol. You may
17245 use this target to communicate with both boards running the Angel
17246 monitor, or with the EmbeddedICE JTAG debug device.
17249 @item target rdp @var{dev}
17254 @value{GDBN} provides the following ARM-specific commands:
17257 @item set arm disassembler
17259 This commands selects from a list of disassembly styles. The
17260 @code{"std"} style is the standard style.
17262 @item show arm disassembler
17264 Show the current disassembly style.
17266 @item set arm apcs32
17267 @cindex ARM 32-bit mode
17268 This command toggles ARM operation mode between 32-bit and 26-bit.
17270 @item show arm apcs32
17271 Display the current usage of the ARM 32-bit mode.
17273 @item set arm fpu @var{fputype}
17274 This command sets the ARM floating-point unit (FPU) type. The
17275 argument @var{fputype} can be one of these:
17279 Determine the FPU type by querying the OS ABI.
17281 Software FPU, with mixed-endian doubles on little-endian ARM
17284 GCC-compiled FPA co-processor.
17286 Software FPU with pure-endian doubles.
17292 Show the current type of the FPU.
17295 This command forces @value{GDBN} to use the specified ABI.
17298 Show the currently used ABI.
17300 @item set arm fallback-mode (arm|thumb|auto)
17301 @value{GDBN} uses the symbol table, when available, to determine
17302 whether instructions are ARM or Thumb. This command controls
17303 @value{GDBN}'s default behavior when the symbol table is not
17304 available. The default is @samp{auto}, which causes @value{GDBN} to
17305 use the current execution mode (from the @code{T} bit in the @code{CPSR}
17308 @item show arm fallback-mode
17309 Show the current fallback instruction mode.
17311 @item set arm force-mode (arm|thumb|auto)
17312 This command overrides use of the symbol table to determine whether
17313 instructions are ARM or Thumb. The default is @samp{auto}, which
17314 causes @value{GDBN} to use the symbol table and then the setting
17315 of @samp{set arm fallback-mode}.
17317 @item show arm force-mode
17318 Show the current forced instruction mode.
17320 @item set debug arm
17321 Toggle whether to display ARM-specific debugging messages from the ARM
17322 target support subsystem.
17324 @item show debug arm
17325 Show whether ARM-specific debugging messages are enabled.
17328 The following commands are available when an ARM target is debugged
17329 using the RDI interface:
17332 @item rdilogfile @r{[}@var{file}@r{]}
17334 @cindex ADP (Angel Debugger Protocol) logging
17335 Set the filename for the ADP (Angel Debugger Protocol) packet log.
17336 With an argument, sets the log file to the specified @var{file}. With
17337 no argument, show the current log file name. The default log file is
17340 @item rdilogenable @r{[}@var{arg}@r{]}
17341 @kindex rdilogenable
17342 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
17343 enables logging, with an argument 0 or @code{"no"} disables it. With
17344 no arguments displays the current setting. When logging is enabled,
17345 ADP packets exchanged between @value{GDBN} and the RDI target device
17346 are logged to a file.
17348 @item set rdiromatzero
17349 @kindex set rdiromatzero
17350 @cindex ROM at zero address, RDI
17351 Tell @value{GDBN} whether the target has ROM at address 0. If on,
17352 vector catching is disabled, so that zero address can be used. If off
17353 (the default), vector catching is enabled. For this command to take
17354 effect, it needs to be invoked prior to the @code{target rdi} command.
17356 @item show rdiromatzero
17357 @kindex show rdiromatzero
17358 Show the current setting of ROM at zero address.
17360 @item set rdiheartbeat
17361 @kindex set rdiheartbeat
17362 @cindex RDI heartbeat
17363 Enable or disable RDI heartbeat packets. It is not recommended to
17364 turn on this option, since it confuses ARM and EPI JTAG interface, as
17365 well as the Angel monitor.
17367 @item show rdiheartbeat
17368 @kindex show rdiheartbeat
17369 Show the setting of RDI heartbeat packets.
17373 @item target sim @r{[}@var{simargs}@r{]} @dots{}
17374 The @value{GDBN} ARM simulator accepts the following optional arguments.
17377 @item --swi-support=@var{type}
17378 Tell the simulator which SWI interfaces to support.
17379 @var{type} may be a comma separated list of the following values.
17380 The default value is @code{all}.
17393 @subsection Renesas M32R/D and M32R/SDI
17396 @kindex target m32r
17397 @item target m32r @var{dev}
17398 Renesas M32R/D ROM monitor.
17400 @kindex target m32rsdi
17401 @item target m32rsdi @var{dev}
17402 Renesas M32R SDI server, connected via parallel port to the board.
17405 The following @value{GDBN} commands are specific to the M32R monitor:
17408 @item set download-path @var{path}
17409 @kindex set download-path
17410 @cindex find downloadable @sc{srec} files (M32R)
17411 Set the default path for finding downloadable @sc{srec} files.
17413 @item show download-path
17414 @kindex show download-path
17415 Show the default path for downloadable @sc{srec} files.
17417 @item set board-address @var{addr}
17418 @kindex set board-address
17419 @cindex M32-EVA target board address
17420 Set the IP address for the M32R-EVA target board.
17422 @item show board-address
17423 @kindex show board-address
17424 Show the current IP address of the target board.
17426 @item set server-address @var{addr}
17427 @kindex set server-address
17428 @cindex download server address (M32R)
17429 Set the IP address for the download server, which is the @value{GDBN}'s
17432 @item show server-address
17433 @kindex show server-address
17434 Display the IP address of the download server.
17436 @item upload @r{[}@var{file}@r{]}
17437 @kindex upload@r{, M32R}
17438 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
17439 upload capability. If no @var{file} argument is given, the current
17440 executable file is uploaded.
17442 @item tload @r{[}@var{file}@r{]}
17443 @kindex tload@r{, M32R}
17444 Test the @code{upload} command.
17447 The following commands are available for M32R/SDI:
17452 @cindex reset SDI connection, M32R
17453 This command resets the SDI connection.
17457 This command shows the SDI connection status.
17460 @kindex debug_chaos
17461 @cindex M32R/Chaos debugging
17462 Instructs the remote that M32R/Chaos debugging is to be used.
17464 @item use_debug_dma
17465 @kindex use_debug_dma
17466 Instructs the remote to use the DEBUG_DMA method of accessing memory.
17469 @kindex use_mon_code
17470 Instructs the remote to use the MON_CODE method of accessing memory.
17473 @kindex use_ib_break
17474 Instructs the remote to set breakpoints by IB break.
17476 @item use_dbt_break
17477 @kindex use_dbt_break
17478 Instructs the remote to set breakpoints by DBT.
17484 The Motorola m68k configuration includes ColdFire support, and a
17485 target command for the following ROM monitor.
17489 @kindex target dbug
17490 @item target dbug @var{dev}
17491 dBUG ROM monitor for Motorola ColdFire.
17496 @subsection MicroBlaze
17497 @cindex Xilinx MicroBlaze
17498 @cindex XMD, Xilinx Microprocessor Debugger
17500 The MicroBlaze is a soft-core processor supported on various Xilinx
17501 FPGAs, such as Spartan or Virtex series. Boards with these processors
17502 usually have JTAG ports which connect to a host system running the Xilinx
17503 Embedded Development Kit (EDK) or Software Development Kit (SDK).
17504 This host system is used to download the configuration bitstream to
17505 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
17506 communicates with the target board using the JTAG interface and
17507 presents a @code{gdbserver} interface to the board. By default
17508 @code{xmd} uses port @code{1234}. (While it is possible to change
17509 this default port, it requires the use of undocumented @code{xmd}
17510 commands. Contact Xilinx support if you need to do this.)
17512 Use these GDB commands to connect to the MicroBlaze target processor.
17515 @item target remote :1234
17516 Use this command to connect to the target if you are running @value{GDBN}
17517 on the same system as @code{xmd}.
17519 @item target remote @var{xmd-host}:1234
17520 Use this command to connect to the target if it is connected to @code{xmd}
17521 running on a different system named @var{xmd-host}.
17524 Use this command to download a program to the MicroBlaze target.
17526 @item set debug microblaze @var{n}
17527 Enable MicroBlaze-specific debugging messages if non-zero.
17529 @item show debug microblaze @var{n}
17530 Show MicroBlaze-specific debugging level.
17533 @node MIPS Embedded
17534 @subsection MIPS Embedded
17536 @cindex MIPS boards
17537 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
17538 MIPS board attached to a serial line. This is available when
17539 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
17542 Use these @value{GDBN} commands to specify the connection to your target board:
17545 @item target mips @var{port}
17546 @kindex target mips @var{port}
17547 To run a program on the board, start up @code{@value{GDBP}} with the
17548 name of your program as the argument. To connect to the board, use the
17549 command @samp{target mips @var{port}}, where @var{port} is the name of
17550 the serial port connected to the board. If the program has not already
17551 been downloaded to the board, you may use the @code{load} command to
17552 download it. You can then use all the usual @value{GDBN} commands.
17554 For example, this sequence connects to the target board through a serial
17555 port, and loads and runs a program called @var{prog} through the
17559 host$ @value{GDBP} @var{prog}
17560 @value{GDBN} is free software and @dots{}
17561 (@value{GDBP}) target mips /dev/ttyb
17562 (@value{GDBP}) load @var{prog}
17566 @item target mips @var{hostname}:@var{portnumber}
17567 On some @value{GDBN} host configurations, you can specify a TCP
17568 connection (for instance, to a serial line managed by a terminal
17569 concentrator) instead of a serial port, using the syntax
17570 @samp{@var{hostname}:@var{portnumber}}.
17572 @item target pmon @var{port}
17573 @kindex target pmon @var{port}
17576 @item target ddb @var{port}
17577 @kindex target ddb @var{port}
17578 NEC's DDB variant of PMON for Vr4300.
17580 @item target lsi @var{port}
17581 @kindex target lsi @var{port}
17582 LSI variant of PMON.
17584 @kindex target r3900
17585 @item target r3900 @var{dev}
17586 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
17588 @kindex target array
17589 @item target array @var{dev}
17590 Array Tech LSI33K RAID controller board.
17596 @value{GDBN} also supports these special commands for MIPS targets:
17599 @item set mipsfpu double
17600 @itemx set mipsfpu single
17601 @itemx set mipsfpu none
17602 @itemx set mipsfpu auto
17603 @itemx show mipsfpu
17604 @kindex set mipsfpu
17605 @kindex show mipsfpu
17606 @cindex MIPS remote floating point
17607 @cindex floating point, MIPS remote
17608 If your target board does not support the MIPS floating point
17609 coprocessor, you should use the command @samp{set mipsfpu none} (if you
17610 need this, you may wish to put the command in your @value{GDBN} init
17611 file). This tells @value{GDBN} how to find the return value of
17612 functions which return floating point values. It also allows
17613 @value{GDBN} to avoid saving the floating point registers when calling
17614 functions on the board. If you are using a floating point coprocessor
17615 with only single precision floating point support, as on the @sc{r4650}
17616 processor, use the command @samp{set mipsfpu single}. The default
17617 double precision floating point coprocessor may be selected using
17618 @samp{set mipsfpu double}.
17620 In previous versions the only choices were double precision or no
17621 floating point, so @samp{set mipsfpu on} will select double precision
17622 and @samp{set mipsfpu off} will select no floating point.
17624 As usual, you can inquire about the @code{mipsfpu} variable with
17625 @samp{show mipsfpu}.
17627 @item set timeout @var{seconds}
17628 @itemx set retransmit-timeout @var{seconds}
17629 @itemx show timeout
17630 @itemx show retransmit-timeout
17631 @cindex @code{timeout}, MIPS protocol
17632 @cindex @code{retransmit-timeout}, MIPS protocol
17633 @kindex set timeout
17634 @kindex show timeout
17635 @kindex set retransmit-timeout
17636 @kindex show retransmit-timeout
17637 You can control the timeout used while waiting for a packet, in the MIPS
17638 remote protocol, with the @code{set timeout @var{seconds}} command. The
17639 default is 5 seconds. Similarly, you can control the timeout used while
17640 waiting for an acknowledgment of a packet with the @code{set
17641 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
17642 You can inspect both values with @code{show timeout} and @code{show
17643 retransmit-timeout}. (These commands are @emph{only} available when
17644 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
17646 The timeout set by @code{set timeout} does not apply when @value{GDBN}
17647 is waiting for your program to stop. In that case, @value{GDBN} waits
17648 forever because it has no way of knowing how long the program is going
17649 to run before stopping.
17651 @item set syn-garbage-limit @var{num}
17652 @kindex set syn-garbage-limit@r{, MIPS remote}
17653 @cindex synchronize with remote MIPS target
17654 Limit the maximum number of characters @value{GDBN} should ignore when
17655 it tries to synchronize with the remote target. The default is 10
17656 characters. Setting the limit to -1 means there's no limit.
17658 @item show syn-garbage-limit
17659 @kindex show syn-garbage-limit@r{, MIPS remote}
17660 Show the current limit on the number of characters to ignore when
17661 trying to synchronize with the remote system.
17663 @item set monitor-prompt @var{prompt}
17664 @kindex set monitor-prompt@r{, MIPS remote}
17665 @cindex remote monitor prompt
17666 Tell @value{GDBN} to expect the specified @var{prompt} string from the
17667 remote monitor. The default depends on the target:
17677 @item show monitor-prompt
17678 @kindex show monitor-prompt@r{, MIPS remote}
17679 Show the current strings @value{GDBN} expects as the prompt from the
17682 @item set monitor-warnings
17683 @kindex set monitor-warnings@r{, MIPS remote}
17684 Enable or disable monitor warnings about hardware breakpoints. This
17685 has effect only for the @code{lsi} target. When on, @value{GDBN} will
17686 display warning messages whose codes are returned by the @code{lsi}
17687 PMON monitor for breakpoint commands.
17689 @item show monitor-warnings
17690 @kindex show monitor-warnings@r{, MIPS remote}
17691 Show the current setting of printing monitor warnings.
17693 @item pmon @var{command}
17694 @kindex pmon@r{, MIPS remote}
17695 @cindex send PMON command
17696 This command allows sending an arbitrary @var{command} string to the
17697 monitor. The monitor must be in debug mode for this to work.
17700 @node OpenRISC 1000
17701 @subsection OpenRISC 1000
17702 @cindex OpenRISC 1000
17704 @cindex or1k boards
17705 See OR1k Architecture document (@uref{www.opencores.org}) for more information
17706 about platform and commands.
17710 @kindex target jtag
17711 @item target jtag jtag://@var{host}:@var{port}
17713 Connects to remote JTAG server.
17714 JTAG remote server can be either an or1ksim or JTAG server,
17715 connected via parallel port to the board.
17717 Example: @code{target jtag jtag://localhost:9999}
17720 @item or1ksim @var{command}
17721 If connected to @code{or1ksim} OpenRISC 1000 Architectural
17722 Simulator, proprietary commands can be executed.
17724 @kindex info or1k spr
17725 @item info or1k spr
17726 Displays spr groups.
17728 @item info or1k spr @var{group}
17729 @itemx info or1k spr @var{groupno}
17730 Displays register names in selected group.
17732 @item info or1k spr @var{group} @var{register}
17733 @itemx info or1k spr @var{register}
17734 @itemx info or1k spr @var{groupno} @var{registerno}
17735 @itemx info or1k spr @var{registerno}
17736 Shows information about specified spr register.
17739 @item spr @var{group} @var{register} @var{value}
17740 @itemx spr @var{register @var{value}}
17741 @itemx spr @var{groupno} @var{registerno @var{value}}
17742 @itemx spr @var{registerno @var{value}}
17743 Writes @var{value} to specified spr register.
17746 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
17747 It is very similar to @value{GDBN} trace, except it does not interfere with normal
17748 program execution and is thus much faster. Hardware breakpoints/watchpoint
17749 triggers can be set using:
17752 Load effective address/data
17754 Store effective address/data
17756 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
17761 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
17762 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
17764 @code{htrace} commands:
17765 @cindex OpenRISC 1000 htrace
17768 @item hwatch @var{conditional}
17769 Set hardware watchpoint on combination of Load/Store Effective Address(es)
17770 or Data. For example:
17772 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
17774 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
17778 Display information about current HW trace configuration.
17780 @item htrace trigger @var{conditional}
17781 Set starting criteria for HW trace.
17783 @item htrace qualifier @var{conditional}
17784 Set acquisition qualifier for HW trace.
17786 @item htrace stop @var{conditional}
17787 Set HW trace stopping criteria.
17789 @item htrace record [@var{data}]*
17790 Selects the data to be recorded, when qualifier is met and HW trace was
17793 @item htrace enable
17794 @itemx htrace disable
17795 Enables/disables the HW trace.
17797 @item htrace rewind [@var{filename}]
17798 Clears currently recorded trace data.
17800 If filename is specified, new trace file is made and any newly collected data
17801 will be written there.
17803 @item htrace print [@var{start} [@var{len}]]
17804 Prints trace buffer, using current record configuration.
17806 @item htrace mode continuous
17807 Set continuous trace mode.
17809 @item htrace mode suspend
17810 Set suspend trace mode.
17814 @node PowerPC Embedded
17815 @subsection PowerPC Embedded
17817 @value{GDBN} provides the following PowerPC-specific commands:
17820 @kindex set powerpc
17821 @item set powerpc soft-float
17822 @itemx show powerpc soft-float
17823 Force @value{GDBN} to use (or not use) a software floating point calling
17824 convention. By default, @value{GDBN} selects the calling convention based
17825 on the selected architecture and the provided executable file.
17827 @item set powerpc vector-abi
17828 @itemx show powerpc vector-abi
17829 Force @value{GDBN} to use the specified calling convention for vector
17830 arguments and return values. The valid options are @samp{auto};
17831 @samp{generic}, to avoid vector registers even if they are present;
17832 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
17833 registers. By default, @value{GDBN} selects the calling convention
17834 based on the selected architecture and the provided executable file.
17836 @kindex target dink32
17837 @item target dink32 @var{dev}
17838 DINK32 ROM monitor.
17840 @kindex target ppcbug
17841 @item target ppcbug @var{dev}
17842 @kindex target ppcbug1
17843 @item target ppcbug1 @var{dev}
17844 PPCBUG ROM monitor for PowerPC.
17847 @item target sds @var{dev}
17848 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
17851 @cindex SDS protocol
17852 The following commands specific to the SDS protocol are supported
17856 @item set sdstimeout @var{nsec}
17857 @kindex set sdstimeout
17858 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
17859 default is 2 seconds.
17861 @item show sdstimeout
17862 @kindex show sdstimeout
17863 Show the current value of the SDS timeout.
17865 @item sds @var{command}
17866 @kindex sds@r{, a command}
17867 Send the specified @var{command} string to the SDS monitor.
17872 @subsection HP PA Embedded
17876 @kindex target op50n
17877 @item target op50n @var{dev}
17878 OP50N monitor, running on an OKI HPPA board.
17880 @kindex target w89k
17881 @item target w89k @var{dev}
17882 W89K monitor, running on a Winbond HPPA board.
17887 @subsection Tsqware Sparclet
17891 @value{GDBN} enables developers to debug tasks running on
17892 Sparclet targets from a Unix host.
17893 @value{GDBN} uses code that runs on
17894 both the Unix host and on the Sparclet target. The program
17895 @code{@value{GDBP}} is installed and executed on the Unix host.
17898 @item remotetimeout @var{args}
17899 @kindex remotetimeout
17900 @value{GDBN} supports the option @code{remotetimeout}.
17901 This option is set by the user, and @var{args} represents the number of
17902 seconds @value{GDBN} waits for responses.
17905 @cindex compiling, on Sparclet
17906 When compiling for debugging, include the options @samp{-g} to get debug
17907 information and @samp{-Ttext} to relocate the program to where you wish to
17908 load it on the target. You may also want to add the options @samp{-n} or
17909 @samp{-N} in order to reduce the size of the sections. Example:
17912 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
17915 You can use @code{objdump} to verify that the addresses are what you intended:
17918 sparclet-aout-objdump --headers --syms prog
17921 @cindex running, on Sparclet
17923 your Unix execution search path to find @value{GDBN}, you are ready to
17924 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
17925 (or @code{sparclet-aout-gdb}, depending on your installation).
17927 @value{GDBN} comes up showing the prompt:
17934 * Sparclet File:: Setting the file to debug
17935 * Sparclet Connection:: Connecting to Sparclet
17936 * Sparclet Download:: Sparclet download
17937 * Sparclet Execution:: Running and debugging
17940 @node Sparclet File
17941 @subsubsection Setting File to Debug
17943 The @value{GDBN} command @code{file} lets you choose with program to debug.
17946 (gdbslet) file prog
17950 @value{GDBN} then attempts to read the symbol table of @file{prog}.
17951 @value{GDBN} locates
17952 the file by searching the directories listed in the command search
17954 If the file was compiled with debug information (option @samp{-g}), source
17955 files will be searched as well.
17956 @value{GDBN} locates
17957 the source files by searching the directories listed in the directory search
17958 path (@pxref{Environment, ,Your Program's Environment}).
17960 to find a file, it displays a message such as:
17963 prog: No such file or directory.
17966 When this happens, add the appropriate directories to the search paths with
17967 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
17968 @code{target} command again.
17970 @node Sparclet Connection
17971 @subsubsection Connecting to Sparclet
17973 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
17974 To connect to a target on serial port ``@code{ttya}'', type:
17977 (gdbslet) target sparclet /dev/ttya
17978 Remote target sparclet connected to /dev/ttya
17979 main () at ../prog.c:3
17983 @value{GDBN} displays messages like these:
17989 @node Sparclet Download
17990 @subsubsection Sparclet Download
17992 @cindex download to Sparclet
17993 Once connected to the Sparclet target,
17994 you can use the @value{GDBN}
17995 @code{load} command to download the file from the host to the target.
17996 The file name and load offset should be given as arguments to the @code{load}
17998 Since the file format is aout, the program must be loaded to the starting
17999 address. You can use @code{objdump} to find out what this value is. The load
18000 offset is an offset which is added to the VMA (virtual memory address)
18001 of each of the file's sections.
18002 For instance, if the program
18003 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
18004 and bss at 0x12010170, in @value{GDBN}, type:
18007 (gdbslet) load prog 0x12010000
18008 Loading section .text, size 0xdb0 vma 0x12010000
18011 If the code is loaded at a different address then what the program was linked
18012 to, you may need to use the @code{section} and @code{add-symbol-file} commands
18013 to tell @value{GDBN} where to map the symbol table.
18015 @node Sparclet Execution
18016 @subsubsection Running and Debugging
18018 @cindex running and debugging Sparclet programs
18019 You can now begin debugging the task using @value{GDBN}'s execution control
18020 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
18021 manual for the list of commands.
18025 Breakpoint 1 at 0x12010000: file prog.c, line 3.
18027 Starting program: prog
18028 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
18029 3 char *symarg = 0;
18031 4 char *execarg = "hello!";
18036 @subsection Fujitsu Sparclite
18040 @kindex target sparclite
18041 @item target sparclite @var{dev}
18042 Fujitsu sparclite boards, used only for the purpose of loading.
18043 You must use an additional command to debug the program.
18044 For example: target remote @var{dev} using @value{GDBN} standard
18050 @subsection Zilog Z8000
18053 @cindex simulator, Z8000
18054 @cindex Zilog Z8000 simulator
18056 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
18059 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
18060 unsegmented variant of the Z8000 architecture) or the Z8001 (the
18061 segmented variant). The simulator recognizes which architecture is
18062 appropriate by inspecting the object code.
18065 @item target sim @var{args}
18067 @kindex target sim@r{, with Z8000}
18068 Debug programs on a simulated CPU. If the simulator supports setup
18069 options, specify them via @var{args}.
18073 After specifying this target, you can debug programs for the simulated
18074 CPU in the same style as programs for your host computer; use the
18075 @code{file} command to load a new program image, the @code{run} command
18076 to run your program, and so on.
18078 As well as making available all the usual machine registers
18079 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
18080 additional items of information as specially named registers:
18085 Counts clock-ticks in the simulator.
18088 Counts instructions run in the simulator.
18091 Execution time in 60ths of a second.
18095 You can refer to these values in @value{GDBN} expressions with the usual
18096 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
18097 conditional breakpoint that suspends only after at least 5000
18098 simulated clock ticks.
18101 @subsection Atmel AVR
18104 When configured for debugging the Atmel AVR, @value{GDBN} supports the
18105 following AVR-specific commands:
18108 @item info io_registers
18109 @kindex info io_registers@r{, AVR}
18110 @cindex I/O registers (Atmel AVR)
18111 This command displays information about the AVR I/O registers. For
18112 each register, @value{GDBN} prints its number and value.
18119 When configured for debugging CRIS, @value{GDBN} provides the
18120 following CRIS-specific commands:
18123 @item set cris-version @var{ver}
18124 @cindex CRIS version
18125 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
18126 The CRIS version affects register names and sizes. This command is useful in
18127 case autodetection of the CRIS version fails.
18129 @item show cris-version
18130 Show the current CRIS version.
18132 @item set cris-dwarf2-cfi
18133 @cindex DWARF-2 CFI and CRIS
18134 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
18135 Change to @samp{off} when using @code{gcc-cris} whose version is below
18138 @item show cris-dwarf2-cfi
18139 Show the current state of using DWARF-2 CFI.
18141 @item set cris-mode @var{mode}
18143 Set the current CRIS mode to @var{mode}. It should only be changed when
18144 debugging in guru mode, in which case it should be set to
18145 @samp{guru} (the default is @samp{normal}).
18147 @item show cris-mode
18148 Show the current CRIS mode.
18152 @subsection Renesas Super-H
18155 For the Renesas Super-H processor, @value{GDBN} provides these
18160 @kindex regs@r{, Super-H}
18161 Show the values of all Super-H registers.
18163 @item set sh calling-convention @var{convention}
18164 @kindex set sh calling-convention
18165 Set the calling-convention used when calling functions from @value{GDBN}.
18166 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
18167 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
18168 convention. If the DWARF-2 information of the called function specifies
18169 that the function follows the Renesas calling convention, the function
18170 is called using the Renesas calling convention. If the calling convention
18171 is set to @samp{renesas}, the Renesas calling convention is always used,
18172 regardless of the DWARF-2 information. This can be used to override the
18173 default of @samp{gcc} if debug information is missing, or the compiler
18174 does not emit the DWARF-2 calling convention entry for a function.
18176 @item show sh calling-convention
18177 @kindex show sh calling-convention
18178 Show the current calling convention setting.
18183 @node Architectures
18184 @section Architectures
18186 This section describes characteristics of architectures that affect
18187 all uses of @value{GDBN} with the architecture, both native and cross.
18194 * HPPA:: HP PA architecture
18195 * SPU:: Cell Broadband Engine SPU architecture
18200 @subsection x86 Architecture-specific Issues
18203 @item set struct-convention @var{mode}
18204 @kindex set struct-convention
18205 @cindex struct return convention
18206 @cindex struct/union returned in registers
18207 Set the convention used by the inferior to return @code{struct}s and
18208 @code{union}s from functions to @var{mode}. Possible values of
18209 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
18210 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
18211 are returned on the stack, while @code{"reg"} means that a
18212 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
18213 be returned in a register.
18215 @item show struct-convention
18216 @kindex show struct-convention
18217 Show the current setting of the convention to return @code{struct}s
18226 @kindex set rstack_high_address
18227 @cindex AMD 29K register stack
18228 @cindex register stack, AMD29K
18229 @item set rstack_high_address @var{address}
18230 On AMD 29000 family processors, registers are saved in a separate
18231 @dfn{register stack}. There is no way for @value{GDBN} to determine the
18232 extent of this stack. Normally, @value{GDBN} just assumes that the
18233 stack is ``large enough''. This may result in @value{GDBN} referencing
18234 memory locations that do not exist. If necessary, you can get around
18235 this problem by specifying the ending address of the register stack with
18236 the @code{set rstack_high_address} command. The argument should be an
18237 address, which you probably want to precede with @samp{0x} to specify in
18240 @kindex show rstack_high_address
18241 @item show rstack_high_address
18242 Display the current limit of the register stack, on AMD 29000 family
18250 See the following section.
18255 @cindex stack on Alpha
18256 @cindex stack on MIPS
18257 @cindex Alpha stack
18259 Alpha- and MIPS-based computers use an unusual stack frame, which
18260 sometimes requires @value{GDBN} to search backward in the object code to
18261 find the beginning of a function.
18263 @cindex response time, MIPS debugging
18264 To improve response time (especially for embedded applications, where
18265 @value{GDBN} may be restricted to a slow serial line for this search)
18266 you may want to limit the size of this search, using one of these
18270 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
18271 @item set heuristic-fence-post @var{limit}
18272 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
18273 search for the beginning of a function. A value of @var{0} (the
18274 default) means there is no limit. However, except for @var{0}, the
18275 larger the limit the more bytes @code{heuristic-fence-post} must search
18276 and therefore the longer it takes to run. You should only need to use
18277 this command when debugging a stripped executable.
18279 @item show heuristic-fence-post
18280 Display the current limit.
18284 These commands are available @emph{only} when @value{GDBN} is configured
18285 for debugging programs on Alpha or MIPS processors.
18287 Several MIPS-specific commands are available when debugging MIPS
18291 @item set mips abi @var{arg}
18292 @kindex set mips abi
18293 @cindex set ABI for MIPS
18294 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
18295 values of @var{arg} are:
18299 The default ABI associated with the current binary (this is the
18310 @item show mips abi
18311 @kindex show mips abi
18312 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
18315 @itemx show mipsfpu
18316 @xref{MIPS Embedded, set mipsfpu}.
18318 @item set mips mask-address @var{arg}
18319 @kindex set mips mask-address
18320 @cindex MIPS addresses, masking
18321 This command determines whether the most-significant 32 bits of 64-bit
18322 MIPS addresses are masked off. The argument @var{arg} can be
18323 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
18324 setting, which lets @value{GDBN} determine the correct value.
18326 @item show mips mask-address
18327 @kindex show mips mask-address
18328 Show whether the upper 32 bits of MIPS addresses are masked off or
18331 @item set remote-mips64-transfers-32bit-regs
18332 @kindex set remote-mips64-transfers-32bit-regs
18333 This command controls compatibility with 64-bit MIPS targets that
18334 transfer data in 32-bit quantities. If you have an old MIPS 64 target
18335 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
18336 and 64 bits for other registers, set this option to @samp{on}.
18338 @item show remote-mips64-transfers-32bit-regs
18339 @kindex show remote-mips64-transfers-32bit-regs
18340 Show the current setting of compatibility with older MIPS 64 targets.
18342 @item set debug mips
18343 @kindex set debug mips
18344 This command turns on and off debugging messages for the MIPS-specific
18345 target code in @value{GDBN}.
18347 @item show debug mips
18348 @kindex show debug mips
18349 Show the current setting of MIPS debugging messages.
18355 @cindex HPPA support
18357 When @value{GDBN} is debugging the HP PA architecture, it provides the
18358 following special commands:
18361 @item set debug hppa
18362 @kindex set debug hppa
18363 This command determines whether HPPA architecture-specific debugging
18364 messages are to be displayed.
18366 @item show debug hppa
18367 Show whether HPPA debugging messages are displayed.
18369 @item maint print unwind @var{address}
18370 @kindex maint print unwind@r{, HPPA}
18371 This command displays the contents of the unwind table entry at the
18372 given @var{address}.
18378 @subsection Cell Broadband Engine SPU architecture
18379 @cindex Cell Broadband Engine
18382 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
18383 it provides the following special commands:
18386 @item info spu event
18388 Display SPU event facility status. Shows current event mask
18389 and pending event status.
18391 @item info spu signal
18392 Display SPU signal notification facility status. Shows pending
18393 signal-control word and signal notification mode of both signal
18394 notification channels.
18396 @item info spu mailbox
18397 Display SPU mailbox facility status. Shows all pending entries,
18398 in order of processing, in each of the SPU Write Outbound,
18399 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
18402 Display MFC DMA status. Shows all pending commands in the MFC
18403 DMA queue. For each entry, opcode, tag, class IDs, effective
18404 and local store addresses and transfer size are shown.
18406 @item info spu proxydma
18407 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
18408 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
18409 and local store addresses and transfer size are shown.
18413 When @value{GDBN} is debugging a combined PowerPC/SPU application
18414 on the Cell Broadband Engine, it provides in addition the following
18418 @item set spu stop-on-load @var{arg}
18420 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
18421 will give control to the user when a new SPE thread enters its @code{main}
18422 function. The default is @code{off}.
18424 @item show spu stop-on-load
18426 Show whether to stop for new SPE threads.
18428 @item set spu auto-flush-cache @var{arg}
18429 Set whether to automatically flush the software-managed cache. When set to
18430 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
18431 cache to be flushed whenever SPE execution stops. This provides a consistent
18432 view of PowerPC memory that is accessed via the cache. If an application
18433 does not use the software-managed cache, this option has no effect.
18435 @item show spu auto-flush-cache
18436 Show whether to automatically flush the software-managed cache.
18441 @subsection PowerPC
18442 @cindex PowerPC architecture
18444 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
18445 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
18446 numbers stored in the floating point registers. These values must be stored
18447 in two consecutive registers, always starting at an even register like
18448 @code{f0} or @code{f2}.
18450 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
18451 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
18452 @code{f2} and @code{f3} for @code{$dl1} and so on.
18454 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
18455 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
18458 @node Controlling GDB
18459 @chapter Controlling @value{GDBN}
18461 You can alter the way @value{GDBN} interacts with you by using the
18462 @code{set} command. For commands controlling how @value{GDBN} displays
18463 data, see @ref{Print Settings, ,Print Settings}. Other settings are
18468 * Editing:: Command editing
18469 * Command History:: Command history
18470 * Screen Size:: Screen size
18471 * Numbers:: Numbers
18472 * ABI:: Configuring the current ABI
18473 * Messages/Warnings:: Optional warnings and messages
18474 * Debugging Output:: Optional messages about internal happenings
18475 * Other Misc Settings:: Other Miscellaneous Settings
18483 @value{GDBN} indicates its readiness to read a command by printing a string
18484 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
18485 can change the prompt string with the @code{set prompt} command. For
18486 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
18487 the prompt in one of the @value{GDBN} sessions so that you can always tell
18488 which one you are talking to.
18490 @emph{Note:} @code{set prompt} does not add a space for you after the
18491 prompt you set. This allows you to set a prompt which ends in a space
18492 or a prompt that does not.
18496 @item set prompt @var{newprompt}
18497 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
18499 @kindex show prompt
18501 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
18505 @section Command Editing
18507 @cindex command line editing
18509 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
18510 @sc{gnu} library provides consistent behavior for programs which provide a
18511 command line interface to the user. Advantages are @sc{gnu} Emacs-style
18512 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
18513 substitution, and a storage and recall of command history across
18514 debugging sessions.
18516 You may control the behavior of command line editing in @value{GDBN} with the
18517 command @code{set}.
18520 @kindex set editing
18523 @itemx set editing on
18524 Enable command line editing (enabled by default).
18526 @item set editing off
18527 Disable command line editing.
18529 @kindex show editing
18531 Show whether command line editing is enabled.
18534 @xref{Command Line Editing}, for more details about the Readline
18535 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
18536 encouraged to read that chapter.
18538 @node Command History
18539 @section Command History
18540 @cindex command history
18542 @value{GDBN} can keep track of the commands you type during your
18543 debugging sessions, so that you can be certain of precisely what
18544 happened. Use these commands to manage the @value{GDBN} command
18547 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
18548 package, to provide the history facility. @xref{Using History
18549 Interactively}, for the detailed description of the History library.
18551 To issue a command to @value{GDBN} without affecting certain aspects of
18552 the state which is seen by users, prefix it with @samp{server }
18553 (@pxref{Server Prefix}). This
18554 means that this command will not affect the command history, nor will it
18555 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
18556 pressed on a line by itself.
18558 @cindex @code{server}, command prefix
18559 The server prefix does not affect the recording of values into the value
18560 history; to print a value without recording it into the value history,
18561 use the @code{output} command instead of the @code{print} command.
18563 Here is the description of @value{GDBN} commands related to command
18567 @cindex history substitution
18568 @cindex history file
18569 @kindex set history filename
18570 @cindex @env{GDBHISTFILE}, environment variable
18571 @item set history filename @var{fname}
18572 Set the name of the @value{GDBN} command history file to @var{fname}.
18573 This is the file where @value{GDBN} reads an initial command history
18574 list, and where it writes the command history from this session when it
18575 exits. You can access this list through history expansion or through
18576 the history command editing characters listed below. This file defaults
18577 to the value of the environment variable @code{GDBHISTFILE}, or to
18578 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
18581 @cindex save command history
18582 @kindex set history save
18583 @item set history save
18584 @itemx set history save on
18585 Record command history in a file, whose name may be specified with the
18586 @code{set history filename} command. By default, this option is disabled.
18588 @item set history save off
18589 Stop recording command history in a file.
18591 @cindex history size
18592 @kindex set history size
18593 @cindex @env{HISTSIZE}, environment variable
18594 @item set history size @var{size}
18595 Set the number of commands which @value{GDBN} keeps in its history list.
18596 This defaults to the value of the environment variable
18597 @code{HISTSIZE}, or to 256 if this variable is not set.
18600 History expansion assigns special meaning to the character @kbd{!}.
18601 @xref{Event Designators}, for more details.
18603 @cindex history expansion, turn on/off
18604 Since @kbd{!} is also the logical not operator in C, history expansion
18605 is off by default. If you decide to enable history expansion with the
18606 @code{set history expansion on} command, you may sometimes need to
18607 follow @kbd{!} (when it is used as logical not, in an expression) with
18608 a space or a tab to prevent it from being expanded. The readline
18609 history facilities do not attempt substitution on the strings
18610 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
18612 The commands to control history expansion are:
18615 @item set history expansion on
18616 @itemx set history expansion
18617 @kindex set history expansion
18618 Enable history expansion. History expansion is off by default.
18620 @item set history expansion off
18621 Disable history expansion.
18624 @kindex show history
18626 @itemx show history filename
18627 @itemx show history save
18628 @itemx show history size
18629 @itemx show history expansion
18630 These commands display the state of the @value{GDBN} history parameters.
18631 @code{show history} by itself displays all four states.
18636 @kindex show commands
18637 @cindex show last commands
18638 @cindex display command history
18639 @item show commands
18640 Display the last ten commands in the command history.
18642 @item show commands @var{n}
18643 Print ten commands centered on command number @var{n}.
18645 @item show commands +
18646 Print ten commands just after the commands last printed.
18650 @section Screen Size
18651 @cindex size of screen
18652 @cindex pauses in output
18654 Certain commands to @value{GDBN} may produce large amounts of
18655 information output to the screen. To help you read all of it,
18656 @value{GDBN} pauses and asks you for input at the end of each page of
18657 output. Type @key{RET} when you want to continue the output, or @kbd{q}
18658 to discard the remaining output. Also, the screen width setting
18659 determines when to wrap lines of output. Depending on what is being
18660 printed, @value{GDBN} tries to break the line at a readable place,
18661 rather than simply letting it overflow onto the following line.
18663 Normally @value{GDBN} knows the size of the screen from the terminal
18664 driver software. For example, on Unix @value{GDBN} uses the termcap data base
18665 together with the value of the @code{TERM} environment variable and the
18666 @code{stty rows} and @code{stty cols} settings. If this is not correct,
18667 you can override it with the @code{set height} and @code{set
18674 @kindex show height
18675 @item set height @var{lpp}
18677 @itemx set width @var{cpl}
18679 These @code{set} commands specify a screen height of @var{lpp} lines and
18680 a screen width of @var{cpl} characters. The associated @code{show}
18681 commands display the current settings.
18683 If you specify a height of zero lines, @value{GDBN} does not pause during
18684 output no matter how long the output is. This is useful if output is to a
18685 file or to an editor buffer.
18687 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
18688 from wrapping its output.
18690 @item set pagination on
18691 @itemx set pagination off
18692 @kindex set pagination
18693 Turn the output pagination on or off; the default is on. Turning
18694 pagination off is the alternative to @code{set height 0}. Note that
18695 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
18696 Options, -batch}) also automatically disables pagination.
18698 @item show pagination
18699 @kindex show pagination
18700 Show the current pagination mode.
18705 @cindex number representation
18706 @cindex entering numbers
18708 You can always enter numbers in octal, decimal, or hexadecimal in
18709 @value{GDBN} by the usual conventions: octal numbers begin with
18710 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
18711 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
18712 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
18713 10; likewise, the default display for numbers---when no particular
18714 format is specified---is base 10. You can change the default base for
18715 both input and output with the commands described below.
18718 @kindex set input-radix
18719 @item set input-radix @var{base}
18720 Set the default base for numeric input. Supported choices
18721 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
18722 specified either unambiguously or using the current input radix; for
18726 set input-radix 012
18727 set input-radix 10.
18728 set input-radix 0xa
18732 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
18733 leaves the input radix unchanged, no matter what it was, since
18734 @samp{10}, being without any leading or trailing signs of its base, is
18735 interpreted in the current radix. Thus, if the current radix is 16,
18736 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
18739 @kindex set output-radix
18740 @item set output-radix @var{base}
18741 Set the default base for numeric display. Supported choices
18742 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
18743 specified either unambiguously or using the current input radix.
18745 @kindex show input-radix
18746 @item show input-radix
18747 Display the current default base for numeric input.
18749 @kindex show output-radix
18750 @item show output-radix
18751 Display the current default base for numeric display.
18753 @item set radix @r{[}@var{base}@r{]}
18757 These commands set and show the default base for both input and output
18758 of numbers. @code{set radix} sets the radix of input and output to
18759 the same base; without an argument, it resets the radix back to its
18760 default value of 10.
18765 @section Configuring the Current ABI
18767 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
18768 application automatically. However, sometimes you need to override its
18769 conclusions. Use these commands to manage @value{GDBN}'s view of the
18776 One @value{GDBN} configuration can debug binaries for multiple operating
18777 system targets, either via remote debugging or native emulation.
18778 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
18779 but you can override its conclusion using the @code{set osabi} command.
18780 One example where this is useful is in debugging of binaries which use
18781 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
18782 not have the same identifying marks that the standard C library for your
18787 Show the OS ABI currently in use.
18790 With no argument, show the list of registered available OS ABI's.
18792 @item set osabi @var{abi}
18793 Set the current OS ABI to @var{abi}.
18796 @cindex float promotion
18798 Generally, the way that an argument of type @code{float} is passed to a
18799 function depends on whether the function is prototyped. For a prototyped
18800 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
18801 according to the architecture's convention for @code{float}. For unprototyped
18802 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
18803 @code{double} and then passed.
18805 Unfortunately, some forms of debug information do not reliably indicate whether
18806 a function is prototyped. If @value{GDBN} calls a function that is not marked
18807 as prototyped, it consults @kbd{set coerce-float-to-double}.
18810 @kindex set coerce-float-to-double
18811 @item set coerce-float-to-double
18812 @itemx set coerce-float-to-double on
18813 Arguments of type @code{float} will be promoted to @code{double} when passed
18814 to an unprototyped function. This is the default setting.
18816 @item set coerce-float-to-double off
18817 Arguments of type @code{float} will be passed directly to unprototyped
18820 @kindex show coerce-float-to-double
18821 @item show coerce-float-to-double
18822 Show the current setting of promoting @code{float} to @code{double}.
18826 @kindex show cp-abi
18827 @value{GDBN} needs to know the ABI used for your program's C@t{++}
18828 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
18829 used to build your application. @value{GDBN} only fully supports
18830 programs with a single C@t{++} ABI; if your program contains code using
18831 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
18832 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
18833 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
18834 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
18835 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
18836 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
18841 Show the C@t{++} ABI currently in use.
18844 With no argument, show the list of supported C@t{++} ABI's.
18846 @item set cp-abi @var{abi}
18847 @itemx set cp-abi auto
18848 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
18851 @node Messages/Warnings
18852 @section Optional Warnings and Messages
18854 @cindex verbose operation
18855 @cindex optional warnings
18856 By default, @value{GDBN} is silent about its inner workings. If you are
18857 running on a slow machine, you may want to use the @code{set verbose}
18858 command. This makes @value{GDBN} tell you when it does a lengthy
18859 internal operation, so you will not think it has crashed.
18861 Currently, the messages controlled by @code{set verbose} are those
18862 which announce that the symbol table for a source file is being read;
18863 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
18866 @kindex set verbose
18867 @item set verbose on
18868 Enables @value{GDBN} output of certain informational messages.
18870 @item set verbose off
18871 Disables @value{GDBN} output of certain informational messages.
18873 @kindex show verbose
18875 Displays whether @code{set verbose} is on or off.
18878 By default, if @value{GDBN} encounters bugs in the symbol table of an
18879 object file, it is silent; but if you are debugging a compiler, you may
18880 find this information useful (@pxref{Symbol Errors, ,Errors Reading
18885 @kindex set complaints
18886 @item set complaints @var{limit}
18887 Permits @value{GDBN} to output @var{limit} complaints about each type of
18888 unusual symbols before becoming silent about the problem. Set
18889 @var{limit} to zero to suppress all complaints; set it to a large number
18890 to prevent complaints from being suppressed.
18892 @kindex show complaints
18893 @item show complaints
18894 Displays how many symbol complaints @value{GDBN} is permitted to produce.
18898 @anchor{confirmation requests}
18899 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
18900 lot of stupid questions to confirm certain commands. For example, if
18901 you try to run a program which is already running:
18905 The program being debugged has been started already.
18906 Start it from the beginning? (y or n)
18909 If you are willing to unflinchingly face the consequences of your own
18910 commands, you can disable this ``feature'':
18914 @kindex set confirm
18916 @cindex confirmation
18917 @cindex stupid questions
18918 @item set confirm off
18919 Disables confirmation requests. Note that running @value{GDBN} with
18920 the @option{--batch} option (@pxref{Mode Options, -batch}) also
18921 automatically disables confirmation requests.
18923 @item set confirm on
18924 Enables confirmation requests (the default).
18926 @kindex show confirm
18928 Displays state of confirmation requests.
18932 @cindex command tracing
18933 If you need to debug user-defined commands or sourced files you may find it
18934 useful to enable @dfn{command tracing}. In this mode each command will be
18935 printed as it is executed, prefixed with one or more @samp{+} symbols, the
18936 quantity denoting the call depth of each command.
18939 @kindex set trace-commands
18940 @cindex command scripts, debugging
18941 @item set trace-commands on
18942 Enable command tracing.
18943 @item set trace-commands off
18944 Disable command tracing.
18945 @item show trace-commands
18946 Display the current state of command tracing.
18949 @node Debugging Output
18950 @section Optional Messages about Internal Happenings
18951 @cindex optional debugging messages
18953 @value{GDBN} has commands that enable optional debugging messages from
18954 various @value{GDBN} subsystems; normally these commands are of
18955 interest to @value{GDBN} maintainers, or when reporting a bug. This
18956 section documents those commands.
18959 @kindex set exec-done-display
18960 @item set exec-done-display
18961 Turns on or off the notification of asynchronous commands'
18962 completion. When on, @value{GDBN} will print a message when an
18963 asynchronous command finishes its execution. The default is off.
18964 @kindex show exec-done-display
18965 @item show exec-done-display
18966 Displays the current setting of asynchronous command completion
18969 @cindex gdbarch debugging info
18970 @cindex architecture debugging info
18971 @item set debug arch
18972 Turns on or off display of gdbarch debugging info. The default is off
18974 @item show debug arch
18975 Displays the current state of displaying gdbarch debugging info.
18976 @item set debug aix-thread
18977 @cindex AIX threads
18978 Display debugging messages about inner workings of the AIX thread
18980 @item show debug aix-thread
18981 Show the current state of AIX thread debugging info display.
18982 @item set debug dwarf2-die
18983 @cindex DWARF2 DIEs
18984 Dump DWARF2 DIEs after they are read in.
18985 The value is the number of nesting levels to print.
18986 A value of zero turns off the display.
18987 @item show debug dwarf2-die
18988 Show the current state of DWARF2 DIE debugging.
18989 @item set debug displaced
18990 @cindex displaced stepping debugging info
18991 Turns on or off display of @value{GDBN} debugging info for the
18992 displaced stepping support. The default is off.
18993 @item show debug displaced
18994 Displays the current state of displaying @value{GDBN} debugging info
18995 related to displaced stepping.
18996 @item set debug event
18997 @cindex event debugging info
18998 Turns on or off display of @value{GDBN} event debugging info. The
19000 @item show debug event
19001 Displays the current state of displaying @value{GDBN} event debugging
19003 @item set debug expression
19004 @cindex expression debugging info
19005 Turns on or off display of debugging info about @value{GDBN}
19006 expression parsing. The default is off.
19007 @item show debug expression
19008 Displays the current state of displaying debugging info about
19009 @value{GDBN} expression parsing.
19010 @item set debug frame
19011 @cindex frame debugging info
19012 Turns on or off display of @value{GDBN} frame debugging info. The
19014 @item show debug frame
19015 Displays the current state of displaying @value{GDBN} frame debugging
19017 @item set debug gnu-nat
19018 @cindex @sc{gnu}/Hurd debug messages
19019 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
19020 @item show debug gnu-nat
19021 Show the current state of @sc{gnu}/Hurd debugging messages.
19022 @item set debug infrun
19023 @cindex inferior debugging info
19024 Turns on or off display of @value{GDBN} debugging info for running the inferior.
19025 The default is off. @file{infrun.c} contains GDB's runtime state machine used
19026 for implementing operations such as single-stepping the inferior.
19027 @item show debug infrun
19028 Displays the current state of @value{GDBN} inferior debugging.
19029 @item set debug lin-lwp
19030 @cindex @sc{gnu}/Linux LWP debug messages
19031 @cindex Linux lightweight processes
19032 Turns on or off debugging messages from the Linux LWP debug support.
19033 @item show debug lin-lwp
19034 Show the current state of Linux LWP debugging messages.
19035 @item set debug lin-lwp-async
19036 @cindex @sc{gnu}/Linux LWP async debug messages
19037 @cindex Linux lightweight processes
19038 Turns on or off debugging messages from the Linux LWP async debug support.
19039 @item show debug lin-lwp-async
19040 Show the current state of Linux LWP async debugging messages.
19041 @item set debug observer
19042 @cindex observer debugging info
19043 Turns on or off display of @value{GDBN} observer debugging. This
19044 includes info such as the notification of observable events.
19045 @item show debug observer
19046 Displays the current state of observer debugging.
19047 @item set debug overload
19048 @cindex C@t{++} overload debugging info
19049 Turns on or off display of @value{GDBN} C@t{++} overload debugging
19050 info. This includes info such as ranking of functions, etc. The default
19052 @item show debug overload
19053 Displays the current state of displaying @value{GDBN} C@t{++} overload
19055 @cindex expression parser, debugging info
19056 @cindex debug expression parser
19057 @item set debug parser
19058 Turns on or off the display of expression parser debugging output.
19059 Internally, this sets the @code{yydebug} variable in the expression
19060 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
19061 details. The default is off.
19062 @item show debug parser
19063 Show the current state of expression parser debugging.
19064 @cindex packets, reporting on stdout
19065 @cindex serial connections, debugging
19066 @cindex debug remote protocol
19067 @cindex remote protocol debugging
19068 @cindex display remote packets
19069 @item set debug remote
19070 Turns on or off display of reports on all packets sent back and forth across
19071 the serial line to the remote machine. The info is printed on the
19072 @value{GDBN} standard output stream. The default is off.
19073 @item show debug remote
19074 Displays the state of display of remote packets.
19075 @item set debug serial
19076 Turns on or off display of @value{GDBN} serial debugging info. The
19078 @item show debug serial
19079 Displays the current state of displaying @value{GDBN} serial debugging
19081 @item set debug solib-frv
19082 @cindex FR-V shared-library debugging
19083 Turns on or off debugging messages for FR-V shared-library code.
19084 @item show debug solib-frv
19085 Display the current state of FR-V shared-library code debugging
19087 @item set debug target
19088 @cindex target debugging info
19089 Turns on or off display of @value{GDBN} target debugging info. This info
19090 includes what is going on at the target level of GDB, as it happens. The
19091 default is 0. Set it to 1 to track events, and to 2 to also track the
19092 value of large memory transfers. Changes to this flag do not take effect
19093 until the next time you connect to a target or use the @code{run} command.
19094 @item show debug target
19095 Displays the current state of displaying @value{GDBN} target debugging
19097 @item set debug timestamp
19098 @cindex timestampping debugging info
19099 Turns on or off display of timestamps with @value{GDBN} debugging info.
19100 When enabled, seconds and microseconds are displayed before each debugging
19102 @item show debug timestamp
19103 Displays the current state of displaying timestamps with @value{GDBN}
19105 @item set debugvarobj
19106 @cindex variable object debugging info
19107 Turns on or off display of @value{GDBN} variable object debugging
19108 info. The default is off.
19109 @item show debugvarobj
19110 Displays the current state of displaying @value{GDBN} variable object
19112 @item set debug xml
19113 @cindex XML parser debugging
19114 Turns on or off debugging messages for built-in XML parsers.
19115 @item show debug xml
19116 Displays the current state of XML debugging messages.
19119 @node Other Misc Settings
19120 @section Other Miscellaneous Settings
19121 @cindex miscellaneous settings
19124 @kindex set interactive-mode
19125 @item set interactive-mode
19126 If @code{on}, forces @value{GDBN} to operate interactively.
19127 If @code{off}, forces @value{GDBN} to operate non-interactively,
19128 If @code{auto} (the default), @value{GDBN} guesses which mode to use,
19129 based on whether the debugger was started in a terminal or not.
19131 In the vast majority of cases, the debugger should be able to guess
19132 correctly which mode should be used. But this setting can be useful
19133 in certain specific cases, such as running a MinGW @value{GDBN}
19134 inside a cygwin window.
19136 @kindex show interactive-mode
19137 @item show interactive-mode
19138 Displays whether the debugger is operating in interactive mode or not.
19141 @node Extending GDB
19142 @chapter Extending @value{GDBN}
19143 @cindex extending GDB
19145 @value{GDBN} provides two mechanisms for extension. The first is based
19146 on composition of @value{GDBN} commands, and the second is based on the
19147 Python scripting language.
19149 To facilitate the use of these extensions, @value{GDBN} is capable
19150 of evaluating the contents of a file. When doing so, @value{GDBN}
19151 can recognize which scripting language is being used by looking at
19152 the filename extension. Files with an unrecognized filename extension
19153 are always treated as a @value{GDBN} Command Files.
19154 @xref{Command Files,, Command files}.
19156 You can control how @value{GDBN} evaluates these files with the following
19160 @kindex set script-extension
19161 @kindex show script-extension
19162 @item set script-extension off
19163 All scripts are always evaluated as @value{GDBN} Command Files.
19165 @item set script-extension soft
19166 The debugger determines the scripting language based on filename
19167 extension. If this scripting language is supported, @value{GDBN}
19168 evaluates the script using that language. Otherwise, it evaluates
19169 the file as a @value{GDBN} Command File.
19171 @item set script-extension strict
19172 The debugger determines the scripting language based on filename
19173 extension, and evaluates the script using that language. If the
19174 language is not supported, then the evaluation fails.
19176 @item show script-extension
19177 Display the current value of the @code{script-extension} option.
19182 * Sequences:: Canned Sequences of Commands
19183 * Python:: Scripting @value{GDBN} using Python
19187 @section Canned Sequences of Commands
19189 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
19190 Command Lists}), @value{GDBN} provides two ways to store sequences of
19191 commands for execution as a unit: user-defined commands and command
19195 * Define:: How to define your own commands
19196 * Hooks:: Hooks for user-defined commands
19197 * Command Files:: How to write scripts of commands to be stored in a file
19198 * Output:: Commands for controlled output
19202 @subsection User-defined Commands
19204 @cindex user-defined command
19205 @cindex arguments, to user-defined commands
19206 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
19207 which you assign a new name as a command. This is done with the
19208 @code{define} command. User commands may accept up to 10 arguments
19209 separated by whitespace. Arguments are accessed within the user command
19210 via @code{$arg0@dots{}$arg9}. A trivial example:
19214 print $arg0 + $arg1 + $arg2
19219 To execute the command use:
19226 This defines the command @code{adder}, which prints the sum of
19227 its three arguments. Note the arguments are text substitutions, so they may
19228 reference variables, use complex expressions, or even perform inferior
19231 @cindex argument count in user-defined commands
19232 @cindex how many arguments (user-defined commands)
19233 In addition, @code{$argc} may be used to find out how many arguments have
19234 been passed. This expands to a number in the range 0@dots{}10.
19239 print $arg0 + $arg1
19242 print $arg0 + $arg1 + $arg2
19250 @item define @var{commandname}
19251 Define a command named @var{commandname}. If there is already a command
19252 by that name, you are asked to confirm that you want to redefine it.
19253 @var{commandname} may be a bare command name consisting of letters,
19254 numbers, dashes, and underscores. It may also start with any predefined
19255 prefix command. For example, @samp{define target my-target} creates
19256 a user-defined @samp{target my-target} command.
19258 The definition of the command is made up of other @value{GDBN} command lines,
19259 which are given following the @code{define} command. The end of these
19260 commands is marked by a line containing @code{end}.
19263 @kindex end@r{ (user-defined commands)}
19264 @item document @var{commandname}
19265 Document the user-defined command @var{commandname}, so that it can be
19266 accessed by @code{help}. The command @var{commandname} must already be
19267 defined. This command reads lines of documentation just as @code{define}
19268 reads the lines of the command definition, ending with @code{end}.
19269 After the @code{document} command is finished, @code{help} on command
19270 @var{commandname} displays the documentation you have written.
19272 You may use the @code{document} command again to change the
19273 documentation of a command. Redefining the command with @code{define}
19274 does not change the documentation.
19276 @kindex dont-repeat
19277 @cindex don't repeat command
19279 Used inside a user-defined command, this tells @value{GDBN} that this
19280 command should not be repeated when the user hits @key{RET}
19281 (@pxref{Command Syntax, repeat last command}).
19283 @kindex help user-defined
19284 @item help user-defined
19285 List all user-defined commands, with the first line of the documentation
19290 @itemx show user @var{commandname}
19291 Display the @value{GDBN} commands used to define @var{commandname} (but
19292 not its documentation). If no @var{commandname} is given, display the
19293 definitions for all user-defined commands.
19295 @cindex infinite recursion in user-defined commands
19296 @kindex show max-user-call-depth
19297 @kindex set max-user-call-depth
19298 @item show max-user-call-depth
19299 @itemx set max-user-call-depth
19300 The value of @code{max-user-call-depth} controls how many recursion
19301 levels are allowed in user-defined commands before @value{GDBN} suspects an
19302 infinite recursion and aborts the command.
19305 In addition to the above commands, user-defined commands frequently
19306 use control flow commands, described in @ref{Command Files}.
19308 When user-defined commands are executed, the
19309 commands of the definition are not printed. An error in any command
19310 stops execution of the user-defined command.
19312 If used interactively, commands that would ask for confirmation proceed
19313 without asking when used inside a user-defined command. Many @value{GDBN}
19314 commands that normally print messages to say what they are doing omit the
19315 messages when used in a user-defined command.
19318 @subsection User-defined Command Hooks
19319 @cindex command hooks
19320 @cindex hooks, for commands
19321 @cindex hooks, pre-command
19324 You may define @dfn{hooks}, which are a special kind of user-defined
19325 command. Whenever you run the command @samp{foo}, if the user-defined
19326 command @samp{hook-foo} exists, it is executed (with no arguments)
19327 before that command.
19329 @cindex hooks, post-command
19331 A hook may also be defined which is run after the command you executed.
19332 Whenever you run the command @samp{foo}, if the user-defined command
19333 @samp{hookpost-foo} exists, it is executed (with no arguments) after
19334 that command. Post-execution hooks may exist simultaneously with
19335 pre-execution hooks, for the same command.
19337 It is valid for a hook to call the command which it hooks. If this
19338 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
19340 @c It would be nice if hookpost could be passed a parameter indicating
19341 @c if the command it hooks executed properly or not. FIXME!
19343 @kindex stop@r{, a pseudo-command}
19344 In addition, a pseudo-command, @samp{stop} exists. Defining
19345 (@samp{hook-stop}) makes the associated commands execute every time
19346 execution stops in your program: before breakpoint commands are run,
19347 displays are printed, or the stack frame is printed.
19349 For example, to ignore @code{SIGALRM} signals while
19350 single-stepping, but treat them normally during normal execution,
19355 handle SIGALRM nopass
19359 handle SIGALRM pass
19362 define hook-continue
19363 handle SIGALRM pass
19367 As a further example, to hook at the beginning and end of the @code{echo}
19368 command, and to add extra text to the beginning and end of the message,
19376 define hookpost-echo
19380 (@value{GDBP}) echo Hello World
19381 <<<---Hello World--->>>
19386 You can define a hook for any single-word command in @value{GDBN}, but
19387 not for command aliases; you should define a hook for the basic command
19388 name, e.g.@: @code{backtrace} rather than @code{bt}.
19389 @c FIXME! So how does Joe User discover whether a command is an alias
19391 You can hook a multi-word command by adding @code{hook-} or
19392 @code{hookpost-} to the last word of the command, e.g.@:
19393 @samp{define target hook-remote} to add a hook to @samp{target remote}.
19395 If an error occurs during the execution of your hook, execution of
19396 @value{GDBN} commands stops and @value{GDBN} issues a prompt
19397 (before the command that you actually typed had a chance to run).
19399 If you try to define a hook which does not match any known command, you
19400 get a warning from the @code{define} command.
19402 @node Command Files
19403 @subsection Command Files
19405 @cindex command files
19406 @cindex scripting commands
19407 A command file for @value{GDBN} is a text file made of lines that are
19408 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
19409 also be included. An empty line in a command file does nothing; it
19410 does not mean to repeat the last command, as it would from the
19413 You can request the execution of a command file with the @code{source}
19414 command. Note that the @code{source} command is also used to evaluate
19415 scripts that are not Command Files. The exact behavior can be configured
19416 using the @code{script-extension} setting.
19417 @xref{Extending GDB,, Extending GDB}.
19421 @cindex execute commands from a file
19422 @item source [-s] [-v] @var{filename}
19423 Execute the command file @var{filename}.
19426 The lines in a command file are generally executed sequentially,
19427 unless the order of execution is changed by one of the
19428 @emph{flow-control commands} described below. The commands are not
19429 printed as they are executed. An error in any command terminates
19430 execution of the command file and control is returned to the console.
19432 @value{GDBN} first searches for @var{filename} in the current directory.
19433 If the file is not found there, and @var{filename} does not specify a
19434 directory, then @value{GDBN} also looks for the file on the source search path
19435 (specified with the @samp{directory} command);
19436 except that @file{$cdir} is not searched because the compilation directory
19437 is not relevant to scripts.
19439 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
19440 on the search path even if @var{filename} specifies a directory.
19441 The search is done by appending @var{filename} to each element of the
19442 search path. So, for example, if @var{filename} is @file{mylib/myscript}
19443 and the search path contains @file{/home/user} then @value{GDBN} will
19444 look for the script @file{/home/user/mylib/myscript}.
19445 The search is also done if @var{filename} is an absolute path.
19446 For example, if @var{filename} is @file{/tmp/myscript} and
19447 the search path contains @file{/home/user} then @value{GDBN} will
19448 look for the script @file{/home/user/tmp/myscript}.
19449 For DOS-like systems, if @var{filename} contains a drive specification,
19450 it is stripped before concatenation. For example, if @var{filename} is
19451 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
19452 will look for the script @file{c:/tmp/myscript}.
19454 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
19455 each command as it is executed. The option must be given before
19456 @var{filename}, and is interpreted as part of the filename anywhere else.
19458 Commands that would ask for confirmation if used interactively proceed
19459 without asking when used in a command file. Many @value{GDBN} commands that
19460 normally print messages to say what they are doing omit the messages
19461 when called from command files.
19463 @value{GDBN} also accepts command input from standard input. In this
19464 mode, normal output goes to standard output and error output goes to
19465 standard error. Errors in a command file supplied on standard input do
19466 not terminate execution of the command file---execution continues with
19470 gdb < cmds > log 2>&1
19473 (The syntax above will vary depending on the shell used.) This example
19474 will execute commands from the file @file{cmds}. All output and errors
19475 would be directed to @file{log}.
19477 Since commands stored on command files tend to be more general than
19478 commands typed interactively, they frequently need to deal with
19479 complicated situations, such as different or unexpected values of
19480 variables and symbols, changes in how the program being debugged is
19481 built, etc. @value{GDBN} provides a set of flow-control commands to
19482 deal with these complexities. Using these commands, you can write
19483 complex scripts that loop over data structures, execute commands
19484 conditionally, etc.
19491 This command allows to include in your script conditionally executed
19492 commands. The @code{if} command takes a single argument, which is an
19493 expression to evaluate. It is followed by a series of commands that
19494 are executed only if the expression is true (its value is nonzero).
19495 There can then optionally be an @code{else} line, followed by a series
19496 of commands that are only executed if the expression was false. The
19497 end of the list is marked by a line containing @code{end}.
19501 This command allows to write loops. Its syntax is similar to
19502 @code{if}: the command takes a single argument, which is an expression
19503 to evaluate, and must be followed by the commands to execute, one per
19504 line, terminated by an @code{end}. These commands are called the
19505 @dfn{body} of the loop. The commands in the body of @code{while} are
19506 executed repeatedly as long as the expression evaluates to true.
19510 This command exits the @code{while} loop in whose body it is included.
19511 Execution of the script continues after that @code{while}s @code{end}
19514 @kindex loop_continue
19515 @item loop_continue
19516 This command skips the execution of the rest of the body of commands
19517 in the @code{while} loop in whose body it is included. Execution
19518 branches to the beginning of the @code{while} loop, where it evaluates
19519 the controlling expression.
19521 @kindex end@r{ (if/else/while commands)}
19523 Terminate the block of commands that are the body of @code{if},
19524 @code{else}, or @code{while} flow-control commands.
19529 @subsection Commands for Controlled Output
19531 During the execution of a command file or a user-defined command, normal
19532 @value{GDBN} output is suppressed; the only output that appears is what is
19533 explicitly printed by the commands in the definition. This section
19534 describes three commands useful for generating exactly the output you
19539 @item echo @var{text}
19540 @c I do not consider backslash-space a standard C escape sequence
19541 @c because it is not in ANSI.
19542 Print @var{text}. Nonprinting characters can be included in
19543 @var{text} using C escape sequences, such as @samp{\n} to print a
19544 newline. @strong{No newline is printed unless you specify one.}
19545 In addition to the standard C escape sequences, a backslash followed
19546 by a space stands for a space. This is useful for displaying a
19547 string with spaces at the beginning or the end, since leading and
19548 trailing spaces are otherwise trimmed from all arguments.
19549 To print @samp{@w{ }and foo =@w{ }}, use the command
19550 @samp{echo \@w{ }and foo = \@w{ }}.
19552 A backslash at the end of @var{text} can be used, as in C, to continue
19553 the command onto subsequent lines. For example,
19556 echo This is some text\n\
19557 which is continued\n\
19558 onto several lines.\n
19561 produces the same output as
19564 echo This is some text\n
19565 echo which is continued\n
19566 echo onto several lines.\n
19570 @item output @var{expression}
19571 Print the value of @var{expression} and nothing but that value: no
19572 newlines, no @samp{$@var{nn} = }. The value is not entered in the
19573 value history either. @xref{Expressions, ,Expressions}, for more information
19576 @item output/@var{fmt} @var{expression}
19577 Print the value of @var{expression} in format @var{fmt}. You can use
19578 the same formats as for @code{print}. @xref{Output Formats,,Output
19579 Formats}, for more information.
19582 @item printf @var{template}, @var{expressions}@dots{}
19583 Print the values of one or more @var{expressions} under the control of
19584 the string @var{template}. To print several values, make
19585 @var{expressions} be a comma-separated list of individual expressions,
19586 which may be either numbers or pointers. Their values are printed as
19587 specified by @var{template}, exactly as a C program would do by
19588 executing the code below:
19591 printf (@var{template}, @var{expressions}@dots{});
19594 As in @code{C} @code{printf}, ordinary characters in @var{template}
19595 are printed verbatim, while @dfn{conversion specification} introduced
19596 by the @samp{%} character cause subsequent @var{expressions} to be
19597 evaluated, their values converted and formatted according to type and
19598 style information encoded in the conversion specifications, and then
19601 For example, you can print two values in hex like this:
19604 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
19607 @code{printf} supports all the standard @code{C} conversion
19608 specifications, including the flags and modifiers between the @samp{%}
19609 character and the conversion letter, with the following exceptions:
19613 The argument-ordering modifiers, such as @samp{2$}, are not supported.
19616 The modifier @samp{*} is not supported for specifying precision or
19620 The @samp{'} flag (for separation of digits into groups according to
19621 @code{LC_NUMERIC'}) is not supported.
19624 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
19628 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
19631 The conversion letters @samp{a} and @samp{A} are not supported.
19635 Note that the @samp{ll} type modifier is supported only if the
19636 underlying @code{C} implementation used to build @value{GDBN} supports
19637 the @code{long long int} type, and the @samp{L} type modifier is
19638 supported only if @code{long double} type is available.
19640 As in @code{C}, @code{printf} supports simple backslash-escape
19641 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
19642 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
19643 single character. Octal and hexadecimal escape sequences are not
19646 Additionally, @code{printf} supports conversion specifications for DFP
19647 (@dfn{Decimal Floating Point}) types using the following length modifiers
19648 together with a floating point specifier.
19653 @samp{H} for printing @code{Decimal32} types.
19656 @samp{D} for printing @code{Decimal64} types.
19659 @samp{DD} for printing @code{Decimal128} types.
19662 If the underlying @code{C} implementation used to build @value{GDBN} has
19663 support for the three length modifiers for DFP types, other modifiers
19664 such as width and precision will also be available for @value{GDBN} to use.
19666 In case there is no such @code{C} support, no additional modifiers will be
19667 available and the value will be printed in the standard way.
19669 Here's an example of printing DFP types using the above conversion letters:
19671 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
19677 @section Scripting @value{GDBN} using Python
19678 @cindex python scripting
19679 @cindex scripting with python
19681 You can script @value{GDBN} using the @uref{http://www.python.org/,
19682 Python programming language}. This feature is available only if
19683 @value{GDBN} was configured using @option{--with-python}.
19686 * Python Commands:: Accessing Python from @value{GDBN}.
19687 * Python API:: Accessing @value{GDBN} from Python.
19690 @node Python Commands
19691 @subsection Python Commands
19692 @cindex python commands
19693 @cindex commands to access python
19695 @value{GDBN} provides one command for accessing the Python interpreter,
19696 and one related setting:
19700 @item python @r{[}@var{code}@r{]}
19701 The @code{python} command can be used to evaluate Python code.
19703 If given an argument, the @code{python} command will evaluate the
19704 argument as a Python command. For example:
19707 (@value{GDBP}) python print 23
19711 If you do not provide an argument to @code{python}, it will act as a
19712 multi-line command, like @code{define}. In this case, the Python
19713 script is made up of subsequent command lines, given after the
19714 @code{python} command. This command list is terminated using a line
19715 containing @code{end}. For example:
19718 (@value{GDBP}) python
19720 End with a line saying just "end".
19726 @kindex maint set python print-stack
19727 @item maint set python print-stack
19728 By default, @value{GDBN} will print a stack trace when an error occurs
19729 in a Python script. This can be controlled using @code{maint set
19730 python print-stack}: if @code{on}, the default, then Python stack
19731 printing is enabled; if @code{off}, then Python stack printing is
19735 It is also possible to execute a Python script from the @value{GDBN}
19739 @item source @file{script-name}
19740 The script name must end with @samp{.py} and @value{GDBN} must be configured
19741 to recognize the script language based on filename extension using
19742 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
19744 @item python execfile ("script-name")
19745 This method is based on the @code{execfile} Python built-in function,
19746 and thus is always available.
19750 @subsection Python API
19752 @cindex programming in python
19754 @cindex python stdout
19755 @cindex python pagination
19756 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
19757 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
19758 A Python program which outputs to one of these streams may have its
19759 output interrupted by the user (@pxref{Screen Size}). In this
19760 situation, a Python @code{KeyboardInterrupt} exception is thrown.
19763 * Basic Python:: Basic Python Functions.
19764 * Exception Handling::
19765 * Auto-loading:: Automatically loading Python code.
19766 * Values From Inferior::
19767 * Types In Python:: Python representation of types.
19768 * Pretty Printing:: Pretty-printing values.
19769 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
19770 * Commands In Python:: Implementing new commands in Python.
19771 * Functions In Python:: Writing new convenience functions.
19772 * Progspaces In Python:: Program spaces.
19773 * Objfiles In Python:: Object files.
19774 * Frames In Python:: Accessing inferior stack frames from Python.
19775 * Blocks In Python:: Accessing frame blocks from Python.
19776 * Symbols In Python:: Python representation of symbols.
19777 * Symbol Tables In Python:: Python representation of symbol tables.
19778 * Lazy Strings In Python:: Python representation of lazy strings.
19779 * Breakpoints In Python:: Manipulating breakpoints using Python.
19783 @subsubsection Basic Python
19785 @cindex python functions
19786 @cindex python module
19788 @value{GDBN} introduces a new Python module, named @code{gdb}. All
19789 methods and classes added by @value{GDBN} are placed in this module.
19790 @value{GDBN} automatically @code{import}s the @code{gdb} module for
19791 use in all scripts evaluated by the @code{python} command.
19793 @findex gdb.execute
19794 @defun execute command [from_tty]
19795 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
19796 If a GDB exception happens while @var{command} runs, it is
19797 translated as described in @ref{Exception Handling,,Exception Handling}.
19798 If no exceptions occur, this function returns @code{None}.
19800 @var{from_tty} specifies whether @value{GDBN} ought to consider this
19801 command as having originated from the user invoking it interactively.
19802 It must be a boolean value. If omitted, it defaults to @code{False}.
19805 @findex gdb.breakpoints
19807 Return a sequence holding all of @value{GDBN}'s breakpoints.
19808 @xref{Breakpoints In Python}, for more information.
19811 @findex gdb.parameter
19812 @defun parameter parameter
19813 Return the value of a @value{GDBN} parameter. @var{parameter} is a
19814 string naming the parameter to look up; @var{parameter} may contain
19815 spaces if the parameter has a multi-part name. For example,
19816 @samp{print object} is a valid parameter name.
19818 If the named parameter does not exist, this function throws a
19819 @code{RuntimeError}. Otherwise, the parameter's value is converted to
19820 a Python value of the appropriate type, and returned.
19823 @findex gdb.history
19824 @defun history number
19825 Return a value from @value{GDBN}'s value history (@pxref{Value
19826 History}). @var{number} indicates which history element to return.
19827 If @var{number} is negative, then @value{GDBN} will take its absolute value
19828 and count backward from the last element (i.e., the most recent element) to
19829 find the value to return. If @var{number} is zero, then @value{GDBN} will
19830 return the most recent element. If the element specified by @var{number}
19831 doesn't exist in the value history, a @code{RuntimeError} exception will be
19834 If no exception is raised, the return value is always an instance of
19835 @code{gdb.Value} (@pxref{Values From Inferior}).
19838 @findex gdb.parse_and_eval
19839 @defun parse_and_eval expression
19840 Parse @var{expression} as an expression in the current language,
19841 evaluate it, and return the result as a @code{gdb.Value}.
19842 @var{expression} must be a string.
19844 This function can be useful when implementing a new command
19845 (@pxref{Commands In Python}), as it provides a way to parse the
19846 command's argument as an expression. It is also useful simply to
19847 compute values, for example, it is the only way to get the value of a
19848 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
19852 @defun write string
19853 Print a string to @value{GDBN}'s paginated standard output stream.
19854 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
19855 call this function.
19860 Flush @value{GDBN}'s paginated standard output stream. Flushing
19861 @code{sys.stdout} or @code{sys.stderr} will automatically call this
19865 @findex gdb.target_charset
19866 @defun target_charset
19867 Return the name of the current target character set (@pxref{Character
19868 Sets}). This differs from @code{gdb.parameter('target-charset')} in
19869 that @samp{auto} is never returned.
19872 @findex gdb.target_wide_charset
19873 @defun target_wide_charset
19874 Return the name of the current target wide character set
19875 (@pxref{Character Sets}). This differs from
19876 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
19880 @node Exception Handling
19881 @subsubsection Exception Handling
19882 @cindex python exceptions
19883 @cindex exceptions, python
19885 When executing the @code{python} command, Python exceptions
19886 uncaught within the Python code are translated to calls to
19887 @value{GDBN} error-reporting mechanism. If the command that called
19888 @code{python} does not handle the error, @value{GDBN} will
19889 terminate it and print an error message containing the Python
19890 exception name, the associated value, and the Python call stack
19891 backtrace at the point where the exception was raised. Example:
19894 (@value{GDBP}) python print foo
19895 Traceback (most recent call last):
19896 File "<string>", line 1, in <module>
19897 NameError: name 'foo' is not defined
19900 @value{GDBN} errors that happen in @value{GDBN} commands invoked by Python
19901 code are converted to Python @code{RuntimeError} exceptions. User
19902 interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
19903 prompt) is translated to a Python @code{KeyboardInterrupt}
19904 exception. If you catch these exceptions in your Python code, your
19905 exception handler will see @code{RuntimeError} or
19906 @code{KeyboardInterrupt} as the exception type, the @value{GDBN} error
19907 message as its value, and the Python call stack backtrace at the
19908 Python statement closest to where the @value{GDBN} error occured as the
19912 @subsubsection Auto-loading
19913 @cindex auto-loading, Python
19915 When a new object file is read (for example, due to the @code{file}
19916 command, or because the inferior has loaded a shared library),
19917 @value{GDBN} will look for a file named @file{@var{objfile}-gdb.py},
19918 where @var{objfile} is the object file's real name, formed by ensuring
19919 that the file name is absolute, following all symlinks, and resolving
19920 @code{.} and @code{..} components. If this file exists and is
19921 readable, @value{GDBN} will evaluate it as a Python script.
19923 If this file does not exist, and if the parameter
19924 @code{debug-file-directory} is set (@pxref{Separate Debug Files}),
19925 then @value{GDBN} will use for its each separated directory component
19926 @code{component} the file named @file{@code{component}/@var{real-name}}, where
19927 @var{real-name} is the object file's real name, as described above.
19929 Finally, if this file does not exist, then @value{GDBN} will look for
19930 a file named @file{@var{data-directory}/python/auto-load/@var{real-name}}, where
19931 @var{data-directory} is @value{GDBN}'s data directory (available via
19932 @code{show data-directory}, @pxref{Data Files}), and @var{real-name}
19933 is the object file's real name, as described above.
19935 When reading an auto-loaded file, @value{GDBN} sets the ``current
19936 objfile''. This is available via the @code{gdb.current_objfile}
19937 function (@pxref{Objfiles In Python}). This can be useful for
19938 registering objfile-specific pretty-printers.
19940 The auto-loading feature is useful for supplying application-specific
19941 debugging commands and scripts. You can enable or disable this
19942 feature, and view its current state.
19945 @kindex maint set python auto-load
19946 @item maint set python auto-load [yes|no]
19947 Enable or disable the Python auto-loading feature.
19949 @kindex maint show python auto-load
19950 @item maint show python auto-load
19951 Show whether Python auto-loading is enabled or disabled.
19954 @value{GDBN} does not track which files it has already auto-loaded.
19955 So, your @samp{-gdb.py} file should take care to ensure that it may be
19956 evaluated multiple times without error.
19958 @node Values From Inferior
19959 @subsubsection Values From Inferior
19960 @cindex values from inferior, with Python
19961 @cindex python, working with values from inferior
19963 @cindex @code{gdb.Value}
19964 @value{GDBN} provides values it obtains from the inferior program in
19965 an object of type @code{gdb.Value}. @value{GDBN} uses this object
19966 for its internal bookkeeping of the inferior's values, and for
19967 fetching values when necessary.
19969 Inferior values that are simple scalars can be used directly in
19970 Python expressions that are valid for the value's data type. Here's
19971 an example for an integer or floating-point value @code{some_val}:
19978 As result of this, @code{bar} will also be a @code{gdb.Value} object
19979 whose values are of the same type as those of @code{some_val}.
19981 Inferior values that are structures or instances of some class can
19982 be accessed using the Python @dfn{dictionary syntax}. For example, if
19983 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
19984 can access its @code{foo} element with:
19987 bar = some_val['foo']
19990 Again, @code{bar} will also be a @code{gdb.Value} object.
19992 The following attributes are provided:
19995 @defivar Value address
19996 If this object is addressable, this read-only attribute holds a
19997 @code{gdb.Value} object representing the address. Otherwise,
19998 this attribute holds @code{None}.
20001 @cindex optimized out value in Python
20002 @defivar Value is_optimized_out
20003 This read-only boolean attribute is true if the compiler optimized out
20004 this value, thus it is not available for fetching from the inferior.
20007 @defivar Value type
20008 The type of this @code{gdb.Value}. The value of this attribute is a
20009 @code{gdb.Type} object.
20013 The following methods are provided:
20016 @defmethod Value cast type
20017 Return a new instance of @code{gdb.Value} that is the result of
20018 casting this instance to the type described by @var{type}, which must
20019 be a @code{gdb.Type} object. If the cast cannot be performed for some
20020 reason, this method throws an exception.
20023 @defmethod Value dereference
20024 For pointer data types, this method returns a new @code{gdb.Value} object
20025 whose contents is the object pointed to by the pointer. For example, if
20026 @code{foo} is a C pointer to an @code{int}, declared in your C program as
20033 then you can use the corresponding @code{gdb.Value} to access what
20034 @code{foo} points to like this:
20037 bar = foo.dereference ()
20040 The result @code{bar} will be a @code{gdb.Value} object holding the
20041 value pointed to by @code{foo}.
20044 @defmethod Value string @r{[}encoding@r{]} @r{[}errors@r{]} @r{[}length@r{]}
20045 If this @code{gdb.Value} represents a string, then this method
20046 converts the contents to a Python string. Otherwise, this method will
20047 throw an exception.
20049 Strings are recognized in a language-specific way; whether a given
20050 @code{gdb.Value} represents a string is determined by the current
20053 For C-like languages, a value is a string if it is a pointer to or an
20054 array of characters or ints. The string is assumed to be terminated
20055 by a zero of the appropriate width. However if the optional length
20056 argument is given, the string will be converted to that given length,
20057 ignoring any embedded zeros that the string may contain.
20059 If the optional @var{encoding} argument is given, it must be a string
20060 naming the encoding of the string in the @code{gdb.Value}, such as
20061 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
20062 the same encodings as the corresponding argument to Python's
20063 @code{string.decode} method, and the Python codec machinery will be used
20064 to convert the string. If @var{encoding} is not given, or if
20065 @var{encoding} is the empty string, then either the @code{target-charset}
20066 (@pxref{Character Sets}) will be used, or a language-specific encoding
20067 will be used, if the current language is able to supply one.
20069 The optional @var{errors} argument is the same as the corresponding
20070 argument to Python's @code{string.decode} method.
20072 If the optional @var{length} argument is given, the string will be
20073 fetched and converted to the given length.
20076 @defmethod Value lazy_string @r{[}encoding@r{]} @r{[}length@r{]}
20077 If this @code{gdb.Value} represents a string, then this method
20078 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
20079 In Python}). Otherwise, this method will throw an exception.
20081 If the optional @var{encoding} argument is given, it must be a string
20082 naming the encoding of the @code{gdb.LazyString}. Some examples are:
20083 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
20084 @var{encoding} argument is an encoding that @value{GDBN} does
20085 recognize, @value{GDBN} will raise an error.
20087 When a lazy string is printed, the @value{GDBN} encoding machinery is
20088 used to convert the string during printing. If the optional
20089 @var{encoding} argument is not provided, or is an empty string,
20090 @value{GDBN} will automatically select the encoding most suitable for
20091 the string type. For further information on encoding in @value{GDBN}
20092 please see @ref{Character Sets}.
20094 If the optional @var{length} argument is given, the string will be
20095 fetched and encoded to the length of characters specified. If
20096 the @var{length} argument is not provided, the string will be fetched
20097 and encoded until a null of appropriate width is found.
20101 @node Types In Python
20102 @subsubsection Types In Python
20103 @cindex types in Python
20104 @cindex Python, working with types
20107 @value{GDBN} represents types from the inferior using the class
20110 The following type-related functions are available in the @code{gdb}
20113 @findex gdb.lookup_type
20114 @defun lookup_type name [block]
20115 This function looks up a type by name. @var{name} is the name of the
20116 type to look up. It must be a string.
20118 If @var{block} is given, then @var{name} is looked up in that scope.
20119 Otherwise, it is searched for globally.
20121 Ordinarily, this function will return an instance of @code{gdb.Type}.
20122 If the named type cannot be found, it will throw an exception.
20125 An instance of @code{Type} has the following attributes:
20129 The type code for this type. The type code will be one of the
20130 @code{TYPE_CODE_} constants defined below.
20133 @defivar Type sizeof
20134 The size of this type, in target @code{char} units. Usually, a
20135 target's @code{char} type will be an 8-bit byte. However, on some
20136 unusual platforms, this type may have a different size.
20140 The tag name for this type. The tag name is the name after
20141 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
20142 languages have this concept. If this type has no tag name, then
20143 @code{None} is returned.
20147 The following methods are provided:
20150 @defmethod Type fields
20151 For structure and union types, this method returns the fields. Range
20152 types have two fields, the minimum and maximum values. Enum types
20153 have one field per enum constant. Function and method types have one
20154 field per parameter. The base types of C@t{++} classes are also
20155 represented as fields. If the type has no fields, or does not fit
20156 into one of these categories, an empty sequence will be returned.
20158 Each field is an object, with some pre-defined attributes:
20161 This attribute is not available for @code{static} fields (as in
20162 C@t{++} or Java). For non-@code{static} fields, the value is the bit
20163 position of the field.
20166 The name of the field, or @code{None} for anonymous fields.
20169 This is @code{True} if the field is artificial, usually meaning that
20170 it was provided by the compiler and not the user. This attribute is
20171 always provided, and is @code{False} if the field is not artificial.
20173 @item is_base_class
20174 This is @code{True} if the field represents a base class of a C@t{++}
20175 structure. This attribute is always provided, and is @code{False}
20176 if the field is not a base class of the type that is the argument of
20177 @code{fields}, or if that type was not a C@t{++} class.
20180 If the field is packed, or is a bitfield, then this will have a
20181 non-zero value, which is the size of the field in bits. Otherwise,
20182 this will be zero; in this case the field's size is given by its type.
20185 The type of the field. This is usually an instance of @code{Type},
20186 but it can be @code{None} in some situations.
20190 @defmethod Type const
20191 Return a new @code{gdb.Type} object which represents a
20192 @code{const}-qualified variant of this type.
20195 @defmethod Type volatile
20196 Return a new @code{gdb.Type} object which represents a
20197 @code{volatile}-qualified variant of this type.
20200 @defmethod Type unqualified
20201 Return a new @code{gdb.Type} object which represents an unqualified
20202 variant of this type. That is, the result is neither @code{const} nor
20206 @defmethod Type range
20207 Return a Python @code{Tuple} object that contains two elements: the
20208 low bound of the argument type and the high bound of that type. If
20209 the type does not have a range, @value{GDBN} will raise a
20210 @code{RuntimeError} exception.
20213 @defmethod Type reference
20214 Return a new @code{gdb.Type} object which represents a reference to this
20218 @defmethod Type pointer
20219 Return a new @code{gdb.Type} object which represents a pointer to this
20223 @defmethod Type strip_typedefs
20224 Return a new @code{gdb.Type} that represents the real type,
20225 after removing all layers of typedefs.
20228 @defmethod Type target
20229 Return a new @code{gdb.Type} object which represents the target type
20232 For a pointer type, the target type is the type of the pointed-to
20233 object. For an array type (meaning C-like arrays), the target type is
20234 the type of the elements of the array. For a function or method type,
20235 the target type is the type of the return value. For a complex type,
20236 the target type is the type of the elements. For a typedef, the
20237 target type is the aliased type.
20239 If the type does not have a target, this method will throw an
20243 @defmethod Type template_argument n [block]
20244 If this @code{gdb.Type} is an instantiation of a template, this will
20245 return a new @code{gdb.Type} which represents the type of the
20246 @var{n}th template argument.
20248 If this @code{gdb.Type} is not a template type, this will throw an
20249 exception. Ordinarily, only C@t{++} code will have template types.
20251 If @var{block} is given, then @var{name} is looked up in that scope.
20252 Otherwise, it is searched for globally.
20257 Each type has a code, which indicates what category this type falls
20258 into. The available type categories are represented by constants
20259 defined in the @code{gdb} module:
20262 @findex TYPE_CODE_PTR
20263 @findex gdb.TYPE_CODE_PTR
20264 @item TYPE_CODE_PTR
20265 The type is a pointer.
20267 @findex TYPE_CODE_ARRAY
20268 @findex gdb.TYPE_CODE_ARRAY
20269 @item TYPE_CODE_ARRAY
20270 The type is an array.
20272 @findex TYPE_CODE_STRUCT
20273 @findex gdb.TYPE_CODE_STRUCT
20274 @item TYPE_CODE_STRUCT
20275 The type is a structure.
20277 @findex TYPE_CODE_UNION
20278 @findex gdb.TYPE_CODE_UNION
20279 @item TYPE_CODE_UNION
20280 The type is a union.
20282 @findex TYPE_CODE_ENUM
20283 @findex gdb.TYPE_CODE_ENUM
20284 @item TYPE_CODE_ENUM
20285 The type is an enum.
20287 @findex TYPE_CODE_FLAGS
20288 @findex gdb.TYPE_CODE_FLAGS
20289 @item TYPE_CODE_FLAGS
20290 A bit flags type, used for things such as status registers.
20292 @findex TYPE_CODE_FUNC
20293 @findex gdb.TYPE_CODE_FUNC
20294 @item TYPE_CODE_FUNC
20295 The type is a function.
20297 @findex TYPE_CODE_INT
20298 @findex gdb.TYPE_CODE_INT
20299 @item TYPE_CODE_INT
20300 The type is an integer type.
20302 @findex TYPE_CODE_FLT
20303 @findex gdb.TYPE_CODE_FLT
20304 @item TYPE_CODE_FLT
20305 A floating point type.
20307 @findex TYPE_CODE_VOID
20308 @findex gdb.TYPE_CODE_VOID
20309 @item TYPE_CODE_VOID
20310 The special type @code{void}.
20312 @findex TYPE_CODE_SET
20313 @findex gdb.TYPE_CODE_SET
20314 @item TYPE_CODE_SET
20317 @findex TYPE_CODE_RANGE
20318 @findex gdb.TYPE_CODE_RANGE
20319 @item TYPE_CODE_RANGE
20320 A range type, that is, an integer type with bounds.
20322 @findex TYPE_CODE_STRING
20323 @findex gdb.TYPE_CODE_STRING
20324 @item TYPE_CODE_STRING
20325 A string type. Note that this is only used for certain languages with
20326 language-defined string types; C strings are not represented this way.
20328 @findex TYPE_CODE_BITSTRING
20329 @findex gdb.TYPE_CODE_BITSTRING
20330 @item TYPE_CODE_BITSTRING
20333 @findex TYPE_CODE_ERROR
20334 @findex gdb.TYPE_CODE_ERROR
20335 @item TYPE_CODE_ERROR
20336 An unknown or erroneous type.
20338 @findex TYPE_CODE_METHOD
20339 @findex gdb.TYPE_CODE_METHOD
20340 @item TYPE_CODE_METHOD
20341 A method type, as found in C@t{++} or Java.
20343 @findex TYPE_CODE_METHODPTR
20344 @findex gdb.TYPE_CODE_METHODPTR
20345 @item TYPE_CODE_METHODPTR
20346 A pointer-to-member-function.
20348 @findex TYPE_CODE_MEMBERPTR
20349 @findex gdb.TYPE_CODE_MEMBERPTR
20350 @item TYPE_CODE_MEMBERPTR
20351 A pointer-to-member.
20353 @findex TYPE_CODE_REF
20354 @findex gdb.TYPE_CODE_REF
20355 @item TYPE_CODE_REF
20358 @findex TYPE_CODE_CHAR
20359 @findex gdb.TYPE_CODE_CHAR
20360 @item TYPE_CODE_CHAR
20363 @findex TYPE_CODE_BOOL
20364 @findex gdb.TYPE_CODE_BOOL
20365 @item TYPE_CODE_BOOL
20368 @findex TYPE_CODE_COMPLEX
20369 @findex gdb.TYPE_CODE_COMPLEX
20370 @item TYPE_CODE_COMPLEX
20371 A complex float type.
20373 @findex TYPE_CODE_TYPEDEF
20374 @findex gdb.TYPE_CODE_TYPEDEF
20375 @item TYPE_CODE_TYPEDEF
20376 A typedef to some other type.
20378 @findex TYPE_CODE_NAMESPACE
20379 @findex gdb.TYPE_CODE_NAMESPACE
20380 @item TYPE_CODE_NAMESPACE
20381 A C@t{++} namespace.
20383 @findex TYPE_CODE_DECFLOAT
20384 @findex gdb.TYPE_CODE_DECFLOAT
20385 @item TYPE_CODE_DECFLOAT
20386 A decimal floating point type.
20388 @findex TYPE_CODE_INTERNAL_FUNCTION
20389 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
20390 @item TYPE_CODE_INTERNAL_FUNCTION
20391 A function internal to @value{GDBN}. This is the type used to represent
20392 convenience functions.
20395 @node Pretty Printing
20396 @subsubsection Pretty Printing
20398 @value{GDBN} provides a mechanism to allow pretty-printing of values
20399 using Python code. The pretty-printer API allows application-specific
20400 code to greatly simplify the display of complex objects. This
20401 mechanism works for both MI and the CLI.
20403 For example, here is how a C@t{++} @code{std::string} looks without a
20407 (@value{GDBP}) print s
20409 static npos = 4294967295,
20411 <std::allocator<char>> = @{
20412 <__gnu_cxx::new_allocator<char>> = @{<No data fields>@}, <No data fields>@},
20413 members of std::basic_string<char, std::char_traits<char>, std::allocator<char> >::_Alloc_hider:
20414 _M_p = 0x804a014 "abcd"
20419 After a pretty-printer for @code{std::string} has been installed, only
20420 the contents are printed:
20423 (@value{GDBP}) print s
20427 A pretty-printer is just an object that holds a value and implements a
20428 specific interface, defined here.
20430 @defop Operation {pretty printer} children (self)
20431 @value{GDBN} will call this method on a pretty-printer to compute the
20432 children of the pretty-printer's value.
20434 This method must return an object conforming to the Python iterator
20435 protocol. Each item returned by the iterator must be a tuple holding
20436 two elements. The first element is the ``name'' of the child; the
20437 second element is the child's value. The value can be any Python
20438 object which is convertible to a @value{GDBN} value.
20440 This method is optional. If it does not exist, @value{GDBN} will act
20441 as though the value has no children.
20444 @defop Operation {pretty printer} display_hint (self)
20445 The CLI may call this method and use its result to change the
20446 formatting of a value. The result will also be supplied to an MI
20447 consumer as a @samp{displayhint} attribute of the variable being
20450 This method is optional. If it does exist, this method must return a
20453 Some display hints are predefined by @value{GDBN}:
20457 Indicate that the object being printed is ``array-like''. The CLI
20458 uses this to respect parameters such as @code{set print elements} and
20459 @code{set print array}.
20462 Indicate that the object being printed is ``map-like'', and that the
20463 children of this value can be assumed to alternate between keys and
20467 Indicate that the object being printed is ``string-like''. If the
20468 printer's @code{to_string} method returns a Python string of some
20469 kind, then @value{GDBN} will call its internal language-specific
20470 string-printing function to format the string. For the CLI this means
20471 adding quotation marks, possibly escaping some characters, respecting
20472 @code{set print elements}, and the like.
20476 @defop Operation {pretty printer} to_string (self)
20477 @value{GDBN} will call this method to display the string
20478 representation of the value passed to the object's constructor.
20480 When printing from the CLI, if the @code{to_string} method exists,
20481 then @value{GDBN} will prepend its result to the values returned by
20482 @code{children}. Exactly how this formatting is done is dependent on
20483 the display hint, and may change as more hints are added. Also,
20484 depending on the print settings (@pxref{Print Settings}), the CLI may
20485 print just the result of @code{to_string} in a stack trace, omitting
20486 the result of @code{children}.
20488 If this method returns a string, it is printed verbatim.
20490 Otherwise, if this method returns an instance of @code{gdb.Value},
20491 then @value{GDBN} prints this value. This may result in a call to
20492 another pretty-printer.
20494 If instead the method returns a Python value which is convertible to a
20495 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
20496 the resulting value. Again, this may result in a call to another
20497 pretty-printer. Python scalars (integers, floats, and booleans) and
20498 strings are convertible to @code{gdb.Value}; other types are not.
20500 Finally, if this method returns @code{None} then no further operations
20501 are peformed in this method and nothing is printed.
20503 If the result is not one of these types, an exception is raised.
20506 @node Selecting Pretty-Printers
20507 @subsubsection Selecting Pretty-Printers
20509 The Python list @code{gdb.pretty_printers} contains an array of
20510 functions that have been registered via addition as a pretty-printer.
20511 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
20512 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
20515 A function on one of these lists is passed a single @code{gdb.Value}
20516 argument and should return a pretty-printer object conforming to the
20517 interface definition above (@pxref{Pretty Printing}). If a function
20518 cannot create a pretty-printer for the value, it should return
20521 @value{GDBN} first checks the @code{pretty_printers} attribute of each
20522 @code{gdb.Objfile} in the current program space and iteratively calls
20523 each function in the list for that @code{gdb.Objfile} until it receives
20524 a pretty-printer object.
20525 If no pretty-printer is found in the objfile lists, @value{GDBN} then
20526 searches the pretty-printer list of the current program space,
20527 calling each function until an object is returned.
20528 After these lists have been exhausted, it tries the global
20529 @code{gdb.pretty-printers} list, again calling each function until an
20530 object is returned.
20532 The order in which the objfiles are searched is not specified. For a
20533 given list, functions are always invoked from the head of the list,
20534 and iterated over sequentially until the end of the list, or a printer
20535 object is returned.
20537 Here is an example showing how a @code{std::string} printer might be
20541 class StdStringPrinter:
20542 "Print a std::string"
20544 def __init__ (self, val):
20547 def to_string (self):
20548 return self.val['_M_dataplus']['_M_p']
20550 def display_hint (self):
20554 And here is an example showing how a lookup function for the printer
20555 example above might be written.
20558 def str_lookup_function (val):
20560 lookup_tag = val.type.tag
20561 regex = re.compile ("^std::basic_string<char,.*>$")
20562 if lookup_tag == None:
20564 if regex.match (lookup_tag):
20565 return StdStringPrinter (val)
20570 The example lookup function extracts the value's type, and attempts to
20571 match it to a type that it can pretty-print. If it is a type the
20572 printer can pretty-print, it will return a printer object. If not, it
20573 returns @code{None}.
20575 We recommend that you put your core pretty-printers into a Python
20576 package. If your pretty-printers are for use with a library, we
20577 further recommend embedding a version number into the package name.
20578 This practice will enable @value{GDBN} to load multiple versions of
20579 your pretty-printers at the same time, because they will have
20582 You should write auto-loaded code (@pxref{Auto-loading}) such that it
20583 can be evaluated multiple times without changing its meaning. An
20584 ideal auto-load file will consist solely of @code{import}s of your
20585 printer modules, followed by a call to a register pretty-printers with
20586 the current objfile.
20588 Taken as a whole, this approach will scale nicely to multiple
20589 inferiors, each potentially using a different library version.
20590 Embedding a version number in the Python package name will ensure that
20591 @value{GDBN} is able to load both sets of printers simultaneously.
20592 Then, because the search for pretty-printers is done by objfile, and
20593 because your auto-loaded code took care to register your library's
20594 printers with a specific objfile, @value{GDBN} will find the correct
20595 printers for the specific version of the library used by each
20598 To continue the @code{std::string} example (@pxref{Pretty Printing}),
20599 this code might appear in @code{gdb.libstdcxx.v6}:
20602 def register_printers (objfile):
20603 objfile.pretty_printers.add (str_lookup_function)
20607 And then the corresponding contents of the auto-load file would be:
20610 import gdb.libstdcxx.v6
20611 gdb.libstdcxx.v6.register_printers (gdb.current_objfile ())
20614 @node Commands In Python
20615 @subsubsection Commands In Python
20617 @cindex commands in python
20618 @cindex python commands
20619 You can implement new @value{GDBN} CLI commands in Python. A CLI
20620 command is implemented using an instance of the @code{gdb.Command}
20621 class, most commonly using a subclass.
20623 @defmethod Command __init__ name @var{command_class} @r{[}@var{completer_class}@r{]} @r{[}@var{prefix}@r{]}
20624 The object initializer for @code{Command} registers the new command
20625 with @value{GDBN}. This initializer is normally invoked from the
20626 subclass' own @code{__init__} method.
20628 @var{name} is the name of the command. If @var{name} consists of
20629 multiple words, then the initial words are looked for as prefix
20630 commands. In this case, if one of the prefix commands does not exist,
20631 an exception is raised.
20633 There is no support for multi-line commands.
20635 @var{command_class} should be one of the @samp{COMMAND_} constants
20636 defined below. This argument tells @value{GDBN} how to categorize the
20637 new command in the help system.
20639 @var{completer_class} is an optional argument. If given, it should be
20640 one of the @samp{COMPLETE_} constants defined below. This argument
20641 tells @value{GDBN} how to perform completion for this command. If not
20642 given, @value{GDBN} will attempt to complete using the object's
20643 @code{complete} method (see below); if no such method is found, an
20644 error will occur when completion is attempted.
20646 @var{prefix} is an optional argument. If @code{True}, then the new
20647 command is a prefix command; sub-commands of this command may be
20650 The help text for the new command is taken from the Python
20651 documentation string for the command's class, if there is one. If no
20652 documentation string is provided, the default value ``This command is
20653 not documented.'' is used.
20656 @cindex don't repeat Python command
20657 @defmethod Command dont_repeat
20658 By default, a @value{GDBN} command is repeated when the user enters a
20659 blank line at the command prompt. A command can suppress this
20660 behavior by invoking the @code{dont_repeat} method. This is similar
20661 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
20664 @defmethod Command invoke argument from_tty
20665 This method is called by @value{GDBN} when this command is invoked.
20667 @var{argument} is a string. It is the argument to the command, after
20668 leading and trailing whitespace has been stripped.
20670 @var{from_tty} is a boolean argument. When true, this means that the
20671 command was entered by the user at the terminal; when false it means
20672 that the command came from elsewhere.
20674 If this method throws an exception, it is turned into a @value{GDBN}
20675 @code{error} call. Otherwise, the return value is ignored.
20678 @cindex completion of Python commands
20679 @defmethod Command complete text word
20680 This method is called by @value{GDBN} when the user attempts
20681 completion on this command. All forms of completion are handled by
20682 this method, that is, the @key{TAB} and @key{M-?} key bindings
20683 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
20686 The arguments @var{text} and @var{word} are both strings. @var{text}
20687 holds the complete command line up to the cursor's location.
20688 @var{word} holds the last word of the command line; this is computed
20689 using a word-breaking heuristic.
20691 The @code{complete} method can return several values:
20694 If the return value is a sequence, the contents of the sequence are
20695 used as the completions. It is up to @code{complete} to ensure that the
20696 contents actually do complete the word. A zero-length sequence is
20697 allowed, it means that there were no completions available. Only
20698 string elements of the sequence are used; other elements in the
20699 sequence are ignored.
20702 If the return value is one of the @samp{COMPLETE_} constants defined
20703 below, then the corresponding @value{GDBN}-internal completion
20704 function is invoked, and its result is used.
20707 All other results are treated as though there were no available
20712 When a new command is registered, it must be declared as a member of
20713 some general class of commands. This is used to classify top-level
20714 commands in the on-line help system; note that prefix commands are not
20715 listed under their own category but rather that of their top-level
20716 command. The available classifications are represented by constants
20717 defined in the @code{gdb} module:
20720 @findex COMMAND_NONE
20721 @findex gdb.COMMAND_NONE
20723 The command does not belong to any particular class. A command in
20724 this category will not be displayed in any of the help categories.
20726 @findex COMMAND_RUNNING
20727 @findex gdb.COMMAND_RUNNING
20728 @item COMMAND_RUNNING
20729 The command is related to running the inferior. For example,
20730 @code{start}, @code{step}, and @code{continue} are in this category.
20731 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
20732 commands in this category.
20734 @findex COMMAND_DATA
20735 @findex gdb.COMMAND_DATA
20737 The command is related to data or variables. For example,
20738 @code{call}, @code{find}, and @code{print} are in this category. Type
20739 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
20742 @findex COMMAND_STACK
20743 @findex gdb.COMMAND_STACK
20744 @item COMMAND_STACK
20745 The command has to do with manipulation of the stack. For example,
20746 @code{backtrace}, @code{frame}, and @code{return} are in this
20747 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
20748 list of commands in this category.
20750 @findex COMMAND_FILES
20751 @findex gdb.COMMAND_FILES
20752 @item COMMAND_FILES
20753 This class is used for file-related commands. For example,
20754 @code{file}, @code{list} and @code{section} are in this category.
20755 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
20756 commands in this category.
20758 @findex COMMAND_SUPPORT
20759 @findex gdb.COMMAND_SUPPORT
20760 @item COMMAND_SUPPORT
20761 This should be used for ``support facilities'', generally meaning
20762 things that are useful to the user when interacting with @value{GDBN},
20763 but not related to the state of the inferior. For example,
20764 @code{help}, @code{make}, and @code{shell} are in this category. Type
20765 @kbd{help support} at the @value{GDBN} prompt to see a list of
20766 commands in this category.
20768 @findex COMMAND_STATUS
20769 @findex gdb.COMMAND_STATUS
20770 @item COMMAND_STATUS
20771 The command is an @samp{info}-related command, that is, related to the
20772 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
20773 and @code{show} are in this category. Type @kbd{help status} at the
20774 @value{GDBN} prompt to see a list of commands in this category.
20776 @findex COMMAND_BREAKPOINTS
20777 @findex gdb.COMMAND_BREAKPOINTS
20778 @item COMMAND_BREAKPOINTS
20779 The command has to do with breakpoints. For example, @code{break},
20780 @code{clear}, and @code{delete} are in this category. Type @kbd{help
20781 breakpoints} at the @value{GDBN} prompt to see a list of commands in
20784 @findex COMMAND_TRACEPOINTS
20785 @findex gdb.COMMAND_TRACEPOINTS
20786 @item COMMAND_TRACEPOINTS
20787 The command has to do with tracepoints. For example, @code{trace},
20788 @code{actions}, and @code{tfind} are in this category. Type
20789 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
20790 commands in this category.
20792 @findex COMMAND_OBSCURE
20793 @findex gdb.COMMAND_OBSCURE
20794 @item COMMAND_OBSCURE
20795 The command is only used in unusual circumstances, or is not of
20796 general interest to users. For example, @code{checkpoint},
20797 @code{fork}, and @code{stop} are in this category. Type @kbd{help
20798 obscure} at the @value{GDBN} prompt to see a list of commands in this
20801 @findex COMMAND_MAINTENANCE
20802 @findex gdb.COMMAND_MAINTENANCE
20803 @item COMMAND_MAINTENANCE
20804 The command is only useful to @value{GDBN} maintainers. The
20805 @code{maintenance} and @code{flushregs} commands are in this category.
20806 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
20807 commands in this category.
20810 A new command can use a predefined completion function, either by
20811 specifying it via an argument at initialization, or by returning it
20812 from the @code{complete} method. These predefined completion
20813 constants are all defined in the @code{gdb} module:
20816 @findex COMPLETE_NONE
20817 @findex gdb.COMPLETE_NONE
20818 @item COMPLETE_NONE
20819 This constant means that no completion should be done.
20821 @findex COMPLETE_FILENAME
20822 @findex gdb.COMPLETE_FILENAME
20823 @item COMPLETE_FILENAME
20824 This constant means that filename completion should be performed.
20826 @findex COMPLETE_LOCATION
20827 @findex gdb.COMPLETE_LOCATION
20828 @item COMPLETE_LOCATION
20829 This constant means that location completion should be done.
20830 @xref{Specify Location}.
20832 @findex COMPLETE_COMMAND
20833 @findex gdb.COMPLETE_COMMAND
20834 @item COMPLETE_COMMAND
20835 This constant means that completion should examine @value{GDBN}
20838 @findex COMPLETE_SYMBOL
20839 @findex gdb.COMPLETE_SYMBOL
20840 @item COMPLETE_SYMBOL
20841 This constant means that completion should be done using symbol names
20845 The following code snippet shows how a trivial CLI command can be
20846 implemented in Python:
20849 class HelloWorld (gdb.Command):
20850 """Greet the whole world."""
20852 def __init__ (self):
20853 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
20855 def invoke (self, arg, from_tty):
20856 print "Hello, World!"
20861 The last line instantiates the class, and is necessary to trigger the
20862 registration of the command with @value{GDBN}. Depending on how the
20863 Python code is read into @value{GDBN}, you may need to import the
20864 @code{gdb} module explicitly.
20866 @node Functions In Python
20867 @subsubsection Writing new convenience functions
20869 @cindex writing convenience functions
20870 @cindex convenience functions in python
20871 @cindex python convenience functions
20872 @tindex gdb.Function
20874 You can implement new convenience functions (@pxref{Convenience Vars})
20875 in Python. A convenience function is an instance of a subclass of the
20876 class @code{gdb.Function}.
20878 @defmethod Function __init__ name
20879 The initializer for @code{Function} registers the new function with
20880 @value{GDBN}. The argument @var{name} is the name of the function,
20881 a string. The function will be visible to the user as a convenience
20882 variable of type @code{internal function}, whose name is the same as
20883 the given @var{name}.
20885 The documentation for the new function is taken from the documentation
20886 string for the new class.
20889 @defmethod Function invoke @var{*args}
20890 When a convenience function is evaluated, its arguments are converted
20891 to instances of @code{gdb.Value}, and then the function's
20892 @code{invoke} method is called. Note that @value{GDBN} does not
20893 predetermine the arity of convenience functions. Instead, all
20894 available arguments are passed to @code{invoke}, following the
20895 standard Python calling convention. In particular, a convenience
20896 function can have default values for parameters without ill effect.
20898 The return value of this method is used as its value in the enclosing
20899 expression. If an ordinary Python value is returned, it is converted
20900 to a @code{gdb.Value} following the usual rules.
20903 The following code snippet shows how a trivial convenience function can
20904 be implemented in Python:
20907 class Greet (gdb.Function):
20908 """Return string to greet someone.
20909 Takes a name as argument."""
20911 def __init__ (self):
20912 super (Greet, self).__init__ ("greet")
20914 def invoke (self, name):
20915 return "Hello, %s!" % name.string ()
20920 The last line instantiates the class, and is necessary to trigger the
20921 registration of the function with @value{GDBN}. Depending on how the
20922 Python code is read into @value{GDBN}, you may need to import the
20923 @code{gdb} module explicitly.
20925 @node Progspaces In Python
20926 @subsubsection Program Spaces In Python
20928 @cindex progspaces in python
20929 @tindex gdb.Progspace
20931 A program space, or @dfn{progspace}, represents a symbolic view
20932 of an address space.
20933 It consists of all of the objfiles of the program.
20934 @xref{Objfiles In Python}.
20935 @xref{Inferiors and Programs, program spaces}, for more details
20936 about program spaces.
20938 The following progspace-related functions are available in the
20941 @findex gdb.current_progspace
20942 @defun current_progspace
20943 This function returns the program space of the currently selected inferior.
20944 @xref{Inferiors and Programs}.
20947 @findex gdb.progspaces
20949 Return a sequence of all the progspaces currently known to @value{GDBN}.
20952 Each progspace is represented by an instance of the @code{gdb.Progspace}
20955 @defivar Progspace filename
20956 The file name of the progspace as a string.
20959 @defivar Progspace pretty_printers
20960 The @code{pretty_printers} attribute is a list of functions. It is
20961 used to look up pretty-printers. A @code{Value} is passed to each
20962 function in order; if the function returns @code{None}, then the
20963 search continues. Otherwise, the return value should be an object
20964 which is used to format the value. @xref{Pretty Printing}, for more
20968 @node Objfiles In Python
20969 @subsubsection Objfiles In Python
20971 @cindex objfiles in python
20972 @tindex gdb.Objfile
20974 @value{GDBN} loads symbols for an inferior from various
20975 symbol-containing files (@pxref{Files}). These include the primary
20976 executable file, any shared libraries used by the inferior, and any
20977 separate debug info files (@pxref{Separate Debug Files}).
20978 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
20980 The following objfile-related functions are available in the
20983 @findex gdb.current_objfile
20984 @defun current_objfile
20985 When auto-loading a Python script (@pxref{Auto-loading}), @value{GDBN}
20986 sets the ``current objfile'' to the corresponding objfile. This
20987 function returns the current objfile. If there is no current objfile,
20988 this function returns @code{None}.
20991 @findex gdb.objfiles
20993 Return a sequence of all the objfiles current known to @value{GDBN}.
20994 @xref{Objfiles In Python}.
20997 Each objfile is represented by an instance of the @code{gdb.Objfile}
21000 @defivar Objfile filename
21001 The file name of the objfile as a string.
21004 @defivar Objfile pretty_printers
21005 The @code{pretty_printers} attribute is a list of functions. It is
21006 used to look up pretty-printers. A @code{Value} is passed to each
21007 function in order; if the function returns @code{None}, then the
21008 search continues. Otherwise, the return value should be an object
21009 which is used to format the value. @xref{Pretty Printing}, for more
21013 @node Frames In Python
21014 @subsubsection Accessing inferior stack frames from Python.
21016 @cindex frames in python
21017 When the debugged program stops, @value{GDBN} is able to analyze its call
21018 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
21019 represents a frame in the stack. A @code{gdb.Frame} object is only valid
21020 while its corresponding frame exists in the inferior's stack. If you try
21021 to use an invalid frame object, @value{GDBN} will throw a @code{RuntimeError}
21024 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
21028 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
21032 The following frame-related functions are available in the @code{gdb} module:
21034 @findex gdb.selected_frame
21035 @defun selected_frame
21036 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
21039 @defun frame_stop_reason_string reason
21040 Return a string explaining the reason why @value{GDBN} stopped unwinding
21041 frames, as expressed by the given @var{reason} code (an integer, see the
21042 @code{unwind_stop_reason} method further down in this section).
21045 A @code{gdb.Frame} object has the following methods:
21048 @defmethod Frame is_valid
21049 Returns true if the @code{gdb.Frame} object is valid, false if not.
21050 A frame object can become invalid if the frame it refers to doesn't
21051 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
21052 an exception if it is invalid at the time the method is called.
21055 @defmethod Frame name
21056 Returns the function name of the frame, or @code{None} if it can't be
21060 @defmethod Frame type
21061 Returns the type of the frame. The value can be one of
21062 @code{gdb.NORMAL_FRAME}, @code{gdb.DUMMY_FRAME}, @code{gdb.SIGTRAMP_FRAME}
21063 or @code{gdb.SENTINEL_FRAME}.
21066 @defmethod Frame unwind_stop_reason
21067 Return an integer representing the reason why it's not possible to find
21068 more frames toward the outermost frame. Use
21069 @code{gdb.frame_stop_reason_string} to convert the value returned by this
21070 function to a string.
21073 @defmethod Frame pc
21074 Returns the frame's resume address.
21077 @defmethod Frame block
21078 Return the frame's code block. @xref{Blocks In Python}.
21081 @defmethod Frame function
21082 Return the symbol for the function corresponding to this frame.
21083 @xref{Symbols In Python}.
21086 @defmethod Frame older
21087 Return the frame that called this frame.
21090 @defmethod Frame newer
21091 Return the frame called by this frame.
21094 @defmethod Frame find_sal
21095 Return the frame's symtab and line object.
21096 @xref{Symbol Tables In Python}.
21099 @defmethod Frame read_var variable @r{[}block@r{]}
21100 Return the value of @var{variable} in this frame. If the optional
21101 argument @var{block} is provided, search for the variable from that
21102 block; otherwise start at the frame's current block (which is
21103 determined by the frame's current program counter). @var{variable}
21104 must be a string or a @code{gdb.Symbol} object. @var{block} must be a
21105 @code{gdb.Block} object.
21108 @defmethod Frame select
21109 Set this frame to be the selected frame. @xref{Stack, ,Examining the
21114 @node Blocks In Python
21115 @subsubsection Accessing frame blocks from Python.
21117 @cindex blocks in python
21120 Within each frame, @value{GDBN} maintains information on each block
21121 stored in that frame. These blocks are organized hierarchically, and
21122 are represented individually in Python as a @code{gdb.Block}.
21123 Please see @ref{Frames In Python}, for a more in-depth discussion on
21124 frames. Furthermore, see @ref{Stack, ,Examining the Stack}, for more
21125 detailed technical information on @value{GDBN}'s book-keeping of the
21128 The following block-related functions are available in the @code{gdb}
21131 @findex gdb.block_for_pc
21132 @defun block_for_pc pc
21133 Return the @code{gdb.Block} containing the given @var{pc} value. If the
21134 block cannot be found for the @var{pc} value specified, the function
21135 will return @code{None}.
21138 A @code{gdb.Block} object has the following attributes:
21141 @defivar Block start
21142 The start address of the block. This attribute is not writable.
21146 The end address of the block. This attribute is not writable.
21149 @defivar Block function
21150 The name of the block represented as a @code{gdb.Symbol}. If the
21151 block is not named, then this attribute holds @code{None}. This
21152 attribute is not writable.
21155 @defivar Block superblock
21156 The block containing this block. If this parent block does not exist,
21157 this attribute holds @code{None}. This attribute is not writable.
21161 @node Symbols In Python
21162 @subsubsection Python representation of Symbols.
21164 @cindex symbols in python
21167 @value{GDBN} represents every variable, function and type as an
21168 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
21169 Similarly, Python represents these symbols in @value{GDBN} with the
21170 @code{gdb.Symbol} object.
21172 The following symbol-related functions are available in the @code{gdb}
21175 @findex gdb.lookup_symbol
21176 @defun lookup_symbol name [block] [domain]
21177 This function searches for a symbol by name. The search scope can be
21178 restricted to the parameters defined in the optional domain and block
21181 @var{name} is the name of the symbol. It must be a string. The
21182 optional @var{block} argument restricts the search to symbols visible
21183 in that @var{block}. The @var{block} argument must be a
21184 @code{gdb.Block} object. The optional @var{domain} argument restricts
21185 the search to the domain type. The @var{domain} argument must be a
21186 domain constant defined in the @code{gdb} module and described later
21190 A @code{gdb.Symbol} object has the following attributes:
21193 @defivar Symbol symtab
21194 The symbol table in which the symbol appears. This attribute is
21195 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
21196 Python}. This attribute is not writable.
21199 @defivar Symbol name
21200 The name of the symbol as a string. This attribute is not writable.
21203 @defivar Symbol linkage_name
21204 The name of the symbol, as used by the linker (i.e., may be mangled).
21205 This attribute is not writable.
21208 @defivar Symbol print_name
21209 The name of the symbol in a form suitable for output. This is either
21210 @code{name} or @code{linkage_name}, depending on whether the user
21211 asked @value{GDBN} to display demangled or mangled names.
21214 @defivar Symbol addr_class
21215 The address class of the symbol. This classifies how to find the value
21216 of a symbol. Each address class is a constant defined in the
21217 @code{gdb} module and described later in this chapter.
21220 @defivar Symbol is_argument
21221 @code{True} if the symbol is an argument of a function.
21224 @defivar Symbol is_constant
21225 @code{True} if the symbol is a constant.
21228 @defivar Symbol is_function
21229 @code{True} if the symbol is a function or a method.
21232 @defivar Symbol is_variable
21233 @code{True} if the symbol is a variable.
21237 The available domain categories in @code{gdb.Symbol} are represented
21238 as constants in the @code{gdb} module:
21241 @findex SYMBOL_UNDEF_DOMAIN
21242 @findex gdb.SYMBOL_UNDEF_DOMAIN
21243 @item SYMBOL_UNDEF_DOMAIN
21244 This is used when a domain has not been discovered or none of the
21245 following domains apply. This usually indicates an error either
21246 in the symbol information or in @value{GDBN}'s handling of symbols.
21247 @findex SYMBOL_VAR_DOMAIN
21248 @findex gdb.SYMBOL_VAR_DOMAIN
21249 @item SYMBOL_VAR_DOMAIN
21250 This domain contains variables, function names, typedef names and enum
21252 @findex SYMBOL_STRUCT_DOMAIN
21253 @findex gdb.SYMBOL_STRUCT_DOMAIN
21254 @item SYMBOL_STRUCT_DOMAIN
21255 This domain holds struct, union and enum type names.
21256 @findex SYMBOL_LABEL_DOMAIN
21257 @findex gdb.SYMBOL_LABEL_DOMAIN
21258 @item SYMBOL_LABEL_DOMAIN
21259 This domain contains names of labels (for gotos).
21260 @findex SYMBOL_VARIABLES_DOMAIN
21261 @findex gdb.SYMBOL_VARIABLES_DOMAIN
21262 @item SYMBOL_VARIABLES_DOMAIN
21263 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
21264 contains everything minus functions and types.
21265 @findex SYMBOL_FUNCTIONS_DOMAIN
21266 @findex gdb.SYMBOL_FUNCTIONS_DOMAIN
21267 @item SYMBOL_FUNCTION_DOMAIN
21268 This domain contains all functions.
21269 @findex SYMBOL_TYPES_DOMAIN
21270 @findex gdb.SYMBOL_TYPES_DOMAIN
21271 @item SYMBOL_TYPES_DOMAIN
21272 This domain contains all types.
21275 The available address class categories in @code{gdb.Symbol} are represented
21276 as constants in the @code{gdb} module:
21279 @findex SYMBOL_LOC_UNDEF
21280 @findex gdb.SYMBOL_LOC_UNDEF
21281 @item SYMBOL_LOC_UNDEF
21282 If this is returned by address class, it indicates an error either in
21283 the symbol information or in @value{GDBN}'s handling of symbols.
21284 @findex SYMBOL_LOC_CONST
21285 @findex gdb.SYMBOL_LOC_CONST
21286 @item SYMBOL_LOC_CONST
21287 Value is constant int.
21288 @findex SYMBOL_LOC_STATIC
21289 @findex gdb.SYMBOL_LOC_STATIC
21290 @item SYMBOL_LOC_STATIC
21291 Value is at a fixed address.
21292 @findex SYMBOL_LOC_REGISTER
21293 @findex gdb.SYMBOL_LOC_REGISTER
21294 @item SYMBOL_LOC_REGISTER
21295 Value is in a register.
21296 @findex SYMBOL_LOC_ARG
21297 @findex gdb.SYMBOL_LOC_ARG
21298 @item SYMBOL_LOC_ARG
21299 Value is an argument. This value is at the offset stored within the
21300 symbol inside the frame's argument list.
21301 @findex SYMBOL_LOC_REF_ARG
21302 @findex gdb.SYMBOL_LOC_REF_ARG
21303 @item SYMBOL_LOC_REF_ARG
21304 Value address is stored in the frame's argument list. Just like
21305 @code{LOC_ARG} except that the value's address is stored at the
21306 offset, not the value itself.
21307 @findex SYMBOL_LOC_REGPARM_ADDR
21308 @findex gdb.SYMBOL_LOC_REGPARM_ADDR
21309 @item SYMBOL_LOC_REGPARM_ADDR
21310 Value is a specified register. Just like @code{LOC_REGISTER} except
21311 the register holds the address of the argument instead of the argument
21313 @findex SYMBOL_LOC_LOCAL
21314 @findex gdb.SYMBOL_LOC_LOCAL
21315 @item SYMBOL_LOC_LOCAL
21316 Value is a local variable.
21317 @findex SYMBOL_LOC_TYPEDEF
21318 @findex gdb.SYMBOL_LOC_TYPEDEF
21319 @item SYMBOL_LOC_TYPEDEF
21320 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
21322 @findex SYMBOL_LOC_BLOCK
21323 @findex gdb.SYMBOL_LOC_BLOCK
21324 @item SYMBOL_LOC_BLOCK
21326 @findex SYMBOL_LOC_CONST_BYTES
21327 @findex gdb.SYMBOL_LOC_CONST_BYTES
21328 @item SYMBOL_LOC_CONST_BYTES
21329 Value is a byte-sequence.
21330 @findex SYMBOL_LOC_UNRESOLVED
21331 @findex gdb.SYMBOL_LOC_UNRESOLVED
21332 @item SYMBOL_LOC_UNRESOLVED
21333 Value is at a fixed address, but the address of the variable has to be
21334 determined from the minimal symbol table whenever the variable is
21336 @findex SYMBOL_LOC_OPTIMIZED_OUT
21337 @findex gdb.SYMBOL_LOC_OPTIMIZED_OUT
21338 @item SYMBOL_LOC_OPTIMIZED_OUT
21339 The value does not actually exist in the program.
21340 @findex SYMBOL_LOC_COMPUTED
21341 @findex gdb.SYMBOL_LOC_COMPUTED
21342 @item SYMBOL_LOC_COMPUTED
21343 The value's address is a computed location.
21346 @node Symbol Tables In Python
21347 @subsubsection Symbol table representation in Python.
21349 @cindex symbol tables in python
21351 @tindex gdb.Symtab_and_line
21353 Access to symbol table data maintained by @value{GDBN} on the inferior
21354 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
21355 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
21356 from the @code{find_sal} method in @code{gdb.Frame} object.
21357 @xref{Frames In Python}.
21359 For more information on @value{GDBN}'s symbol table management, see
21360 @ref{Symbols, ,Examining the Symbol Table}, for more information.
21362 A @code{gdb.Symtab_and_line} object has the following attributes:
21365 @defivar Symtab_and_line symtab
21366 The symbol table object (@code{gdb.Symtab}) for this frame.
21367 This attribute is not writable.
21370 @defivar Symtab_and_line pc
21371 Indicates the current program counter address. This attribute is not
21375 @defivar Symtab_and_line line
21376 Indicates the current line number for this object. This
21377 attribute is not writable.
21381 A @code{gdb.Symtab} object has the following attributes:
21384 @defivar Symtab filename
21385 The symbol table's source filename. This attribute is not writable.
21388 @defivar Symtab objfile
21389 The symbol table's backing object file. @xref{Objfiles In Python}.
21390 This attribute is not writable.
21394 The following methods are provided:
21397 @defmethod Symtab fullname
21398 Return the symbol table's source absolute file name.
21402 @node Breakpoints In Python
21403 @subsubsection Manipulating breakpoints using Python
21405 @cindex breakpoints in python
21406 @tindex gdb.Breakpoint
21408 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
21411 @defmethod Breakpoint __init__ spec @r{[}type@r{]} @r{[}wp_class@r{]}
21412 Create a new breakpoint. @var{spec} is a string naming the
21413 location of the breakpoint, or an expression that defines a
21414 watchpoint. The contents can be any location recognized by the
21415 @code{break} command, or in the case of a watchpoint, by the @code{watch}
21416 command. The optional @var{type} denotes the breakpoint to create
21417 from the types defined later in this chapter. This argument can be
21418 either: @code{BP_BREAKPOINT} or @code{BP_WATCHPOINT}. @var{type}
21419 defaults to @code{BP_BREAKPOINT}. The optional @var{wp_class}
21420 argument defines the class of watchpoint to create, if @var{type} is
21421 defined as @code{BP_WATCHPOINT}. If a watchpoint class is not
21422 provided, it is assumed to be a @var{WP_WRITE} class.
21425 The available watchpoint types represented by constants are defined in the
21430 @findex gdb.WP_READ
21432 Read only watchpoint.
21435 @findex gdb.WP_WRITE
21437 Write only watchpoint.
21440 @findex gdb.WP_ACCESS
21442 Read/Write watchpoint.
21445 @defmethod Breakpoint is_valid
21446 Return @code{True} if this @code{Breakpoint} object is valid,
21447 @code{False} otherwise. A @code{Breakpoint} object can become invalid
21448 if the user deletes the breakpoint. In this case, the object still
21449 exists, but the underlying breakpoint does not. In the cases of
21450 watchpoint scope, the watchpoint remains valid even if execution of the
21451 inferior leaves the scope of that watchpoint.
21454 @defivar Breakpoint enabled
21455 This attribute is @code{True} if the breakpoint is enabled, and
21456 @code{False} otherwise. This attribute is writable.
21459 @defivar Breakpoint silent
21460 This attribute is @code{True} if the breakpoint is silent, and
21461 @code{False} otherwise. This attribute is writable.
21463 Note that a breakpoint can also be silent if it has commands and the
21464 first command is @code{silent}. This is not reported by the
21465 @code{silent} attribute.
21468 @defivar Breakpoint thread
21469 If the breakpoint is thread-specific, this attribute holds the thread
21470 id. If the breakpoint is not thread-specific, this attribute is
21471 @code{None}. This attribute is writable.
21474 @defivar Breakpoint task
21475 If the breakpoint is Ada task-specific, this attribute holds the Ada task
21476 id. If the breakpoint is not task-specific (or the underlying
21477 language is not Ada), this attribute is @code{None}. This attribute
21481 @defivar Breakpoint ignore_count
21482 This attribute holds the ignore count for the breakpoint, an integer.
21483 This attribute is writable.
21486 @defivar Breakpoint number
21487 This attribute holds the breakpoint's number --- the identifier used by
21488 the user to manipulate the breakpoint. This attribute is not writable.
21491 @defivar Breakpoint type
21492 This attribute holds the breakpoint's type --- the identifier used to
21493 determine the actual breakpoint type or use-case. This attribute is not
21497 The available types are represented by constants defined in the @code{gdb}
21501 @findex BP_BREAKPOINT
21502 @findex gdb.BP_BREAKPOINT
21503 @item BP_BREAKPOINT
21504 Normal code breakpoint.
21506 @findex BP_WATCHPOINT
21507 @findex gdb.BP_WATCHPOINT
21508 @item BP_WATCHPOINT
21509 Watchpoint breakpoint.
21511 @findex BP_HARDWARE_WATCHPOINT
21512 @findex gdb.BP_HARDWARE_WATCHPOINT
21513 @item BP_HARDWARE_WATCHPOINT
21514 Hardware assisted watchpoint.
21516 @findex BP_READ_WATCHPOINT
21517 @findex gdb.BP_READ_WATCHPOINT
21518 @item BP_READ_WATCHPOINT
21519 Hardware assisted read watchpoint.
21521 @findex BP_ACCESS_WATCHPOINT
21522 @findex gdb.BP_ACCESS_WATCHPOINT
21523 @item BP_ACCESS_WATCHPOINT
21524 Hardware assisted access watchpoint.
21527 @defivar Breakpoint hit_count
21528 This attribute holds the hit count for the breakpoint, an integer.
21529 This attribute is writable, but currently it can only be set to zero.
21532 @defivar Breakpoint location
21533 This attribute holds the location of the breakpoint, as specified by
21534 the user. It is a string. If the breakpoint does not have a location
21535 (that is, it is a watchpoint) the attribute's value is @code{None}. This
21536 attribute is not writable.
21539 @defivar Breakpoint expression
21540 This attribute holds a breakpoint expression, as specified by
21541 the user. It is a string. If the breakpoint does not have an
21542 expression (the breakpoint is not a watchpoint) the attribute's value
21543 is @code{None}. This attribute is not writable.
21546 @defivar Breakpoint condition
21547 This attribute holds the condition of the breakpoint, as specified by
21548 the user. It is a string. If there is no condition, this attribute's
21549 value is @code{None}. This attribute is writable.
21552 @defivar Breakpoint commands
21553 This attribute holds the commands attached to the breakpoint. If
21554 there are commands, this attribute's value is a string holding all the
21555 commands, separated by newlines. If there are no commands, this
21556 attribute is @code{None}. This attribute is not writable.
21559 @node Lazy Strings In Python
21560 @subsubsection Python representation of lazy strings.
21562 @cindex lazy strings in python
21563 @tindex gdb.LazyString
21565 A @dfn{lazy string} is a string whose contents is not retrieved or
21566 encoded until it is needed.
21568 A @code{gdb.LazyString} is represented in @value{GDBN} as an
21569 @code{address} that points to a region of memory, an @code{encoding}
21570 that will be used to encode that region of memory, and a @code{length}
21571 to delimit the region of memory that represents the string. The
21572 difference between a @code{gdb.LazyString} and a string wrapped within
21573 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
21574 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
21575 retrieved and encoded during printing, while a @code{gdb.Value}
21576 wrapping a string is immediately retrieved and encoded on creation.
21578 A @code{gdb.LazyString} object has the following functions:
21580 @defmethod LazyString value
21581 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
21582 will point to the string in memory, but will lose all the delayed
21583 retrieval, encoding and handling that @value{GDBN} applies to a
21584 @code{gdb.LazyString}.
21587 @defivar LazyString address
21588 This attribute holds the address of the string. This attribute is not
21592 @defivar LazyString length
21593 This attribute holds the length of the string in characters. If the
21594 length is -1, then the string will be fetched and encoded up to the
21595 first null of appropriate width. This attribute is not writable.
21598 @defivar LazyString encoding
21599 This attribute holds the encoding that will be applied to the string
21600 when the string is printed by @value{GDBN}. If the encoding is not
21601 set, or contains an empty string, then @value{GDBN} will select the
21602 most appropriate encoding when the string is printed. This attribute
21606 @defivar LazyString type
21607 This attribute holds the type that is represented by the lazy string's
21608 type. For a lazy string this will always be a pointer type. To
21609 resolve this to the lazy string's character type, use the type's
21610 @code{target} method. @xref{Types In Python}. This attribute is not
21615 @chapter Command Interpreters
21616 @cindex command interpreters
21618 @value{GDBN} supports multiple command interpreters, and some command
21619 infrastructure to allow users or user interface writers to switch
21620 between interpreters or run commands in other interpreters.
21622 @value{GDBN} currently supports two command interpreters, the console
21623 interpreter (sometimes called the command-line interpreter or @sc{cli})
21624 and the machine interface interpreter (or @sc{gdb/mi}). This manual
21625 describes both of these interfaces in great detail.
21627 By default, @value{GDBN} will start with the console interpreter.
21628 However, the user may choose to start @value{GDBN} with another
21629 interpreter by specifying the @option{-i} or @option{--interpreter}
21630 startup options. Defined interpreters include:
21634 @cindex console interpreter
21635 The traditional console or command-line interpreter. This is the most often
21636 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
21637 @value{GDBN} will use this interpreter.
21640 @cindex mi interpreter
21641 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
21642 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
21643 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
21647 @cindex mi2 interpreter
21648 The current @sc{gdb/mi} interface.
21651 @cindex mi1 interpreter
21652 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
21656 @cindex invoke another interpreter
21657 The interpreter being used by @value{GDBN} may not be dynamically
21658 switched at runtime. Although possible, this could lead to a very
21659 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
21660 enters the command "interpreter-set console" in a console view,
21661 @value{GDBN} would switch to using the console interpreter, rendering
21662 the IDE inoperable!
21664 @kindex interpreter-exec
21665 Although you may only choose a single interpreter at startup, you may execute
21666 commands in any interpreter from the current interpreter using the appropriate
21667 command. If you are running the console interpreter, simply use the
21668 @code{interpreter-exec} command:
21671 interpreter-exec mi "-data-list-register-names"
21674 @sc{gdb/mi} has a similar command, although it is only available in versions of
21675 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
21678 @chapter @value{GDBN} Text User Interface
21680 @cindex Text User Interface
21683 * TUI Overview:: TUI overview
21684 * TUI Keys:: TUI key bindings
21685 * TUI Single Key Mode:: TUI single key mode
21686 * TUI Commands:: TUI-specific commands
21687 * TUI Configuration:: TUI configuration variables
21690 The @value{GDBN} Text User Interface (TUI) is a terminal
21691 interface which uses the @code{curses} library to show the source
21692 file, the assembly output, the program registers and @value{GDBN}
21693 commands in separate text windows. The TUI mode is supported only
21694 on platforms where a suitable version of the @code{curses} library
21697 @pindex @value{GDBTUI}
21698 The TUI mode is enabled by default when you invoke @value{GDBN} as
21699 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
21700 You can also switch in and out of TUI mode while @value{GDBN} runs by
21701 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
21702 @xref{TUI Keys, ,TUI Key Bindings}.
21705 @section TUI Overview
21707 In TUI mode, @value{GDBN} can display several text windows:
21711 This window is the @value{GDBN} command window with the @value{GDBN}
21712 prompt and the @value{GDBN} output. The @value{GDBN} input is still
21713 managed using readline.
21716 The source window shows the source file of the program. The current
21717 line and active breakpoints are displayed in this window.
21720 The assembly window shows the disassembly output of the program.
21723 This window shows the processor registers. Registers are highlighted
21724 when their values change.
21727 The source and assembly windows show the current program position
21728 by highlighting the current line and marking it with a @samp{>} marker.
21729 Breakpoints are indicated with two markers. The first marker
21730 indicates the breakpoint type:
21734 Breakpoint which was hit at least once.
21737 Breakpoint which was never hit.
21740 Hardware breakpoint which was hit at least once.
21743 Hardware breakpoint which was never hit.
21746 The second marker indicates whether the breakpoint is enabled or not:
21750 Breakpoint is enabled.
21753 Breakpoint is disabled.
21756 The source, assembly and register windows are updated when the current
21757 thread changes, when the frame changes, or when the program counter
21760 These windows are not all visible at the same time. The command
21761 window is always visible. The others can be arranged in several
21772 source and assembly,
21775 source and registers, or
21778 assembly and registers.
21781 A status line above the command window shows the following information:
21785 Indicates the current @value{GDBN} target.
21786 (@pxref{Targets, ,Specifying a Debugging Target}).
21789 Gives the current process or thread number.
21790 When no process is being debugged, this field is set to @code{No process}.
21793 Gives the current function name for the selected frame.
21794 The name is demangled if demangling is turned on (@pxref{Print Settings}).
21795 When there is no symbol corresponding to the current program counter,
21796 the string @code{??} is displayed.
21799 Indicates the current line number for the selected frame.
21800 When the current line number is not known, the string @code{??} is displayed.
21803 Indicates the current program counter address.
21807 @section TUI Key Bindings
21808 @cindex TUI key bindings
21810 The TUI installs several key bindings in the readline keymaps
21811 (@pxref{Command Line Editing}). The following key bindings
21812 are installed for both TUI mode and the @value{GDBN} standard mode.
21821 Enter or leave the TUI mode. When leaving the TUI mode,
21822 the curses window management stops and @value{GDBN} operates using
21823 its standard mode, writing on the terminal directly. When reentering
21824 the TUI mode, control is given back to the curses windows.
21825 The screen is then refreshed.
21829 Use a TUI layout with only one window. The layout will
21830 either be @samp{source} or @samp{assembly}. When the TUI mode
21831 is not active, it will switch to the TUI mode.
21833 Think of this key binding as the Emacs @kbd{C-x 1} binding.
21837 Use a TUI layout with at least two windows. When the current
21838 layout already has two windows, the next layout with two windows is used.
21839 When a new layout is chosen, one window will always be common to the
21840 previous layout and the new one.
21842 Think of it as the Emacs @kbd{C-x 2} binding.
21846 Change the active window. The TUI associates several key bindings
21847 (like scrolling and arrow keys) with the active window. This command
21848 gives the focus to the next TUI window.
21850 Think of it as the Emacs @kbd{C-x o} binding.
21854 Switch in and out of the TUI SingleKey mode that binds single
21855 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
21858 The following key bindings only work in the TUI mode:
21863 Scroll the active window one page up.
21867 Scroll the active window one page down.
21871 Scroll the active window one line up.
21875 Scroll the active window one line down.
21879 Scroll the active window one column left.
21883 Scroll the active window one column right.
21887 Refresh the screen.
21890 Because the arrow keys scroll the active window in the TUI mode, they
21891 are not available for their normal use by readline unless the command
21892 window has the focus. When another window is active, you must use
21893 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
21894 and @kbd{C-f} to control the command window.
21896 @node TUI Single Key Mode
21897 @section TUI Single Key Mode
21898 @cindex TUI single key mode
21900 The TUI also provides a @dfn{SingleKey} mode, which binds several
21901 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
21902 switch into this mode, where the following key bindings are used:
21905 @kindex c @r{(SingleKey TUI key)}
21909 @kindex d @r{(SingleKey TUI key)}
21913 @kindex f @r{(SingleKey TUI key)}
21917 @kindex n @r{(SingleKey TUI key)}
21921 @kindex q @r{(SingleKey TUI key)}
21923 exit the SingleKey mode.
21925 @kindex r @r{(SingleKey TUI key)}
21929 @kindex s @r{(SingleKey TUI key)}
21933 @kindex u @r{(SingleKey TUI key)}
21937 @kindex v @r{(SingleKey TUI key)}
21941 @kindex w @r{(SingleKey TUI key)}
21946 Other keys temporarily switch to the @value{GDBN} command prompt.
21947 The key that was pressed is inserted in the editing buffer so that
21948 it is possible to type most @value{GDBN} commands without interaction
21949 with the TUI SingleKey mode. Once the command is entered the TUI
21950 SingleKey mode is restored. The only way to permanently leave
21951 this mode is by typing @kbd{q} or @kbd{C-x s}.
21955 @section TUI-specific Commands
21956 @cindex TUI commands
21958 The TUI has specific commands to control the text windows.
21959 These commands are always available, even when @value{GDBN} is not in
21960 the TUI mode. When @value{GDBN} is in the standard mode, most
21961 of these commands will automatically switch to the TUI mode.
21963 Note that if @value{GDBN}'s @code{stdout} is not connected to a
21964 terminal, or @value{GDBN} has been started with the machine interface
21965 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
21966 these commands will fail with an error, because it would not be
21967 possible or desirable to enable curses window management.
21972 List and give the size of all displayed windows.
21976 Display the next layout.
21979 Display the previous layout.
21982 Display the source window only.
21985 Display the assembly window only.
21988 Display the source and assembly window.
21991 Display the register window together with the source or assembly window.
21995 Make the next window active for scrolling.
21998 Make the previous window active for scrolling.
22001 Make the source window active for scrolling.
22004 Make the assembly window active for scrolling.
22007 Make the register window active for scrolling.
22010 Make the command window active for scrolling.
22014 Refresh the screen. This is similar to typing @kbd{C-L}.
22016 @item tui reg float
22018 Show the floating point registers in the register window.
22020 @item tui reg general
22021 Show the general registers in the register window.
22024 Show the next register group. The list of register groups as well as
22025 their order is target specific. The predefined register groups are the
22026 following: @code{general}, @code{float}, @code{system}, @code{vector},
22027 @code{all}, @code{save}, @code{restore}.
22029 @item tui reg system
22030 Show the system registers in the register window.
22034 Update the source window and the current execution point.
22036 @item winheight @var{name} +@var{count}
22037 @itemx winheight @var{name} -@var{count}
22039 Change the height of the window @var{name} by @var{count}
22040 lines. Positive counts increase the height, while negative counts
22043 @item tabset @var{nchars}
22045 Set the width of tab stops to be @var{nchars} characters.
22048 @node TUI Configuration
22049 @section TUI Configuration Variables
22050 @cindex TUI configuration variables
22052 Several configuration variables control the appearance of TUI windows.
22055 @item set tui border-kind @var{kind}
22056 @kindex set tui border-kind
22057 Select the border appearance for the source, assembly and register windows.
22058 The possible values are the following:
22061 Use a space character to draw the border.
22064 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
22067 Use the Alternate Character Set to draw the border. The border is
22068 drawn using character line graphics if the terminal supports them.
22071 @item set tui border-mode @var{mode}
22072 @kindex set tui border-mode
22073 @itemx set tui active-border-mode @var{mode}
22074 @kindex set tui active-border-mode
22075 Select the display attributes for the borders of the inactive windows
22076 or the active window. The @var{mode} can be one of the following:
22079 Use normal attributes to display the border.
22085 Use reverse video mode.
22088 Use half bright mode.
22090 @item half-standout
22091 Use half bright and standout mode.
22094 Use extra bright or bold mode.
22096 @item bold-standout
22097 Use extra bright or bold and standout mode.
22102 @chapter Using @value{GDBN} under @sc{gnu} Emacs
22105 @cindex @sc{gnu} Emacs
22106 A special interface allows you to use @sc{gnu} Emacs to view (and
22107 edit) the source files for the program you are debugging with
22110 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
22111 executable file you want to debug as an argument. This command starts
22112 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
22113 created Emacs buffer.
22114 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
22116 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
22121 All ``terminal'' input and output goes through an Emacs buffer, called
22124 This applies both to @value{GDBN} commands and their output, and to the input
22125 and output done by the program you are debugging.
22127 This is useful because it means that you can copy the text of previous
22128 commands and input them again; you can even use parts of the output
22131 All the facilities of Emacs' Shell mode are available for interacting
22132 with your program. In particular, you can send signals the usual
22133 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
22137 @value{GDBN} displays source code through Emacs.
22139 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
22140 source file for that frame and puts an arrow (@samp{=>}) at the
22141 left margin of the current line. Emacs uses a separate buffer for
22142 source display, and splits the screen to show both your @value{GDBN} session
22145 Explicit @value{GDBN} @code{list} or search commands still produce output as
22146 usual, but you probably have no reason to use them from Emacs.
22149 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
22150 a graphical mode, enabled by default, which provides further buffers
22151 that can control the execution and describe the state of your program.
22152 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
22154 If you specify an absolute file name when prompted for the @kbd{M-x
22155 gdb} argument, then Emacs sets your current working directory to where
22156 your program resides. If you only specify the file name, then Emacs
22157 sets your current working directory to to the directory associated
22158 with the previous buffer. In this case, @value{GDBN} may find your
22159 program by searching your environment's @code{PATH} variable, but on
22160 some operating systems it might not find the source. So, although the
22161 @value{GDBN} input and output session proceeds normally, the auxiliary
22162 buffer does not display the current source and line of execution.
22164 The initial working directory of @value{GDBN} is printed on the top
22165 line of the GUD buffer and this serves as a default for the commands
22166 that specify files for @value{GDBN} to operate on. @xref{Files,
22167 ,Commands to Specify Files}.
22169 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
22170 need to call @value{GDBN} by a different name (for example, if you
22171 keep several configurations around, with different names) you can
22172 customize the Emacs variable @code{gud-gdb-command-name} to run the
22175 In the GUD buffer, you can use these special Emacs commands in
22176 addition to the standard Shell mode commands:
22180 Describe the features of Emacs' GUD Mode.
22183 Execute to another source line, like the @value{GDBN} @code{step} command; also
22184 update the display window to show the current file and location.
22187 Execute to next source line in this function, skipping all function
22188 calls, like the @value{GDBN} @code{next} command. Then update the display window
22189 to show the current file and location.
22192 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
22193 display window accordingly.
22196 Execute until exit from the selected stack frame, like the @value{GDBN}
22197 @code{finish} command.
22200 Continue execution of your program, like the @value{GDBN} @code{continue}
22204 Go up the number of frames indicated by the numeric argument
22205 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
22206 like the @value{GDBN} @code{up} command.
22209 Go down the number of frames indicated by the numeric argument, like the
22210 @value{GDBN} @code{down} command.
22213 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
22214 tells @value{GDBN} to set a breakpoint on the source line point is on.
22216 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
22217 separate frame which shows a backtrace when the GUD buffer is current.
22218 Move point to any frame in the stack and type @key{RET} to make it
22219 become the current frame and display the associated source in the
22220 source buffer. Alternatively, click @kbd{Mouse-2} to make the
22221 selected frame become the current one. In graphical mode, the
22222 speedbar displays watch expressions.
22224 If you accidentally delete the source-display buffer, an easy way to get
22225 it back is to type the command @code{f} in the @value{GDBN} buffer, to
22226 request a frame display; when you run under Emacs, this recreates
22227 the source buffer if necessary to show you the context of the current
22230 The source files displayed in Emacs are in ordinary Emacs buffers
22231 which are visiting the source files in the usual way. You can edit
22232 the files with these buffers if you wish; but keep in mind that @value{GDBN}
22233 communicates with Emacs in terms of line numbers. If you add or
22234 delete lines from the text, the line numbers that @value{GDBN} knows cease
22235 to correspond properly with the code.
22237 A more detailed description of Emacs' interaction with @value{GDBN} is
22238 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
22241 @c The following dropped because Epoch is nonstandard. Reactivate
22242 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
22244 @kindex Emacs Epoch environment
22248 Version 18 of @sc{gnu} Emacs has a built-in window system
22249 called the @code{epoch}
22250 environment. Users of this environment can use a new command,
22251 @code{inspect} which performs identically to @code{print} except that
22252 each value is printed in its own window.
22257 @chapter The @sc{gdb/mi} Interface
22259 @unnumberedsec Function and Purpose
22261 @cindex @sc{gdb/mi}, its purpose
22262 @sc{gdb/mi} is a line based machine oriented text interface to
22263 @value{GDBN} and is activated by specifying using the
22264 @option{--interpreter} command line option (@pxref{Mode Options}). It
22265 is specifically intended to support the development of systems which
22266 use the debugger as just one small component of a larger system.
22268 This chapter is a specification of the @sc{gdb/mi} interface. It is written
22269 in the form of a reference manual.
22271 Note that @sc{gdb/mi} is still under construction, so some of the
22272 features described below are incomplete and subject to change
22273 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
22275 @unnumberedsec Notation and Terminology
22277 @cindex notational conventions, for @sc{gdb/mi}
22278 This chapter uses the following notation:
22282 @code{|} separates two alternatives.
22285 @code{[ @var{something} ]} indicates that @var{something} is optional:
22286 it may or may not be given.
22289 @code{( @var{group} )*} means that @var{group} inside the parentheses
22290 may repeat zero or more times.
22293 @code{( @var{group} )+} means that @var{group} inside the parentheses
22294 may repeat one or more times.
22297 @code{"@var{string}"} means a literal @var{string}.
22301 @heading Dependencies
22305 * GDB/MI General Design::
22306 * GDB/MI Command Syntax::
22307 * GDB/MI Compatibility with CLI::
22308 * GDB/MI Development and Front Ends::
22309 * GDB/MI Output Records::
22310 * GDB/MI Simple Examples::
22311 * GDB/MI Command Description Format::
22312 * GDB/MI Breakpoint Commands::
22313 * GDB/MI Program Context::
22314 * GDB/MI Thread Commands::
22315 * GDB/MI Program Execution::
22316 * GDB/MI Stack Manipulation::
22317 * GDB/MI Variable Objects::
22318 * GDB/MI Data Manipulation::
22319 * GDB/MI Tracepoint Commands::
22320 * GDB/MI Symbol Query::
22321 * GDB/MI File Commands::
22323 * GDB/MI Kod Commands::
22324 * GDB/MI Memory Overlay Commands::
22325 * GDB/MI Signal Handling Commands::
22327 * GDB/MI Target Manipulation::
22328 * GDB/MI File Transfer Commands::
22329 * GDB/MI Miscellaneous Commands::
22332 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22333 @node GDB/MI General Design
22334 @section @sc{gdb/mi} General Design
22335 @cindex GDB/MI General Design
22337 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
22338 parts---commands sent to @value{GDBN}, responses to those commands
22339 and notifications. Each command results in exactly one response,
22340 indicating either successful completion of the command, or an error.
22341 For the commands that do not resume the target, the response contains the
22342 requested information. For the commands that resume the target, the
22343 response only indicates whether the target was successfully resumed.
22344 Notifications is the mechanism for reporting changes in the state of the
22345 target, or in @value{GDBN} state, that cannot conveniently be associated with
22346 a command and reported as part of that command response.
22348 The important examples of notifications are:
22352 Exec notifications. These are used to report changes in
22353 target state---when a target is resumed, or stopped. It would not
22354 be feasible to include this information in response of resuming
22355 commands, because one resume commands can result in multiple events in
22356 different threads. Also, quite some time may pass before any event
22357 happens in the target, while a frontend needs to know whether the resuming
22358 command itself was successfully executed.
22361 Console output, and status notifications. Console output
22362 notifications are used to report output of CLI commands, as well as
22363 diagnostics for other commands. Status notifications are used to
22364 report the progress of a long-running operation. Naturally, including
22365 this information in command response would mean no output is produced
22366 until the command is finished, which is undesirable.
22369 General notifications. Commands may have various side effects on
22370 the @value{GDBN} or target state beyond their official purpose. For example,
22371 a command may change the selected thread. Although such changes can
22372 be included in command response, using notification allows for more
22373 orthogonal frontend design.
22377 There's no guarantee that whenever an MI command reports an error,
22378 @value{GDBN} or the target are in any specific state, and especially,
22379 the state is not reverted to the state before the MI command was
22380 processed. Therefore, whenever an MI command results in an error,
22381 we recommend that the frontend refreshes all the information shown in
22382 the user interface.
22386 * Context management::
22387 * Asynchronous and non-stop modes::
22391 @node Context management
22392 @subsection Context management
22394 In most cases when @value{GDBN} accesses the target, this access is
22395 done in context of a specific thread and frame (@pxref{Frames}).
22396 Often, even when accessing global data, the target requires that a thread
22397 be specified. The CLI interface maintains the selected thread and frame,
22398 and supplies them to target on each command. This is convenient,
22399 because a command line user would not want to specify that information
22400 explicitly on each command, and because user interacts with
22401 @value{GDBN} via a single terminal, so no confusion is possible as
22402 to what thread and frame are the current ones.
22404 In the case of MI, the concept of selected thread and frame is less
22405 useful. First, a frontend can easily remember this information
22406 itself. Second, a graphical frontend can have more than one window,
22407 each one used for debugging a different thread, and the frontend might
22408 want to access additional threads for internal purposes. This
22409 increases the risk that by relying on implicitly selected thread, the
22410 frontend may be operating on a wrong one. Therefore, each MI command
22411 should explicitly specify which thread and frame to operate on. To
22412 make it possible, each MI command accepts the @samp{--thread} and
22413 @samp{--frame} options, the value to each is @value{GDBN} identifier
22414 for thread and frame to operate on.
22416 Usually, each top-level window in a frontend allows the user to select
22417 a thread and a frame, and remembers the user selection for further
22418 operations. However, in some cases @value{GDBN} may suggest that the
22419 current thread be changed. For example, when stopping on a breakpoint
22420 it is reasonable to switch to the thread where breakpoint is hit. For
22421 another example, if the user issues the CLI @samp{thread} command via
22422 the frontend, it is desirable to change the frontend's selected thread to the
22423 one specified by user. @value{GDBN} communicates the suggestion to
22424 change current thread using the @samp{=thread-selected} notification.
22425 No such notification is available for the selected frame at the moment.
22427 Note that historically, MI shares the selected thread with CLI, so
22428 frontends used the @code{-thread-select} to execute commands in the
22429 right context. However, getting this to work right is cumbersome. The
22430 simplest way is for frontend to emit @code{-thread-select} command
22431 before every command. This doubles the number of commands that need
22432 to be sent. The alternative approach is to suppress @code{-thread-select}
22433 if the selected thread in @value{GDBN} is supposed to be identical to the
22434 thread the frontend wants to operate on. However, getting this
22435 optimization right can be tricky. In particular, if the frontend
22436 sends several commands to @value{GDBN}, and one of the commands changes the
22437 selected thread, then the behaviour of subsequent commands will
22438 change. So, a frontend should either wait for response from such
22439 problematic commands, or explicitly add @code{-thread-select} for
22440 all subsequent commands. No frontend is known to do this exactly
22441 right, so it is suggested to just always pass the @samp{--thread} and
22442 @samp{--frame} options.
22444 @node Asynchronous and non-stop modes
22445 @subsection Asynchronous command execution and non-stop mode
22447 On some targets, @value{GDBN} is capable of processing MI commands
22448 even while the target is running. This is called @dfn{asynchronous
22449 command execution} (@pxref{Background Execution}). The frontend may
22450 specify a preferrence for asynchronous execution using the
22451 @code{-gdb-set target-async 1} command, which should be emitted before
22452 either running the executable or attaching to the target. After the
22453 frontend has started the executable or attached to the target, it can
22454 find if asynchronous execution is enabled using the
22455 @code{-list-target-features} command.
22457 Even if @value{GDBN} can accept a command while target is running,
22458 many commands that access the target do not work when the target is
22459 running. Therefore, asynchronous command execution is most useful
22460 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
22461 it is possible to examine the state of one thread, while other threads
22464 When a given thread is running, MI commands that try to access the
22465 target in the context of that thread may not work, or may work only on
22466 some targets. In particular, commands that try to operate on thread's
22467 stack will not work, on any target. Commands that read memory, or
22468 modify breakpoints, may work or not work, depending on the target. Note
22469 that even commands that operate on global state, such as @code{print},
22470 @code{set}, and breakpoint commands, still access the target in the
22471 context of a specific thread, so frontend should try to find a
22472 stopped thread and perform the operation on that thread (using the
22473 @samp{--thread} option).
22475 Which commands will work in the context of a running thread is
22476 highly target dependent. However, the two commands
22477 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
22478 to find the state of a thread, will always work.
22480 @node Thread groups
22481 @subsection Thread groups
22482 @value{GDBN} may be used to debug several processes at the same time.
22483 On some platfroms, @value{GDBN} may support debugging of several
22484 hardware systems, each one having several cores with several different
22485 processes running on each core. This section describes the MI
22486 mechanism to support such debugging scenarios.
22488 The key observation is that regardless of the structure of the
22489 target, MI can have a global list of threads, because most commands that
22490 accept the @samp{--thread} option do not need to know what process that
22491 thread belongs to. Therefore, it is not necessary to introduce
22492 neither additional @samp{--process} option, nor an notion of the
22493 current process in the MI interface. The only strictly new feature
22494 that is required is the ability to find how the threads are grouped
22497 To allow the user to discover such grouping, and to support arbitrary
22498 hierarchy of machines/cores/processes, MI introduces the concept of a
22499 @dfn{thread group}. Thread group is a collection of threads and other
22500 thread groups. A thread group always has a string identifier, a type,
22501 and may have additional attributes specific to the type. A new
22502 command, @code{-list-thread-groups}, returns the list of top-level
22503 thread groups, which correspond to processes that @value{GDBN} is
22504 debugging at the moment. By passing an identifier of a thread group
22505 to the @code{-list-thread-groups} command, it is possible to obtain
22506 the members of specific thread group.
22508 To allow the user to easily discover processes, and other objects, he
22509 wishes to debug, a concept of @dfn{available thread group} is
22510 introduced. Available thread group is an thread group that
22511 @value{GDBN} is not debugging, but that can be attached to, using the
22512 @code{-target-attach} command. The list of available top-level thread
22513 groups can be obtained using @samp{-list-thread-groups --available}.
22514 In general, the content of a thread group may be only retrieved only
22515 after attaching to that thread group.
22517 Thread groups are related to inferiors (@pxref{Inferiors and
22518 Programs}). Each inferior corresponds to a thread group of a special
22519 type @samp{process}, and some additional operations are permitted on
22520 such thread groups.
22522 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22523 @node GDB/MI Command Syntax
22524 @section @sc{gdb/mi} Command Syntax
22527 * GDB/MI Input Syntax::
22528 * GDB/MI Output Syntax::
22531 @node GDB/MI Input Syntax
22532 @subsection @sc{gdb/mi} Input Syntax
22534 @cindex input syntax for @sc{gdb/mi}
22535 @cindex @sc{gdb/mi}, input syntax
22537 @item @var{command} @expansion{}
22538 @code{@var{cli-command} | @var{mi-command}}
22540 @item @var{cli-command} @expansion{}
22541 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
22542 @var{cli-command} is any existing @value{GDBN} CLI command.
22544 @item @var{mi-command} @expansion{}
22545 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
22546 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
22548 @item @var{token} @expansion{}
22549 "any sequence of digits"
22551 @item @var{option} @expansion{}
22552 @code{"-" @var{parameter} [ " " @var{parameter} ]}
22554 @item @var{parameter} @expansion{}
22555 @code{@var{non-blank-sequence} | @var{c-string}}
22557 @item @var{operation} @expansion{}
22558 @emph{any of the operations described in this chapter}
22560 @item @var{non-blank-sequence} @expansion{}
22561 @emph{anything, provided it doesn't contain special characters such as
22562 "-", @var{nl}, """ and of course " "}
22564 @item @var{c-string} @expansion{}
22565 @code{""" @var{seven-bit-iso-c-string-content} """}
22567 @item @var{nl} @expansion{}
22576 The CLI commands are still handled by the @sc{mi} interpreter; their
22577 output is described below.
22580 The @code{@var{token}}, when present, is passed back when the command
22584 Some @sc{mi} commands accept optional arguments as part of the parameter
22585 list. Each option is identified by a leading @samp{-} (dash) and may be
22586 followed by an optional argument parameter. Options occur first in the
22587 parameter list and can be delimited from normal parameters using
22588 @samp{--} (this is useful when some parameters begin with a dash).
22595 We want easy access to the existing CLI syntax (for debugging).
22598 We want it to be easy to spot a @sc{mi} operation.
22601 @node GDB/MI Output Syntax
22602 @subsection @sc{gdb/mi} Output Syntax
22604 @cindex output syntax of @sc{gdb/mi}
22605 @cindex @sc{gdb/mi}, output syntax
22606 The output from @sc{gdb/mi} consists of zero or more out-of-band records
22607 followed, optionally, by a single result record. This result record
22608 is for the most recent command. The sequence of output records is
22609 terminated by @samp{(gdb)}.
22611 If an input command was prefixed with a @code{@var{token}} then the
22612 corresponding output for that command will also be prefixed by that same
22616 @item @var{output} @expansion{}
22617 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
22619 @item @var{result-record} @expansion{}
22620 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
22622 @item @var{out-of-band-record} @expansion{}
22623 @code{@var{async-record} | @var{stream-record}}
22625 @item @var{async-record} @expansion{}
22626 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
22628 @item @var{exec-async-output} @expansion{}
22629 @code{[ @var{token} ] "*" @var{async-output}}
22631 @item @var{status-async-output} @expansion{}
22632 @code{[ @var{token} ] "+" @var{async-output}}
22634 @item @var{notify-async-output} @expansion{}
22635 @code{[ @var{token} ] "=" @var{async-output}}
22637 @item @var{async-output} @expansion{}
22638 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
22640 @item @var{result-class} @expansion{}
22641 @code{"done" | "running" | "connected" | "error" | "exit"}
22643 @item @var{async-class} @expansion{}
22644 @code{"stopped" | @var{others}} (where @var{others} will be added
22645 depending on the needs---this is still in development).
22647 @item @var{result} @expansion{}
22648 @code{ @var{variable} "=" @var{value}}
22650 @item @var{variable} @expansion{}
22651 @code{ @var{string} }
22653 @item @var{value} @expansion{}
22654 @code{ @var{const} | @var{tuple} | @var{list} }
22656 @item @var{const} @expansion{}
22657 @code{@var{c-string}}
22659 @item @var{tuple} @expansion{}
22660 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
22662 @item @var{list} @expansion{}
22663 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
22664 @var{result} ( "," @var{result} )* "]" }
22666 @item @var{stream-record} @expansion{}
22667 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
22669 @item @var{console-stream-output} @expansion{}
22670 @code{"~" @var{c-string}}
22672 @item @var{target-stream-output} @expansion{}
22673 @code{"@@" @var{c-string}}
22675 @item @var{log-stream-output} @expansion{}
22676 @code{"&" @var{c-string}}
22678 @item @var{nl} @expansion{}
22681 @item @var{token} @expansion{}
22682 @emph{any sequence of digits}.
22690 All output sequences end in a single line containing a period.
22693 The @code{@var{token}} is from the corresponding request. Note that
22694 for all async output, while the token is allowed by the grammar and
22695 may be output by future versions of @value{GDBN} for select async
22696 output messages, it is generally omitted. Frontends should treat
22697 all async output as reporting general changes in the state of the
22698 target and there should be no need to associate async output to any
22702 @cindex status output in @sc{gdb/mi}
22703 @var{status-async-output} contains on-going status information about the
22704 progress of a slow operation. It can be discarded. All status output is
22705 prefixed by @samp{+}.
22708 @cindex async output in @sc{gdb/mi}
22709 @var{exec-async-output} contains asynchronous state change on the target
22710 (stopped, started, disappeared). All async output is prefixed by
22714 @cindex notify output in @sc{gdb/mi}
22715 @var{notify-async-output} contains supplementary information that the
22716 client should handle (e.g., a new breakpoint information). All notify
22717 output is prefixed by @samp{=}.
22720 @cindex console output in @sc{gdb/mi}
22721 @var{console-stream-output} is output that should be displayed as is in the
22722 console. It is the textual response to a CLI command. All the console
22723 output is prefixed by @samp{~}.
22726 @cindex target output in @sc{gdb/mi}
22727 @var{target-stream-output} is the output produced by the target program.
22728 All the target output is prefixed by @samp{@@}.
22731 @cindex log output in @sc{gdb/mi}
22732 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
22733 instance messages that should be displayed as part of an error log. All
22734 the log output is prefixed by @samp{&}.
22737 @cindex list output in @sc{gdb/mi}
22738 New @sc{gdb/mi} commands should only output @var{lists} containing
22744 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
22745 details about the various output records.
22747 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22748 @node GDB/MI Compatibility with CLI
22749 @section @sc{gdb/mi} Compatibility with CLI
22751 @cindex compatibility, @sc{gdb/mi} and CLI
22752 @cindex @sc{gdb/mi}, compatibility with CLI
22754 For the developers convenience CLI commands can be entered directly,
22755 but there may be some unexpected behaviour. For example, commands
22756 that query the user will behave as if the user replied yes, breakpoint
22757 command lists are not executed and some CLI commands, such as
22758 @code{if}, @code{when} and @code{define}, prompt for further input with
22759 @samp{>}, which is not valid MI output.
22761 This feature may be removed at some stage in the future and it is
22762 recommended that front ends use the @code{-interpreter-exec} command
22763 (@pxref{-interpreter-exec}).
22765 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22766 @node GDB/MI Development and Front Ends
22767 @section @sc{gdb/mi} Development and Front Ends
22768 @cindex @sc{gdb/mi} development
22770 The application which takes the MI output and presents the state of the
22771 program being debugged to the user is called a @dfn{front end}.
22773 Although @sc{gdb/mi} is still incomplete, it is currently being used
22774 by a variety of front ends to @value{GDBN}. This makes it difficult
22775 to introduce new functionality without breaking existing usage. This
22776 section tries to minimize the problems by describing how the protocol
22779 Some changes in MI need not break a carefully designed front end, and
22780 for these the MI version will remain unchanged. The following is a
22781 list of changes that may occur within one level, so front ends should
22782 parse MI output in a way that can handle them:
22786 New MI commands may be added.
22789 New fields may be added to the output of any MI command.
22792 The range of values for fields with specified values, e.g.,
22793 @code{in_scope} (@pxref{-var-update}) may be extended.
22795 @c The format of field's content e.g type prefix, may change so parse it
22796 @c at your own risk. Yes, in general?
22798 @c The order of fields may change? Shouldn't really matter but it might
22799 @c resolve inconsistencies.
22802 If the changes are likely to break front ends, the MI version level
22803 will be increased by one. This will allow the front end to parse the
22804 output according to the MI version. Apart from mi0, new versions of
22805 @value{GDBN} will not support old versions of MI and it will be the
22806 responsibility of the front end to work with the new one.
22808 @c Starting with mi3, add a new command -mi-version that prints the MI
22811 The best way to avoid unexpected changes in MI that might break your front
22812 end is to make your project known to @value{GDBN} developers and
22813 follow development on @email{gdb@@sourceware.org} and
22814 @email{gdb-patches@@sourceware.org}.
22815 @cindex mailing lists
22817 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22818 @node GDB/MI Output Records
22819 @section @sc{gdb/mi} Output Records
22822 * GDB/MI Result Records::
22823 * GDB/MI Stream Records::
22824 * GDB/MI Async Records::
22825 * GDB/MI Frame Information::
22826 * GDB/MI Thread Information::
22829 @node GDB/MI Result Records
22830 @subsection @sc{gdb/mi} Result Records
22832 @cindex result records in @sc{gdb/mi}
22833 @cindex @sc{gdb/mi}, result records
22834 In addition to a number of out-of-band notifications, the response to a
22835 @sc{gdb/mi} command includes one of the following result indications:
22839 @item "^done" [ "," @var{results} ]
22840 The synchronous operation was successful, @code{@var{results}} are the return
22845 This result record is equivalent to @samp{^done}. Historically, it
22846 was output instead of @samp{^done} if the command has resumed the
22847 target. This behaviour is maintained for backward compatibility, but
22848 all frontends should treat @samp{^done} and @samp{^running}
22849 identically and rely on the @samp{*running} output record to determine
22850 which threads are resumed.
22854 @value{GDBN} has connected to a remote target.
22856 @item "^error" "," @var{c-string}
22858 The operation failed. The @code{@var{c-string}} contains the corresponding
22863 @value{GDBN} has terminated.
22867 @node GDB/MI Stream Records
22868 @subsection @sc{gdb/mi} Stream Records
22870 @cindex @sc{gdb/mi}, stream records
22871 @cindex stream records in @sc{gdb/mi}
22872 @value{GDBN} internally maintains a number of output streams: the console, the
22873 target, and the log. The output intended for each of these streams is
22874 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
22876 Each stream record begins with a unique @dfn{prefix character} which
22877 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
22878 Syntax}). In addition to the prefix, each stream record contains a
22879 @code{@var{string-output}}. This is either raw text (with an implicit new
22880 line) or a quoted C string (which does not contain an implicit newline).
22883 @item "~" @var{string-output}
22884 The console output stream contains text that should be displayed in the
22885 CLI console window. It contains the textual responses to CLI commands.
22887 @item "@@" @var{string-output}
22888 The target output stream contains any textual output from the running
22889 target. This is only present when GDB's event loop is truly
22890 asynchronous, which is currently only the case for remote targets.
22892 @item "&" @var{string-output}
22893 The log stream contains debugging messages being produced by @value{GDBN}'s
22897 @node GDB/MI Async Records
22898 @subsection @sc{gdb/mi} Async Records
22900 @cindex async records in @sc{gdb/mi}
22901 @cindex @sc{gdb/mi}, async records
22902 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
22903 additional changes that have occurred. Those changes can either be a
22904 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
22905 target activity (e.g., target stopped).
22907 The following is the list of possible async records:
22911 @item *running,thread-id="@var{thread}"
22912 The target is now running. The @var{thread} field tells which
22913 specific thread is now running, and can be @samp{all} if all threads
22914 are running. The frontend should assume that no interaction with a
22915 running thread is possible after this notification is produced.
22916 The frontend should not assume that this notification is output
22917 only once for any command. @value{GDBN} may emit this notification
22918 several times, either for different threads, because it cannot resume
22919 all threads together, or even for a single thread, if the thread must
22920 be stepped though some code before letting it run freely.
22922 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
22923 The target has stopped. The @var{reason} field can have one of the
22927 @item breakpoint-hit
22928 A breakpoint was reached.
22929 @item watchpoint-trigger
22930 A watchpoint was triggered.
22931 @item read-watchpoint-trigger
22932 A read watchpoint was triggered.
22933 @item access-watchpoint-trigger
22934 An access watchpoint was triggered.
22935 @item function-finished
22936 An -exec-finish or similar CLI command was accomplished.
22937 @item location-reached
22938 An -exec-until or similar CLI command was accomplished.
22939 @item watchpoint-scope
22940 A watchpoint has gone out of scope.
22941 @item end-stepping-range
22942 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
22943 similar CLI command was accomplished.
22944 @item exited-signalled
22945 The inferior exited because of a signal.
22947 The inferior exited.
22948 @item exited-normally
22949 The inferior exited normally.
22950 @item signal-received
22951 A signal was received by the inferior.
22954 The @var{id} field identifies the thread that directly caused the stop
22955 -- for example by hitting a breakpoint. Depending on whether all-stop
22956 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
22957 stop all threads, or only the thread that directly triggered the stop.
22958 If all threads are stopped, the @var{stopped} field will have the
22959 value of @code{"all"}. Otherwise, the value of the @var{stopped}
22960 field will be a list of thread identifiers. Presently, this list will
22961 always include a single thread, but frontend should be prepared to see
22962 several threads in the list. The @var{core} field reports the
22963 processor core on which the stop event has happened. This field may be absent
22964 if such information is not available.
22966 @item =thread-group-added,id="@var{id}"
22967 @itemx =thread-group-removed,id="@var{id}"
22968 A thread group was either added or removed. The @var{id} field
22969 contains the @value{GDBN} identifier of the thread group. When a thread
22970 group is added, it generally might not be associated with a running
22971 process. When a thread group is removed, its id becomes invalid and
22972 cannot be used in any way.
22974 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
22975 A thread group became associated with a running program,
22976 either because the program was just started or the thread group
22977 was attached to a program. The @var{id} field contains the
22978 @value{GDBN} identifier of the thread group. The @var{pid} field
22979 contains process identifier, specific to the operating system.
22981 @itemx =thread-group-exited,id="@var{id}"
22982 A thread group is no longer associated with a running program,
22983 either because the program has exited, or because it was detached
22984 from. The @var{id} field contains the @value{GDBN} identifier of the
22987 @item =thread-created,id="@var{id}",group-id="@var{gid}"
22988 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
22989 A thread either was created, or has exited. The @var{id} field
22990 contains the @value{GDBN} identifier of the thread. The @var{gid}
22991 field identifies the thread group this thread belongs to.
22993 @item =thread-selected,id="@var{id}"
22994 Informs that the selected thread was changed as result of the last
22995 command. This notification is not emitted as result of @code{-thread-select}
22996 command but is emitted whenever an MI command that is not documented
22997 to change the selected thread actually changes it. In particular,
22998 invoking, directly or indirectly (via user-defined command), the CLI
22999 @code{thread} command, will generate this notification.
23001 We suggest that in response to this notification, front ends
23002 highlight the selected thread and cause subsequent commands to apply to
23005 @item =library-loaded,...
23006 Reports that a new library file was loaded by the program. This
23007 notification has 4 fields---@var{id}, @var{target-name},
23008 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
23009 opaque identifier of the library. For remote debugging case,
23010 @var{target-name} and @var{host-name} fields give the name of the
23011 library file on the target, and on the host respectively. For native
23012 debugging, both those fields have the same value. The
23013 @var{symbols-loaded} field reports if the debug symbols for this
23014 library are loaded. The @var{thread-group} field, if present,
23015 specifies the id of the thread group in whose context the library was loaded.
23016 If the field is absent, it means the library was loaded in the context
23017 of all present thread groups.
23019 @item =library-unloaded,...
23020 Reports that a library was unloaded by the program. This notification
23021 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
23022 the same meaning as for the @code{=library-loaded} notification.
23023 The @var{thread-group} field, if present, specifies the id of the
23024 thread group in whose context the library was unloaded. If the field is
23025 absent, it means the library was unloaded in the context of all present
23030 @node GDB/MI Frame Information
23031 @subsection @sc{gdb/mi} Frame Information
23033 Response from many MI commands includes an information about stack
23034 frame. This information is a tuple that may have the following
23039 The level of the stack frame. The innermost frame has the level of
23040 zero. This field is always present.
23043 The name of the function corresponding to the frame. This field may
23044 be absent if @value{GDBN} is unable to determine the function name.
23047 The code address for the frame. This field is always present.
23050 The name of the source files that correspond to the frame's code
23051 address. This field may be absent.
23054 The source line corresponding to the frames' code address. This field
23058 The name of the binary file (either executable or shared library) the
23059 corresponds to the frame's code address. This field may be absent.
23063 @node GDB/MI Thread Information
23064 @subsection @sc{gdb/mi} Thread Information
23066 Whenever @value{GDBN} has to report an information about a thread, it
23067 uses a tuple with the following fields:
23071 The numeric id assigned to the thread by @value{GDBN}. This field is
23075 Target-specific string identifying the thread. This field is always present.
23078 Additional information about the thread provided by the target.
23079 It is supposed to be human-readable and not interpreted by the
23080 frontend. This field is optional.
23083 Either @samp{stopped} or @samp{running}, depending on whether the
23084 thread is presently running. This field is always present.
23087 The value of this field is an integer number of the processor core the
23088 thread was last seen on. This field is optional.
23092 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23093 @node GDB/MI Simple Examples
23094 @section Simple Examples of @sc{gdb/mi} Interaction
23095 @cindex @sc{gdb/mi}, simple examples
23097 This subsection presents several simple examples of interaction using
23098 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
23099 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
23100 the output received from @sc{gdb/mi}.
23102 Note the line breaks shown in the examples are here only for
23103 readability, they don't appear in the real output.
23105 @subheading Setting a Breakpoint
23107 Setting a breakpoint generates synchronous output which contains detailed
23108 information of the breakpoint.
23111 -> -break-insert main
23112 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
23113 enabled="y",addr="0x08048564",func="main",file="myprog.c",
23114 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
23118 @subheading Program Execution
23120 Program execution generates asynchronous records and MI gives the
23121 reason that execution stopped.
23127 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
23128 frame=@{addr="0x08048564",func="main",
23129 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
23130 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
23135 <- *stopped,reason="exited-normally"
23139 @subheading Quitting @value{GDBN}
23141 Quitting @value{GDBN} just prints the result class @samp{^exit}.
23149 Please note that @samp{^exit} is printed immediately, but it might
23150 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
23151 performs necessary cleanups, including killing programs being debugged
23152 or disconnecting from debug hardware, so the frontend should wait till
23153 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
23154 fails to exit in reasonable time.
23156 @subheading A Bad Command
23158 Here's what happens if you pass a non-existent command:
23162 <- ^error,msg="Undefined MI command: rubbish"
23167 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23168 @node GDB/MI Command Description Format
23169 @section @sc{gdb/mi} Command Description Format
23171 The remaining sections describe blocks of commands. Each block of
23172 commands is laid out in a fashion similar to this section.
23174 @subheading Motivation
23176 The motivation for this collection of commands.
23178 @subheading Introduction
23180 A brief introduction to this collection of commands as a whole.
23182 @subheading Commands
23184 For each command in the block, the following is described:
23186 @subsubheading Synopsis
23189 -command @var{args}@dots{}
23192 @subsubheading Result
23194 @subsubheading @value{GDBN} Command
23196 The corresponding @value{GDBN} CLI command(s), if any.
23198 @subsubheading Example
23200 Example(s) formatted for readability. Some of the described commands have
23201 not been implemented yet and these are labeled N.A.@: (not available).
23204 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23205 @node GDB/MI Breakpoint Commands
23206 @section @sc{gdb/mi} Breakpoint Commands
23208 @cindex breakpoint commands for @sc{gdb/mi}
23209 @cindex @sc{gdb/mi}, breakpoint commands
23210 This section documents @sc{gdb/mi} commands for manipulating
23213 @subheading The @code{-break-after} Command
23214 @findex -break-after
23216 @subsubheading Synopsis
23219 -break-after @var{number} @var{count}
23222 The breakpoint number @var{number} is not in effect until it has been
23223 hit @var{count} times. To see how this is reflected in the output of
23224 the @samp{-break-list} command, see the description of the
23225 @samp{-break-list} command below.
23227 @subsubheading @value{GDBN} Command
23229 The corresponding @value{GDBN} command is @samp{ignore}.
23231 @subsubheading Example
23236 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
23237 enabled="y",addr="0x000100d0",func="main",file="hello.c",
23238 fullname="/home/foo/hello.c",line="5",times="0"@}
23245 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
23246 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23247 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23248 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23249 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23250 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23251 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23252 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
23253 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
23254 line="5",times="0",ignore="3"@}]@}
23259 @subheading The @code{-break-catch} Command
23260 @findex -break-catch
23263 @subheading The @code{-break-commands} Command
23264 @findex -break-commands
23266 @subsubheading Synopsis
23269 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
23272 Specifies the CLI commands that should be executed when breakpoint
23273 @var{number} is hit. The parameters @var{command1} to @var{commandN}
23274 are the commands. If no command is specified, any previously-set
23275 commands are cleared. @xref{Break Commands}. Typical use of this
23276 functionality is tracing a program, that is, printing of values of
23277 some variables whenever breakpoint is hit and then continuing.
23279 @subsubheading @value{GDBN} Command
23281 The corresponding @value{GDBN} command is @samp{commands}.
23283 @subsubheading Example
23288 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
23289 enabled="y",addr="0x000100d0",func="main",file="hello.c",
23290 fullname="/home/foo/hello.c",line="5",times="0"@}
23292 -break-commands 1 "print v" "continue"
23297 @subheading The @code{-break-condition} Command
23298 @findex -break-condition
23300 @subsubheading Synopsis
23303 -break-condition @var{number} @var{expr}
23306 Breakpoint @var{number} will stop the program only if the condition in
23307 @var{expr} is true. The condition becomes part of the
23308 @samp{-break-list} output (see the description of the @samp{-break-list}
23311 @subsubheading @value{GDBN} Command
23313 The corresponding @value{GDBN} command is @samp{condition}.
23315 @subsubheading Example
23319 -break-condition 1 1
23323 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
23324 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23325 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23326 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23327 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23328 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23329 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23330 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
23331 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
23332 line="5",cond="1",times="0",ignore="3"@}]@}
23336 @subheading The @code{-break-delete} Command
23337 @findex -break-delete
23339 @subsubheading Synopsis
23342 -break-delete ( @var{breakpoint} )+
23345 Delete the breakpoint(s) whose number(s) are specified in the argument
23346 list. This is obviously reflected in the breakpoint list.
23348 @subsubheading @value{GDBN} Command
23350 The corresponding @value{GDBN} command is @samp{delete}.
23352 @subsubheading Example
23360 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
23361 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23362 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23363 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23364 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23365 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23366 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23371 @subheading The @code{-break-disable} Command
23372 @findex -break-disable
23374 @subsubheading Synopsis
23377 -break-disable ( @var{breakpoint} )+
23380 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
23381 break list is now set to @samp{n} for the named @var{breakpoint}(s).
23383 @subsubheading @value{GDBN} Command
23385 The corresponding @value{GDBN} command is @samp{disable}.
23387 @subsubheading Example
23395 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
23396 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23397 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23398 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23399 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23400 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23401 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23402 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
23403 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
23404 line="5",times="0"@}]@}
23408 @subheading The @code{-break-enable} Command
23409 @findex -break-enable
23411 @subsubheading Synopsis
23414 -break-enable ( @var{breakpoint} )+
23417 Enable (previously disabled) @var{breakpoint}(s).
23419 @subsubheading @value{GDBN} Command
23421 The corresponding @value{GDBN} command is @samp{enable}.
23423 @subsubheading Example
23431 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
23432 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23433 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23434 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23435 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23436 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23437 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23438 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
23439 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
23440 line="5",times="0"@}]@}
23444 @subheading The @code{-break-info} Command
23445 @findex -break-info
23447 @subsubheading Synopsis
23450 -break-info @var{breakpoint}
23454 Get information about a single breakpoint.
23456 @subsubheading @value{GDBN} Command
23458 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
23460 @subsubheading Example
23463 @subheading The @code{-break-insert} Command
23464 @findex -break-insert
23466 @subsubheading Synopsis
23469 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
23470 [ -c @var{condition} ] [ -i @var{ignore-count} ]
23471 [ -p @var{thread} ] [ @var{location} ]
23475 If specified, @var{location}, can be one of:
23482 @item filename:linenum
23483 @item filename:function
23487 The possible optional parameters of this command are:
23491 Insert a temporary breakpoint.
23493 Insert a hardware breakpoint.
23494 @item -c @var{condition}
23495 Make the breakpoint conditional on @var{condition}.
23496 @item -i @var{ignore-count}
23497 Initialize the @var{ignore-count}.
23499 If @var{location} cannot be parsed (for example if it
23500 refers to unknown files or functions), create a pending
23501 breakpoint. Without this flag, @value{GDBN} will report
23502 an error, and won't create a breakpoint, if @var{location}
23505 Create a disabled breakpoint.
23507 Create a tracepoint. @xref{Tracepoints}. When this parameter
23508 is used together with @samp{-h}, a fast tracepoint is created.
23511 @subsubheading Result
23513 The result is in the form:
23516 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
23517 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
23518 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
23519 times="@var{times}"@}
23523 where @var{number} is the @value{GDBN} number for this breakpoint,
23524 @var{funcname} is the name of the function where the breakpoint was
23525 inserted, @var{filename} is the name of the source file which contains
23526 this function, @var{lineno} is the source line number within that file
23527 and @var{times} the number of times that the breakpoint has been hit
23528 (always 0 for -break-insert but may be greater for -break-info or -break-list
23529 which use the same output).
23531 Note: this format is open to change.
23532 @c An out-of-band breakpoint instead of part of the result?
23534 @subsubheading @value{GDBN} Command
23536 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
23537 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
23539 @subsubheading Example
23544 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
23545 fullname="/home/foo/recursive2.c,line="4",times="0"@}
23547 -break-insert -t foo
23548 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
23549 fullname="/home/foo/recursive2.c,line="11",times="0"@}
23552 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
23553 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23554 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23555 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23556 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23557 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23558 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23559 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
23560 addr="0x0001072c", func="main",file="recursive2.c",
23561 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
23562 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
23563 addr="0x00010774",func="foo",file="recursive2.c",
23564 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
23566 -break-insert -r foo.*
23567 ~int foo(int, int);
23568 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
23569 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
23573 @subheading The @code{-break-list} Command
23574 @findex -break-list
23576 @subsubheading Synopsis
23582 Displays the list of inserted breakpoints, showing the following fields:
23586 number of the breakpoint
23588 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
23590 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
23593 is the breakpoint enabled or no: @samp{y} or @samp{n}
23595 memory location at which the breakpoint is set
23597 logical location of the breakpoint, expressed by function name, file
23600 number of times the breakpoint has been hit
23603 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
23604 @code{body} field is an empty list.
23606 @subsubheading @value{GDBN} Command
23608 The corresponding @value{GDBN} command is @samp{info break}.
23610 @subsubheading Example
23615 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
23616 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23617 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23618 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23619 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23620 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23621 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23622 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
23623 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
23624 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
23625 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
23626 line="13",times="0"@}]@}
23630 Here's an example of the result when there are no breakpoints:
23635 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
23636 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23637 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23638 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23639 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23640 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23641 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23646 @subheading The @code{-break-passcount} Command
23647 @findex -break-passcount
23649 @subsubheading Synopsis
23652 -break-passcount @var{tracepoint-number} @var{passcount}
23655 Set the passcount for tracepoint @var{tracepoint-number} to
23656 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
23657 is not a tracepoint, error is emitted. This corresponds to CLI
23658 command @samp{passcount}.
23660 @subheading The @code{-break-watch} Command
23661 @findex -break-watch
23663 @subsubheading Synopsis
23666 -break-watch [ -a | -r ]
23669 Create a watchpoint. With the @samp{-a} option it will create an
23670 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
23671 read from or on a write to the memory location. With the @samp{-r}
23672 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
23673 trigger only when the memory location is accessed for reading. Without
23674 either of the options, the watchpoint created is a regular watchpoint,
23675 i.e., it will trigger when the memory location is accessed for writing.
23676 @xref{Set Watchpoints, , Setting Watchpoints}.
23678 Note that @samp{-break-list} will report a single list of watchpoints and
23679 breakpoints inserted.
23681 @subsubheading @value{GDBN} Command
23683 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
23686 @subsubheading Example
23688 Setting a watchpoint on a variable in the @code{main} function:
23693 ^done,wpt=@{number="2",exp="x"@}
23698 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
23699 value=@{old="-268439212",new="55"@},
23700 frame=@{func="main",args=[],file="recursive2.c",
23701 fullname="/home/foo/bar/recursive2.c",line="5"@}
23705 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
23706 the program execution twice: first for the variable changing value, then
23707 for the watchpoint going out of scope.
23712 ^done,wpt=@{number="5",exp="C"@}
23717 *stopped,reason="watchpoint-trigger",
23718 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
23719 frame=@{func="callee4",args=[],
23720 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23721 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
23726 *stopped,reason="watchpoint-scope",wpnum="5",
23727 frame=@{func="callee3",args=[@{name="strarg",
23728 value="0x11940 \"A string argument.\""@}],
23729 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23730 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
23734 Listing breakpoints and watchpoints, at different points in the program
23735 execution. Note that once the watchpoint goes out of scope, it is
23741 ^done,wpt=@{number="2",exp="C"@}
23744 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
23745 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23746 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23747 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23748 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23749 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23750 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23751 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
23752 addr="0x00010734",func="callee4",
23753 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23754 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
23755 bkpt=@{number="2",type="watchpoint",disp="keep",
23756 enabled="y",addr="",what="C",times="0"@}]@}
23761 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
23762 value=@{old="-276895068",new="3"@},
23763 frame=@{func="callee4",args=[],
23764 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23765 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
23768 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
23769 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23770 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23771 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23772 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23773 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23774 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23775 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
23776 addr="0x00010734",func="callee4",
23777 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23778 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
23779 bkpt=@{number="2",type="watchpoint",disp="keep",
23780 enabled="y",addr="",what="C",times="-5"@}]@}
23784 ^done,reason="watchpoint-scope",wpnum="2",
23785 frame=@{func="callee3",args=[@{name="strarg",
23786 value="0x11940 \"A string argument.\""@}],
23787 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23788 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
23791 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
23792 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
23793 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
23794 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
23795 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
23796 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
23797 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
23798 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
23799 addr="0x00010734",func="callee4",
23800 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23801 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
23806 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23807 @node GDB/MI Program Context
23808 @section @sc{gdb/mi} Program Context
23810 @subheading The @code{-exec-arguments} Command
23811 @findex -exec-arguments
23814 @subsubheading Synopsis
23817 -exec-arguments @var{args}
23820 Set the inferior program arguments, to be used in the next
23823 @subsubheading @value{GDBN} Command
23825 The corresponding @value{GDBN} command is @samp{set args}.
23827 @subsubheading Example
23831 -exec-arguments -v word
23838 @subheading The @code{-exec-show-arguments} Command
23839 @findex -exec-show-arguments
23841 @subsubheading Synopsis
23844 -exec-show-arguments
23847 Print the arguments of the program.
23849 @subsubheading @value{GDBN} Command
23851 The corresponding @value{GDBN} command is @samp{show args}.
23853 @subsubheading Example
23858 @subheading The @code{-environment-cd} Command
23859 @findex -environment-cd
23861 @subsubheading Synopsis
23864 -environment-cd @var{pathdir}
23867 Set @value{GDBN}'s working directory.
23869 @subsubheading @value{GDBN} Command
23871 The corresponding @value{GDBN} command is @samp{cd}.
23873 @subsubheading Example
23877 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
23883 @subheading The @code{-environment-directory} Command
23884 @findex -environment-directory
23886 @subsubheading Synopsis
23889 -environment-directory [ -r ] [ @var{pathdir} ]+
23892 Add directories @var{pathdir} to beginning of search path for source files.
23893 If the @samp{-r} option is used, the search path is reset to the default
23894 search path. If directories @var{pathdir} are supplied in addition to the
23895 @samp{-r} option, the search path is first reset and then addition
23897 Multiple directories may be specified, separated by blanks. Specifying
23898 multiple directories in a single command
23899 results in the directories added to the beginning of the
23900 search path in the same order they were presented in the command.
23901 If blanks are needed as
23902 part of a directory name, double-quotes should be used around
23903 the name. In the command output, the path will show up separated
23904 by the system directory-separator character. The directory-separator
23905 character must not be used
23906 in any directory name.
23907 If no directories are specified, the current search path is displayed.
23909 @subsubheading @value{GDBN} Command
23911 The corresponding @value{GDBN} command is @samp{dir}.
23913 @subsubheading Example
23917 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
23918 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
23920 -environment-directory ""
23921 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
23923 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
23924 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
23926 -environment-directory -r
23927 ^done,source-path="$cdir:$cwd"
23932 @subheading The @code{-environment-path} Command
23933 @findex -environment-path
23935 @subsubheading Synopsis
23938 -environment-path [ -r ] [ @var{pathdir} ]+
23941 Add directories @var{pathdir} to beginning of search path for object files.
23942 If the @samp{-r} option is used, the search path is reset to the original
23943 search path that existed at gdb start-up. If directories @var{pathdir} are
23944 supplied in addition to the
23945 @samp{-r} option, the search path is first reset and then addition
23947 Multiple directories may be specified, separated by blanks. Specifying
23948 multiple directories in a single command
23949 results in the directories added to the beginning of the
23950 search path in the same order they were presented in the command.
23951 If blanks are needed as
23952 part of a directory name, double-quotes should be used around
23953 the name. In the command output, the path will show up separated
23954 by the system directory-separator character. The directory-separator
23955 character must not be used
23956 in any directory name.
23957 If no directories are specified, the current path is displayed.
23960 @subsubheading @value{GDBN} Command
23962 The corresponding @value{GDBN} command is @samp{path}.
23964 @subsubheading Example
23969 ^done,path="/usr/bin"
23971 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
23972 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
23974 -environment-path -r /usr/local/bin
23975 ^done,path="/usr/local/bin:/usr/bin"
23980 @subheading The @code{-environment-pwd} Command
23981 @findex -environment-pwd
23983 @subsubheading Synopsis
23989 Show the current working directory.
23991 @subsubheading @value{GDBN} Command
23993 The corresponding @value{GDBN} command is @samp{pwd}.
23995 @subsubheading Example
24000 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
24004 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24005 @node GDB/MI Thread Commands
24006 @section @sc{gdb/mi} Thread Commands
24009 @subheading The @code{-thread-info} Command
24010 @findex -thread-info
24012 @subsubheading Synopsis
24015 -thread-info [ @var{thread-id} ]
24018 Reports information about either a specific thread, if
24019 the @var{thread-id} parameter is present, or about all
24020 threads. When printing information about all threads,
24021 also reports the current thread.
24023 @subsubheading @value{GDBN} Command
24025 The @samp{info thread} command prints the same information
24028 @subsubheading Example
24033 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
24034 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
24035 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
24036 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
24037 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}],
24038 current-thread-id="1"
24042 The @samp{state} field may have the following values:
24046 The thread is stopped. Frame information is available for stopped
24050 The thread is running. There's no frame information for running
24055 @subheading The @code{-thread-list-ids} Command
24056 @findex -thread-list-ids
24058 @subsubheading Synopsis
24064 Produces a list of the currently known @value{GDBN} thread ids. At the
24065 end of the list it also prints the total number of such threads.
24067 This command is retained for historical reasons, the
24068 @code{-thread-info} command should be used instead.
24070 @subsubheading @value{GDBN} Command
24072 Part of @samp{info threads} supplies the same information.
24074 @subsubheading Example
24079 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
24080 current-thread-id="1",number-of-threads="3"
24085 @subheading The @code{-thread-select} Command
24086 @findex -thread-select
24088 @subsubheading Synopsis
24091 -thread-select @var{threadnum}
24094 Make @var{threadnum} the current thread. It prints the number of the new
24095 current thread, and the topmost frame for that thread.
24097 This command is deprecated in favor of explicitly using the
24098 @samp{--thread} option to each command.
24100 @subsubheading @value{GDBN} Command
24102 The corresponding @value{GDBN} command is @samp{thread}.
24104 @subsubheading Example
24111 *stopped,reason="end-stepping-range",thread-id="2",line="187",
24112 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
24116 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
24117 number-of-threads="3"
24120 ^done,new-thread-id="3",
24121 frame=@{level="0",func="vprintf",
24122 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
24123 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
24127 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24128 @node GDB/MI Program Execution
24129 @section @sc{gdb/mi} Program Execution
24131 These are the asynchronous commands which generate the out-of-band
24132 record @samp{*stopped}. Currently @value{GDBN} only really executes
24133 asynchronously with remote targets and this interaction is mimicked in
24136 @subheading The @code{-exec-continue} Command
24137 @findex -exec-continue
24139 @subsubheading Synopsis
24142 -exec-continue [--reverse] [--all|--thread-group N]
24145 Resumes the execution of the inferior program, which will continue
24146 to execute until it reaches a debugger stop event. If the
24147 @samp{--reverse} option is specified, execution resumes in reverse until
24148 it reaches a stop event. Stop events may include
24151 breakpoints or watchpoints
24153 signals or exceptions
24155 the end of the process (or its beginning under @samp{--reverse})
24157 the end or beginning of a replay log if one is being used.
24159 In all-stop mode (@pxref{All-Stop
24160 Mode}), may resume only one thread, or all threads, depending on the
24161 value of the @samp{scheduler-locking} variable. If @samp{--all} is
24162 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
24163 ignored in all-stop mode. If the @samp{--thread-group} options is
24164 specified, then all threads in that thread group are resumed.
24166 @subsubheading @value{GDBN} Command
24168 The corresponding @value{GDBN} corresponding is @samp{continue}.
24170 @subsubheading Example
24177 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
24178 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
24184 @subheading The @code{-exec-finish} Command
24185 @findex -exec-finish
24187 @subsubheading Synopsis
24190 -exec-finish [--reverse]
24193 Resumes the execution of the inferior program until the current
24194 function is exited. Displays the results returned by the function.
24195 If the @samp{--reverse} option is specified, resumes the reverse
24196 execution of the inferior program until the point where current
24197 function was called.
24199 @subsubheading @value{GDBN} Command
24201 The corresponding @value{GDBN} command is @samp{finish}.
24203 @subsubheading Example
24205 Function returning @code{void}.
24212 *stopped,reason="function-finished",frame=@{func="main",args=[],
24213 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
24217 Function returning other than @code{void}. The name of the internal
24218 @value{GDBN} variable storing the result is printed, together with the
24225 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
24226 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
24227 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24228 gdb-result-var="$1",return-value="0"
24233 @subheading The @code{-exec-interrupt} Command
24234 @findex -exec-interrupt
24236 @subsubheading Synopsis
24239 -exec-interrupt [--all|--thread-group N]
24242 Interrupts the background execution of the target. Note how the token
24243 associated with the stop message is the one for the execution command
24244 that has been interrupted. The token for the interrupt itself only
24245 appears in the @samp{^done} output. If the user is trying to
24246 interrupt a non-running program, an error message will be printed.
24248 Note that when asynchronous execution is enabled, this command is
24249 asynchronous just like other execution commands. That is, first the
24250 @samp{^done} response will be printed, and the target stop will be
24251 reported after that using the @samp{*stopped} notification.
24253 In non-stop mode, only the context thread is interrupted by default.
24254 All threads (in all inferiors) will be interrupted if the
24255 @samp{--all} option is specified. If the @samp{--thread-group}
24256 option is specified, all threads in that group will be interrupted.
24258 @subsubheading @value{GDBN} Command
24260 The corresponding @value{GDBN} command is @samp{interrupt}.
24262 @subsubheading Example
24273 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
24274 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
24275 fullname="/home/foo/bar/try.c",line="13"@}
24280 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
24284 @subheading The @code{-exec-jump} Command
24287 @subsubheading Synopsis
24290 -exec-jump @var{location}
24293 Resumes execution of the inferior program at the location specified by
24294 parameter. @xref{Specify Location}, for a description of the
24295 different forms of @var{location}.
24297 @subsubheading @value{GDBN} Command
24299 The corresponding @value{GDBN} command is @samp{jump}.
24301 @subsubheading Example
24304 -exec-jump foo.c:10
24305 *running,thread-id="all"
24310 @subheading The @code{-exec-next} Command
24313 @subsubheading Synopsis
24316 -exec-next [--reverse]
24319 Resumes execution of the inferior program, stopping when the beginning
24320 of the next source line is reached.
24322 If the @samp{--reverse} option is specified, resumes reverse execution
24323 of the inferior program, stopping at the beginning of the previous
24324 source line. If you issue this command on the first line of a
24325 function, it will take you back to the caller of that function, to the
24326 source line where the function was called.
24329 @subsubheading @value{GDBN} Command
24331 The corresponding @value{GDBN} command is @samp{next}.
24333 @subsubheading Example
24339 *stopped,reason="end-stepping-range",line="8",file="hello.c"
24344 @subheading The @code{-exec-next-instruction} Command
24345 @findex -exec-next-instruction
24347 @subsubheading Synopsis
24350 -exec-next-instruction [--reverse]
24353 Executes one machine instruction. If the instruction is a function
24354 call, continues until the function returns. If the program stops at an
24355 instruction in the middle of a source line, the address will be
24358 If the @samp{--reverse} option is specified, resumes reverse execution
24359 of the inferior program, stopping at the previous instruction. If the
24360 previously executed instruction was a return from another function,
24361 it will continue to execute in reverse until the call to that function
24362 (from the current stack frame) is reached.
24364 @subsubheading @value{GDBN} Command
24366 The corresponding @value{GDBN} command is @samp{nexti}.
24368 @subsubheading Example
24372 -exec-next-instruction
24376 *stopped,reason="end-stepping-range",
24377 addr="0x000100d4",line="5",file="hello.c"
24382 @subheading The @code{-exec-return} Command
24383 @findex -exec-return
24385 @subsubheading Synopsis
24391 Makes current function return immediately. Doesn't execute the inferior.
24392 Displays the new current frame.
24394 @subsubheading @value{GDBN} Command
24396 The corresponding @value{GDBN} command is @samp{return}.
24398 @subsubheading Example
24402 200-break-insert callee4
24403 200^done,bkpt=@{number="1",addr="0x00010734",
24404 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
24409 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
24410 frame=@{func="callee4",args=[],
24411 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24412 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
24418 111^done,frame=@{level="0",func="callee3",
24419 args=[@{name="strarg",
24420 value="0x11940 \"A string argument.\""@}],
24421 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24422 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
24427 @subheading The @code{-exec-run} Command
24430 @subsubheading Synopsis
24433 -exec-run [--all | --thread-group N]
24436 Starts execution of the inferior from the beginning. The inferior
24437 executes until either a breakpoint is encountered or the program
24438 exits. In the latter case the output will include an exit code, if
24439 the program has exited exceptionally.
24441 When no option is specified, the current inferior is started. If the
24442 @samp{--thread-group} option is specified, it should refer to a thread
24443 group of type @samp{process}, and that thread group will be started.
24444 If the @samp{--all} option is specified, then all inferiors will be started.
24446 @subsubheading @value{GDBN} Command
24448 The corresponding @value{GDBN} command is @samp{run}.
24450 @subsubheading Examples
24455 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
24460 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
24461 frame=@{func="main",args=[],file="recursive2.c",
24462 fullname="/home/foo/bar/recursive2.c",line="4"@}
24467 Program exited normally:
24475 *stopped,reason="exited-normally"
24480 Program exited exceptionally:
24488 *stopped,reason="exited",exit-code="01"
24492 Another way the program can terminate is if it receives a signal such as
24493 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
24497 *stopped,reason="exited-signalled",signal-name="SIGINT",
24498 signal-meaning="Interrupt"
24502 @c @subheading -exec-signal
24505 @subheading The @code{-exec-step} Command
24508 @subsubheading Synopsis
24511 -exec-step [--reverse]
24514 Resumes execution of the inferior program, stopping when the beginning
24515 of the next source line is reached, if the next source line is not a
24516 function call. If it is, stop at the first instruction of the called
24517 function. If the @samp{--reverse} option is specified, resumes reverse
24518 execution of the inferior program, stopping at the beginning of the
24519 previously executed source line.
24521 @subsubheading @value{GDBN} Command
24523 The corresponding @value{GDBN} command is @samp{step}.
24525 @subsubheading Example
24527 Stepping into a function:
24533 *stopped,reason="end-stepping-range",
24534 frame=@{func="foo",args=[@{name="a",value="10"@},
24535 @{name="b",value="0"@}],file="recursive2.c",
24536 fullname="/home/foo/bar/recursive2.c",line="11"@}
24546 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
24551 @subheading The @code{-exec-step-instruction} Command
24552 @findex -exec-step-instruction
24554 @subsubheading Synopsis
24557 -exec-step-instruction [--reverse]
24560 Resumes the inferior which executes one machine instruction. If the
24561 @samp{--reverse} option is specified, resumes reverse execution of the
24562 inferior program, stopping at the previously executed instruction.
24563 The output, once @value{GDBN} has stopped, will vary depending on
24564 whether we have stopped in the middle of a source line or not. In the
24565 former case, the address at which the program stopped will be printed
24568 @subsubheading @value{GDBN} Command
24570 The corresponding @value{GDBN} command is @samp{stepi}.
24572 @subsubheading Example
24576 -exec-step-instruction
24580 *stopped,reason="end-stepping-range",
24581 frame=@{func="foo",args=[],file="try.c",
24582 fullname="/home/foo/bar/try.c",line="10"@}
24584 -exec-step-instruction
24588 *stopped,reason="end-stepping-range",
24589 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
24590 fullname="/home/foo/bar/try.c",line="10"@}
24595 @subheading The @code{-exec-until} Command
24596 @findex -exec-until
24598 @subsubheading Synopsis
24601 -exec-until [ @var{location} ]
24604 Executes the inferior until the @var{location} specified in the
24605 argument is reached. If there is no argument, the inferior executes
24606 until a source line greater than the current one is reached. The
24607 reason for stopping in this case will be @samp{location-reached}.
24609 @subsubheading @value{GDBN} Command
24611 The corresponding @value{GDBN} command is @samp{until}.
24613 @subsubheading Example
24617 -exec-until recursive2.c:6
24621 *stopped,reason="location-reached",frame=@{func="main",args=[],
24622 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
24627 @subheading -file-clear
24628 Is this going away????
24631 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24632 @node GDB/MI Stack Manipulation
24633 @section @sc{gdb/mi} Stack Manipulation Commands
24636 @subheading The @code{-stack-info-frame} Command
24637 @findex -stack-info-frame
24639 @subsubheading Synopsis
24645 Get info on the selected frame.
24647 @subsubheading @value{GDBN} Command
24649 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
24650 (without arguments).
24652 @subsubheading Example
24657 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
24658 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24659 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
24663 @subheading The @code{-stack-info-depth} Command
24664 @findex -stack-info-depth
24666 @subsubheading Synopsis
24669 -stack-info-depth [ @var{max-depth} ]
24672 Return the depth of the stack. If the integer argument @var{max-depth}
24673 is specified, do not count beyond @var{max-depth} frames.
24675 @subsubheading @value{GDBN} Command
24677 There's no equivalent @value{GDBN} command.
24679 @subsubheading Example
24681 For a stack with frame levels 0 through 11:
24688 -stack-info-depth 4
24691 -stack-info-depth 12
24694 -stack-info-depth 11
24697 -stack-info-depth 13
24702 @subheading The @code{-stack-list-arguments} Command
24703 @findex -stack-list-arguments
24705 @subsubheading Synopsis
24708 -stack-list-arguments @var{print-values}
24709 [ @var{low-frame} @var{high-frame} ]
24712 Display a list of the arguments for the frames between @var{low-frame}
24713 and @var{high-frame} (inclusive). If @var{low-frame} and
24714 @var{high-frame} are not provided, list the arguments for the whole
24715 call stack. If the two arguments are equal, show the single frame
24716 at the corresponding level. It is an error if @var{low-frame} is
24717 larger than the actual number of frames. On the other hand,
24718 @var{high-frame} may be larger than the actual number of frames, in
24719 which case only existing frames will be returned.
24721 If @var{print-values} is 0 or @code{--no-values}, print only the names of
24722 the variables; if it is 1 or @code{--all-values}, print also their
24723 values; and if it is 2 or @code{--simple-values}, print the name,
24724 type and value for simple data types, and the name and type for arrays,
24725 structures and unions.
24727 Use of this command to obtain arguments in a single frame is
24728 deprecated in favor of the @samp{-stack-list-variables} command.
24730 @subsubheading @value{GDBN} Command
24732 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
24733 @samp{gdb_get_args} command which partially overlaps with the
24734 functionality of @samp{-stack-list-arguments}.
24736 @subsubheading Example
24743 frame=@{level="0",addr="0x00010734",func="callee4",
24744 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24745 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
24746 frame=@{level="1",addr="0x0001076c",func="callee3",
24747 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24748 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
24749 frame=@{level="2",addr="0x0001078c",func="callee2",
24750 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24751 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
24752 frame=@{level="3",addr="0x000107b4",func="callee1",
24753 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24754 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
24755 frame=@{level="4",addr="0x000107e0",func="main",
24756 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24757 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
24759 -stack-list-arguments 0
24762 frame=@{level="0",args=[]@},
24763 frame=@{level="1",args=[name="strarg"]@},
24764 frame=@{level="2",args=[name="intarg",name="strarg"]@},
24765 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
24766 frame=@{level="4",args=[]@}]
24768 -stack-list-arguments 1
24771 frame=@{level="0",args=[]@},
24773 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
24774 frame=@{level="2",args=[
24775 @{name="intarg",value="2"@},
24776 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
24777 @{frame=@{level="3",args=[
24778 @{name="intarg",value="2"@},
24779 @{name="strarg",value="0x11940 \"A string argument.\""@},
24780 @{name="fltarg",value="3.5"@}]@},
24781 frame=@{level="4",args=[]@}]
24783 -stack-list-arguments 0 2 2
24784 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
24786 -stack-list-arguments 1 2 2
24787 ^done,stack-args=[frame=@{level="2",
24788 args=[@{name="intarg",value="2"@},
24789 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
24793 @c @subheading -stack-list-exception-handlers
24796 @subheading The @code{-stack-list-frames} Command
24797 @findex -stack-list-frames
24799 @subsubheading Synopsis
24802 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
24805 List the frames currently on the stack. For each frame it displays the
24810 The frame number, 0 being the topmost frame, i.e., the innermost function.
24812 The @code{$pc} value for that frame.
24816 File name of the source file where the function lives.
24818 Line number corresponding to the @code{$pc}.
24821 If invoked without arguments, this command prints a backtrace for the
24822 whole stack. If given two integer arguments, it shows the frames whose
24823 levels are between the two arguments (inclusive). If the two arguments
24824 are equal, it shows the single frame at the corresponding level. It is
24825 an error if @var{low-frame} is larger than the actual number of
24826 frames. On the other hand, @var{high-frame} may be larger than the
24827 actual number of frames, in which case only existing frames will be returned.
24829 @subsubheading @value{GDBN} Command
24831 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
24833 @subsubheading Example
24835 Full stack backtrace:
24841 [frame=@{level="0",addr="0x0001076c",func="foo",
24842 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
24843 frame=@{level="1",addr="0x000107a4",func="foo",
24844 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24845 frame=@{level="2",addr="0x000107a4",func="foo",
24846 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24847 frame=@{level="3",addr="0x000107a4",func="foo",
24848 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24849 frame=@{level="4",addr="0x000107a4",func="foo",
24850 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24851 frame=@{level="5",addr="0x000107a4",func="foo",
24852 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24853 frame=@{level="6",addr="0x000107a4",func="foo",
24854 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24855 frame=@{level="7",addr="0x000107a4",func="foo",
24856 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24857 frame=@{level="8",addr="0x000107a4",func="foo",
24858 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24859 frame=@{level="9",addr="0x000107a4",func="foo",
24860 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24861 frame=@{level="10",addr="0x000107a4",func="foo",
24862 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24863 frame=@{level="11",addr="0x00010738",func="main",
24864 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
24868 Show frames between @var{low_frame} and @var{high_frame}:
24872 -stack-list-frames 3 5
24874 [frame=@{level="3",addr="0x000107a4",func="foo",
24875 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24876 frame=@{level="4",addr="0x000107a4",func="foo",
24877 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24878 frame=@{level="5",addr="0x000107a4",func="foo",
24879 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
24883 Show a single frame:
24887 -stack-list-frames 3 3
24889 [frame=@{level="3",addr="0x000107a4",func="foo",
24890 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
24895 @subheading The @code{-stack-list-locals} Command
24896 @findex -stack-list-locals
24898 @subsubheading Synopsis
24901 -stack-list-locals @var{print-values}
24904 Display the local variable names for the selected frame. If
24905 @var{print-values} is 0 or @code{--no-values}, print only the names of
24906 the variables; if it is 1 or @code{--all-values}, print also their
24907 values; and if it is 2 or @code{--simple-values}, print the name,
24908 type and value for simple data types, and the name and type for arrays,
24909 structures and unions. In this last case, a frontend can immediately
24910 display the value of simple data types and create variable objects for
24911 other data types when the user wishes to explore their values in
24914 This command is deprecated in favor of the
24915 @samp{-stack-list-variables} command.
24917 @subsubheading @value{GDBN} Command
24919 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
24921 @subsubheading Example
24925 -stack-list-locals 0
24926 ^done,locals=[name="A",name="B",name="C"]
24928 -stack-list-locals --all-values
24929 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
24930 @{name="C",value="@{1, 2, 3@}"@}]
24931 -stack-list-locals --simple-values
24932 ^done,locals=[@{name="A",type="int",value="1"@},
24933 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
24937 @subheading The @code{-stack-list-variables} Command
24938 @findex -stack-list-variables
24940 @subsubheading Synopsis
24943 -stack-list-variables @var{print-values}
24946 Display the names of local variables and function arguments for the selected frame. If
24947 @var{print-values} is 0 or @code{--no-values}, print only the names of
24948 the variables; if it is 1 or @code{--all-values}, print also their
24949 values; and if it is 2 or @code{--simple-values}, print the name,
24950 type and value for simple data types, and the name and type for arrays,
24951 structures and unions.
24953 @subsubheading Example
24957 -stack-list-variables --thread 1 --frame 0 --all-values
24958 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
24963 @subheading The @code{-stack-select-frame} Command
24964 @findex -stack-select-frame
24966 @subsubheading Synopsis
24969 -stack-select-frame @var{framenum}
24972 Change the selected frame. Select a different frame @var{framenum} on
24975 This command in deprecated in favor of passing the @samp{--frame}
24976 option to every command.
24978 @subsubheading @value{GDBN} Command
24980 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
24981 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
24983 @subsubheading Example
24987 -stack-select-frame 2
24992 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24993 @node GDB/MI Variable Objects
24994 @section @sc{gdb/mi} Variable Objects
24998 @subheading Motivation for Variable Objects in @sc{gdb/mi}
25000 For the implementation of a variable debugger window (locals, watched
25001 expressions, etc.), we are proposing the adaptation of the existing code
25002 used by @code{Insight}.
25004 The two main reasons for that are:
25008 It has been proven in practice (it is already on its second generation).
25011 It will shorten development time (needless to say how important it is
25015 The original interface was designed to be used by Tcl code, so it was
25016 slightly changed so it could be used through @sc{gdb/mi}. This section
25017 describes the @sc{gdb/mi} operations that will be available and gives some
25018 hints about their use.
25020 @emph{Note}: In addition to the set of operations described here, we
25021 expect the @sc{gui} implementation of a variable window to require, at
25022 least, the following operations:
25025 @item @code{-gdb-show} @code{output-radix}
25026 @item @code{-stack-list-arguments}
25027 @item @code{-stack-list-locals}
25028 @item @code{-stack-select-frame}
25033 @subheading Introduction to Variable Objects
25035 @cindex variable objects in @sc{gdb/mi}
25037 Variable objects are "object-oriented" MI interface for examining and
25038 changing values of expressions. Unlike some other MI interfaces that
25039 work with expressions, variable objects are specifically designed for
25040 simple and efficient presentation in the frontend. A variable object
25041 is identified by string name. When a variable object is created, the
25042 frontend specifies the expression for that variable object. The
25043 expression can be a simple variable, or it can be an arbitrary complex
25044 expression, and can even involve CPU registers. After creating a
25045 variable object, the frontend can invoke other variable object
25046 operations---for example to obtain or change the value of a variable
25047 object, or to change display format.
25049 Variable objects have hierarchical tree structure. Any variable object
25050 that corresponds to a composite type, such as structure in C, has
25051 a number of child variable objects, for example corresponding to each
25052 element of a structure. A child variable object can itself have
25053 children, recursively. Recursion ends when we reach
25054 leaf variable objects, which always have built-in types. Child variable
25055 objects are created only by explicit request, so if a frontend
25056 is not interested in the children of a particular variable object, no
25057 child will be created.
25059 For a leaf variable object it is possible to obtain its value as a
25060 string, or set the value from a string. String value can be also
25061 obtained for a non-leaf variable object, but it's generally a string
25062 that only indicates the type of the object, and does not list its
25063 contents. Assignment to a non-leaf variable object is not allowed.
25065 A frontend does not need to read the values of all variable objects each time
25066 the program stops. Instead, MI provides an update command that lists all
25067 variable objects whose values has changed since the last update
25068 operation. This considerably reduces the amount of data that must
25069 be transferred to the frontend. As noted above, children variable
25070 objects are created on demand, and only leaf variable objects have a
25071 real value. As result, gdb will read target memory only for leaf
25072 variables that frontend has created.
25074 The automatic update is not always desirable. For example, a frontend
25075 might want to keep a value of some expression for future reference,
25076 and never update it. For another example, fetching memory is
25077 relatively slow for embedded targets, so a frontend might want
25078 to disable automatic update for the variables that are either not
25079 visible on the screen, or ``closed''. This is possible using so
25080 called ``frozen variable objects''. Such variable objects are never
25081 implicitly updated.
25083 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
25084 fixed variable object, the expression is parsed when the variable
25085 object is created, including associating identifiers to specific
25086 variables. The meaning of expression never changes. For a floating
25087 variable object the values of variables whose names appear in the
25088 expressions are re-evaluated every time in the context of the current
25089 frame. Consider this example:
25094 struct work_state state;
25101 If a fixed variable object for the @code{state} variable is created in
25102 this function, and we enter the recursive call, the the variable
25103 object will report the value of @code{state} in the top-level
25104 @code{do_work} invocation. On the other hand, a floating variable
25105 object will report the value of @code{state} in the current frame.
25107 If an expression specified when creating a fixed variable object
25108 refers to a local variable, the variable object becomes bound to the
25109 thread and frame in which the variable object is created. When such
25110 variable object is updated, @value{GDBN} makes sure that the
25111 thread/frame combination the variable object is bound to still exists,
25112 and re-evaluates the variable object in context of that thread/frame.
25114 The following is the complete set of @sc{gdb/mi} operations defined to
25115 access this functionality:
25117 @multitable @columnfractions .4 .6
25118 @item @strong{Operation}
25119 @tab @strong{Description}
25121 @item @code{-enable-pretty-printing}
25122 @tab enable Python-based pretty-printing
25123 @item @code{-var-create}
25124 @tab create a variable object
25125 @item @code{-var-delete}
25126 @tab delete the variable object and/or its children
25127 @item @code{-var-set-format}
25128 @tab set the display format of this variable
25129 @item @code{-var-show-format}
25130 @tab show the display format of this variable
25131 @item @code{-var-info-num-children}
25132 @tab tells how many children this object has
25133 @item @code{-var-list-children}
25134 @tab return a list of the object's children
25135 @item @code{-var-info-type}
25136 @tab show the type of this variable object
25137 @item @code{-var-info-expression}
25138 @tab print parent-relative expression that this variable object represents
25139 @item @code{-var-info-path-expression}
25140 @tab print full expression that this variable object represents
25141 @item @code{-var-show-attributes}
25142 @tab is this variable editable? does it exist here?
25143 @item @code{-var-evaluate-expression}
25144 @tab get the value of this variable
25145 @item @code{-var-assign}
25146 @tab set the value of this variable
25147 @item @code{-var-update}
25148 @tab update the variable and its children
25149 @item @code{-var-set-frozen}
25150 @tab set frozeness attribute
25151 @item @code{-var-set-update-range}
25152 @tab set range of children to display on update
25155 In the next subsection we describe each operation in detail and suggest
25156 how it can be used.
25158 @subheading Description And Use of Operations on Variable Objects
25160 @subheading The @code{-enable-pretty-printing} Command
25161 @findex -enable-pretty-printing
25164 -enable-pretty-printing
25167 @value{GDBN} allows Python-based visualizers to affect the output of the
25168 MI variable object commands. However, because there was no way to
25169 implement this in a fully backward-compatible way, a front end must
25170 request that this functionality be enabled.
25172 Once enabled, this feature cannot be disabled.
25174 Note that if Python support has not been compiled into @value{GDBN},
25175 this command will still succeed (and do nothing).
25177 This feature is currently (as of @value{GDBN} 7.0) experimental, and
25178 may work differently in future versions of @value{GDBN}.
25180 @subheading The @code{-var-create} Command
25181 @findex -var-create
25183 @subsubheading Synopsis
25186 -var-create @{@var{name} | "-"@}
25187 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
25190 This operation creates a variable object, which allows the monitoring of
25191 a variable, the result of an expression, a memory cell or a CPU
25194 The @var{name} parameter is the string by which the object can be
25195 referenced. It must be unique. If @samp{-} is specified, the varobj
25196 system will generate a string ``varNNNNNN'' automatically. It will be
25197 unique provided that one does not specify @var{name} of that format.
25198 The command fails if a duplicate name is found.
25200 The frame under which the expression should be evaluated can be
25201 specified by @var{frame-addr}. A @samp{*} indicates that the current
25202 frame should be used. A @samp{@@} indicates that a floating variable
25203 object must be created.
25205 @var{expression} is any expression valid on the current language set (must not
25206 begin with a @samp{*}), or one of the following:
25210 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
25213 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
25216 @samp{$@var{regname}} --- a CPU register name
25219 @cindex dynamic varobj
25220 A varobj's contents may be provided by a Python-based pretty-printer. In this
25221 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
25222 have slightly different semantics in some cases. If the
25223 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
25224 will never create a dynamic varobj. This ensures backward
25225 compatibility for existing clients.
25227 @subsubheading Result
25229 This operation returns attributes of the newly-created varobj. These
25234 The name of the varobj.
25237 The number of children of the varobj. This number is not necessarily
25238 reliable for a dynamic varobj. Instead, you must examine the
25239 @samp{has_more} attribute.
25242 The varobj's scalar value. For a varobj whose type is some sort of
25243 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
25244 will not be interesting.
25247 The varobj's type. This is a string representation of the type, as
25248 would be printed by the @value{GDBN} CLI.
25251 If a variable object is bound to a specific thread, then this is the
25252 thread's identifier.
25255 For a dynamic varobj, this indicates whether there appear to be any
25256 children available. For a non-dynamic varobj, this will be 0.
25259 This attribute will be present and have the value @samp{1} if the
25260 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
25261 then this attribute will not be present.
25264 A dynamic varobj can supply a display hint to the front end. The
25265 value comes directly from the Python pretty-printer object's
25266 @code{display_hint} method. @xref{Pretty Printing}.
25269 Typical output will look like this:
25272 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
25273 has_more="@var{has_more}"
25277 @subheading The @code{-var-delete} Command
25278 @findex -var-delete
25280 @subsubheading Synopsis
25283 -var-delete [ -c ] @var{name}
25286 Deletes a previously created variable object and all of its children.
25287 With the @samp{-c} option, just deletes the children.
25289 Returns an error if the object @var{name} is not found.
25292 @subheading The @code{-var-set-format} Command
25293 @findex -var-set-format
25295 @subsubheading Synopsis
25298 -var-set-format @var{name} @var{format-spec}
25301 Sets the output format for the value of the object @var{name} to be
25304 @anchor{-var-set-format}
25305 The syntax for the @var{format-spec} is as follows:
25308 @var{format-spec} @expansion{}
25309 @{binary | decimal | hexadecimal | octal | natural@}
25312 The natural format is the default format choosen automatically
25313 based on the variable type (like decimal for an @code{int}, hex
25314 for pointers, etc.).
25316 For a variable with children, the format is set only on the
25317 variable itself, and the children are not affected.
25319 @subheading The @code{-var-show-format} Command
25320 @findex -var-show-format
25322 @subsubheading Synopsis
25325 -var-show-format @var{name}
25328 Returns the format used to display the value of the object @var{name}.
25331 @var{format} @expansion{}
25336 @subheading The @code{-var-info-num-children} Command
25337 @findex -var-info-num-children
25339 @subsubheading Synopsis
25342 -var-info-num-children @var{name}
25345 Returns the number of children of a variable object @var{name}:
25351 Note that this number is not completely reliable for a dynamic varobj.
25352 It will return the current number of children, but more children may
25356 @subheading The @code{-var-list-children} Command
25357 @findex -var-list-children
25359 @subsubheading Synopsis
25362 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
25364 @anchor{-var-list-children}
25366 Return a list of the children of the specified variable object and
25367 create variable objects for them, if they do not already exist. With
25368 a single argument or if @var{print-values} has a value for of 0 or
25369 @code{--no-values}, print only the names of the variables; if
25370 @var{print-values} is 1 or @code{--all-values}, also print their
25371 values; and if it is 2 or @code{--simple-values} print the name and
25372 value for simple data types and just the name for arrays, structures
25375 @var{from} and @var{to}, if specified, indicate the range of children
25376 to report. If @var{from} or @var{to} is less than zero, the range is
25377 reset and all children will be reported. Otherwise, children starting
25378 at @var{from} (zero-based) and up to and excluding @var{to} will be
25381 If a child range is requested, it will only affect the current call to
25382 @code{-var-list-children}, but not future calls to @code{-var-update}.
25383 For this, you must instead use @code{-var-set-update-range}. The
25384 intent of this approach is to enable a front end to implement any
25385 update approach it likes; for example, scrolling a view may cause the
25386 front end to request more children with @code{-var-list-children}, and
25387 then the front end could call @code{-var-set-update-range} with a
25388 different range to ensure that future updates are restricted to just
25391 For each child the following results are returned:
25396 Name of the variable object created for this child.
25399 The expression to be shown to the user by the front end to designate this child.
25400 For example this may be the name of a structure member.
25402 For a dynamic varobj, this value cannot be used to form an
25403 expression. There is no way to do this at all with a dynamic varobj.
25405 For C/C@t{++} structures there are several pseudo children returned to
25406 designate access qualifiers. For these pseudo children @var{exp} is
25407 @samp{public}, @samp{private}, or @samp{protected}. In this case the
25408 type and value are not present.
25410 A dynamic varobj will not report the access qualifying
25411 pseudo-children, regardless of the language. This information is not
25412 available at all with a dynamic varobj.
25415 Number of children this child has. For a dynamic varobj, this will be
25419 The type of the child.
25422 If values were requested, this is the value.
25425 If this variable object is associated with a thread, this is the thread id.
25426 Otherwise this result is not present.
25429 If the variable object is frozen, this variable will be present with a value of 1.
25432 The result may have its own attributes:
25436 A dynamic varobj can supply a display hint to the front end. The
25437 value comes directly from the Python pretty-printer object's
25438 @code{display_hint} method. @xref{Pretty Printing}.
25441 This is an integer attribute which is nonzero if there are children
25442 remaining after the end of the selected range.
25445 @subsubheading Example
25449 -var-list-children n
25450 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
25451 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
25453 -var-list-children --all-values n
25454 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
25455 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
25459 @subheading The @code{-var-info-type} Command
25460 @findex -var-info-type
25462 @subsubheading Synopsis
25465 -var-info-type @var{name}
25468 Returns the type of the specified variable @var{name}. The type is
25469 returned as a string in the same format as it is output by the
25473 type=@var{typename}
25477 @subheading The @code{-var-info-expression} Command
25478 @findex -var-info-expression
25480 @subsubheading Synopsis
25483 -var-info-expression @var{name}
25486 Returns a string that is suitable for presenting this
25487 variable object in user interface. The string is generally
25488 not valid expression in the current language, and cannot be evaluated.
25490 For example, if @code{a} is an array, and variable object
25491 @code{A} was created for @code{a}, then we'll get this output:
25494 (gdb) -var-info-expression A.1
25495 ^done,lang="C",exp="1"
25499 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
25501 Note that the output of the @code{-var-list-children} command also
25502 includes those expressions, so the @code{-var-info-expression} command
25505 @subheading The @code{-var-info-path-expression} Command
25506 @findex -var-info-path-expression
25508 @subsubheading Synopsis
25511 -var-info-path-expression @var{name}
25514 Returns an expression that can be evaluated in the current
25515 context and will yield the same value that a variable object has.
25516 Compare this with the @code{-var-info-expression} command, which
25517 result can be used only for UI presentation. Typical use of
25518 the @code{-var-info-path-expression} command is creating a
25519 watchpoint from a variable object.
25521 This command is currently not valid for children of a dynamic varobj,
25522 and will give an error when invoked on one.
25524 For example, suppose @code{C} is a C@t{++} class, derived from class
25525 @code{Base}, and that the @code{Base} class has a member called
25526 @code{m_size}. Assume a variable @code{c} is has the type of
25527 @code{C} and a variable object @code{C} was created for variable
25528 @code{c}. Then, we'll get this output:
25530 (gdb) -var-info-path-expression C.Base.public.m_size
25531 ^done,path_expr=((Base)c).m_size)
25534 @subheading The @code{-var-show-attributes} Command
25535 @findex -var-show-attributes
25537 @subsubheading Synopsis
25540 -var-show-attributes @var{name}
25543 List attributes of the specified variable object @var{name}:
25546 status=@var{attr} [ ( ,@var{attr} )* ]
25550 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
25552 @subheading The @code{-var-evaluate-expression} Command
25553 @findex -var-evaluate-expression
25555 @subsubheading Synopsis
25558 -var-evaluate-expression [-f @var{format-spec}] @var{name}
25561 Evaluates the expression that is represented by the specified variable
25562 object and returns its value as a string. The format of the string
25563 can be specified with the @samp{-f} option. The possible values of
25564 this option are the same as for @code{-var-set-format}
25565 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
25566 the current display format will be used. The current display format
25567 can be changed using the @code{-var-set-format} command.
25573 Note that one must invoke @code{-var-list-children} for a variable
25574 before the value of a child variable can be evaluated.
25576 @subheading The @code{-var-assign} Command
25577 @findex -var-assign
25579 @subsubheading Synopsis
25582 -var-assign @var{name} @var{expression}
25585 Assigns the value of @var{expression} to the variable object specified
25586 by @var{name}. The object must be @samp{editable}. If the variable's
25587 value is altered by the assign, the variable will show up in any
25588 subsequent @code{-var-update} list.
25590 @subsubheading Example
25598 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
25602 @subheading The @code{-var-update} Command
25603 @findex -var-update
25605 @subsubheading Synopsis
25608 -var-update [@var{print-values}] @{@var{name} | "*"@}
25611 Reevaluate the expressions corresponding to the variable object
25612 @var{name} and all its direct and indirect children, and return the
25613 list of variable objects whose values have changed; @var{name} must
25614 be a root variable object. Here, ``changed'' means that the result of
25615 @code{-var-evaluate-expression} before and after the
25616 @code{-var-update} is different. If @samp{*} is used as the variable
25617 object names, all existing variable objects are updated, except
25618 for frozen ones (@pxref{-var-set-frozen}). The option
25619 @var{print-values} determines whether both names and values, or just
25620 names are printed. The possible values of this option are the same
25621 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
25622 recommended to use the @samp{--all-values} option, to reduce the
25623 number of MI commands needed on each program stop.
25625 With the @samp{*} parameter, if a variable object is bound to a
25626 currently running thread, it will not be updated, without any
25629 If @code{-var-set-update-range} was previously used on a varobj, then
25630 only the selected range of children will be reported.
25632 @code{-var-update} reports all the changed varobjs in a tuple named
25635 Each item in the change list is itself a tuple holding:
25639 The name of the varobj.
25642 If values were requested for this update, then this field will be
25643 present and will hold the value of the varobj.
25646 @anchor{-var-update}
25647 This field is a string which may take one of three values:
25651 The variable object's current value is valid.
25654 The variable object does not currently hold a valid value but it may
25655 hold one in the future if its associated expression comes back into
25659 The variable object no longer holds a valid value.
25660 This can occur when the executable file being debugged has changed,
25661 either through recompilation or by using the @value{GDBN} @code{file}
25662 command. The front end should normally choose to delete these variable
25666 In the future new values may be added to this list so the front should
25667 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
25670 This is only present if the varobj is still valid. If the type
25671 changed, then this will be the string @samp{true}; otherwise it will
25675 If the varobj's type changed, then this field will be present and will
25678 @item new_num_children
25679 For a dynamic varobj, if the number of children changed, or if the
25680 type changed, this will be the new number of children.
25682 The @samp{numchild} field in other varobj responses is generally not
25683 valid for a dynamic varobj -- it will show the number of children that
25684 @value{GDBN} knows about, but because dynamic varobjs lazily
25685 instantiate their children, this will not reflect the number of
25686 children which may be available.
25688 The @samp{new_num_children} attribute only reports changes to the
25689 number of children known by @value{GDBN}. This is the only way to
25690 detect whether an update has removed children (which necessarily can
25691 only happen at the end of the update range).
25694 The display hint, if any.
25697 This is an integer value, which will be 1 if there are more children
25698 available outside the varobj's update range.
25701 This attribute will be present and have the value @samp{1} if the
25702 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
25703 then this attribute will not be present.
25706 If new children were added to a dynamic varobj within the selected
25707 update range (as set by @code{-var-set-update-range}), then they will
25708 be listed in this attribute.
25711 @subsubheading Example
25718 -var-update --all-values var1
25719 ^done,changelist=[@{name="var1",value="3",in_scope="true",
25720 type_changed="false"@}]
25724 @subheading The @code{-var-set-frozen} Command
25725 @findex -var-set-frozen
25726 @anchor{-var-set-frozen}
25728 @subsubheading Synopsis
25731 -var-set-frozen @var{name} @var{flag}
25734 Set the frozenness flag on the variable object @var{name}. The
25735 @var{flag} parameter should be either @samp{1} to make the variable
25736 frozen or @samp{0} to make it unfrozen. If a variable object is
25737 frozen, then neither itself, nor any of its children, are
25738 implicitly updated by @code{-var-update} of
25739 a parent variable or by @code{-var-update *}. Only
25740 @code{-var-update} of the variable itself will update its value and
25741 values of its children. After a variable object is unfrozen, it is
25742 implicitly updated by all subsequent @code{-var-update} operations.
25743 Unfreezing a variable does not update it, only subsequent
25744 @code{-var-update} does.
25746 @subsubheading Example
25750 -var-set-frozen V 1
25755 @subheading The @code{-var-set-update-range} command
25756 @findex -var-set-update-range
25757 @anchor{-var-set-update-range}
25759 @subsubheading Synopsis
25762 -var-set-update-range @var{name} @var{from} @var{to}
25765 Set the range of children to be returned by future invocations of
25766 @code{-var-update}.
25768 @var{from} and @var{to} indicate the range of children to report. If
25769 @var{from} or @var{to} is less than zero, the range is reset and all
25770 children will be reported. Otherwise, children starting at @var{from}
25771 (zero-based) and up to and excluding @var{to} will be reported.
25773 @subsubheading Example
25777 -var-set-update-range V 1 2
25781 @subheading The @code{-var-set-visualizer} command
25782 @findex -var-set-visualizer
25783 @anchor{-var-set-visualizer}
25785 @subsubheading Synopsis
25788 -var-set-visualizer @var{name} @var{visualizer}
25791 Set a visualizer for the variable object @var{name}.
25793 @var{visualizer} is the visualizer to use. The special value
25794 @samp{None} means to disable any visualizer in use.
25796 If not @samp{None}, @var{visualizer} must be a Python expression.
25797 This expression must evaluate to a callable object which accepts a
25798 single argument. @value{GDBN} will call this object with the value of
25799 the varobj @var{name} as an argument (this is done so that the same
25800 Python pretty-printing code can be used for both the CLI and MI).
25801 When called, this object must return an object which conforms to the
25802 pretty-printing interface (@pxref{Pretty Printing}).
25804 The pre-defined function @code{gdb.default_visualizer} may be used to
25805 select a visualizer by following the built-in process
25806 (@pxref{Selecting Pretty-Printers}). This is done automatically when
25807 a varobj is created, and so ordinarily is not needed.
25809 This feature is only available if Python support is enabled. The MI
25810 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
25811 can be used to check this.
25813 @subsubheading Example
25815 Resetting the visualizer:
25819 -var-set-visualizer V None
25823 Reselecting the default (type-based) visualizer:
25827 -var-set-visualizer V gdb.default_visualizer
25831 Suppose @code{SomeClass} is a visualizer class. A lambda expression
25832 can be used to instantiate this class for a varobj:
25836 -var-set-visualizer V "lambda val: SomeClass()"
25840 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25841 @node GDB/MI Data Manipulation
25842 @section @sc{gdb/mi} Data Manipulation
25844 @cindex data manipulation, in @sc{gdb/mi}
25845 @cindex @sc{gdb/mi}, data manipulation
25846 This section describes the @sc{gdb/mi} commands that manipulate data:
25847 examine memory and registers, evaluate expressions, etc.
25849 @c REMOVED FROM THE INTERFACE.
25850 @c @subheading -data-assign
25851 @c Change the value of a program variable. Plenty of side effects.
25852 @c @subsubheading GDB Command
25854 @c @subsubheading Example
25857 @subheading The @code{-data-disassemble} Command
25858 @findex -data-disassemble
25860 @subsubheading Synopsis
25864 [ -s @var{start-addr} -e @var{end-addr} ]
25865 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
25873 @item @var{start-addr}
25874 is the beginning address (or @code{$pc})
25875 @item @var{end-addr}
25877 @item @var{filename}
25878 is the name of the file to disassemble
25879 @item @var{linenum}
25880 is the line number to disassemble around
25882 is the number of disassembly lines to be produced. If it is -1,
25883 the whole function will be disassembled, in case no @var{end-addr} is
25884 specified. If @var{end-addr} is specified as a non-zero value, and
25885 @var{lines} is lower than the number of disassembly lines between
25886 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
25887 displayed; if @var{lines} is higher than the number of lines between
25888 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
25891 is either 0 (meaning only disassembly) or 1 (meaning mixed source and
25895 @subsubheading Result
25897 The output for each instruction is composed of four fields:
25906 Note that whatever included in the instruction field, is not manipulated
25907 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
25909 @subsubheading @value{GDBN} Command
25911 There's no direct mapping from this command to the CLI.
25913 @subsubheading Example
25915 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
25919 -data-disassemble -s $pc -e "$pc + 20" -- 0
25922 @{address="0x000107c0",func-name="main",offset="4",
25923 inst="mov 2, %o0"@},
25924 @{address="0x000107c4",func-name="main",offset="8",
25925 inst="sethi %hi(0x11800), %o2"@},
25926 @{address="0x000107c8",func-name="main",offset="12",
25927 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
25928 @{address="0x000107cc",func-name="main",offset="16",
25929 inst="sethi %hi(0x11800), %o2"@},
25930 @{address="0x000107d0",func-name="main",offset="20",
25931 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
25935 Disassemble the whole @code{main} function. Line 32 is part of
25939 -data-disassemble -f basics.c -l 32 -- 0
25941 @{address="0x000107bc",func-name="main",offset="0",
25942 inst="save %sp, -112, %sp"@},
25943 @{address="0x000107c0",func-name="main",offset="4",
25944 inst="mov 2, %o0"@},
25945 @{address="0x000107c4",func-name="main",offset="8",
25946 inst="sethi %hi(0x11800), %o2"@},
25948 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
25949 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
25953 Disassemble 3 instructions from the start of @code{main}:
25957 -data-disassemble -f basics.c -l 32 -n 3 -- 0
25959 @{address="0x000107bc",func-name="main",offset="0",
25960 inst="save %sp, -112, %sp"@},
25961 @{address="0x000107c0",func-name="main",offset="4",
25962 inst="mov 2, %o0"@},
25963 @{address="0x000107c4",func-name="main",offset="8",
25964 inst="sethi %hi(0x11800), %o2"@}]
25968 Disassemble 3 instructions from the start of @code{main} in mixed mode:
25972 -data-disassemble -f basics.c -l 32 -n 3 -- 1
25974 src_and_asm_line=@{line="31",
25975 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
25976 testsuite/gdb.mi/basics.c",line_asm_insn=[
25977 @{address="0x000107bc",func-name="main",offset="0",
25978 inst="save %sp, -112, %sp"@}]@},
25979 src_and_asm_line=@{line="32",
25980 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
25981 testsuite/gdb.mi/basics.c",line_asm_insn=[
25982 @{address="0x000107c0",func-name="main",offset="4",
25983 inst="mov 2, %o0"@},
25984 @{address="0x000107c4",func-name="main",offset="8",
25985 inst="sethi %hi(0x11800), %o2"@}]@}]
25990 @subheading The @code{-data-evaluate-expression} Command
25991 @findex -data-evaluate-expression
25993 @subsubheading Synopsis
25996 -data-evaluate-expression @var{expr}
25999 Evaluate @var{expr} as an expression. The expression could contain an
26000 inferior function call. The function call will execute synchronously.
26001 If the expression contains spaces, it must be enclosed in double quotes.
26003 @subsubheading @value{GDBN} Command
26005 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
26006 @samp{call}. In @code{gdbtk} only, there's a corresponding
26007 @samp{gdb_eval} command.
26009 @subsubheading Example
26011 In the following example, the numbers that precede the commands are the
26012 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
26013 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
26017 211-data-evaluate-expression A
26020 311-data-evaluate-expression &A
26021 311^done,value="0xefffeb7c"
26023 411-data-evaluate-expression A+3
26026 511-data-evaluate-expression "A + 3"
26032 @subheading The @code{-data-list-changed-registers} Command
26033 @findex -data-list-changed-registers
26035 @subsubheading Synopsis
26038 -data-list-changed-registers
26041 Display a list of the registers that have changed.
26043 @subsubheading @value{GDBN} Command
26045 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
26046 has the corresponding command @samp{gdb_changed_register_list}.
26048 @subsubheading Example
26050 On a PPC MBX board:
26058 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
26059 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
26062 -data-list-changed-registers
26063 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
26064 "10","11","13","14","15","16","17","18","19","20","21","22","23",
26065 "24","25","26","27","28","30","31","64","65","66","67","69"]
26070 @subheading The @code{-data-list-register-names} Command
26071 @findex -data-list-register-names
26073 @subsubheading Synopsis
26076 -data-list-register-names [ ( @var{regno} )+ ]
26079 Show a list of register names for the current target. If no arguments
26080 are given, it shows a list of the names of all the registers. If
26081 integer numbers are given as arguments, it will print a list of the
26082 names of the registers corresponding to the arguments. To ensure
26083 consistency between a register name and its number, the output list may
26084 include empty register names.
26086 @subsubheading @value{GDBN} Command
26088 @value{GDBN} does not have a command which corresponds to
26089 @samp{-data-list-register-names}. In @code{gdbtk} there is a
26090 corresponding command @samp{gdb_regnames}.
26092 @subsubheading Example
26094 For the PPC MBX board:
26097 -data-list-register-names
26098 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
26099 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
26100 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
26101 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
26102 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
26103 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
26104 "", "pc","ps","cr","lr","ctr","xer"]
26106 -data-list-register-names 1 2 3
26107 ^done,register-names=["r1","r2","r3"]
26111 @subheading The @code{-data-list-register-values} Command
26112 @findex -data-list-register-values
26114 @subsubheading Synopsis
26117 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
26120 Display the registers' contents. @var{fmt} is the format according to
26121 which the registers' contents are to be returned, followed by an optional
26122 list of numbers specifying the registers to display. A missing list of
26123 numbers indicates that the contents of all the registers must be returned.
26125 Allowed formats for @var{fmt} are:
26142 @subsubheading @value{GDBN} Command
26144 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
26145 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
26147 @subsubheading Example
26149 For a PPC MBX board (note: line breaks are for readability only, they
26150 don't appear in the actual output):
26154 -data-list-register-values r 64 65
26155 ^done,register-values=[@{number="64",value="0xfe00a300"@},
26156 @{number="65",value="0x00029002"@}]
26158 -data-list-register-values x
26159 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
26160 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
26161 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
26162 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
26163 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
26164 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
26165 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
26166 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
26167 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
26168 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
26169 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
26170 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
26171 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
26172 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
26173 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
26174 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
26175 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
26176 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
26177 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
26178 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
26179 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
26180 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
26181 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
26182 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
26183 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
26184 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
26185 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
26186 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
26187 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
26188 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
26189 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
26190 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
26191 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
26192 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
26193 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
26194 @{number="69",value="0x20002b03"@}]
26199 @subheading The @code{-data-read-memory} Command
26200 @findex -data-read-memory
26202 @subsubheading Synopsis
26205 -data-read-memory [ -o @var{byte-offset} ]
26206 @var{address} @var{word-format} @var{word-size}
26207 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
26214 @item @var{address}
26215 An expression specifying the address of the first memory word to be
26216 read. Complex expressions containing embedded white space should be
26217 quoted using the C convention.
26219 @item @var{word-format}
26220 The format to be used to print the memory words. The notation is the
26221 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
26224 @item @var{word-size}
26225 The size of each memory word in bytes.
26227 @item @var{nr-rows}
26228 The number of rows in the output table.
26230 @item @var{nr-cols}
26231 The number of columns in the output table.
26234 If present, indicates that each row should include an @sc{ascii} dump. The
26235 value of @var{aschar} is used as a padding character when a byte is not a
26236 member of the printable @sc{ascii} character set (printable @sc{ascii}
26237 characters are those whose code is between 32 and 126, inclusively).
26239 @item @var{byte-offset}
26240 An offset to add to the @var{address} before fetching memory.
26243 This command displays memory contents as a table of @var{nr-rows} by
26244 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
26245 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
26246 (returned as @samp{total-bytes}). Should less than the requested number
26247 of bytes be returned by the target, the missing words are identified
26248 using @samp{N/A}. The number of bytes read from the target is returned
26249 in @samp{nr-bytes} and the starting address used to read memory in
26252 The address of the next/previous row or page is available in
26253 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
26256 @subsubheading @value{GDBN} Command
26258 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
26259 @samp{gdb_get_mem} memory read command.
26261 @subsubheading Example
26263 Read six bytes of memory starting at @code{bytes+6} but then offset by
26264 @code{-6} bytes. Format as three rows of two columns. One byte per
26265 word. Display each word in hex.
26269 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
26270 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
26271 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
26272 prev-page="0x0000138a",memory=[
26273 @{addr="0x00001390",data=["0x00","0x01"]@},
26274 @{addr="0x00001392",data=["0x02","0x03"]@},
26275 @{addr="0x00001394",data=["0x04","0x05"]@}]
26279 Read two bytes of memory starting at address @code{shorts + 64} and
26280 display as a single word formatted in decimal.
26284 5-data-read-memory shorts+64 d 2 1 1
26285 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
26286 next-row="0x00001512",prev-row="0x0000150e",
26287 next-page="0x00001512",prev-page="0x0000150e",memory=[
26288 @{addr="0x00001510",data=["128"]@}]
26292 Read thirty two bytes of memory starting at @code{bytes+16} and format
26293 as eight rows of four columns. Include a string encoding with @samp{x}
26294 used as the non-printable character.
26298 4-data-read-memory bytes+16 x 1 8 4 x
26299 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
26300 next-row="0x000013c0",prev-row="0x0000139c",
26301 next-page="0x000013c0",prev-page="0x00001380",memory=[
26302 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
26303 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
26304 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
26305 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
26306 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
26307 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
26308 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
26309 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
26313 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26314 @node GDB/MI Tracepoint Commands
26315 @section @sc{gdb/mi} Tracepoint Commands
26317 The commands defined in this section implement MI support for
26318 tracepoints. For detailed introduction, see @ref{Tracepoints}.
26320 @subheading The @code{-trace-find} Command
26321 @findex -trace-find
26323 @subsubheading Synopsis
26326 -trace-find @var{mode} [@var{parameters}@dots{}]
26329 Find a trace frame using criteria defined by @var{mode} and
26330 @var{parameters}. The following table lists permissible
26331 modes and their parameters. For details of operation, see @ref{tfind}.
26336 No parameters are required. Stops examining trace frames.
26339 An integer is required as parameter. Selects tracepoint frame with
26342 @item tracepoint-number
26343 An integer is required as parameter. Finds next
26344 trace frame that corresponds to tracepoint with the specified number.
26347 An address is required as parameter. Finds
26348 next trace frame that corresponds to any tracepoint at the specified
26351 @item pc-inside-range
26352 Two addresses are required as parameters. Finds next trace
26353 frame that corresponds to a tracepoint at an address inside the
26354 specified range. Both bounds are considered to be inside the range.
26356 @item pc-outside-range
26357 Two addresses are required as parameters. Finds
26358 next trace frame that corresponds to a tracepoint at an address outside
26359 the specified range. Both bounds are considered to be inside the range.
26362 Line specification is required as parameter. @xref{Specify Location}.
26363 Finds next trace frame that corresponds to a tracepoint at
26364 the specified location.
26368 If @samp{none} was passed as @var{mode}, the response does not
26369 have fields. Otherwise, the response may have the following fields:
26373 This field has either @samp{0} or @samp{1} as the value, depending
26374 on whether a matching tracepoint was found.
26377 The index of the found traceframe. This field is present iff
26378 the @samp{found} field has value of @samp{1}.
26381 The index of the found tracepoint. This field is present iff
26382 the @samp{found} field has value of @samp{1}.
26385 The information about the frame corresponding to the found trace
26386 frame. This field is present only if a trace frame was found.
26387 @xref{GDB/MI Frame Information}, for description of this field.
26391 @subsubheading @value{GDBN} Command
26393 The corresponding @value{GDBN} command is @samp{tfind}.
26395 @subheading -trace-define-variable
26396 @findex -trace-define-variable
26398 @subsubheading Synopsis
26401 -trace-define-variable @var{name} [ @var{value} ]
26404 Create trace variable @var{name} if it does not exist. If
26405 @var{value} is specified, sets the initial value of the specified
26406 trace variable to that value. Note that the @var{name} should start
26407 with the @samp{$} character.
26409 @subsubheading @value{GDBN} Command
26411 The corresponding @value{GDBN} command is @samp{tvariable}.
26413 @subheading -trace-list-variables
26414 @findex -trace-list-variables
26416 @subsubheading Synopsis
26419 -trace-list-variables
26422 Return a table of all defined trace variables. Each element of the
26423 table has the following fields:
26427 The name of the trace variable. This field is always present.
26430 The initial value. This is a 64-bit signed integer. This
26431 field is always present.
26434 The value the trace variable has at the moment. This is a 64-bit
26435 signed integer. This field is absent iff current value is
26436 not defined, for example if the trace was never run, or is
26441 @subsubheading @value{GDBN} Command
26443 The corresponding @value{GDBN} command is @samp{tvariables}.
26445 @subsubheading Example
26449 -trace-list-variables
26450 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
26451 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
26452 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
26453 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
26454 body=[variable=@{name="$trace_timestamp",initial="0"@}
26455 variable=@{name="$foo",initial="10",current="15"@}]@}
26459 @subheading -trace-save
26460 @findex -trace-save
26462 @subsubheading Synopsis
26465 -trace-save [-r ] @var{filename}
26468 Saves the collected trace data to @var{filename}. Without the
26469 @samp{-r} option, the data is downloaded from the target and saved
26470 in a local file. With the @samp{-r} option the target is asked
26471 to perform the save.
26473 @subsubheading @value{GDBN} Command
26475 The corresponding @value{GDBN} command is @samp{tsave}.
26478 @subheading -trace-start
26479 @findex -trace-start
26481 @subsubheading Synopsis
26487 Starts a tracing experiments. The result of this command does not
26490 @subsubheading @value{GDBN} Command
26492 The corresponding @value{GDBN} command is @samp{tstart}.
26494 @subheading -trace-status
26495 @findex -trace-status
26497 @subsubheading Synopsis
26503 Obtains the status of a tracing experiment. The result may include
26504 the following fields:
26509 May have a value of either @samp{0}, when no tracing operations are
26510 supported, @samp{1}, when all tracing operations are supported, or
26511 @samp{file} when examining trace file. In the latter case, examining
26512 of trace frame is possible but new tracing experiement cannot be
26513 started. This field is always present.
26516 May have a value of either @samp{0} or @samp{1} depending on whether
26517 tracing experiement is in progress on target. This field is present
26518 if @samp{supported} field is not @samp{0}.
26521 Report the reason why the tracing was stopped last time. This field
26522 may be absent iff tracing was never stopped on target yet. The
26523 value of @samp{request} means the tracing was stopped as result of
26524 the @code{-trace-stop} command. The value of @samp{overflow} means
26525 the tracing buffer is full. The value of @samp{disconnection} means
26526 tracing was automatically stopped when @value{GDBN} has disconnected.
26527 The value of @samp{passcount} means tracing was stopped when a
26528 tracepoint was passed a maximal number of times for that tracepoint.
26529 This field is present if @samp{supported} field is not @samp{0}.
26531 @item stopping-tracepoint
26532 The number of tracepoint whose passcount as exceeded. This field is
26533 present iff the @samp{stop-reason} field has the value of
26537 @itemx frames-created
26538 The @samp{frames} field is a count of the total number of trace frames
26539 in the trace buffer, while @samp{frames-created} is the total created
26540 during the run, including ones that were discarded, such as when a
26541 circular trace buffer filled up. Both fields are optional.
26545 These fields tell the current size of the tracing buffer and the
26546 remaining space. These fields are optional.
26549 The value of the circular trace buffer flag. @code{1} means that the
26550 trace buffer is circular and old trace frames will be discarded if
26551 necessary to make room, @code{0} means that the trace buffer is linear
26555 The value of the disconnected tracing flag. @code{1} means that
26556 tracing will continue after @value{GDBN} disconnects, @code{0} means
26557 that the trace run will stop.
26561 @subsubheading @value{GDBN} Command
26563 The corresponding @value{GDBN} command is @samp{tstatus}.
26565 @subheading -trace-stop
26566 @findex -trace-stop
26568 @subsubheading Synopsis
26574 Stops a tracing experiment. The result of this command has the same
26575 fields as @code{-trace-status}, except that the @samp{supported} and
26576 @samp{running} fields are not output.
26578 @subsubheading @value{GDBN} Command
26580 The corresponding @value{GDBN} command is @samp{tstop}.
26583 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26584 @node GDB/MI Symbol Query
26585 @section @sc{gdb/mi} Symbol Query Commands
26589 @subheading The @code{-symbol-info-address} Command
26590 @findex -symbol-info-address
26592 @subsubheading Synopsis
26595 -symbol-info-address @var{symbol}
26598 Describe where @var{symbol} is stored.
26600 @subsubheading @value{GDBN} Command
26602 The corresponding @value{GDBN} command is @samp{info address}.
26604 @subsubheading Example
26608 @subheading The @code{-symbol-info-file} Command
26609 @findex -symbol-info-file
26611 @subsubheading Synopsis
26617 Show the file for the symbol.
26619 @subsubheading @value{GDBN} Command
26621 There's no equivalent @value{GDBN} command. @code{gdbtk} has
26622 @samp{gdb_find_file}.
26624 @subsubheading Example
26628 @subheading The @code{-symbol-info-function} Command
26629 @findex -symbol-info-function
26631 @subsubheading Synopsis
26634 -symbol-info-function
26637 Show which function the symbol lives in.
26639 @subsubheading @value{GDBN} Command
26641 @samp{gdb_get_function} in @code{gdbtk}.
26643 @subsubheading Example
26647 @subheading The @code{-symbol-info-line} Command
26648 @findex -symbol-info-line
26650 @subsubheading Synopsis
26656 Show the core addresses of the code for a source line.
26658 @subsubheading @value{GDBN} Command
26660 The corresponding @value{GDBN} command is @samp{info line}.
26661 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
26663 @subsubheading Example
26667 @subheading The @code{-symbol-info-symbol} Command
26668 @findex -symbol-info-symbol
26670 @subsubheading Synopsis
26673 -symbol-info-symbol @var{addr}
26676 Describe what symbol is at location @var{addr}.
26678 @subsubheading @value{GDBN} Command
26680 The corresponding @value{GDBN} command is @samp{info symbol}.
26682 @subsubheading Example
26686 @subheading The @code{-symbol-list-functions} Command
26687 @findex -symbol-list-functions
26689 @subsubheading Synopsis
26692 -symbol-list-functions
26695 List the functions in the executable.
26697 @subsubheading @value{GDBN} Command
26699 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
26700 @samp{gdb_search} in @code{gdbtk}.
26702 @subsubheading Example
26707 @subheading The @code{-symbol-list-lines} Command
26708 @findex -symbol-list-lines
26710 @subsubheading Synopsis
26713 -symbol-list-lines @var{filename}
26716 Print the list of lines that contain code and their associated program
26717 addresses for the given source filename. The entries are sorted in
26718 ascending PC order.
26720 @subsubheading @value{GDBN} Command
26722 There is no corresponding @value{GDBN} command.
26724 @subsubheading Example
26727 -symbol-list-lines basics.c
26728 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
26734 @subheading The @code{-symbol-list-types} Command
26735 @findex -symbol-list-types
26737 @subsubheading Synopsis
26743 List all the type names.
26745 @subsubheading @value{GDBN} Command
26747 The corresponding commands are @samp{info types} in @value{GDBN},
26748 @samp{gdb_search} in @code{gdbtk}.
26750 @subsubheading Example
26754 @subheading The @code{-symbol-list-variables} Command
26755 @findex -symbol-list-variables
26757 @subsubheading Synopsis
26760 -symbol-list-variables
26763 List all the global and static variable names.
26765 @subsubheading @value{GDBN} Command
26767 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
26769 @subsubheading Example
26773 @subheading The @code{-symbol-locate} Command
26774 @findex -symbol-locate
26776 @subsubheading Synopsis
26782 @subsubheading @value{GDBN} Command
26784 @samp{gdb_loc} in @code{gdbtk}.
26786 @subsubheading Example
26790 @subheading The @code{-symbol-type} Command
26791 @findex -symbol-type
26793 @subsubheading Synopsis
26796 -symbol-type @var{variable}
26799 Show type of @var{variable}.
26801 @subsubheading @value{GDBN} Command
26803 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
26804 @samp{gdb_obj_variable}.
26806 @subsubheading Example
26811 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26812 @node GDB/MI File Commands
26813 @section @sc{gdb/mi} File Commands
26815 This section describes the GDB/MI commands to specify executable file names
26816 and to read in and obtain symbol table information.
26818 @subheading The @code{-file-exec-and-symbols} Command
26819 @findex -file-exec-and-symbols
26821 @subsubheading Synopsis
26824 -file-exec-and-symbols @var{file}
26827 Specify the executable file to be debugged. This file is the one from
26828 which the symbol table is also read. If no file is specified, the
26829 command clears the executable and symbol information. If breakpoints
26830 are set when using this command with no arguments, @value{GDBN} will produce
26831 error messages. Otherwise, no output is produced, except a completion
26834 @subsubheading @value{GDBN} Command
26836 The corresponding @value{GDBN} command is @samp{file}.
26838 @subsubheading Example
26842 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
26848 @subheading The @code{-file-exec-file} Command
26849 @findex -file-exec-file
26851 @subsubheading Synopsis
26854 -file-exec-file @var{file}
26857 Specify the executable file to be debugged. Unlike
26858 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
26859 from this file. If used without argument, @value{GDBN} clears the information
26860 about the executable file. No output is produced, except a completion
26863 @subsubheading @value{GDBN} Command
26865 The corresponding @value{GDBN} command is @samp{exec-file}.
26867 @subsubheading Example
26871 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
26878 @subheading The @code{-file-list-exec-sections} Command
26879 @findex -file-list-exec-sections
26881 @subsubheading Synopsis
26884 -file-list-exec-sections
26887 List the sections of the current executable file.
26889 @subsubheading @value{GDBN} Command
26891 The @value{GDBN} command @samp{info file} shows, among the rest, the same
26892 information as this command. @code{gdbtk} has a corresponding command
26893 @samp{gdb_load_info}.
26895 @subsubheading Example
26900 @subheading The @code{-file-list-exec-source-file} Command
26901 @findex -file-list-exec-source-file
26903 @subsubheading Synopsis
26906 -file-list-exec-source-file
26909 List the line number, the current source file, and the absolute path
26910 to the current source file for the current executable. The macro
26911 information field has a value of @samp{1} or @samp{0} depending on
26912 whether or not the file includes preprocessor macro information.
26914 @subsubheading @value{GDBN} Command
26916 The @value{GDBN} equivalent is @samp{info source}
26918 @subsubheading Example
26922 123-file-list-exec-source-file
26923 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
26928 @subheading The @code{-file-list-exec-source-files} Command
26929 @findex -file-list-exec-source-files
26931 @subsubheading Synopsis
26934 -file-list-exec-source-files
26937 List the source files for the current executable.
26939 It will always output the filename, but only when @value{GDBN} can find
26940 the absolute file name of a source file, will it output the fullname.
26942 @subsubheading @value{GDBN} Command
26944 The @value{GDBN} equivalent is @samp{info sources}.
26945 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
26947 @subsubheading Example
26950 -file-list-exec-source-files
26952 @{file=foo.c,fullname=/home/foo.c@},
26953 @{file=/home/bar.c,fullname=/home/bar.c@},
26954 @{file=gdb_could_not_find_fullpath.c@}]
26959 @subheading The @code{-file-list-shared-libraries} Command
26960 @findex -file-list-shared-libraries
26962 @subsubheading Synopsis
26965 -file-list-shared-libraries
26968 List the shared libraries in the program.
26970 @subsubheading @value{GDBN} Command
26972 The corresponding @value{GDBN} command is @samp{info shared}.
26974 @subsubheading Example
26978 @subheading The @code{-file-list-symbol-files} Command
26979 @findex -file-list-symbol-files
26981 @subsubheading Synopsis
26984 -file-list-symbol-files
26989 @subsubheading @value{GDBN} Command
26991 The corresponding @value{GDBN} command is @samp{info file} (part of it).
26993 @subsubheading Example
26998 @subheading The @code{-file-symbol-file} Command
26999 @findex -file-symbol-file
27001 @subsubheading Synopsis
27004 -file-symbol-file @var{file}
27007 Read symbol table info from the specified @var{file} argument. When
27008 used without arguments, clears @value{GDBN}'s symbol table info. No output is
27009 produced, except for a completion notification.
27011 @subsubheading @value{GDBN} Command
27013 The corresponding @value{GDBN} command is @samp{symbol-file}.
27015 @subsubheading Example
27019 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
27025 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27026 @node GDB/MI Memory Overlay Commands
27027 @section @sc{gdb/mi} Memory Overlay Commands
27029 The memory overlay commands are not implemented.
27031 @c @subheading -overlay-auto
27033 @c @subheading -overlay-list-mapping-state
27035 @c @subheading -overlay-list-overlays
27037 @c @subheading -overlay-map
27039 @c @subheading -overlay-off
27041 @c @subheading -overlay-on
27043 @c @subheading -overlay-unmap
27045 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27046 @node GDB/MI Signal Handling Commands
27047 @section @sc{gdb/mi} Signal Handling Commands
27049 Signal handling commands are not implemented.
27051 @c @subheading -signal-handle
27053 @c @subheading -signal-list-handle-actions
27055 @c @subheading -signal-list-signal-types
27059 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27060 @node GDB/MI Target Manipulation
27061 @section @sc{gdb/mi} Target Manipulation Commands
27064 @subheading The @code{-target-attach} Command
27065 @findex -target-attach
27067 @subsubheading Synopsis
27070 -target-attach @var{pid} | @var{gid} | @var{file}
27073 Attach to a process @var{pid} or a file @var{file} outside of
27074 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
27075 group, the id previously returned by
27076 @samp{-list-thread-groups --available} must be used.
27078 @subsubheading @value{GDBN} Command
27080 The corresponding @value{GDBN} command is @samp{attach}.
27082 @subsubheading Example
27086 =thread-created,id="1"
27087 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
27093 @subheading The @code{-target-compare-sections} Command
27094 @findex -target-compare-sections
27096 @subsubheading Synopsis
27099 -target-compare-sections [ @var{section} ]
27102 Compare data of section @var{section} on target to the exec file.
27103 Without the argument, all sections are compared.
27105 @subsubheading @value{GDBN} Command
27107 The @value{GDBN} equivalent is @samp{compare-sections}.
27109 @subsubheading Example
27114 @subheading The @code{-target-detach} Command
27115 @findex -target-detach
27117 @subsubheading Synopsis
27120 -target-detach [ @var{pid} | @var{gid} ]
27123 Detach from the remote target which normally resumes its execution.
27124 If either @var{pid} or @var{gid} is specified, detaches from either
27125 the specified process, or specified thread group. There's no output.
27127 @subsubheading @value{GDBN} Command
27129 The corresponding @value{GDBN} command is @samp{detach}.
27131 @subsubheading Example
27141 @subheading The @code{-target-disconnect} Command
27142 @findex -target-disconnect
27144 @subsubheading Synopsis
27150 Disconnect from the remote target. There's no output and the target is
27151 generally not resumed.
27153 @subsubheading @value{GDBN} Command
27155 The corresponding @value{GDBN} command is @samp{disconnect}.
27157 @subsubheading Example
27167 @subheading The @code{-target-download} Command
27168 @findex -target-download
27170 @subsubheading Synopsis
27176 Loads the executable onto the remote target.
27177 It prints out an update message every half second, which includes the fields:
27181 The name of the section.
27183 The size of what has been sent so far for that section.
27185 The size of the section.
27187 The total size of what was sent so far (the current and the previous sections).
27189 The size of the overall executable to download.
27193 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
27194 @sc{gdb/mi} Output Syntax}).
27196 In addition, it prints the name and size of the sections, as they are
27197 downloaded. These messages include the following fields:
27201 The name of the section.
27203 The size of the section.
27205 The size of the overall executable to download.
27209 At the end, a summary is printed.
27211 @subsubheading @value{GDBN} Command
27213 The corresponding @value{GDBN} command is @samp{load}.
27215 @subsubheading Example
27217 Note: each status message appears on a single line. Here the messages
27218 have been broken down so that they can fit onto a page.
27223 +download,@{section=".text",section-size="6668",total-size="9880"@}
27224 +download,@{section=".text",section-sent="512",section-size="6668",
27225 total-sent="512",total-size="9880"@}
27226 +download,@{section=".text",section-sent="1024",section-size="6668",
27227 total-sent="1024",total-size="9880"@}
27228 +download,@{section=".text",section-sent="1536",section-size="6668",
27229 total-sent="1536",total-size="9880"@}
27230 +download,@{section=".text",section-sent="2048",section-size="6668",
27231 total-sent="2048",total-size="9880"@}
27232 +download,@{section=".text",section-sent="2560",section-size="6668",
27233 total-sent="2560",total-size="9880"@}
27234 +download,@{section=".text",section-sent="3072",section-size="6668",
27235 total-sent="3072",total-size="9880"@}
27236 +download,@{section=".text",section-sent="3584",section-size="6668",
27237 total-sent="3584",total-size="9880"@}
27238 +download,@{section=".text",section-sent="4096",section-size="6668",
27239 total-sent="4096",total-size="9880"@}
27240 +download,@{section=".text",section-sent="4608",section-size="6668",
27241 total-sent="4608",total-size="9880"@}
27242 +download,@{section=".text",section-sent="5120",section-size="6668",
27243 total-sent="5120",total-size="9880"@}
27244 +download,@{section=".text",section-sent="5632",section-size="6668",
27245 total-sent="5632",total-size="9880"@}
27246 +download,@{section=".text",section-sent="6144",section-size="6668",
27247 total-sent="6144",total-size="9880"@}
27248 +download,@{section=".text",section-sent="6656",section-size="6668",
27249 total-sent="6656",total-size="9880"@}
27250 +download,@{section=".init",section-size="28",total-size="9880"@}
27251 +download,@{section=".fini",section-size="28",total-size="9880"@}
27252 +download,@{section=".data",section-size="3156",total-size="9880"@}
27253 +download,@{section=".data",section-sent="512",section-size="3156",
27254 total-sent="7236",total-size="9880"@}
27255 +download,@{section=".data",section-sent="1024",section-size="3156",
27256 total-sent="7748",total-size="9880"@}
27257 +download,@{section=".data",section-sent="1536",section-size="3156",
27258 total-sent="8260",total-size="9880"@}
27259 +download,@{section=".data",section-sent="2048",section-size="3156",
27260 total-sent="8772",total-size="9880"@}
27261 +download,@{section=".data",section-sent="2560",section-size="3156",
27262 total-sent="9284",total-size="9880"@}
27263 +download,@{section=".data",section-sent="3072",section-size="3156",
27264 total-sent="9796",total-size="9880"@}
27265 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
27272 @subheading The @code{-target-exec-status} Command
27273 @findex -target-exec-status
27275 @subsubheading Synopsis
27278 -target-exec-status
27281 Provide information on the state of the target (whether it is running or
27282 not, for instance).
27284 @subsubheading @value{GDBN} Command
27286 There's no equivalent @value{GDBN} command.
27288 @subsubheading Example
27292 @subheading The @code{-target-list-available-targets} Command
27293 @findex -target-list-available-targets
27295 @subsubheading Synopsis
27298 -target-list-available-targets
27301 List the possible targets to connect to.
27303 @subsubheading @value{GDBN} Command
27305 The corresponding @value{GDBN} command is @samp{help target}.
27307 @subsubheading Example
27311 @subheading The @code{-target-list-current-targets} Command
27312 @findex -target-list-current-targets
27314 @subsubheading Synopsis
27317 -target-list-current-targets
27320 Describe the current target.
27322 @subsubheading @value{GDBN} Command
27324 The corresponding information is printed by @samp{info file} (among
27327 @subsubheading Example
27331 @subheading The @code{-target-list-parameters} Command
27332 @findex -target-list-parameters
27334 @subsubheading Synopsis
27337 -target-list-parameters
27343 @subsubheading @value{GDBN} Command
27347 @subsubheading Example
27351 @subheading The @code{-target-select} Command
27352 @findex -target-select
27354 @subsubheading Synopsis
27357 -target-select @var{type} @var{parameters @dots{}}
27360 Connect @value{GDBN} to the remote target. This command takes two args:
27364 The type of target, for instance @samp{remote}, etc.
27365 @item @var{parameters}
27366 Device names, host names and the like. @xref{Target Commands, ,
27367 Commands for Managing Targets}, for more details.
27370 The output is a connection notification, followed by the address at
27371 which the target program is, in the following form:
27374 ^connected,addr="@var{address}",func="@var{function name}",
27375 args=[@var{arg list}]
27378 @subsubheading @value{GDBN} Command
27380 The corresponding @value{GDBN} command is @samp{target}.
27382 @subsubheading Example
27386 -target-select remote /dev/ttya
27387 ^connected,addr="0xfe00a300",func="??",args=[]
27391 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27392 @node GDB/MI File Transfer Commands
27393 @section @sc{gdb/mi} File Transfer Commands
27396 @subheading The @code{-target-file-put} Command
27397 @findex -target-file-put
27399 @subsubheading Synopsis
27402 -target-file-put @var{hostfile} @var{targetfile}
27405 Copy file @var{hostfile} from the host system (the machine running
27406 @value{GDBN}) to @var{targetfile} on the target system.
27408 @subsubheading @value{GDBN} Command
27410 The corresponding @value{GDBN} command is @samp{remote put}.
27412 @subsubheading Example
27416 -target-file-put localfile remotefile
27422 @subheading The @code{-target-file-get} Command
27423 @findex -target-file-get
27425 @subsubheading Synopsis
27428 -target-file-get @var{targetfile} @var{hostfile}
27431 Copy file @var{targetfile} from the target system to @var{hostfile}
27432 on the host system.
27434 @subsubheading @value{GDBN} Command
27436 The corresponding @value{GDBN} command is @samp{remote get}.
27438 @subsubheading Example
27442 -target-file-get remotefile localfile
27448 @subheading The @code{-target-file-delete} Command
27449 @findex -target-file-delete
27451 @subsubheading Synopsis
27454 -target-file-delete @var{targetfile}
27457 Delete @var{targetfile} from the target system.
27459 @subsubheading @value{GDBN} Command
27461 The corresponding @value{GDBN} command is @samp{remote delete}.
27463 @subsubheading Example
27467 -target-file-delete remotefile
27473 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27474 @node GDB/MI Miscellaneous Commands
27475 @section Miscellaneous @sc{gdb/mi} Commands
27477 @c @subheading -gdb-complete
27479 @subheading The @code{-gdb-exit} Command
27482 @subsubheading Synopsis
27488 Exit @value{GDBN} immediately.
27490 @subsubheading @value{GDBN} Command
27492 Approximately corresponds to @samp{quit}.
27494 @subsubheading Example
27504 @subheading The @code{-exec-abort} Command
27505 @findex -exec-abort
27507 @subsubheading Synopsis
27513 Kill the inferior running program.
27515 @subsubheading @value{GDBN} Command
27517 The corresponding @value{GDBN} command is @samp{kill}.
27519 @subsubheading Example
27524 @subheading The @code{-gdb-set} Command
27527 @subsubheading Synopsis
27533 Set an internal @value{GDBN} variable.
27534 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
27536 @subsubheading @value{GDBN} Command
27538 The corresponding @value{GDBN} command is @samp{set}.
27540 @subsubheading Example
27550 @subheading The @code{-gdb-show} Command
27553 @subsubheading Synopsis
27559 Show the current value of a @value{GDBN} variable.
27561 @subsubheading @value{GDBN} Command
27563 The corresponding @value{GDBN} command is @samp{show}.
27565 @subsubheading Example
27574 @c @subheading -gdb-source
27577 @subheading The @code{-gdb-version} Command
27578 @findex -gdb-version
27580 @subsubheading Synopsis
27586 Show version information for @value{GDBN}. Used mostly in testing.
27588 @subsubheading @value{GDBN} Command
27590 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
27591 default shows this information when you start an interactive session.
27593 @subsubheading Example
27595 @c This example modifies the actual output from GDB to avoid overfull
27601 ~Copyright 2000 Free Software Foundation, Inc.
27602 ~GDB is free software, covered by the GNU General Public License, and
27603 ~you are welcome to change it and/or distribute copies of it under
27604 ~ certain conditions.
27605 ~Type "show copying" to see the conditions.
27606 ~There is absolutely no warranty for GDB. Type "show warranty" for
27608 ~This GDB was configured as
27609 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
27614 @subheading The @code{-list-features} Command
27615 @findex -list-features
27617 Returns a list of particular features of the MI protocol that
27618 this version of gdb implements. A feature can be a command,
27619 or a new field in an output of some command, or even an
27620 important bugfix. While a frontend can sometimes detect presence
27621 of a feature at runtime, it is easier to perform detection at debugger
27624 The command returns a list of strings, with each string naming an
27625 available feature. Each returned string is just a name, it does not
27626 have any internal structure. The list of possible feature names
27632 (gdb) -list-features
27633 ^done,result=["feature1","feature2"]
27636 The current list of features is:
27639 @item frozen-varobjs
27640 Indicates presence of the @code{-var-set-frozen} command, as well
27641 as possible presense of the @code{frozen} field in the output
27642 of @code{-varobj-create}.
27643 @item pending-breakpoints
27644 Indicates presence of the @option{-f} option to the @code{-break-insert} command.
27646 Indicates presence of Python scripting support, Python-based
27647 pretty-printing commands, and possible presence of the
27648 @samp{display_hint} field in the output of @code{-var-list-children}
27650 Indicates presence of the @code{-thread-info} command.
27654 @subheading The @code{-list-target-features} Command
27655 @findex -list-target-features
27657 Returns a list of particular features that are supported by the
27658 target. Those features affect the permitted MI commands, but
27659 unlike the features reported by the @code{-list-features} command, the
27660 features depend on which target GDB is using at the moment. Whenever
27661 a target can change, due to commands such as @code{-target-select},
27662 @code{-target-attach} or @code{-exec-run}, the list of target features
27663 may change, and the frontend should obtain it again.
27667 (gdb) -list-features
27668 ^done,result=["async"]
27671 The current list of features is:
27675 Indicates that the target is capable of asynchronous command
27676 execution, which means that @value{GDBN} will accept further commands
27677 while the target is running.
27681 @subheading The @code{-list-thread-groups} Command
27682 @findex -list-thread-groups
27684 @subheading Synopsis
27687 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
27690 Lists thread groups (@pxref{Thread groups}). When a single thread
27691 group is passed as the argument, lists the children of that group.
27692 When several thread group are passed, lists information about those
27693 thread groups. Without any parameters, lists information about all
27694 top-level thread groups.
27696 Normally, thread groups that are being debugged are reported.
27697 With the @samp{--available} option, @value{GDBN} reports thread groups
27698 available on the target.
27700 The output of this command may have either a @samp{threads} result or
27701 a @samp{groups} result. The @samp{thread} result has a list of tuples
27702 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
27703 Information}). The @samp{groups} result has a list of tuples as value,
27704 each tuple describing a thread group. If top-level groups are
27705 requested (that is, no parameter is passed), or when several groups
27706 are passed, the output always has a @samp{groups} result. The format
27707 of the @samp{group} result is described below.
27709 To reduce the number of roundtrips it's possible to list thread groups
27710 together with their children, by passing the @samp{--recurse} option
27711 and the recursion depth. Presently, only recursion depth of 1 is
27712 permitted. If this option is present, then every reported thread group
27713 will also include its children, either as @samp{group} or
27714 @samp{threads} field.
27716 In general, any combination of option and parameters is permitted, with
27717 the following caveats:
27721 When a single thread group is passed, the output will typically
27722 be the @samp{threads} result. Because threads may not contain
27723 anything, the @samp{recurse} option will be ignored.
27726 When the @samp{--available} option is passed, limited information may
27727 be available. In particular, the list of threads of a process might
27728 be inaccessible. Further, specifying specific thread groups might
27729 not give any performance advantage over listing all thread groups.
27730 The frontend should assume that @samp{-list-thread-groups --available}
27731 is always an expensive operation and cache the results.
27735 The @samp{groups} result is a list of tuples, where each tuple may
27736 have the following fields:
27740 Identifier of the thread group. This field is always present.
27741 The identifier is an opaque string; frontends should not try to
27742 convert it to an integer, even though it might look like one.
27745 The type of the thread group. At present, only @samp{process} is a
27749 The target-specific process identifier. This field is only present
27750 for thread groups of type @samp{process} and only if the process exists.
27753 The number of children this thread group has. This field may be
27754 absent for an available thread group.
27757 This field has a list of tuples as value, each tuple describing a
27758 thread. It may be present if the @samp{--recurse} option is
27759 specified, and it's actually possible to obtain the threads.
27762 This field is a list of integers, each identifying a core that one
27763 thread of the group is running on. This field may be absent if
27764 such information is not available.
27767 The name of the executable file that corresponds to this thread group.
27768 The field is only present for thread groups of type @samp{process},
27769 and only if there is a corresponding executable file.
27773 @subheading Example
27777 -list-thread-groups
27778 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
27779 -list-thread-groups 17
27780 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
27781 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
27782 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
27783 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
27784 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
27785 -list-thread-groups --available
27786 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
27787 -list-thread-groups --available --recurse 1
27788 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
27789 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
27790 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
27791 -list-thread-groups --available --recurse 1 17 18
27792 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
27793 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
27794 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
27798 @subheading The @code{-add-inferior} Command
27799 @findex -add-inferior
27801 @subheading Synopsis
27807 Creates a new inferior (@pxref{Inferiors and Programs}). The created
27808 inferior is not associated with any executable. Such association may
27809 be established with the @samp{-file-exec-and-symbols} command
27810 (@pxref{GDB/MI File Commands}). The command response has a single
27811 field, @samp{thread-group}, whose value is the identifier of the
27812 thread group corresponding to the new inferior.
27814 @subheading Example
27819 ^done,thread-group="i3"
27822 @subheading The @code{-interpreter-exec} Command
27823 @findex -interpreter-exec
27825 @subheading Synopsis
27828 -interpreter-exec @var{interpreter} @var{command}
27830 @anchor{-interpreter-exec}
27832 Execute the specified @var{command} in the given @var{interpreter}.
27834 @subheading @value{GDBN} Command
27836 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
27838 @subheading Example
27842 -interpreter-exec console "break main"
27843 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
27844 &"During symbol reading, bad structure-type format.\n"
27845 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
27850 @subheading The @code{-inferior-tty-set} Command
27851 @findex -inferior-tty-set
27853 @subheading Synopsis
27856 -inferior-tty-set /dev/pts/1
27859 Set terminal for future runs of the program being debugged.
27861 @subheading @value{GDBN} Command
27863 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
27865 @subheading Example
27869 -inferior-tty-set /dev/pts/1
27874 @subheading The @code{-inferior-tty-show} Command
27875 @findex -inferior-tty-show
27877 @subheading Synopsis
27883 Show terminal for future runs of program being debugged.
27885 @subheading @value{GDBN} Command
27887 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
27889 @subheading Example
27893 -inferior-tty-set /dev/pts/1
27897 ^done,inferior_tty_terminal="/dev/pts/1"
27901 @subheading The @code{-enable-timings} Command
27902 @findex -enable-timings
27904 @subheading Synopsis
27907 -enable-timings [yes | no]
27910 Toggle the printing of the wallclock, user and system times for an MI
27911 command as a field in its output. This command is to help frontend
27912 developers optimize the performance of their code. No argument is
27913 equivalent to @samp{yes}.
27915 @subheading @value{GDBN} Command
27919 @subheading Example
27927 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27928 addr="0x080484ed",func="main",file="myprog.c",
27929 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
27930 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
27938 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
27939 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
27940 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
27941 fullname="/home/nickrob/myprog.c",line="73"@}
27946 @chapter @value{GDBN} Annotations
27948 This chapter describes annotations in @value{GDBN}. Annotations were
27949 designed to interface @value{GDBN} to graphical user interfaces or other
27950 similar programs which want to interact with @value{GDBN} at a
27951 relatively high level.
27953 The annotation mechanism has largely been superseded by @sc{gdb/mi}
27957 This is Edition @value{EDITION}, @value{DATE}.
27961 * Annotations Overview:: What annotations are; the general syntax.
27962 * Server Prefix:: Issuing a command without affecting user state.
27963 * Prompting:: Annotations marking @value{GDBN}'s need for input.
27964 * Errors:: Annotations for error messages.
27965 * Invalidation:: Some annotations describe things now invalid.
27966 * Annotations for Running::
27967 Whether the program is running, how it stopped, etc.
27968 * Source Annotations:: Annotations describing source code.
27971 @node Annotations Overview
27972 @section What is an Annotation?
27973 @cindex annotations
27975 Annotations start with a newline character, two @samp{control-z}
27976 characters, and the name of the annotation. If there is no additional
27977 information associated with this annotation, the name of the annotation
27978 is followed immediately by a newline. If there is additional
27979 information, the name of the annotation is followed by a space, the
27980 additional information, and a newline. The additional information
27981 cannot contain newline characters.
27983 Any output not beginning with a newline and two @samp{control-z}
27984 characters denotes literal output from @value{GDBN}. Currently there is
27985 no need for @value{GDBN} to output a newline followed by two
27986 @samp{control-z} characters, but if there was such a need, the
27987 annotations could be extended with an @samp{escape} annotation which
27988 means those three characters as output.
27990 The annotation @var{level}, which is specified using the
27991 @option{--annotate} command line option (@pxref{Mode Options}), controls
27992 how much information @value{GDBN} prints together with its prompt,
27993 values of expressions, source lines, and other types of output. Level 0
27994 is for no annotations, level 1 is for use when @value{GDBN} is run as a
27995 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
27996 for programs that control @value{GDBN}, and level 2 annotations have
27997 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
27998 Interface, annotate, GDB's Obsolete Annotations}).
28001 @kindex set annotate
28002 @item set annotate @var{level}
28003 The @value{GDBN} command @code{set annotate} sets the level of
28004 annotations to the specified @var{level}.
28006 @item show annotate
28007 @kindex show annotate
28008 Show the current annotation level.
28011 This chapter describes level 3 annotations.
28013 A simple example of starting up @value{GDBN} with annotations is:
28016 $ @kbd{gdb --annotate=3}
28018 Copyright 2003 Free Software Foundation, Inc.
28019 GDB is free software, covered by the GNU General Public License,
28020 and you are welcome to change it and/or distribute copies of it
28021 under certain conditions.
28022 Type "show copying" to see the conditions.
28023 There is absolutely no warranty for GDB. Type "show warranty"
28025 This GDB was configured as "i386-pc-linux-gnu"
28036 Here @samp{quit} is input to @value{GDBN}; the rest is output from
28037 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
28038 denotes a @samp{control-z} character) are annotations; the rest is
28039 output from @value{GDBN}.
28041 @node Server Prefix
28042 @section The Server Prefix
28043 @cindex server prefix
28045 If you prefix a command with @samp{server } then it will not affect
28046 the command history, nor will it affect @value{GDBN}'s notion of which
28047 command to repeat if @key{RET} is pressed on a line by itself. This
28048 means that commands can be run behind a user's back by a front-end in
28049 a transparent manner.
28051 The @code{server } prefix does not affect the recording of values into
28052 the value history; to print a value without recording it into the
28053 value history, use the @code{output} command instead of the
28054 @code{print} command.
28056 Using this prefix also disables confirmation requests
28057 (@pxref{confirmation requests}).
28060 @section Annotation for @value{GDBN} Input
28062 @cindex annotations for prompts
28063 When @value{GDBN} prompts for input, it annotates this fact so it is possible
28064 to know when to send output, when the output from a given command is
28067 Different kinds of input each have a different @dfn{input type}. Each
28068 input type has three annotations: a @code{pre-} annotation, which
28069 denotes the beginning of any prompt which is being output, a plain
28070 annotation, which denotes the end of the prompt, and then a @code{post-}
28071 annotation which denotes the end of any echo which may (or may not) be
28072 associated with the input. For example, the @code{prompt} input type
28073 features the following annotations:
28081 The input types are
28084 @findex pre-prompt annotation
28085 @findex prompt annotation
28086 @findex post-prompt annotation
28088 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
28090 @findex pre-commands annotation
28091 @findex commands annotation
28092 @findex post-commands annotation
28094 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
28095 command. The annotations are repeated for each command which is input.
28097 @findex pre-overload-choice annotation
28098 @findex overload-choice annotation
28099 @findex post-overload-choice annotation
28100 @item overload-choice
28101 When @value{GDBN} wants the user to select between various overloaded functions.
28103 @findex pre-query annotation
28104 @findex query annotation
28105 @findex post-query annotation
28107 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
28109 @findex pre-prompt-for-continue annotation
28110 @findex prompt-for-continue annotation
28111 @findex post-prompt-for-continue annotation
28112 @item prompt-for-continue
28113 When @value{GDBN} is asking the user to press return to continue. Note: Don't
28114 expect this to work well; instead use @code{set height 0} to disable
28115 prompting. This is because the counting of lines is buggy in the
28116 presence of annotations.
28121 @cindex annotations for errors, warnings and interrupts
28123 @findex quit annotation
28128 This annotation occurs right before @value{GDBN} responds to an interrupt.
28130 @findex error annotation
28135 This annotation occurs right before @value{GDBN} responds to an error.
28137 Quit and error annotations indicate that any annotations which @value{GDBN} was
28138 in the middle of may end abruptly. For example, if a
28139 @code{value-history-begin} annotation is followed by a @code{error}, one
28140 cannot expect to receive the matching @code{value-history-end}. One
28141 cannot expect not to receive it either, however; an error annotation
28142 does not necessarily mean that @value{GDBN} is immediately returning all the way
28145 @findex error-begin annotation
28146 A quit or error annotation may be preceded by
28152 Any output between that and the quit or error annotation is the error
28155 Warning messages are not yet annotated.
28156 @c If we want to change that, need to fix warning(), type_error(),
28157 @c range_error(), and possibly other places.
28160 @section Invalidation Notices
28162 @cindex annotations for invalidation messages
28163 The following annotations say that certain pieces of state may have
28167 @findex frames-invalid annotation
28168 @item ^Z^Zframes-invalid
28170 The frames (for example, output from the @code{backtrace} command) may
28173 @findex breakpoints-invalid annotation
28174 @item ^Z^Zbreakpoints-invalid
28176 The breakpoints may have changed. For example, the user just added or
28177 deleted a breakpoint.
28180 @node Annotations for Running
28181 @section Running the Program
28182 @cindex annotations for running programs
28184 @findex starting annotation
28185 @findex stopping annotation
28186 When the program starts executing due to a @value{GDBN} command such as
28187 @code{step} or @code{continue},
28193 is output. When the program stops,
28199 is output. Before the @code{stopped} annotation, a variety of
28200 annotations describe how the program stopped.
28203 @findex exited annotation
28204 @item ^Z^Zexited @var{exit-status}
28205 The program exited, and @var{exit-status} is the exit status (zero for
28206 successful exit, otherwise nonzero).
28208 @findex signalled annotation
28209 @findex signal-name annotation
28210 @findex signal-name-end annotation
28211 @findex signal-string annotation
28212 @findex signal-string-end annotation
28213 @item ^Z^Zsignalled
28214 The program exited with a signal. After the @code{^Z^Zsignalled}, the
28215 annotation continues:
28221 ^Z^Zsignal-name-end
28225 ^Z^Zsignal-string-end
28230 where @var{name} is the name of the signal, such as @code{SIGILL} or
28231 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
28232 as @code{Illegal Instruction} or @code{Segmentation fault}.
28233 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
28234 user's benefit and have no particular format.
28236 @findex signal annotation
28238 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
28239 just saying that the program received the signal, not that it was
28240 terminated with it.
28242 @findex breakpoint annotation
28243 @item ^Z^Zbreakpoint @var{number}
28244 The program hit breakpoint number @var{number}.
28246 @findex watchpoint annotation
28247 @item ^Z^Zwatchpoint @var{number}
28248 The program hit watchpoint number @var{number}.
28251 @node Source Annotations
28252 @section Displaying Source
28253 @cindex annotations for source display
28255 @findex source annotation
28256 The following annotation is used instead of displaying source code:
28259 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
28262 where @var{filename} is an absolute file name indicating which source
28263 file, @var{line} is the line number within that file (where 1 is the
28264 first line in the file), @var{character} is the character position
28265 within the file (where 0 is the first character in the file) (for most
28266 debug formats this will necessarily point to the beginning of a line),
28267 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
28268 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
28269 @var{addr} is the address in the target program associated with the
28270 source which is being displayed. @var{addr} is in the form @samp{0x}
28271 followed by one or more lowercase hex digits (note that this does not
28272 depend on the language).
28274 @node JIT Interface
28275 @chapter JIT Compilation Interface
28276 @cindex just-in-time compilation
28277 @cindex JIT compilation interface
28279 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
28280 interface. A JIT compiler is a program or library that generates native
28281 executable code at runtime and executes it, usually in order to achieve good
28282 performance while maintaining platform independence.
28284 Programs that use JIT compilation are normally difficult to debug because
28285 portions of their code are generated at runtime, instead of being loaded from
28286 object files, which is where @value{GDBN} normally finds the program's symbols
28287 and debug information. In order to debug programs that use JIT compilation,
28288 @value{GDBN} has an interface that allows the program to register in-memory
28289 symbol files with @value{GDBN} at runtime.
28291 If you are using @value{GDBN} to debug a program that uses this interface, then
28292 it should work transparently so long as you have not stripped the binary. If
28293 you are developing a JIT compiler, then the interface is documented in the rest
28294 of this chapter. At this time, the only known client of this interface is the
28297 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
28298 JIT compiler communicates with @value{GDBN} by writing data into a global
28299 variable and calling a fuction at a well-known symbol. When @value{GDBN}
28300 attaches, it reads a linked list of symbol files from the global variable to
28301 find existing code, and puts a breakpoint in the function so that it can find
28302 out about additional code.
28305 * Declarations:: Relevant C struct declarations
28306 * Registering Code:: Steps to register code
28307 * Unregistering Code:: Steps to unregister code
28311 @section JIT Declarations
28313 These are the relevant struct declarations that a C program should include to
28314 implement the interface:
28324 struct jit_code_entry
28326 struct jit_code_entry *next_entry;
28327 struct jit_code_entry *prev_entry;
28328 const char *symfile_addr;
28329 uint64_t symfile_size;
28332 struct jit_descriptor
28335 /* This type should be jit_actions_t, but we use uint32_t
28336 to be explicit about the bitwidth. */
28337 uint32_t action_flag;
28338 struct jit_code_entry *relevant_entry;
28339 struct jit_code_entry *first_entry;
28342 /* GDB puts a breakpoint in this function. */
28343 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
28345 /* Make sure to specify the version statically, because the
28346 debugger may check the version before we can set it. */
28347 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
28350 If the JIT is multi-threaded, then it is important that the JIT synchronize any
28351 modifications to this global data properly, which can easily be done by putting
28352 a global mutex around modifications to these structures.
28354 @node Registering Code
28355 @section Registering Code
28357 To register code with @value{GDBN}, the JIT should follow this protocol:
28361 Generate an object file in memory with symbols and other desired debug
28362 information. The file must include the virtual addresses of the sections.
28365 Create a code entry for the file, which gives the start and size of the symbol
28369 Add it to the linked list in the JIT descriptor.
28372 Point the relevant_entry field of the descriptor at the entry.
28375 Set @code{action_flag} to @code{JIT_REGISTER} and call
28376 @code{__jit_debug_register_code}.
28379 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
28380 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
28381 new code. However, the linked list must still be maintained in order to allow
28382 @value{GDBN} to attach to a running process and still find the symbol files.
28384 @node Unregistering Code
28385 @section Unregistering Code
28387 If code is freed, then the JIT should use the following protocol:
28391 Remove the code entry corresponding to the code from the linked list.
28394 Point the @code{relevant_entry} field of the descriptor at the code entry.
28397 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
28398 @code{__jit_debug_register_code}.
28401 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
28402 and the JIT will leak the memory used for the associated symbol files.
28405 @chapter Reporting Bugs in @value{GDBN}
28406 @cindex bugs in @value{GDBN}
28407 @cindex reporting bugs in @value{GDBN}
28409 Your bug reports play an essential role in making @value{GDBN} reliable.
28411 Reporting a bug may help you by bringing a solution to your problem, or it
28412 may not. But in any case the principal function of a bug report is to help
28413 the entire community by making the next version of @value{GDBN} work better. Bug
28414 reports are your contribution to the maintenance of @value{GDBN}.
28416 In order for a bug report to serve its purpose, you must include the
28417 information that enables us to fix the bug.
28420 * Bug Criteria:: Have you found a bug?
28421 * Bug Reporting:: How to report bugs
28425 @section Have You Found a Bug?
28426 @cindex bug criteria
28428 If you are not sure whether you have found a bug, here are some guidelines:
28431 @cindex fatal signal
28432 @cindex debugger crash
28433 @cindex crash of debugger
28435 If the debugger gets a fatal signal, for any input whatever, that is a
28436 @value{GDBN} bug. Reliable debuggers never crash.
28438 @cindex error on valid input
28440 If @value{GDBN} produces an error message for valid input, that is a
28441 bug. (Note that if you're cross debugging, the problem may also be
28442 somewhere in the connection to the target.)
28444 @cindex invalid input
28446 If @value{GDBN} does not produce an error message for invalid input,
28447 that is a bug. However, you should note that your idea of
28448 ``invalid input'' might be our idea of ``an extension'' or ``support
28449 for traditional practice''.
28452 If you are an experienced user of debugging tools, your suggestions
28453 for improvement of @value{GDBN} are welcome in any case.
28456 @node Bug Reporting
28457 @section How to Report Bugs
28458 @cindex bug reports
28459 @cindex @value{GDBN} bugs, reporting
28461 A number of companies and individuals offer support for @sc{gnu} products.
28462 If you obtained @value{GDBN} from a support organization, we recommend you
28463 contact that organization first.
28465 You can find contact information for many support companies and
28466 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
28468 @c should add a web page ref...
28471 @ifset BUGURL_DEFAULT
28472 In any event, we also recommend that you submit bug reports for
28473 @value{GDBN}. The preferred method is to submit them directly using
28474 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
28475 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
28478 @strong{Do not send bug reports to @samp{info-gdb}, or to
28479 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
28480 not want to receive bug reports. Those that do have arranged to receive
28483 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
28484 serves as a repeater. The mailing list and the newsgroup carry exactly
28485 the same messages. Often people think of posting bug reports to the
28486 newsgroup instead of mailing them. This appears to work, but it has one
28487 problem which can be crucial: a newsgroup posting often lacks a mail
28488 path back to the sender. Thus, if we need to ask for more information,
28489 we may be unable to reach you. For this reason, it is better to send
28490 bug reports to the mailing list.
28492 @ifclear BUGURL_DEFAULT
28493 In any event, we also recommend that you submit bug reports for
28494 @value{GDBN} to @value{BUGURL}.
28498 The fundamental principle of reporting bugs usefully is this:
28499 @strong{report all the facts}. If you are not sure whether to state a
28500 fact or leave it out, state it!
28502 Often people omit facts because they think they know what causes the
28503 problem and assume that some details do not matter. Thus, you might
28504 assume that the name of the variable you use in an example does not matter.
28505 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
28506 stray memory reference which happens to fetch from the location where that
28507 name is stored in memory; perhaps, if the name were different, the contents
28508 of that location would fool the debugger into doing the right thing despite
28509 the bug. Play it safe and give a specific, complete example. That is the
28510 easiest thing for you to do, and the most helpful.
28512 Keep in mind that the purpose of a bug report is to enable us to fix the
28513 bug. It may be that the bug has been reported previously, but neither
28514 you nor we can know that unless your bug report is complete and
28517 Sometimes people give a few sketchy facts and ask, ``Does this ring a
28518 bell?'' Those bug reports are useless, and we urge everyone to
28519 @emph{refuse to respond to them} except to chide the sender to report
28522 To enable us to fix the bug, you should include all these things:
28526 The version of @value{GDBN}. @value{GDBN} announces it if you start
28527 with no arguments; you can also print it at any time using @code{show
28530 Without this, we will not know whether there is any point in looking for
28531 the bug in the current version of @value{GDBN}.
28534 The type of machine you are using, and the operating system name and
28538 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
28539 ``@value{GCC}--2.8.1''.
28542 What compiler (and its version) was used to compile the program you are
28543 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
28544 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
28545 to get this information; for other compilers, see the documentation for
28549 The command arguments you gave the compiler to compile your example and
28550 observe the bug. For example, did you use @samp{-O}? To guarantee
28551 you will not omit something important, list them all. A copy of the
28552 Makefile (or the output from make) is sufficient.
28554 If we were to try to guess the arguments, we would probably guess wrong
28555 and then we might not encounter the bug.
28558 A complete input script, and all necessary source files, that will
28562 A description of what behavior you observe that you believe is
28563 incorrect. For example, ``It gets a fatal signal.''
28565 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
28566 will certainly notice it. But if the bug is incorrect output, we might
28567 not notice unless it is glaringly wrong. You might as well not give us
28568 a chance to make a mistake.
28570 Even if the problem you experience is a fatal signal, you should still
28571 say so explicitly. Suppose something strange is going on, such as, your
28572 copy of @value{GDBN} is out of synch, or you have encountered a bug in
28573 the C library on your system. (This has happened!) Your copy might
28574 crash and ours would not. If you told us to expect a crash, then when
28575 ours fails to crash, we would know that the bug was not happening for
28576 us. If you had not told us to expect a crash, then we would not be able
28577 to draw any conclusion from our observations.
28580 @cindex recording a session script
28581 To collect all this information, you can use a session recording program
28582 such as @command{script}, which is available on many Unix systems.
28583 Just run your @value{GDBN} session inside @command{script} and then
28584 include the @file{typescript} file with your bug report.
28586 Another way to record a @value{GDBN} session is to run @value{GDBN}
28587 inside Emacs and then save the entire buffer to a file.
28590 If you wish to suggest changes to the @value{GDBN} source, send us context
28591 diffs. If you even discuss something in the @value{GDBN} source, refer to
28592 it by context, not by line number.
28594 The line numbers in our development sources will not match those in your
28595 sources. Your line numbers would convey no useful information to us.
28599 Here are some things that are not necessary:
28603 A description of the envelope of the bug.
28605 Often people who encounter a bug spend a lot of time investigating
28606 which changes to the input file will make the bug go away and which
28607 changes will not affect it.
28609 This is often time consuming and not very useful, because the way we
28610 will find the bug is by running a single example under the debugger
28611 with breakpoints, not by pure deduction from a series of examples.
28612 We recommend that you save your time for something else.
28614 Of course, if you can find a simpler example to report @emph{instead}
28615 of the original one, that is a convenience for us. Errors in the
28616 output will be easier to spot, running under the debugger will take
28617 less time, and so on.
28619 However, simplification is not vital; if you do not want to do this,
28620 report the bug anyway and send us the entire test case you used.
28623 A patch for the bug.
28625 A patch for the bug does help us if it is a good one. But do not omit
28626 the necessary information, such as the test case, on the assumption that
28627 a patch is all we need. We might see problems with your patch and decide
28628 to fix the problem another way, or we might not understand it at all.
28630 Sometimes with a program as complicated as @value{GDBN} it is very hard to
28631 construct an example that will make the program follow a certain path
28632 through the code. If you do not send us the example, we will not be able
28633 to construct one, so we will not be able to verify that the bug is fixed.
28635 And if we cannot understand what bug you are trying to fix, or why your
28636 patch should be an improvement, we will not install it. A test case will
28637 help us to understand.
28640 A guess about what the bug is or what it depends on.
28642 Such guesses are usually wrong. Even we cannot guess right about such
28643 things without first using the debugger to find the facts.
28646 @c The readline documentation is distributed with the readline code
28647 @c and consists of the two following files:
28649 @c inc-hist.texinfo
28650 @c Use -I with makeinfo to point to the appropriate directory,
28651 @c environment var TEXINPUTS with TeX.
28652 @include rluser.texi
28653 @include inc-hist.texinfo
28656 @node Formatting Documentation
28657 @appendix Formatting Documentation
28659 @cindex @value{GDBN} reference card
28660 @cindex reference card
28661 The @value{GDBN} 4 release includes an already-formatted reference card, ready
28662 for printing with PostScript or Ghostscript, in the @file{gdb}
28663 subdirectory of the main source directory@footnote{In
28664 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
28665 release.}. If you can use PostScript or Ghostscript with your printer,
28666 you can print the reference card immediately with @file{refcard.ps}.
28668 The release also includes the source for the reference card. You
28669 can format it, using @TeX{}, by typing:
28675 The @value{GDBN} reference card is designed to print in @dfn{landscape}
28676 mode on US ``letter'' size paper;
28677 that is, on a sheet 11 inches wide by 8.5 inches
28678 high. You will need to specify this form of printing as an option to
28679 your @sc{dvi} output program.
28681 @cindex documentation
28683 All the documentation for @value{GDBN} comes as part of the machine-readable
28684 distribution. The documentation is written in Texinfo format, which is
28685 a documentation system that uses a single source file to produce both
28686 on-line information and a printed manual. You can use one of the Info
28687 formatting commands to create the on-line version of the documentation
28688 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
28690 @value{GDBN} includes an already formatted copy of the on-line Info
28691 version of this manual in the @file{gdb} subdirectory. The main Info
28692 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
28693 subordinate files matching @samp{gdb.info*} in the same directory. If
28694 necessary, you can print out these files, or read them with any editor;
28695 but they are easier to read using the @code{info} subsystem in @sc{gnu}
28696 Emacs or the standalone @code{info} program, available as part of the
28697 @sc{gnu} Texinfo distribution.
28699 If you want to format these Info files yourself, you need one of the
28700 Info formatting programs, such as @code{texinfo-format-buffer} or
28703 If you have @code{makeinfo} installed, and are in the top level
28704 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
28705 version @value{GDBVN}), you can make the Info file by typing:
28712 If you want to typeset and print copies of this manual, you need @TeX{},
28713 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
28714 Texinfo definitions file.
28716 @TeX{} is a typesetting program; it does not print files directly, but
28717 produces output files called @sc{dvi} files. To print a typeset
28718 document, you need a program to print @sc{dvi} files. If your system
28719 has @TeX{} installed, chances are it has such a program. The precise
28720 command to use depends on your system; @kbd{lpr -d} is common; another
28721 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
28722 require a file name without any extension or a @samp{.dvi} extension.
28724 @TeX{} also requires a macro definitions file called
28725 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
28726 written in Texinfo format. On its own, @TeX{} cannot either read or
28727 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
28728 and is located in the @file{gdb-@var{version-number}/texinfo}
28731 If you have @TeX{} and a @sc{dvi} printer program installed, you can
28732 typeset and print this manual. First switch to the @file{gdb}
28733 subdirectory of the main source directory (for example, to
28734 @file{gdb-@value{GDBVN}/gdb}) and type:
28740 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
28742 @node Installing GDB
28743 @appendix Installing @value{GDBN}
28744 @cindex installation
28747 * Requirements:: Requirements for building @value{GDBN}
28748 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
28749 * Separate Objdir:: Compiling @value{GDBN} in another directory
28750 * Config Names:: Specifying names for hosts and targets
28751 * Configure Options:: Summary of options for configure
28752 * System-wide configuration:: Having a system-wide init file
28756 @section Requirements for Building @value{GDBN}
28757 @cindex building @value{GDBN}, requirements for
28759 Building @value{GDBN} requires various tools and packages to be available.
28760 Other packages will be used only if they are found.
28762 @heading Tools/Packages Necessary for Building @value{GDBN}
28764 @item ISO C90 compiler
28765 @value{GDBN} is written in ISO C90. It should be buildable with any
28766 working C90 compiler, e.g.@: GCC.
28770 @heading Tools/Packages Optional for Building @value{GDBN}
28774 @value{GDBN} can use the Expat XML parsing library. This library may be
28775 included with your operating system distribution; if it is not, you
28776 can get the latest version from @url{http://expat.sourceforge.net}.
28777 The @file{configure} script will search for this library in several
28778 standard locations; if it is installed in an unusual path, you can
28779 use the @option{--with-libexpat-prefix} option to specify its location.
28785 Remote protocol memory maps (@pxref{Memory Map Format})
28787 Target descriptions (@pxref{Target Descriptions})
28789 Remote shared library lists (@pxref{Library List Format})
28791 MS-Windows shared libraries (@pxref{Shared Libraries})
28795 @cindex compressed debug sections
28796 @value{GDBN} will use the @samp{zlib} library, if available, to read
28797 compressed debug sections. Some linkers, such as GNU gold, are capable
28798 of producing binaries with compressed debug sections. If @value{GDBN}
28799 is compiled with @samp{zlib}, it will be able to read the debug
28800 information in such binaries.
28802 The @samp{zlib} library is likely included with your operating system
28803 distribution; if it is not, you can get the latest version from
28804 @url{http://zlib.net}.
28807 @value{GDBN}'s features related to character sets (@pxref{Character
28808 Sets}) require a functioning @code{iconv} implementation. If you are
28809 on a GNU system, then this is provided by the GNU C Library. Some
28810 other systems also provide a working @code{iconv}.
28812 On systems with @code{iconv}, you can install GNU Libiconv. If you
28813 have previously installed Libiconv, you can use the
28814 @option{--with-libiconv-prefix} option to configure.
28816 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
28817 arrange to build Libiconv if a directory named @file{libiconv} appears
28818 in the top-most source directory. If Libiconv is built this way, and
28819 if the operating system does not provide a suitable @code{iconv}
28820 implementation, then the just-built library will automatically be used
28821 by @value{GDBN}. One easy way to set this up is to download GNU
28822 Libiconv, unpack it, and then rename the directory holding the
28823 Libiconv source code to @samp{libiconv}.
28826 @node Running Configure
28827 @section Invoking the @value{GDBN} @file{configure} Script
28828 @cindex configuring @value{GDBN}
28829 @value{GDBN} comes with a @file{configure} script that automates the process
28830 of preparing @value{GDBN} for installation; you can then use @code{make} to
28831 build the @code{gdb} program.
28833 @c irrelevant in info file; it's as current as the code it lives with.
28834 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
28835 look at the @file{README} file in the sources; we may have improved the
28836 installation procedures since publishing this manual.}
28839 The @value{GDBN} distribution includes all the source code you need for
28840 @value{GDBN} in a single directory, whose name is usually composed by
28841 appending the version number to @samp{gdb}.
28843 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
28844 @file{gdb-@value{GDBVN}} directory. That directory contains:
28847 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
28848 script for configuring @value{GDBN} and all its supporting libraries
28850 @item gdb-@value{GDBVN}/gdb
28851 the source specific to @value{GDBN} itself
28853 @item gdb-@value{GDBVN}/bfd
28854 source for the Binary File Descriptor library
28856 @item gdb-@value{GDBVN}/include
28857 @sc{gnu} include files
28859 @item gdb-@value{GDBVN}/libiberty
28860 source for the @samp{-liberty} free software library
28862 @item gdb-@value{GDBVN}/opcodes
28863 source for the library of opcode tables and disassemblers
28865 @item gdb-@value{GDBVN}/readline
28866 source for the @sc{gnu} command-line interface
28868 @item gdb-@value{GDBVN}/glob
28869 source for the @sc{gnu} filename pattern-matching subroutine
28871 @item gdb-@value{GDBVN}/mmalloc
28872 source for the @sc{gnu} memory-mapped malloc package
28875 The simplest way to configure and build @value{GDBN} is to run @file{configure}
28876 from the @file{gdb-@var{version-number}} source directory, which in
28877 this example is the @file{gdb-@value{GDBVN}} directory.
28879 First switch to the @file{gdb-@var{version-number}} source directory
28880 if you are not already in it; then run @file{configure}. Pass the
28881 identifier for the platform on which @value{GDBN} will run as an
28887 cd gdb-@value{GDBVN}
28888 ./configure @var{host}
28893 where @var{host} is an identifier such as @samp{sun4} or
28894 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
28895 (You can often leave off @var{host}; @file{configure} tries to guess the
28896 correct value by examining your system.)
28898 Running @samp{configure @var{host}} and then running @code{make} builds the
28899 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
28900 libraries, then @code{gdb} itself. The configured source files, and the
28901 binaries, are left in the corresponding source directories.
28904 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
28905 system does not recognize this automatically when you run a different
28906 shell, you may need to run @code{sh} on it explicitly:
28909 sh configure @var{host}
28912 If you run @file{configure} from a directory that contains source
28913 directories for multiple libraries or programs, such as the
28914 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
28916 creates configuration files for every directory level underneath (unless
28917 you tell it not to, with the @samp{--norecursion} option).
28919 You should run the @file{configure} script from the top directory in the
28920 source tree, the @file{gdb-@var{version-number}} directory. If you run
28921 @file{configure} from one of the subdirectories, you will configure only
28922 that subdirectory. That is usually not what you want. In particular,
28923 if you run the first @file{configure} from the @file{gdb} subdirectory
28924 of the @file{gdb-@var{version-number}} directory, you will omit the
28925 configuration of @file{bfd}, @file{readline}, and other sibling
28926 directories of the @file{gdb} subdirectory. This leads to build errors
28927 about missing include files such as @file{bfd/bfd.h}.
28929 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
28930 However, you should make sure that the shell on your path (named by
28931 the @samp{SHELL} environment variable) is publicly readable. Remember
28932 that @value{GDBN} uses the shell to start your program---some systems refuse to
28933 let @value{GDBN} debug child processes whose programs are not readable.
28935 @node Separate Objdir
28936 @section Compiling @value{GDBN} in Another Directory
28938 If you want to run @value{GDBN} versions for several host or target machines,
28939 you need a different @code{gdb} compiled for each combination of
28940 host and target. @file{configure} is designed to make this easy by
28941 allowing you to generate each configuration in a separate subdirectory,
28942 rather than in the source directory. If your @code{make} program
28943 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
28944 @code{make} in each of these directories builds the @code{gdb}
28945 program specified there.
28947 To build @code{gdb} in a separate directory, run @file{configure}
28948 with the @samp{--srcdir} option to specify where to find the source.
28949 (You also need to specify a path to find @file{configure}
28950 itself from your working directory. If the path to @file{configure}
28951 would be the same as the argument to @samp{--srcdir}, you can leave out
28952 the @samp{--srcdir} option; it is assumed.)
28954 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
28955 separate directory for a Sun 4 like this:
28959 cd gdb-@value{GDBVN}
28962 ../gdb-@value{GDBVN}/configure sun4
28967 When @file{configure} builds a configuration using a remote source
28968 directory, it creates a tree for the binaries with the same structure
28969 (and using the same names) as the tree under the source directory. In
28970 the example, you'd find the Sun 4 library @file{libiberty.a} in the
28971 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
28972 @file{gdb-sun4/gdb}.
28974 Make sure that your path to the @file{configure} script has just one
28975 instance of @file{gdb} in it. If your path to @file{configure} looks
28976 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
28977 one subdirectory of @value{GDBN}, not the whole package. This leads to
28978 build errors about missing include files such as @file{bfd/bfd.h}.
28980 One popular reason to build several @value{GDBN} configurations in separate
28981 directories is to configure @value{GDBN} for cross-compiling (where
28982 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
28983 programs that run on another machine---the @dfn{target}).
28984 You specify a cross-debugging target by
28985 giving the @samp{--target=@var{target}} option to @file{configure}.
28987 When you run @code{make} to build a program or library, you must run
28988 it in a configured directory---whatever directory you were in when you
28989 called @file{configure} (or one of its subdirectories).
28991 The @code{Makefile} that @file{configure} generates in each source
28992 directory also runs recursively. If you type @code{make} in a source
28993 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
28994 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
28995 will build all the required libraries, and then build GDB.
28997 When you have multiple hosts or targets configured in separate
28998 directories, you can run @code{make} on them in parallel (for example,
28999 if they are NFS-mounted on each of the hosts); they will not interfere
29003 @section Specifying Names for Hosts and Targets
29005 The specifications used for hosts and targets in the @file{configure}
29006 script are based on a three-part naming scheme, but some short predefined
29007 aliases are also supported. The full naming scheme encodes three pieces
29008 of information in the following pattern:
29011 @var{architecture}-@var{vendor}-@var{os}
29014 For example, you can use the alias @code{sun4} as a @var{host} argument,
29015 or as the value for @var{target} in a @code{--target=@var{target}}
29016 option. The equivalent full name is @samp{sparc-sun-sunos4}.
29018 The @file{configure} script accompanying @value{GDBN} does not provide
29019 any query facility to list all supported host and target names or
29020 aliases. @file{configure} calls the Bourne shell script
29021 @code{config.sub} to map abbreviations to full names; you can read the
29022 script, if you wish, or you can use it to test your guesses on
29023 abbreviations---for example:
29026 % sh config.sub i386-linux
29028 % sh config.sub alpha-linux
29029 alpha-unknown-linux-gnu
29030 % sh config.sub hp9k700
29032 % sh config.sub sun4
29033 sparc-sun-sunos4.1.1
29034 % sh config.sub sun3
29035 m68k-sun-sunos4.1.1
29036 % sh config.sub i986v
29037 Invalid configuration `i986v': machine `i986v' not recognized
29041 @code{config.sub} is also distributed in the @value{GDBN} source
29042 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
29044 @node Configure Options
29045 @section @file{configure} Options
29047 Here is a summary of the @file{configure} options and arguments that
29048 are most often useful for building @value{GDBN}. @file{configure} also has
29049 several other options not listed here. @inforef{What Configure
29050 Does,,configure.info}, for a full explanation of @file{configure}.
29053 configure @r{[}--help@r{]}
29054 @r{[}--prefix=@var{dir}@r{]}
29055 @r{[}--exec-prefix=@var{dir}@r{]}
29056 @r{[}--srcdir=@var{dirname}@r{]}
29057 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
29058 @r{[}--target=@var{target}@r{]}
29063 You may introduce options with a single @samp{-} rather than
29064 @samp{--} if you prefer; but you may abbreviate option names if you use
29069 Display a quick summary of how to invoke @file{configure}.
29071 @item --prefix=@var{dir}
29072 Configure the source to install programs and files under directory
29075 @item --exec-prefix=@var{dir}
29076 Configure the source to install programs under directory
29079 @c avoid splitting the warning from the explanation:
29081 @item --srcdir=@var{dirname}
29082 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
29083 @code{make} that implements the @code{VPATH} feature.}@*
29084 Use this option to make configurations in directories separate from the
29085 @value{GDBN} source directories. Among other things, you can use this to
29086 build (or maintain) several configurations simultaneously, in separate
29087 directories. @file{configure} writes configuration-specific files in
29088 the current directory, but arranges for them to use the source in the
29089 directory @var{dirname}. @file{configure} creates directories under
29090 the working directory in parallel to the source directories below
29093 @item --norecursion
29094 Configure only the directory level where @file{configure} is executed; do not
29095 propagate configuration to subdirectories.
29097 @item --target=@var{target}
29098 Configure @value{GDBN} for cross-debugging programs running on the specified
29099 @var{target}. Without this option, @value{GDBN} is configured to debug
29100 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
29102 There is no convenient way to generate a list of all available targets.
29104 @item @var{host} @dots{}
29105 Configure @value{GDBN} to run on the specified @var{host}.
29107 There is no convenient way to generate a list of all available hosts.
29110 There are many other options available as well, but they are generally
29111 needed for special purposes only.
29113 @node System-wide configuration
29114 @section System-wide configuration and settings
29115 @cindex system-wide init file
29117 @value{GDBN} can be configured to have a system-wide init file;
29118 this file will be read and executed at startup (@pxref{Startup, , What
29119 @value{GDBN} does during startup}).
29121 Here is the corresponding configure option:
29124 @item --with-system-gdbinit=@var{file}
29125 Specify that the default location of the system-wide init file is
29129 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
29130 it may be subject to relocation. Two possible cases:
29134 If the default location of this init file contains @file{$prefix},
29135 it will be subject to relocation. Suppose that the configure options
29136 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
29137 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
29138 init file is looked for as @file{$install/etc/gdbinit} instead of
29139 @file{$prefix/etc/gdbinit}.
29142 By contrast, if the default location does not contain the prefix,
29143 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
29144 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
29145 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
29146 wherever @value{GDBN} is installed.
29149 @node Maintenance Commands
29150 @appendix Maintenance Commands
29151 @cindex maintenance commands
29152 @cindex internal commands
29154 In addition to commands intended for @value{GDBN} users, @value{GDBN}
29155 includes a number of commands intended for @value{GDBN} developers,
29156 that are not documented elsewhere in this manual. These commands are
29157 provided here for reference. (For commands that turn on debugging
29158 messages, see @ref{Debugging Output}.)
29161 @kindex maint agent
29162 @kindex maint agent-eval
29163 @item maint agent @var{expression}
29164 @itemx maint agent-eval @var{expression}
29165 Translate the given @var{expression} into remote agent bytecodes.
29166 This command is useful for debugging the Agent Expression mechanism
29167 (@pxref{Agent Expressions}). The @samp{agent} version produces an
29168 expression useful for data collection, such as by tracepoints, while
29169 @samp{maint agent-eval} produces an expression that evaluates directly
29170 to a result. For instance, a collection expression for @code{globa +
29171 globb} will include bytecodes to record four bytes of memory at each
29172 of the addresses of @code{globa} and @code{globb}, while discarding
29173 the result of the addition, while an evaluation expression will do the
29174 addition and return the sum.
29176 @kindex maint info breakpoints
29177 @item @anchor{maint info breakpoints}maint info breakpoints
29178 Using the same format as @samp{info breakpoints}, display both the
29179 breakpoints you've set explicitly, and those @value{GDBN} is using for
29180 internal purposes. Internal breakpoints are shown with negative
29181 breakpoint numbers. The type column identifies what kind of breakpoint
29186 Normal, explicitly set breakpoint.
29189 Normal, explicitly set watchpoint.
29192 Internal breakpoint, used to handle correctly stepping through
29193 @code{longjmp} calls.
29195 @item longjmp resume
29196 Internal breakpoint at the target of a @code{longjmp}.
29199 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
29202 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
29205 Shared library events.
29209 @kindex set displaced-stepping
29210 @kindex show displaced-stepping
29211 @cindex displaced stepping support
29212 @cindex out-of-line single-stepping
29213 @item set displaced-stepping
29214 @itemx show displaced-stepping
29215 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
29216 if the target supports it. Displaced stepping is a way to single-step
29217 over breakpoints without removing them from the inferior, by executing
29218 an out-of-line copy of the instruction that was originally at the
29219 breakpoint location. It is also known as out-of-line single-stepping.
29222 @item set displaced-stepping on
29223 If the target architecture supports it, @value{GDBN} will use
29224 displaced stepping to step over breakpoints.
29226 @item set displaced-stepping off
29227 @value{GDBN} will not use displaced stepping to step over breakpoints,
29228 even if such is supported by the target architecture.
29230 @cindex non-stop mode, and @samp{set displaced-stepping}
29231 @item set displaced-stepping auto
29232 This is the default mode. @value{GDBN} will use displaced stepping
29233 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
29234 architecture supports displaced stepping.
29237 @kindex maint check-symtabs
29238 @item maint check-symtabs
29239 Check the consistency of psymtabs and symtabs.
29241 @kindex maint cplus first_component
29242 @item maint cplus first_component @var{name}
29243 Print the first C@t{++} class/namespace component of @var{name}.
29245 @kindex maint cplus namespace
29246 @item maint cplus namespace
29247 Print the list of possible C@t{++} namespaces.
29249 @kindex maint demangle
29250 @item maint demangle @var{name}
29251 Demangle a C@t{++} or Objective-C mangled @var{name}.
29253 @kindex maint deprecate
29254 @kindex maint undeprecate
29255 @cindex deprecated commands
29256 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
29257 @itemx maint undeprecate @var{command}
29258 Deprecate or undeprecate the named @var{command}. Deprecated commands
29259 cause @value{GDBN} to issue a warning when you use them. The optional
29260 argument @var{replacement} says which newer command should be used in
29261 favor of the deprecated one; if it is given, @value{GDBN} will mention
29262 the replacement as part of the warning.
29264 @kindex maint dump-me
29265 @item maint dump-me
29266 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
29267 Cause a fatal signal in the debugger and force it to dump its core.
29268 This is supported only on systems which support aborting a program
29269 with the @code{SIGQUIT} signal.
29271 @kindex maint internal-error
29272 @kindex maint internal-warning
29273 @item maint internal-error @r{[}@var{message-text}@r{]}
29274 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
29275 Cause @value{GDBN} to call the internal function @code{internal_error}
29276 or @code{internal_warning} and hence behave as though an internal error
29277 or internal warning has been detected. In addition to reporting the
29278 internal problem, these functions give the user the opportunity to
29279 either quit @value{GDBN} or create a core file of the current
29280 @value{GDBN} session.
29282 These commands take an optional parameter @var{message-text} that is
29283 used as the text of the error or warning message.
29285 Here's an example of using @code{internal-error}:
29288 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
29289 @dots{}/maint.c:121: internal-error: testing, 1, 2
29290 A problem internal to GDB has been detected. Further
29291 debugging may prove unreliable.
29292 Quit this debugging session? (y or n) @kbd{n}
29293 Create a core file? (y or n) @kbd{n}
29297 @cindex @value{GDBN} internal error
29298 @cindex internal errors, control of @value{GDBN} behavior
29300 @kindex maint set internal-error
29301 @kindex maint show internal-error
29302 @kindex maint set internal-warning
29303 @kindex maint show internal-warning
29304 @item maint set internal-error @var{action} [ask|yes|no]
29305 @itemx maint show internal-error @var{action}
29306 @itemx maint set internal-warning @var{action} [ask|yes|no]
29307 @itemx maint show internal-warning @var{action}
29308 When @value{GDBN} reports an internal problem (error or warning) it
29309 gives the user the opportunity to both quit @value{GDBN} and create a
29310 core file of the current @value{GDBN} session. These commands let you
29311 override the default behaviour for each particular @var{action},
29312 described in the table below.
29316 You can specify that @value{GDBN} should always (yes) or never (no)
29317 quit. The default is to ask the user what to do.
29320 You can specify that @value{GDBN} should always (yes) or never (no)
29321 create a core file. The default is to ask the user what to do.
29324 @kindex maint packet
29325 @item maint packet @var{text}
29326 If @value{GDBN} is talking to an inferior via the serial protocol,
29327 then this command sends the string @var{text} to the inferior, and
29328 displays the response packet. @value{GDBN} supplies the initial
29329 @samp{$} character, the terminating @samp{#} character, and the
29332 @kindex maint print architecture
29333 @item maint print architecture @r{[}@var{file}@r{]}
29334 Print the entire architecture configuration. The optional argument
29335 @var{file} names the file where the output goes.
29337 @kindex maint print c-tdesc
29338 @item maint print c-tdesc
29339 Print the current target description (@pxref{Target Descriptions}) as
29340 a C source file. The created source file can be used in @value{GDBN}
29341 when an XML parser is not available to parse the description.
29343 @kindex maint print dummy-frames
29344 @item maint print dummy-frames
29345 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
29348 (@value{GDBP}) @kbd{b add}
29350 (@value{GDBP}) @kbd{print add(2,3)}
29351 Breakpoint 2, add (a=2, b=3) at @dots{}
29353 The program being debugged stopped while in a function called from GDB.
29355 (@value{GDBP}) @kbd{maint print dummy-frames}
29356 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
29357 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
29358 call_lo=0x01014000 call_hi=0x01014001
29362 Takes an optional file parameter.
29364 @kindex maint print registers
29365 @kindex maint print raw-registers
29366 @kindex maint print cooked-registers
29367 @kindex maint print register-groups
29368 @item maint print registers @r{[}@var{file}@r{]}
29369 @itemx maint print raw-registers @r{[}@var{file}@r{]}
29370 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
29371 @itemx maint print register-groups @r{[}@var{file}@r{]}
29372 Print @value{GDBN}'s internal register data structures.
29374 The command @code{maint print raw-registers} includes the contents of
29375 the raw register cache; the command @code{maint print cooked-registers}
29376 includes the (cooked) value of all registers, including registers which
29377 aren't available on the target nor visible to user; and the
29378 command @code{maint print register-groups} includes the groups that each
29379 register is a member of. @xref{Registers,, Registers, gdbint,
29380 @value{GDBN} Internals}.
29382 These commands take an optional parameter, a file name to which to
29383 write the information.
29385 @kindex maint print reggroups
29386 @item maint print reggroups @r{[}@var{file}@r{]}
29387 Print @value{GDBN}'s internal register group data structures. The
29388 optional argument @var{file} tells to what file to write the
29391 The register groups info looks like this:
29394 (@value{GDBP}) @kbd{maint print reggroups}
29407 This command forces @value{GDBN} to flush its internal register cache.
29409 @kindex maint print objfiles
29410 @cindex info for known object files
29411 @item maint print objfiles
29412 Print a dump of all known object files. For each object file, this
29413 command prints its name, address in memory, and all of its psymtabs
29416 @kindex maint print statistics
29417 @cindex bcache statistics
29418 @item maint print statistics
29419 This command prints, for each object file in the program, various data
29420 about that object file followed by the byte cache (@dfn{bcache})
29421 statistics for the object file. The objfile data includes the number
29422 of minimal, partial, full, and stabs symbols, the number of types
29423 defined by the objfile, the number of as yet unexpanded psym tables,
29424 the number of line tables and string tables, and the amount of memory
29425 used by the various tables. The bcache statistics include the counts,
29426 sizes, and counts of duplicates of all and unique objects, max,
29427 average, and median entry size, total memory used and its overhead and
29428 savings, and various measures of the hash table size and chain
29431 @kindex maint print target-stack
29432 @cindex target stack description
29433 @item maint print target-stack
29434 A @dfn{target} is an interface between the debugger and a particular
29435 kind of file or process. Targets can be stacked in @dfn{strata},
29436 so that more than one target can potentially respond to a request.
29437 In particular, memory accesses will walk down the stack of targets
29438 until they find a target that is interested in handling that particular
29441 This command prints a short description of each layer that was pushed on
29442 the @dfn{target stack}, starting from the top layer down to the bottom one.
29444 @kindex maint print type
29445 @cindex type chain of a data type
29446 @item maint print type @var{expr}
29447 Print the type chain for a type specified by @var{expr}. The argument
29448 can be either a type name or a symbol. If it is a symbol, the type of
29449 that symbol is described. The type chain produced by this command is
29450 a recursive definition of the data type as stored in @value{GDBN}'s
29451 data structures, including its flags and contained types.
29453 @kindex maint set dwarf2 max-cache-age
29454 @kindex maint show dwarf2 max-cache-age
29455 @item maint set dwarf2 max-cache-age
29456 @itemx maint show dwarf2 max-cache-age
29457 Control the DWARF 2 compilation unit cache.
29459 @cindex DWARF 2 compilation units cache
29460 In object files with inter-compilation-unit references, such as those
29461 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
29462 reader needs to frequently refer to previously read compilation units.
29463 This setting controls how long a compilation unit will remain in the
29464 cache if it is not referenced. A higher limit means that cached
29465 compilation units will be stored in memory longer, and more total
29466 memory will be used. Setting it to zero disables caching, which will
29467 slow down @value{GDBN} startup, but reduce memory consumption.
29469 @kindex maint set profile
29470 @kindex maint show profile
29471 @cindex profiling GDB
29472 @item maint set profile
29473 @itemx maint show profile
29474 Control profiling of @value{GDBN}.
29476 Profiling will be disabled until you use the @samp{maint set profile}
29477 command to enable it. When you enable profiling, the system will begin
29478 collecting timing and execution count data; when you disable profiling or
29479 exit @value{GDBN}, the results will be written to a log file. Remember that
29480 if you use profiling, @value{GDBN} will overwrite the profiling log file
29481 (often called @file{gmon.out}). If you have a record of important profiling
29482 data in a @file{gmon.out} file, be sure to move it to a safe location.
29484 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
29485 compiled with the @samp{-pg} compiler option.
29487 @kindex maint set show-debug-regs
29488 @kindex maint show show-debug-regs
29489 @cindex hardware debug registers
29490 @item maint set show-debug-regs
29491 @itemx maint show show-debug-regs
29492 Control whether to show variables that mirror the hardware debug
29493 registers. Use @code{ON} to enable, @code{OFF} to disable. If
29494 enabled, the debug registers values are shown when @value{GDBN} inserts or
29495 removes a hardware breakpoint or watchpoint, and when the inferior
29496 triggers a hardware-assisted breakpoint or watchpoint.
29498 @kindex maint set show-all-tib
29499 @kindex maint show show-all-tib
29500 @item maint set show-all-tib
29501 @itemx maint show show-all-tib
29502 Control whether to show all non zero areas within a 1k block starting
29503 at thread local base, when using the @samp{info w32 thread-information-block}
29506 @kindex maint space
29507 @cindex memory used by commands
29509 Control whether to display memory usage for each command. If set to a
29510 nonzero value, @value{GDBN} will display how much memory each command
29511 took, following the command's own output. This can also be requested
29512 by invoking @value{GDBN} with the @option{--statistics} command-line
29513 switch (@pxref{Mode Options}).
29516 @cindex time of command execution
29518 Control whether to display the execution time for each command. If
29519 set to a nonzero value, @value{GDBN} will display how much time it
29520 took to execute each command, following the command's own output.
29521 The time is not printed for the commands that run the target, since
29522 there's no mechanism currently to compute how much time was spend
29523 by @value{GDBN} and how much time was spend by the program been debugged.
29524 it's not possibly currently
29525 This can also be requested by invoking @value{GDBN} with the
29526 @option{--statistics} command-line switch (@pxref{Mode Options}).
29528 @kindex maint translate-address
29529 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
29530 Find the symbol stored at the location specified by the address
29531 @var{addr} and an optional section name @var{section}. If found,
29532 @value{GDBN} prints the name of the closest symbol and an offset from
29533 the symbol's location to the specified address. This is similar to
29534 the @code{info address} command (@pxref{Symbols}), except that this
29535 command also allows to find symbols in other sections.
29537 If section was not specified, the section in which the symbol was found
29538 is also printed. For dynamically linked executables, the name of
29539 executable or shared library containing the symbol is printed as well.
29543 The following command is useful for non-interactive invocations of
29544 @value{GDBN}, such as in the test suite.
29547 @item set watchdog @var{nsec}
29548 @kindex set watchdog
29549 @cindex watchdog timer
29550 @cindex timeout for commands
29551 Set the maximum number of seconds @value{GDBN} will wait for the
29552 target operation to finish. If this time expires, @value{GDBN}
29553 reports and error and the command is aborted.
29555 @item show watchdog
29556 Show the current setting of the target wait timeout.
29559 @node Remote Protocol
29560 @appendix @value{GDBN} Remote Serial Protocol
29565 * Stop Reply Packets::
29566 * General Query Packets::
29567 * Architecture-Specific Protocol Details::
29568 * Tracepoint Packets::
29569 * Host I/O Packets::
29571 * Notification Packets::
29572 * Remote Non-Stop::
29573 * Packet Acknowledgment::
29575 * File-I/O Remote Protocol Extension::
29576 * Library List Format::
29577 * Memory Map Format::
29578 * Thread List Format::
29584 There may be occasions when you need to know something about the
29585 protocol---for example, if there is only one serial port to your target
29586 machine, you might want your program to do something special if it
29587 recognizes a packet meant for @value{GDBN}.
29589 In the examples below, @samp{->} and @samp{<-} are used to indicate
29590 transmitted and received data, respectively.
29592 @cindex protocol, @value{GDBN} remote serial
29593 @cindex serial protocol, @value{GDBN} remote
29594 @cindex remote serial protocol
29595 All @value{GDBN} commands and responses (other than acknowledgments
29596 and notifications, see @ref{Notification Packets}) are sent as a
29597 @var{packet}. A @var{packet} is introduced with the character
29598 @samp{$}, the actual @var{packet-data}, and the terminating character
29599 @samp{#} followed by a two-digit @var{checksum}:
29602 @code{$}@var{packet-data}@code{#}@var{checksum}
29606 @cindex checksum, for @value{GDBN} remote
29608 The two-digit @var{checksum} is computed as the modulo 256 sum of all
29609 characters between the leading @samp{$} and the trailing @samp{#} (an
29610 eight bit unsigned checksum).
29612 Implementors should note that prior to @value{GDBN} 5.0 the protocol
29613 specification also included an optional two-digit @var{sequence-id}:
29616 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
29619 @cindex sequence-id, for @value{GDBN} remote
29621 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
29622 has never output @var{sequence-id}s. Stubs that handle packets added
29623 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
29625 When either the host or the target machine receives a packet, the first
29626 response expected is an acknowledgment: either @samp{+} (to indicate
29627 the package was received correctly) or @samp{-} (to request
29631 -> @code{$}@var{packet-data}@code{#}@var{checksum}
29636 The @samp{+}/@samp{-} acknowledgments can be disabled
29637 once a connection is established.
29638 @xref{Packet Acknowledgment}, for details.
29640 The host (@value{GDBN}) sends @var{command}s, and the target (the
29641 debugging stub incorporated in your program) sends a @var{response}. In
29642 the case of step and continue @var{command}s, the response is only sent
29643 when the operation has completed, and the target has again stopped all
29644 threads in all attached processes. This is the default all-stop mode
29645 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
29646 execution mode; see @ref{Remote Non-Stop}, for details.
29648 @var{packet-data} consists of a sequence of characters with the
29649 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
29652 @cindex remote protocol, field separator
29653 Fields within the packet should be separated using @samp{,} @samp{;} or
29654 @samp{:}. Except where otherwise noted all numbers are represented in
29655 @sc{hex} with leading zeros suppressed.
29657 Implementors should note that prior to @value{GDBN} 5.0, the character
29658 @samp{:} could not appear as the third character in a packet (as it
29659 would potentially conflict with the @var{sequence-id}).
29661 @cindex remote protocol, binary data
29662 @anchor{Binary Data}
29663 Binary data in most packets is encoded either as two hexadecimal
29664 digits per byte of binary data. This allowed the traditional remote
29665 protocol to work over connections which were only seven-bit clean.
29666 Some packets designed more recently assume an eight-bit clean
29667 connection, and use a more efficient encoding to send and receive
29670 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
29671 as an escape character. Any escaped byte is transmitted as the escape
29672 character followed by the original character XORed with @code{0x20}.
29673 For example, the byte @code{0x7d} would be transmitted as the two
29674 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
29675 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
29676 @samp{@}}) must always be escaped. Responses sent by the stub
29677 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
29678 is not interpreted as the start of a run-length encoded sequence
29681 Response @var{data} can be run-length encoded to save space.
29682 Run-length encoding replaces runs of identical characters with one
29683 instance of the repeated character, followed by a @samp{*} and a
29684 repeat count. The repeat count is itself sent encoded, to avoid
29685 binary characters in @var{data}: a value of @var{n} is sent as
29686 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
29687 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
29688 code 32) for a repeat count of 3. (This is because run-length
29689 encoding starts to win for counts 3 or more.) Thus, for example,
29690 @samp{0* } is a run-length encoding of ``0000'': the space character
29691 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
29694 The printable characters @samp{#} and @samp{$} or with a numeric value
29695 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
29696 seven repeats (@samp{$}) can be expanded using a repeat count of only
29697 five (@samp{"}). For example, @samp{00000000} can be encoded as
29700 The error response returned for some packets includes a two character
29701 error number. That number is not well defined.
29703 @cindex empty response, for unsupported packets
29704 For any @var{command} not supported by the stub, an empty response
29705 (@samp{$#00}) should be returned. That way it is possible to extend the
29706 protocol. A newer @value{GDBN} can tell if a packet is supported based
29709 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
29710 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
29716 The following table provides a complete list of all currently defined
29717 @var{command}s and their corresponding response @var{data}.
29718 @xref{File-I/O Remote Protocol Extension}, for details about the File
29719 I/O extension of the remote protocol.
29721 Each packet's description has a template showing the packet's overall
29722 syntax, followed by an explanation of the packet's meaning. We
29723 include spaces in some of the templates for clarity; these are not
29724 part of the packet's syntax. No @value{GDBN} packet uses spaces to
29725 separate its components. For example, a template like @samp{foo
29726 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
29727 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
29728 @var{baz}. @value{GDBN} does not transmit a space character between the
29729 @samp{foo} and the @var{bar}, or between the @var{bar} and the
29732 @cindex @var{thread-id}, in remote protocol
29733 @anchor{thread-id syntax}
29734 Several packets and replies include a @var{thread-id} field to identify
29735 a thread. Normally these are positive numbers with a target-specific
29736 interpretation, formatted as big-endian hex strings. A @var{thread-id}
29737 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
29740 In addition, the remote protocol supports a multiprocess feature in
29741 which the @var{thread-id} syntax is extended to optionally include both
29742 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
29743 The @var{pid} (process) and @var{tid} (thread) components each have the
29744 format described above: a positive number with target-specific
29745 interpretation formatted as a big-endian hex string, literal @samp{-1}
29746 to indicate all processes or threads (respectively), or @samp{0} to
29747 indicate an arbitrary process or thread. Specifying just a process, as
29748 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
29749 error to specify all processes but a specific thread, such as
29750 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
29751 for those packets and replies explicitly documented to include a process
29752 ID, rather than a @var{thread-id}.
29754 The multiprocess @var{thread-id} syntax extensions are only used if both
29755 @value{GDBN} and the stub report support for the @samp{multiprocess}
29756 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
29759 Note that all packet forms beginning with an upper- or lower-case
29760 letter, other than those described here, are reserved for future use.
29762 Here are the packet descriptions.
29767 @cindex @samp{!} packet
29768 @anchor{extended mode}
29769 Enable extended mode. In extended mode, the remote server is made
29770 persistent. The @samp{R} packet is used to restart the program being
29776 The remote target both supports and has enabled extended mode.
29780 @cindex @samp{?} packet
29781 Indicate the reason the target halted. The reply is the same as for
29782 step and continue. This packet has a special interpretation when the
29783 target is in non-stop mode; see @ref{Remote Non-Stop}.
29786 @xref{Stop Reply Packets}, for the reply specifications.
29788 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
29789 @cindex @samp{A} packet
29790 Initialized @code{argv[]} array passed into program. @var{arglen}
29791 specifies the number of bytes in the hex encoded byte stream
29792 @var{arg}. See @code{gdbserver} for more details.
29797 The arguments were set.
29803 @cindex @samp{b} packet
29804 (Don't use this packet; its behavior is not well-defined.)
29805 Change the serial line speed to @var{baud}.
29807 JTC: @emph{When does the transport layer state change? When it's
29808 received, or after the ACK is transmitted. In either case, there are
29809 problems if the command or the acknowledgment packet is dropped.}
29811 Stan: @emph{If people really wanted to add something like this, and get
29812 it working for the first time, they ought to modify ser-unix.c to send
29813 some kind of out-of-band message to a specially-setup stub and have the
29814 switch happen "in between" packets, so that from remote protocol's point
29815 of view, nothing actually happened.}
29817 @item B @var{addr},@var{mode}
29818 @cindex @samp{B} packet
29819 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
29820 breakpoint at @var{addr}.
29822 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
29823 (@pxref{insert breakpoint or watchpoint packet}).
29825 @cindex @samp{bc} packet
29828 Backward continue. Execute the target system in reverse. No parameter.
29829 @xref{Reverse Execution}, for more information.
29832 @xref{Stop Reply Packets}, for the reply specifications.
29834 @cindex @samp{bs} packet
29837 Backward single step. Execute one instruction in reverse. No parameter.
29838 @xref{Reverse Execution}, for more information.
29841 @xref{Stop Reply Packets}, for the reply specifications.
29843 @item c @r{[}@var{addr}@r{]}
29844 @cindex @samp{c} packet
29845 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
29846 resume at current address.
29849 @xref{Stop Reply Packets}, for the reply specifications.
29851 @item C @var{sig}@r{[};@var{addr}@r{]}
29852 @cindex @samp{C} packet
29853 Continue with signal @var{sig} (hex signal number). If
29854 @samp{;@var{addr}} is omitted, resume at same address.
29857 @xref{Stop Reply Packets}, for the reply specifications.
29860 @cindex @samp{d} packet
29863 Don't use this packet; instead, define a general set packet
29864 (@pxref{General Query Packets}).
29868 @cindex @samp{D} packet
29869 The first form of the packet is used to detach @value{GDBN} from the
29870 remote system. It is sent to the remote target
29871 before @value{GDBN} disconnects via the @code{detach} command.
29873 The second form, including a process ID, is used when multiprocess
29874 protocol extensions are enabled (@pxref{multiprocess extensions}), to
29875 detach only a specific process. The @var{pid} is specified as a
29876 big-endian hex string.
29886 @item F @var{RC},@var{EE},@var{CF};@var{XX}
29887 @cindex @samp{F} packet
29888 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
29889 This is part of the File-I/O protocol extension. @xref{File-I/O
29890 Remote Protocol Extension}, for the specification.
29893 @anchor{read registers packet}
29894 @cindex @samp{g} packet
29895 Read general registers.
29899 @item @var{XX@dots{}}
29900 Each byte of register data is described by two hex digits. The bytes
29901 with the register are transmitted in target byte order. The size of
29902 each register and their position within the @samp{g} packet are
29903 determined by the @value{GDBN} internal gdbarch functions
29904 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
29905 specification of several standard @samp{g} packets is specified below.
29910 @item G @var{XX@dots{}}
29911 @cindex @samp{G} packet
29912 Write general registers. @xref{read registers packet}, for a
29913 description of the @var{XX@dots{}} data.
29923 @item H @var{c} @var{thread-id}
29924 @cindex @samp{H} packet
29925 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
29926 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
29927 should be @samp{c} for step and continue operations, @samp{g} for other
29928 operations. The thread designator @var{thread-id} has the format and
29929 interpretation described in @ref{thread-id syntax}.
29940 @c 'H': How restrictive (or permissive) is the thread model. If a
29941 @c thread is selected and stopped, are other threads allowed
29942 @c to continue to execute? As I mentioned above, I think the
29943 @c semantics of each command when a thread is selected must be
29944 @c described. For example:
29946 @c 'g': If the stub supports threads and a specific thread is
29947 @c selected, returns the register block from that thread;
29948 @c otherwise returns current registers.
29950 @c 'G' If the stub supports threads and a specific thread is
29951 @c selected, sets the registers of the register block of
29952 @c that thread; otherwise sets current registers.
29954 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
29955 @anchor{cycle step packet}
29956 @cindex @samp{i} packet
29957 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
29958 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
29959 step starting at that address.
29962 @cindex @samp{I} packet
29963 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
29967 @cindex @samp{k} packet
29970 FIXME: @emph{There is no description of how to operate when a specific
29971 thread context has been selected (i.e.@: does 'k' kill only that
29974 @item m @var{addr},@var{length}
29975 @cindex @samp{m} packet
29976 Read @var{length} bytes of memory starting at address @var{addr}.
29977 Note that @var{addr} may not be aligned to any particular boundary.
29979 The stub need not use any particular size or alignment when gathering
29980 data from memory for the response; even if @var{addr} is word-aligned
29981 and @var{length} is a multiple of the word size, the stub is free to
29982 use byte accesses, or not. For this reason, this packet may not be
29983 suitable for accessing memory-mapped I/O devices.
29984 @cindex alignment of remote memory accesses
29985 @cindex size of remote memory accesses
29986 @cindex memory, alignment and size of remote accesses
29990 @item @var{XX@dots{}}
29991 Memory contents; each byte is transmitted as a two-digit hexadecimal
29992 number. The reply may contain fewer bytes than requested if the
29993 server was able to read only part of the region of memory.
29998 @item M @var{addr},@var{length}:@var{XX@dots{}}
29999 @cindex @samp{M} packet
30000 Write @var{length} bytes of memory starting at address @var{addr}.
30001 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
30002 hexadecimal number.
30009 for an error (this includes the case where only part of the data was
30014 @cindex @samp{p} packet
30015 Read the value of register @var{n}; @var{n} is in hex.
30016 @xref{read registers packet}, for a description of how the returned
30017 register value is encoded.
30021 @item @var{XX@dots{}}
30022 the register's value
30026 Indicating an unrecognized @var{query}.
30029 @item P @var{n@dots{}}=@var{r@dots{}}
30030 @anchor{write register packet}
30031 @cindex @samp{P} packet
30032 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
30033 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
30034 digits for each byte in the register (target byte order).
30044 @item q @var{name} @var{params}@dots{}
30045 @itemx Q @var{name} @var{params}@dots{}
30046 @cindex @samp{q} packet
30047 @cindex @samp{Q} packet
30048 General query (@samp{q}) and set (@samp{Q}). These packets are
30049 described fully in @ref{General Query Packets}.
30052 @cindex @samp{r} packet
30053 Reset the entire system.
30055 Don't use this packet; use the @samp{R} packet instead.
30058 @cindex @samp{R} packet
30059 Restart the program being debugged. @var{XX}, while needed, is ignored.
30060 This packet is only available in extended mode (@pxref{extended mode}).
30062 The @samp{R} packet has no reply.
30064 @item s @r{[}@var{addr}@r{]}
30065 @cindex @samp{s} packet
30066 Single step. @var{addr} is the address at which to resume. If
30067 @var{addr} is omitted, resume at same address.
30070 @xref{Stop Reply Packets}, for the reply specifications.
30072 @item S @var{sig}@r{[};@var{addr}@r{]}
30073 @anchor{step with signal packet}
30074 @cindex @samp{S} packet
30075 Step with signal. This is analogous to the @samp{C} packet, but
30076 requests a single-step, rather than a normal resumption of execution.
30079 @xref{Stop Reply Packets}, for the reply specifications.
30081 @item t @var{addr}:@var{PP},@var{MM}
30082 @cindex @samp{t} packet
30083 Search backwards starting at address @var{addr} for a match with pattern
30084 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
30085 @var{addr} must be at least 3 digits.
30087 @item T @var{thread-id}
30088 @cindex @samp{T} packet
30089 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
30094 thread is still alive
30100 Packets starting with @samp{v} are identified by a multi-letter name,
30101 up to the first @samp{;} or @samp{?} (or the end of the packet).
30103 @item vAttach;@var{pid}
30104 @cindex @samp{vAttach} packet
30105 Attach to a new process with the specified process ID @var{pid}.
30106 The process ID is a
30107 hexadecimal integer identifying the process. In all-stop mode, all
30108 threads in the attached process are stopped; in non-stop mode, it may be
30109 attached without being stopped if that is supported by the target.
30111 @c In non-stop mode, on a successful vAttach, the stub should set the
30112 @c current thread to a thread of the newly-attached process. After
30113 @c attaching, GDB queries for the attached process's thread ID with qC.
30114 @c Also note that, from a user perspective, whether or not the
30115 @c target is stopped on attach in non-stop mode depends on whether you
30116 @c use the foreground or background version of the attach command, not
30117 @c on what vAttach does; GDB does the right thing with respect to either
30118 @c stopping or restarting threads.
30120 This packet is only available in extended mode (@pxref{extended mode}).
30126 @item @r{Any stop packet}
30127 for success in all-stop mode (@pxref{Stop Reply Packets})
30129 for success in non-stop mode (@pxref{Remote Non-Stop})
30132 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
30133 @cindex @samp{vCont} packet
30134 Resume the inferior, specifying different actions for each thread.
30135 If an action is specified with no @var{thread-id}, then it is applied to any
30136 threads that don't have a specific action specified; if no default action is
30137 specified then other threads should remain stopped in all-stop mode and
30138 in their current state in non-stop mode.
30139 Specifying multiple
30140 default actions is an error; specifying no actions is also an error.
30141 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
30143 Currently supported actions are:
30149 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
30153 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
30158 The optional argument @var{addr} normally associated with the
30159 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
30160 not supported in @samp{vCont}.
30162 The @samp{t} action is only relevant in non-stop mode
30163 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
30164 A stop reply should be generated for any affected thread not already stopped.
30165 When a thread is stopped by means of a @samp{t} action,
30166 the corresponding stop reply should indicate that the thread has stopped with
30167 signal @samp{0}, regardless of whether the target uses some other signal
30168 as an implementation detail.
30171 @xref{Stop Reply Packets}, for the reply specifications.
30174 @cindex @samp{vCont?} packet
30175 Request a list of actions supported by the @samp{vCont} packet.
30179 @item vCont@r{[};@var{action}@dots{}@r{]}
30180 The @samp{vCont} packet is supported. Each @var{action} is a supported
30181 command in the @samp{vCont} packet.
30183 The @samp{vCont} packet is not supported.
30186 @item vFile:@var{operation}:@var{parameter}@dots{}
30187 @cindex @samp{vFile} packet
30188 Perform a file operation on the target system. For details,
30189 see @ref{Host I/O Packets}.
30191 @item vFlashErase:@var{addr},@var{length}
30192 @cindex @samp{vFlashErase} packet
30193 Direct the stub to erase @var{length} bytes of flash starting at
30194 @var{addr}. The region may enclose any number of flash blocks, but
30195 its start and end must fall on block boundaries, as indicated by the
30196 flash block size appearing in the memory map (@pxref{Memory Map
30197 Format}). @value{GDBN} groups flash memory programming operations
30198 together, and sends a @samp{vFlashDone} request after each group; the
30199 stub is allowed to delay erase operation until the @samp{vFlashDone}
30200 packet is received.
30202 The stub must support @samp{vCont} if it reports support for
30203 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
30204 this case @samp{vCont} actions can be specified to apply to all threads
30205 in a process by using the @samp{p@var{pid}.-1} form of the
30216 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
30217 @cindex @samp{vFlashWrite} packet
30218 Direct the stub to write data to flash address @var{addr}. The data
30219 is passed in binary form using the same encoding as for the @samp{X}
30220 packet (@pxref{Binary Data}). The memory ranges specified by
30221 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
30222 not overlap, and must appear in order of increasing addresses
30223 (although @samp{vFlashErase} packets for higher addresses may already
30224 have been received; the ordering is guaranteed only between
30225 @samp{vFlashWrite} packets). If a packet writes to an address that was
30226 neither erased by a preceding @samp{vFlashErase} packet nor by some other
30227 target-specific method, the results are unpredictable.
30235 for vFlashWrite addressing non-flash memory
30241 @cindex @samp{vFlashDone} packet
30242 Indicate to the stub that flash programming operation is finished.
30243 The stub is permitted to delay or batch the effects of a group of
30244 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
30245 @samp{vFlashDone} packet is received. The contents of the affected
30246 regions of flash memory are unpredictable until the @samp{vFlashDone}
30247 request is completed.
30249 @item vKill;@var{pid}
30250 @cindex @samp{vKill} packet
30251 Kill the process with the specified process ID. @var{pid} is a
30252 hexadecimal integer identifying the process. This packet is used in
30253 preference to @samp{k} when multiprocess protocol extensions are
30254 supported; see @ref{multiprocess extensions}.
30264 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
30265 @cindex @samp{vRun} packet
30266 Run the program @var{filename}, passing it each @var{argument} on its
30267 command line. The file and arguments are hex-encoded strings. If
30268 @var{filename} is an empty string, the stub may use a default program
30269 (e.g.@: the last program run). The program is created in the stopped
30272 @c FIXME: What about non-stop mode?
30274 This packet is only available in extended mode (@pxref{extended mode}).
30280 @item @r{Any stop packet}
30281 for success (@pxref{Stop Reply Packets})
30285 @anchor{vStopped packet}
30286 @cindex @samp{vStopped} packet
30288 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
30289 reply and prompt for the stub to report another one.
30293 @item @r{Any stop packet}
30294 if there is another unreported stop event (@pxref{Stop Reply Packets})
30296 if there are no unreported stop events
30299 @item X @var{addr},@var{length}:@var{XX@dots{}}
30301 @cindex @samp{X} packet
30302 Write data to memory, where the data is transmitted in binary.
30303 @var{addr} is address, @var{length} is number of bytes,
30304 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
30314 @item z @var{type},@var{addr},@var{kind}
30315 @itemx Z @var{type},@var{addr},@var{kind}
30316 @anchor{insert breakpoint or watchpoint packet}
30317 @cindex @samp{z} packet
30318 @cindex @samp{Z} packets
30319 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
30320 watchpoint starting at address @var{address} of kind @var{kind}.
30322 Each breakpoint and watchpoint packet @var{type} is documented
30325 @emph{Implementation notes: A remote target shall return an empty string
30326 for an unrecognized breakpoint or watchpoint packet @var{type}. A
30327 remote target shall support either both or neither of a given
30328 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
30329 avoid potential problems with duplicate packets, the operations should
30330 be implemented in an idempotent way.}
30332 @item z0,@var{addr},@var{kind}
30333 @itemx Z0,@var{addr},@var{kind}
30334 @cindex @samp{z0} packet
30335 @cindex @samp{Z0} packet
30336 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
30337 @var{addr} of type @var{kind}.
30339 A memory breakpoint is implemented by replacing the instruction at
30340 @var{addr} with a software breakpoint or trap instruction. The
30341 @var{kind} is target-specific and typically indicates the size of
30342 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
30343 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
30344 architectures have additional meanings for @var{kind};
30345 see @ref{Architecture-Specific Protocol Details}.
30347 @emph{Implementation note: It is possible for a target to copy or move
30348 code that contains memory breakpoints (e.g., when implementing
30349 overlays). The behavior of this packet, in the presence of such a
30350 target, is not defined.}
30362 @item z1,@var{addr},@var{kind}
30363 @itemx Z1,@var{addr},@var{kind}
30364 @cindex @samp{z1} packet
30365 @cindex @samp{Z1} packet
30366 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
30367 address @var{addr}.
30369 A hardware breakpoint is implemented using a mechanism that is not
30370 dependant on being able to modify the target's memory. @var{kind}
30371 has the same meaning as in @samp{Z0} packets.
30373 @emph{Implementation note: A hardware breakpoint is not affected by code
30386 @item z2,@var{addr},@var{kind}
30387 @itemx Z2,@var{addr},@var{kind}
30388 @cindex @samp{z2} packet
30389 @cindex @samp{Z2} packet
30390 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
30391 @var{kind} is interpreted as the number of bytes to watch.
30403 @item z3,@var{addr},@var{kind}
30404 @itemx Z3,@var{addr},@var{kind}
30405 @cindex @samp{z3} packet
30406 @cindex @samp{Z3} packet
30407 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
30408 @var{kind} is interpreted as the number of bytes to watch.
30420 @item z4,@var{addr},@var{kind}
30421 @itemx Z4,@var{addr},@var{kind}
30422 @cindex @samp{z4} packet
30423 @cindex @samp{Z4} packet
30424 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
30425 @var{kind} is interpreted as the number of bytes to watch.
30439 @node Stop Reply Packets
30440 @section Stop Reply Packets
30441 @cindex stop reply packets
30443 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
30444 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
30445 receive any of the below as a reply. Except for @samp{?}
30446 and @samp{vStopped}, that reply is only returned
30447 when the target halts. In the below the exact meaning of @dfn{signal
30448 number} is defined by the header @file{include/gdb/signals.h} in the
30449 @value{GDBN} source code.
30451 As in the description of request packets, we include spaces in the
30452 reply templates for clarity; these are not part of the reply packet's
30453 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
30459 The program received signal number @var{AA} (a two-digit hexadecimal
30460 number). This is equivalent to a @samp{T} response with no
30461 @var{n}:@var{r} pairs.
30463 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
30464 @cindex @samp{T} packet reply
30465 The program received signal number @var{AA} (a two-digit hexadecimal
30466 number). This is equivalent to an @samp{S} response, except that the
30467 @samp{@var{n}:@var{r}} pairs can carry values of important registers
30468 and other information directly in the stop reply packet, reducing
30469 round-trip latency. Single-step and breakpoint traps are reported
30470 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
30474 If @var{n} is a hexadecimal number, it is a register number, and the
30475 corresponding @var{r} gives that register's value. @var{r} is a
30476 series of bytes in target byte order, with each byte given by a
30477 two-digit hex number.
30480 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
30481 the stopped thread, as specified in @ref{thread-id syntax}.
30484 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
30485 the core on which the stop event was detected.
30488 If @var{n} is a recognized @dfn{stop reason}, it describes a more
30489 specific event that stopped the target. The currently defined stop
30490 reasons are listed below. @var{aa} should be @samp{05}, the trap
30491 signal. At most one stop reason should be present.
30494 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
30495 and go on to the next; this allows us to extend the protocol in the
30499 The currently defined stop reasons are:
30505 The packet indicates a watchpoint hit, and @var{r} is the data address, in
30508 @cindex shared library events, remote reply
30510 The packet indicates that the loaded libraries have changed.
30511 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
30512 list of loaded libraries. @var{r} is ignored.
30514 @cindex replay log events, remote reply
30516 The packet indicates that the target cannot continue replaying
30517 logged execution events, because it has reached the end (or the
30518 beginning when executing backward) of the log. The value of @var{r}
30519 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
30520 for more information.
30524 @itemx W @var{AA} ; process:@var{pid}
30525 The process exited, and @var{AA} is the exit status. This is only
30526 applicable to certain targets.
30528 The second form of the response, including the process ID of the exited
30529 process, can be used only when @value{GDBN} has reported support for
30530 multiprocess protocol extensions; see @ref{multiprocess extensions}.
30531 The @var{pid} is formatted as a big-endian hex string.
30534 @itemx X @var{AA} ; process:@var{pid}
30535 The process terminated with signal @var{AA}.
30537 The second form of the response, including the process ID of the
30538 terminated process, can be used only when @value{GDBN} has reported
30539 support for multiprocess protocol extensions; see @ref{multiprocess
30540 extensions}. The @var{pid} is formatted as a big-endian hex string.
30542 @item O @var{XX}@dots{}
30543 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
30544 written as the program's console output. This can happen at any time
30545 while the program is running and the debugger should continue to wait
30546 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
30548 @item F @var{call-id},@var{parameter}@dots{}
30549 @var{call-id} is the identifier which says which host system call should
30550 be called. This is just the name of the function. Translation into the
30551 correct system call is only applicable as it's defined in @value{GDBN}.
30552 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
30555 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
30556 this very system call.
30558 The target replies with this packet when it expects @value{GDBN} to
30559 call a host system call on behalf of the target. @value{GDBN} replies
30560 with an appropriate @samp{F} packet and keeps up waiting for the next
30561 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
30562 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
30563 Protocol Extension}, for more details.
30567 @node General Query Packets
30568 @section General Query Packets
30569 @cindex remote query requests
30571 Packets starting with @samp{q} are @dfn{general query packets};
30572 packets starting with @samp{Q} are @dfn{general set packets}. General
30573 query and set packets are a semi-unified form for retrieving and
30574 sending information to and from the stub.
30576 The initial letter of a query or set packet is followed by a name
30577 indicating what sort of thing the packet applies to. For example,
30578 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
30579 definitions with the stub. These packet names follow some
30584 The name must not contain commas, colons or semicolons.
30586 Most @value{GDBN} query and set packets have a leading upper case
30589 The names of custom vendor packets should use a company prefix, in
30590 lower case, followed by a period. For example, packets designed at
30591 the Acme Corporation might begin with @samp{qacme.foo} (for querying
30592 foos) or @samp{Qacme.bar} (for setting bars).
30595 The name of a query or set packet should be separated from any
30596 parameters by a @samp{:}; the parameters themselves should be
30597 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
30598 full packet name, and check for a separator or the end of the packet,
30599 in case two packet names share a common prefix. New packets should not begin
30600 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
30601 packets predate these conventions, and have arguments without any terminator
30602 for the packet name; we suspect they are in widespread use in places that
30603 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
30604 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
30607 Like the descriptions of the other packets, each description here
30608 has a template showing the packet's overall syntax, followed by an
30609 explanation of the packet's meaning. We include spaces in some of the
30610 templates for clarity; these are not part of the packet's syntax. No
30611 @value{GDBN} packet uses spaces to separate its components.
30613 Here are the currently defined query and set packets:
30618 @cindex current thread, remote request
30619 @cindex @samp{qC} packet
30620 Return the current thread ID.
30624 @item QC @var{thread-id}
30625 Where @var{thread-id} is a thread ID as documented in
30626 @ref{thread-id syntax}.
30627 @item @r{(anything else)}
30628 Any other reply implies the old thread ID.
30631 @item qCRC:@var{addr},@var{length}
30632 @cindex CRC of memory block, remote request
30633 @cindex @samp{qCRC} packet
30634 Compute the CRC checksum of a block of memory using CRC-32 defined in
30635 IEEE 802.3. The CRC is computed byte at a time, taking the most
30636 significant bit of each byte first. The initial pattern code
30637 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
30639 @emph{Note:} This is the same CRC used in validating separate debug
30640 files (@pxref{Separate Debug Files, , Debugging Information in Separate
30641 Files}). However the algorithm is slightly different. When validating
30642 separate debug files, the CRC is computed taking the @emph{least}
30643 significant bit of each byte first, and the final result is inverted to
30644 detect trailing zeros.
30649 An error (such as memory fault)
30650 @item C @var{crc32}
30651 The specified memory region's checksum is @var{crc32}.
30655 @itemx qsThreadInfo
30656 @cindex list active threads, remote request
30657 @cindex @samp{qfThreadInfo} packet
30658 @cindex @samp{qsThreadInfo} packet
30659 Obtain a list of all active thread IDs from the target (OS). Since there
30660 may be too many active threads to fit into one reply packet, this query
30661 works iteratively: it may require more than one query/reply sequence to
30662 obtain the entire list of threads. The first query of the sequence will
30663 be the @samp{qfThreadInfo} query; subsequent queries in the
30664 sequence will be the @samp{qsThreadInfo} query.
30666 NOTE: This packet replaces the @samp{qL} query (see below).
30670 @item m @var{thread-id}
30672 @item m @var{thread-id},@var{thread-id}@dots{}
30673 a comma-separated list of thread IDs
30675 (lower case letter @samp{L}) denotes end of list.
30678 In response to each query, the target will reply with a list of one or
30679 more thread IDs, separated by commas.
30680 @value{GDBN} will respond to each reply with a request for more thread
30681 ids (using the @samp{qs} form of the query), until the target responds
30682 with @samp{l} (lower-case el, for @dfn{last}).
30683 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
30686 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
30687 @cindex get thread-local storage address, remote request
30688 @cindex @samp{qGetTLSAddr} packet
30689 Fetch the address associated with thread local storage specified
30690 by @var{thread-id}, @var{offset}, and @var{lm}.
30692 @var{thread-id} is the thread ID associated with the
30693 thread for which to fetch the TLS address. @xref{thread-id syntax}.
30695 @var{offset} is the (big endian, hex encoded) offset associated with the
30696 thread local variable. (This offset is obtained from the debug
30697 information associated with the variable.)
30699 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
30700 the load module associated with the thread local storage. For example,
30701 a @sc{gnu}/Linux system will pass the link map address of the shared
30702 object associated with the thread local storage under consideration.
30703 Other operating environments may choose to represent the load module
30704 differently, so the precise meaning of this parameter will vary.
30708 @item @var{XX}@dots{}
30709 Hex encoded (big endian) bytes representing the address of the thread
30710 local storage requested.
30713 An error occurred. @var{nn} are hex digits.
30716 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
30719 @item qGetTIBAddr:@var{thread-id}
30720 @cindex get thread information block address
30721 @cindex @samp{qGetTIBAddr} packet
30722 Fetch address of the Windows OS specific Thread Information Block.
30724 @var{thread-id} is the thread ID associated with the thread.
30728 @item @var{XX}@dots{}
30729 Hex encoded (big endian) bytes representing the linear address of the
30730 thread information block.
30733 An error occured. This means that either the thread was not found, or the
30734 address could not be retrieved.
30737 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
30740 @item qL @var{startflag} @var{threadcount} @var{nextthread}
30741 Obtain thread information from RTOS. Where: @var{startflag} (one hex
30742 digit) is one to indicate the first query and zero to indicate a
30743 subsequent query; @var{threadcount} (two hex digits) is the maximum
30744 number of threads the response packet can contain; and @var{nextthread}
30745 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
30746 returned in the response as @var{argthread}.
30748 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
30752 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
30753 Where: @var{count} (two hex digits) is the number of threads being
30754 returned; @var{done} (one hex digit) is zero to indicate more threads
30755 and one indicates no further threads; @var{argthreadid} (eight hex
30756 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
30757 is a sequence of thread IDs from the target. @var{threadid} (eight hex
30758 digits). See @code{remote.c:parse_threadlist_response()}.
30762 @cindex section offsets, remote request
30763 @cindex @samp{qOffsets} packet
30764 Get section offsets that the target used when relocating the downloaded
30769 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
30770 Relocate the @code{Text} section by @var{xxx} from its original address.
30771 Relocate the @code{Data} section by @var{yyy} from its original address.
30772 If the object file format provides segment information (e.g.@: @sc{elf}
30773 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
30774 segments by the supplied offsets.
30776 @emph{Note: while a @code{Bss} offset may be included in the response,
30777 @value{GDBN} ignores this and instead applies the @code{Data} offset
30778 to the @code{Bss} section.}
30780 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
30781 Relocate the first segment of the object file, which conventionally
30782 contains program code, to a starting address of @var{xxx}. If
30783 @samp{DataSeg} is specified, relocate the second segment, which
30784 conventionally contains modifiable data, to a starting address of
30785 @var{yyy}. @value{GDBN} will report an error if the object file
30786 does not contain segment information, or does not contain at least
30787 as many segments as mentioned in the reply. Extra segments are
30788 kept at fixed offsets relative to the last relocated segment.
30791 @item qP @var{mode} @var{thread-id}
30792 @cindex thread information, remote request
30793 @cindex @samp{qP} packet
30794 Returns information on @var{thread-id}. Where: @var{mode} is a hex
30795 encoded 32 bit mode; @var{thread-id} is a thread ID
30796 (@pxref{thread-id syntax}).
30798 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
30801 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
30805 @cindex non-stop mode, remote request
30806 @cindex @samp{QNonStop} packet
30808 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
30809 @xref{Remote Non-Stop}, for more information.
30814 The request succeeded.
30817 An error occurred. @var{nn} are hex digits.
30820 An empty reply indicates that @samp{QNonStop} is not supported by
30824 This packet is not probed by default; the remote stub must request it,
30825 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
30826 Use of this packet is controlled by the @code{set non-stop} command;
30827 @pxref{Non-Stop Mode}.
30829 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
30830 @cindex pass signals to inferior, remote request
30831 @cindex @samp{QPassSignals} packet
30832 @anchor{QPassSignals}
30833 Each listed @var{signal} should be passed directly to the inferior process.
30834 Signals are numbered identically to continue packets and stop replies
30835 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
30836 strictly greater than the previous item. These signals do not need to stop
30837 the inferior, or be reported to @value{GDBN}. All other signals should be
30838 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
30839 combine; any earlier @samp{QPassSignals} list is completely replaced by the
30840 new list. This packet improves performance when using @samp{handle
30841 @var{signal} nostop noprint pass}.
30846 The request succeeded.
30849 An error occurred. @var{nn} are hex digits.
30852 An empty reply indicates that @samp{QPassSignals} is not supported by
30856 Use of this packet is controlled by the @code{set remote pass-signals}
30857 command (@pxref{Remote Configuration, set remote pass-signals}).
30858 This packet is not probed by default; the remote stub must request it,
30859 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
30861 @item qRcmd,@var{command}
30862 @cindex execute remote command, remote request
30863 @cindex @samp{qRcmd} packet
30864 @var{command} (hex encoded) is passed to the local interpreter for
30865 execution. Invalid commands should be reported using the output
30866 string. Before the final result packet, the target may also respond
30867 with a number of intermediate @samp{O@var{output}} console output
30868 packets. @emph{Implementors should note that providing access to a
30869 stubs's interpreter may have security implications}.
30874 A command response with no output.
30876 A command response with the hex encoded output string @var{OUTPUT}.
30878 Indicate a badly formed request.
30880 An empty reply indicates that @samp{qRcmd} is not recognized.
30883 (Note that the @code{qRcmd} packet's name is separated from the
30884 command by a @samp{,}, not a @samp{:}, contrary to the naming
30885 conventions above. Please don't use this packet as a model for new
30888 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
30889 @cindex searching memory, in remote debugging
30890 @cindex @samp{qSearch:memory} packet
30891 @anchor{qSearch memory}
30892 Search @var{length} bytes at @var{address} for @var{search-pattern}.
30893 @var{address} and @var{length} are encoded in hex.
30894 @var{search-pattern} is a sequence of bytes, hex encoded.
30899 The pattern was not found.
30901 The pattern was found at @var{address}.
30903 A badly formed request or an error was encountered while searching memory.
30905 An empty reply indicates that @samp{qSearch:memory} is not recognized.
30908 @item QStartNoAckMode
30909 @cindex @samp{QStartNoAckMode} packet
30910 @anchor{QStartNoAckMode}
30911 Request that the remote stub disable the normal @samp{+}/@samp{-}
30912 protocol acknowledgments (@pxref{Packet Acknowledgment}).
30917 The stub has switched to no-acknowledgment mode.
30918 @value{GDBN} acknowledges this reponse,
30919 but neither the stub nor @value{GDBN} shall send or expect further
30920 @samp{+}/@samp{-} acknowledgments in the current connection.
30922 An empty reply indicates that the stub does not support no-acknowledgment mode.
30925 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
30926 @cindex supported packets, remote query
30927 @cindex features of the remote protocol
30928 @cindex @samp{qSupported} packet
30929 @anchor{qSupported}
30930 Tell the remote stub about features supported by @value{GDBN}, and
30931 query the stub for features it supports. This packet allows
30932 @value{GDBN} and the remote stub to take advantage of each others'
30933 features. @samp{qSupported} also consolidates multiple feature probes
30934 at startup, to improve @value{GDBN} performance---a single larger
30935 packet performs better than multiple smaller probe packets on
30936 high-latency links. Some features may enable behavior which must not
30937 be on by default, e.g.@: because it would confuse older clients or
30938 stubs. Other features may describe packets which could be
30939 automatically probed for, but are not. These features must be
30940 reported before @value{GDBN} will use them. This ``default
30941 unsupported'' behavior is not appropriate for all packets, but it
30942 helps to keep the initial connection time under control with new
30943 versions of @value{GDBN} which support increasing numbers of packets.
30947 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
30948 The stub supports or does not support each returned @var{stubfeature},
30949 depending on the form of each @var{stubfeature} (see below for the
30952 An empty reply indicates that @samp{qSupported} is not recognized,
30953 or that no features needed to be reported to @value{GDBN}.
30956 The allowed forms for each feature (either a @var{gdbfeature} in the
30957 @samp{qSupported} packet, or a @var{stubfeature} in the response)
30961 @item @var{name}=@var{value}
30962 The remote protocol feature @var{name} is supported, and associated
30963 with the specified @var{value}. The format of @var{value} depends
30964 on the feature, but it must not include a semicolon.
30966 The remote protocol feature @var{name} is supported, and does not
30967 need an associated value.
30969 The remote protocol feature @var{name} is not supported.
30971 The remote protocol feature @var{name} may be supported, and
30972 @value{GDBN} should auto-detect support in some other way when it is
30973 needed. This form will not be used for @var{gdbfeature} notifications,
30974 but may be used for @var{stubfeature} responses.
30977 Whenever the stub receives a @samp{qSupported} request, the
30978 supplied set of @value{GDBN} features should override any previous
30979 request. This allows @value{GDBN} to put the stub in a known
30980 state, even if the stub had previously been communicating with
30981 a different version of @value{GDBN}.
30983 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
30988 This feature indicates whether @value{GDBN} supports multiprocess
30989 extensions to the remote protocol. @value{GDBN} does not use such
30990 extensions unless the stub also reports that it supports them by
30991 including @samp{multiprocess+} in its @samp{qSupported} reply.
30992 @xref{multiprocess extensions}, for details.
30995 This feature indicates that @value{GDBN} supports the XML target
30996 description. If the stub sees @samp{xmlRegisters=} with target
30997 specific strings separated by a comma, it will report register
31001 Stubs should ignore any unknown values for
31002 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
31003 packet supports receiving packets of unlimited length (earlier
31004 versions of @value{GDBN} may reject overly long responses). Additional values
31005 for @var{gdbfeature} may be defined in the future to let the stub take
31006 advantage of new features in @value{GDBN}, e.g.@: incompatible
31007 improvements in the remote protocol---the @samp{multiprocess} feature is
31008 an example of such a feature. The stub's reply should be independent
31009 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
31010 describes all the features it supports, and then the stub replies with
31011 all the features it supports.
31013 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
31014 responses, as long as each response uses one of the standard forms.
31016 Some features are flags. A stub which supports a flag feature
31017 should respond with a @samp{+} form response. Other features
31018 require values, and the stub should respond with an @samp{=}
31021 Each feature has a default value, which @value{GDBN} will use if
31022 @samp{qSupported} is not available or if the feature is not mentioned
31023 in the @samp{qSupported} response. The default values are fixed; a
31024 stub is free to omit any feature responses that match the defaults.
31026 Not all features can be probed, but for those which can, the probing
31027 mechanism is useful: in some cases, a stub's internal
31028 architecture may not allow the protocol layer to know some information
31029 about the underlying target in advance. This is especially common in
31030 stubs which may be configured for multiple targets.
31032 These are the currently defined stub features and their properties:
31034 @multitable @columnfractions 0.35 0.2 0.12 0.2
31035 @c NOTE: The first row should be @headitem, but we do not yet require
31036 @c a new enough version of Texinfo (4.7) to use @headitem.
31038 @tab Value Required
31042 @item @samp{PacketSize}
31047 @item @samp{qXfer:auxv:read}
31052 @item @samp{qXfer:features:read}
31057 @item @samp{qXfer:libraries:read}
31062 @item @samp{qXfer:memory-map:read}
31067 @item @samp{qXfer:spu:read}
31072 @item @samp{qXfer:spu:write}
31077 @item @samp{qXfer:siginfo:read}
31082 @item @samp{qXfer:siginfo:write}
31087 @item @samp{qXfer:threads:read}
31093 @item @samp{QNonStop}
31098 @item @samp{QPassSignals}
31103 @item @samp{QStartNoAckMode}
31108 @item @samp{multiprocess}
31113 @item @samp{ConditionalTracepoints}
31118 @item @samp{ReverseContinue}
31123 @item @samp{ReverseStep}
31128 @item @samp{TracepointSource}
31135 These are the currently defined stub features, in more detail:
31138 @cindex packet size, remote protocol
31139 @item PacketSize=@var{bytes}
31140 The remote stub can accept packets up to at least @var{bytes} in
31141 length. @value{GDBN} will send packets up to this size for bulk
31142 transfers, and will never send larger packets. This is a limit on the
31143 data characters in the packet, including the frame and checksum.
31144 There is no trailing NUL byte in a remote protocol packet; if the stub
31145 stores packets in a NUL-terminated format, it should allow an extra
31146 byte in its buffer for the NUL. If this stub feature is not supported,
31147 @value{GDBN} guesses based on the size of the @samp{g} packet response.
31149 @item qXfer:auxv:read
31150 The remote stub understands the @samp{qXfer:auxv:read} packet
31151 (@pxref{qXfer auxiliary vector read}).
31153 @item qXfer:features:read
31154 The remote stub understands the @samp{qXfer:features:read} packet
31155 (@pxref{qXfer target description read}).
31157 @item qXfer:libraries:read
31158 The remote stub understands the @samp{qXfer:libraries:read} packet
31159 (@pxref{qXfer library list read}).
31161 @item qXfer:memory-map:read
31162 The remote stub understands the @samp{qXfer:memory-map:read} packet
31163 (@pxref{qXfer memory map read}).
31165 @item qXfer:spu:read
31166 The remote stub understands the @samp{qXfer:spu:read} packet
31167 (@pxref{qXfer spu read}).
31169 @item qXfer:spu:write
31170 The remote stub understands the @samp{qXfer:spu:write} packet
31171 (@pxref{qXfer spu write}).
31173 @item qXfer:siginfo:read
31174 The remote stub understands the @samp{qXfer:siginfo:read} packet
31175 (@pxref{qXfer siginfo read}).
31177 @item qXfer:siginfo:write
31178 The remote stub understands the @samp{qXfer:siginfo:write} packet
31179 (@pxref{qXfer siginfo write}).
31181 @item qXfer:threads:read
31182 The remote stub understands the @samp{qXfer:threads:read} packet
31183 (@pxref{qXfer threads read}).
31186 The remote stub understands the @samp{QNonStop} packet
31187 (@pxref{QNonStop}).
31190 The remote stub understands the @samp{QPassSignals} packet
31191 (@pxref{QPassSignals}).
31193 @item QStartNoAckMode
31194 The remote stub understands the @samp{QStartNoAckMode} packet and
31195 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
31198 @anchor{multiprocess extensions}
31199 @cindex multiprocess extensions, in remote protocol
31200 The remote stub understands the multiprocess extensions to the remote
31201 protocol syntax. The multiprocess extensions affect the syntax of
31202 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
31203 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
31204 replies. Note that reporting this feature indicates support for the
31205 syntactic extensions only, not that the stub necessarily supports
31206 debugging of more than one process at a time. The stub must not use
31207 multiprocess extensions in packet replies unless @value{GDBN} has also
31208 indicated it supports them in its @samp{qSupported} request.
31210 @item qXfer:osdata:read
31211 The remote stub understands the @samp{qXfer:osdata:read} packet
31212 ((@pxref{qXfer osdata read}).
31214 @item ConditionalTracepoints
31215 The remote stub accepts and implements conditional expressions defined
31216 for tracepoints (@pxref{Tracepoint Conditions}).
31218 @item ReverseContinue
31219 The remote stub accepts and implements the reverse continue packet
31223 The remote stub accepts and implements the reverse step packet
31226 @item TracepointSource
31227 The remote stub understands the @samp{QTDPsrc} packet that supplies
31228 the source form of tracepoint definitions.
31233 @cindex symbol lookup, remote request
31234 @cindex @samp{qSymbol} packet
31235 Notify the target that @value{GDBN} is prepared to serve symbol lookup
31236 requests. Accept requests from the target for the values of symbols.
31241 The target does not need to look up any (more) symbols.
31242 @item qSymbol:@var{sym_name}
31243 The target requests the value of symbol @var{sym_name} (hex encoded).
31244 @value{GDBN} may provide the value by using the
31245 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
31249 @item qSymbol:@var{sym_value}:@var{sym_name}
31250 Set the value of @var{sym_name} to @var{sym_value}.
31252 @var{sym_name} (hex encoded) is the name of a symbol whose value the
31253 target has previously requested.
31255 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
31256 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
31262 The target does not need to look up any (more) symbols.
31263 @item qSymbol:@var{sym_name}
31264 The target requests the value of a new symbol @var{sym_name} (hex
31265 encoded). @value{GDBN} will continue to supply the values of symbols
31266 (if available), until the target ceases to request them.
31271 @item QTDisconnected
31278 @xref{Tracepoint Packets}.
31280 @item qThreadExtraInfo,@var{thread-id}
31281 @cindex thread attributes info, remote request
31282 @cindex @samp{qThreadExtraInfo} packet
31283 Obtain a printable string description of a thread's attributes from
31284 the target OS. @var{thread-id} is a thread ID;
31285 see @ref{thread-id syntax}. This
31286 string may contain anything that the target OS thinks is interesting
31287 for @value{GDBN} to tell the user about the thread. The string is
31288 displayed in @value{GDBN}'s @code{info threads} display. Some
31289 examples of possible thread extra info strings are @samp{Runnable}, or
31290 @samp{Blocked on Mutex}.
31294 @item @var{XX}@dots{}
31295 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
31296 comprising the printable string containing the extra information about
31297 the thread's attributes.
31300 (Note that the @code{qThreadExtraInfo} packet's name is separated from
31301 the command by a @samp{,}, not a @samp{:}, contrary to the naming
31302 conventions above. Please don't use this packet as a model for new
31314 @xref{Tracepoint Packets}.
31316 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
31317 @cindex read special object, remote request
31318 @cindex @samp{qXfer} packet
31319 @anchor{qXfer read}
31320 Read uninterpreted bytes from the target's special data area
31321 identified by the keyword @var{object}. Request @var{length} bytes
31322 starting at @var{offset} bytes into the data. The content and
31323 encoding of @var{annex} is specific to @var{object}; it can supply
31324 additional details about what data to access.
31326 Here are the specific requests of this form defined so far. All
31327 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
31328 formats, listed below.
31331 @item qXfer:auxv:read::@var{offset},@var{length}
31332 @anchor{qXfer auxiliary vector read}
31333 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
31334 auxiliary vector}. Note @var{annex} must be empty.
31336 This packet is not probed by default; the remote stub must request it,
31337 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
31339 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
31340 @anchor{qXfer target description read}
31341 Access the @dfn{target description}. @xref{Target Descriptions}. The
31342 annex specifies which XML document to access. The main description is
31343 always loaded from the @samp{target.xml} annex.
31345 This packet is not probed by default; the remote stub must request it,
31346 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
31348 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
31349 @anchor{qXfer library list read}
31350 Access the target's list of loaded libraries. @xref{Library List Format}.
31351 The annex part of the generic @samp{qXfer} packet must be empty
31352 (@pxref{qXfer read}).
31354 Targets which maintain a list of libraries in the program's memory do
31355 not need to implement this packet; it is designed for platforms where
31356 the operating system manages the list of loaded libraries.
31358 This packet is not probed by default; the remote stub must request it,
31359 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
31361 @item qXfer:memory-map:read::@var{offset},@var{length}
31362 @anchor{qXfer memory map read}
31363 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
31364 annex part of the generic @samp{qXfer} packet must be empty
31365 (@pxref{qXfer read}).
31367 This packet is not probed by default; the remote stub must request it,
31368 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
31370 @item qXfer:siginfo:read::@var{offset},@var{length}
31371 @anchor{qXfer siginfo read}
31372 Read contents of the extra signal information on the target
31373 system. The annex part of the generic @samp{qXfer} packet must be
31374 empty (@pxref{qXfer read}).
31376 This packet is not probed by default; the remote stub must request it,
31377 by supplying an appropriate @samp{qSupported} response
31378 (@pxref{qSupported}).
31380 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
31381 @anchor{qXfer spu read}
31382 Read contents of an @code{spufs} file on the target system. The
31383 annex specifies which file to read; it must be of the form
31384 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
31385 in the target process, and @var{name} identifes the @code{spufs} file
31386 in that context to be accessed.
31388 This packet is not probed by default; the remote stub must request it,
31389 by supplying an appropriate @samp{qSupported} response
31390 (@pxref{qSupported}).
31392 @item qXfer:threads:read::@var{offset},@var{length}
31393 @anchor{qXfer threads read}
31394 Access the list of threads on target. @xref{Thread List Format}. The
31395 annex part of the generic @samp{qXfer} packet must be empty
31396 (@pxref{qXfer read}).
31398 This packet is not probed by default; the remote stub must request it,
31399 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
31401 @item qXfer:osdata:read::@var{offset},@var{length}
31402 @anchor{qXfer osdata read}
31403 Access the target's @dfn{operating system information}.
31404 @xref{Operating System Information}.
31411 Data @var{data} (@pxref{Binary Data}) has been read from the
31412 target. There may be more data at a higher address (although
31413 it is permitted to return @samp{m} even for the last valid
31414 block of data, as long as at least one byte of data was read).
31415 @var{data} may have fewer bytes than the @var{length} in the
31419 Data @var{data} (@pxref{Binary Data}) has been read from the target.
31420 There is no more data to be read. @var{data} may have fewer bytes
31421 than the @var{length} in the request.
31424 The @var{offset} in the request is at the end of the data.
31425 There is no more data to be read.
31428 The request was malformed, or @var{annex} was invalid.
31431 The offset was invalid, or there was an error encountered reading the data.
31432 @var{nn} is a hex-encoded @code{errno} value.
31435 An empty reply indicates the @var{object} string was not recognized by
31436 the stub, or that the object does not support reading.
31439 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
31440 @cindex write data into object, remote request
31441 @anchor{qXfer write}
31442 Write uninterpreted bytes into the target's special data area
31443 identified by the keyword @var{object}, starting at @var{offset} bytes
31444 into the data. @var{data}@dots{} is the binary-encoded data
31445 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
31446 is specific to @var{object}; it can supply additional details about what data
31449 Here are the specific requests of this form defined so far. All
31450 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
31451 formats, listed below.
31454 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
31455 @anchor{qXfer siginfo write}
31456 Write @var{data} to the extra signal information on the target system.
31457 The annex part of the generic @samp{qXfer} packet must be
31458 empty (@pxref{qXfer write}).
31460 This packet is not probed by default; the remote stub must request it,
31461 by supplying an appropriate @samp{qSupported} response
31462 (@pxref{qSupported}).
31464 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
31465 @anchor{qXfer spu write}
31466 Write @var{data} to an @code{spufs} file on the target system. The
31467 annex specifies which file to write; it must be of the form
31468 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
31469 in the target process, and @var{name} identifes the @code{spufs} file
31470 in that context to be accessed.
31472 This packet is not probed by default; the remote stub must request it,
31473 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
31479 @var{nn} (hex encoded) is the number of bytes written.
31480 This may be fewer bytes than supplied in the request.
31483 The request was malformed, or @var{annex} was invalid.
31486 The offset was invalid, or there was an error encountered writing the data.
31487 @var{nn} is a hex-encoded @code{errno} value.
31490 An empty reply indicates the @var{object} string was not
31491 recognized by the stub, or that the object does not support writing.
31494 @item qXfer:@var{object}:@var{operation}:@dots{}
31495 Requests of this form may be added in the future. When a stub does
31496 not recognize the @var{object} keyword, or its support for
31497 @var{object} does not recognize the @var{operation} keyword, the stub
31498 must respond with an empty packet.
31500 @item qAttached:@var{pid}
31501 @cindex query attached, remote request
31502 @cindex @samp{qAttached} packet
31503 Return an indication of whether the remote server attached to an
31504 existing process or created a new process. When the multiprocess
31505 protocol extensions are supported (@pxref{multiprocess extensions}),
31506 @var{pid} is an integer in hexadecimal format identifying the target
31507 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
31508 the query packet will be simplified as @samp{qAttached}.
31510 This query is used, for example, to know whether the remote process
31511 should be detached or killed when a @value{GDBN} session is ended with
31512 the @code{quit} command.
31517 The remote server attached to an existing process.
31519 The remote server created a new process.
31521 A badly formed request or an error was encountered.
31526 @node Architecture-Specific Protocol Details
31527 @section Architecture-Specific Protocol Details
31529 This section describes how the remote protocol is applied to specific
31530 target architectures. Also see @ref{Standard Target Features}, for
31531 details of XML target descriptions for each architecture.
31535 @subsubsection Breakpoint Kinds
31537 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
31542 16-bit Thumb mode breakpoint.
31545 32-bit Thumb mode (Thumb-2) breakpoint.
31548 32-bit ARM mode breakpoint.
31554 @subsubsection Register Packet Format
31556 The following @code{g}/@code{G} packets have previously been defined.
31557 In the below, some thirty-two bit registers are transferred as
31558 sixty-four bits. Those registers should be zero/sign extended (which?)
31559 to fill the space allocated. Register bytes are transferred in target
31560 byte order. The two nibbles within a register byte are transferred
31561 most-significant - least-significant.
31567 All registers are transferred as thirty-two bit quantities in the order:
31568 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
31569 registers; fsr; fir; fp.
31573 All registers are transferred as sixty-four bit quantities (including
31574 thirty-two bit registers such as @code{sr}). The ordering is the same
31579 @node Tracepoint Packets
31580 @section Tracepoint Packets
31581 @cindex tracepoint packets
31582 @cindex packets, tracepoint
31584 Here we describe the packets @value{GDBN} uses to implement
31585 tracepoints (@pxref{Tracepoints}).
31589 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
31590 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
31591 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
31592 the tracepoint is disabled. @var{step} is the tracepoint's step
31593 count, and @var{pass} is its pass count. If an @samp{F} is present,
31594 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
31595 the number of bytes that the target should copy elsewhere to make room
31596 for the tracepoint. If an @samp{X} is present, it introduces a
31597 tracepoint condition, which consists of a hexadecimal length, followed
31598 by a comma and hex-encoded bytes, in a manner similar to action
31599 encodings as described below. If the trailing @samp{-} is present,
31600 further @samp{QTDP} packets will follow to specify this tracepoint's
31606 The packet was understood and carried out.
31608 The packet was not recognized.
31611 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
31612 Define actions to be taken when a tracepoint is hit. @var{n} and
31613 @var{addr} must be the same as in the initial @samp{QTDP} packet for
31614 this tracepoint. This packet may only be sent immediately after
31615 another @samp{QTDP} packet that ended with a @samp{-}. If the
31616 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
31617 specifying more actions for this tracepoint.
31619 In the series of action packets for a given tracepoint, at most one
31620 can have an @samp{S} before its first @var{action}. If such a packet
31621 is sent, it and the following packets define ``while-stepping''
31622 actions. Any prior packets define ordinary actions --- that is, those
31623 taken when the tracepoint is first hit. If no action packet has an
31624 @samp{S}, then all the packets in the series specify ordinary
31625 tracepoint actions.
31627 The @samp{@var{action}@dots{}} portion of the packet is a series of
31628 actions, concatenated without separators. Each action has one of the
31634 Collect the registers whose bits are set in @var{mask}. @var{mask} is
31635 a hexadecimal number whose @var{i}'th bit is set if register number
31636 @var{i} should be collected. (The least significant bit is numbered
31637 zero.) Note that @var{mask} may be any number of digits long; it may
31638 not fit in a 32-bit word.
31640 @item M @var{basereg},@var{offset},@var{len}
31641 Collect @var{len} bytes of memory starting at the address in register
31642 number @var{basereg}, plus @var{offset}. If @var{basereg} is
31643 @samp{-1}, then the range has a fixed address: @var{offset} is the
31644 address of the lowest byte to collect. The @var{basereg},
31645 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
31646 values (the @samp{-1} value for @var{basereg} is a special case).
31648 @item X @var{len},@var{expr}
31649 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
31650 it directs. @var{expr} is an agent expression, as described in
31651 @ref{Agent Expressions}. Each byte of the expression is encoded as a
31652 two-digit hex number in the packet; @var{len} is the number of bytes
31653 in the expression (and thus one-half the number of hex digits in the
31658 Any number of actions may be packed together in a single @samp{QTDP}
31659 packet, as long as the packet does not exceed the maximum packet
31660 length (400 bytes, for many stubs). There may be only one @samp{R}
31661 action per tracepoint, and it must precede any @samp{M} or @samp{X}
31662 actions. Any registers referred to by @samp{M} and @samp{X} actions
31663 must be collected by a preceding @samp{R} action. (The
31664 ``while-stepping'' actions are treated as if they were attached to a
31665 separate tracepoint, as far as these restrictions are concerned.)
31670 The packet was understood and carried out.
31672 The packet was not recognized.
31675 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
31676 @cindex @samp{QTDPsrc} packet
31677 Specify a source string of tracepoint @var{n} at address @var{addr}.
31678 This is useful to get accurate reproduction of the tracepoints
31679 originally downloaded at the beginning of the trace run. @var{type}
31680 is the name of the tracepoint part, such as @samp{cond} for the
31681 tracepoint's conditional expression (see below for a list of types), while
31682 @var{bytes} is the string, encoded in hexadecimal.
31684 @var{start} is the offset of the @var{bytes} within the overall source
31685 string, while @var{slen} is the total length of the source string.
31686 This is intended for handling source strings that are longer than will
31687 fit in a single packet.
31688 @c Add detailed example when this info is moved into a dedicated
31689 @c tracepoint descriptions section.
31691 The available string types are @samp{at} for the location,
31692 @samp{cond} for the conditional, and @samp{cmd} for an action command.
31693 @value{GDBN} sends a separate packet for each command in the action
31694 list, in the same order in which the commands are stored in the list.
31696 The target does not need to do anything with source strings except
31697 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
31700 Although this packet is optional, and @value{GDBN} will only send it
31701 if the target replies with @samp{TracepointSource} @xref{General
31702 Query Packets}, it makes both disconnected tracing and trace files
31703 much easier to use. Otherwise the user must be careful that the
31704 tracepoints in effect while looking at trace frames are identical to
31705 the ones in effect during the trace run; even a small discrepancy
31706 could cause @samp{tdump} not to work, or a particular trace frame not
31709 @item QTDV:@var{n}:@var{value}
31710 @cindex define trace state variable, remote request
31711 @cindex @samp{QTDV} packet
31712 Create a new trace state variable, number @var{n}, with an initial
31713 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
31714 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
31715 the option of not using this packet for initial values of zero; the
31716 target should simply create the trace state variables as they are
31717 mentioned in expressions.
31719 @item QTFrame:@var{n}
31720 Select the @var{n}'th tracepoint frame from the buffer, and use the
31721 register and memory contents recorded there to answer subsequent
31722 request packets from @value{GDBN}.
31724 A successful reply from the stub indicates that the stub has found the
31725 requested frame. The response is a series of parts, concatenated
31726 without separators, describing the frame we selected. Each part has
31727 one of the following forms:
31731 The selected frame is number @var{n} in the trace frame buffer;
31732 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
31733 was no frame matching the criteria in the request packet.
31736 The selected trace frame records a hit of tracepoint number @var{t};
31737 @var{t} is a hexadecimal number.
31741 @item QTFrame:pc:@var{addr}
31742 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
31743 currently selected frame whose PC is @var{addr};
31744 @var{addr} is a hexadecimal number.
31746 @item QTFrame:tdp:@var{t}
31747 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
31748 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
31749 is a hexadecimal number.
31751 @item QTFrame:range:@var{start}:@var{end}
31752 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
31753 currently selected frame whose PC is between @var{start} (inclusive)
31754 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
31757 @item QTFrame:outside:@var{start}:@var{end}
31758 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
31759 frame @emph{outside} the given range of addresses (exclusive).
31762 Begin the tracepoint experiment. Begin collecting data from tracepoint
31763 hits in the trace frame buffer.
31766 End the tracepoint experiment. Stop collecting trace frames.
31769 Clear the table of tracepoints, and empty the trace frame buffer.
31771 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
31772 Establish the given ranges of memory as ``transparent''. The stub
31773 will answer requests for these ranges from memory's current contents,
31774 if they were not collected as part of the tracepoint hit.
31776 @value{GDBN} uses this to mark read-only regions of memory, like those
31777 containing program code. Since these areas never change, they should
31778 still have the same contents they did when the tracepoint was hit, so
31779 there's no reason for the stub to refuse to provide their contents.
31781 @item QTDisconnected:@var{value}
31782 Set the choice to what to do with the tracing run when @value{GDBN}
31783 disconnects from the target. A @var{value} of 1 directs the target to
31784 continue the tracing run, while 0 tells the target to stop tracing if
31785 @value{GDBN} is no longer in the picture.
31788 Ask the stub if there is a trace experiment running right now.
31790 The reply has the form:
31794 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
31795 @var{running} is a single digit @code{1} if the trace is presently
31796 running, or @code{0} if not. It is followed by semicolon-separated
31797 optional fields that an agent may use to report additional status.
31801 If the trace is not running, the agent may report any of several
31802 explanations as one of the optional fields:
31807 No trace has been run yet.
31810 The trace was stopped by a user-originated stop command.
31813 The trace stopped because the trace buffer filled up.
31815 @item tdisconnected:0
31816 The trace stopped because @value{GDBN} disconnected from the target.
31818 @item tpasscount:@var{tpnum}
31819 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
31821 @item terror:@var{text}:@var{tpnum}
31822 The trace stopped because tracepoint @var{tpnum} had an error. The
31823 string @var{text} is available to describe the nature of the error
31824 (for instance, a divide by zero in the condition expression).
31825 @var{text} is hex encoded.
31828 The trace stopped for some other reason.
31832 Additional optional fields supply statistical and other information.
31833 Although not required, they are extremely useful for users monitoring
31834 the progress of a trace run. If a trace has stopped, and these
31835 numbers are reported, they must reflect the state of the just-stopped
31840 @item tframes:@var{n}
31841 The number of trace frames in the buffer.
31843 @item tcreated:@var{n}
31844 The total number of trace frames created during the run. This may
31845 be larger than the trace frame count, if the buffer is circular.
31847 @item tsize:@var{n}
31848 The total size of the trace buffer, in bytes.
31850 @item tfree:@var{n}
31851 The number of bytes still unused in the buffer.
31853 @item circular:@var{n}
31854 The value of the circular trace buffer flag. @code{1} means that the
31855 trace buffer is circular and old trace frames will be discarded if
31856 necessary to make room, @code{0} means that the trace buffer is linear
31859 @item disconn:@var{n}
31860 The value of the disconnected tracing flag. @code{1} means that
31861 tracing will continue after @value{GDBN} disconnects, @code{0} means
31862 that the trace run will stop.
31866 @item qTV:@var{var}
31867 @cindex trace state variable value, remote request
31868 @cindex @samp{qTV} packet
31869 Ask the stub for the value of the trace state variable number @var{var}.
31874 The value of the variable is @var{value}. This will be the current
31875 value of the variable if the user is examining a running target, or a
31876 saved value if the variable was collected in the trace frame that the
31877 user is looking at. Note that multiple requests may result in
31878 different reply values, such as when requesting values while the
31879 program is running.
31882 The value of the variable is unknown. This would occur, for example,
31883 if the user is examining a trace frame in which the requested variable
31889 These packets request data about tracepoints that are being used by
31890 the target. @value{GDBN} sends @code{qTfP} to get the first piece
31891 of data, and multiple @code{qTsP} to get additional pieces. Replies
31892 to these packets generally take the form of the @code{QTDP} packets
31893 that define tracepoints. (FIXME add detailed syntax)
31897 These packets request data about trace state variables that are on the
31898 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
31899 and multiple @code{qTsV} to get additional variables. Replies to
31900 these packets follow the syntax of the @code{QTDV} packets that define
31901 trace state variables.
31903 @item QTSave:@var{filename}
31904 This packet directs the target to save trace data to the file name
31905 @var{filename} in the target's filesystem. @var{filename} is encoded
31906 as a hex string; the interpretation of the file name (relative vs
31907 absolute, wild cards, etc) is up to the target.
31909 @item qTBuffer:@var{offset},@var{len}
31910 Return up to @var{len} bytes of the current contents of trace buffer,
31911 starting at @var{offset}. The trace buffer is treated as if it were
31912 a contiguous collection of traceframes, as per the trace file format.
31913 The reply consists as many hex-encoded bytes as the target can deliver
31914 in a packet; it is not an error to return fewer than were asked for.
31915 A reply consisting of just @code{l} indicates that no bytes are
31918 @item QTBuffer:circular:@var{value}
31919 This packet directs the target to use a circular trace buffer if
31920 @var{value} is 1, or a linear buffer if the value is 0.
31924 @node Host I/O Packets
31925 @section Host I/O Packets
31926 @cindex Host I/O, remote protocol
31927 @cindex file transfer, remote protocol
31929 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
31930 operations on the far side of a remote link. For example, Host I/O is
31931 used to upload and download files to a remote target with its own
31932 filesystem. Host I/O uses the same constant values and data structure
31933 layout as the target-initiated File-I/O protocol. However, the
31934 Host I/O packets are structured differently. The target-initiated
31935 protocol relies on target memory to store parameters and buffers.
31936 Host I/O requests are initiated by @value{GDBN}, and the
31937 target's memory is not involved. @xref{File-I/O Remote Protocol
31938 Extension}, for more details on the target-initiated protocol.
31940 The Host I/O request packets all encode a single operation along with
31941 its arguments. They have this format:
31945 @item vFile:@var{operation}: @var{parameter}@dots{}
31946 @var{operation} is the name of the particular request; the target
31947 should compare the entire packet name up to the second colon when checking
31948 for a supported operation. The format of @var{parameter} depends on
31949 the operation. Numbers are always passed in hexadecimal. Negative
31950 numbers have an explicit minus sign (i.e.@: two's complement is not
31951 used). Strings (e.g.@: filenames) are encoded as a series of
31952 hexadecimal bytes. The last argument to a system call may be a
31953 buffer of escaped binary data (@pxref{Binary Data}).
31957 The valid responses to Host I/O packets are:
31961 @item F @var{result} [, @var{errno}] [; @var{attachment}]
31962 @var{result} is the integer value returned by this operation, usually
31963 non-negative for success and -1 for errors. If an error has occured,
31964 @var{errno} will be included in the result. @var{errno} will have a
31965 value defined by the File-I/O protocol (@pxref{Errno Values}). For
31966 operations which return data, @var{attachment} supplies the data as a
31967 binary buffer. Binary buffers in response packets are escaped in the
31968 normal way (@pxref{Binary Data}). See the individual packet
31969 documentation for the interpretation of @var{result} and
31973 An empty response indicates that this operation is not recognized.
31977 These are the supported Host I/O operations:
31980 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
31981 Open a file at @var{pathname} and return a file descriptor for it, or
31982 return -1 if an error occurs. @var{pathname} is a string,
31983 @var{flags} is an integer indicating a mask of open flags
31984 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
31985 of mode bits to use if the file is created (@pxref{mode_t Values}).
31986 @xref{open}, for details of the open flags and mode values.
31988 @item vFile:close: @var{fd}
31989 Close the open file corresponding to @var{fd} and return 0, or
31990 -1 if an error occurs.
31992 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
31993 Read data from the open file corresponding to @var{fd}. Up to
31994 @var{count} bytes will be read from the file, starting at @var{offset}
31995 relative to the start of the file. The target may read fewer bytes;
31996 common reasons include packet size limits and an end-of-file
31997 condition. The number of bytes read is returned. Zero should only be
31998 returned for a successful read at the end of the file, or if
31999 @var{count} was zero.
32001 The data read should be returned as a binary attachment on success.
32002 If zero bytes were read, the response should include an empty binary
32003 attachment (i.e.@: a trailing semicolon). The return value is the
32004 number of target bytes read; the binary attachment may be longer if
32005 some characters were escaped.
32007 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
32008 Write @var{data} (a binary buffer) to the open file corresponding
32009 to @var{fd}. Start the write at @var{offset} from the start of the
32010 file. Unlike many @code{write} system calls, there is no
32011 separate @var{count} argument; the length of @var{data} in the
32012 packet is used. @samp{vFile:write} returns the number of bytes written,
32013 which may be shorter than the length of @var{data}, or -1 if an
32016 @item vFile:unlink: @var{pathname}
32017 Delete the file at @var{pathname} on the target. Return 0,
32018 or -1 if an error occurs. @var{pathname} is a string.
32023 @section Interrupts
32024 @cindex interrupts (remote protocol)
32026 When a program on the remote target is running, @value{GDBN} may
32027 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
32028 a @code{BREAK} followed by @code{g},
32029 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
32031 The precise meaning of @code{BREAK} is defined by the transport
32032 mechanism and may, in fact, be undefined. @value{GDBN} does not
32033 currently define a @code{BREAK} mechanism for any of the network
32034 interfaces except for TCP, in which case @value{GDBN} sends the
32035 @code{telnet} BREAK sequence.
32037 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
32038 transport mechanisms. It is represented by sending the single byte
32039 @code{0x03} without any of the usual packet overhead described in
32040 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
32041 transmitted as part of a packet, it is considered to be packet data
32042 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
32043 (@pxref{X packet}), used for binary downloads, may include an unescaped
32044 @code{0x03} as part of its packet.
32046 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
32047 When Linux kernel receives this sequence from serial port,
32048 it stops execution and connects to gdb.
32050 Stubs are not required to recognize these interrupt mechanisms and the
32051 precise meaning associated with receipt of the interrupt is
32052 implementation defined. If the target supports debugging of multiple
32053 threads and/or processes, it should attempt to interrupt all
32054 currently-executing threads and processes.
32055 If the stub is successful at interrupting the
32056 running program, it should send one of the stop
32057 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
32058 of successfully stopping the program in all-stop mode, and a stop reply
32059 for each stopped thread in non-stop mode.
32060 Interrupts received while the
32061 program is stopped are discarded.
32063 @node Notification Packets
32064 @section Notification Packets
32065 @cindex notification packets
32066 @cindex packets, notification
32068 The @value{GDBN} remote serial protocol includes @dfn{notifications},
32069 packets that require no acknowledgment. Both the GDB and the stub
32070 may send notifications (although the only notifications defined at
32071 present are sent by the stub). Notifications carry information
32072 without incurring the round-trip latency of an acknowledgment, and so
32073 are useful for low-impact communications where occasional packet loss
32076 A notification packet has the form @samp{% @var{data} #
32077 @var{checksum}}, where @var{data} is the content of the notification,
32078 and @var{checksum} is a checksum of @var{data}, computed and formatted
32079 as for ordinary @value{GDBN} packets. A notification's @var{data}
32080 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
32081 receiving a notification, the recipient sends no @samp{+} or @samp{-}
32082 to acknowledge the notification's receipt or to report its corruption.
32084 Every notification's @var{data} begins with a name, which contains no
32085 colon characters, followed by a colon character.
32087 Recipients should silently ignore corrupted notifications and
32088 notifications they do not understand. Recipients should restart
32089 timeout periods on receipt of a well-formed notification, whether or
32090 not they understand it.
32092 Senders should only send the notifications described here when this
32093 protocol description specifies that they are permitted. In the
32094 future, we may extend the protocol to permit existing notifications in
32095 new contexts; this rule helps older senders avoid confusing newer
32098 (Older versions of @value{GDBN} ignore bytes received until they see
32099 the @samp{$} byte that begins an ordinary packet, so new stubs may
32100 transmit notifications without fear of confusing older clients. There
32101 are no notifications defined for @value{GDBN} to send at the moment, but we
32102 assume that most older stubs would ignore them, as well.)
32104 The following notification packets from the stub to @value{GDBN} are
32108 @item Stop: @var{reply}
32109 Report an asynchronous stop event in non-stop mode.
32110 The @var{reply} has the form of a stop reply, as
32111 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
32112 for information on how these notifications are acknowledged by
32116 @node Remote Non-Stop
32117 @section Remote Protocol Support for Non-Stop Mode
32119 @value{GDBN}'s remote protocol supports non-stop debugging of
32120 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
32121 supports non-stop mode, it should report that to @value{GDBN} by including
32122 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
32124 @value{GDBN} typically sends a @samp{QNonStop} packet only when
32125 establishing a new connection with the stub. Entering non-stop mode
32126 does not alter the state of any currently-running threads, but targets
32127 must stop all threads in any already-attached processes when entering
32128 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
32129 probe the target state after a mode change.
32131 In non-stop mode, when an attached process encounters an event that
32132 would otherwise be reported with a stop reply, it uses the
32133 asynchronous notification mechanism (@pxref{Notification Packets}) to
32134 inform @value{GDBN}. In contrast to all-stop mode, where all threads
32135 in all processes are stopped when a stop reply is sent, in non-stop
32136 mode only the thread reporting the stop event is stopped. That is,
32137 when reporting a @samp{S} or @samp{T} response to indicate completion
32138 of a step operation, hitting a breakpoint, or a fault, only the
32139 affected thread is stopped; any other still-running threads continue
32140 to run. When reporting a @samp{W} or @samp{X} response, all running
32141 threads belonging to other attached processes continue to run.
32143 Only one stop reply notification at a time may be pending; if
32144 additional stop events occur before @value{GDBN} has acknowledged the
32145 previous notification, they must be queued by the stub for later
32146 synchronous transmission in response to @samp{vStopped} packets from
32147 @value{GDBN}. Because the notification mechanism is unreliable,
32148 the stub is permitted to resend a stop reply notification
32149 if it believes @value{GDBN} may not have received it. @value{GDBN}
32150 ignores additional stop reply notifications received before it has
32151 finished processing a previous notification and the stub has completed
32152 sending any queued stop events.
32154 Otherwise, @value{GDBN} must be prepared to receive a stop reply
32155 notification at any time. Specifically, they may appear when
32156 @value{GDBN} is not otherwise reading input from the stub, or when
32157 @value{GDBN} is expecting to read a normal synchronous response or a
32158 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
32159 Notification packets are distinct from any other communication from
32160 the stub so there is no ambiguity.
32162 After receiving a stop reply notification, @value{GDBN} shall
32163 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
32164 as a regular, synchronous request to the stub. Such acknowledgment
32165 is not required to happen immediately, as @value{GDBN} is permitted to
32166 send other, unrelated packets to the stub first, which the stub should
32169 Upon receiving a @samp{vStopped} packet, if the stub has other queued
32170 stop events to report to @value{GDBN}, it shall respond by sending a
32171 normal stop reply response. @value{GDBN} shall then send another
32172 @samp{vStopped} packet to solicit further responses; again, it is
32173 permitted to send other, unrelated packets as well which the stub
32174 should process normally.
32176 If the stub receives a @samp{vStopped} packet and there are no
32177 additional stop events to report, the stub shall return an @samp{OK}
32178 response. At this point, if further stop events occur, the stub shall
32179 send a new stop reply notification, @value{GDBN} shall accept the
32180 notification, and the process shall be repeated.
32182 In non-stop mode, the target shall respond to the @samp{?} packet as
32183 follows. First, any incomplete stop reply notification/@samp{vStopped}
32184 sequence in progress is abandoned. The target must begin a new
32185 sequence reporting stop events for all stopped threads, whether or not
32186 it has previously reported those events to @value{GDBN}. The first
32187 stop reply is sent as a synchronous reply to the @samp{?} packet, and
32188 subsequent stop replies are sent as responses to @samp{vStopped} packets
32189 using the mechanism described above. The target must not send
32190 asynchronous stop reply notifications until the sequence is complete.
32191 If all threads are running when the target receives the @samp{?} packet,
32192 or if the target is not attached to any process, it shall respond
32195 @node Packet Acknowledgment
32196 @section Packet Acknowledgment
32198 @cindex acknowledgment, for @value{GDBN} remote
32199 @cindex packet acknowledgment, for @value{GDBN} remote
32200 By default, when either the host or the target machine receives a packet,
32201 the first response expected is an acknowledgment: either @samp{+} (to indicate
32202 the package was received correctly) or @samp{-} (to request retransmission).
32203 This mechanism allows the @value{GDBN} remote protocol to operate over
32204 unreliable transport mechanisms, such as a serial line.
32206 In cases where the transport mechanism is itself reliable (such as a pipe or
32207 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
32208 It may be desirable to disable them in that case to reduce communication
32209 overhead, or for other reasons. This can be accomplished by means of the
32210 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
32212 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
32213 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
32214 and response format still includes the normal checksum, as described in
32215 @ref{Overview}, but the checksum may be ignored by the receiver.
32217 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
32218 no-acknowledgment mode, it should report that to @value{GDBN}
32219 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
32220 @pxref{qSupported}.
32221 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
32222 disabled via the @code{set remote noack-packet off} command
32223 (@pxref{Remote Configuration}),
32224 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
32225 Only then may the stub actually turn off packet acknowledgments.
32226 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
32227 response, which can be safely ignored by the stub.
32229 Note that @code{set remote noack-packet} command only affects negotiation
32230 between @value{GDBN} and the stub when subsequent connections are made;
32231 it does not affect the protocol acknowledgment state for any current
32233 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
32234 new connection is established,
32235 there is also no protocol request to re-enable the acknowledgments
32236 for the current connection, once disabled.
32241 Example sequence of a target being re-started. Notice how the restart
32242 does not get any direct output:
32247 @emph{target restarts}
32250 <- @code{T001:1234123412341234}
32254 Example sequence of a target being stepped by a single instruction:
32257 -> @code{G1445@dots{}}
32262 <- @code{T001:1234123412341234}
32266 <- @code{1455@dots{}}
32270 @node File-I/O Remote Protocol Extension
32271 @section File-I/O Remote Protocol Extension
32272 @cindex File-I/O remote protocol extension
32275 * File-I/O Overview::
32276 * Protocol Basics::
32277 * The F Request Packet::
32278 * The F Reply Packet::
32279 * The Ctrl-C Message::
32281 * List of Supported Calls::
32282 * Protocol-specific Representation of Datatypes::
32284 * File-I/O Examples::
32287 @node File-I/O Overview
32288 @subsection File-I/O Overview
32289 @cindex file-i/o overview
32291 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
32292 target to use the host's file system and console I/O to perform various
32293 system calls. System calls on the target system are translated into a
32294 remote protocol packet to the host system, which then performs the needed
32295 actions and returns a response packet to the target system.
32296 This simulates file system operations even on targets that lack file systems.
32298 The protocol is defined to be independent of both the host and target systems.
32299 It uses its own internal representation of datatypes and values. Both
32300 @value{GDBN} and the target's @value{GDBN} stub are responsible for
32301 translating the system-dependent value representations into the internal
32302 protocol representations when data is transmitted.
32304 The communication is synchronous. A system call is possible only when
32305 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
32306 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
32307 the target is stopped to allow deterministic access to the target's
32308 memory. Therefore File-I/O is not interruptible by target signals. On
32309 the other hand, it is possible to interrupt File-I/O by a user interrupt
32310 (@samp{Ctrl-C}) within @value{GDBN}.
32312 The target's request to perform a host system call does not finish
32313 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
32314 after finishing the system call, the target returns to continuing the
32315 previous activity (continue, step). No additional continue or step
32316 request from @value{GDBN} is required.
32319 (@value{GDBP}) continue
32320 <- target requests 'system call X'
32321 target is stopped, @value{GDBN} executes system call
32322 -> @value{GDBN} returns result
32323 ... target continues, @value{GDBN} returns to wait for the target
32324 <- target hits breakpoint and sends a Txx packet
32327 The protocol only supports I/O on the console and to regular files on
32328 the host file system. Character or block special devices, pipes,
32329 named pipes, sockets or any other communication method on the host
32330 system are not supported by this protocol.
32332 File I/O is not supported in non-stop mode.
32334 @node Protocol Basics
32335 @subsection Protocol Basics
32336 @cindex protocol basics, file-i/o
32338 The File-I/O protocol uses the @code{F} packet as the request as well
32339 as reply packet. Since a File-I/O system call can only occur when
32340 @value{GDBN} is waiting for a response from the continuing or stepping target,
32341 the File-I/O request is a reply that @value{GDBN} has to expect as a result
32342 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
32343 This @code{F} packet contains all information needed to allow @value{GDBN}
32344 to call the appropriate host system call:
32348 A unique identifier for the requested system call.
32351 All parameters to the system call. Pointers are given as addresses
32352 in the target memory address space. Pointers to strings are given as
32353 pointer/length pair. Numerical values are given as they are.
32354 Numerical control flags are given in a protocol-specific representation.
32358 At this point, @value{GDBN} has to perform the following actions.
32362 If the parameters include pointer values to data needed as input to a
32363 system call, @value{GDBN} requests this data from the target with a
32364 standard @code{m} packet request. This additional communication has to be
32365 expected by the target implementation and is handled as any other @code{m}
32369 @value{GDBN} translates all value from protocol representation to host
32370 representation as needed. Datatypes are coerced into the host types.
32373 @value{GDBN} calls the system call.
32376 It then coerces datatypes back to protocol representation.
32379 If the system call is expected to return data in buffer space specified
32380 by pointer parameters to the call, the data is transmitted to the
32381 target using a @code{M} or @code{X} packet. This packet has to be expected
32382 by the target implementation and is handled as any other @code{M} or @code{X}
32387 Eventually @value{GDBN} replies with another @code{F} packet which contains all
32388 necessary information for the target to continue. This at least contains
32395 @code{errno}, if has been changed by the system call.
32402 After having done the needed type and value coercion, the target continues
32403 the latest continue or step action.
32405 @node The F Request Packet
32406 @subsection The @code{F} Request Packet
32407 @cindex file-i/o request packet
32408 @cindex @code{F} request packet
32410 The @code{F} request packet has the following format:
32413 @item F@var{call-id},@var{parameter@dots{}}
32415 @var{call-id} is the identifier to indicate the host system call to be called.
32416 This is just the name of the function.
32418 @var{parameter@dots{}} are the parameters to the system call.
32419 Parameters are hexadecimal integer values, either the actual values in case
32420 of scalar datatypes, pointers to target buffer space in case of compound
32421 datatypes and unspecified memory areas, or pointer/length pairs in case
32422 of string parameters. These are appended to the @var{call-id} as a
32423 comma-delimited list. All values are transmitted in ASCII
32424 string representation, pointer/length pairs separated by a slash.
32430 @node The F Reply Packet
32431 @subsection The @code{F} Reply Packet
32432 @cindex file-i/o reply packet
32433 @cindex @code{F} reply packet
32435 The @code{F} reply packet has the following format:
32439 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
32441 @var{retcode} is the return code of the system call as hexadecimal value.
32443 @var{errno} is the @code{errno} set by the call, in protocol-specific
32445 This parameter can be omitted if the call was successful.
32447 @var{Ctrl-C flag} is only sent if the user requested a break. In this
32448 case, @var{errno} must be sent as well, even if the call was successful.
32449 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
32456 or, if the call was interrupted before the host call has been performed:
32463 assuming 4 is the protocol-specific representation of @code{EINTR}.
32468 @node The Ctrl-C Message
32469 @subsection The @samp{Ctrl-C} Message
32470 @cindex ctrl-c message, in file-i/o protocol
32472 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
32473 reply packet (@pxref{The F Reply Packet}),
32474 the target should behave as if it had
32475 gotten a break message. The meaning for the target is ``system call
32476 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
32477 (as with a break message) and return to @value{GDBN} with a @code{T02}
32480 It's important for the target to know in which
32481 state the system call was interrupted. There are two possible cases:
32485 The system call hasn't been performed on the host yet.
32488 The system call on the host has been finished.
32492 These two states can be distinguished by the target by the value of the
32493 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
32494 call hasn't been performed. This is equivalent to the @code{EINTR} handling
32495 on POSIX systems. In any other case, the target may presume that the
32496 system call has been finished --- successfully or not --- and should behave
32497 as if the break message arrived right after the system call.
32499 @value{GDBN} must behave reliably. If the system call has not been called
32500 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
32501 @code{errno} in the packet. If the system call on the host has been finished
32502 before the user requests a break, the full action must be finished by
32503 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
32504 The @code{F} packet may only be sent when either nothing has happened
32505 or the full action has been completed.
32508 @subsection Console I/O
32509 @cindex console i/o as part of file-i/o
32511 By default and if not explicitly closed by the target system, the file
32512 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
32513 on the @value{GDBN} console is handled as any other file output operation
32514 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
32515 by @value{GDBN} so that after the target read request from file descriptor
32516 0 all following typing is buffered until either one of the following
32521 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
32523 system call is treated as finished.
32526 The user presses @key{RET}. This is treated as end of input with a trailing
32530 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
32531 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
32535 If the user has typed more characters than fit in the buffer given to
32536 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
32537 either another @code{read(0, @dots{})} is requested by the target, or debugging
32538 is stopped at the user's request.
32541 @node List of Supported Calls
32542 @subsection List of Supported Calls
32543 @cindex list of supported file-i/o calls
32560 @unnumberedsubsubsec open
32561 @cindex open, file-i/o system call
32566 int open(const char *pathname, int flags);
32567 int open(const char *pathname, int flags, mode_t mode);
32571 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
32574 @var{flags} is the bitwise @code{OR} of the following values:
32578 If the file does not exist it will be created. The host
32579 rules apply as far as file ownership and time stamps
32583 When used with @code{O_CREAT}, if the file already exists it is
32584 an error and open() fails.
32587 If the file already exists and the open mode allows
32588 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
32589 truncated to zero length.
32592 The file is opened in append mode.
32595 The file is opened for reading only.
32598 The file is opened for writing only.
32601 The file is opened for reading and writing.
32605 Other bits are silently ignored.
32609 @var{mode} is the bitwise @code{OR} of the following values:
32613 User has read permission.
32616 User has write permission.
32619 Group has read permission.
32622 Group has write permission.
32625 Others have read permission.
32628 Others have write permission.
32632 Other bits are silently ignored.
32635 @item Return value:
32636 @code{open} returns the new file descriptor or -1 if an error
32643 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
32646 @var{pathname} refers to a directory.
32649 The requested access is not allowed.
32652 @var{pathname} was too long.
32655 A directory component in @var{pathname} does not exist.
32658 @var{pathname} refers to a device, pipe, named pipe or socket.
32661 @var{pathname} refers to a file on a read-only filesystem and
32662 write access was requested.
32665 @var{pathname} is an invalid pointer value.
32668 No space on device to create the file.
32671 The process already has the maximum number of files open.
32674 The limit on the total number of files open on the system
32678 The call was interrupted by the user.
32684 @unnumberedsubsubsec close
32685 @cindex close, file-i/o system call
32694 @samp{Fclose,@var{fd}}
32696 @item Return value:
32697 @code{close} returns zero on success, or -1 if an error occurred.
32703 @var{fd} isn't a valid open file descriptor.
32706 The call was interrupted by the user.
32712 @unnumberedsubsubsec read
32713 @cindex read, file-i/o system call
32718 int read(int fd, void *buf, unsigned int count);
32722 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
32724 @item Return value:
32725 On success, the number of bytes read is returned.
32726 Zero indicates end of file. If count is zero, read
32727 returns zero as well. On error, -1 is returned.
32733 @var{fd} is not a valid file descriptor or is not open for
32737 @var{bufptr} is an invalid pointer value.
32740 The call was interrupted by the user.
32746 @unnumberedsubsubsec write
32747 @cindex write, file-i/o system call
32752 int write(int fd, const void *buf, unsigned int count);
32756 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
32758 @item Return value:
32759 On success, the number of bytes written are returned.
32760 Zero indicates nothing was written. On error, -1
32767 @var{fd} is not a valid file descriptor or is not open for
32771 @var{bufptr} is an invalid pointer value.
32774 An attempt was made to write a file that exceeds the
32775 host-specific maximum file size allowed.
32778 No space on device to write the data.
32781 The call was interrupted by the user.
32787 @unnumberedsubsubsec lseek
32788 @cindex lseek, file-i/o system call
32793 long lseek (int fd, long offset, int flag);
32797 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
32799 @var{flag} is one of:
32803 The offset is set to @var{offset} bytes.
32806 The offset is set to its current location plus @var{offset}
32810 The offset is set to the size of the file plus @var{offset}
32814 @item Return value:
32815 On success, the resulting unsigned offset in bytes from
32816 the beginning of the file is returned. Otherwise, a
32817 value of -1 is returned.
32823 @var{fd} is not a valid open file descriptor.
32826 @var{fd} is associated with the @value{GDBN} console.
32829 @var{flag} is not a proper value.
32832 The call was interrupted by the user.
32838 @unnumberedsubsubsec rename
32839 @cindex rename, file-i/o system call
32844 int rename(const char *oldpath, const char *newpath);
32848 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
32850 @item Return value:
32851 On success, zero is returned. On error, -1 is returned.
32857 @var{newpath} is an existing directory, but @var{oldpath} is not a
32861 @var{newpath} is a non-empty directory.
32864 @var{oldpath} or @var{newpath} is a directory that is in use by some
32868 An attempt was made to make a directory a subdirectory
32872 A component used as a directory in @var{oldpath} or new
32873 path is not a directory. Or @var{oldpath} is a directory
32874 and @var{newpath} exists but is not a directory.
32877 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
32880 No access to the file or the path of the file.
32884 @var{oldpath} or @var{newpath} was too long.
32887 A directory component in @var{oldpath} or @var{newpath} does not exist.
32890 The file is on a read-only filesystem.
32893 The device containing the file has no room for the new
32897 The call was interrupted by the user.
32903 @unnumberedsubsubsec unlink
32904 @cindex unlink, file-i/o system call
32909 int unlink(const char *pathname);
32913 @samp{Funlink,@var{pathnameptr}/@var{len}}
32915 @item Return value:
32916 On success, zero is returned. On error, -1 is returned.
32922 No access to the file or the path of the file.
32925 The system does not allow unlinking of directories.
32928 The file @var{pathname} cannot be unlinked because it's
32929 being used by another process.
32932 @var{pathnameptr} is an invalid pointer value.
32935 @var{pathname} was too long.
32938 A directory component in @var{pathname} does not exist.
32941 A component of the path is not a directory.
32944 The file is on a read-only filesystem.
32947 The call was interrupted by the user.
32953 @unnumberedsubsubsec stat/fstat
32954 @cindex fstat, file-i/o system call
32955 @cindex stat, file-i/o system call
32960 int stat(const char *pathname, struct stat *buf);
32961 int fstat(int fd, struct stat *buf);
32965 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
32966 @samp{Ffstat,@var{fd},@var{bufptr}}
32968 @item Return value:
32969 On success, zero is returned. On error, -1 is returned.
32975 @var{fd} is not a valid open file.
32978 A directory component in @var{pathname} does not exist or the
32979 path is an empty string.
32982 A component of the path is not a directory.
32985 @var{pathnameptr} is an invalid pointer value.
32988 No access to the file or the path of the file.
32991 @var{pathname} was too long.
32994 The call was interrupted by the user.
33000 @unnumberedsubsubsec gettimeofday
33001 @cindex gettimeofday, file-i/o system call
33006 int gettimeofday(struct timeval *tv, void *tz);
33010 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
33012 @item Return value:
33013 On success, 0 is returned, -1 otherwise.
33019 @var{tz} is a non-NULL pointer.
33022 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
33028 @unnumberedsubsubsec isatty
33029 @cindex isatty, file-i/o system call
33034 int isatty(int fd);
33038 @samp{Fisatty,@var{fd}}
33040 @item Return value:
33041 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
33047 The call was interrupted by the user.
33052 Note that the @code{isatty} call is treated as a special case: it returns
33053 1 to the target if the file descriptor is attached
33054 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
33055 would require implementing @code{ioctl} and would be more complex than
33060 @unnumberedsubsubsec system
33061 @cindex system, file-i/o system call
33066 int system(const char *command);
33070 @samp{Fsystem,@var{commandptr}/@var{len}}
33072 @item Return value:
33073 If @var{len} is zero, the return value indicates whether a shell is
33074 available. A zero return value indicates a shell is not available.
33075 For non-zero @var{len}, the value returned is -1 on error and the
33076 return status of the command otherwise. Only the exit status of the
33077 command is returned, which is extracted from the host's @code{system}
33078 return value by calling @code{WEXITSTATUS(retval)}. In case
33079 @file{/bin/sh} could not be executed, 127 is returned.
33085 The call was interrupted by the user.
33090 @value{GDBN} takes over the full task of calling the necessary host calls
33091 to perform the @code{system} call. The return value of @code{system} on
33092 the host is simplified before it's returned
33093 to the target. Any termination signal information from the child process
33094 is discarded, and the return value consists
33095 entirely of the exit status of the called command.
33097 Due to security concerns, the @code{system} call is by default refused
33098 by @value{GDBN}. The user has to allow this call explicitly with the
33099 @code{set remote system-call-allowed 1} command.
33102 @item set remote system-call-allowed
33103 @kindex set remote system-call-allowed
33104 Control whether to allow the @code{system} calls in the File I/O
33105 protocol for the remote target. The default is zero (disabled).
33107 @item show remote system-call-allowed
33108 @kindex show remote system-call-allowed
33109 Show whether the @code{system} calls are allowed in the File I/O
33113 @node Protocol-specific Representation of Datatypes
33114 @subsection Protocol-specific Representation of Datatypes
33115 @cindex protocol-specific representation of datatypes, in file-i/o protocol
33118 * Integral Datatypes::
33120 * Memory Transfer::
33125 @node Integral Datatypes
33126 @unnumberedsubsubsec Integral Datatypes
33127 @cindex integral datatypes, in file-i/o protocol
33129 The integral datatypes used in the system calls are @code{int},
33130 @code{unsigned int}, @code{long}, @code{unsigned long},
33131 @code{mode_t}, and @code{time_t}.
33133 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
33134 implemented as 32 bit values in this protocol.
33136 @code{long} and @code{unsigned long} are implemented as 64 bit types.
33138 @xref{Limits}, for corresponding MIN and MAX values (similar to those
33139 in @file{limits.h}) to allow range checking on host and target.
33141 @code{time_t} datatypes are defined as seconds since the Epoch.
33143 All integral datatypes transferred as part of a memory read or write of a
33144 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
33147 @node Pointer Values
33148 @unnumberedsubsubsec Pointer Values
33149 @cindex pointer values, in file-i/o protocol
33151 Pointers to target data are transmitted as they are. An exception
33152 is made for pointers to buffers for which the length isn't
33153 transmitted as part of the function call, namely strings. Strings
33154 are transmitted as a pointer/length pair, both as hex values, e.g.@:
33161 which is a pointer to data of length 18 bytes at position 0x1aaf.
33162 The length is defined as the full string length in bytes, including
33163 the trailing null byte. For example, the string @code{"hello world"}
33164 at address 0x123456 is transmitted as
33170 @node Memory Transfer
33171 @unnumberedsubsubsec Memory Transfer
33172 @cindex memory transfer, in file-i/o protocol
33174 Structured data which is transferred using a memory read or write (for
33175 example, a @code{struct stat}) is expected to be in a protocol-specific format
33176 with all scalar multibyte datatypes being big endian. Translation to
33177 this representation needs to be done both by the target before the @code{F}
33178 packet is sent, and by @value{GDBN} before
33179 it transfers memory to the target. Transferred pointers to structured
33180 data should point to the already-coerced data at any time.
33184 @unnumberedsubsubsec struct stat
33185 @cindex struct stat, in file-i/o protocol
33187 The buffer of type @code{struct stat} used by the target and @value{GDBN}
33188 is defined as follows:
33192 unsigned int st_dev; /* device */
33193 unsigned int st_ino; /* inode */
33194 mode_t st_mode; /* protection */
33195 unsigned int st_nlink; /* number of hard links */
33196 unsigned int st_uid; /* user ID of owner */
33197 unsigned int st_gid; /* group ID of owner */
33198 unsigned int st_rdev; /* device type (if inode device) */
33199 unsigned long st_size; /* total size, in bytes */
33200 unsigned long st_blksize; /* blocksize for filesystem I/O */
33201 unsigned long st_blocks; /* number of blocks allocated */
33202 time_t st_atime; /* time of last access */
33203 time_t st_mtime; /* time of last modification */
33204 time_t st_ctime; /* time of last change */
33208 The integral datatypes conform to the definitions given in the
33209 appropriate section (see @ref{Integral Datatypes}, for details) so this
33210 structure is of size 64 bytes.
33212 The values of several fields have a restricted meaning and/or
33218 A value of 0 represents a file, 1 the console.
33221 No valid meaning for the target. Transmitted unchanged.
33224 Valid mode bits are described in @ref{Constants}. Any other
33225 bits have currently no meaning for the target.
33230 No valid meaning for the target. Transmitted unchanged.
33235 These values have a host and file system dependent
33236 accuracy. Especially on Windows hosts, the file system may not
33237 support exact timing values.
33240 The target gets a @code{struct stat} of the above representation and is
33241 responsible for coercing it to the target representation before
33244 Note that due to size differences between the host, target, and protocol
33245 representations of @code{struct stat} members, these members could eventually
33246 get truncated on the target.
33248 @node struct timeval
33249 @unnumberedsubsubsec struct timeval
33250 @cindex struct timeval, in file-i/o protocol
33252 The buffer of type @code{struct timeval} used by the File-I/O protocol
33253 is defined as follows:
33257 time_t tv_sec; /* second */
33258 long tv_usec; /* microsecond */
33262 The integral datatypes conform to the definitions given in the
33263 appropriate section (see @ref{Integral Datatypes}, for details) so this
33264 structure is of size 8 bytes.
33267 @subsection Constants
33268 @cindex constants, in file-i/o protocol
33270 The following values are used for the constants inside of the
33271 protocol. @value{GDBN} and target are responsible for translating these
33272 values before and after the call as needed.
33283 @unnumberedsubsubsec Open Flags
33284 @cindex open flags, in file-i/o protocol
33286 All values are given in hexadecimal representation.
33298 @node mode_t Values
33299 @unnumberedsubsubsec mode_t Values
33300 @cindex mode_t values, in file-i/o protocol
33302 All values are given in octal representation.
33319 @unnumberedsubsubsec Errno Values
33320 @cindex errno values, in file-i/o protocol
33322 All values are given in decimal representation.
33347 @code{EUNKNOWN} is used as a fallback error value if a host system returns
33348 any error value not in the list of supported error numbers.
33351 @unnumberedsubsubsec Lseek Flags
33352 @cindex lseek flags, in file-i/o protocol
33361 @unnumberedsubsubsec Limits
33362 @cindex limits, in file-i/o protocol
33364 All values are given in decimal representation.
33367 INT_MIN -2147483648
33369 UINT_MAX 4294967295
33370 LONG_MIN -9223372036854775808
33371 LONG_MAX 9223372036854775807
33372 ULONG_MAX 18446744073709551615
33375 @node File-I/O Examples
33376 @subsection File-I/O Examples
33377 @cindex file-i/o examples
33379 Example sequence of a write call, file descriptor 3, buffer is at target
33380 address 0x1234, 6 bytes should be written:
33383 <- @code{Fwrite,3,1234,6}
33384 @emph{request memory read from target}
33387 @emph{return "6 bytes written"}
33391 Example sequence of a read call, file descriptor 3, buffer is at target
33392 address 0x1234, 6 bytes should be read:
33395 <- @code{Fread,3,1234,6}
33396 @emph{request memory write to target}
33397 -> @code{X1234,6:XXXXXX}
33398 @emph{return "6 bytes read"}
33402 Example sequence of a read call, call fails on the host due to invalid
33403 file descriptor (@code{EBADF}):
33406 <- @code{Fread,3,1234,6}
33410 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
33414 <- @code{Fread,3,1234,6}
33419 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
33423 <- @code{Fread,3,1234,6}
33424 -> @code{X1234,6:XXXXXX}
33428 @node Library List Format
33429 @section Library List Format
33430 @cindex library list format, remote protocol
33432 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
33433 same process as your application to manage libraries. In this case,
33434 @value{GDBN} can use the loader's symbol table and normal memory
33435 operations to maintain a list of shared libraries. On other
33436 platforms, the operating system manages loaded libraries.
33437 @value{GDBN} can not retrieve the list of currently loaded libraries
33438 through memory operations, so it uses the @samp{qXfer:libraries:read}
33439 packet (@pxref{qXfer library list read}) instead. The remote stub
33440 queries the target's operating system and reports which libraries
33443 The @samp{qXfer:libraries:read} packet returns an XML document which
33444 lists loaded libraries and their offsets. Each library has an
33445 associated name and one or more segment or section base addresses,
33446 which report where the library was loaded in memory.
33448 For the common case of libraries that are fully linked binaries, the
33449 library should have a list of segments. If the target supports
33450 dynamic linking of a relocatable object file, its library XML element
33451 should instead include a list of allocated sections. The segment or
33452 section bases are start addresses, not relocation offsets; they do not
33453 depend on the library's link-time base addresses.
33455 @value{GDBN} must be linked with the Expat library to support XML
33456 library lists. @xref{Expat}.
33458 A simple memory map, with one loaded library relocated by a single
33459 offset, looks like this:
33463 <library name="/lib/libc.so.6">
33464 <segment address="0x10000000"/>
33469 Another simple memory map, with one loaded library with three
33470 allocated sections (.text, .data, .bss), looks like this:
33474 <library name="sharedlib.o">
33475 <section address="0x10000000"/>
33476 <section address="0x20000000"/>
33477 <section address="0x30000000"/>
33482 The format of a library list is described by this DTD:
33485 <!-- library-list: Root element with versioning -->
33486 <!ELEMENT library-list (library)*>
33487 <!ATTLIST library-list version CDATA #FIXED "1.0">
33488 <!ELEMENT library (segment*, section*)>
33489 <!ATTLIST library name CDATA #REQUIRED>
33490 <!ELEMENT segment EMPTY>
33491 <!ATTLIST segment address CDATA #REQUIRED>
33492 <!ELEMENT section EMPTY>
33493 <!ATTLIST section address CDATA #REQUIRED>
33496 In addition, segments and section descriptors cannot be mixed within a
33497 single library element, and you must supply at least one segment or
33498 section for each library.
33500 @node Memory Map Format
33501 @section Memory Map Format
33502 @cindex memory map format
33504 To be able to write into flash memory, @value{GDBN} needs to obtain a
33505 memory map from the target. This section describes the format of the
33508 The memory map is obtained using the @samp{qXfer:memory-map:read}
33509 (@pxref{qXfer memory map read}) packet and is an XML document that
33510 lists memory regions.
33512 @value{GDBN} must be linked with the Expat library to support XML
33513 memory maps. @xref{Expat}.
33515 The top-level structure of the document is shown below:
33518 <?xml version="1.0"?>
33519 <!DOCTYPE memory-map
33520 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
33521 "http://sourceware.org/gdb/gdb-memory-map.dtd">
33527 Each region can be either:
33532 A region of RAM starting at @var{addr} and extending for @var{length}
33536 <memory type="ram" start="@var{addr}" length="@var{length}"/>
33541 A region of read-only memory:
33544 <memory type="rom" start="@var{addr}" length="@var{length}"/>
33549 A region of flash memory, with erasure blocks @var{blocksize}
33553 <memory type="flash" start="@var{addr}" length="@var{length}">
33554 <property name="blocksize">@var{blocksize}</property>
33560 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
33561 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
33562 packets to write to addresses in such ranges.
33564 The formal DTD for memory map format is given below:
33567 <!-- ................................................... -->
33568 <!-- Memory Map XML DTD ................................ -->
33569 <!-- File: memory-map.dtd .............................. -->
33570 <!-- .................................... .............. -->
33571 <!-- memory-map.dtd -->
33572 <!-- memory-map: Root element with versioning -->
33573 <!ELEMENT memory-map (memory | property)>
33574 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
33575 <!ELEMENT memory (property)>
33576 <!-- memory: Specifies a memory region,
33577 and its type, or device. -->
33578 <!ATTLIST memory type CDATA #REQUIRED
33579 start CDATA #REQUIRED
33580 length CDATA #REQUIRED
33581 device CDATA #IMPLIED>
33582 <!-- property: Generic attribute tag -->
33583 <!ELEMENT property (#PCDATA | property)*>
33584 <!ATTLIST property name CDATA #REQUIRED>
33587 @node Thread List Format
33588 @section Thread List Format
33589 @cindex thread list format
33591 To efficiently update the list of threads and their attributes,
33592 @value{GDBN} issues the @samp{qXfer:threads:read} packet
33593 (@pxref{qXfer threads read}) and obtains the XML document with
33594 the following structure:
33597 <?xml version="1.0"?>
33599 <thread id="id" core="0">
33600 ... description ...
33605 Each @samp{thread} element must have the @samp{id} attribute that
33606 identifies the thread (@pxref{thread-id syntax}). The
33607 @samp{core} attribute, if present, specifies which processor core
33608 the thread was last executing on. The content of the of @samp{thread}
33609 element is interpreted as human-readable auxilliary information.
33611 @include agentexpr.texi
33613 @node Trace File Format
33614 @appendix Trace File Format
33615 @cindex trace file format
33617 The trace file comes in three parts: a header, a textual description
33618 section, and a trace frame section with binary data.
33620 The header has the form @code{\x7fTRACE0\n}. The first byte is
33621 @code{0x7f} so as to indicate that the file contains binary data,
33622 while the @code{0} is a version number that may have different values
33625 The description section consists of multiple lines of @sc{ascii} text
33626 separated by newline characters (@code{0xa}). The lines may include a
33627 variety of optional descriptive or context-setting information, such
33628 as tracepoint definitions or register set size. @value{GDBN} will
33629 ignore any line that it does not recognize. An empty line marks the end
33632 @c FIXME add some specific types of data
33634 The trace frame section consists of a number of consecutive frames.
33635 Each frame begins with a two-byte tracepoint number, followed by a
33636 four-byte size giving the amount of data in the frame. The data in
33637 the frame consists of a number of blocks, each introduced by a
33638 character indicating its type (at least register, memory, and trace
33639 state variable). The data in this section is raw binary, not a
33640 hexadecimal or other encoding; its endianness matches the target's
33643 @c FIXME bi-arch may require endianness/arch info in description section
33646 @item R @var{bytes}
33647 Register block. The number and ordering of bytes matches that of a
33648 @code{g} packet in the remote protocol. Note that these are the
33649 actual bytes, in target order and @value{GDBN} register order, not a
33650 hexadecimal encoding.
33652 @item M @var{address} @var{length} @var{bytes}...
33653 Memory block. This is a contiguous block of memory, at the 8-byte
33654 address @var{address}, with a 2-byte length @var{length}, followed by
33655 @var{length} bytes.
33657 @item V @var{number} @var{value}
33658 Trace state variable block. This records the 8-byte signed value
33659 @var{value} of trace state variable numbered @var{number}.
33663 Future enhancements of the trace file format may include additional types
33666 @node Target Descriptions
33667 @appendix Target Descriptions
33668 @cindex target descriptions
33670 @strong{Warning:} target descriptions are still under active development,
33671 and the contents and format may change between @value{GDBN} releases.
33672 The format is expected to stabilize in the future.
33674 One of the challenges of using @value{GDBN} to debug embedded systems
33675 is that there are so many minor variants of each processor
33676 architecture in use. It is common practice for vendors to start with
33677 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
33678 and then make changes to adapt it to a particular market niche. Some
33679 architectures have hundreds of variants, available from dozens of
33680 vendors. This leads to a number of problems:
33684 With so many different customized processors, it is difficult for
33685 the @value{GDBN} maintainers to keep up with the changes.
33687 Since individual variants may have short lifetimes or limited
33688 audiences, it may not be worthwhile to carry information about every
33689 variant in the @value{GDBN} source tree.
33691 When @value{GDBN} does support the architecture of the embedded system
33692 at hand, the task of finding the correct architecture name to give the
33693 @command{set architecture} command can be error-prone.
33696 To address these problems, the @value{GDBN} remote protocol allows a
33697 target system to not only identify itself to @value{GDBN}, but to
33698 actually describe its own features. This lets @value{GDBN} support
33699 processor variants it has never seen before --- to the extent that the
33700 descriptions are accurate, and that @value{GDBN} understands them.
33702 @value{GDBN} must be linked with the Expat library to support XML
33703 target descriptions. @xref{Expat}.
33706 * Retrieving Descriptions:: How descriptions are fetched from a target.
33707 * Target Description Format:: The contents of a target description.
33708 * Predefined Target Types:: Standard types available for target
33710 * Standard Target Features:: Features @value{GDBN} knows about.
33713 @node Retrieving Descriptions
33714 @section Retrieving Descriptions
33716 Target descriptions can be read from the target automatically, or
33717 specified by the user manually. The default behavior is to read the
33718 description from the target. @value{GDBN} retrieves it via the remote
33719 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
33720 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
33721 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
33722 XML document, of the form described in @ref{Target Description
33725 Alternatively, you can specify a file to read for the target description.
33726 If a file is set, the target will not be queried. The commands to
33727 specify a file are:
33730 @cindex set tdesc filename
33731 @item set tdesc filename @var{path}
33732 Read the target description from @var{path}.
33734 @cindex unset tdesc filename
33735 @item unset tdesc filename
33736 Do not read the XML target description from a file. @value{GDBN}
33737 will use the description supplied by the current target.
33739 @cindex show tdesc filename
33740 @item show tdesc filename
33741 Show the filename to read for a target description, if any.
33745 @node Target Description Format
33746 @section Target Description Format
33747 @cindex target descriptions, XML format
33749 A target description annex is an @uref{http://www.w3.org/XML/, XML}
33750 document which complies with the Document Type Definition provided in
33751 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
33752 means you can use generally available tools like @command{xmllint} to
33753 check that your feature descriptions are well-formed and valid.
33754 However, to help people unfamiliar with XML write descriptions for
33755 their targets, we also describe the grammar here.
33757 Target descriptions can identify the architecture of the remote target
33758 and (for some architectures) provide information about custom register
33759 sets. They can also identify the OS ABI of the remote target.
33760 @value{GDBN} can use this information to autoconfigure for your
33761 target, or to warn you if you connect to an unsupported target.
33763 Here is a simple target description:
33766 <target version="1.0">
33767 <architecture>i386:x86-64</architecture>
33772 This minimal description only says that the target uses
33773 the x86-64 architecture.
33775 A target description has the following overall form, with [ ] marking
33776 optional elements and @dots{} marking repeatable elements. The elements
33777 are explained further below.
33780 <?xml version="1.0"?>
33781 <!DOCTYPE target SYSTEM "gdb-target.dtd">
33782 <target version="1.0">
33783 @r{[}@var{architecture}@r{]}
33784 @r{[}@var{osabi}@r{]}
33785 @r{[}@var{compatible}@r{]}
33786 @r{[}@var{feature}@dots{}@r{]}
33791 The description is generally insensitive to whitespace and line
33792 breaks, under the usual common-sense rules. The XML version
33793 declaration and document type declaration can generally be omitted
33794 (@value{GDBN} does not require them), but specifying them may be
33795 useful for XML validation tools. The @samp{version} attribute for
33796 @samp{<target>} may also be omitted, but we recommend
33797 including it; if future versions of @value{GDBN} use an incompatible
33798 revision of @file{gdb-target.dtd}, they will detect and report
33799 the version mismatch.
33801 @subsection Inclusion
33802 @cindex target descriptions, inclusion
33805 @cindex <xi:include>
33808 It can sometimes be valuable to split a target description up into
33809 several different annexes, either for organizational purposes, or to
33810 share files between different possible target descriptions. You can
33811 divide a description into multiple files by replacing any element of
33812 the target description with an inclusion directive of the form:
33815 <xi:include href="@var{document}"/>
33819 When @value{GDBN} encounters an element of this form, it will retrieve
33820 the named XML @var{document}, and replace the inclusion directive with
33821 the contents of that document. If the current description was read
33822 using @samp{qXfer}, then so will be the included document;
33823 @var{document} will be interpreted as the name of an annex. If the
33824 current description was read from a file, @value{GDBN} will look for
33825 @var{document} as a file in the same directory where it found the
33826 original description.
33828 @subsection Architecture
33829 @cindex <architecture>
33831 An @samp{<architecture>} element has this form:
33834 <architecture>@var{arch}</architecture>
33837 @var{arch} is one of the architectures from the set accepted by
33838 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
33841 @cindex @code{<osabi>}
33843 This optional field was introduced in @value{GDBN} version 7.0.
33844 Previous versions of @value{GDBN} ignore it.
33846 An @samp{<osabi>} element has this form:
33849 <osabi>@var{abi-name}</osabi>
33852 @var{abi-name} is an OS ABI name from the same selection accepted by
33853 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
33855 @subsection Compatible Architecture
33856 @cindex @code{<compatible>}
33858 This optional field was introduced in @value{GDBN} version 7.0.
33859 Previous versions of @value{GDBN} ignore it.
33861 A @samp{<compatible>} element has this form:
33864 <compatible>@var{arch}</compatible>
33867 @var{arch} is one of the architectures from the set accepted by
33868 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
33870 A @samp{<compatible>} element is used to specify that the target
33871 is able to run binaries in some other than the main target architecture
33872 given by the @samp{<architecture>} element. For example, on the
33873 Cell Broadband Engine, the main architecture is @code{powerpc:common}
33874 or @code{powerpc:common64}, but the system is able to run binaries
33875 in the @code{spu} architecture as well. The way to describe this
33876 capability with @samp{<compatible>} is as follows:
33879 <architecture>powerpc:common</architecture>
33880 <compatible>spu</compatible>
33883 @subsection Features
33886 Each @samp{<feature>} describes some logical portion of the target
33887 system. Features are currently used to describe available CPU
33888 registers and the types of their contents. A @samp{<feature>} element
33892 <feature name="@var{name}">
33893 @r{[}@var{type}@dots{}@r{]}
33899 Each feature's name should be unique within the description. The name
33900 of a feature does not matter unless @value{GDBN} has some special
33901 knowledge of the contents of that feature; if it does, the feature
33902 should have its standard name. @xref{Standard Target Features}.
33906 Any register's value is a collection of bits which @value{GDBN} must
33907 interpret. The default interpretation is a two's complement integer,
33908 but other types can be requested by name in the register description.
33909 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
33910 Target Types}), and the description can define additional composite types.
33912 Each type element must have an @samp{id} attribute, which gives
33913 a unique (within the containing @samp{<feature>}) name to the type.
33914 Types must be defined before they are used.
33917 Some targets offer vector registers, which can be treated as arrays
33918 of scalar elements. These types are written as @samp{<vector>} elements,
33919 specifying the array element type, @var{type}, and the number of elements,
33923 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
33927 If a register's value is usefully viewed in multiple ways, define it
33928 with a union type containing the useful representations. The
33929 @samp{<union>} element contains one or more @samp{<field>} elements,
33930 each of which has a @var{name} and a @var{type}:
33933 <union id="@var{id}">
33934 <field name="@var{name}" type="@var{type}"/>
33940 If a register's value is composed from several separate values, define
33941 it with a structure type. There are two forms of the @samp{<struct>}
33942 element; a @samp{<struct>} element must either contain only bitfields
33943 or contain no bitfields. If the structure contains only bitfields,
33944 its total size in bytes must be specified, each bitfield must have an
33945 explicit start and end, and bitfields are automatically assigned an
33946 integer type. The field's @var{start} should be less than or
33947 equal to its @var{end}, and zero represents the least significant bit.
33950 <struct id="@var{id}" size="@var{size}">
33951 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
33956 If the structure contains no bitfields, then each field has an
33957 explicit type, and no implicit padding is added.
33960 <struct id="@var{id}">
33961 <field name="@var{name}" type="@var{type}"/>
33967 If a register's value is a series of single-bit flags, define it with
33968 a flags type. The @samp{<flags>} element has an explicit @var{size}
33969 and contains one or more @samp{<field>} elements. Each field has a
33970 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
33974 <flags id="@var{id}" size="@var{size}">
33975 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
33980 @subsection Registers
33983 Each register is represented as an element with this form:
33986 <reg name="@var{name}"
33987 bitsize="@var{size}"
33988 @r{[}regnum="@var{num}"@r{]}
33989 @r{[}save-restore="@var{save-restore}"@r{]}
33990 @r{[}type="@var{type}"@r{]}
33991 @r{[}group="@var{group}"@r{]}/>
33995 The components are as follows:
34000 The register's name; it must be unique within the target description.
34003 The register's size, in bits.
34006 The register's number. If omitted, a register's number is one greater
34007 than that of the previous register (either in the current feature or in
34008 a preceeding feature); the first register in the target description
34009 defaults to zero. This register number is used to read or write
34010 the register; e.g.@: it is used in the remote @code{p} and @code{P}
34011 packets, and registers appear in the @code{g} and @code{G} packets
34012 in order of increasing register number.
34015 Whether the register should be preserved across inferior function
34016 calls; this must be either @code{yes} or @code{no}. The default is
34017 @code{yes}, which is appropriate for most registers except for
34018 some system control registers; this is not related to the target's
34022 The type of the register. @var{type} may be a predefined type, a type
34023 defined in the current feature, or one of the special types @code{int}
34024 and @code{float}. @code{int} is an integer type of the correct size
34025 for @var{bitsize}, and @code{float} is a floating point type (in the
34026 architecture's normal floating point format) of the correct size for
34027 @var{bitsize}. The default is @code{int}.
34030 The register group to which this register belongs. @var{group} must
34031 be either @code{general}, @code{float}, or @code{vector}. If no
34032 @var{group} is specified, @value{GDBN} will not display the register
34033 in @code{info registers}.
34037 @node Predefined Target Types
34038 @section Predefined Target Types
34039 @cindex target descriptions, predefined types
34041 Type definitions in the self-description can build up composite types
34042 from basic building blocks, but can not define fundamental types. Instead,
34043 standard identifiers are provided by @value{GDBN} for the fundamental
34044 types. The currently supported types are:
34053 Signed integer types holding the specified number of bits.
34060 Unsigned integer types holding the specified number of bits.
34064 Pointers to unspecified code and data. The program counter and
34065 any dedicated return address register may be marked as code
34066 pointers; printing a code pointer converts it into a symbolic
34067 address. The stack pointer and any dedicated address registers
34068 may be marked as data pointers.
34071 Single precision IEEE floating point.
34074 Double precision IEEE floating point.
34077 The 12-byte extended precision format used by ARM FPA registers.
34080 The 10-byte extended precision format used by x87 registers.
34083 32bit @sc{eflags} register used by x86.
34086 32bit @sc{mxcsr} register used by x86.
34090 @node Standard Target Features
34091 @section Standard Target Features
34092 @cindex target descriptions, standard features
34094 A target description must contain either no registers or all the
34095 target's registers. If the description contains no registers, then
34096 @value{GDBN} will assume a default register layout, selected based on
34097 the architecture. If the description contains any registers, the
34098 default layout will not be used; the standard registers must be
34099 described in the target description, in such a way that @value{GDBN}
34100 can recognize them.
34102 This is accomplished by giving specific names to feature elements
34103 which contain standard registers. @value{GDBN} will look for features
34104 with those names and verify that they contain the expected registers;
34105 if any known feature is missing required registers, or if any required
34106 feature is missing, @value{GDBN} will reject the target
34107 description. You can add additional registers to any of the
34108 standard features --- @value{GDBN} will display them just as if
34109 they were added to an unrecognized feature.
34111 This section lists the known features and their expected contents.
34112 Sample XML documents for these features are included in the
34113 @value{GDBN} source tree, in the directory @file{gdb/features}.
34115 Names recognized by @value{GDBN} should include the name of the
34116 company or organization which selected the name, and the overall
34117 architecture to which the feature applies; so e.g.@: the feature
34118 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
34120 The names of registers are not case sensitive for the purpose
34121 of recognizing standard features, but @value{GDBN} will only display
34122 registers using the capitalization used in the description.
34129 * PowerPC Features::
34134 @subsection ARM Features
34135 @cindex target descriptions, ARM features
34137 The @samp{org.gnu.gdb.arm.core} feature is required for ARM targets.
34138 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
34139 @samp{lr}, @samp{pc}, and @samp{cpsr}.
34141 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
34142 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
34144 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
34145 it should contain at least registers @samp{wR0} through @samp{wR15} and
34146 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
34147 @samp{wCSSF}, and @samp{wCASF} registers are optional.
34149 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
34150 should contain at least registers @samp{d0} through @samp{d15}. If
34151 they are present, @samp{d16} through @samp{d31} should also be included.
34152 @value{GDBN} will synthesize the single-precision registers from
34153 halves of the double-precision registers.
34155 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
34156 need to contain registers; it instructs @value{GDBN} to display the
34157 VFP double-precision registers as vectors and to synthesize the
34158 quad-precision registers from pairs of double-precision registers.
34159 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
34160 be present and include 32 double-precision registers.
34162 @node i386 Features
34163 @subsection i386 Features
34164 @cindex target descriptions, i386 features
34166 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
34167 targets. It should describe the following registers:
34171 @samp{eax} through @samp{edi} plus @samp{eip} for i386
34173 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
34175 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
34176 @samp{fs}, @samp{gs}
34178 @samp{st0} through @samp{st7}
34180 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
34181 @samp{foseg}, @samp{fooff} and @samp{fop}
34184 The register sets may be different, depending on the target.
34186 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
34187 describe registers:
34191 @samp{xmm0} through @samp{xmm7} for i386
34193 @samp{xmm0} through @samp{xmm15} for amd64
34198 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
34199 @samp{org.gnu.gdb.i386.sse} feature. It should
34200 describe the upper 128 bits of @sc{ymm} registers:
34204 @samp{ymm0h} through @samp{ymm7h} for i386
34206 @samp{ymm0h} through @samp{ymm15h} for amd64
34210 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
34211 describe a single register, @samp{orig_eax}.
34213 @node MIPS Features
34214 @subsection MIPS Features
34215 @cindex target descriptions, MIPS features
34217 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
34218 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
34219 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
34222 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
34223 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
34224 registers. They may be 32-bit or 64-bit depending on the target.
34226 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
34227 it may be optional in a future version of @value{GDBN}. It should
34228 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
34229 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
34231 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
34232 contain a single register, @samp{restart}, which is used by the
34233 Linux kernel to control restartable syscalls.
34235 @node M68K Features
34236 @subsection M68K Features
34237 @cindex target descriptions, M68K features
34240 @item @samp{org.gnu.gdb.m68k.core}
34241 @itemx @samp{org.gnu.gdb.coldfire.core}
34242 @itemx @samp{org.gnu.gdb.fido.core}
34243 One of those features must be always present.
34244 The feature that is present determines which flavor of m68k is
34245 used. The feature that is present should contain registers
34246 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
34247 @samp{sp}, @samp{ps} and @samp{pc}.
34249 @item @samp{org.gnu.gdb.coldfire.fp}
34250 This feature is optional. If present, it should contain registers
34251 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
34255 @node PowerPC Features
34256 @subsection PowerPC Features
34257 @cindex target descriptions, PowerPC features
34259 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
34260 targets. It should contain registers @samp{r0} through @samp{r31},
34261 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
34262 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
34264 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
34265 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
34267 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
34268 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
34271 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
34272 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
34273 will combine these registers with the floating point registers
34274 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
34275 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
34276 through @samp{vs63}, the set of vector registers for POWER7.
34278 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
34279 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
34280 @samp{spefscr}. SPE targets should provide 32-bit registers in
34281 @samp{org.gnu.gdb.power.core} and provide the upper halves in
34282 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
34283 these to present registers @samp{ev0} through @samp{ev31} to the
34286 @node Operating System Information
34287 @appendix Operating System Information
34288 @cindex operating system information
34294 Users of @value{GDBN} often wish to obtain information about the state of
34295 the operating system running on the target---for example the list of
34296 processes, or the list of open files. This section describes the
34297 mechanism that makes it possible. This mechanism is similar to the
34298 target features mechanism (@pxref{Target Descriptions}), but focuses
34299 on a different aspect of target.
34301 Operating system information is retrived from the target via the
34302 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
34303 read}). The object name in the request should be @samp{osdata}, and
34304 the @var{annex} identifies the data to be fetched.
34307 @appendixsection Process list
34308 @cindex operating system information, process list
34310 When requesting the process list, the @var{annex} field in the
34311 @samp{qXfer} request should be @samp{processes}. The returned data is
34312 an XML document. The formal syntax of this document is defined in
34313 @file{gdb/features/osdata.dtd}.
34315 An example document is:
34318 <?xml version="1.0"?>
34319 <!DOCTYPE target SYSTEM "osdata.dtd">
34320 <osdata type="processes">
34322 <column name="pid">1</column>
34323 <column name="user">root</column>
34324 <column name="command">/sbin/init</column>
34325 <column name="cores">1,2,3</column>
34330 Each item should include a column whose name is @samp{pid}. The value
34331 of that column should identify the process on the target. The
34332 @samp{user} and @samp{command} columns are optional, and will be
34333 displayed by @value{GDBN}. The @samp{cores} column, if present,
34334 should contain a comma-separated list of cores that this process
34335 is running on. Target may provide additional columns,
34336 which @value{GDBN} currently ignores.
34350 % I think something like @colophon should be in texinfo. In the
34352 \long\def\colophon{\hbox to0pt{}\vfill
34353 \centerline{The body of this manual is set in}
34354 \centerline{\fontname\tenrm,}
34355 \centerline{with headings in {\bf\fontname\tenbf}}
34356 \centerline{and examples in {\tt\fontname\tentt}.}
34357 \centerline{{\it\fontname\tenit\/},}
34358 \centerline{{\bf\fontname\tenbf}, and}
34359 \centerline{{\sl\fontname\tensl\/}}
34360 \centerline{are used for emphasis.}\vfill}
34362 % Blame: doc@cygnus.com, 1991.