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.
1033 Batch mode may be useful for running @value{GDBN} as a filter, for
1034 example to download and run a program on another computer; in order to
1035 make this more useful, the message
1038 Program exited normally.
1042 (which is ordinarily issued whenever a program running under
1043 @value{GDBN} control terminates) is not issued when running in batch
1047 @cindex @code{--batch-silent}
1048 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1049 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1050 unaffected). This is much quieter than @samp{-silent} and would be useless
1051 for an interactive session.
1053 This is particularly useful when using targets that give @samp{Loading section}
1054 messages, for example.
1056 Note that targets that give their output via @value{GDBN}, as opposed to
1057 writing directly to @code{stdout}, will also be made silent.
1059 @item -return-child-result
1060 @cindex @code{--return-child-result}
1061 The return code from @value{GDBN} will be the return code from the child
1062 process (the process being debugged), with the following exceptions:
1066 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1067 internal error. In this case the exit code is the same as it would have been
1068 without @samp{-return-child-result}.
1070 The user quits with an explicit value. E.g., @samp{quit 1}.
1072 The child process never runs, or is not allowed to terminate, in which case
1073 the exit code will be -1.
1076 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1077 when @value{GDBN} is being used as a remote program loader or simulator
1082 @cindex @code{--nowindows}
1084 ``No windows''. If @value{GDBN} comes with a graphical user interface
1085 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1086 interface. If no GUI is available, this option has no effect.
1090 @cindex @code{--windows}
1092 If @value{GDBN} includes a GUI, then this option requires it to be
1095 @item -cd @var{directory}
1097 Run @value{GDBN} using @var{directory} as its working directory,
1098 instead of the current directory.
1102 @cindex @code{--fullname}
1104 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1105 subprocess. It tells @value{GDBN} to output the full file name and line
1106 number in a standard, recognizable fashion each time a stack frame is
1107 displayed (which includes each time your program stops). This
1108 recognizable format looks like two @samp{\032} characters, followed by
1109 the file name, line number and character position separated by colons,
1110 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1111 @samp{\032} characters as a signal to display the source code for the
1115 @cindex @code{--epoch}
1116 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1117 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1118 routines so as to allow Epoch to display values of expressions in a
1121 @item -annotate @var{level}
1122 @cindex @code{--annotate}
1123 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1124 effect is identical to using @samp{set annotate @var{level}}
1125 (@pxref{Annotations}). The annotation @var{level} controls how much
1126 information @value{GDBN} prints together with its prompt, values of
1127 expressions, source lines, and other types of output. Level 0 is the
1128 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1129 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1130 that control @value{GDBN}, and level 2 has been deprecated.
1132 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1136 @cindex @code{--args}
1137 Change interpretation of command line so that arguments following the
1138 executable file are passed as command line arguments to the inferior.
1139 This option stops option processing.
1141 @item -baud @var{bps}
1143 @cindex @code{--baud}
1145 Set the line speed (baud rate or bits per second) of any serial
1146 interface used by @value{GDBN} for remote debugging.
1148 @item -l @var{timeout}
1150 Set the timeout (in seconds) of any communication used by @value{GDBN}
1151 for remote debugging.
1153 @item -tty @var{device}
1154 @itemx -t @var{device}
1155 @cindex @code{--tty}
1157 Run using @var{device} for your program's standard input and output.
1158 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1160 @c resolve the situation of these eventually
1162 @cindex @code{--tui}
1163 Activate the @dfn{Text User Interface} when starting. The Text User
1164 Interface manages several text windows on the terminal, showing
1165 source, assembly, registers and @value{GDBN} command outputs
1166 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1167 Text User Interface can be enabled by invoking the program
1168 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1169 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1172 @c @cindex @code{--xdb}
1173 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1174 @c For information, see the file @file{xdb_trans.html}, which is usually
1175 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1178 @item -interpreter @var{interp}
1179 @cindex @code{--interpreter}
1180 Use the interpreter @var{interp} for interface with the controlling
1181 program or device. This option is meant to be set by programs which
1182 communicate with @value{GDBN} using it as a back end.
1183 @xref{Interpreters, , Command Interpreters}.
1185 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1186 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1187 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1188 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1189 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1190 @sc{gdb/mi} interfaces are no longer supported.
1193 @cindex @code{--write}
1194 Open the executable and core files for both reading and writing. This
1195 is equivalent to the @samp{set write on} command inside @value{GDBN}
1199 @cindex @code{--statistics}
1200 This option causes @value{GDBN} to print statistics about time and
1201 memory usage after it completes each command and returns to the prompt.
1204 @cindex @code{--version}
1205 This option causes @value{GDBN} to print its version number and
1206 no-warranty blurb, and exit.
1211 @subsection What @value{GDBN} Does During Startup
1212 @cindex @value{GDBN} startup
1214 Here's the description of what @value{GDBN} does during session startup:
1218 Sets up the command interpreter as specified by the command line
1219 (@pxref{Mode Options, interpreter}).
1223 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1224 used when building @value{GDBN}; @pxref{System-wide configuration,
1225 ,System-wide configuration and settings}) and executes all the commands in
1229 Reads the init file (if any) in your home directory@footnote{On
1230 DOS/Windows systems, the home directory is the one pointed to by the
1231 @code{HOME} environment variable.} and executes all the commands in
1235 Processes command line options and operands.
1238 Reads and executes the commands from init file (if any) in the current
1239 working directory. This is only done if the current directory is
1240 different from your home directory. Thus, you can have more than one
1241 init file, one generic in your home directory, and another, specific
1242 to the program you are debugging, in the directory where you invoke
1246 Reads command files specified by the @samp{-x} option. @xref{Command
1247 Files}, for more details about @value{GDBN} command files.
1250 Reads the command history recorded in the @dfn{history file}.
1251 @xref{Command History}, for more details about the command history and the
1252 files where @value{GDBN} records it.
1255 Init files use the same syntax as @dfn{command files} (@pxref{Command
1256 Files}) and are processed by @value{GDBN} in the same way. The init
1257 file in your home directory can set options (such as @samp{set
1258 complaints}) that affect subsequent processing of command line options
1259 and operands. Init files are not executed if you use the @samp{-nx}
1260 option (@pxref{Mode Options, ,Choosing Modes}).
1262 To display the list of init files loaded by gdb at startup, you
1263 can use @kbd{gdb --help}.
1265 @cindex init file name
1266 @cindex @file{.gdbinit}
1267 @cindex @file{gdb.ini}
1268 The @value{GDBN} init files are normally called @file{.gdbinit}.
1269 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1270 the limitations of file names imposed by DOS filesystems. The Windows
1271 ports of @value{GDBN} use the standard name, but if they find a
1272 @file{gdb.ini} file, they warn you about that and suggest to rename
1273 the file to the standard name.
1277 @section Quitting @value{GDBN}
1278 @cindex exiting @value{GDBN}
1279 @cindex leaving @value{GDBN}
1282 @kindex quit @r{[}@var{expression}@r{]}
1283 @kindex q @r{(@code{quit})}
1284 @item quit @r{[}@var{expression}@r{]}
1286 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1287 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1288 do not supply @var{expression}, @value{GDBN} will terminate normally;
1289 otherwise it will terminate using the result of @var{expression} as the
1294 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1295 terminates the action of any @value{GDBN} command that is in progress and
1296 returns to @value{GDBN} command level. It is safe to type the interrupt
1297 character at any time because @value{GDBN} does not allow it to take effect
1298 until a time when it is safe.
1300 If you have been using @value{GDBN} to control an attached process or
1301 device, you can release it with the @code{detach} command
1302 (@pxref{Attach, ,Debugging an Already-running Process}).
1304 @node Shell Commands
1305 @section Shell Commands
1307 If you need to execute occasional shell commands during your
1308 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1309 just use the @code{shell} command.
1313 @cindex shell escape
1314 @item shell @var{command string}
1315 Invoke a standard shell to execute @var{command string}.
1316 If it exists, the environment variable @code{SHELL} determines which
1317 shell to run. Otherwise @value{GDBN} uses the default shell
1318 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1321 The utility @code{make} is often needed in development environments.
1322 You do not have to use the @code{shell} command for this purpose in
1327 @cindex calling make
1328 @item make @var{make-args}
1329 Execute the @code{make} program with the specified
1330 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1333 @node Logging Output
1334 @section Logging Output
1335 @cindex logging @value{GDBN} output
1336 @cindex save @value{GDBN} output to a file
1338 You may want to save the output of @value{GDBN} commands to a file.
1339 There are several commands to control @value{GDBN}'s logging.
1343 @item set logging on
1345 @item set logging off
1347 @cindex logging file name
1348 @item set logging file @var{file}
1349 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1350 @item set logging overwrite [on|off]
1351 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1352 you want @code{set logging on} to overwrite the logfile instead.
1353 @item set logging redirect [on|off]
1354 By default, @value{GDBN} output will go to both the terminal and the logfile.
1355 Set @code{redirect} if you want output to go only to the log file.
1356 @kindex show logging
1358 Show the current values of the logging settings.
1362 @chapter @value{GDBN} Commands
1364 You can abbreviate a @value{GDBN} command to the first few letters of the command
1365 name, if that abbreviation is unambiguous; and you can repeat certain
1366 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1367 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1368 show you the alternatives available, if there is more than one possibility).
1371 * Command Syntax:: How to give commands to @value{GDBN}
1372 * Completion:: Command completion
1373 * Help:: How to ask @value{GDBN} for help
1376 @node Command Syntax
1377 @section Command Syntax
1379 A @value{GDBN} command is a single line of input. There is no limit on
1380 how long it can be. It starts with a command name, which is followed by
1381 arguments whose meaning depends on the command name. For example, the
1382 command @code{step} accepts an argument which is the number of times to
1383 step, as in @samp{step 5}. You can also use the @code{step} command
1384 with no arguments. Some commands do not allow any arguments.
1386 @cindex abbreviation
1387 @value{GDBN} command names may always be truncated if that abbreviation is
1388 unambiguous. Other possible command abbreviations are listed in the
1389 documentation for individual commands. In some cases, even ambiguous
1390 abbreviations are allowed; for example, @code{s} is specially defined as
1391 equivalent to @code{step} even though there are other commands whose
1392 names start with @code{s}. You can test abbreviations by using them as
1393 arguments to the @code{help} command.
1395 @cindex repeating commands
1396 @kindex RET @r{(repeat last command)}
1397 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1398 repeat the previous command. Certain commands (for example, @code{run})
1399 will not repeat this way; these are commands whose unintentional
1400 repetition might cause trouble and which you are unlikely to want to
1401 repeat. User-defined commands can disable this feature; see
1402 @ref{Define, dont-repeat}.
1404 The @code{list} and @code{x} commands, when you repeat them with
1405 @key{RET}, construct new arguments rather than repeating
1406 exactly as typed. This permits easy scanning of source or memory.
1408 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1409 output, in a way similar to the common utility @code{more}
1410 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1411 @key{RET} too many in this situation, @value{GDBN} disables command
1412 repetition after any command that generates this sort of display.
1414 @kindex # @r{(a comment)}
1416 Any text from a @kbd{#} to the end of the line is a comment; it does
1417 nothing. This is useful mainly in command files (@pxref{Command
1418 Files,,Command Files}).
1420 @cindex repeating command sequences
1421 @kindex Ctrl-o @r{(operate-and-get-next)}
1422 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1423 commands. This command accepts the current line, like @key{RET}, and
1424 then fetches the next line relative to the current line from the history
1428 @section Command Completion
1431 @cindex word completion
1432 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1433 only one possibility; it can also show you what the valid possibilities
1434 are for the next word in a command, at any time. This works for @value{GDBN}
1435 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1437 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1438 of a word. If there is only one possibility, @value{GDBN} fills in the
1439 word, and waits for you to finish the command (or press @key{RET} to
1440 enter it). For example, if you type
1442 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1443 @c complete accuracy in these examples; space introduced for clarity.
1444 @c If texinfo enhancements make it unnecessary, it would be nice to
1445 @c replace " @key" by "@key" in the following...
1447 (@value{GDBP}) info bre @key{TAB}
1451 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1452 the only @code{info} subcommand beginning with @samp{bre}:
1455 (@value{GDBP}) info breakpoints
1459 You can either press @key{RET} at this point, to run the @code{info
1460 breakpoints} command, or backspace and enter something else, if
1461 @samp{breakpoints} does not look like the command you expected. (If you
1462 were sure you wanted @code{info breakpoints} in the first place, you
1463 might as well just type @key{RET} immediately after @samp{info bre},
1464 to exploit command abbreviations rather than command completion).
1466 If there is more than one possibility for the next word when you press
1467 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1468 characters and try again, or just press @key{TAB} a second time;
1469 @value{GDBN} displays all the possible completions for that word. For
1470 example, you might want to set a breakpoint on a subroutine whose name
1471 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1472 just sounds the bell. Typing @key{TAB} again displays all the
1473 function names in your program that begin with those characters, for
1477 (@value{GDBP}) b make_ @key{TAB}
1478 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1479 make_a_section_from_file make_environ
1480 make_abs_section make_function_type
1481 make_blockvector make_pointer_type
1482 make_cleanup make_reference_type
1483 make_command make_symbol_completion_list
1484 (@value{GDBP}) b make_
1488 After displaying the available possibilities, @value{GDBN} copies your
1489 partial input (@samp{b make_} in the example) so you can finish the
1492 If you just want to see the list of alternatives in the first place, you
1493 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1494 means @kbd{@key{META} ?}. You can type this either by holding down a
1495 key designated as the @key{META} shift on your keyboard (if there is
1496 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1498 @cindex quotes in commands
1499 @cindex completion of quoted strings
1500 Sometimes the string you need, while logically a ``word'', may contain
1501 parentheses or other characters that @value{GDBN} normally excludes from
1502 its notion of a word. To permit word completion to work in this
1503 situation, you may enclose words in @code{'} (single quote marks) in
1504 @value{GDBN} commands.
1506 The most likely situation where you might need this is in typing the
1507 name of a C@t{++} function. This is because C@t{++} allows function
1508 overloading (multiple definitions of the same function, distinguished
1509 by argument type). For example, when you want to set a breakpoint you
1510 may need to distinguish whether you mean the version of @code{name}
1511 that takes an @code{int} parameter, @code{name(int)}, or the version
1512 that takes a @code{float} parameter, @code{name(float)}. To use the
1513 word-completion facilities in this situation, type a single quote
1514 @code{'} at the beginning of the function name. This alerts
1515 @value{GDBN} that it may need to consider more information than usual
1516 when you press @key{TAB} or @kbd{M-?} to request word completion:
1519 (@value{GDBP}) b 'bubble( @kbd{M-?}
1520 bubble(double,double) bubble(int,int)
1521 (@value{GDBP}) b 'bubble(
1524 In some cases, @value{GDBN} can tell that completing a name requires using
1525 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1526 completing as much as it can) if you do not type the quote in the first
1530 (@value{GDBP}) b bub @key{TAB}
1531 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1532 (@value{GDBP}) b 'bubble(
1536 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1537 you have not yet started typing the argument list when you ask for
1538 completion on an overloaded symbol.
1540 For more information about overloaded functions, see @ref{C Plus Plus
1541 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1542 overload-resolution off} to disable overload resolution;
1543 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1545 @cindex completion of structure field names
1546 @cindex structure field name completion
1547 @cindex completion of union field names
1548 @cindex union field name completion
1549 When completing in an expression which looks up a field in a
1550 structure, @value{GDBN} also tries@footnote{The completer can be
1551 confused by certain kinds of invalid expressions. Also, it only
1552 examines the static type of the expression, not the dynamic type.} to
1553 limit completions to the field names available in the type of the
1557 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1558 magic to_delete to_fputs to_put to_rewind
1559 to_data to_flush to_isatty to_read to_write
1563 This is because the @code{gdb_stdout} is a variable of the type
1564 @code{struct ui_file} that is defined in @value{GDBN} sources as
1571 ui_file_flush_ftype *to_flush;
1572 ui_file_write_ftype *to_write;
1573 ui_file_fputs_ftype *to_fputs;
1574 ui_file_read_ftype *to_read;
1575 ui_file_delete_ftype *to_delete;
1576 ui_file_isatty_ftype *to_isatty;
1577 ui_file_rewind_ftype *to_rewind;
1578 ui_file_put_ftype *to_put;
1585 @section Getting Help
1586 @cindex online documentation
1589 You can always ask @value{GDBN} itself for information on its commands,
1590 using the command @code{help}.
1593 @kindex h @r{(@code{help})}
1596 You can use @code{help} (abbreviated @code{h}) with no arguments to
1597 display a short list of named classes of commands:
1601 List of classes of commands:
1603 aliases -- Aliases of other commands
1604 breakpoints -- Making program stop at certain points
1605 data -- Examining data
1606 files -- Specifying and examining files
1607 internals -- Maintenance commands
1608 obscure -- Obscure features
1609 running -- Running the program
1610 stack -- Examining the stack
1611 status -- Status inquiries
1612 support -- Support facilities
1613 tracepoints -- Tracing of program execution without
1614 stopping the program
1615 user-defined -- User-defined commands
1617 Type "help" followed by a class name for a list of
1618 commands in that class.
1619 Type "help" followed by command name for full
1621 Command name abbreviations are allowed if unambiguous.
1624 @c the above line break eliminates huge line overfull...
1626 @item help @var{class}
1627 Using one of the general help classes as an argument, you can get a
1628 list of the individual commands in that class. For example, here is the
1629 help display for the class @code{status}:
1632 (@value{GDBP}) help status
1637 @c Line break in "show" line falsifies real output, but needed
1638 @c to fit in smallbook page size.
1639 info -- Generic command for showing things
1640 about the program being debugged
1641 show -- Generic command for showing things
1644 Type "help" followed by command name for full
1646 Command name abbreviations are allowed if unambiguous.
1650 @item help @var{command}
1651 With a command name as @code{help} argument, @value{GDBN} displays a
1652 short paragraph on how to use that command.
1655 @item apropos @var{args}
1656 The @code{apropos} command searches through all of the @value{GDBN}
1657 commands, and their documentation, for the regular expression specified in
1658 @var{args}. It prints out all matches found. For example:
1669 set symbol-reloading -- Set dynamic symbol table reloading
1670 multiple times in one run
1671 show symbol-reloading -- Show dynamic symbol table reloading
1672 multiple times in one run
1677 @item complete @var{args}
1678 The @code{complete @var{args}} command lists all the possible completions
1679 for the beginning of a command. Use @var{args} to specify the beginning of the
1680 command you want completed. For example:
1686 @noindent results in:
1697 @noindent This is intended for use by @sc{gnu} Emacs.
1700 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1701 and @code{show} to inquire about the state of your program, or the state
1702 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1703 manual introduces each of them in the appropriate context. The listings
1704 under @code{info} and under @code{show} in the Index point to
1705 all the sub-commands. @xref{Index}.
1710 @kindex i @r{(@code{info})}
1712 This command (abbreviated @code{i}) is for describing the state of your
1713 program. For example, you can show the arguments passed to a function
1714 with @code{info args}, list the registers currently in use with @code{info
1715 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1716 You can get a complete list of the @code{info} sub-commands with
1717 @w{@code{help info}}.
1721 You can assign the result of an expression to an environment variable with
1722 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1723 @code{set prompt $}.
1727 In contrast to @code{info}, @code{show} is for describing the state of
1728 @value{GDBN} itself.
1729 You can change most of the things you can @code{show}, by using the
1730 related command @code{set}; for example, you can control what number
1731 system is used for displays with @code{set radix}, or simply inquire
1732 which is currently in use with @code{show radix}.
1735 To display all the settable parameters and their current
1736 values, you can use @code{show} with no arguments; you may also use
1737 @code{info set}. Both commands produce the same display.
1738 @c FIXME: "info set" violates the rule that "info" is for state of
1739 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1740 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1744 Here are three miscellaneous @code{show} subcommands, all of which are
1745 exceptional in lacking corresponding @code{set} commands:
1748 @kindex show version
1749 @cindex @value{GDBN} version number
1751 Show what version of @value{GDBN} is running. You should include this
1752 information in @value{GDBN} bug-reports. If multiple versions of
1753 @value{GDBN} are in use at your site, you may need to determine which
1754 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1755 commands are introduced, and old ones may wither away. Also, many
1756 system vendors ship variant versions of @value{GDBN}, and there are
1757 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1758 The version number is the same as the one announced when you start
1761 @kindex show copying
1762 @kindex info copying
1763 @cindex display @value{GDBN} copyright
1766 Display information about permission for copying @value{GDBN}.
1768 @kindex show warranty
1769 @kindex info warranty
1771 @itemx info warranty
1772 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1773 if your version of @value{GDBN} comes with one.
1778 @chapter Running Programs Under @value{GDBN}
1780 When you run a program under @value{GDBN}, you must first generate
1781 debugging information when you compile it.
1783 You may start @value{GDBN} with its arguments, if any, in an environment
1784 of your choice. If you are doing native debugging, you may redirect
1785 your program's input and output, debug an already running process, or
1786 kill a child process.
1789 * Compilation:: Compiling for debugging
1790 * Starting:: Starting your program
1791 * Arguments:: Your program's arguments
1792 * Environment:: Your program's environment
1794 * Working Directory:: Your program's working directory
1795 * Input/Output:: Your program's input and output
1796 * Attach:: Debugging an already-running process
1797 * Kill Process:: Killing the child process
1799 * Inferiors and Programs:: Debugging multiple inferiors and programs
1800 * Threads:: Debugging programs with multiple threads
1801 * Forks:: Debugging forks
1802 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1806 @section Compiling for Debugging
1808 In order to debug a program effectively, you need to generate
1809 debugging information when you compile it. This debugging information
1810 is stored in the object file; it describes the data type of each
1811 variable or function and the correspondence between source line numbers
1812 and addresses in the executable code.
1814 To request debugging information, specify the @samp{-g} option when you run
1817 Programs that are to be shipped to your customers are compiled with
1818 optimizations, using the @samp{-O} compiler option. However, some
1819 compilers are unable to handle the @samp{-g} and @samp{-O} options
1820 together. Using those compilers, you cannot generate optimized
1821 executables containing debugging information.
1823 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1824 without @samp{-O}, making it possible to debug optimized code. We
1825 recommend that you @emph{always} use @samp{-g} whenever you compile a
1826 program. You may think your program is correct, but there is no sense
1827 in pushing your luck. For more information, see @ref{Optimized Code}.
1829 Older versions of the @sc{gnu} C compiler permitted a variant option
1830 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1831 format; if your @sc{gnu} C compiler has this option, do not use it.
1833 @value{GDBN} knows about preprocessor macros and can show you their
1834 expansion (@pxref{Macros}). Most compilers do not include information
1835 about preprocessor macros in the debugging information if you specify
1836 the @option{-g} flag alone, because this information is rather large.
1837 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1838 provides macro information if you specify the options
1839 @option{-gdwarf-2} and @option{-g3}; the former option requests
1840 debugging information in the Dwarf 2 format, and the latter requests
1841 ``extra information''. In the future, we hope to find more compact
1842 ways to represent macro information, so that it can be included with
1847 @section Starting your Program
1853 @kindex r @r{(@code{run})}
1856 Use the @code{run} command to start your program under @value{GDBN}.
1857 You must first specify the program name (except on VxWorks) with an
1858 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1859 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1860 (@pxref{Files, ,Commands to Specify Files}).
1864 If you are running your program in an execution environment that
1865 supports processes, @code{run} creates an inferior process and makes
1866 that process run your program. In some environments without processes,
1867 @code{run} jumps to the start of your program. Other targets,
1868 like @samp{remote}, are always running. If you get an error
1869 message like this one:
1872 The "remote" target does not support "run".
1873 Try "help target" or "continue".
1877 then use @code{continue} to run your program. You may need @code{load}
1878 first (@pxref{load}).
1880 The execution of a program is affected by certain information it
1881 receives from its superior. @value{GDBN} provides ways to specify this
1882 information, which you must do @emph{before} starting your program. (You
1883 can change it after starting your program, but such changes only affect
1884 your program the next time you start it.) This information may be
1885 divided into four categories:
1888 @item The @emph{arguments.}
1889 Specify the arguments to give your program as the arguments of the
1890 @code{run} command. If a shell is available on your target, the shell
1891 is used to pass the arguments, so that you may use normal conventions
1892 (such as wildcard expansion or variable substitution) in describing
1894 In Unix systems, you can control which shell is used with the
1895 @code{SHELL} environment variable.
1896 @xref{Arguments, ,Your Program's Arguments}.
1898 @item The @emph{environment.}
1899 Your program normally inherits its environment from @value{GDBN}, but you can
1900 use the @value{GDBN} commands @code{set environment} and @code{unset
1901 environment} to change parts of the environment that affect
1902 your program. @xref{Environment, ,Your Program's Environment}.
1904 @item The @emph{working directory.}
1905 Your program inherits its working directory from @value{GDBN}. You can set
1906 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1907 @xref{Working Directory, ,Your Program's Working Directory}.
1909 @item The @emph{standard input and output.}
1910 Your program normally uses the same device for standard input and
1911 standard output as @value{GDBN} is using. You can redirect input and output
1912 in the @code{run} command line, or you can use the @code{tty} command to
1913 set a different device for your program.
1914 @xref{Input/Output, ,Your Program's Input and Output}.
1917 @emph{Warning:} While input and output redirection work, you cannot use
1918 pipes to pass the output of the program you are debugging to another
1919 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1923 When you issue the @code{run} command, your program begins to execute
1924 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1925 of how to arrange for your program to stop. Once your program has
1926 stopped, you may call functions in your program, using the @code{print}
1927 or @code{call} commands. @xref{Data, ,Examining Data}.
1929 If the modification time of your symbol file has changed since the last
1930 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1931 table, and reads it again. When it does this, @value{GDBN} tries to retain
1932 your current breakpoints.
1937 @cindex run to main procedure
1938 The name of the main procedure can vary from language to language.
1939 With C or C@t{++}, the main procedure name is always @code{main}, but
1940 other languages such as Ada do not require a specific name for their
1941 main procedure. The debugger provides a convenient way to start the
1942 execution of the program and to stop at the beginning of the main
1943 procedure, depending on the language used.
1945 The @samp{start} command does the equivalent of setting a temporary
1946 breakpoint at the beginning of the main procedure and then invoking
1947 the @samp{run} command.
1949 @cindex elaboration phase
1950 Some programs contain an @dfn{elaboration} phase where some startup code is
1951 executed before the main procedure is called. This depends on the
1952 languages used to write your program. In C@t{++}, for instance,
1953 constructors for static and global objects are executed before
1954 @code{main} is called. It is therefore possible that the debugger stops
1955 before reaching the main procedure. However, the temporary breakpoint
1956 will remain to halt execution.
1958 Specify the arguments to give to your program as arguments to the
1959 @samp{start} command. These arguments will be given verbatim to the
1960 underlying @samp{run} command. Note that the same arguments will be
1961 reused if no argument is provided during subsequent calls to
1962 @samp{start} or @samp{run}.
1964 It is sometimes necessary to debug the program during elaboration. In
1965 these cases, using the @code{start} command would stop the execution of
1966 your program too late, as the program would have already completed the
1967 elaboration phase. Under these circumstances, insert breakpoints in your
1968 elaboration code before running your program.
1970 @kindex set exec-wrapper
1971 @item set exec-wrapper @var{wrapper}
1972 @itemx show exec-wrapper
1973 @itemx unset exec-wrapper
1974 When @samp{exec-wrapper} is set, the specified wrapper is used to
1975 launch programs for debugging. @value{GDBN} starts your program
1976 with a shell command of the form @kbd{exec @var{wrapper}
1977 @var{program}}. Quoting is added to @var{program} and its
1978 arguments, but not to @var{wrapper}, so you should add quotes if
1979 appropriate for your shell. The wrapper runs until it executes
1980 your program, and then @value{GDBN} takes control.
1982 You can use any program that eventually calls @code{execve} with
1983 its arguments as a wrapper. Several standard Unix utilities do
1984 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
1985 with @code{exec "$@@"} will also work.
1987 For example, you can use @code{env} to pass an environment variable to
1988 the debugged program, without setting the variable in your shell's
1992 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
1996 This command is available when debugging locally on most targets, excluding
1997 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
1999 @kindex set disable-randomization
2000 @item set disable-randomization
2001 @itemx set disable-randomization on
2002 This option (enabled by default in @value{GDBN}) will turn off the native
2003 randomization of the virtual address space of the started program. This option
2004 is useful for multiple debugging sessions to make the execution better
2005 reproducible and memory addresses reusable across debugging sessions.
2007 This feature is implemented only on @sc{gnu}/Linux. You can get the same
2011 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2014 @item set disable-randomization off
2015 Leave the behavior of the started executable unchanged. Some bugs rear their
2016 ugly heads only when the program is loaded at certain addresses. If your bug
2017 disappears when you run the program under @value{GDBN}, that might be because
2018 @value{GDBN} by default disables the address randomization on platforms, such
2019 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2020 disable-randomization off} to try to reproduce such elusive bugs.
2022 The virtual address space randomization is implemented only on @sc{gnu}/Linux.
2023 It protects the programs against some kinds of security attacks. In these
2024 cases the attacker needs to know the exact location of a concrete executable
2025 code. Randomizing its location makes it impossible to inject jumps misusing
2026 a code at its expected addresses.
2028 Prelinking shared libraries provides a startup performance advantage but it
2029 makes addresses in these libraries predictable for privileged processes by
2030 having just unprivileged access at the target system. Reading the shared
2031 library binary gives enough information for assembling the malicious code
2032 misusing it. Still even a prelinked shared library can get loaded at a new
2033 random address just requiring the regular relocation process during the
2034 startup. Shared libraries not already prelinked are always loaded at
2035 a randomly chosen address.
2037 Position independent executables (PIE) contain position independent code
2038 similar to the shared libraries and therefore such executables get loaded at
2039 a randomly chosen address upon startup. PIE executables always load even
2040 already prelinked shared libraries at a random address. You can build such
2041 executable using @command{gcc -fPIE -pie}.
2043 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2044 (as long as the randomization is enabled).
2046 @item show disable-randomization
2047 Show the current setting of the explicit disable of the native randomization of
2048 the virtual address space of the started program.
2053 @section Your Program's Arguments
2055 @cindex arguments (to your program)
2056 The arguments to your program can be specified by the arguments of the
2058 They are passed to a shell, which expands wildcard characters and
2059 performs redirection of I/O, and thence to your program. Your
2060 @code{SHELL} environment variable (if it exists) specifies what shell
2061 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2062 the default shell (@file{/bin/sh} on Unix).
2064 On non-Unix systems, the program is usually invoked directly by
2065 @value{GDBN}, which emulates I/O redirection via the appropriate system
2066 calls, and the wildcard characters are expanded by the startup code of
2067 the program, not by the shell.
2069 @code{run} with no arguments uses the same arguments used by the previous
2070 @code{run}, or those set by the @code{set args} command.
2075 Specify the arguments to be used the next time your program is run. If
2076 @code{set args} has no arguments, @code{run} executes your program
2077 with no arguments. Once you have run your program with arguments,
2078 using @code{set args} before the next @code{run} is the only way to run
2079 it again without arguments.
2083 Show the arguments to give your program when it is started.
2087 @section Your Program's Environment
2089 @cindex environment (of your program)
2090 The @dfn{environment} consists of a set of environment variables and
2091 their values. Environment variables conventionally record such things as
2092 your user name, your home directory, your terminal type, and your search
2093 path for programs to run. Usually you set up environment variables with
2094 the shell and they are inherited by all the other programs you run. When
2095 debugging, it can be useful to try running your program with a modified
2096 environment without having to start @value{GDBN} over again.
2100 @item path @var{directory}
2101 Add @var{directory} to the front of the @code{PATH} environment variable
2102 (the search path for executables) that will be passed to your program.
2103 The value of @code{PATH} used by @value{GDBN} does not change.
2104 You may specify several directory names, separated by whitespace or by a
2105 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2106 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2107 is moved to the front, so it is searched sooner.
2109 You can use the string @samp{$cwd} to refer to whatever is the current
2110 working directory at the time @value{GDBN} searches the path. If you
2111 use @samp{.} instead, it refers to the directory where you executed the
2112 @code{path} command. @value{GDBN} replaces @samp{.} in the
2113 @var{directory} argument (with the current path) before adding
2114 @var{directory} to the search path.
2115 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2116 @c document that, since repeating it would be a no-op.
2120 Display the list of search paths for executables (the @code{PATH}
2121 environment variable).
2123 @kindex show environment
2124 @item show environment @r{[}@var{varname}@r{]}
2125 Print the value of environment variable @var{varname} to be given to
2126 your program when it starts. If you do not supply @var{varname},
2127 print the names and values of all environment variables to be given to
2128 your program. You can abbreviate @code{environment} as @code{env}.
2130 @kindex set environment
2131 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2132 Set environment variable @var{varname} to @var{value}. The value
2133 changes for your program only, not for @value{GDBN} itself. @var{value} may
2134 be any string; the values of environment variables are just strings, and
2135 any interpretation is supplied by your program itself. The @var{value}
2136 parameter is optional; if it is eliminated, the variable is set to a
2138 @c "any string" here does not include leading, trailing
2139 @c blanks. Gnu asks: does anyone care?
2141 For example, this command:
2148 tells the debugged program, when subsequently run, that its user is named
2149 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2150 are not actually required.)
2152 @kindex unset environment
2153 @item unset environment @var{varname}
2154 Remove variable @var{varname} from the environment to be passed to your
2155 program. This is different from @samp{set env @var{varname} =};
2156 @code{unset environment} removes the variable from the environment,
2157 rather than assigning it an empty value.
2160 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2162 by your @code{SHELL} environment variable if it exists (or
2163 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2164 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2165 @file{.bashrc} for BASH---any variables you set in that file affect
2166 your program. You may wish to move setting of environment variables to
2167 files that are only run when you sign on, such as @file{.login} or
2170 @node Working Directory
2171 @section Your Program's Working Directory
2173 @cindex working directory (of your program)
2174 Each time you start your program with @code{run}, it inherits its
2175 working directory from the current working directory of @value{GDBN}.
2176 The @value{GDBN} working directory is initially whatever it inherited
2177 from its parent process (typically the shell), but you can specify a new
2178 working directory in @value{GDBN} with the @code{cd} command.
2180 The @value{GDBN} working directory also serves as a default for the commands
2181 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2186 @cindex change working directory
2187 @item cd @var{directory}
2188 Set the @value{GDBN} working directory to @var{directory}.
2192 Print the @value{GDBN} working directory.
2195 It is generally impossible to find the current working directory of
2196 the process being debugged (since a program can change its directory
2197 during its run). If you work on a system where @value{GDBN} is
2198 configured with the @file{/proc} support, you can use the @code{info
2199 proc} command (@pxref{SVR4 Process Information}) to find out the
2200 current working directory of the debuggee.
2203 @section Your Program's Input and Output
2208 By default, the program you run under @value{GDBN} does input and output to
2209 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2210 to its own terminal modes to interact with you, but it records the terminal
2211 modes your program was using and switches back to them when you continue
2212 running your program.
2215 @kindex info terminal
2217 Displays information recorded by @value{GDBN} about the terminal modes your
2221 You can redirect your program's input and/or output using shell
2222 redirection with the @code{run} command. For example,
2229 starts your program, diverting its output to the file @file{outfile}.
2232 @cindex controlling terminal
2233 Another way to specify where your program should do input and output is
2234 with the @code{tty} command. This command accepts a file name as
2235 argument, and causes this file to be the default for future @code{run}
2236 commands. It also resets the controlling terminal for the child
2237 process, for future @code{run} commands. For example,
2244 directs that processes started with subsequent @code{run} commands
2245 default to do input and output on the terminal @file{/dev/ttyb} and have
2246 that as their controlling terminal.
2248 An explicit redirection in @code{run} overrides the @code{tty} command's
2249 effect on the input/output device, but not its effect on the controlling
2252 When you use the @code{tty} command or redirect input in the @code{run}
2253 command, only the input @emph{for your program} is affected. The input
2254 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2255 for @code{set inferior-tty}.
2257 @cindex inferior tty
2258 @cindex set inferior controlling terminal
2259 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2260 display the name of the terminal that will be used for future runs of your
2264 @item set inferior-tty /dev/ttyb
2265 @kindex set inferior-tty
2266 Set the tty for the program being debugged to /dev/ttyb.
2268 @item show inferior-tty
2269 @kindex show inferior-tty
2270 Show the current tty for the program being debugged.
2274 @section Debugging an Already-running Process
2279 @item attach @var{process-id}
2280 This command attaches to a running process---one that was started
2281 outside @value{GDBN}. (@code{info files} shows your active
2282 targets.) The command takes as argument a process ID. The usual way to
2283 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2284 or with the @samp{jobs -l} shell command.
2286 @code{attach} does not repeat if you press @key{RET} a second time after
2287 executing the command.
2290 To use @code{attach}, your program must be running in an environment
2291 which supports processes; for example, @code{attach} does not work for
2292 programs on bare-board targets that lack an operating system. You must
2293 also have permission to send the process a signal.
2295 When you use @code{attach}, the debugger finds the program running in
2296 the process first by looking in the current working directory, then (if
2297 the program is not found) by using the source file search path
2298 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2299 the @code{file} command to load the program. @xref{Files, ,Commands to
2302 The first thing @value{GDBN} does after arranging to debug the specified
2303 process is to stop it. You can examine and modify an attached process
2304 with all the @value{GDBN} commands that are ordinarily available when
2305 you start processes with @code{run}. You can insert breakpoints; you
2306 can step and continue; you can modify storage. If you would rather the
2307 process continue running, you may use the @code{continue} command after
2308 attaching @value{GDBN} to the process.
2313 When you have finished debugging the attached process, you can use the
2314 @code{detach} command to release it from @value{GDBN} control. Detaching
2315 the process continues its execution. After the @code{detach} command,
2316 that process and @value{GDBN} become completely independent once more, and you
2317 are ready to @code{attach} another process or start one with @code{run}.
2318 @code{detach} does not repeat if you press @key{RET} again after
2319 executing the command.
2322 If you exit @value{GDBN} while you have an attached process, you detach
2323 that process. If you use the @code{run} command, you kill that process.
2324 By default, @value{GDBN} asks for confirmation if you try to do either of these
2325 things; you can control whether or not you need to confirm by using the
2326 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2330 @section Killing the Child Process
2335 Kill the child process in which your program is running under @value{GDBN}.
2338 This command is useful if you wish to debug a core dump instead of a
2339 running process. @value{GDBN} ignores any core dump file while your program
2342 On some operating systems, a program cannot be executed outside @value{GDBN}
2343 while you have breakpoints set on it inside @value{GDBN}. You can use the
2344 @code{kill} command in this situation to permit running your program
2345 outside the debugger.
2347 The @code{kill} command is also useful if you wish to recompile and
2348 relink your program, since on many systems it is impossible to modify an
2349 executable file while it is running in a process. In this case, when you
2350 next type @code{run}, @value{GDBN} notices that the file has changed, and
2351 reads the symbol table again (while trying to preserve your current
2352 breakpoint settings).
2354 @node Inferiors and Programs
2355 @section Debugging Multiple Inferiors and Programs
2357 @value{GDBN} lets you run and debug multiple programs in a single
2358 session. In addition, @value{GDBN} on some systems may let you run
2359 several programs simultaneously (otherwise you have to exit from one
2360 before starting another). In the most general case, you can have
2361 multiple threads of execution in each of multiple processes, launched
2362 from multiple executables.
2365 @value{GDBN} represents the state of each program execution with an
2366 object called an @dfn{inferior}. An inferior typically corresponds to
2367 a process, but is more general and applies also to targets that do not
2368 have processes. Inferiors may be created before a process runs, and
2369 may be retained after a process exits. Inferiors have unique
2370 identifiers that are different from process ids. Usually each
2371 inferior will also have its own distinct address space, although some
2372 embedded targets may have several inferiors running in different parts
2373 of a single address space. Each inferior may in turn have multiple
2374 threads running in it.
2376 To find out what inferiors exist at any moment, use @w{@code{info
2380 @kindex info inferiors
2381 @item info inferiors
2382 Print a list of all inferiors currently being managed by @value{GDBN}.
2384 @value{GDBN} displays for each inferior (in this order):
2388 the inferior number assigned by @value{GDBN}
2391 the target system's inferior identifier
2394 the name of the executable the inferior is running.
2399 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2400 indicates the current inferior.
2404 @c end table here to get a little more width for example
2407 (@value{GDBP}) info inferiors
2408 Num Description Executable
2409 2 process 2307 hello
2410 * 1 process 3401 goodbye
2413 To switch focus between inferiors, use the @code{inferior} command:
2416 @kindex inferior @var{infno}
2417 @item inferior @var{infno}
2418 Make inferior number @var{infno} the current inferior. The argument
2419 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2420 in the first field of the @samp{info inferiors} display.
2424 You can get multiple executables into a debugging session via the
2425 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2426 systems @value{GDBN} can add inferiors to the debug session
2427 automatically by following calls to @code{fork} and @code{exec}. To
2428 remove inferiors from the debugging session use the
2429 @w{@code{remove-inferior}} command.
2432 @kindex add-inferior
2433 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2434 Adds @var{n} inferiors to be run using @var{executable} as the
2435 executable. @var{n} defaults to 1. If no executable is specified,
2436 the inferiors begins empty, with no program. You can still assign or
2437 change the program assigned to the inferior at any time by using the
2438 @code{file} command with the executable name as its argument.
2440 @kindex clone-inferior
2441 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2442 Adds @var{n} inferiors ready to execute the same program as inferior
2443 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2444 number of the current inferior. This is a convenient command when you
2445 want to run another instance of the inferior you are debugging.
2448 (@value{GDBP}) info inferiors
2449 Num Description Executable
2450 * 1 process 29964 helloworld
2451 (@value{GDBP}) clone-inferior
2454 (@value{GDBP}) info inferiors
2455 Num Description Executable
2457 * 1 process 29964 helloworld
2460 You can now simply switch focus to inferior 2 and run it.
2462 @kindex remove-inferior
2463 @item remove-inferior @var{infno}
2464 Removes the inferior @var{infno}. It is not possible to remove an
2465 inferior that is running with this command. For those, use the
2466 @code{kill} or @code{detach} command first.
2470 To quit debugging one of the running inferiors that is not the current
2471 inferior, you can either detach from it by using the @w{@code{detach
2472 inferior}} command (allowing it to run independently), or kill it
2473 using the @w{@code{kill inferior}} command:
2476 @kindex detach inferior @var{infno}
2477 @item detach inferior @var{infno}
2478 Detach from the inferior identified by @value{GDBN} inferior number
2479 @var{infno}, and remove it from the inferior list.
2481 @kindex kill inferior @var{infno}
2482 @item kill inferior @var{infno}
2483 Kill the inferior identified by @value{GDBN} inferior number
2484 @var{infno}, and remove it from the inferior list.
2487 After the successful completion of a command such as @code{detach},
2488 @code{detach inferior}, @code{kill} or @code{kill inferior}, or after
2489 a normal process exit, the inferior is still valid and listed with
2490 @code{info inferiors}, ready to be restarted.
2493 To be notified when inferiors are started or exit under @value{GDBN}'s
2494 control use @w{@code{set print inferior-events}}:
2497 @kindex set print inferior-events
2498 @cindex print messages on inferior start and exit
2499 @item set print inferior-events
2500 @itemx set print inferior-events on
2501 @itemx set print inferior-events off
2502 The @code{set print inferior-events} command allows you to enable or
2503 disable printing of messages when @value{GDBN} notices that new
2504 inferiors have started or that inferiors have exited or have been
2505 detached. By default, these messages will not be printed.
2507 @kindex show print inferior-events
2508 @item show print inferior-events
2509 Show whether messages will be printed when @value{GDBN} detects that
2510 inferiors have started, exited or have been detached.
2513 Many commands will work the same with multiple programs as with a
2514 single program: e.g., @code{print myglobal} will simply display the
2515 value of @code{myglobal} in the current inferior.
2518 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2519 get more info about the relationship of inferiors, programs, address
2520 spaces in a debug session. You can do that with the @w{@code{maint
2521 info program-spaces}} command.
2524 @kindex maint info program-spaces
2525 @item maint info program-spaces
2526 Print a list of all program spaces currently being managed by
2529 @value{GDBN} displays for each program space (in this order):
2533 the program space number assigned by @value{GDBN}
2536 the name of the executable loaded into the program space, with e.g.,
2537 the @code{file} command.
2542 An asterisk @samp{*} preceding the @value{GDBN} program space number
2543 indicates the current program space.
2545 In addition, below each program space line, @value{GDBN} prints extra
2546 information that isn't suitable to display in tabular form. For
2547 example, the list of inferiors bound to the program space.
2550 (@value{GDBP}) maint info program-spaces
2553 Bound inferiors: ID 1 (process 21561)
2557 Here we can see that no inferior is running the program @code{hello},
2558 while @code{process 21561} is running the program @code{goodbye}. On
2559 some targets, it is possible that multiple inferiors are bound to the
2560 same program space. The most common example is that of debugging both
2561 the parent and child processes of a @code{vfork} call. For example,
2564 (@value{GDBP}) maint info program-spaces
2567 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2570 Here, both inferior 2 and inferior 1 are running in the same program
2571 space as a result of inferior 1 having executed a @code{vfork} call.
2575 @section Debugging Programs with Multiple Threads
2577 @cindex threads of execution
2578 @cindex multiple threads
2579 @cindex switching threads
2580 In some operating systems, such as HP-UX and Solaris, a single program
2581 may have more than one @dfn{thread} of execution. The precise semantics
2582 of threads differ from one operating system to another, but in general
2583 the threads of a single program are akin to multiple processes---except
2584 that they share one address space (that is, they can all examine and
2585 modify the same variables). On the other hand, each thread has its own
2586 registers and execution stack, and perhaps private memory.
2588 @value{GDBN} provides these facilities for debugging multi-thread
2592 @item automatic notification of new threads
2593 @item @samp{thread @var{threadno}}, a command to switch among threads
2594 @item @samp{info threads}, a command to inquire about existing threads
2595 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2596 a command to apply a command to a list of threads
2597 @item thread-specific breakpoints
2598 @item @samp{set print thread-events}, which controls printing of
2599 messages on thread start and exit.
2600 @item @samp{set libthread-db-search-path @var{path}}, which lets
2601 the user specify which @code{libthread_db} to use if the default choice
2602 isn't compatible with the program.
2606 @emph{Warning:} These facilities are not yet available on every
2607 @value{GDBN} configuration where the operating system supports threads.
2608 If your @value{GDBN} does not support threads, these commands have no
2609 effect. For example, a system without thread support shows no output
2610 from @samp{info threads}, and always rejects the @code{thread} command,
2614 (@value{GDBP}) info threads
2615 (@value{GDBP}) thread 1
2616 Thread ID 1 not known. Use the "info threads" command to
2617 see the IDs of currently known threads.
2619 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2620 @c doesn't support threads"?
2623 @cindex focus of debugging
2624 @cindex current thread
2625 The @value{GDBN} thread debugging facility allows you to observe all
2626 threads while your program runs---but whenever @value{GDBN} takes
2627 control, one thread in particular is always the focus of debugging.
2628 This thread is called the @dfn{current thread}. Debugging commands show
2629 program information from the perspective of the current thread.
2631 @cindex @code{New} @var{systag} message
2632 @cindex thread identifier (system)
2633 @c FIXME-implementors!! It would be more helpful if the [New...] message
2634 @c included GDB's numeric thread handle, so you could just go to that
2635 @c thread without first checking `info threads'.
2636 Whenever @value{GDBN} detects a new thread in your program, it displays
2637 the target system's identification for the thread with a message in the
2638 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2639 whose form varies depending on the particular system. For example, on
2640 @sc{gnu}/Linux, you might see
2643 [New Thread 46912507313328 (LWP 25582)]
2647 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2648 the @var{systag} is simply something like @samp{process 368}, with no
2651 @c FIXME!! (1) Does the [New...] message appear even for the very first
2652 @c thread of a program, or does it only appear for the
2653 @c second---i.e.@: when it becomes obvious we have a multithread
2655 @c (2) *Is* there necessarily a first thread always? Or do some
2656 @c multithread systems permit starting a program with multiple
2657 @c threads ab initio?
2659 @cindex thread number
2660 @cindex thread identifier (GDB)
2661 For debugging purposes, @value{GDBN} associates its own thread
2662 number---always a single integer---with each thread in your program.
2665 @kindex info threads
2667 Display a summary of all threads currently in your
2668 program. @value{GDBN} displays for each thread (in this order):
2672 the thread number assigned by @value{GDBN}
2675 the target system's thread identifier (@var{systag})
2678 the current stack frame summary for that thread
2682 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2683 indicates the current thread.
2687 @c end table here to get a little more width for example
2690 (@value{GDBP}) info threads
2691 3 process 35 thread 27 0x34e5 in sigpause ()
2692 2 process 35 thread 23 0x34e5 in sigpause ()
2693 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2699 @cindex debugging multithreaded programs (on HP-UX)
2700 @cindex thread identifier (GDB), on HP-UX
2701 For debugging purposes, @value{GDBN} associates its own thread
2702 number---a small integer assigned in thread-creation order---with each
2703 thread in your program.
2705 @cindex @code{New} @var{systag} message, on HP-UX
2706 @cindex thread identifier (system), on HP-UX
2707 @c FIXME-implementors!! It would be more helpful if the [New...] message
2708 @c included GDB's numeric thread handle, so you could just go to that
2709 @c thread without first checking `info threads'.
2710 Whenever @value{GDBN} detects a new thread in your program, it displays
2711 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2712 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2713 whose form varies depending on the particular system. For example, on
2717 [New thread 2 (system thread 26594)]
2721 when @value{GDBN} notices a new thread.
2724 @kindex info threads (HP-UX)
2726 Display a summary of all threads currently in your
2727 program. @value{GDBN} displays for each thread (in this order):
2730 @item the thread number assigned by @value{GDBN}
2732 @item the target system's thread identifier (@var{systag})
2734 @item the current stack frame summary for that thread
2738 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2739 indicates the current thread.
2743 @c end table here to get a little more width for example
2746 (@value{GDBP}) info threads
2747 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2749 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2750 from /usr/lib/libc.2
2751 1 system thread 27905 0x7b003498 in _brk () \@*
2752 from /usr/lib/libc.2
2755 On Solaris, you can display more information about user threads with a
2756 Solaris-specific command:
2759 @item maint info sol-threads
2760 @kindex maint info sol-threads
2761 @cindex thread info (Solaris)
2762 Display info on Solaris user threads.
2766 @kindex thread @var{threadno}
2767 @item thread @var{threadno}
2768 Make thread number @var{threadno} the current thread. The command
2769 argument @var{threadno} is the internal @value{GDBN} thread number, as
2770 shown in the first field of the @samp{info threads} display.
2771 @value{GDBN} responds by displaying the system identifier of the thread
2772 you selected, and its current stack frame summary:
2775 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2776 (@value{GDBP}) thread 2
2777 [Switching to process 35 thread 23]
2778 0x34e5 in sigpause ()
2782 As with the @samp{[New @dots{}]} message, the form of the text after
2783 @samp{Switching to} depends on your system's conventions for identifying
2786 @kindex thread apply
2787 @cindex apply command to several threads
2788 @item thread apply [@var{threadno}] [@var{all}] @var{command}
2789 The @code{thread apply} command allows you to apply the named
2790 @var{command} to one or more threads. Specify the numbers of the
2791 threads that you want affected with the command argument
2792 @var{threadno}. It can be a single thread number, one of the numbers
2793 shown in the first field of the @samp{info threads} display; or it
2794 could be a range of thread numbers, as in @code{2-4}. To apply a
2795 command to all threads, type @kbd{thread apply all @var{command}}.
2797 @kindex set print thread-events
2798 @cindex print messages on thread start and exit
2799 @item set print thread-events
2800 @itemx set print thread-events on
2801 @itemx set print thread-events off
2802 The @code{set print thread-events} command allows you to enable or
2803 disable printing of messages when @value{GDBN} notices that new threads have
2804 started or that threads have exited. By default, these messages will
2805 be printed if detection of these events is supported by the target.
2806 Note that these messages cannot be disabled on all targets.
2808 @kindex show print thread-events
2809 @item show print thread-events
2810 Show whether messages will be printed when @value{GDBN} detects that threads
2811 have started and exited.
2814 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2815 more information about how @value{GDBN} behaves when you stop and start
2816 programs with multiple threads.
2818 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2819 watchpoints in programs with multiple threads.
2822 @kindex set libthread-db-search-path
2823 @cindex search path for @code{libthread_db}
2824 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2825 If this variable is set, @var{path} is a colon-separated list of
2826 directories @value{GDBN} will use to search for @code{libthread_db}.
2827 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2830 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2831 @code{libthread_db} library to obtain information about threads in the
2832 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2833 to find @code{libthread_db}. If that fails, @value{GDBN} will continue
2834 with default system shared library directories, and finally the directory
2835 from which @code{libpthread} was loaded in the inferior process.
2837 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2838 @value{GDBN} attempts to initialize it with the current inferior process.
2839 If this initialization fails (which could happen because of a version
2840 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2841 will unload @code{libthread_db}, and continue with the next directory.
2842 If none of @code{libthread_db} libraries initialize successfully,
2843 @value{GDBN} will issue a warning and thread debugging will be disabled.
2845 Setting @code{libthread-db-search-path} is currently implemented
2846 only on some platforms.
2848 @kindex show libthread-db-search-path
2849 @item show libthread-db-search-path
2850 Display current libthread_db search path.
2854 @section Debugging Forks
2856 @cindex fork, debugging programs which call
2857 @cindex multiple processes
2858 @cindex processes, multiple
2859 On most systems, @value{GDBN} has no special support for debugging
2860 programs which create additional processes using the @code{fork}
2861 function. When a program forks, @value{GDBN} will continue to debug the
2862 parent process and the child process will run unimpeded. If you have
2863 set a breakpoint in any code which the child then executes, the child
2864 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2865 will cause it to terminate.
2867 However, if you want to debug the child process there is a workaround
2868 which isn't too painful. Put a call to @code{sleep} in the code which
2869 the child process executes after the fork. It may be useful to sleep
2870 only if a certain environment variable is set, or a certain file exists,
2871 so that the delay need not occur when you don't want to run @value{GDBN}
2872 on the child. While the child is sleeping, use the @code{ps} program to
2873 get its process ID. Then tell @value{GDBN} (a new invocation of
2874 @value{GDBN} if you are also debugging the parent process) to attach to
2875 the child process (@pxref{Attach}). From that point on you can debug
2876 the child process just like any other process which you attached to.
2878 On some systems, @value{GDBN} provides support for debugging programs that
2879 create additional processes using the @code{fork} or @code{vfork} functions.
2880 Currently, the only platforms with this feature are HP-UX (11.x and later
2881 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2883 By default, when a program forks, @value{GDBN} will continue to debug
2884 the parent process and the child process will run unimpeded.
2886 If you want to follow the child process instead of the parent process,
2887 use the command @w{@code{set follow-fork-mode}}.
2890 @kindex set follow-fork-mode
2891 @item set follow-fork-mode @var{mode}
2892 Set the debugger response to a program call of @code{fork} or
2893 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2894 process. The @var{mode} argument can be:
2898 The original process is debugged after a fork. The child process runs
2899 unimpeded. This is the default.
2902 The new process is debugged after a fork. The parent process runs
2907 @kindex show follow-fork-mode
2908 @item show follow-fork-mode
2909 Display the current debugger response to a @code{fork} or @code{vfork} call.
2912 @cindex debugging multiple processes
2913 On Linux, if you want to debug both the parent and child processes, use the
2914 command @w{@code{set detach-on-fork}}.
2917 @kindex set detach-on-fork
2918 @item set detach-on-fork @var{mode}
2919 Tells gdb whether to detach one of the processes after a fork, or
2920 retain debugger control over them both.
2924 The child process (or parent process, depending on the value of
2925 @code{follow-fork-mode}) will be detached and allowed to run
2926 independently. This is the default.
2929 Both processes will be held under the control of @value{GDBN}.
2930 One process (child or parent, depending on the value of
2931 @code{follow-fork-mode}) is debugged as usual, while the other
2936 @kindex show detach-on-fork
2937 @item show detach-on-fork
2938 Show whether detach-on-fork mode is on/off.
2941 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
2942 will retain control of all forked processes (including nested forks).
2943 You can list the forked processes under the control of @value{GDBN} by
2944 using the @w{@code{info inferiors}} command, and switch from one fork
2945 to another by using the @code{inferior} command (@pxref{Inferiors and
2946 Programs, ,Debugging Multiple Inferiors and Programs}).
2948 To quit debugging one of the forked processes, you can either detach
2949 from it by using the @w{@code{detach inferior}} command (allowing it
2950 to run independently), or kill it using the @w{@code{kill inferior}}
2951 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
2954 If you ask to debug a child process and a @code{vfork} is followed by an
2955 @code{exec}, @value{GDBN} executes the new target up to the first
2956 breakpoint in the new target. If you have a breakpoint set on
2957 @code{main} in your original program, the breakpoint will also be set on
2958 the child process's @code{main}.
2960 On some systems, when a child process is spawned by @code{vfork}, you
2961 cannot debug the child or parent until an @code{exec} call completes.
2963 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2964 call executes, the new target restarts. To restart the parent
2965 process, use the @code{file} command with the parent executable name
2966 as its argument. By default, after an @code{exec} call executes,
2967 @value{GDBN} discards the symbols of the previous executable image.
2968 You can change this behaviour with the @w{@code{set follow-exec-mode}}
2972 @kindex set follow-exec-mode
2973 @item set follow-exec-mode @var{mode}
2975 Set debugger response to a program call of @code{exec}. An
2976 @code{exec} call replaces the program image of a process.
2978 @code{follow-exec-mode} can be:
2982 @value{GDBN} creates a new inferior and rebinds the process to this
2983 new inferior. The program the process was running before the
2984 @code{exec} call can be restarted afterwards by restarting the
2990 (@value{GDBP}) info inferiors
2992 Id Description Executable
2995 process 12020 is executing new program: prog2
2996 Program exited normally.
2997 (@value{GDBP}) info inferiors
2998 Id Description Executable
3004 @value{GDBN} keeps the process bound to the same inferior. The new
3005 executable image replaces the previous executable loaded in the
3006 inferior. Restarting the inferior after the @code{exec} call, with
3007 e.g., the @code{run} command, restarts the executable the process was
3008 running after the @code{exec} call. This is the default mode.
3013 (@value{GDBP}) info inferiors
3014 Id Description Executable
3017 process 12020 is executing new program: prog2
3018 Program exited normally.
3019 (@value{GDBP}) info inferiors
3020 Id Description Executable
3027 You can use the @code{catch} command to make @value{GDBN} stop whenever
3028 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3029 Catchpoints, ,Setting Catchpoints}.
3031 @node Checkpoint/Restart
3032 @section Setting a @emph{Bookmark} to Return to Later
3037 @cindex snapshot of a process
3038 @cindex rewind program state
3040 On certain operating systems@footnote{Currently, only
3041 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3042 program's state, called a @dfn{checkpoint}, and come back to it
3045 Returning to a checkpoint effectively undoes everything that has
3046 happened in the program since the @code{checkpoint} was saved. This
3047 includes changes in memory, registers, and even (within some limits)
3048 system state. Effectively, it is like going back in time to the
3049 moment when the checkpoint was saved.
3051 Thus, if you're stepping thru a program and you think you're
3052 getting close to the point where things go wrong, you can save
3053 a checkpoint. Then, if you accidentally go too far and miss
3054 the critical statement, instead of having to restart your program
3055 from the beginning, you can just go back to the checkpoint and
3056 start again from there.
3058 This can be especially useful if it takes a lot of time or
3059 steps to reach the point where you think the bug occurs.
3061 To use the @code{checkpoint}/@code{restart} method of debugging:
3066 Save a snapshot of the debugged program's current execution state.
3067 The @code{checkpoint} command takes no arguments, but each checkpoint
3068 is assigned a small integer id, similar to a breakpoint id.
3070 @kindex info checkpoints
3071 @item info checkpoints
3072 List the checkpoints that have been saved in the current debugging
3073 session. For each checkpoint, the following information will be
3080 @item Source line, or label
3083 @kindex restart @var{checkpoint-id}
3084 @item restart @var{checkpoint-id}
3085 Restore the program state that was saved as checkpoint number
3086 @var{checkpoint-id}. All program variables, registers, stack frames
3087 etc.@: will be returned to the values that they had when the checkpoint
3088 was saved. In essence, gdb will ``wind back the clock'' to the point
3089 in time when the checkpoint was saved.
3091 Note that breakpoints, @value{GDBN} variables, command history etc.
3092 are not affected by restoring a checkpoint. In general, a checkpoint
3093 only restores things that reside in the program being debugged, not in
3096 @kindex delete checkpoint @var{checkpoint-id}
3097 @item delete checkpoint @var{checkpoint-id}
3098 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3102 Returning to a previously saved checkpoint will restore the user state
3103 of the program being debugged, plus a significant subset of the system
3104 (OS) state, including file pointers. It won't ``un-write'' data from
3105 a file, but it will rewind the file pointer to the previous location,
3106 so that the previously written data can be overwritten. For files
3107 opened in read mode, the pointer will also be restored so that the
3108 previously read data can be read again.
3110 Of course, characters that have been sent to a printer (or other
3111 external device) cannot be ``snatched back'', and characters received
3112 from eg.@: a serial device can be removed from internal program buffers,
3113 but they cannot be ``pushed back'' into the serial pipeline, ready to
3114 be received again. Similarly, the actual contents of files that have
3115 been changed cannot be restored (at this time).
3117 However, within those constraints, you actually can ``rewind'' your
3118 program to a previously saved point in time, and begin debugging it
3119 again --- and you can change the course of events so as to debug a
3120 different execution path this time.
3122 @cindex checkpoints and process id
3123 Finally, there is one bit of internal program state that will be
3124 different when you return to a checkpoint --- the program's process
3125 id. Each checkpoint will have a unique process id (or @var{pid}),
3126 and each will be different from the program's original @var{pid}.
3127 If your program has saved a local copy of its process id, this could
3128 potentially pose a problem.
3130 @subsection A Non-obvious Benefit of Using Checkpoints
3132 On some systems such as @sc{gnu}/Linux, address space randomization
3133 is performed on new processes for security reasons. This makes it
3134 difficult or impossible to set a breakpoint, or watchpoint, on an
3135 absolute address if you have to restart the program, since the
3136 absolute location of a symbol will change from one execution to the
3139 A checkpoint, however, is an @emph{identical} copy of a process.
3140 Therefore if you create a checkpoint at (eg.@:) the start of main,
3141 and simply return to that checkpoint instead of restarting the
3142 process, you can avoid the effects of address randomization and
3143 your symbols will all stay in the same place.
3146 @chapter Stopping and Continuing
3148 The principal purposes of using a debugger are so that you can stop your
3149 program before it terminates; or so that, if your program runs into
3150 trouble, you can investigate and find out why.
3152 Inside @value{GDBN}, your program may stop for any of several reasons,
3153 such as a signal, a breakpoint, or reaching a new line after a
3154 @value{GDBN} command such as @code{step}. You may then examine and
3155 change variables, set new breakpoints or remove old ones, and then
3156 continue execution. Usually, the messages shown by @value{GDBN} provide
3157 ample explanation of the status of your program---but you can also
3158 explicitly request this information at any time.
3161 @kindex info program
3163 Display information about the status of your program: whether it is
3164 running or not, what process it is, and why it stopped.
3168 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3169 * Continuing and Stepping:: Resuming execution
3171 * Thread Stops:: Stopping and starting multi-thread programs
3175 @section Breakpoints, Watchpoints, and Catchpoints
3178 A @dfn{breakpoint} makes your program stop whenever a certain point in
3179 the program is reached. For each breakpoint, you can add conditions to
3180 control in finer detail whether your program stops. You can set
3181 breakpoints with the @code{break} command and its variants (@pxref{Set
3182 Breaks, ,Setting Breakpoints}), to specify the place where your program
3183 should stop by line number, function name or exact address in the
3186 On some systems, you can set breakpoints in shared libraries before
3187 the executable is run. There is a minor limitation on HP-UX systems:
3188 you must wait until the executable is run in order to set breakpoints
3189 in shared library routines that are not called directly by the program
3190 (for example, routines that are arguments in a @code{pthread_create}
3194 @cindex data breakpoints
3195 @cindex memory tracing
3196 @cindex breakpoint on memory address
3197 @cindex breakpoint on variable modification
3198 A @dfn{watchpoint} is a special breakpoint that stops your program
3199 when the value of an expression changes. The expression may be a value
3200 of a variable, or it could involve values of one or more variables
3201 combined by operators, such as @samp{a + b}. This is sometimes called
3202 @dfn{data breakpoints}. You must use a different command to set
3203 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3204 from that, you can manage a watchpoint like any other breakpoint: you
3205 enable, disable, and delete both breakpoints and watchpoints using the
3208 You can arrange to have values from your program displayed automatically
3209 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3213 @cindex breakpoint on events
3214 A @dfn{catchpoint} is another special breakpoint that stops your program
3215 when a certain kind of event occurs, such as the throwing of a C@t{++}
3216 exception or the loading of a library. As with watchpoints, you use a
3217 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3218 Catchpoints}), but aside from that, you can manage a catchpoint like any
3219 other breakpoint. (To stop when your program receives a signal, use the
3220 @code{handle} command; see @ref{Signals, ,Signals}.)
3222 @cindex breakpoint numbers
3223 @cindex numbers for breakpoints
3224 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3225 catchpoint when you create it; these numbers are successive integers
3226 starting with one. In many of the commands for controlling various
3227 features of breakpoints you use the breakpoint number to say which
3228 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3229 @dfn{disabled}; if disabled, it has no effect on your program until you
3232 @cindex breakpoint ranges
3233 @cindex ranges of breakpoints
3234 Some @value{GDBN} commands accept a range of breakpoints on which to
3235 operate. A breakpoint range is either a single breakpoint number, like
3236 @samp{5}, or two such numbers, in increasing order, separated by a
3237 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3238 all breakpoints in that range are operated on.
3241 * Set Breaks:: Setting breakpoints
3242 * Set Watchpoints:: Setting watchpoints
3243 * Set Catchpoints:: Setting catchpoints
3244 * Delete Breaks:: Deleting breakpoints
3245 * Disabling:: Disabling breakpoints
3246 * Conditions:: Break conditions
3247 * Break Commands:: Breakpoint command lists
3248 * Error in Breakpoints:: ``Cannot insert breakpoints''
3249 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3253 @subsection Setting Breakpoints
3255 @c FIXME LMB what does GDB do if no code on line of breakpt?
3256 @c consider in particular declaration with/without initialization.
3258 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3261 @kindex b @r{(@code{break})}
3262 @vindex $bpnum@r{, convenience variable}
3263 @cindex latest breakpoint
3264 Breakpoints are set with the @code{break} command (abbreviated
3265 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3266 number of the breakpoint you've set most recently; see @ref{Convenience
3267 Vars,, Convenience Variables}, for a discussion of what you can do with
3268 convenience variables.
3271 @item break @var{location}
3272 Set a breakpoint at the given @var{location}, which can specify a
3273 function name, a line number, or an address of an instruction.
3274 (@xref{Specify Location}, for a list of all the possible ways to
3275 specify a @var{location}.) The breakpoint will stop your program just
3276 before it executes any of the code in the specified @var{location}.
3278 When using source languages that permit overloading of symbols, such as
3279 C@t{++}, a function name may refer to more than one possible place to break.
3280 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3283 It is also possible to insert a breakpoint that will stop the program
3284 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3285 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3288 When called without any arguments, @code{break} sets a breakpoint at
3289 the next instruction to be executed in the selected stack frame
3290 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3291 innermost, this makes your program stop as soon as control
3292 returns to that frame. This is similar to the effect of a
3293 @code{finish} command in the frame inside the selected frame---except
3294 that @code{finish} does not leave an active breakpoint. If you use
3295 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3296 the next time it reaches the current location; this may be useful
3299 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3300 least one instruction has been executed. If it did not do this, you
3301 would be unable to proceed past a breakpoint without first disabling the
3302 breakpoint. This rule applies whether or not the breakpoint already
3303 existed when your program stopped.
3305 @item break @dots{} if @var{cond}
3306 Set a breakpoint with condition @var{cond}; evaluate the expression
3307 @var{cond} each time the breakpoint is reached, and stop only if the
3308 value is nonzero---that is, if @var{cond} evaluates as true.
3309 @samp{@dots{}} stands for one of the possible arguments described
3310 above (or no argument) specifying where to break. @xref{Conditions,
3311 ,Break Conditions}, for more information on breakpoint conditions.
3314 @item tbreak @var{args}
3315 Set a breakpoint enabled only for one stop. @var{args} are the
3316 same as for the @code{break} command, and the breakpoint is set in the same
3317 way, but the breakpoint is automatically deleted after the first time your
3318 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3321 @cindex hardware breakpoints
3322 @item hbreak @var{args}
3323 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3324 @code{break} command and the breakpoint is set in the same way, but the
3325 breakpoint requires hardware support and some target hardware may not
3326 have this support. The main purpose of this is EPROM/ROM code
3327 debugging, so you can set a breakpoint at an instruction without
3328 changing the instruction. This can be used with the new trap-generation
3329 provided by SPARClite DSU and most x86-based targets. These targets
3330 will generate traps when a program accesses some data or instruction
3331 address that is assigned to the debug registers. However the hardware
3332 breakpoint registers can take a limited number of breakpoints. For
3333 example, on the DSU, only two data breakpoints can be set at a time, and
3334 @value{GDBN} will reject this command if more than two are used. Delete
3335 or disable unused hardware breakpoints before setting new ones
3336 (@pxref{Disabling, ,Disabling Breakpoints}).
3337 @xref{Conditions, ,Break Conditions}.
3338 For remote targets, you can restrict the number of hardware
3339 breakpoints @value{GDBN} will use, see @ref{set remote
3340 hardware-breakpoint-limit}.
3343 @item thbreak @var{args}
3344 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3345 are the same as for the @code{hbreak} command and the breakpoint is set in
3346 the same way. However, like the @code{tbreak} command,
3347 the breakpoint is automatically deleted after the
3348 first time your program stops there. Also, like the @code{hbreak}
3349 command, the breakpoint requires hardware support and some target hardware
3350 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3351 See also @ref{Conditions, ,Break Conditions}.
3354 @cindex regular expression
3355 @cindex breakpoints in functions matching a regexp
3356 @cindex set breakpoints in many functions
3357 @item rbreak @var{regex}
3358 Set breakpoints on all functions matching the regular expression
3359 @var{regex}. This command sets an unconditional breakpoint on all
3360 matches, printing a list of all breakpoints it set. Once these
3361 breakpoints are set, they are treated just like the breakpoints set with
3362 the @code{break} command. You can delete them, disable them, or make
3363 them conditional the same way as any other breakpoint.
3365 The syntax of the regular expression is the standard one used with tools
3366 like @file{grep}. Note that this is different from the syntax used by
3367 shells, so for instance @code{foo*} matches all functions that include
3368 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3369 @code{.*} leading and trailing the regular expression you supply, so to
3370 match only functions that begin with @code{foo}, use @code{^foo}.
3372 @cindex non-member C@t{++} functions, set breakpoint in
3373 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3374 breakpoints on overloaded functions that are not members of any special
3377 @cindex set breakpoints on all functions
3378 The @code{rbreak} command can be used to set breakpoints in
3379 @strong{all} the functions in a program, like this:
3382 (@value{GDBP}) rbreak .
3385 @kindex info breakpoints
3386 @cindex @code{$_} and @code{info breakpoints}
3387 @item info breakpoints @r{[}@var{n}@r{]}
3388 @itemx info break @r{[}@var{n}@r{]}
3389 @itemx info watchpoints @r{[}@var{n}@r{]}
3390 Print a table of all breakpoints, watchpoints, and catchpoints set and
3391 not deleted. Optional argument @var{n} means print information only
3392 about the specified breakpoint (or watchpoint or catchpoint). For
3393 each breakpoint, following columns are printed:
3396 @item Breakpoint Numbers
3398 Breakpoint, watchpoint, or catchpoint.
3400 Whether the breakpoint is marked to be disabled or deleted when hit.
3401 @item Enabled or Disabled
3402 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3403 that are not enabled.
3405 Where the breakpoint is in your program, as a memory address. For a
3406 pending breakpoint whose address is not yet known, this field will
3407 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3408 library that has the symbol or line referred by breakpoint is loaded.
3409 See below for details. A breakpoint with several locations will
3410 have @samp{<MULTIPLE>} in this field---see below for details.
3412 Where the breakpoint is in the source for your program, as a file and
3413 line number. For a pending breakpoint, the original string passed to
3414 the breakpoint command will be listed as it cannot be resolved until
3415 the appropriate shared library is loaded in the future.
3419 If a breakpoint is conditional, @code{info break} shows the condition on
3420 the line following the affected breakpoint; breakpoint commands, if any,
3421 are listed after that. A pending breakpoint is allowed to have a condition
3422 specified for it. The condition is not parsed for validity until a shared
3423 library is loaded that allows the pending breakpoint to resolve to a
3427 @code{info break} with a breakpoint
3428 number @var{n} as argument lists only that breakpoint. The
3429 convenience variable @code{$_} and the default examining-address for
3430 the @code{x} command are set to the address of the last breakpoint
3431 listed (@pxref{Memory, ,Examining Memory}).
3434 @code{info break} displays a count of the number of times the breakpoint
3435 has been hit. This is especially useful in conjunction with the
3436 @code{ignore} command. You can ignore a large number of breakpoint
3437 hits, look at the breakpoint info to see how many times the breakpoint
3438 was hit, and then run again, ignoring one less than that number. This
3439 will get you quickly to the last hit of that breakpoint.
3442 @value{GDBN} allows you to set any number of breakpoints at the same place in
3443 your program. There is nothing silly or meaningless about this. When
3444 the breakpoints are conditional, this is even useful
3445 (@pxref{Conditions, ,Break Conditions}).
3447 @cindex multiple locations, breakpoints
3448 @cindex breakpoints, multiple locations
3449 It is possible that a breakpoint corresponds to several locations
3450 in your program. Examples of this situation are:
3454 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3455 instances of the function body, used in different cases.
3458 For a C@t{++} template function, a given line in the function can
3459 correspond to any number of instantiations.
3462 For an inlined function, a given source line can correspond to
3463 several places where that function is inlined.
3466 In all those cases, @value{GDBN} will insert a breakpoint at all
3467 the relevant locations@footnote{
3468 As of this writing, multiple-location breakpoints work only if there's
3469 line number information for all the locations. This means that they
3470 will generally not work in system libraries, unless you have debug
3471 info with line numbers for them.}.
3473 A breakpoint with multiple locations is displayed in the breakpoint
3474 table using several rows---one header row, followed by one row for
3475 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3476 address column. The rows for individual locations contain the actual
3477 addresses for locations, and show the functions to which those
3478 locations belong. The number column for a location is of the form
3479 @var{breakpoint-number}.@var{location-number}.
3484 Num Type Disp Enb Address What
3485 1 breakpoint keep y <MULTIPLE>
3487 breakpoint already hit 1 time
3488 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3489 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3492 Each location can be individually enabled or disabled by passing
3493 @var{breakpoint-number}.@var{location-number} as argument to the
3494 @code{enable} and @code{disable} commands. Note that you cannot
3495 delete the individual locations from the list, you can only delete the
3496 entire list of locations that belong to their parent breakpoint (with
3497 the @kbd{delete @var{num}} command, where @var{num} is the number of
3498 the parent breakpoint, 1 in the above example). Disabling or enabling
3499 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3500 that belong to that breakpoint.
3502 @cindex pending breakpoints
3503 It's quite common to have a breakpoint inside a shared library.
3504 Shared libraries can be loaded and unloaded explicitly,
3505 and possibly repeatedly, as the program is executed. To support
3506 this use case, @value{GDBN} updates breakpoint locations whenever
3507 any shared library is loaded or unloaded. Typically, you would
3508 set a breakpoint in a shared library at the beginning of your
3509 debugging session, when the library is not loaded, and when the
3510 symbols from the library are not available. When you try to set
3511 breakpoint, @value{GDBN} will ask you if you want to set
3512 a so called @dfn{pending breakpoint}---breakpoint whose address
3513 is not yet resolved.
3515 After the program is run, whenever a new shared library is loaded,
3516 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3517 shared library contains the symbol or line referred to by some
3518 pending breakpoint, that breakpoint is resolved and becomes an
3519 ordinary breakpoint. When a library is unloaded, all breakpoints
3520 that refer to its symbols or source lines become pending again.
3522 This logic works for breakpoints with multiple locations, too. For
3523 example, if you have a breakpoint in a C@t{++} template function, and
3524 a newly loaded shared library has an instantiation of that template,
3525 a new location is added to the list of locations for the breakpoint.
3527 Except for having unresolved address, pending breakpoints do not
3528 differ from regular breakpoints. You can set conditions or commands,
3529 enable and disable them and perform other breakpoint operations.
3531 @value{GDBN} provides some additional commands for controlling what
3532 happens when the @samp{break} command cannot resolve breakpoint
3533 address specification to an address:
3535 @kindex set breakpoint pending
3536 @kindex show breakpoint pending
3538 @item set breakpoint pending auto
3539 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3540 location, it queries you whether a pending breakpoint should be created.
3542 @item set breakpoint pending on
3543 This indicates that an unrecognized breakpoint location should automatically
3544 result in a pending breakpoint being created.
3546 @item set breakpoint pending off
3547 This indicates that pending breakpoints are not to be created. Any
3548 unrecognized breakpoint location results in an error. This setting does
3549 not affect any pending breakpoints previously created.
3551 @item show breakpoint pending
3552 Show the current behavior setting for creating pending breakpoints.
3555 The settings above only affect the @code{break} command and its
3556 variants. Once breakpoint is set, it will be automatically updated
3557 as shared libraries are loaded and unloaded.
3559 @cindex automatic hardware breakpoints
3560 For some targets, @value{GDBN} can automatically decide if hardware or
3561 software breakpoints should be used, depending on whether the
3562 breakpoint address is read-only or read-write. This applies to
3563 breakpoints set with the @code{break} command as well as to internal
3564 breakpoints set by commands like @code{next} and @code{finish}. For
3565 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3568 You can control this automatic behaviour with the following commands::
3570 @kindex set breakpoint auto-hw
3571 @kindex show breakpoint auto-hw
3573 @item set breakpoint auto-hw on
3574 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3575 will try to use the target memory map to decide if software or hardware
3576 breakpoint must be used.
3578 @item set breakpoint auto-hw off
3579 This indicates @value{GDBN} should not automatically select breakpoint
3580 type. If the target provides a memory map, @value{GDBN} will warn when
3581 trying to set software breakpoint at a read-only address.
3584 @value{GDBN} normally implements breakpoints by replacing the program code
3585 at the breakpoint address with a special instruction, which, when
3586 executed, given control to the debugger. By default, the program
3587 code is so modified only when the program is resumed. As soon as
3588 the program stops, @value{GDBN} restores the original instructions. This
3589 behaviour guards against leaving breakpoints inserted in the
3590 target should gdb abrubptly disconnect. However, with slow remote
3591 targets, inserting and removing breakpoint can reduce the performance.
3592 This behavior can be controlled with the following commands::
3594 @kindex set breakpoint always-inserted
3595 @kindex show breakpoint always-inserted
3597 @item set breakpoint always-inserted off
3598 All breakpoints, including newly added by the user, are inserted in
3599 the target only when the target is resumed. All breakpoints are
3600 removed from the target when it stops.
3602 @item set breakpoint always-inserted on
3603 Causes all breakpoints to be inserted in the target at all times. If
3604 the user adds a new breakpoint, or changes an existing breakpoint, the
3605 breakpoints in the target are updated immediately. A breakpoint is
3606 removed from the target only when breakpoint itself is removed.
3608 @cindex non-stop mode, and @code{breakpoint always-inserted}
3609 @item set breakpoint always-inserted auto
3610 This is the default mode. If @value{GDBN} is controlling the inferior
3611 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3612 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3613 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3614 @code{breakpoint always-inserted} mode is off.
3617 @cindex negative breakpoint numbers
3618 @cindex internal @value{GDBN} breakpoints
3619 @value{GDBN} itself sometimes sets breakpoints in your program for
3620 special purposes, such as proper handling of @code{longjmp} (in C
3621 programs). These internal breakpoints are assigned negative numbers,
3622 starting with @code{-1}; @samp{info breakpoints} does not display them.
3623 You can see these breakpoints with the @value{GDBN} maintenance command
3624 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3627 @node Set Watchpoints
3628 @subsection Setting Watchpoints
3630 @cindex setting watchpoints
3631 You can use a watchpoint to stop execution whenever the value of an
3632 expression changes, without having to predict a particular place where
3633 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3634 The expression may be as simple as the value of a single variable, or
3635 as complex as many variables combined by operators. Examples include:
3639 A reference to the value of a single variable.
3642 An address cast to an appropriate data type. For example,
3643 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3644 address (assuming an @code{int} occupies 4 bytes).
3647 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3648 expression can use any operators valid in the program's native
3649 language (@pxref{Languages}).
3652 You can set a watchpoint on an expression even if the expression can
3653 not be evaluated yet. For instance, you can set a watchpoint on
3654 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3655 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3656 the expression produces a valid value. If the expression becomes
3657 valid in some other way than changing a variable (e.g.@: if the memory
3658 pointed to by @samp{*global_ptr} becomes readable as the result of a
3659 @code{malloc} call), @value{GDBN} may not stop until the next time
3660 the expression changes.
3662 @cindex software watchpoints
3663 @cindex hardware watchpoints
3664 Depending on your system, watchpoints may be implemented in software or
3665 hardware. @value{GDBN} does software watchpointing by single-stepping your
3666 program and testing the variable's value each time, which is hundreds of
3667 times slower than normal execution. (But this may still be worth it, to
3668 catch errors where you have no clue what part of your program is the
3671 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3672 x86-based targets, @value{GDBN} includes support for hardware
3673 watchpoints, which do not slow down the running of your program.
3677 @item watch @var{expr} @r{[}thread @var{threadnum}@r{]}
3678 Set a watchpoint for an expression. @value{GDBN} will break when the
3679 expression @var{expr} is written into by the program and its value
3680 changes. The simplest (and the most popular) use of this command is
3681 to watch the value of a single variable:
3684 (@value{GDBP}) watch foo
3687 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3688 clause, @value{GDBN} breaks only when the thread identified by
3689 @var{threadnum} changes the value of @var{expr}. If any other threads
3690 change the value of @var{expr}, @value{GDBN} will not break. Note
3691 that watchpoints restricted to a single thread in this way only work
3692 with Hardware Watchpoints.
3695 @item rwatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3696 Set a watchpoint that will break when the value of @var{expr} is read
3700 @item awatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3701 Set a watchpoint that will break when @var{expr} is either read from
3702 or written into by the program.
3704 @kindex info watchpoints @r{[}@var{n}@r{]}
3705 @item info watchpoints
3706 This command prints a list of watchpoints, breakpoints, and catchpoints;
3707 it is the same as @code{info break} (@pxref{Set Breaks}).
3710 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3711 watchpoints execute very quickly, and the debugger reports a change in
3712 value at the exact instruction where the change occurs. If @value{GDBN}
3713 cannot set a hardware watchpoint, it sets a software watchpoint, which
3714 executes more slowly and reports the change in value at the next
3715 @emph{statement}, not the instruction, after the change occurs.
3717 @cindex use only software watchpoints
3718 You can force @value{GDBN} to use only software watchpoints with the
3719 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3720 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3721 the underlying system supports them. (Note that hardware-assisted
3722 watchpoints that were set @emph{before} setting
3723 @code{can-use-hw-watchpoints} to zero will still use the hardware
3724 mechanism of watching expression values.)
3727 @item set can-use-hw-watchpoints
3728 @kindex set can-use-hw-watchpoints
3729 Set whether or not to use hardware watchpoints.
3731 @item show can-use-hw-watchpoints
3732 @kindex show can-use-hw-watchpoints
3733 Show the current mode of using hardware watchpoints.
3736 For remote targets, you can restrict the number of hardware
3737 watchpoints @value{GDBN} will use, see @ref{set remote
3738 hardware-breakpoint-limit}.
3740 When you issue the @code{watch} command, @value{GDBN} reports
3743 Hardware watchpoint @var{num}: @var{expr}
3747 if it was able to set a hardware watchpoint.
3749 Currently, the @code{awatch} and @code{rwatch} commands can only set
3750 hardware watchpoints, because accesses to data that don't change the
3751 value of the watched expression cannot be detected without examining
3752 every instruction as it is being executed, and @value{GDBN} does not do
3753 that currently. If @value{GDBN} finds that it is unable to set a
3754 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3755 will print a message like this:
3758 Expression cannot be implemented with read/access watchpoint.
3761 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3762 data type of the watched expression is wider than what a hardware
3763 watchpoint on the target machine can handle. For example, some systems
3764 can only watch regions that are up to 4 bytes wide; on such systems you
3765 cannot set hardware watchpoints for an expression that yields a
3766 double-precision floating-point number (which is typically 8 bytes
3767 wide). As a work-around, it might be possible to break the large region
3768 into a series of smaller ones and watch them with separate watchpoints.
3770 If you set too many hardware watchpoints, @value{GDBN} might be unable
3771 to insert all of them when you resume the execution of your program.
3772 Since the precise number of active watchpoints is unknown until such
3773 time as the program is about to be resumed, @value{GDBN} might not be
3774 able to warn you about this when you set the watchpoints, and the
3775 warning will be printed only when the program is resumed:
3778 Hardware watchpoint @var{num}: Could not insert watchpoint
3782 If this happens, delete or disable some of the watchpoints.
3784 Watching complex expressions that reference many variables can also
3785 exhaust the resources available for hardware-assisted watchpoints.
3786 That's because @value{GDBN} needs to watch every variable in the
3787 expression with separately allocated resources.
3789 If you call a function interactively using @code{print} or @code{call},
3790 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3791 kind of breakpoint or the call completes.
3793 @value{GDBN} automatically deletes watchpoints that watch local
3794 (automatic) variables, or expressions that involve such variables, when
3795 they go out of scope, that is, when the execution leaves the block in
3796 which these variables were defined. In particular, when the program
3797 being debugged terminates, @emph{all} local variables go out of scope,
3798 and so only watchpoints that watch global variables remain set. If you
3799 rerun the program, you will need to set all such watchpoints again. One
3800 way of doing that would be to set a code breakpoint at the entry to the
3801 @code{main} function and when it breaks, set all the watchpoints.
3803 @cindex watchpoints and threads
3804 @cindex threads and watchpoints
3805 In multi-threaded programs, watchpoints will detect changes to the
3806 watched expression from every thread.
3809 @emph{Warning:} In multi-threaded programs, software watchpoints
3810 have only limited usefulness. If @value{GDBN} creates a software
3811 watchpoint, it can only watch the value of an expression @emph{in a
3812 single thread}. If you are confident that the expression can only
3813 change due to the current thread's activity (and if you are also
3814 confident that no other thread can become current), then you can use
3815 software watchpoints as usual. However, @value{GDBN} may not notice
3816 when a non-current thread's activity changes the expression. (Hardware
3817 watchpoints, in contrast, watch an expression in all threads.)
3820 @xref{set remote hardware-watchpoint-limit}.
3822 @node Set Catchpoints
3823 @subsection Setting Catchpoints
3824 @cindex catchpoints, setting
3825 @cindex exception handlers
3826 @cindex event handling
3828 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3829 kinds of program events, such as C@t{++} exceptions or the loading of a
3830 shared library. Use the @code{catch} command to set a catchpoint.
3834 @item catch @var{event}
3835 Stop when @var{event} occurs. @var{event} can be any of the following:
3838 @cindex stop on C@t{++} exceptions
3839 The throwing of a C@t{++} exception.
3842 The catching of a C@t{++} exception.
3845 @cindex Ada exception catching
3846 @cindex catch Ada exceptions
3847 An Ada exception being raised. If an exception name is specified
3848 at the end of the command (eg @code{catch exception Program_Error}),
3849 the debugger will stop only when this specific exception is raised.
3850 Otherwise, the debugger stops execution when any Ada exception is raised.
3852 When inserting an exception catchpoint on a user-defined exception whose
3853 name is identical to one of the exceptions defined by the language, the
3854 fully qualified name must be used as the exception name. Otherwise,
3855 @value{GDBN} will assume that it should stop on the pre-defined exception
3856 rather than the user-defined one. For instance, assuming an exception
3857 called @code{Constraint_Error} is defined in package @code{Pck}, then
3858 the command to use to catch such exceptions is @kbd{catch exception
3859 Pck.Constraint_Error}.
3861 @item exception unhandled
3862 An exception that was raised but is not handled by the program.
3865 A failed Ada assertion.
3868 @cindex break on fork/exec
3869 A call to @code{exec}. This is currently only available for HP-UX
3873 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @r{...}
3874 @cindex break on a system call.
3875 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
3876 syscall is a mechanism for application programs to request a service
3877 from the operating system (OS) or one of the OS system services.
3878 @value{GDBN} can catch some or all of the syscalls issued by the
3879 debuggee, and show the related information for each syscall. If no
3880 argument is specified, calls to and returns from all system calls
3883 @var{name} can be any system call name that is valid for the
3884 underlying OS. Just what syscalls are valid depends on the OS. On
3885 GNU and Unix systems, you can find the full list of valid syscall
3886 names on @file{/usr/include/asm/unistd.h}.
3888 @c For MS-Windows, the syscall names and the corresponding numbers
3889 @c can be found, e.g., on this URL:
3890 @c http://www.metasploit.com/users/opcode/syscalls.html
3891 @c but we don't support Windows syscalls yet.
3893 Normally, @value{GDBN} knows in advance which syscalls are valid for
3894 each OS, so you can use the @value{GDBN} command-line completion
3895 facilities (@pxref{Completion,, command completion}) to list the
3898 You may also specify the system call numerically. A syscall's
3899 number is the value passed to the OS's syscall dispatcher to
3900 identify the requested service. When you specify the syscall by its
3901 name, @value{GDBN} uses its database of syscalls to convert the name
3902 into the corresponding numeric code, but using the number directly
3903 may be useful if @value{GDBN}'s database does not have the complete
3904 list of syscalls on your system (e.g., because @value{GDBN} lags
3905 behind the OS upgrades).
3907 The example below illustrates how this command works if you don't provide
3911 (@value{GDBP}) catch syscall
3912 Catchpoint 1 (syscall)
3914 Starting program: /tmp/catch-syscall
3916 Catchpoint 1 (call to syscall 'close'), \
3917 0xffffe424 in __kernel_vsyscall ()
3921 Catchpoint 1 (returned from syscall 'close'), \
3922 0xffffe424 in __kernel_vsyscall ()
3926 Here is an example of catching a system call by name:
3929 (@value{GDBP}) catch syscall chroot
3930 Catchpoint 1 (syscall 'chroot' [61])
3932 Starting program: /tmp/catch-syscall
3934 Catchpoint 1 (call to syscall 'chroot'), \
3935 0xffffe424 in __kernel_vsyscall ()
3939 Catchpoint 1 (returned from syscall 'chroot'), \
3940 0xffffe424 in __kernel_vsyscall ()
3944 An example of specifying a system call numerically. In the case
3945 below, the syscall number has a corresponding entry in the XML
3946 file, so @value{GDBN} finds its name and prints it:
3949 (@value{GDBP}) catch syscall 252
3950 Catchpoint 1 (syscall(s) 'exit_group')
3952 Starting program: /tmp/catch-syscall
3954 Catchpoint 1 (call to syscall 'exit_group'), \
3955 0xffffe424 in __kernel_vsyscall ()
3959 Program exited normally.
3963 However, there can be situations when there is no corresponding name
3964 in XML file for that syscall number. In this case, @value{GDBN} prints
3965 a warning message saying that it was not able to find the syscall name,
3966 but the catchpoint will be set anyway. See the example below:
3969 (@value{GDBP}) catch syscall 764
3970 warning: The number '764' does not represent a known syscall.
3971 Catchpoint 2 (syscall 764)
3975 If you configure @value{GDBN} using the @samp{--without-expat} option,
3976 it will not be able to display syscall names. Also, if your
3977 architecture does not have an XML file describing its system calls,
3978 you will not be able to see the syscall names. It is important to
3979 notice that these two features are used for accessing the syscall
3980 name database. In either case, you will see a warning like this:
3983 (@value{GDBP}) catch syscall
3984 warning: Could not open "syscalls/i386-linux.xml"
3985 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
3986 GDB will not be able to display syscall names.
3987 Catchpoint 1 (syscall)
3991 Of course, the file name will change depending on your architecture and system.
3993 Still using the example above, you can also try to catch a syscall by its
3994 number. In this case, you would see something like:
3997 (@value{GDBP}) catch syscall 252
3998 Catchpoint 1 (syscall(s) 252)
4001 Again, in this case @value{GDBN} would not be able to display syscall's names.
4004 A call to @code{fork}. This is currently only available for HP-UX
4008 A call to @code{vfork}. This is currently only available for HP-UX
4013 @item tcatch @var{event}
4014 Set a catchpoint that is enabled only for one stop. The catchpoint is
4015 automatically deleted after the first time the event is caught.
4019 Use the @code{info break} command to list the current catchpoints.
4021 There are currently some limitations to C@t{++} exception handling
4022 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
4026 If you call a function interactively, @value{GDBN} normally returns
4027 control to you when the function has finished executing. If the call
4028 raises an exception, however, the call may bypass the mechanism that
4029 returns control to you and cause your program either to abort or to
4030 simply continue running until it hits a breakpoint, catches a signal
4031 that @value{GDBN} is listening for, or exits. This is the case even if
4032 you set a catchpoint for the exception; catchpoints on exceptions are
4033 disabled within interactive calls.
4036 You cannot raise an exception interactively.
4039 You cannot install an exception handler interactively.
4042 @cindex raise exceptions
4043 Sometimes @code{catch} is not the best way to debug exception handling:
4044 if you need to know exactly where an exception is raised, it is better to
4045 stop @emph{before} the exception handler is called, since that way you
4046 can see the stack before any unwinding takes place. If you set a
4047 breakpoint in an exception handler instead, it may not be easy to find
4048 out where the exception was raised.
4050 To stop just before an exception handler is called, you need some
4051 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
4052 raised by calling a library function named @code{__raise_exception}
4053 which has the following ANSI C interface:
4056 /* @var{addr} is where the exception identifier is stored.
4057 @var{id} is the exception identifier. */
4058 void __raise_exception (void **addr, void *id);
4062 To make the debugger catch all exceptions before any stack
4063 unwinding takes place, set a breakpoint on @code{__raise_exception}
4064 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
4066 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
4067 that depends on the value of @var{id}, you can stop your program when
4068 a specific exception is raised. You can use multiple conditional
4069 breakpoints to stop your program when any of a number of exceptions are
4074 @subsection Deleting Breakpoints
4076 @cindex clearing breakpoints, watchpoints, catchpoints
4077 @cindex deleting breakpoints, watchpoints, catchpoints
4078 It is often necessary to eliminate a breakpoint, watchpoint, or
4079 catchpoint once it has done its job and you no longer want your program
4080 to stop there. This is called @dfn{deleting} the breakpoint. A
4081 breakpoint that has been deleted no longer exists; it is forgotten.
4083 With the @code{clear} command you can delete breakpoints according to
4084 where they are in your program. With the @code{delete} command you can
4085 delete individual breakpoints, watchpoints, or catchpoints by specifying
4086 their breakpoint numbers.
4088 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4089 automatically ignores breakpoints on the first instruction to be executed
4090 when you continue execution without changing the execution address.
4095 Delete any breakpoints at the next instruction to be executed in the
4096 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4097 the innermost frame is selected, this is a good way to delete a
4098 breakpoint where your program just stopped.
4100 @item clear @var{location}
4101 Delete any breakpoints set at the specified @var{location}.
4102 @xref{Specify Location}, for the various forms of @var{location}; the
4103 most useful ones are listed below:
4106 @item clear @var{function}
4107 @itemx clear @var{filename}:@var{function}
4108 Delete any breakpoints set at entry to the named @var{function}.
4110 @item clear @var{linenum}
4111 @itemx clear @var{filename}:@var{linenum}
4112 Delete any breakpoints set at or within the code of the specified
4113 @var{linenum} of the specified @var{filename}.
4116 @cindex delete breakpoints
4118 @kindex d @r{(@code{delete})}
4119 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4120 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4121 ranges specified as arguments. If no argument is specified, delete all
4122 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4123 confirm off}). You can abbreviate this command as @code{d}.
4127 @subsection Disabling Breakpoints
4129 @cindex enable/disable a breakpoint
4130 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4131 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4132 it had been deleted, but remembers the information on the breakpoint so
4133 that you can @dfn{enable} it again later.
4135 You disable and enable breakpoints, watchpoints, and catchpoints with
4136 the @code{enable} and @code{disable} commands, optionally specifying one
4137 or more breakpoint numbers as arguments. Use @code{info break} or
4138 @code{info watch} to print a list of breakpoints, watchpoints, and
4139 catchpoints if you do not know which numbers to use.
4141 Disabling and enabling a breakpoint that has multiple locations
4142 affects all of its locations.
4144 A breakpoint, watchpoint, or catchpoint can have any of four different
4145 states of enablement:
4149 Enabled. The breakpoint stops your program. A breakpoint set
4150 with the @code{break} command starts out in this state.
4152 Disabled. The breakpoint has no effect on your program.
4154 Enabled once. The breakpoint stops your program, but then becomes
4157 Enabled for deletion. The breakpoint stops your program, but
4158 immediately after it does so it is deleted permanently. A breakpoint
4159 set with the @code{tbreak} command starts out in this state.
4162 You can use the following commands to enable or disable breakpoints,
4163 watchpoints, and catchpoints:
4167 @kindex dis @r{(@code{disable})}
4168 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4169 Disable the specified breakpoints---or all breakpoints, if none are
4170 listed. A disabled breakpoint has no effect but is not forgotten. All
4171 options such as ignore-counts, conditions and commands are remembered in
4172 case the breakpoint is enabled again later. You may abbreviate
4173 @code{disable} as @code{dis}.
4176 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4177 Enable the specified breakpoints (or all defined breakpoints). They
4178 become effective once again in stopping your program.
4180 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4181 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4182 of these breakpoints immediately after stopping your program.
4184 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4185 Enable the specified breakpoints to work once, then die. @value{GDBN}
4186 deletes any of these breakpoints as soon as your program stops there.
4187 Breakpoints set by the @code{tbreak} command start out in this state.
4190 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4191 @c confusing: tbreak is also initially enabled.
4192 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4193 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4194 subsequently, they become disabled or enabled only when you use one of
4195 the commands above. (The command @code{until} can set and delete a
4196 breakpoint of its own, but it does not change the state of your other
4197 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4201 @subsection Break Conditions
4202 @cindex conditional breakpoints
4203 @cindex breakpoint conditions
4205 @c FIXME what is scope of break condition expr? Context where wanted?
4206 @c in particular for a watchpoint?
4207 The simplest sort of breakpoint breaks every time your program reaches a
4208 specified place. You can also specify a @dfn{condition} for a
4209 breakpoint. A condition is just a Boolean expression in your
4210 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4211 a condition evaluates the expression each time your program reaches it,
4212 and your program stops only if the condition is @emph{true}.
4214 This is the converse of using assertions for program validation; in that
4215 situation, you want to stop when the assertion is violated---that is,
4216 when the condition is false. In C, if you want to test an assertion expressed
4217 by the condition @var{assert}, you should set the condition
4218 @samp{! @var{assert}} on the appropriate breakpoint.
4220 Conditions are also accepted for watchpoints; you may not need them,
4221 since a watchpoint is inspecting the value of an expression anyhow---but
4222 it might be simpler, say, to just set a watchpoint on a variable name,
4223 and specify a condition that tests whether the new value is an interesting
4226 Break conditions can have side effects, and may even call functions in
4227 your program. This can be useful, for example, to activate functions
4228 that log program progress, or to use your own print functions to
4229 format special data structures. The effects are completely predictable
4230 unless there is another enabled breakpoint at the same address. (In
4231 that case, @value{GDBN} might see the other breakpoint first and stop your
4232 program without checking the condition of this one.) Note that
4233 breakpoint commands are usually more convenient and flexible than break
4235 purpose of performing side effects when a breakpoint is reached
4236 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4238 Break conditions can be specified when a breakpoint is set, by using
4239 @samp{if} in the arguments to the @code{break} command. @xref{Set
4240 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4241 with the @code{condition} command.
4243 You can also use the @code{if} keyword with the @code{watch} command.
4244 The @code{catch} command does not recognize the @code{if} keyword;
4245 @code{condition} is the only way to impose a further condition on a
4250 @item condition @var{bnum} @var{expression}
4251 Specify @var{expression} as the break condition for breakpoint,
4252 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4253 breakpoint @var{bnum} stops your program only if the value of
4254 @var{expression} is true (nonzero, in C). When you use
4255 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4256 syntactic correctness, and to determine whether symbols in it have
4257 referents in the context of your breakpoint. If @var{expression} uses
4258 symbols not referenced in the context of the breakpoint, @value{GDBN}
4259 prints an error message:
4262 No symbol "foo" in current context.
4267 not actually evaluate @var{expression} at the time the @code{condition}
4268 command (or a command that sets a breakpoint with a condition, like
4269 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4271 @item condition @var{bnum}
4272 Remove the condition from breakpoint number @var{bnum}. It becomes
4273 an ordinary unconditional breakpoint.
4276 @cindex ignore count (of breakpoint)
4277 A special case of a breakpoint condition is to stop only when the
4278 breakpoint has been reached a certain number of times. This is so
4279 useful that there is a special way to do it, using the @dfn{ignore
4280 count} of the breakpoint. Every breakpoint has an ignore count, which
4281 is an integer. Most of the time, the ignore count is zero, and
4282 therefore has no effect. But if your program reaches a breakpoint whose
4283 ignore count is positive, then instead of stopping, it just decrements
4284 the ignore count by one and continues. As a result, if the ignore count
4285 value is @var{n}, the breakpoint does not stop the next @var{n} times
4286 your program reaches it.
4290 @item ignore @var{bnum} @var{count}
4291 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4292 The next @var{count} times the breakpoint is reached, your program's
4293 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4296 To make the breakpoint stop the next time it is reached, specify
4299 When you use @code{continue} to resume execution of your program from a
4300 breakpoint, you can specify an ignore count directly as an argument to
4301 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4302 Stepping,,Continuing and Stepping}.
4304 If a breakpoint has a positive ignore count and a condition, the
4305 condition is not checked. Once the ignore count reaches zero,
4306 @value{GDBN} resumes checking the condition.
4308 You could achieve the effect of the ignore count with a condition such
4309 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4310 is decremented each time. @xref{Convenience Vars, ,Convenience
4314 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4317 @node Break Commands
4318 @subsection Breakpoint Command Lists
4320 @cindex breakpoint commands
4321 You can give any breakpoint (or watchpoint or catchpoint) a series of
4322 commands to execute when your program stops due to that breakpoint. For
4323 example, you might want to print the values of certain expressions, or
4324 enable other breakpoints.
4328 @kindex end@r{ (breakpoint commands)}
4329 @item commands @r{[}@var{bnum}@r{]}
4330 @itemx @dots{} @var{command-list} @dots{}
4332 Specify a list of commands for breakpoint number @var{bnum}. The commands
4333 themselves appear on the following lines. Type a line containing just
4334 @code{end} to terminate the commands.
4336 To remove all commands from a breakpoint, type @code{commands} and
4337 follow it immediately with @code{end}; that is, give no commands.
4339 With no @var{bnum} argument, @code{commands} refers to the last
4340 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
4341 recently encountered).
4344 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4345 disabled within a @var{command-list}.
4347 You can use breakpoint commands to start your program up again. Simply
4348 use the @code{continue} command, or @code{step}, or any other command
4349 that resumes execution.
4351 Any other commands in the command list, after a command that resumes
4352 execution, are ignored. This is because any time you resume execution
4353 (even with a simple @code{next} or @code{step}), you may encounter
4354 another breakpoint---which could have its own command list, leading to
4355 ambiguities about which list to execute.
4358 If the first command you specify in a command list is @code{silent}, the
4359 usual message about stopping at a breakpoint is not printed. This may
4360 be desirable for breakpoints that are to print a specific message and
4361 then continue. If none of the remaining commands print anything, you
4362 see no sign that the breakpoint was reached. @code{silent} is
4363 meaningful only at the beginning of a breakpoint command list.
4365 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4366 print precisely controlled output, and are often useful in silent
4367 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4369 For example, here is how you could use breakpoint commands to print the
4370 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4376 printf "x is %d\n",x
4381 One application for breakpoint commands is to compensate for one bug so
4382 you can test for another. Put a breakpoint just after the erroneous line
4383 of code, give it a condition to detect the case in which something
4384 erroneous has been done, and give it commands to assign correct values
4385 to any variables that need them. End with the @code{continue} command
4386 so that your program does not stop, and start with the @code{silent}
4387 command so that no output is produced. Here is an example:
4398 @c @ifclear BARETARGET
4399 @node Error in Breakpoints
4400 @subsection ``Cannot insert breakpoints''
4402 If you request too many active hardware-assisted breakpoints and
4403 watchpoints, you will see this error message:
4405 @c FIXME: the precise wording of this message may change; the relevant
4406 @c source change is not committed yet (Sep 3, 1999).
4408 Stopped; cannot insert breakpoints.
4409 You may have requested too many hardware breakpoints and watchpoints.
4413 This message is printed when you attempt to resume the program, since
4414 only then @value{GDBN} knows exactly how many hardware breakpoints and
4415 watchpoints it needs to insert.
4417 When this message is printed, you need to disable or remove some of the
4418 hardware-assisted breakpoints and watchpoints, and then continue.
4420 @node Breakpoint-related Warnings
4421 @subsection ``Breakpoint address adjusted...''
4422 @cindex breakpoint address adjusted
4424 Some processor architectures place constraints on the addresses at
4425 which breakpoints may be placed. For architectures thus constrained,
4426 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4427 with the constraints dictated by the architecture.
4429 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4430 a VLIW architecture in which a number of RISC-like instructions may be
4431 bundled together for parallel execution. The FR-V architecture
4432 constrains the location of a breakpoint instruction within such a
4433 bundle to the instruction with the lowest address. @value{GDBN}
4434 honors this constraint by adjusting a breakpoint's address to the
4435 first in the bundle.
4437 It is not uncommon for optimized code to have bundles which contain
4438 instructions from different source statements, thus it may happen that
4439 a breakpoint's address will be adjusted from one source statement to
4440 another. Since this adjustment may significantly alter @value{GDBN}'s
4441 breakpoint related behavior from what the user expects, a warning is
4442 printed when the breakpoint is first set and also when the breakpoint
4445 A warning like the one below is printed when setting a breakpoint
4446 that's been subject to address adjustment:
4449 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4452 Such warnings are printed both for user settable and @value{GDBN}'s
4453 internal breakpoints. If you see one of these warnings, you should
4454 verify that a breakpoint set at the adjusted address will have the
4455 desired affect. If not, the breakpoint in question may be removed and
4456 other breakpoints may be set which will have the desired behavior.
4457 E.g., it may be sufficient to place the breakpoint at a later
4458 instruction. A conditional breakpoint may also be useful in some
4459 cases to prevent the breakpoint from triggering too often.
4461 @value{GDBN} will also issue a warning when stopping at one of these
4462 adjusted breakpoints:
4465 warning: Breakpoint 1 address previously adjusted from 0x00010414
4469 When this warning is encountered, it may be too late to take remedial
4470 action except in cases where the breakpoint is hit earlier or more
4471 frequently than expected.
4473 @node Continuing and Stepping
4474 @section Continuing and Stepping
4478 @cindex resuming execution
4479 @dfn{Continuing} means resuming program execution until your program
4480 completes normally. In contrast, @dfn{stepping} means executing just
4481 one more ``step'' of your program, where ``step'' may mean either one
4482 line of source code, or one machine instruction (depending on what
4483 particular command you use). Either when continuing or when stepping,
4484 your program may stop even sooner, due to a breakpoint or a signal. (If
4485 it stops due to a signal, you may want to use @code{handle}, or use
4486 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4490 @kindex c @r{(@code{continue})}
4491 @kindex fg @r{(resume foreground execution)}
4492 @item continue @r{[}@var{ignore-count}@r{]}
4493 @itemx c @r{[}@var{ignore-count}@r{]}
4494 @itemx fg @r{[}@var{ignore-count}@r{]}
4495 Resume program execution, at the address where your program last stopped;
4496 any breakpoints set at that address are bypassed. The optional argument
4497 @var{ignore-count} allows you to specify a further number of times to
4498 ignore a breakpoint at this location; its effect is like that of
4499 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4501 The argument @var{ignore-count} is meaningful only when your program
4502 stopped due to a breakpoint. At other times, the argument to
4503 @code{continue} is ignored.
4505 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4506 debugged program is deemed to be the foreground program) are provided
4507 purely for convenience, and have exactly the same behavior as
4511 To resume execution at a different place, you can use @code{return}
4512 (@pxref{Returning, ,Returning from a Function}) to go back to the
4513 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4514 Different Address}) to go to an arbitrary location in your program.
4516 A typical technique for using stepping is to set a breakpoint
4517 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4518 beginning of the function or the section of your program where a problem
4519 is believed to lie, run your program until it stops at that breakpoint,
4520 and then step through the suspect area, examining the variables that are
4521 interesting, until you see the problem happen.
4525 @kindex s @r{(@code{step})}
4527 Continue running your program until control reaches a different source
4528 line, then stop it and return control to @value{GDBN}. This command is
4529 abbreviated @code{s}.
4532 @c "without debugging information" is imprecise; actually "without line
4533 @c numbers in the debugging information". (gcc -g1 has debugging info but
4534 @c not line numbers). But it seems complex to try to make that
4535 @c distinction here.
4536 @emph{Warning:} If you use the @code{step} command while control is
4537 within a function that was compiled without debugging information,
4538 execution proceeds until control reaches a function that does have
4539 debugging information. Likewise, it will not step into a function which
4540 is compiled without debugging information. To step through functions
4541 without debugging information, use the @code{stepi} command, described
4545 The @code{step} command only stops at the first instruction of a source
4546 line. This prevents the multiple stops that could otherwise occur in
4547 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4548 to stop if a function that has debugging information is called within
4549 the line. In other words, @code{step} @emph{steps inside} any functions
4550 called within the line.
4552 Also, the @code{step} command only enters a function if there is line
4553 number information for the function. Otherwise it acts like the
4554 @code{next} command. This avoids problems when using @code{cc -gl}
4555 on MIPS machines. Previously, @code{step} entered subroutines if there
4556 was any debugging information about the routine.
4558 @item step @var{count}
4559 Continue running as in @code{step}, but do so @var{count} times. If a
4560 breakpoint is reached, or a signal not related to stepping occurs before
4561 @var{count} steps, stepping stops right away.
4564 @kindex n @r{(@code{next})}
4565 @item next @r{[}@var{count}@r{]}
4566 Continue to the next source line in the current (innermost) stack frame.
4567 This is similar to @code{step}, but function calls that appear within
4568 the line of code are executed without stopping. Execution stops when
4569 control reaches a different line of code at the original stack level
4570 that was executing when you gave the @code{next} command. This command
4571 is abbreviated @code{n}.
4573 An argument @var{count} is a repeat count, as for @code{step}.
4576 @c FIX ME!! Do we delete this, or is there a way it fits in with
4577 @c the following paragraph? --- Vctoria
4579 @c @code{next} within a function that lacks debugging information acts like
4580 @c @code{step}, but any function calls appearing within the code of the
4581 @c function are executed without stopping.
4583 The @code{next} command only stops at the first instruction of a
4584 source line. This prevents multiple stops that could otherwise occur in
4585 @code{switch} statements, @code{for} loops, etc.
4587 @kindex set step-mode
4589 @cindex functions without line info, and stepping
4590 @cindex stepping into functions with no line info
4591 @itemx set step-mode on
4592 The @code{set step-mode on} command causes the @code{step} command to
4593 stop at the first instruction of a function which contains no debug line
4594 information rather than stepping over it.
4596 This is useful in cases where you may be interested in inspecting the
4597 machine instructions of a function which has no symbolic info and do not
4598 want @value{GDBN} to automatically skip over this function.
4600 @item set step-mode off
4601 Causes the @code{step} command to step over any functions which contains no
4602 debug information. This is the default.
4604 @item show step-mode
4605 Show whether @value{GDBN} will stop in or step over functions without
4606 source line debug information.
4609 @kindex fin @r{(@code{finish})}
4611 Continue running until just after function in the selected stack frame
4612 returns. Print the returned value (if any). This command can be
4613 abbreviated as @code{fin}.
4615 Contrast this with the @code{return} command (@pxref{Returning,
4616 ,Returning from a Function}).
4619 @kindex u @r{(@code{until})}
4620 @cindex run until specified location
4623 Continue running until a source line past the current line, in the
4624 current stack frame, is reached. This command is used to avoid single
4625 stepping through a loop more than once. It is like the @code{next}
4626 command, except that when @code{until} encounters a jump, it
4627 automatically continues execution until the program counter is greater
4628 than the address of the jump.
4630 This means that when you reach the end of a loop after single stepping
4631 though it, @code{until} makes your program continue execution until it
4632 exits the loop. In contrast, a @code{next} command at the end of a loop
4633 simply steps back to the beginning of the loop, which forces you to step
4634 through the next iteration.
4636 @code{until} always stops your program if it attempts to exit the current
4639 @code{until} may produce somewhat counterintuitive results if the order
4640 of machine code does not match the order of the source lines. For
4641 example, in the following excerpt from a debugging session, the @code{f}
4642 (@code{frame}) command shows that execution is stopped at line
4643 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4647 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4649 (@value{GDBP}) until
4650 195 for ( ; argc > 0; NEXTARG) @{
4653 This happened because, for execution efficiency, the compiler had
4654 generated code for the loop closure test at the end, rather than the
4655 start, of the loop---even though the test in a C @code{for}-loop is
4656 written before the body of the loop. The @code{until} command appeared
4657 to step back to the beginning of the loop when it advanced to this
4658 expression; however, it has not really gone to an earlier
4659 statement---not in terms of the actual machine code.
4661 @code{until} with no argument works by means of single
4662 instruction stepping, and hence is slower than @code{until} with an
4665 @item until @var{location}
4666 @itemx u @var{location}
4667 Continue running your program until either the specified location is
4668 reached, or the current stack frame returns. @var{location} is any of
4669 the forms described in @ref{Specify Location}.
4670 This form of the command uses temporary breakpoints, and
4671 hence is quicker than @code{until} without an argument. The specified
4672 location is actually reached only if it is in the current frame. This
4673 implies that @code{until} can be used to skip over recursive function
4674 invocations. For instance in the code below, if the current location is
4675 line @code{96}, issuing @code{until 99} will execute the program up to
4676 line @code{99} in the same invocation of factorial, i.e., after the inner
4677 invocations have returned.
4680 94 int factorial (int value)
4682 96 if (value > 1) @{
4683 97 value *= factorial (value - 1);
4690 @kindex advance @var{location}
4691 @itemx advance @var{location}
4692 Continue running the program up to the given @var{location}. An argument is
4693 required, which should be of one of the forms described in
4694 @ref{Specify Location}.
4695 Execution will also stop upon exit from the current stack
4696 frame. This command is similar to @code{until}, but @code{advance} will
4697 not skip over recursive function calls, and the target location doesn't
4698 have to be in the same frame as the current one.
4702 @kindex si @r{(@code{stepi})}
4704 @itemx stepi @var{arg}
4706 Execute one machine instruction, then stop and return to the debugger.
4708 It is often useful to do @samp{display/i $pc} when stepping by machine
4709 instructions. This makes @value{GDBN} automatically display the next
4710 instruction to be executed, each time your program stops. @xref{Auto
4711 Display,, Automatic Display}.
4713 An argument is a repeat count, as in @code{step}.
4717 @kindex ni @r{(@code{nexti})}
4719 @itemx nexti @var{arg}
4721 Execute one machine instruction, but if it is a function call,
4722 proceed until the function returns.
4724 An argument is a repeat count, as in @code{next}.
4731 A signal is an asynchronous event that can happen in a program. The
4732 operating system defines the possible kinds of signals, and gives each
4733 kind a name and a number. For example, in Unix @code{SIGINT} is the
4734 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4735 @code{SIGSEGV} is the signal a program gets from referencing a place in
4736 memory far away from all the areas in use; @code{SIGALRM} occurs when
4737 the alarm clock timer goes off (which happens only if your program has
4738 requested an alarm).
4740 @cindex fatal signals
4741 Some signals, including @code{SIGALRM}, are a normal part of the
4742 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4743 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4744 program has not specified in advance some other way to handle the signal.
4745 @code{SIGINT} does not indicate an error in your program, but it is normally
4746 fatal so it can carry out the purpose of the interrupt: to kill the program.
4748 @value{GDBN} has the ability to detect any occurrence of a signal in your
4749 program. You can tell @value{GDBN} in advance what to do for each kind of
4752 @cindex handling signals
4753 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4754 @code{SIGALRM} be silently passed to your program
4755 (so as not to interfere with their role in the program's functioning)
4756 but to stop your program immediately whenever an error signal happens.
4757 You can change these settings with the @code{handle} command.
4760 @kindex info signals
4764 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4765 handle each one. You can use this to see the signal numbers of all
4766 the defined types of signals.
4768 @item info signals @var{sig}
4769 Similar, but print information only about the specified signal number.
4771 @code{info handle} is an alias for @code{info signals}.
4774 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4775 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4776 can be the number of a signal or its name (with or without the
4777 @samp{SIG} at the beginning); a list of signal numbers of the form
4778 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4779 known signals. Optional arguments @var{keywords}, described below,
4780 say what change to make.
4784 The keywords allowed by the @code{handle} command can be abbreviated.
4785 Their full names are:
4789 @value{GDBN} should not stop your program when this signal happens. It may
4790 still print a message telling you that the signal has come in.
4793 @value{GDBN} should stop your program when this signal happens. This implies
4794 the @code{print} keyword as well.
4797 @value{GDBN} should print a message when this signal happens.
4800 @value{GDBN} should not mention the occurrence of the signal at all. This
4801 implies the @code{nostop} keyword as well.
4805 @value{GDBN} should allow your program to see this signal; your program
4806 can handle the signal, or else it may terminate if the signal is fatal
4807 and not handled. @code{pass} and @code{noignore} are synonyms.
4811 @value{GDBN} should not allow your program to see this signal.
4812 @code{nopass} and @code{ignore} are synonyms.
4816 When a signal stops your program, the signal is not visible to the
4818 continue. Your program sees the signal then, if @code{pass} is in
4819 effect for the signal in question @emph{at that time}. In other words,
4820 after @value{GDBN} reports a signal, you can use the @code{handle}
4821 command with @code{pass} or @code{nopass} to control whether your
4822 program sees that signal when you continue.
4824 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4825 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4826 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4829 You can also use the @code{signal} command to prevent your program from
4830 seeing a signal, or cause it to see a signal it normally would not see,
4831 or to give it any signal at any time. For example, if your program stopped
4832 due to some sort of memory reference error, you might store correct
4833 values into the erroneous variables and continue, hoping to see more
4834 execution; but your program would probably terminate immediately as
4835 a result of the fatal signal once it saw the signal. To prevent this,
4836 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4839 @cindex extra signal information
4840 @anchor{extra signal information}
4842 On some targets, @value{GDBN} can inspect extra signal information
4843 associated with the intercepted signal, before it is actually
4844 delivered to the program being debugged. This information is exported
4845 by the convenience variable @code{$_siginfo}, and consists of data
4846 that is passed by the kernel to the signal handler at the time of the
4847 receipt of a signal. The data type of the information itself is
4848 target dependent. You can see the data type using the @code{ptype
4849 $_siginfo} command. On Unix systems, it typically corresponds to the
4850 standard @code{siginfo_t} type, as defined in the @file{signal.h}
4853 Here's an example, on a @sc{gnu}/Linux system, printing the stray
4854 referenced address that raised a segmentation fault.
4858 (@value{GDBP}) continue
4859 Program received signal SIGSEGV, Segmentation fault.
4860 0x0000000000400766 in main ()
4862 (@value{GDBP}) ptype $_siginfo
4869 struct @{...@} _kill;
4870 struct @{...@} _timer;
4872 struct @{...@} _sigchld;
4873 struct @{...@} _sigfault;
4874 struct @{...@} _sigpoll;
4877 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
4881 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
4882 $1 = (void *) 0x7ffff7ff7000
4886 Depending on target support, @code{$_siginfo} may also be writable.
4889 @section Stopping and Starting Multi-thread Programs
4891 @cindex stopped threads
4892 @cindex threads, stopped
4894 @cindex continuing threads
4895 @cindex threads, continuing
4897 @value{GDBN} supports debugging programs with multiple threads
4898 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
4899 are two modes of controlling execution of your program within the
4900 debugger. In the default mode, referred to as @dfn{all-stop mode},
4901 when any thread in your program stops (for example, at a breakpoint
4902 or while being stepped), all other threads in the program are also stopped by
4903 @value{GDBN}. On some targets, @value{GDBN} also supports
4904 @dfn{non-stop mode}, in which other threads can continue to run freely while
4905 you examine the stopped thread in the debugger.
4908 * All-Stop Mode:: All threads stop when GDB takes control
4909 * Non-Stop Mode:: Other threads continue to execute
4910 * Background Execution:: Running your program asynchronously
4911 * Thread-Specific Breakpoints:: Controlling breakpoints
4912 * Interrupted System Calls:: GDB may interfere with system calls
4916 @subsection All-Stop Mode
4918 @cindex all-stop mode
4920 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
4921 @emph{all} threads of execution stop, not just the current thread. This
4922 allows you to examine the overall state of the program, including
4923 switching between threads, without worrying that things may change
4926 Conversely, whenever you restart the program, @emph{all} threads start
4927 executing. @emph{This is true even when single-stepping} with commands
4928 like @code{step} or @code{next}.
4930 In particular, @value{GDBN} cannot single-step all threads in lockstep.
4931 Since thread scheduling is up to your debugging target's operating
4932 system (not controlled by @value{GDBN}), other threads may
4933 execute more than one statement while the current thread completes a
4934 single step. Moreover, in general other threads stop in the middle of a
4935 statement, rather than at a clean statement boundary, when the program
4938 You might even find your program stopped in another thread after
4939 continuing or even single-stepping. This happens whenever some other
4940 thread runs into a breakpoint, a signal, or an exception before the
4941 first thread completes whatever you requested.
4943 @cindex automatic thread selection
4944 @cindex switching threads automatically
4945 @cindex threads, automatic switching
4946 Whenever @value{GDBN} stops your program, due to a breakpoint or a
4947 signal, it automatically selects the thread where that breakpoint or
4948 signal happened. @value{GDBN} alerts you to the context switch with a
4949 message such as @samp{[Switching to Thread @var{n}]} to identify the
4952 On some OSes, you can modify @value{GDBN}'s default behavior by
4953 locking the OS scheduler to allow only a single thread to run.
4956 @item set scheduler-locking @var{mode}
4957 @cindex scheduler locking mode
4958 @cindex lock scheduler
4959 Set the scheduler locking mode. If it is @code{off}, then there is no
4960 locking and any thread may run at any time. If @code{on}, then only the
4961 current thread may run when the inferior is resumed. The @code{step}
4962 mode optimizes for single-stepping; it prevents other threads
4963 from preempting the current thread while you are stepping, so that
4964 the focus of debugging does not change unexpectedly.
4965 Other threads only rarely (or never) get a chance to run
4966 when you step. They are more likely to run when you @samp{next} over a
4967 function call, and they are completely free to run when you use commands
4968 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
4969 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
4970 the current thread away from the thread that you are debugging.
4972 @item show scheduler-locking
4973 Display the current scheduler locking mode.
4976 @cindex resume threads of multiple processes simultaneously
4977 By default, when you issue one of the execution commands such as
4978 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
4979 threads of the current inferior to run. For example, if @value{GDBN}
4980 is attached to two inferiors, each with two threads, the
4981 @code{continue} command resumes only the two threads of the current
4982 inferior. This is useful, for example, when you debug a program that
4983 forks and you want to hold the parent stopped (so that, for instance,
4984 it doesn't run to exit), while you debug the child. In other
4985 situations, you may not be interested in inspecting the current state
4986 of any of the processes @value{GDBN} is attached to, and you may want
4987 to resume them all until some breakpoint is hit. In the latter case,
4988 you can instruct @value{GDBN} to allow all threads of all the
4989 inferiors to run with the @w{@code{set schedule-multiple}} command.
4992 @kindex set schedule-multiple
4993 @item set schedule-multiple
4994 Set the mode for allowing threads of multiple processes to be resumed
4995 when an execution command is issued. When @code{on}, all threads of
4996 all processes are allowed to run. When @code{off}, only the threads
4997 of the current process are resumed. The default is @code{off}. The
4998 @code{scheduler-locking} mode takes precedence when set to @code{on},
4999 or while you are stepping and set to @code{step}.
5001 @item show schedule-multiple
5002 Display the current mode for resuming the execution of threads of
5007 @subsection Non-Stop Mode
5009 @cindex non-stop mode
5011 @c This section is really only a place-holder, and needs to be expanded
5012 @c with more details.
5014 For some multi-threaded targets, @value{GDBN} supports an optional
5015 mode of operation in which you can examine stopped program threads in
5016 the debugger while other threads continue to execute freely. This
5017 minimizes intrusion when debugging live systems, such as programs
5018 where some threads have real-time constraints or must continue to
5019 respond to external events. This is referred to as @dfn{non-stop} mode.
5021 In non-stop mode, when a thread stops to report a debugging event,
5022 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5023 threads as well, in contrast to the all-stop mode behavior. Additionally,
5024 execution commands such as @code{continue} and @code{step} apply by default
5025 only to the current thread in non-stop mode, rather than all threads as
5026 in all-stop mode. This allows you to control threads explicitly in
5027 ways that are not possible in all-stop mode --- for example, stepping
5028 one thread while allowing others to run freely, stepping
5029 one thread while holding all others stopped, or stepping several threads
5030 independently and simultaneously.
5032 To enter non-stop mode, use this sequence of commands before you run
5033 or attach to your program:
5036 # Enable the async interface.
5039 # If using the CLI, pagination breaks non-stop.
5042 # Finally, turn it on!
5046 You can use these commands to manipulate the non-stop mode setting:
5049 @kindex set non-stop
5050 @item set non-stop on
5051 Enable selection of non-stop mode.
5052 @item set non-stop off
5053 Disable selection of non-stop mode.
5054 @kindex show non-stop
5056 Show the current non-stop enablement setting.
5059 Note these commands only reflect whether non-stop mode is enabled,
5060 not whether the currently-executing program is being run in non-stop mode.
5061 In particular, the @code{set non-stop} preference is only consulted when
5062 @value{GDBN} starts or connects to the target program, and it is generally
5063 not possible to switch modes once debugging has started. Furthermore,
5064 since not all targets support non-stop mode, even when you have enabled
5065 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5068 In non-stop mode, all execution commands apply only to the current thread
5069 by default. That is, @code{continue} only continues one thread.
5070 To continue all threads, issue @code{continue -a} or @code{c -a}.
5072 You can use @value{GDBN}'s background execution commands
5073 (@pxref{Background Execution}) to run some threads in the background
5074 while you continue to examine or step others from @value{GDBN}.
5075 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5076 always executed asynchronously in non-stop mode.
5078 Suspending execution is done with the @code{interrupt} command when
5079 running in the background, or @kbd{Ctrl-c} during foreground execution.
5080 In all-stop mode, this stops the whole process;
5081 but in non-stop mode the interrupt applies only to the current thread.
5082 To stop the whole program, use @code{interrupt -a}.
5084 Other execution commands do not currently support the @code{-a} option.
5086 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5087 that thread current, as it does in all-stop mode. This is because the
5088 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5089 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5090 changed to a different thread just as you entered a command to operate on the
5091 previously current thread.
5093 @node Background Execution
5094 @subsection Background Execution
5096 @cindex foreground execution
5097 @cindex background execution
5098 @cindex asynchronous execution
5099 @cindex execution, foreground, background and asynchronous
5101 @value{GDBN}'s execution commands have two variants: the normal
5102 foreground (synchronous) behavior, and a background
5103 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5104 the program to report that some thread has stopped before prompting for
5105 another command. In background execution, @value{GDBN} immediately gives
5106 a command prompt so that you can issue other commands while your program runs.
5108 You need to explicitly enable asynchronous mode before you can use
5109 background execution commands. You can use these commands to
5110 manipulate the asynchronous mode setting:
5113 @kindex set target-async
5114 @item set target-async on
5115 Enable asynchronous mode.
5116 @item set target-async off
5117 Disable asynchronous mode.
5118 @kindex show target-async
5119 @item show target-async
5120 Show the current target-async setting.
5123 If the target doesn't support async mode, @value{GDBN} issues an error
5124 message if you attempt to use the background execution commands.
5126 To specify background execution, add a @code{&} to the command. For example,
5127 the background form of the @code{continue} command is @code{continue&}, or
5128 just @code{c&}. The execution commands that accept background execution
5134 @xref{Starting, , Starting your Program}.
5138 @xref{Attach, , Debugging an Already-running Process}.
5142 @xref{Continuing and Stepping, step}.
5146 @xref{Continuing and Stepping, stepi}.
5150 @xref{Continuing and Stepping, next}.
5154 @xref{Continuing and Stepping, nexti}.
5158 @xref{Continuing and Stepping, continue}.
5162 @xref{Continuing and Stepping, finish}.
5166 @xref{Continuing and Stepping, until}.
5170 Background execution is especially useful in conjunction with non-stop
5171 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5172 However, you can also use these commands in the normal all-stop mode with
5173 the restriction that you cannot issue another execution command until the
5174 previous one finishes. Examples of commands that are valid in all-stop
5175 mode while the program is running include @code{help} and @code{info break}.
5177 You can interrupt your program while it is running in the background by
5178 using the @code{interrupt} command.
5185 Suspend execution of the running program. In all-stop mode,
5186 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5187 only the current thread. To stop the whole program in non-stop mode,
5188 use @code{interrupt -a}.
5191 @node Thread-Specific Breakpoints
5192 @subsection Thread-Specific Breakpoints
5194 When your program has multiple threads (@pxref{Threads,, Debugging
5195 Programs with Multiple Threads}), you can choose whether to set
5196 breakpoints on all threads, or on a particular thread.
5199 @cindex breakpoints and threads
5200 @cindex thread breakpoints
5201 @kindex break @dots{} thread @var{threadno}
5202 @item break @var{linespec} thread @var{threadno}
5203 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5204 @var{linespec} specifies source lines; there are several ways of
5205 writing them (@pxref{Specify Location}), but the effect is always to
5206 specify some source line.
5208 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5209 to specify that you only want @value{GDBN} to stop the program when a
5210 particular thread reaches this breakpoint. @var{threadno} is one of the
5211 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5212 column of the @samp{info threads} display.
5214 If you do not specify @samp{thread @var{threadno}} when you set a
5215 breakpoint, the breakpoint applies to @emph{all} threads of your
5218 You can use the @code{thread} qualifier on conditional breakpoints as
5219 well; in this case, place @samp{thread @var{threadno}} before or
5220 after the breakpoint condition, like this:
5223 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5228 @node Interrupted System Calls
5229 @subsection Interrupted System Calls
5231 @cindex thread breakpoints and system calls
5232 @cindex system calls and thread breakpoints
5233 @cindex premature return from system calls
5234 There is an unfortunate side effect when using @value{GDBN} to debug
5235 multi-threaded programs. If one thread stops for a
5236 breakpoint, or for some other reason, and another thread is blocked in a
5237 system call, then the system call may return prematurely. This is a
5238 consequence of the interaction between multiple threads and the signals
5239 that @value{GDBN} uses to implement breakpoints and other events that
5242 To handle this problem, your program should check the return value of
5243 each system call and react appropriately. This is good programming
5246 For example, do not write code like this:
5252 The call to @code{sleep} will return early if a different thread stops
5253 at a breakpoint or for some other reason.
5255 Instead, write this:
5260 unslept = sleep (unslept);
5263 A system call is allowed to return early, so the system is still
5264 conforming to its specification. But @value{GDBN} does cause your
5265 multi-threaded program to behave differently than it would without
5268 Also, @value{GDBN} uses internal breakpoints in the thread library to
5269 monitor certain events such as thread creation and thread destruction.
5270 When such an event happens, a system call in another thread may return
5271 prematurely, even though your program does not appear to stop.
5274 @node Reverse Execution
5275 @chapter Running programs backward
5276 @cindex reverse execution
5277 @cindex running programs backward
5279 When you are debugging a program, it is not unusual to realize that
5280 you have gone too far, and some event of interest has already happened.
5281 If the target environment supports it, @value{GDBN} can allow you to
5282 ``rewind'' the program by running it backward.
5284 A target environment that supports reverse execution should be able
5285 to ``undo'' the changes in machine state that have taken place as the
5286 program was executing normally. Variables, registers etc.@: should
5287 revert to their previous values. Obviously this requires a great
5288 deal of sophistication on the part of the target environment; not
5289 all target environments can support reverse execution.
5291 When a program is executed in reverse, the instructions that
5292 have most recently been executed are ``un-executed'', in reverse
5293 order. The program counter runs backward, following the previous
5294 thread of execution in reverse. As each instruction is ``un-executed'',
5295 the values of memory and/or registers that were changed by that
5296 instruction are reverted to their previous states. After executing
5297 a piece of source code in reverse, all side effects of that code
5298 should be ``undone'', and all variables should be returned to their
5299 prior values@footnote{
5300 Note that some side effects are easier to undo than others. For instance,
5301 memory and registers are relatively easy, but device I/O is hard. Some
5302 targets may be able undo things like device I/O, and some may not.
5304 The contract between @value{GDBN} and the reverse executing target
5305 requires only that the target do something reasonable when
5306 @value{GDBN} tells it to execute backwards, and then report the
5307 results back to @value{GDBN}. Whatever the target reports back to
5308 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
5309 assumes that the memory and registers that the target reports are in a
5310 consistant state, but @value{GDBN} accepts whatever it is given.
5313 If you are debugging in a target environment that supports
5314 reverse execution, @value{GDBN} provides the following commands.
5317 @kindex reverse-continue
5318 @kindex rc @r{(@code{reverse-continue})}
5319 @item reverse-continue @r{[}@var{ignore-count}@r{]}
5320 @itemx rc @r{[}@var{ignore-count}@r{]}
5321 Beginning at the point where your program last stopped, start executing
5322 in reverse. Reverse execution will stop for breakpoints and synchronous
5323 exceptions (signals), just like normal execution. Behavior of
5324 asynchronous signals depends on the target environment.
5326 @kindex reverse-step
5327 @kindex rs @r{(@code{step})}
5328 @item reverse-step @r{[}@var{count}@r{]}
5329 Run the program backward until control reaches the start of a
5330 different source line; then stop it, and return control to @value{GDBN}.
5332 Like the @code{step} command, @code{reverse-step} will only stop
5333 at the beginning of a source line. It ``un-executes'' the previously
5334 executed source line. If the previous source line included calls to
5335 debuggable functions, @code{reverse-step} will step (backward) into
5336 the called function, stopping at the beginning of the @emph{last}
5337 statement in the called function (typically a return statement).
5339 Also, as with the @code{step} command, if non-debuggable functions are
5340 called, @code{reverse-step} will run thru them backward without stopping.
5342 @kindex reverse-stepi
5343 @kindex rsi @r{(@code{reverse-stepi})}
5344 @item reverse-stepi @r{[}@var{count}@r{]}
5345 Reverse-execute one machine instruction. Note that the instruction
5346 to be reverse-executed is @emph{not} the one pointed to by the program
5347 counter, but the instruction executed prior to that one. For instance,
5348 if the last instruction was a jump, @code{reverse-stepi} will take you
5349 back from the destination of the jump to the jump instruction itself.
5351 @kindex reverse-next
5352 @kindex rn @r{(@code{reverse-next})}
5353 @item reverse-next @r{[}@var{count}@r{]}
5354 Run backward to the beginning of the previous line executed in
5355 the current (innermost) stack frame. If the line contains function
5356 calls, they will be ``un-executed'' without stopping. Starting from
5357 the first line of a function, @code{reverse-next} will take you back
5358 to the caller of that function, @emph{before} the function was called,
5359 just as the normal @code{next} command would take you from the last
5360 line of a function back to its return to its caller
5361 @footnote{Unless the code is too heavily optimized.}.
5363 @kindex reverse-nexti
5364 @kindex rni @r{(@code{reverse-nexti})}
5365 @item reverse-nexti @r{[}@var{count}@r{]}
5366 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
5367 in reverse, except that called functions are ``un-executed'' atomically.
5368 That is, if the previously executed instruction was a return from
5369 another instruction, @code{reverse-nexti} will continue to execute
5370 in reverse until the call to that function (from the current stack
5373 @kindex reverse-finish
5374 @item reverse-finish
5375 Just as the @code{finish} command takes you to the point where the
5376 current function returns, @code{reverse-finish} takes you to the point
5377 where it was called. Instead of ending up at the end of the current
5378 function invocation, you end up at the beginning.
5380 @kindex set exec-direction
5381 @item set exec-direction
5382 Set the direction of target execution.
5383 @itemx set exec-direction reverse
5384 @cindex execute forward or backward in time
5385 @value{GDBN} will perform all execution commands in reverse, until the
5386 exec-direction mode is changed to ``forward''. Affected commands include
5387 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
5388 command cannot be used in reverse mode.
5389 @item set exec-direction forward
5390 @value{GDBN} will perform all execution commands in the normal fashion.
5391 This is the default.
5395 @node Process Record and Replay
5396 @chapter Recording Inferior's Execution and Replaying It
5397 @cindex process record and replay
5398 @cindex recording inferior's execution and replaying it
5400 On some platforms, @value{GDBN} provides a special @dfn{process record
5401 and replay} target that can record a log of the process execution, and
5402 replay it later with both forward and reverse execution commands.
5405 When this target is in use, if the execution log includes the record
5406 for the next instruction, @value{GDBN} will debug in @dfn{replay
5407 mode}. In the replay mode, the inferior does not really execute code
5408 instructions. Instead, all the events that normally happen during
5409 code execution are taken from the execution log. While code is not
5410 really executed in replay mode, the values of registers (including the
5411 program counter register) and the memory of the inferior are still
5412 changed as they normally would. Their contents are taken from the
5416 If the record for the next instruction is not in the execution log,
5417 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
5418 inferior executes normally, and @value{GDBN} records the execution log
5421 The process record and replay target supports reverse execution
5422 (@pxref{Reverse Execution}), even if the platform on which the
5423 inferior runs does not. However, the reverse execution is limited in
5424 this case by the range of the instructions recorded in the execution
5425 log. In other words, reverse execution on platforms that don't
5426 support it directly can only be done in the replay mode.
5428 When debugging in the reverse direction, @value{GDBN} will work in
5429 replay mode as long as the execution log includes the record for the
5430 previous instruction; otherwise, it will work in record mode, if the
5431 platform supports reverse execution, or stop if not.
5433 For architecture environments that support process record and replay,
5434 @value{GDBN} provides the following commands:
5437 @kindex target record
5441 This command starts the process record and replay target. The process
5442 record and replay target can only debug a process that is already
5443 running. Therefore, you need first to start the process with the
5444 @kbd{run} or @kbd{start} commands, and then start the recording with
5445 the @kbd{target record} command.
5447 Both @code{record} and @code{rec} are aliases of @code{target record}.
5449 @cindex displaced stepping, and process record and replay
5450 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
5451 will be automatically disabled when process record and replay target
5452 is started. That's because the process record and replay target
5453 doesn't support displaced stepping.
5455 @cindex non-stop mode, and process record and replay
5456 @cindex asynchronous execution, and process record and replay
5457 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
5458 the asynchronous execution mode (@pxref{Background Execution}), the
5459 process record and replay target cannot be started because it doesn't
5460 support these two modes.
5465 Stop the process record and replay target. When process record and
5466 replay target stops, the entire execution log will be deleted and the
5467 inferior will either be terminated, or will remain in its final state.
5469 When you stop the process record and replay target in record mode (at
5470 the end of the execution log), the inferior will be stopped at the
5471 next instruction that would have been recorded. In other words, if
5472 you record for a while and then stop recording, the inferior process
5473 will be left in the same state as if the recording never happened.
5475 On the other hand, if the process record and replay target is stopped
5476 while in replay mode (that is, not at the end of the execution log,
5477 but at some earlier point), the inferior process will become ``live''
5478 at that earlier state, and it will then be possible to continue the
5479 usual ``live'' debugging of the process from that state.
5481 When the inferior process exits, or @value{GDBN} detaches from it,
5482 process record and replay target will automatically stop itself.
5484 @kindex set record insn-number-max
5485 @item set record insn-number-max @var{limit}
5486 Set the limit of instructions to be recorded. Default value is 200000.
5488 If @var{limit} is a positive number, then @value{GDBN} will start
5489 deleting instructions from the log once the number of the record
5490 instructions becomes greater than @var{limit}. For every new recorded
5491 instruction, @value{GDBN} will delete the earliest recorded
5492 instruction to keep the number of recorded instructions at the limit.
5493 (Since deleting recorded instructions loses information, @value{GDBN}
5494 lets you control what happens when the limit is reached, by means of
5495 the @code{stop-at-limit} option, described below.)
5497 If @var{limit} is zero, @value{GDBN} will never delete recorded
5498 instructions from the execution log. The number of recorded
5499 instructions is unlimited in this case.
5501 @kindex show record insn-number-max
5502 @item show record insn-number-max
5503 Show the limit of instructions to be recorded.
5505 @kindex set record stop-at-limit
5506 @item set record stop-at-limit
5507 Control the behavior when the number of recorded instructions reaches
5508 the limit. If ON (the default), @value{GDBN} will stop when the limit
5509 is reached for the first time and ask you whether you want to stop the
5510 inferior or continue running it and recording the execution log. If
5511 you decide to continue recording, each new recorded instruction will
5512 cause the oldest one to be deleted.
5514 If this option is OFF, @value{GDBN} will automatically delete the
5515 oldest record to make room for each new one, without asking.
5517 @kindex show record stop-at-limit
5518 @item show record stop-at-limit
5519 Show the current setting of @code{stop-at-limit}.
5523 Show various statistics about the state of process record and its
5524 in-memory execution log buffer, including:
5528 Whether in record mode or replay mode.
5530 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
5532 Highest recorded instruction number.
5534 Current instruction about to be replayed (if in replay mode).
5536 Number of instructions contained in the execution log.
5538 Maximum number of instructions that may be contained in the execution log.
5541 @kindex record delete
5544 When record target runs in replay mode (``in the past''), delete the
5545 subsequent execution log and begin to record a new execution log starting
5546 from the current address. This means you will abandon the previously
5547 recorded ``future'' and begin recording a new ``future''.
5552 @chapter Examining the Stack
5554 When your program has stopped, the first thing you need to know is where it
5555 stopped and how it got there.
5558 Each time your program performs a function call, information about the call
5560 That information includes the location of the call in your program,
5561 the arguments of the call,
5562 and the local variables of the function being called.
5563 The information is saved in a block of data called a @dfn{stack frame}.
5564 The stack frames are allocated in a region of memory called the @dfn{call
5567 When your program stops, the @value{GDBN} commands for examining the
5568 stack allow you to see all of this information.
5570 @cindex selected frame
5571 One of the stack frames is @dfn{selected} by @value{GDBN} and many
5572 @value{GDBN} commands refer implicitly to the selected frame. In
5573 particular, whenever you ask @value{GDBN} for the value of a variable in
5574 your program, the value is found in the selected frame. There are
5575 special @value{GDBN} commands to select whichever frame you are
5576 interested in. @xref{Selection, ,Selecting a Frame}.
5578 When your program stops, @value{GDBN} automatically selects the
5579 currently executing frame and describes it briefly, similar to the
5580 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
5583 * Frames:: Stack frames
5584 * Backtrace:: Backtraces
5585 * Selection:: Selecting a frame
5586 * Frame Info:: Information on a frame
5591 @section Stack Frames
5593 @cindex frame, definition
5595 The call stack is divided up into contiguous pieces called @dfn{stack
5596 frames}, or @dfn{frames} for short; each frame is the data associated
5597 with one call to one function. The frame contains the arguments given
5598 to the function, the function's local variables, and the address at
5599 which the function is executing.
5601 @cindex initial frame
5602 @cindex outermost frame
5603 @cindex innermost frame
5604 When your program is started, the stack has only one frame, that of the
5605 function @code{main}. This is called the @dfn{initial} frame or the
5606 @dfn{outermost} frame. Each time a function is called, a new frame is
5607 made. Each time a function returns, the frame for that function invocation
5608 is eliminated. If a function is recursive, there can be many frames for
5609 the same function. The frame for the function in which execution is
5610 actually occurring is called the @dfn{innermost} frame. This is the most
5611 recently created of all the stack frames that still exist.
5613 @cindex frame pointer
5614 Inside your program, stack frames are identified by their addresses. A
5615 stack frame consists of many bytes, each of which has its own address; each
5616 kind of computer has a convention for choosing one byte whose
5617 address serves as the address of the frame. Usually this address is kept
5618 in a register called the @dfn{frame pointer register}
5619 (@pxref{Registers, $fp}) while execution is going on in that frame.
5621 @cindex frame number
5622 @value{GDBN} assigns numbers to all existing stack frames, starting with
5623 zero for the innermost frame, one for the frame that called it,
5624 and so on upward. These numbers do not really exist in your program;
5625 they are assigned by @value{GDBN} to give you a way of designating stack
5626 frames in @value{GDBN} commands.
5628 @c The -fomit-frame-pointer below perennially causes hbox overflow
5629 @c underflow problems.
5630 @cindex frameless execution
5631 Some compilers provide a way to compile functions so that they operate
5632 without stack frames. (For example, the @value{NGCC} option
5634 @samp{-fomit-frame-pointer}
5636 generates functions without a frame.)
5637 This is occasionally done with heavily used library functions to save
5638 the frame setup time. @value{GDBN} has limited facilities for dealing
5639 with these function invocations. If the innermost function invocation
5640 has no stack frame, @value{GDBN} nevertheless regards it as though
5641 it had a separate frame, which is numbered zero as usual, allowing
5642 correct tracing of the function call chain. However, @value{GDBN} has
5643 no provision for frameless functions elsewhere in the stack.
5646 @kindex frame@r{, command}
5647 @cindex current stack frame
5648 @item frame @var{args}
5649 The @code{frame} command allows you to move from one stack frame to another,
5650 and to print the stack frame you select. @var{args} may be either the
5651 address of the frame or the stack frame number. Without an argument,
5652 @code{frame} prints the current stack frame.
5654 @kindex select-frame
5655 @cindex selecting frame silently
5657 The @code{select-frame} command allows you to move from one stack frame
5658 to another without printing the frame. This is the silent version of
5666 @cindex call stack traces
5667 A backtrace is a summary of how your program got where it is. It shows one
5668 line per frame, for many frames, starting with the currently executing
5669 frame (frame zero), followed by its caller (frame one), and on up the
5674 @kindex bt @r{(@code{backtrace})}
5677 Print a backtrace of the entire stack: one line per frame for all
5678 frames in the stack.
5680 You can stop the backtrace at any time by typing the system interrupt
5681 character, normally @kbd{Ctrl-c}.
5683 @item backtrace @var{n}
5685 Similar, but print only the innermost @var{n} frames.
5687 @item backtrace -@var{n}
5689 Similar, but print only the outermost @var{n} frames.
5691 @item backtrace full
5693 @itemx bt full @var{n}
5694 @itemx bt full -@var{n}
5695 Print the values of the local variables also. @var{n} specifies the
5696 number of frames to print, as described above.
5701 The names @code{where} and @code{info stack} (abbreviated @code{info s})
5702 are additional aliases for @code{backtrace}.
5704 @cindex multiple threads, backtrace
5705 In a multi-threaded program, @value{GDBN} by default shows the
5706 backtrace only for the current thread. To display the backtrace for
5707 several or all of the threads, use the command @code{thread apply}
5708 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
5709 apply all backtrace}, @value{GDBN} will display the backtrace for all
5710 the threads; this is handy when you debug a core dump of a
5711 multi-threaded program.
5713 Each line in the backtrace shows the frame number and the function name.
5714 The program counter value is also shown---unless you use @code{set
5715 print address off}. The backtrace also shows the source file name and
5716 line number, as well as the arguments to the function. The program
5717 counter value is omitted if it is at the beginning of the code for that
5720 Here is an example of a backtrace. It was made with the command
5721 @samp{bt 3}, so it shows the innermost three frames.
5725 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5727 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
5728 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
5730 (More stack frames follow...)
5735 The display for frame zero does not begin with a program counter
5736 value, indicating that your program has stopped at the beginning of the
5737 code for line @code{993} of @code{builtin.c}.
5740 The value of parameter @code{data} in frame 1 has been replaced by
5741 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
5742 only if it is a scalar (integer, pointer, enumeration, etc). See command
5743 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
5744 on how to configure the way function parameter values are printed.
5746 @cindex value optimized out, in backtrace
5747 @cindex function call arguments, optimized out
5748 If your program was compiled with optimizations, some compilers will
5749 optimize away arguments passed to functions if those arguments are
5750 never used after the call. Such optimizations generate code that
5751 passes arguments through registers, but doesn't store those arguments
5752 in the stack frame. @value{GDBN} has no way of displaying such
5753 arguments in stack frames other than the innermost one. Here's what
5754 such a backtrace might look like:
5758 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5760 #1 0x6e38 in expand_macro (sym=<value optimized out>) at macro.c:242
5761 #2 0x6840 in expand_token (obs=0x0, t=<value optimized out>, td=0xf7fffb08)
5763 (More stack frames follow...)
5768 The values of arguments that were not saved in their stack frames are
5769 shown as @samp{<value optimized out>}.
5771 If you need to display the values of such optimized-out arguments,
5772 either deduce that from other variables whose values depend on the one
5773 you are interested in, or recompile without optimizations.
5775 @cindex backtrace beyond @code{main} function
5776 @cindex program entry point
5777 @cindex startup code, and backtrace
5778 Most programs have a standard user entry point---a place where system
5779 libraries and startup code transition into user code. For C this is
5780 @code{main}@footnote{
5781 Note that embedded programs (the so-called ``free-standing''
5782 environment) are not required to have a @code{main} function as the
5783 entry point. They could even have multiple entry points.}.
5784 When @value{GDBN} finds the entry function in a backtrace
5785 it will terminate the backtrace, to avoid tracing into highly
5786 system-specific (and generally uninteresting) code.
5788 If you need to examine the startup code, or limit the number of levels
5789 in a backtrace, you can change this behavior:
5792 @item set backtrace past-main
5793 @itemx set backtrace past-main on
5794 @kindex set backtrace
5795 Backtraces will continue past the user entry point.
5797 @item set backtrace past-main off
5798 Backtraces will stop when they encounter the user entry point. This is the
5801 @item show backtrace past-main
5802 @kindex show backtrace
5803 Display the current user entry point backtrace policy.
5805 @item set backtrace past-entry
5806 @itemx set backtrace past-entry on
5807 Backtraces will continue past the internal entry point of an application.
5808 This entry point is encoded by the linker when the application is built,
5809 and is likely before the user entry point @code{main} (or equivalent) is called.
5811 @item set backtrace past-entry off
5812 Backtraces will stop when they encounter the internal entry point of an
5813 application. This is the default.
5815 @item show backtrace past-entry
5816 Display the current internal entry point backtrace policy.
5818 @item set backtrace limit @var{n}
5819 @itemx set backtrace limit 0
5820 @cindex backtrace limit
5821 Limit the backtrace to @var{n} levels. A value of zero means
5824 @item show backtrace limit
5825 Display the current limit on backtrace levels.
5829 @section Selecting a Frame
5831 Most commands for examining the stack and other data in your program work on
5832 whichever stack frame is selected at the moment. Here are the commands for
5833 selecting a stack frame; all of them finish by printing a brief description
5834 of the stack frame just selected.
5837 @kindex frame@r{, selecting}
5838 @kindex f @r{(@code{frame})}
5841 Select frame number @var{n}. Recall that frame zero is the innermost
5842 (currently executing) frame, frame one is the frame that called the
5843 innermost one, and so on. The highest-numbered frame is the one for
5846 @item frame @var{addr}
5848 Select the frame at address @var{addr}. This is useful mainly if the
5849 chaining of stack frames has been damaged by a bug, making it
5850 impossible for @value{GDBN} to assign numbers properly to all frames. In
5851 addition, this can be useful when your program has multiple stacks and
5852 switches between them.
5854 On the SPARC architecture, @code{frame} needs two addresses to
5855 select an arbitrary frame: a frame pointer and a stack pointer.
5857 On the MIPS and Alpha architecture, it needs two addresses: a stack
5858 pointer and a program counter.
5860 On the 29k architecture, it needs three addresses: a register stack
5861 pointer, a program counter, and a memory stack pointer.
5865 Move @var{n} frames up the stack. For positive numbers @var{n}, this
5866 advances toward the outermost frame, to higher frame numbers, to frames
5867 that have existed longer. @var{n} defaults to one.
5870 @kindex do @r{(@code{down})}
5872 Move @var{n} frames down the stack. For positive numbers @var{n}, this
5873 advances toward the innermost frame, to lower frame numbers, to frames
5874 that were created more recently. @var{n} defaults to one. You may
5875 abbreviate @code{down} as @code{do}.
5878 All of these commands end by printing two lines of output describing the
5879 frame. The first line shows the frame number, the function name, the
5880 arguments, and the source file and line number of execution in that
5881 frame. The second line shows the text of that source line.
5889 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
5891 10 read_input_file (argv[i]);
5895 After such a printout, the @code{list} command with no arguments
5896 prints ten lines centered on the point of execution in the frame.
5897 You can also edit the program at the point of execution with your favorite
5898 editing program by typing @code{edit}.
5899 @xref{List, ,Printing Source Lines},
5903 @kindex down-silently
5905 @item up-silently @var{n}
5906 @itemx down-silently @var{n}
5907 These two commands are variants of @code{up} and @code{down},
5908 respectively; they differ in that they do their work silently, without
5909 causing display of the new frame. They are intended primarily for use
5910 in @value{GDBN} command scripts, where the output might be unnecessary and
5915 @section Information About a Frame
5917 There are several other commands to print information about the selected
5923 When used without any argument, this command does not change which
5924 frame is selected, but prints a brief description of the currently
5925 selected stack frame. It can be abbreviated @code{f}. With an
5926 argument, this command is used to select a stack frame.
5927 @xref{Selection, ,Selecting a Frame}.
5930 @kindex info f @r{(@code{info frame})}
5933 This command prints a verbose description of the selected stack frame,
5938 the address of the frame
5940 the address of the next frame down (called by this frame)
5942 the address of the next frame up (caller of this frame)
5944 the language in which the source code corresponding to this frame is written
5946 the address of the frame's arguments
5948 the address of the frame's local variables
5950 the program counter saved in it (the address of execution in the caller frame)
5952 which registers were saved in the frame
5955 @noindent The verbose description is useful when
5956 something has gone wrong that has made the stack format fail to fit
5957 the usual conventions.
5959 @item info frame @var{addr}
5960 @itemx info f @var{addr}
5961 Print a verbose description of the frame at address @var{addr}, without
5962 selecting that frame. The selected frame remains unchanged by this
5963 command. This requires the same kind of address (more than one for some
5964 architectures) that you specify in the @code{frame} command.
5965 @xref{Selection, ,Selecting a Frame}.
5969 Print the arguments of the selected frame, each on a separate line.
5973 Print the local variables of the selected frame, each on a separate
5974 line. These are all variables (declared either static or automatic)
5975 accessible at the point of execution of the selected frame.
5978 @cindex catch exceptions, list active handlers
5979 @cindex exception handlers, how to list
5981 Print a list of all the exception handlers that are active in the
5982 current stack frame at the current point of execution. To see other
5983 exception handlers, visit the associated frame (using the @code{up},
5984 @code{down}, or @code{frame} commands); then type @code{info catch}.
5985 @xref{Set Catchpoints, , Setting Catchpoints}.
5991 @chapter Examining Source Files
5993 @value{GDBN} can print parts of your program's source, since the debugging
5994 information recorded in the program tells @value{GDBN} what source files were
5995 used to build it. When your program stops, @value{GDBN} spontaneously prints
5996 the line where it stopped. Likewise, when you select a stack frame
5997 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
5998 execution in that frame has stopped. You can print other portions of
5999 source files by explicit command.
6001 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
6002 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
6003 @value{GDBN} under @sc{gnu} Emacs}.
6006 * List:: Printing source lines
6007 * Specify Location:: How to specify code locations
6008 * Edit:: Editing source files
6009 * Search:: Searching source files
6010 * Source Path:: Specifying source directories
6011 * Machine Code:: Source and machine code
6015 @section Printing Source Lines
6018 @kindex l @r{(@code{list})}
6019 To print lines from a source file, use the @code{list} command
6020 (abbreviated @code{l}). By default, ten lines are printed.
6021 There are several ways to specify what part of the file you want to
6022 print; see @ref{Specify Location}, for the full list.
6024 Here are the forms of the @code{list} command most commonly used:
6027 @item list @var{linenum}
6028 Print lines centered around line number @var{linenum} in the
6029 current source file.
6031 @item list @var{function}
6032 Print lines centered around the beginning of function
6036 Print more lines. If the last lines printed were printed with a
6037 @code{list} command, this prints lines following the last lines
6038 printed; however, if the last line printed was a solitary line printed
6039 as part of displaying a stack frame (@pxref{Stack, ,Examining the
6040 Stack}), this prints lines centered around that line.
6043 Print lines just before the lines last printed.
6046 @cindex @code{list}, how many lines to display
6047 By default, @value{GDBN} prints ten source lines with any of these forms of
6048 the @code{list} command. You can change this using @code{set listsize}:
6051 @kindex set listsize
6052 @item set listsize @var{count}
6053 Make the @code{list} command display @var{count} source lines (unless
6054 the @code{list} argument explicitly specifies some other number).
6056 @kindex show listsize
6058 Display the number of lines that @code{list} prints.
6061 Repeating a @code{list} command with @key{RET} discards the argument,
6062 so it is equivalent to typing just @code{list}. This is more useful
6063 than listing the same lines again. An exception is made for an
6064 argument of @samp{-}; that argument is preserved in repetition so that
6065 each repetition moves up in the source file.
6067 In general, the @code{list} command expects you to supply zero, one or two
6068 @dfn{linespecs}. Linespecs specify source lines; there are several ways
6069 of writing them (@pxref{Specify Location}), but the effect is always
6070 to specify some source line.
6072 Here is a complete description of the possible arguments for @code{list}:
6075 @item list @var{linespec}
6076 Print lines centered around the line specified by @var{linespec}.
6078 @item list @var{first},@var{last}
6079 Print lines from @var{first} to @var{last}. Both arguments are
6080 linespecs. When a @code{list} command has two linespecs, and the
6081 source file of the second linespec is omitted, this refers to
6082 the same source file as the first linespec.
6084 @item list ,@var{last}
6085 Print lines ending with @var{last}.
6087 @item list @var{first},
6088 Print lines starting with @var{first}.
6091 Print lines just after the lines last printed.
6094 Print lines just before the lines last printed.
6097 As described in the preceding table.
6100 @node Specify Location
6101 @section Specifying a Location
6102 @cindex specifying location
6105 Several @value{GDBN} commands accept arguments that specify a location
6106 of your program's code. Since @value{GDBN} is a source-level
6107 debugger, a location usually specifies some line in the source code;
6108 for that reason, locations are also known as @dfn{linespecs}.
6110 Here are all the different ways of specifying a code location that
6111 @value{GDBN} understands:
6115 Specifies the line number @var{linenum} of the current source file.
6118 @itemx +@var{offset}
6119 Specifies the line @var{offset} lines before or after the @dfn{current
6120 line}. For the @code{list} command, the current line is the last one
6121 printed; for the breakpoint commands, this is the line at which
6122 execution stopped in the currently selected @dfn{stack frame}
6123 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
6124 used as the second of the two linespecs in a @code{list} command,
6125 this specifies the line @var{offset} lines up or down from the first
6128 @item @var{filename}:@var{linenum}
6129 Specifies the line @var{linenum} in the source file @var{filename}.
6131 @item @var{function}
6132 Specifies the line that begins the body of the function @var{function}.
6133 For example, in C, this is the line with the open brace.
6135 @item @var{filename}:@var{function}
6136 Specifies the line that begins the body of the function @var{function}
6137 in the file @var{filename}. You only need the file name with a
6138 function name to avoid ambiguity when there are identically named
6139 functions in different source files.
6141 @item *@var{address}
6142 Specifies the program address @var{address}. For line-oriented
6143 commands, such as @code{list} and @code{edit}, this specifies a source
6144 line that contains @var{address}. For @code{break} and other
6145 breakpoint oriented commands, this can be used to set breakpoints in
6146 parts of your program which do not have debugging information or
6149 Here @var{address} may be any expression valid in the current working
6150 language (@pxref{Languages, working language}) that specifies a code
6151 address. In addition, as a convenience, @value{GDBN} extends the
6152 semantics of expressions used in locations to cover the situations
6153 that frequently happen during debugging. Here are the various forms
6157 @item @var{expression}
6158 Any expression valid in the current working language.
6160 @item @var{funcaddr}
6161 An address of a function or procedure derived from its name. In C,
6162 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
6163 simply the function's name @var{function} (and actually a special case
6164 of a valid expression). In Pascal and Modula-2, this is
6165 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
6166 (although the Pascal form also works).
6168 This form specifies the address of the function's first instruction,
6169 before the stack frame and arguments have been set up.
6171 @item '@var{filename}'::@var{funcaddr}
6172 Like @var{funcaddr} above, but also specifies the name of the source
6173 file explicitly. This is useful if the name of the function does not
6174 specify the function unambiguously, e.g., if there are several
6175 functions with identical names in different source files.
6182 @section Editing Source Files
6183 @cindex editing source files
6186 @kindex e @r{(@code{edit})}
6187 To edit the lines in a source file, use the @code{edit} command.
6188 The editing program of your choice
6189 is invoked with the current line set to
6190 the active line in the program.
6191 Alternatively, there are several ways to specify what part of the file you
6192 want to print if you want to see other parts of the program:
6195 @item edit @var{location}
6196 Edit the source file specified by @code{location}. Editing starts at
6197 that @var{location}, e.g., at the specified source line of the
6198 specified file. @xref{Specify Location}, for all the possible forms
6199 of the @var{location} argument; here are the forms of the @code{edit}
6200 command most commonly used:
6203 @item edit @var{number}
6204 Edit the current source file with @var{number} as the active line number.
6206 @item edit @var{function}
6207 Edit the file containing @var{function} at the beginning of its definition.
6212 @subsection Choosing your Editor
6213 You can customize @value{GDBN} to use any editor you want
6215 The only restriction is that your editor (say @code{ex}), recognizes the
6216 following command-line syntax:
6218 ex +@var{number} file
6220 The optional numeric value +@var{number} specifies the number of the line in
6221 the file where to start editing.}.
6222 By default, it is @file{@value{EDITOR}}, but you can change this
6223 by setting the environment variable @code{EDITOR} before using
6224 @value{GDBN}. For example, to configure @value{GDBN} to use the
6225 @code{vi} editor, you could use these commands with the @code{sh} shell:
6231 or in the @code{csh} shell,
6233 setenv EDITOR /usr/bin/vi
6238 @section Searching Source Files
6239 @cindex searching source files
6241 There are two commands for searching through the current source file for a
6246 @kindex forward-search
6247 @item forward-search @var{regexp}
6248 @itemx search @var{regexp}
6249 The command @samp{forward-search @var{regexp}} checks each line,
6250 starting with the one following the last line listed, for a match for
6251 @var{regexp}. It lists the line that is found. You can use the
6252 synonym @samp{search @var{regexp}} or abbreviate the command name as
6255 @kindex reverse-search
6256 @item reverse-search @var{regexp}
6257 The command @samp{reverse-search @var{regexp}} checks each line, starting
6258 with the one before the last line listed and going backward, for a match
6259 for @var{regexp}. It lists the line that is found. You can abbreviate
6260 this command as @code{rev}.
6264 @section Specifying Source Directories
6267 @cindex directories for source files
6268 Executable programs sometimes do not record the directories of the source
6269 files from which they were compiled, just the names. Even when they do,
6270 the directories could be moved between the compilation and your debugging
6271 session. @value{GDBN} has a list of directories to search for source files;
6272 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
6273 it tries all the directories in the list, in the order they are present
6274 in the list, until it finds a file with the desired name.
6276 For example, suppose an executable references the file
6277 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
6278 @file{/mnt/cross}. The file is first looked up literally; if this
6279 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
6280 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
6281 message is printed. @value{GDBN} does not look up the parts of the
6282 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
6283 Likewise, the subdirectories of the source path are not searched: if
6284 the source path is @file{/mnt/cross}, and the binary refers to
6285 @file{foo.c}, @value{GDBN} would not find it under
6286 @file{/mnt/cross/usr/src/foo-1.0/lib}.
6288 Plain file names, relative file names with leading directories, file
6289 names containing dots, etc.@: are all treated as described above; for
6290 instance, if the source path is @file{/mnt/cross}, and the source file
6291 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
6292 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
6293 that---@file{/mnt/cross/foo.c}.
6295 Note that the executable search path is @emph{not} used to locate the
6298 Whenever you reset or rearrange the source path, @value{GDBN} clears out
6299 any information it has cached about where source files are found and where
6300 each line is in the file.
6304 When you start @value{GDBN}, its source path includes only @samp{cdir}
6305 and @samp{cwd}, in that order.
6306 To add other directories, use the @code{directory} command.
6308 The search path is used to find both program source files and @value{GDBN}
6309 script files (read using the @samp{-command} option and @samp{source} command).
6311 In addition to the source path, @value{GDBN} provides a set of commands
6312 that manage a list of source path substitution rules. A @dfn{substitution
6313 rule} specifies how to rewrite source directories stored in the program's
6314 debug information in case the sources were moved to a different
6315 directory between compilation and debugging. A rule is made of
6316 two strings, the first specifying what needs to be rewritten in
6317 the path, and the second specifying how it should be rewritten.
6318 In @ref{set substitute-path}, we name these two parts @var{from} and
6319 @var{to} respectively. @value{GDBN} does a simple string replacement
6320 of @var{from} with @var{to} at the start of the directory part of the
6321 source file name, and uses that result instead of the original file
6322 name to look up the sources.
6324 Using the previous example, suppose the @file{foo-1.0} tree has been
6325 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
6326 @value{GDBN} to replace @file{/usr/src} in all source path names with
6327 @file{/mnt/cross}. The first lookup will then be
6328 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
6329 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
6330 substitution rule, use the @code{set substitute-path} command
6331 (@pxref{set substitute-path}).
6333 To avoid unexpected substitution results, a rule is applied only if the
6334 @var{from} part of the directory name ends at a directory separator.
6335 For instance, a rule substituting @file{/usr/source} into
6336 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
6337 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
6338 is applied only at the beginning of the directory name, this rule will
6339 not be applied to @file{/root/usr/source/baz.c} either.
6341 In many cases, you can achieve the same result using the @code{directory}
6342 command. However, @code{set substitute-path} can be more efficient in
6343 the case where the sources are organized in a complex tree with multiple
6344 subdirectories. With the @code{directory} command, you need to add each
6345 subdirectory of your project. If you moved the entire tree while
6346 preserving its internal organization, then @code{set substitute-path}
6347 allows you to direct the debugger to all the sources with one single
6350 @code{set substitute-path} is also more than just a shortcut command.
6351 The source path is only used if the file at the original location no
6352 longer exists. On the other hand, @code{set substitute-path} modifies
6353 the debugger behavior to look at the rewritten location instead. So, if
6354 for any reason a source file that is not relevant to your executable is
6355 located at the original location, a substitution rule is the only
6356 method available to point @value{GDBN} at the new location.
6358 @cindex @samp{--with-relocated-sources}
6359 @cindex default source path substitution
6360 You can configure a default source path substitution rule by
6361 configuring @value{GDBN} with the
6362 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
6363 should be the name of a directory under @value{GDBN}'s configured
6364 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
6365 directory names in debug information under @var{dir} will be adjusted
6366 automatically if the installed @value{GDBN} is moved to a new
6367 location. This is useful if @value{GDBN}, libraries or executables
6368 with debug information and corresponding source code are being moved
6372 @item directory @var{dirname} @dots{}
6373 @item dir @var{dirname} @dots{}
6374 Add directory @var{dirname} to the front of the source path. Several
6375 directory names may be given to this command, separated by @samp{:}
6376 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
6377 part of absolute file names) or
6378 whitespace. You may specify a directory that is already in the source
6379 path; this moves it forward, so @value{GDBN} searches it sooner.
6383 @vindex $cdir@r{, convenience variable}
6384 @vindex $cwd@r{, convenience variable}
6385 @cindex compilation directory
6386 @cindex current directory
6387 @cindex working directory
6388 @cindex directory, current
6389 @cindex directory, compilation
6390 You can use the string @samp{$cdir} to refer to the compilation
6391 directory (if one is recorded), and @samp{$cwd} to refer to the current
6392 working directory. @samp{$cwd} is not the same as @samp{.}---the former
6393 tracks the current working directory as it changes during your @value{GDBN}
6394 session, while the latter is immediately expanded to the current
6395 directory at the time you add an entry to the source path.
6398 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
6400 @c RET-repeat for @code{directory} is explicitly disabled, but since
6401 @c repeating it would be a no-op we do not say that. (thanks to RMS)
6403 @item show directories
6404 @kindex show directories
6405 Print the source path: show which directories it contains.
6407 @anchor{set substitute-path}
6408 @item set substitute-path @var{from} @var{to}
6409 @kindex set substitute-path
6410 Define a source path substitution rule, and add it at the end of the
6411 current list of existing substitution rules. If a rule with the same
6412 @var{from} was already defined, then the old rule is also deleted.
6414 For example, if the file @file{/foo/bar/baz.c} was moved to
6415 @file{/mnt/cross/baz.c}, then the command
6418 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
6422 will tell @value{GDBN} to replace @samp{/usr/src} with
6423 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
6424 @file{baz.c} even though it was moved.
6426 In the case when more than one substitution rule have been defined,
6427 the rules are evaluated one by one in the order where they have been
6428 defined. The first one matching, if any, is selected to perform
6431 For instance, if we had entered the following commands:
6434 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
6435 (@value{GDBP}) set substitute-path /usr/src /mnt/src
6439 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
6440 @file{/mnt/include/defs.h} by using the first rule. However, it would
6441 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
6442 @file{/mnt/src/lib/foo.c}.
6445 @item unset substitute-path [path]
6446 @kindex unset substitute-path
6447 If a path is specified, search the current list of substitution rules
6448 for a rule that would rewrite that path. Delete that rule if found.
6449 A warning is emitted by the debugger if no rule could be found.
6451 If no path is specified, then all substitution rules are deleted.
6453 @item show substitute-path [path]
6454 @kindex show substitute-path
6455 If a path is specified, then print the source path substitution rule
6456 which would rewrite that path, if any.
6458 If no path is specified, then print all existing source path substitution
6463 If your source path is cluttered with directories that are no longer of
6464 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
6465 versions of source. You can correct the situation as follows:
6469 Use @code{directory} with no argument to reset the source path to its default value.
6472 Use @code{directory} with suitable arguments to reinstall the
6473 directories you want in the source path. You can add all the
6474 directories in one command.
6478 @section Source and Machine Code
6479 @cindex source line and its code address
6481 You can use the command @code{info line} to map source lines to program
6482 addresses (and vice versa), and the command @code{disassemble} to display
6483 a range of addresses as machine instructions. You can use the command
6484 @code{set disassemble-next-line} to set whether to disassemble next
6485 source line when execution stops. When run under @sc{gnu} Emacs
6486 mode, the @code{info line} command causes the arrow to point to the
6487 line specified. Also, @code{info line} prints addresses in symbolic form as
6492 @item info line @var{linespec}
6493 Print the starting and ending addresses of the compiled code for
6494 source line @var{linespec}. You can specify source lines in any of
6495 the ways documented in @ref{Specify Location}.
6498 For example, we can use @code{info line} to discover the location of
6499 the object code for the first line of function
6500 @code{m4_changequote}:
6502 @c FIXME: I think this example should also show the addresses in
6503 @c symbolic form, as they usually would be displayed.
6505 (@value{GDBP}) info line m4_changequote
6506 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
6510 @cindex code address and its source line
6511 We can also inquire (using @code{*@var{addr}} as the form for
6512 @var{linespec}) what source line covers a particular address:
6514 (@value{GDBP}) info line *0x63ff
6515 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
6518 @cindex @code{$_} and @code{info line}
6519 @cindex @code{x} command, default address
6520 @kindex x@r{(examine), and} info line
6521 After @code{info line}, the default address for the @code{x} command
6522 is changed to the starting address of the line, so that @samp{x/i} is
6523 sufficient to begin examining the machine code (@pxref{Memory,
6524 ,Examining Memory}). Also, this address is saved as the value of the
6525 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
6530 @cindex assembly instructions
6531 @cindex instructions, assembly
6532 @cindex machine instructions
6533 @cindex listing machine instructions
6535 @itemx disassemble /m
6536 @itemx disassemble /r
6537 This specialized command dumps a range of memory as machine
6538 instructions. It can also print mixed source+disassembly by specifying
6539 the @code{/m} modifier and print the raw instructions in hex as well as
6540 in symbolic form by specifying the @code{/r}.
6541 The default memory range is the function surrounding the
6542 program counter of the selected frame. A single argument to this
6543 command is a program counter value; @value{GDBN} dumps the function
6544 surrounding this value. When two arguments are given, they should
6545 be separated by a comma, possibly surrounded by whitespace. The
6546 arguments specify a range of addresses (first inclusive, second exclusive)
6547 to dump. In that case, the name of the function is also printed (since
6548 there could be several functions in the given range).
6550 The argument(s) can be any expression yielding a numeric value, such as
6551 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
6553 If the range of memory being disassembled contains current program counter,
6554 the instruction at that location is shown with a @code{=>} marker.
6557 The following example shows the disassembly of a range of addresses of
6558 HP PA-RISC 2.0 code:
6561 (@value{GDBP}) disas 0x32c4, 0x32e4
6562 Dump of assembler code from 0x32c4 to 0x32e4:
6563 0x32c4 <main+204>: addil 0,dp
6564 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
6565 0x32cc <main+212>: ldil 0x3000,r31
6566 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
6567 0x32d4 <main+220>: ldo 0(r31),rp
6568 0x32d8 <main+224>: addil -0x800,dp
6569 0x32dc <main+228>: ldo 0x588(r1),r26
6570 0x32e0 <main+232>: ldil 0x3000,r31
6571 End of assembler dump.
6574 Here is an example showing mixed source+assembly for Intel x86, when the
6575 program is stopped just after function prologue:
6578 (@value{GDBP}) disas /m main
6579 Dump of assembler code for function main:
6581 0x08048330 <+0>: push %ebp
6582 0x08048331 <+1>: mov %esp,%ebp
6583 0x08048333 <+3>: sub $0x8,%esp
6584 0x08048336 <+6>: and $0xfffffff0,%esp
6585 0x08048339 <+9>: sub $0x10,%esp
6587 6 printf ("Hello.\n");
6588 => 0x0804833c <+12>: movl $0x8048440,(%esp)
6589 0x08048343 <+19>: call 0x8048284 <puts@@plt>
6593 0x08048348 <+24>: mov $0x0,%eax
6594 0x0804834d <+29>: leave
6595 0x0804834e <+30>: ret
6597 End of assembler dump.
6600 Some architectures have more than one commonly-used set of instruction
6601 mnemonics or other syntax.
6603 For programs that were dynamically linked and use shared libraries,
6604 instructions that call functions or branch to locations in the shared
6605 libraries might show a seemingly bogus location---it's actually a
6606 location of the relocation table. On some architectures, @value{GDBN}
6607 might be able to resolve these to actual function names.
6610 @kindex set disassembly-flavor
6611 @cindex Intel disassembly flavor
6612 @cindex AT&T disassembly flavor
6613 @item set disassembly-flavor @var{instruction-set}
6614 Select the instruction set to use when disassembling the
6615 program via the @code{disassemble} or @code{x/i} commands.
6617 Currently this command is only defined for the Intel x86 family. You
6618 can set @var{instruction-set} to either @code{intel} or @code{att}.
6619 The default is @code{att}, the AT&T flavor used by default by Unix
6620 assemblers for x86-based targets.
6622 @kindex show disassembly-flavor
6623 @item show disassembly-flavor
6624 Show the current setting of the disassembly flavor.
6628 @kindex set disassemble-next-line
6629 @kindex show disassemble-next-line
6630 @item set disassemble-next-line
6631 @itemx show disassemble-next-line
6632 Control whether or not @value{GDBN} will disassemble the next source
6633 line or instruction when execution stops. If ON, @value{GDBN} will
6634 display disassembly of the next source line when execution of the
6635 program being debugged stops. This is @emph{in addition} to
6636 displaying the source line itself, which @value{GDBN} always does if
6637 possible. If the next source line cannot be displayed for some reason
6638 (e.g., if @value{GDBN} cannot find the source file, or there's no line
6639 info in the debug info), @value{GDBN} will display disassembly of the
6640 next @emph{instruction} instead of showing the next source line. If
6641 AUTO, @value{GDBN} will display disassembly of next instruction only
6642 if the source line cannot be displayed. This setting causes
6643 @value{GDBN} to display some feedback when you step through a function
6644 with no line info or whose source file is unavailable. The default is
6645 OFF, which means never display the disassembly of the next line or
6651 @chapter Examining Data
6653 @cindex printing data
6654 @cindex examining data
6657 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
6658 @c document because it is nonstandard... Under Epoch it displays in a
6659 @c different window or something like that.
6660 The usual way to examine data in your program is with the @code{print}
6661 command (abbreviated @code{p}), or its synonym @code{inspect}. It
6662 evaluates and prints the value of an expression of the language your
6663 program is written in (@pxref{Languages, ,Using @value{GDBN} with
6664 Different Languages}).
6667 @item print @var{expr}
6668 @itemx print /@var{f} @var{expr}
6669 @var{expr} is an expression (in the source language). By default the
6670 value of @var{expr} is printed in a format appropriate to its data type;
6671 you can choose a different format by specifying @samp{/@var{f}}, where
6672 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
6676 @itemx print /@var{f}
6677 @cindex reprint the last value
6678 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
6679 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
6680 conveniently inspect the same value in an alternative format.
6683 A more low-level way of examining data is with the @code{x} command.
6684 It examines data in memory at a specified address and prints it in a
6685 specified format. @xref{Memory, ,Examining Memory}.
6687 If you are interested in information about types, or about how the
6688 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
6689 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
6693 * Expressions:: Expressions
6694 * Ambiguous Expressions:: Ambiguous Expressions
6695 * Variables:: Program variables
6696 * Arrays:: Artificial arrays
6697 * Output Formats:: Output formats
6698 * Memory:: Examining memory
6699 * Auto Display:: Automatic display
6700 * Print Settings:: Print settings
6701 * Value History:: Value history
6702 * Convenience Vars:: Convenience variables
6703 * Registers:: Registers
6704 * Floating Point Hardware:: Floating point hardware
6705 * Vector Unit:: Vector Unit
6706 * OS Information:: Auxiliary data provided by operating system
6707 * Memory Region Attributes:: Memory region attributes
6708 * Dump/Restore Files:: Copy between memory and a file
6709 * Core File Generation:: Cause a program dump its core
6710 * Character Sets:: Debugging programs that use a different
6711 character set than GDB does
6712 * Caching Remote Data:: Data caching for remote targets
6713 * Searching Memory:: Searching memory for a sequence of bytes
6717 @section Expressions
6720 @code{print} and many other @value{GDBN} commands accept an expression and
6721 compute its value. Any kind of constant, variable or operator defined
6722 by the programming language you are using is valid in an expression in
6723 @value{GDBN}. This includes conditional expressions, function calls,
6724 casts, and string constants. It also includes preprocessor macros, if
6725 you compiled your program to include this information; see
6728 @cindex arrays in expressions
6729 @value{GDBN} supports array constants in expressions input by
6730 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
6731 you can use the command @code{print @{1, 2, 3@}} to create an array
6732 of three integers. If you pass an array to a function or assign it
6733 to a program variable, @value{GDBN} copies the array to memory that
6734 is @code{malloc}ed in the target program.
6736 Because C is so widespread, most of the expressions shown in examples in
6737 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
6738 Languages}, for information on how to use expressions in other
6741 In this section, we discuss operators that you can use in @value{GDBN}
6742 expressions regardless of your programming language.
6744 @cindex casts, in expressions
6745 Casts are supported in all languages, not just in C, because it is so
6746 useful to cast a number into a pointer in order to examine a structure
6747 at that address in memory.
6748 @c FIXME: casts supported---Mod2 true?
6750 @value{GDBN} supports these operators, in addition to those common
6751 to programming languages:
6755 @samp{@@} is a binary operator for treating parts of memory as arrays.
6756 @xref{Arrays, ,Artificial Arrays}, for more information.
6759 @samp{::} allows you to specify a variable in terms of the file or
6760 function where it is defined. @xref{Variables, ,Program Variables}.
6762 @cindex @{@var{type}@}
6763 @cindex type casting memory
6764 @cindex memory, viewing as typed object
6765 @cindex casts, to view memory
6766 @item @{@var{type}@} @var{addr}
6767 Refers to an object of type @var{type} stored at address @var{addr} in
6768 memory. @var{addr} may be any expression whose value is an integer or
6769 pointer (but parentheses are required around binary operators, just as in
6770 a cast). This construct is allowed regardless of what kind of data is
6771 normally supposed to reside at @var{addr}.
6774 @node Ambiguous Expressions
6775 @section Ambiguous Expressions
6776 @cindex ambiguous expressions
6778 Expressions can sometimes contain some ambiguous elements. For instance,
6779 some programming languages (notably Ada, C@t{++} and Objective-C) permit
6780 a single function name to be defined several times, for application in
6781 different contexts. This is called @dfn{overloading}. Another example
6782 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
6783 templates and is typically instantiated several times, resulting in
6784 the same function name being defined in different contexts.
6786 In some cases and depending on the language, it is possible to adjust
6787 the expression to remove the ambiguity. For instance in C@t{++}, you
6788 can specify the signature of the function you want to break on, as in
6789 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
6790 qualified name of your function often makes the expression unambiguous
6793 When an ambiguity that needs to be resolved is detected, the debugger
6794 has the capability to display a menu of numbered choices for each
6795 possibility, and then waits for the selection with the prompt @samp{>}.
6796 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
6797 aborts the current command. If the command in which the expression was
6798 used allows more than one choice to be selected, the next option in the
6799 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
6802 For example, the following session excerpt shows an attempt to set a
6803 breakpoint at the overloaded symbol @code{String::after}.
6804 We choose three particular definitions of that function name:
6806 @c FIXME! This is likely to change to show arg type lists, at least
6809 (@value{GDBP}) b String::after
6812 [2] file:String.cc; line number:867
6813 [3] file:String.cc; line number:860
6814 [4] file:String.cc; line number:875
6815 [5] file:String.cc; line number:853
6816 [6] file:String.cc; line number:846
6817 [7] file:String.cc; line number:735
6819 Breakpoint 1 at 0xb26c: file String.cc, line 867.
6820 Breakpoint 2 at 0xb344: file String.cc, line 875.
6821 Breakpoint 3 at 0xafcc: file String.cc, line 846.
6822 Multiple breakpoints were set.
6823 Use the "delete" command to delete unwanted
6830 @kindex set multiple-symbols
6831 @item set multiple-symbols @var{mode}
6832 @cindex multiple-symbols menu
6834 This option allows you to adjust the debugger behavior when an expression
6837 By default, @var{mode} is set to @code{all}. If the command with which
6838 the expression is used allows more than one choice, then @value{GDBN}
6839 automatically selects all possible choices. For instance, inserting
6840 a breakpoint on a function using an ambiguous name results in a breakpoint
6841 inserted on each possible match. However, if a unique choice must be made,
6842 then @value{GDBN} uses the menu to help you disambiguate the expression.
6843 For instance, printing the address of an overloaded function will result
6844 in the use of the menu.
6846 When @var{mode} is set to @code{ask}, the debugger always uses the menu
6847 when an ambiguity is detected.
6849 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
6850 an error due to the ambiguity and the command is aborted.
6852 @kindex show multiple-symbols
6853 @item show multiple-symbols
6854 Show the current value of the @code{multiple-symbols} setting.
6858 @section Program Variables
6860 The most common kind of expression to use is the name of a variable
6863 Variables in expressions are understood in the selected stack frame
6864 (@pxref{Selection, ,Selecting a Frame}); they must be either:
6868 global (or file-static)
6875 visible according to the scope rules of the
6876 programming language from the point of execution in that frame
6879 @noindent This means that in the function
6894 you can examine and use the variable @code{a} whenever your program is
6895 executing within the function @code{foo}, but you can only use or
6896 examine the variable @code{b} while your program is executing inside
6897 the block where @code{b} is declared.
6899 @cindex variable name conflict
6900 There is an exception: you can refer to a variable or function whose
6901 scope is a single source file even if the current execution point is not
6902 in this file. But it is possible to have more than one such variable or
6903 function with the same name (in different source files). If that
6904 happens, referring to that name has unpredictable effects. If you wish,
6905 you can specify a static variable in a particular function or file,
6906 using the colon-colon (@code{::}) notation:
6908 @cindex colon-colon, context for variables/functions
6910 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
6911 @cindex @code{::}, context for variables/functions
6914 @var{file}::@var{variable}
6915 @var{function}::@var{variable}
6919 Here @var{file} or @var{function} is the name of the context for the
6920 static @var{variable}. In the case of file names, you can use quotes to
6921 make sure @value{GDBN} parses the file name as a single word---for example,
6922 to print a global value of @code{x} defined in @file{f2.c}:
6925 (@value{GDBP}) p 'f2.c'::x
6928 @cindex C@t{++} scope resolution
6929 This use of @samp{::} is very rarely in conflict with the very similar
6930 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
6931 scope resolution operator in @value{GDBN} expressions.
6932 @c FIXME: Um, so what happens in one of those rare cases where it's in
6935 @cindex wrong values
6936 @cindex variable values, wrong
6937 @cindex function entry/exit, wrong values of variables
6938 @cindex optimized code, wrong values of variables
6940 @emph{Warning:} Occasionally, a local variable may appear to have the
6941 wrong value at certain points in a function---just after entry to a new
6942 scope, and just before exit.
6944 You may see this problem when you are stepping by machine instructions.
6945 This is because, on most machines, it takes more than one instruction to
6946 set up a stack frame (including local variable definitions); if you are
6947 stepping by machine instructions, variables may appear to have the wrong
6948 values until the stack frame is completely built. On exit, it usually
6949 also takes more than one machine instruction to destroy a stack frame;
6950 after you begin stepping through that group of instructions, local
6951 variable definitions may be gone.
6953 This may also happen when the compiler does significant optimizations.
6954 To be sure of always seeing accurate values, turn off all optimization
6957 @cindex ``No symbol "foo" in current context''
6958 Another possible effect of compiler optimizations is to optimize
6959 unused variables out of existence, or assign variables to registers (as
6960 opposed to memory addresses). Depending on the support for such cases
6961 offered by the debug info format used by the compiler, @value{GDBN}
6962 might not be able to display values for such local variables. If that
6963 happens, @value{GDBN} will print a message like this:
6966 No symbol "foo" in current context.
6969 To solve such problems, either recompile without optimizations, or use a
6970 different debug info format, if the compiler supports several such
6971 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
6972 usually supports the @option{-gstabs+} option. @option{-gstabs+}
6973 produces debug info in a format that is superior to formats such as
6974 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
6975 an effective form for debug info. @xref{Debugging Options,,Options
6976 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
6977 Compiler Collection (GCC)}.
6978 @xref{C, ,C and C@t{++}}, for more information about debug info formats
6979 that are best suited to C@t{++} programs.
6981 If you ask to print an object whose contents are unknown to
6982 @value{GDBN}, e.g., because its data type is not completely specified
6983 by the debug information, @value{GDBN} will say @samp{<incomplete
6984 type>}. @xref{Symbols, incomplete type}, for more about this.
6986 Strings are identified as arrays of @code{char} values without specified
6987 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
6988 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
6989 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
6990 defines literal string type @code{"char"} as @code{char} without a sign.
6995 signed char var1[] = "A";
6998 You get during debugging
7003 $2 = @{65 'A', 0 '\0'@}
7007 @section Artificial Arrays
7009 @cindex artificial array
7011 @kindex @@@r{, referencing memory as an array}
7012 It is often useful to print out several successive objects of the
7013 same type in memory; a section of an array, or an array of
7014 dynamically determined size for which only a pointer exists in the
7017 You can do this by referring to a contiguous span of memory as an
7018 @dfn{artificial array}, using the binary operator @samp{@@}. The left
7019 operand of @samp{@@} should be the first element of the desired array
7020 and be an individual object. The right operand should be the desired length
7021 of the array. The result is an array value whose elements are all of
7022 the type of the left argument. The first element is actually the left
7023 argument; the second element comes from bytes of memory immediately
7024 following those that hold the first element, and so on. Here is an
7025 example. If a program says
7028 int *array = (int *) malloc (len * sizeof (int));
7032 you can print the contents of @code{array} with
7038 The left operand of @samp{@@} must reside in memory. Array values made
7039 with @samp{@@} in this way behave just like other arrays in terms of
7040 subscripting, and are coerced to pointers when used in expressions.
7041 Artificial arrays most often appear in expressions via the value history
7042 (@pxref{Value History, ,Value History}), after printing one out.
7044 Another way to create an artificial array is to use a cast.
7045 This re-interprets a value as if it were an array.
7046 The value need not be in memory:
7048 (@value{GDBP}) p/x (short[2])0x12345678
7049 $1 = @{0x1234, 0x5678@}
7052 As a convenience, if you leave the array length out (as in
7053 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
7054 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
7056 (@value{GDBP}) p/x (short[])0x12345678
7057 $2 = @{0x1234, 0x5678@}
7060 Sometimes the artificial array mechanism is not quite enough; in
7061 moderately complex data structures, the elements of interest may not
7062 actually be adjacent---for example, if you are interested in the values
7063 of pointers in an array. One useful work-around in this situation is
7064 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
7065 Variables}) as a counter in an expression that prints the first
7066 interesting value, and then repeat that expression via @key{RET}. For
7067 instance, suppose you have an array @code{dtab} of pointers to
7068 structures, and you are interested in the values of a field @code{fv}
7069 in each structure. Here is an example of what you might type:
7079 @node Output Formats
7080 @section Output Formats
7082 @cindex formatted output
7083 @cindex output formats
7084 By default, @value{GDBN} prints a value according to its data type. Sometimes
7085 this is not what you want. For example, you might want to print a number
7086 in hex, or a pointer in decimal. Or you might want to view data in memory
7087 at a certain address as a character string or as an instruction. To do
7088 these things, specify an @dfn{output format} when you print a value.
7090 The simplest use of output formats is to say how to print a value
7091 already computed. This is done by starting the arguments of the
7092 @code{print} command with a slash and a format letter. The format
7093 letters supported are:
7097 Regard the bits of the value as an integer, and print the integer in
7101 Print as integer in signed decimal.
7104 Print as integer in unsigned decimal.
7107 Print as integer in octal.
7110 Print as integer in binary. The letter @samp{t} stands for ``two''.
7111 @footnote{@samp{b} cannot be used because these format letters are also
7112 used with the @code{x} command, where @samp{b} stands for ``byte'';
7113 see @ref{Memory,,Examining Memory}.}
7116 @cindex unknown address, locating
7117 @cindex locate address
7118 Print as an address, both absolute in hexadecimal and as an offset from
7119 the nearest preceding symbol. You can use this format used to discover
7120 where (in what function) an unknown address is located:
7123 (@value{GDBP}) p/a 0x54320
7124 $3 = 0x54320 <_initialize_vx+396>
7128 The command @code{info symbol 0x54320} yields similar results.
7129 @xref{Symbols, info symbol}.
7132 Regard as an integer and print it as a character constant. This
7133 prints both the numerical value and its character representation. The
7134 character representation is replaced with the octal escape @samp{\nnn}
7135 for characters outside the 7-bit @sc{ascii} range.
7137 Without this format, @value{GDBN} displays @code{char},
7138 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
7139 constants. Single-byte members of vectors are displayed as integer
7143 Regard the bits of the value as a floating point number and print
7144 using typical floating point syntax.
7147 @cindex printing strings
7148 @cindex printing byte arrays
7149 Regard as a string, if possible. With this format, pointers to single-byte
7150 data are displayed as null-terminated strings and arrays of single-byte data
7151 are displayed as fixed-length strings. Other values are displayed in their
7154 Without this format, @value{GDBN} displays pointers to and arrays of
7155 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
7156 strings. Single-byte members of a vector are displayed as an integer
7160 @cindex raw printing
7161 Print using the @samp{raw} formatting. By default, @value{GDBN} will
7162 use a type-specific pretty-printer. The @samp{r} format bypasses any
7163 pretty-printer which might exist for the value's type.
7166 For example, to print the program counter in hex (@pxref{Registers}), type
7173 Note that no space is required before the slash; this is because command
7174 names in @value{GDBN} cannot contain a slash.
7176 To reprint the last value in the value history with a different format,
7177 you can use the @code{print} command with just a format and no
7178 expression. For example, @samp{p/x} reprints the last value in hex.
7181 @section Examining Memory
7183 You can use the command @code{x} (for ``examine'') to examine memory in
7184 any of several formats, independently of your program's data types.
7186 @cindex examining memory
7188 @kindex x @r{(examine memory)}
7189 @item x/@var{nfu} @var{addr}
7192 Use the @code{x} command to examine memory.
7195 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
7196 much memory to display and how to format it; @var{addr} is an
7197 expression giving the address where you want to start displaying memory.
7198 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
7199 Several commands set convenient defaults for @var{addr}.
7202 @item @var{n}, the repeat count
7203 The repeat count is a decimal integer; the default is 1. It specifies
7204 how much memory (counting by units @var{u}) to display.
7205 @c This really is **decimal**; unaffected by 'set radix' as of GDB
7208 @item @var{f}, the display format
7209 The display format is one of the formats used by @code{print}
7210 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
7211 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
7212 The default is @samp{x} (hexadecimal) initially. The default changes
7213 each time you use either @code{x} or @code{print}.
7215 @item @var{u}, the unit size
7216 The unit size is any of
7222 Halfwords (two bytes).
7224 Words (four bytes). This is the initial default.
7226 Giant words (eight bytes).
7229 Each time you specify a unit size with @code{x}, that size becomes the
7230 default unit the next time you use @code{x}. (For the @samp{s} and
7231 @samp{i} formats, the unit size is ignored and is normally not written.)
7233 @item @var{addr}, starting display address
7234 @var{addr} is the address where you want @value{GDBN} to begin displaying
7235 memory. The expression need not have a pointer value (though it may);
7236 it is always interpreted as an integer address of a byte of memory.
7237 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
7238 @var{addr} is usually just after the last address examined---but several
7239 other commands also set the default address: @code{info breakpoints} (to
7240 the address of the last breakpoint listed), @code{info line} (to the
7241 starting address of a line), and @code{print} (if you use it to display
7242 a value from memory).
7245 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
7246 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
7247 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
7248 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
7249 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
7251 Since the letters indicating unit sizes are all distinct from the
7252 letters specifying output formats, you do not have to remember whether
7253 unit size or format comes first; either order works. The output
7254 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
7255 (However, the count @var{n} must come first; @samp{wx4} does not work.)
7257 Even though the unit size @var{u} is ignored for the formats @samp{s}
7258 and @samp{i}, you might still want to use a count @var{n}; for example,
7259 @samp{3i} specifies that you want to see three machine instructions,
7260 including any operands. For convenience, especially when used with
7261 the @code{display} command, the @samp{i} format also prints branch delay
7262 slot instructions, if any, beyond the count specified, which immediately
7263 follow the last instruction that is within the count. The command
7264 @code{disassemble} gives an alternative way of inspecting machine
7265 instructions; see @ref{Machine Code,,Source and Machine Code}.
7267 All the defaults for the arguments to @code{x} are designed to make it
7268 easy to continue scanning memory with minimal specifications each time
7269 you use @code{x}. For example, after you have inspected three machine
7270 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
7271 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
7272 the repeat count @var{n} is used again; the other arguments default as
7273 for successive uses of @code{x}.
7275 When examining machine instructions, the instruction at current program
7276 counter is shown with a @code{=>} marker. For example:
7279 (@value{GDBP}) x/5i $pc-6
7280 0x804837f <main+11>: mov %esp,%ebp
7281 0x8048381 <main+13>: push %ecx
7282 0x8048382 <main+14>: sub $0x4,%esp
7283 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
7284 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
7287 @cindex @code{$_}, @code{$__}, and value history
7288 The addresses and contents printed by the @code{x} command are not saved
7289 in the value history because there is often too much of them and they
7290 would get in the way. Instead, @value{GDBN} makes these values available for
7291 subsequent use in expressions as values of the convenience variables
7292 @code{$_} and @code{$__}. After an @code{x} command, the last address
7293 examined is available for use in expressions in the convenience variable
7294 @code{$_}. The contents of that address, as examined, are available in
7295 the convenience variable @code{$__}.
7297 If the @code{x} command has a repeat count, the address and contents saved
7298 are from the last memory unit printed; this is not the same as the last
7299 address printed if several units were printed on the last line of output.
7301 @cindex remote memory comparison
7302 @cindex verify remote memory image
7303 When you are debugging a program running on a remote target machine
7304 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
7305 remote machine's memory against the executable file you downloaded to
7306 the target. The @code{compare-sections} command is provided for such
7310 @kindex compare-sections
7311 @item compare-sections @r{[}@var{section-name}@r{]}
7312 Compare the data of a loadable section @var{section-name} in the
7313 executable file of the program being debugged with the same section in
7314 the remote machine's memory, and report any mismatches. With no
7315 arguments, compares all loadable sections. This command's
7316 availability depends on the target's support for the @code{"qCRC"}
7321 @section Automatic Display
7322 @cindex automatic display
7323 @cindex display of expressions
7325 If you find that you want to print the value of an expression frequently
7326 (to see how it changes), you might want to add it to the @dfn{automatic
7327 display list} so that @value{GDBN} prints its value each time your program stops.
7328 Each expression added to the list is given a number to identify it;
7329 to remove an expression from the list, you specify that number.
7330 The automatic display looks like this:
7334 3: bar[5] = (struct hack *) 0x3804
7338 This display shows item numbers, expressions and their current values. As with
7339 displays you request manually using @code{x} or @code{print}, you can
7340 specify the output format you prefer; in fact, @code{display} decides
7341 whether to use @code{print} or @code{x} depending your format
7342 specification---it uses @code{x} if you specify either the @samp{i}
7343 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
7347 @item display @var{expr}
7348 Add the expression @var{expr} to the list of expressions to display
7349 each time your program stops. @xref{Expressions, ,Expressions}.
7351 @code{display} does not repeat if you press @key{RET} again after using it.
7353 @item display/@var{fmt} @var{expr}
7354 For @var{fmt} specifying only a display format and not a size or
7355 count, add the expression @var{expr} to the auto-display list but
7356 arrange to display it each time in the specified format @var{fmt}.
7357 @xref{Output Formats,,Output Formats}.
7359 @item display/@var{fmt} @var{addr}
7360 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
7361 number of units, add the expression @var{addr} as a memory address to
7362 be examined each time your program stops. Examining means in effect
7363 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
7366 For example, @samp{display/i $pc} can be helpful, to see the machine
7367 instruction about to be executed each time execution stops (@samp{$pc}
7368 is a common name for the program counter; @pxref{Registers, ,Registers}).
7371 @kindex delete display
7373 @item undisplay @var{dnums}@dots{}
7374 @itemx delete display @var{dnums}@dots{}
7375 Remove item numbers @var{dnums} from the list of expressions to display.
7377 @code{undisplay} does not repeat if you press @key{RET} after using it.
7378 (Otherwise you would just get the error @samp{No display number @dots{}}.)
7380 @kindex disable display
7381 @item disable display @var{dnums}@dots{}
7382 Disable the display of item numbers @var{dnums}. A disabled display
7383 item is not printed automatically, but is not forgotten. It may be
7384 enabled again later.
7386 @kindex enable display
7387 @item enable display @var{dnums}@dots{}
7388 Enable display of item numbers @var{dnums}. It becomes effective once
7389 again in auto display of its expression, until you specify otherwise.
7392 Display the current values of the expressions on the list, just as is
7393 done when your program stops.
7395 @kindex info display
7397 Print the list of expressions previously set up to display
7398 automatically, each one with its item number, but without showing the
7399 values. This includes disabled expressions, which are marked as such.
7400 It also includes expressions which would not be displayed right now
7401 because they refer to automatic variables not currently available.
7404 @cindex display disabled out of scope
7405 If a display expression refers to local variables, then it does not make
7406 sense outside the lexical context for which it was set up. Such an
7407 expression is disabled when execution enters a context where one of its
7408 variables is not defined. For example, if you give the command
7409 @code{display last_char} while inside a function with an argument
7410 @code{last_char}, @value{GDBN} displays this argument while your program
7411 continues to stop inside that function. When it stops elsewhere---where
7412 there is no variable @code{last_char}---the display is disabled
7413 automatically. The next time your program stops where @code{last_char}
7414 is meaningful, you can enable the display expression once again.
7416 @node Print Settings
7417 @section Print Settings
7419 @cindex format options
7420 @cindex print settings
7421 @value{GDBN} provides the following ways to control how arrays, structures,
7422 and symbols are printed.
7425 These settings are useful for debugging programs in any language:
7429 @item set print address
7430 @itemx set print address on
7431 @cindex print/don't print memory addresses
7432 @value{GDBN} prints memory addresses showing the location of stack
7433 traces, structure values, pointer values, breakpoints, and so forth,
7434 even when it also displays the contents of those addresses. The default
7435 is @code{on}. For example, this is what a stack frame display looks like with
7436 @code{set print address on}:
7441 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
7443 530 if (lquote != def_lquote)
7447 @item set print address off
7448 Do not print addresses when displaying their contents. For example,
7449 this is the same stack frame displayed with @code{set print address off}:
7453 (@value{GDBP}) set print addr off
7455 #0 set_quotes (lq="<<", rq=">>") at input.c:530
7456 530 if (lquote != def_lquote)
7460 You can use @samp{set print address off} to eliminate all machine
7461 dependent displays from the @value{GDBN} interface. For example, with
7462 @code{print address off}, you should get the same text for backtraces on
7463 all machines---whether or not they involve pointer arguments.
7466 @item show print address
7467 Show whether or not addresses are to be printed.
7470 When @value{GDBN} prints a symbolic address, it normally prints the
7471 closest earlier symbol plus an offset. If that symbol does not uniquely
7472 identify the address (for example, it is a name whose scope is a single
7473 source file), you may need to clarify. One way to do this is with
7474 @code{info line}, for example @samp{info line *0x4537}. Alternately,
7475 you can set @value{GDBN} to print the source file and line number when
7476 it prints a symbolic address:
7479 @item set print symbol-filename on
7480 @cindex source file and line of a symbol
7481 @cindex symbol, source file and line
7482 Tell @value{GDBN} to print the source file name and line number of a
7483 symbol in the symbolic form of an address.
7485 @item set print symbol-filename off
7486 Do not print source file name and line number of a symbol. This is the
7489 @item show print symbol-filename
7490 Show whether or not @value{GDBN} will print the source file name and
7491 line number of a symbol in the symbolic form of an address.
7494 Another situation where it is helpful to show symbol filenames and line
7495 numbers is when disassembling code; @value{GDBN} shows you the line
7496 number and source file that corresponds to each instruction.
7498 Also, you may wish to see the symbolic form only if the address being
7499 printed is reasonably close to the closest earlier symbol:
7502 @item set print max-symbolic-offset @var{max-offset}
7503 @cindex maximum value for offset of closest symbol
7504 Tell @value{GDBN} to only display the symbolic form of an address if the
7505 offset between the closest earlier symbol and the address is less than
7506 @var{max-offset}. The default is 0, which tells @value{GDBN}
7507 to always print the symbolic form of an address if any symbol precedes it.
7509 @item show print max-symbolic-offset
7510 Ask how large the maximum offset is that @value{GDBN} prints in a
7514 @cindex wild pointer, interpreting
7515 @cindex pointer, finding referent
7516 If you have a pointer and you are not sure where it points, try
7517 @samp{set print symbol-filename on}. Then you can determine the name
7518 and source file location of the variable where it points, using
7519 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
7520 For example, here @value{GDBN} shows that a variable @code{ptt} points
7521 at another variable @code{t}, defined in @file{hi2.c}:
7524 (@value{GDBP}) set print symbol-filename on
7525 (@value{GDBP}) p/a ptt
7526 $4 = 0xe008 <t in hi2.c>
7530 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
7531 does not show the symbol name and filename of the referent, even with
7532 the appropriate @code{set print} options turned on.
7535 Other settings control how different kinds of objects are printed:
7538 @item set print array
7539 @itemx set print array on
7540 @cindex pretty print arrays
7541 Pretty print arrays. This format is more convenient to read,
7542 but uses more space. The default is off.
7544 @item set print array off
7545 Return to compressed format for arrays.
7547 @item show print array
7548 Show whether compressed or pretty format is selected for displaying
7551 @cindex print array indexes
7552 @item set print array-indexes
7553 @itemx set print array-indexes on
7554 Print the index of each element when displaying arrays. May be more
7555 convenient to locate a given element in the array or quickly find the
7556 index of a given element in that printed array. The default is off.
7558 @item set print array-indexes off
7559 Stop printing element indexes when displaying arrays.
7561 @item show print array-indexes
7562 Show whether the index of each element is printed when displaying
7565 @item set print elements @var{number-of-elements}
7566 @cindex number of array elements to print
7567 @cindex limit on number of printed array elements
7568 Set a limit on how many elements of an array @value{GDBN} will print.
7569 If @value{GDBN} is printing a large array, it stops printing after it has
7570 printed the number of elements set by the @code{set print elements} command.
7571 This limit also applies to the display of strings.
7572 When @value{GDBN} starts, this limit is set to 200.
7573 Setting @var{number-of-elements} to zero means that the printing is unlimited.
7575 @item show print elements
7576 Display the number of elements of a large array that @value{GDBN} will print.
7577 If the number is 0, then the printing is unlimited.
7579 @item set print frame-arguments @var{value}
7580 @kindex set print frame-arguments
7581 @cindex printing frame argument values
7582 @cindex print all frame argument values
7583 @cindex print frame argument values for scalars only
7584 @cindex do not print frame argument values
7585 This command allows to control how the values of arguments are printed
7586 when the debugger prints a frame (@pxref{Frames}). The possible
7591 The values of all arguments are printed.
7594 Print the value of an argument only if it is a scalar. The value of more
7595 complex arguments such as arrays, structures, unions, etc, is replaced
7596 by @code{@dots{}}. This is the default. Here is an example where
7597 only scalar arguments are shown:
7600 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
7605 None of the argument values are printed. Instead, the value of each argument
7606 is replaced by @code{@dots{}}. In this case, the example above now becomes:
7609 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
7614 By default, only scalar arguments are printed. This command can be used
7615 to configure the debugger to print the value of all arguments, regardless
7616 of their type. However, it is often advantageous to not print the value
7617 of more complex parameters. For instance, it reduces the amount of
7618 information printed in each frame, making the backtrace more readable.
7619 Also, it improves performance when displaying Ada frames, because
7620 the computation of large arguments can sometimes be CPU-intensive,
7621 especially in large applications. Setting @code{print frame-arguments}
7622 to @code{scalars} (the default) or @code{none} avoids this computation,
7623 thus speeding up the display of each Ada frame.
7625 @item show print frame-arguments
7626 Show how the value of arguments should be displayed when printing a frame.
7628 @item set print repeats
7629 @cindex repeated array elements
7630 Set the threshold for suppressing display of repeated array
7631 elements. When the number of consecutive identical elements of an
7632 array exceeds the threshold, @value{GDBN} prints the string
7633 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
7634 identical repetitions, instead of displaying the identical elements
7635 themselves. Setting the threshold to zero will cause all elements to
7636 be individually printed. The default threshold is 10.
7638 @item show print repeats
7639 Display the current threshold for printing repeated identical
7642 @item set print null-stop
7643 @cindex @sc{null} elements in arrays
7644 Cause @value{GDBN} to stop printing the characters of an array when the first
7645 @sc{null} is encountered. This is useful when large arrays actually
7646 contain only short strings.
7649 @item show print null-stop
7650 Show whether @value{GDBN} stops printing an array on the first
7651 @sc{null} character.
7653 @item set print pretty on
7654 @cindex print structures in indented form
7655 @cindex indentation in structure display
7656 Cause @value{GDBN} to print structures in an indented format with one member
7657 per line, like this:
7672 @item set print pretty off
7673 Cause @value{GDBN} to print structures in a compact format, like this:
7677 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
7678 meat = 0x54 "Pork"@}
7683 This is the default format.
7685 @item show print pretty
7686 Show which format @value{GDBN} is using to print structures.
7688 @item set print sevenbit-strings on
7689 @cindex eight-bit characters in strings
7690 @cindex octal escapes in strings
7691 Print using only seven-bit characters; if this option is set,
7692 @value{GDBN} displays any eight-bit characters (in strings or
7693 character values) using the notation @code{\}@var{nnn}. This setting is
7694 best if you are working in English (@sc{ascii}) and you use the
7695 high-order bit of characters as a marker or ``meta'' bit.
7697 @item set print sevenbit-strings off
7698 Print full eight-bit characters. This allows the use of more
7699 international character sets, and is the default.
7701 @item show print sevenbit-strings
7702 Show whether or not @value{GDBN} is printing only seven-bit characters.
7704 @item set print union on
7705 @cindex unions in structures, printing
7706 Tell @value{GDBN} to print unions which are contained in structures
7707 and other unions. This is the default setting.
7709 @item set print union off
7710 Tell @value{GDBN} not to print unions which are contained in
7711 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
7714 @item show print union
7715 Ask @value{GDBN} whether or not it will print unions which are contained in
7716 structures and other unions.
7718 For example, given the declarations
7721 typedef enum @{Tree, Bug@} Species;
7722 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
7723 typedef enum @{Caterpillar, Cocoon, Butterfly@}
7734 struct thing foo = @{Tree, @{Acorn@}@};
7738 with @code{set print union on} in effect @samp{p foo} would print
7741 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
7745 and with @code{set print union off} in effect it would print
7748 $1 = @{it = Tree, form = @{...@}@}
7752 @code{set print union} affects programs written in C-like languages
7758 These settings are of interest when debugging C@t{++} programs:
7761 @cindex demangling C@t{++} names
7762 @item set print demangle
7763 @itemx set print demangle on
7764 Print C@t{++} names in their source form rather than in the encoded
7765 (``mangled'') form passed to the assembler and linker for type-safe
7766 linkage. The default is on.
7768 @item show print demangle
7769 Show whether C@t{++} names are printed in mangled or demangled form.
7771 @item set print asm-demangle
7772 @itemx set print asm-demangle on
7773 Print C@t{++} names in their source form rather than their mangled form, even
7774 in assembler code printouts such as instruction disassemblies.
7777 @item show print asm-demangle
7778 Show whether C@t{++} names in assembly listings are printed in mangled
7781 @cindex C@t{++} symbol decoding style
7782 @cindex symbol decoding style, C@t{++}
7783 @kindex set demangle-style
7784 @item set demangle-style @var{style}
7785 Choose among several encoding schemes used by different compilers to
7786 represent C@t{++} names. The choices for @var{style} are currently:
7790 Allow @value{GDBN} to choose a decoding style by inspecting your program.
7793 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
7794 This is the default.
7797 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
7800 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
7803 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
7804 @strong{Warning:} this setting alone is not sufficient to allow
7805 debugging @code{cfront}-generated executables. @value{GDBN} would
7806 require further enhancement to permit that.
7809 If you omit @var{style}, you will see a list of possible formats.
7811 @item show demangle-style
7812 Display the encoding style currently in use for decoding C@t{++} symbols.
7814 @item set print object
7815 @itemx set print object on
7816 @cindex derived type of an object, printing
7817 @cindex display derived types
7818 When displaying a pointer to an object, identify the @emph{actual}
7819 (derived) type of the object rather than the @emph{declared} type, using
7820 the virtual function table.
7822 @item set print object off
7823 Display only the declared type of objects, without reference to the
7824 virtual function table. This is the default setting.
7826 @item show print object
7827 Show whether actual, or declared, object types are displayed.
7829 @item set print static-members
7830 @itemx set print static-members on
7831 @cindex static members of C@t{++} objects
7832 Print static members when displaying a C@t{++} object. The default is on.
7834 @item set print static-members off
7835 Do not print static members when displaying a C@t{++} object.
7837 @item show print static-members
7838 Show whether C@t{++} static members are printed or not.
7840 @item set print pascal_static-members
7841 @itemx set print pascal_static-members on
7842 @cindex static members of Pascal objects
7843 @cindex Pascal objects, static members display
7844 Print static members when displaying a Pascal object. The default is on.
7846 @item set print pascal_static-members off
7847 Do not print static members when displaying a Pascal object.
7849 @item show print pascal_static-members
7850 Show whether Pascal static members are printed or not.
7852 @c These don't work with HP ANSI C++ yet.
7853 @item set print vtbl
7854 @itemx set print vtbl on
7855 @cindex pretty print C@t{++} virtual function tables
7856 @cindex virtual functions (C@t{++}) display
7857 @cindex VTBL display
7858 Pretty print C@t{++} virtual function tables. The default is off.
7859 (The @code{vtbl} commands do not work on programs compiled with the HP
7860 ANSI C@t{++} compiler (@code{aCC}).)
7862 @item set print vtbl off
7863 Do not pretty print C@t{++} virtual function tables.
7865 @item show print vtbl
7866 Show whether C@t{++} virtual function tables are pretty printed, or not.
7870 @section Value History
7872 @cindex value history
7873 @cindex history of values printed by @value{GDBN}
7874 Values printed by the @code{print} command are saved in the @value{GDBN}
7875 @dfn{value history}. This allows you to refer to them in other expressions.
7876 Values are kept until the symbol table is re-read or discarded
7877 (for example with the @code{file} or @code{symbol-file} commands).
7878 When the symbol table changes, the value history is discarded,
7879 since the values may contain pointers back to the types defined in the
7884 @cindex history number
7885 The values printed are given @dfn{history numbers} by which you can
7886 refer to them. These are successive integers starting with one.
7887 @code{print} shows you the history number assigned to a value by
7888 printing @samp{$@var{num} = } before the value; here @var{num} is the
7891 To refer to any previous value, use @samp{$} followed by the value's
7892 history number. The way @code{print} labels its output is designed to
7893 remind you of this. Just @code{$} refers to the most recent value in
7894 the history, and @code{$$} refers to the value before that.
7895 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
7896 is the value just prior to @code{$$}, @code{$$1} is equivalent to
7897 @code{$$}, and @code{$$0} is equivalent to @code{$}.
7899 For example, suppose you have just printed a pointer to a structure and
7900 want to see the contents of the structure. It suffices to type
7906 If you have a chain of structures where the component @code{next} points
7907 to the next one, you can print the contents of the next one with this:
7914 You can print successive links in the chain by repeating this
7915 command---which you can do by just typing @key{RET}.
7917 Note that the history records values, not expressions. If the value of
7918 @code{x} is 4 and you type these commands:
7926 then the value recorded in the value history by the @code{print} command
7927 remains 4 even though the value of @code{x} has changed.
7932 Print the last ten values in the value history, with their item numbers.
7933 This is like @samp{p@ $$9} repeated ten times, except that @code{show
7934 values} does not change the history.
7936 @item show values @var{n}
7937 Print ten history values centered on history item number @var{n}.
7940 Print ten history values just after the values last printed. If no more
7941 values are available, @code{show values +} produces no display.
7944 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
7945 same effect as @samp{show values +}.
7947 @node Convenience Vars
7948 @section Convenience Variables
7950 @cindex convenience variables
7951 @cindex user-defined variables
7952 @value{GDBN} provides @dfn{convenience variables} that you can use within
7953 @value{GDBN} to hold on to a value and refer to it later. These variables
7954 exist entirely within @value{GDBN}; they are not part of your program, and
7955 setting a convenience variable has no direct effect on further execution
7956 of your program. That is why you can use them freely.
7958 Convenience variables are prefixed with @samp{$}. Any name preceded by
7959 @samp{$} can be used for a convenience variable, unless it is one of
7960 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
7961 (Value history references, in contrast, are @emph{numbers} preceded
7962 by @samp{$}. @xref{Value History, ,Value History}.)
7964 You can save a value in a convenience variable with an assignment
7965 expression, just as you would set a variable in your program.
7969 set $foo = *object_ptr
7973 would save in @code{$foo} the value contained in the object pointed to by
7976 Using a convenience variable for the first time creates it, but its
7977 value is @code{void} until you assign a new value. You can alter the
7978 value with another assignment at any time.
7980 Convenience variables have no fixed types. You can assign a convenience
7981 variable any type of value, including structures and arrays, even if
7982 that variable already has a value of a different type. The convenience
7983 variable, when used as an expression, has the type of its current value.
7986 @kindex show convenience
7987 @cindex show all user variables
7988 @item show convenience
7989 Print a list of convenience variables used so far, and their values.
7990 Abbreviated @code{show conv}.
7992 @kindex init-if-undefined
7993 @cindex convenience variables, initializing
7994 @item init-if-undefined $@var{variable} = @var{expression}
7995 Set a convenience variable if it has not already been set. This is useful
7996 for user-defined commands that keep some state. It is similar, in concept,
7997 to using local static variables with initializers in C (except that
7998 convenience variables are global). It can also be used to allow users to
7999 override default values used in a command script.
8001 If the variable is already defined then the expression is not evaluated so
8002 any side-effects do not occur.
8005 One of the ways to use a convenience variable is as a counter to be
8006 incremented or a pointer to be advanced. For example, to print
8007 a field from successive elements of an array of structures:
8011 print bar[$i++]->contents
8015 Repeat that command by typing @key{RET}.
8017 Some convenience variables are created automatically by @value{GDBN} and given
8018 values likely to be useful.
8021 @vindex $_@r{, convenience variable}
8023 The variable @code{$_} is automatically set by the @code{x} command to
8024 the last address examined (@pxref{Memory, ,Examining Memory}). Other
8025 commands which provide a default address for @code{x} to examine also
8026 set @code{$_} to that address; these commands include @code{info line}
8027 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
8028 except when set by the @code{x} command, in which case it is a pointer
8029 to the type of @code{$__}.
8031 @vindex $__@r{, convenience variable}
8033 The variable @code{$__} is automatically set by the @code{x} command
8034 to the value found in the last address examined. Its type is chosen
8035 to match the format in which the data was printed.
8038 @vindex $_exitcode@r{, convenience variable}
8039 The variable @code{$_exitcode} is automatically set to the exit code when
8040 the program being debugged terminates.
8043 @vindex $_siginfo@r{, convenience variable}
8044 The variable @code{$_siginfo} contains extra signal information
8045 (@pxref{extra signal information}). Note that @code{$_siginfo}
8046 could be empty, if the application has not yet received any signals.
8047 For example, it will be empty before you execute the @code{run} command.
8050 On HP-UX systems, if you refer to a function or variable name that
8051 begins with a dollar sign, @value{GDBN} searches for a user or system
8052 name first, before it searches for a convenience variable.
8054 @cindex convenience functions
8055 @value{GDBN} also supplies some @dfn{convenience functions}. These
8056 have a syntax similar to convenience variables. A convenience
8057 function can be used in an expression just like an ordinary function;
8058 however, a convenience function is implemented internally to
8063 @kindex help function
8064 @cindex show all convenience functions
8065 Print a list of all convenience functions.
8072 You can refer to machine register contents, in expressions, as variables
8073 with names starting with @samp{$}. The names of registers are different
8074 for each machine; use @code{info registers} to see the names used on
8078 @kindex info registers
8079 @item info registers
8080 Print the names and values of all registers except floating-point
8081 and vector registers (in the selected stack frame).
8083 @kindex info all-registers
8084 @cindex floating point registers
8085 @item info all-registers
8086 Print the names and values of all registers, including floating-point
8087 and vector registers (in the selected stack frame).
8089 @item info registers @var{regname} @dots{}
8090 Print the @dfn{relativized} value of each specified register @var{regname}.
8091 As discussed in detail below, register values are normally relative to
8092 the selected stack frame. @var{regname} may be any register name valid on
8093 the machine you are using, with or without the initial @samp{$}.
8096 @cindex stack pointer register
8097 @cindex program counter register
8098 @cindex process status register
8099 @cindex frame pointer register
8100 @cindex standard registers
8101 @value{GDBN} has four ``standard'' register names that are available (in
8102 expressions) on most machines---whenever they do not conflict with an
8103 architecture's canonical mnemonics for registers. The register names
8104 @code{$pc} and @code{$sp} are used for the program counter register and
8105 the stack pointer. @code{$fp} is used for a register that contains a
8106 pointer to the current stack frame, and @code{$ps} is used for a
8107 register that contains the processor status. For example,
8108 you could print the program counter in hex with
8115 or print the instruction to be executed next with
8122 or add four to the stack pointer@footnote{This is a way of removing
8123 one word from the stack, on machines where stacks grow downward in
8124 memory (most machines, nowadays). This assumes that the innermost
8125 stack frame is selected; setting @code{$sp} is not allowed when other
8126 stack frames are selected. To pop entire frames off the stack,
8127 regardless of machine architecture, use @code{return};
8128 see @ref{Returning, ,Returning from a Function}.} with
8134 Whenever possible, these four standard register names are available on
8135 your machine even though the machine has different canonical mnemonics,
8136 so long as there is no conflict. The @code{info registers} command
8137 shows the canonical names. For example, on the SPARC, @code{info
8138 registers} displays the processor status register as @code{$psr} but you
8139 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
8140 is an alias for the @sc{eflags} register.
8142 @value{GDBN} always considers the contents of an ordinary register as an
8143 integer when the register is examined in this way. Some machines have
8144 special registers which can hold nothing but floating point; these
8145 registers are considered to have floating point values. There is no way
8146 to refer to the contents of an ordinary register as floating point value
8147 (although you can @emph{print} it as a floating point value with
8148 @samp{print/f $@var{regname}}).
8150 Some registers have distinct ``raw'' and ``virtual'' data formats. This
8151 means that the data format in which the register contents are saved by
8152 the operating system is not the same one that your program normally
8153 sees. For example, the registers of the 68881 floating point
8154 coprocessor are always saved in ``extended'' (raw) format, but all C
8155 programs expect to work with ``double'' (virtual) format. In such
8156 cases, @value{GDBN} normally works with the virtual format only (the format
8157 that makes sense for your program), but the @code{info registers} command
8158 prints the data in both formats.
8160 @cindex SSE registers (x86)
8161 @cindex MMX registers (x86)
8162 Some machines have special registers whose contents can be interpreted
8163 in several different ways. For example, modern x86-based machines
8164 have SSE and MMX registers that can hold several values packed
8165 together in several different formats. @value{GDBN} refers to such
8166 registers in @code{struct} notation:
8169 (@value{GDBP}) print $xmm1
8171 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
8172 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
8173 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
8174 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
8175 v4_int32 = @{0, 20657912, 11, 13@},
8176 v2_int64 = @{88725056443645952, 55834574859@},
8177 uint128 = 0x0000000d0000000b013b36f800000000
8182 To set values of such registers, you need to tell @value{GDBN} which
8183 view of the register you wish to change, as if you were assigning
8184 value to a @code{struct} member:
8187 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
8190 Normally, register values are relative to the selected stack frame
8191 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
8192 value that the register would contain if all stack frames farther in
8193 were exited and their saved registers restored. In order to see the
8194 true contents of hardware registers, you must select the innermost
8195 frame (with @samp{frame 0}).
8197 However, @value{GDBN} must deduce where registers are saved, from the machine
8198 code generated by your compiler. If some registers are not saved, or if
8199 @value{GDBN} is unable to locate the saved registers, the selected stack
8200 frame makes no difference.
8202 @node Floating Point Hardware
8203 @section Floating Point Hardware
8204 @cindex floating point
8206 Depending on the configuration, @value{GDBN} may be able to give
8207 you more information about the status of the floating point hardware.
8212 Display hardware-dependent information about the floating
8213 point unit. The exact contents and layout vary depending on the
8214 floating point chip. Currently, @samp{info float} is supported on
8215 the ARM and x86 machines.
8219 @section Vector Unit
8222 Depending on the configuration, @value{GDBN} may be able to give you
8223 more information about the status of the vector unit.
8228 Display information about the vector unit. The exact contents and
8229 layout vary depending on the hardware.
8232 @node OS Information
8233 @section Operating System Auxiliary Information
8234 @cindex OS information
8236 @value{GDBN} provides interfaces to useful OS facilities that can help
8237 you debug your program.
8239 @cindex @code{ptrace} system call
8240 @cindex @code{struct user} contents
8241 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
8242 machines), it interfaces with the inferior via the @code{ptrace}
8243 system call. The operating system creates a special sata structure,
8244 called @code{struct user}, for this interface. You can use the
8245 command @code{info udot} to display the contents of this data
8251 Display the contents of the @code{struct user} maintained by the OS
8252 kernel for the program being debugged. @value{GDBN} displays the
8253 contents of @code{struct user} as a list of hex numbers, similar to
8254 the @code{examine} command.
8257 @cindex auxiliary vector
8258 @cindex vector, auxiliary
8259 Some operating systems supply an @dfn{auxiliary vector} to programs at
8260 startup. This is akin to the arguments and environment that you
8261 specify for a program, but contains a system-dependent variety of
8262 binary values that tell system libraries important details about the
8263 hardware, operating system, and process. Each value's purpose is
8264 identified by an integer tag; the meanings are well-known but system-specific.
8265 Depending on the configuration and operating system facilities,
8266 @value{GDBN} may be able to show you this information. For remote
8267 targets, this functionality may further depend on the remote stub's
8268 support of the @samp{qXfer:auxv:read} packet, see
8269 @ref{qXfer auxiliary vector read}.
8274 Display the auxiliary vector of the inferior, which can be either a
8275 live process or a core dump file. @value{GDBN} prints each tag value
8276 numerically, and also shows names and text descriptions for recognized
8277 tags. Some values in the vector are numbers, some bit masks, and some
8278 pointers to strings or other data. @value{GDBN} displays each value in the
8279 most appropriate form for a recognized tag, and in hexadecimal for
8280 an unrecognized tag.
8283 On some targets, @value{GDBN} can access operating-system-specific information
8284 and display it to user, without interpretation. For remote targets,
8285 this functionality depends on the remote stub's support of the
8286 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
8289 @kindex info os processes
8290 @item info os processes
8291 Display the list of processes on the target. For each process,
8292 @value{GDBN} prints the process identifier, the name of the user, and
8293 the command corresponding to the process.
8296 @node Memory Region Attributes
8297 @section Memory Region Attributes
8298 @cindex memory region attributes
8300 @dfn{Memory region attributes} allow you to describe special handling
8301 required by regions of your target's memory. @value{GDBN} uses
8302 attributes to determine whether to allow certain types of memory
8303 accesses; whether to use specific width accesses; and whether to cache
8304 target memory. By default the description of memory regions is
8305 fetched from the target (if the current target supports this), but the
8306 user can override the fetched regions.
8308 Defined memory regions can be individually enabled and disabled. When a
8309 memory region is disabled, @value{GDBN} uses the default attributes when
8310 accessing memory in that region. Similarly, if no memory regions have
8311 been defined, @value{GDBN} uses the default attributes when accessing
8314 When a memory region is defined, it is given a number to identify it;
8315 to enable, disable, or remove a memory region, you specify that number.
8319 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
8320 Define a memory region bounded by @var{lower} and @var{upper} with
8321 attributes @var{attributes}@dots{}, and add it to the list of regions
8322 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
8323 case: it is treated as the target's maximum memory address.
8324 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
8327 Discard any user changes to the memory regions and use target-supplied
8328 regions, if available, or no regions if the target does not support.
8331 @item delete mem @var{nums}@dots{}
8332 Remove memory regions @var{nums}@dots{} from the list of regions
8333 monitored by @value{GDBN}.
8336 @item disable mem @var{nums}@dots{}
8337 Disable monitoring of memory regions @var{nums}@dots{}.
8338 A disabled memory region is not forgotten.
8339 It may be enabled again later.
8342 @item enable mem @var{nums}@dots{}
8343 Enable monitoring of memory regions @var{nums}@dots{}.
8347 Print a table of all defined memory regions, with the following columns
8351 @item Memory Region Number
8352 @item Enabled or Disabled.
8353 Enabled memory regions are marked with @samp{y}.
8354 Disabled memory regions are marked with @samp{n}.
8357 The address defining the inclusive lower bound of the memory region.
8360 The address defining the exclusive upper bound of the memory region.
8363 The list of attributes set for this memory region.
8368 @subsection Attributes
8370 @subsubsection Memory Access Mode
8371 The access mode attributes set whether @value{GDBN} may make read or
8372 write accesses to a memory region.
8374 While these attributes prevent @value{GDBN} from performing invalid
8375 memory accesses, they do nothing to prevent the target system, I/O DMA,
8376 etc.@: from accessing memory.
8380 Memory is read only.
8382 Memory is write only.
8384 Memory is read/write. This is the default.
8387 @subsubsection Memory Access Size
8388 The access size attribute tells @value{GDBN} to use specific sized
8389 accesses in the memory region. Often memory mapped device registers
8390 require specific sized accesses. If no access size attribute is
8391 specified, @value{GDBN} may use accesses of any size.
8395 Use 8 bit memory accesses.
8397 Use 16 bit memory accesses.
8399 Use 32 bit memory accesses.
8401 Use 64 bit memory accesses.
8404 @c @subsubsection Hardware/Software Breakpoints
8405 @c The hardware/software breakpoint attributes set whether @value{GDBN}
8406 @c will use hardware or software breakpoints for the internal breakpoints
8407 @c used by the step, next, finish, until, etc. commands.
8411 @c Always use hardware breakpoints
8412 @c @item swbreak (default)
8415 @subsubsection Data Cache
8416 The data cache attributes set whether @value{GDBN} will cache target
8417 memory. While this generally improves performance by reducing debug
8418 protocol overhead, it can lead to incorrect results because @value{GDBN}
8419 does not know about volatile variables or memory mapped device
8424 Enable @value{GDBN} to cache target memory.
8426 Disable @value{GDBN} from caching target memory. This is the default.
8429 @subsection Memory Access Checking
8430 @value{GDBN} can be instructed to refuse accesses to memory that is
8431 not explicitly described. This can be useful if accessing such
8432 regions has undesired effects for a specific target, or to provide
8433 better error checking. The following commands control this behaviour.
8436 @kindex set mem inaccessible-by-default
8437 @item set mem inaccessible-by-default [on|off]
8438 If @code{on} is specified, make @value{GDBN} treat memory not
8439 explicitly described by the memory ranges as non-existent and refuse accesses
8440 to such memory. The checks are only performed if there's at least one
8441 memory range defined. If @code{off} is specified, make @value{GDBN}
8442 treat the memory not explicitly described by the memory ranges as RAM.
8443 The default value is @code{on}.
8444 @kindex show mem inaccessible-by-default
8445 @item show mem inaccessible-by-default
8446 Show the current handling of accesses to unknown memory.
8450 @c @subsubsection Memory Write Verification
8451 @c The memory write verification attributes set whether @value{GDBN}
8452 @c will re-reads data after each write to verify the write was successful.
8456 @c @item noverify (default)
8459 @node Dump/Restore Files
8460 @section Copy Between Memory and a File
8461 @cindex dump/restore files
8462 @cindex append data to a file
8463 @cindex dump data to a file
8464 @cindex restore data from a file
8466 You can use the commands @code{dump}, @code{append}, and
8467 @code{restore} to copy data between target memory and a file. The
8468 @code{dump} and @code{append} commands write data to a file, and the
8469 @code{restore} command reads data from a file back into the inferior's
8470 memory. Files may be in binary, Motorola S-record, Intel hex, or
8471 Tektronix Hex format; however, @value{GDBN} can only append to binary
8477 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8478 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
8479 Dump the contents of memory from @var{start_addr} to @var{end_addr},
8480 or the value of @var{expr}, to @var{filename} in the given format.
8482 The @var{format} parameter may be any one of:
8489 Motorola S-record format.
8491 Tektronix Hex format.
8494 @value{GDBN} uses the same definitions of these formats as the
8495 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
8496 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
8500 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8501 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
8502 Append the contents of memory from @var{start_addr} to @var{end_addr},
8503 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
8504 (@value{GDBN} can only append data to files in raw binary form.)
8507 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
8508 Restore the contents of file @var{filename} into memory. The
8509 @code{restore} command can automatically recognize any known @sc{bfd}
8510 file format, except for raw binary. To restore a raw binary file you
8511 must specify the optional keyword @code{binary} after the filename.
8513 If @var{bias} is non-zero, its value will be added to the addresses
8514 contained in the file. Binary files always start at address zero, so
8515 they will be restored at address @var{bias}. Other bfd files have
8516 a built-in location; they will be restored at offset @var{bias}
8519 If @var{start} and/or @var{end} are non-zero, then only data between
8520 file offset @var{start} and file offset @var{end} will be restored.
8521 These offsets are relative to the addresses in the file, before
8522 the @var{bias} argument is applied.
8526 @node Core File Generation
8527 @section How to Produce a Core File from Your Program
8528 @cindex dump core from inferior
8530 A @dfn{core file} or @dfn{core dump} is a file that records the memory
8531 image of a running process and its process status (register values
8532 etc.). Its primary use is post-mortem debugging of a program that
8533 crashed while it ran outside a debugger. A program that crashes
8534 automatically produces a core file, unless this feature is disabled by
8535 the user. @xref{Files}, for information on invoking @value{GDBN} in
8536 the post-mortem debugging mode.
8538 Occasionally, you may wish to produce a core file of the program you
8539 are debugging in order to preserve a snapshot of its state.
8540 @value{GDBN} has a special command for that.
8544 @kindex generate-core-file
8545 @item generate-core-file [@var{file}]
8546 @itemx gcore [@var{file}]
8547 Produce a core dump of the inferior process. The optional argument
8548 @var{file} specifies the file name where to put the core dump. If not
8549 specified, the file name defaults to @file{core.@var{pid}}, where
8550 @var{pid} is the inferior process ID.
8552 Note that this command is implemented only for some systems (as of
8553 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
8556 @node Character Sets
8557 @section Character Sets
8558 @cindex character sets
8560 @cindex translating between character sets
8561 @cindex host character set
8562 @cindex target character set
8564 If the program you are debugging uses a different character set to
8565 represent characters and strings than the one @value{GDBN} uses itself,
8566 @value{GDBN} can automatically translate between the character sets for
8567 you. The character set @value{GDBN} uses we call the @dfn{host
8568 character set}; the one the inferior program uses we call the
8569 @dfn{target character set}.
8571 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
8572 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
8573 remote protocol (@pxref{Remote Debugging}) to debug a program
8574 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
8575 then the host character set is Latin-1, and the target character set is
8576 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
8577 target-charset EBCDIC-US}, then @value{GDBN} translates between
8578 @sc{ebcdic} and Latin 1 as you print character or string values, or use
8579 character and string literals in expressions.
8581 @value{GDBN} has no way to automatically recognize which character set
8582 the inferior program uses; you must tell it, using the @code{set
8583 target-charset} command, described below.
8585 Here are the commands for controlling @value{GDBN}'s character set
8589 @item set target-charset @var{charset}
8590 @kindex set target-charset
8591 Set the current target character set to @var{charset}. To display the
8592 list of supported target character sets, type
8593 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
8595 @item set host-charset @var{charset}
8596 @kindex set host-charset
8597 Set the current host character set to @var{charset}.
8599 By default, @value{GDBN} uses a host character set appropriate to the
8600 system it is running on; you can override that default using the
8601 @code{set host-charset} command. On some systems, @value{GDBN} cannot
8602 automatically determine the appropriate host character set. In this
8603 case, @value{GDBN} uses @samp{UTF-8}.
8605 @value{GDBN} can only use certain character sets as its host character
8606 set. If you type @kbd{@w{set target-charset @key{TAB}@key{TAB}}},
8607 @value{GDBN} will list the host character sets it supports.
8609 @item set charset @var{charset}
8611 Set the current host and target character sets to @var{charset}. As
8612 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
8613 @value{GDBN} will list the names of the character sets that can be used
8614 for both host and target.
8617 @kindex show charset
8618 Show the names of the current host and target character sets.
8620 @item show host-charset
8621 @kindex show host-charset
8622 Show the name of the current host character set.
8624 @item show target-charset
8625 @kindex show target-charset
8626 Show the name of the current target character set.
8628 @item set target-wide-charset @var{charset}
8629 @kindex set target-wide-charset
8630 Set the current target's wide character set to @var{charset}. This is
8631 the character set used by the target's @code{wchar_t} type. To
8632 display the list of supported wide character sets, type
8633 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
8635 @item show target-wide-charset
8636 @kindex show target-wide-charset
8637 Show the name of the current target's wide character set.
8640 Here is an example of @value{GDBN}'s character set support in action.
8641 Assume that the following source code has been placed in the file
8642 @file{charset-test.c}:
8648 = @{72, 101, 108, 108, 111, 44, 32, 119,
8649 111, 114, 108, 100, 33, 10, 0@};
8650 char ibm1047_hello[]
8651 = @{200, 133, 147, 147, 150, 107, 64, 166,
8652 150, 153, 147, 132, 90, 37, 0@};
8656 printf ("Hello, world!\n");
8660 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
8661 containing the string @samp{Hello, world!} followed by a newline,
8662 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
8664 We compile the program, and invoke the debugger on it:
8667 $ gcc -g charset-test.c -o charset-test
8668 $ gdb -nw charset-test
8669 GNU gdb 2001-12-19-cvs
8670 Copyright 2001 Free Software Foundation, Inc.
8675 We can use the @code{show charset} command to see what character sets
8676 @value{GDBN} is currently using to interpret and display characters and
8680 (@value{GDBP}) show charset
8681 The current host and target character set is `ISO-8859-1'.
8685 For the sake of printing this manual, let's use @sc{ascii} as our
8686 initial character set:
8688 (@value{GDBP}) set charset ASCII
8689 (@value{GDBP}) show charset
8690 The current host and target character set is `ASCII'.
8694 Let's assume that @sc{ascii} is indeed the correct character set for our
8695 host system --- in other words, let's assume that if @value{GDBN} prints
8696 characters using the @sc{ascii} character set, our terminal will display
8697 them properly. Since our current target character set is also
8698 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
8701 (@value{GDBP}) print ascii_hello
8702 $1 = 0x401698 "Hello, world!\n"
8703 (@value{GDBP}) print ascii_hello[0]
8708 @value{GDBN} uses the target character set for character and string
8709 literals you use in expressions:
8712 (@value{GDBP}) print '+'
8717 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
8720 @value{GDBN} relies on the user to tell it which character set the
8721 target program uses. If we print @code{ibm1047_hello} while our target
8722 character set is still @sc{ascii}, we get jibberish:
8725 (@value{GDBP}) print ibm1047_hello
8726 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
8727 (@value{GDBP}) print ibm1047_hello[0]
8732 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
8733 @value{GDBN} tells us the character sets it supports:
8736 (@value{GDBP}) set target-charset
8737 ASCII EBCDIC-US IBM1047 ISO-8859-1
8738 (@value{GDBP}) set target-charset
8741 We can select @sc{ibm1047} as our target character set, and examine the
8742 program's strings again. Now the @sc{ascii} string is wrong, but
8743 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
8744 target character set, @sc{ibm1047}, to the host character set,
8745 @sc{ascii}, and they display correctly:
8748 (@value{GDBP}) set target-charset IBM1047
8749 (@value{GDBP}) show charset
8750 The current host character set is `ASCII'.
8751 The current target character set is `IBM1047'.
8752 (@value{GDBP}) print ascii_hello
8753 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
8754 (@value{GDBP}) print ascii_hello[0]
8756 (@value{GDBP}) print ibm1047_hello
8757 $8 = 0x4016a8 "Hello, world!\n"
8758 (@value{GDBP}) print ibm1047_hello[0]
8763 As above, @value{GDBN} uses the target character set for character and
8764 string literals you use in expressions:
8767 (@value{GDBP}) print '+'
8772 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
8775 @node Caching Remote Data
8776 @section Caching Data of Remote Targets
8777 @cindex caching data of remote targets
8779 @value{GDBN} caches data exchanged between the debugger and a
8780 remote target (@pxref{Remote Debugging}). Such caching generally improves
8781 performance, because it reduces the overhead of the remote protocol by
8782 bundling memory reads and writes into large chunks. Unfortunately, simply
8783 caching everything would lead to incorrect results, since @value{GDBN}
8784 does not necessarily know anything about volatile values, memory-mapped I/O
8785 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
8786 memory can be changed @emph{while} a gdb command is executing.
8787 Therefore, by default, @value{GDBN} only caches data
8788 known to be on the stack@footnote{In non-stop mode, it is moderately
8789 rare for a running thread to modify the stack of a stopped thread
8790 in a way that would interfere with a backtrace, and caching of
8791 stack reads provides a significant speed up of remote backtraces.}.
8792 Other regions of memory can be explicitly marked as
8793 cacheable; see @pxref{Memory Region Attributes}.
8796 @kindex set remotecache
8797 @item set remotecache on
8798 @itemx set remotecache off
8799 This option no longer does anything; it exists for compatibility
8802 @kindex show remotecache
8803 @item show remotecache
8804 Show the current state of the obsolete remotecache flag.
8806 @kindex set stack-cache
8807 @item set stack-cache on
8808 @itemx set stack-cache off
8809 Enable or disable caching of stack accesses. When @code{ON}, use
8810 caching. By default, this option is @code{ON}.
8812 @kindex show stack-cache
8813 @item show stack-cache
8814 Show the current state of data caching for memory accesses.
8817 @item info dcache @r{[}line@r{]}
8818 Print the information about the data cache performance. The
8819 information displayed includes the dcache width and depth, and for
8820 each cache line, its number, address, and how many times it was
8821 referenced. This command is useful for debugging the data cache
8824 If a line number is specified, the contents of that line will be
8828 @node Searching Memory
8829 @section Search Memory
8830 @cindex searching memory
8832 Memory can be searched for a particular sequence of bytes with the
8833 @code{find} command.
8837 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8838 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8839 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
8840 etc. The search begins at address @var{start_addr} and continues for either
8841 @var{len} bytes or through to @var{end_addr} inclusive.
8844 @var{s} and @var{n} are optional parameters.
8845 They may be specified in either order, apart or together.
8848 @item @var{s}, search query size
8849 The size of each search query value.
8855 halfwords (two bytes)
8859 giant words (eight bytes)
8862 All values are interpreted in the current language.
8863 This means, for example, that if the current source language is C/C@t{++}
8864 then searching for the string ``hello'' includes the trailing '\0'.
8866 If the value size is not specified, it is taken from the
8867 value's type in the current language.
8868 This is useful when one wants to specify the search
8869 pattern as a mixture of types.
8870 Note that this means, for example, that in the case of C-like languages
8871 a search for an untyped 0x42 will search for @samp{(int) 0x42}
8872 which is typically four bytes.
8874 @item @var{n}, maximum number of finds
8875 The maximum number of matches to print. The default is to print all finds.
8878 You can use strings as search values. Quote them with double-quotes
8880 The string value is copied into the search pattern byte by byte,
8881 regardless of the endianness of the target and the size specification.
8883 The address of each match found is printed as well as a count of the
8884 number of matches found.
8886 The address of the last value found is stored in convenience variable
8888 A count of the number of matches is stored in @samp{$numfound}.
8890 For example, if stopped at the @code{printf} in this function:
8896 static char hello[] = "hello-hello";
8897 static struct @{ char c; short s; int i; @}
8898 __attribute__ ((packed)) mixed
8899 = @{ 'c', 0x1234, 0x87654321 @};
8900 printf ("%s\n", hello);
8905 you get during debugging:
8908 (gdb) find &hello[0], +sizeof(hello), "hello"
8909 0x804956d <hello.1620+6>
8911 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
8912 0x8049567 <hello.1620>
8913 0x804956d <hello.1620+6>
8915 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
8916 0x8049567 <hello.1620>
8918 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
8919 0x8049560 <mixed.1625>
8921 (gdb) print $numfound
8924 $2 = (void *) 0x8049560
8927 @node Optimized Code
8928 @chapter Debugging Optimized Code
8929 @cindex optimized code, debugging
8930 @cindex debugging optimized code
8932 Almost all compilers support optimization. With optimization
8933 disabled, the compiler generates assembly code that corresponds
8934 directly to your source code, in a simplistic way. As the compiler
8935 applies more powerful optimizations, the generated assembly code
8936 diverges from your original source code. With help from debugging
8937 information generated by the compiler, @value{GDBN} can map from
8938 the running program back to constructs from your original source.
8940 @value{GDBN} is more accurate with optimization disabled. If you
8941 can recompile without optimization, it is easier to follow the
8942 progress of your program during debugging. But, there are many cases
8943 where you may need to debug an optimized version.
8945 When you debug a program compiled with @samp{-g -O}, remember that the
8946 optimizer has rearranged your code; the debugger shows you what is
8947 really there. Do not be too surprised when the execution path does not
8948 exactly match your source file! An extreme example: if you define a
8949 variable, but never use it, @value{GDBN} never sees that
8950 variable---because the compiler optimizes it out of existence.
8952 Some things do not work as well with @samp{-g -O} as with just
8953 @samp{-g}, particularly on machines with instruction scheduling. If in
8954 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
8955 please report it to us as a bug (including a test case!).
8956 @xref{Variables}, for more information about debugging optimized code.
8959 * Inline Functions:: How @value{GDBN} presents inlining
8962 @node Inline Functions
8963 @section Inline Functions
8964 @cindex inline functions, debugging
8966 @dfn{Inlining} is an optimization that inserts a copy of the function
8967 body directly at each call site, instead of jumping to a shared
8968 routine. @value{GDBN} displays inlined functions just like
8969 non-inlined functions. They appear in backtraces. You can view their
8970 arguments and local variables, step into them with @code{step}, skip
8971 them with @code{next}, and escape from them with @code{finish}.
8972 You can check whether a function was inlined by using the
8973 @code{info frame} command.
8975 For @value{GDBN} to support inlined functions, the compiler must
8976 record information about inlining in the debug information ---
8977 @value{NGCC} using the @sc{dwarf 2} format does this, and several
8978 other compilers do also. @value{GDBN} only supports inlined functions
8979 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
8980 do not emit two required attributes (@samp{DW_AT_call_file} and
8981 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
8982 function calls with earlier versions of @value{NGCC}. It instead
8983 displays the arguments and local variables of inlined functions as
8984 local variables in the caller.
8986 The body of an inlined function is directly included at its call site;
8987 unlike a non-inlined function, there are no instructions devoted to
8988 the call. @value{GDBN} still pretends that the call site and the
8989 start of the inlined function are different instructions. Stepping to
8990 the call site shows the call site, and then stepping again shows
8991 the first line of the inlined function, even though no additional
8992 instructions are executed.
8994 This makes source-level debugging much clearer; you can see both the
8995 context of the call and then the effect of the call. Only stepping by
8996 a single instruction using @code{stepi} or @code{nexti} does not do
8997 this; single instruction steps always show the inlined body.
8999 There are some ways that @value{GDBN} does not pretend that inlined
9000 function calls are the same as normal calls:
9004 You cannot set breakpoints on inlined functions. @value{GDBN}
9005 either reports that there is no symbol with that name, or else sets the
9006 breakpoint only on non-inlined copies of the function. This limitation
9007 will be removed in a future version of @value{GDBN}; until then,
9008 set a breakpoint by line number on the first line of the inlined
9012 Setting breakpoints at the call site of an inlined function may not
9013 work, because the call site does not contain any code. @value{GDBN}
9014 may incorrectly move the breakpoint to the next line of the enclosing
9015 function, after the call. This limitation will be removed in a future
9016 version of @value{GDBN}; until then, set a breakpoint on an earlier line
9017 or inside the inlined function instead.
9020 @value{GDBN} cannot locate the return value of inlined calls after
9021 using the @code{finish} command. This is a limitation of compiler-generated
9022 debugging information; after @code{finish}, you can step to the next line
9023 and print a variable where your program stored the return value.
9029 @chapter C Preprocessor Macros
9031 Some languages, such as C and C@t{++}, provide a way to define and invoke
9032 ``preprocessor macros'' which expand into strings of tokens.
9033 @value{GDBN} can evaluate expressions containing macro invocations, show
9034 the result of macro expansion, and show a macro's definition, including
9035 where it was defined.
9037 You may need to compile your program specially to provide @value{GDBN}
9038 with information about preprocessor macros. Most compilers do not
9039 include macros in their debugging information, even when you compile
9040 with the @option{-g} flag. @xref{Compilation}.
9042 A program may define a macro at one point, remove that definition later,
9043 and then provide a different definition after that. Thus, at different
9044 points in the program, a macro may have different definitions, or have
9045 no definition at all. If there is a current stack frame, @value{GDBN}
9046 uses the macros in scope at that frame's source code line. Otherwise,
9047 @value{GDBN} uses the macros in scope at the current listing location;
9050 Whenever @value{GDBN} evaluates an expression, it always expands any
9051 macro invocations present in the expression. @value{GDBN} also provides
9052 the following commands for working with macros explicitly.
9056 @kindex macro expand
9057 @cindex macro expansion, showing the results of preprocessor
9058 @cindex preprocessor macro expansion, showing the results of
9059 @cindex expanding preprocessor macros
9060 @item macro expand @var{expression}
9061 @itemx macro exp @var{expression}
9062 Show the results of expanding all preprocessor macro invocations in
9063 @var{expression}. Since @value{GDBN} simply expands macros, but does
9064 not parse the result, @var{expression} need not be a valid expression;
9065 it can be any string of tokens.
9068 @item macro expand-once @var{expression}
9069 @itemx macro exp1 @var{expression}
9070 @cindex expand macro once
9071 @i{(This command is not yet implemented.)} Show the results of
9072 expanding those preprocessor macro invocations that appear explicitly in
9073 @var{expression}. Macro invocations appearing in that expansion are
9074 left unchanged. This command allows you to see the effect of a
9075 particular macro more clearly, without being confused by further
9076 expansions. Since @value{GDBN} simply expands macros, but does not
9077 parse the result, @var{expression} need not be a valid expression; it
9078 can be any string of tokens.
9081 @cindex macro definition, showing
9082 @cindex definition, showing a macro's
9083 @item info macro @var{macro}
9084 Show the definition of the macro named @var{macro}, and describe the
9085 source location or compiler command-line where that definition was established.
9087 @kindex macro define
9088 @cindex user-defined macros
9089 @cindex defining macros interactively
9090 @cindex macros, user-defined
9091 @item macro define @var{macro} @var{replacement-list}
9092 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
9093 Introduce a definition for a preprocessor macro named @var{macro},
9094 invocations of which are replaced by the tokens given in
9095 @var{replacement-list}. The first form of this command defines an
9096 ``object-like'' macro, which takes no arguments; the second form
9097 defines a ``function-like'' macro, which takes the arguments given in
9100 A definition introduced by this command is in scope in every
9101 expression evaluated in @value{GDBN}, until it is removed with the
9102 @code{macro undef} command, described below. The definition overrides
9103 all definitions for @var{macro} present in the program being debugged,
9104 as well as any previous user-supplied definition.
9107 @item macro undef @var{macro}
9108 Remove any user-supplied definition for the macro named @var{macro}.
9109 This command only affects definitions provided with the @code{macro
9110 define} command, described above; it cannot remove definitions present
9111 in the program being debugged.
9115 List all the macros defined using the @code{macro define} command.
9118 @cindex macros, example of debugging with
9119 Here is a transcript showing the above commands in action. First, we
9120 show our source files:
9128 #define ADD(x) (M + x)
9133 printf ("Hello, world!\n");
9135 printf ("We're so creative.\n");
9137 printf ("Goodbye, world!\n");
9144 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
9145 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
9146 compiler includes information about preprocessor macros in the debugging
9150 $ gcc -gdwarf-2 -g3 sample.c -o sample
9154 Now, we start @value{GDBN} on our sample program:
9158 GNU gdb 2002-05-06-cvs
9159 Copyright 2002 Free Software Foundation, Inc.
9160 GDB is free software, @dots{}
9164 We can expand macros and examine their definitions, even when the
9165 program is not running. @value{GDBN} uses the current listing position
9166 to decide which macro definitions are in scope:
9169 (@value{GDBP}) list main
9172 5 #define ADD(x) (M + x)
9177 10 printf ("Hello, world!\n");
9179 12 printf ("We're so creative.\n");
9180 (@value{GDBP}) info macro ADD
9181 Defined at /home/jimb/gdb/macros/play/sample.c:5
9182 #define ADD(x) (M + x)
9183 (@value{GDBP}) info macro Q
9184 Defined at /home/jimb/gdb/macros/play/sample.h:1
9185 included at /home/jimb/gdb/macros/play/sample.c:2
9187 (@value{GDBP}) macro expand ADD(1)
9188 expands to: (42 + 1)
9189 (@value{GDBP}) macro expand-once ADD(1)
9190 expands to: once (M + 1)
9194 In the example above, note that @code{macro expand-once} expands only
9195 the macro invocation explicit in the original text --- the invocation of
9196 @code{ADD} --- but does not expand the invocation of the macro @code{M},
9197 which was introduced by @code{ADD}.
9199 Once the program is running, @value{GDBN} uses the macro definitions in
9200 force at the source line of the current stack frame:
9203 (@value{GDBP}) break main
9204 Breakpoint 1 at 0x8048370: file sample.c, line 10.
9206 Starting program: /home/jimb/gdb/macros/play/sample
9208 Breakpoint 1, main () at sample.c:10
9209 10 printf ("Hello, world!\n");
9213 At line 10, the definition of the macro @code{N} at line 9 is in force:
9216 (@value{GDBP}) info macro N
9217 Defined at /home/jimb/gdb/macros/play/sample.c:9
9219 (@value{GDBP}) macro expand N Q M
9221 (@value{GDBP}) print N Q M
9226 As we step over directives that remove @code{N}'s definition, and then
9227 give it a new definition, @value{GDBN} finds the definition (or lack
9228 thereof) in force at each point:
9233 12 printf ("We're so creative.\n");
9234 (@value{GDBP}) info macro N
9235 The symbol `N' has no definition as a C/C++ preprocessor macro
9236 at /home/jimb/gdb/macros/play/sample.c:12
9239 14 printf ("Goodbye, world!\n");
9240 (@value{GDBP}) info macro N
9241 Defined at /home/jimb/gdb/macros/play/sample.c:13
9243 (@value{GDBP}) macro expand N Q M
9244 expands to: 1729 < 42
9245 (@value{GDBP}) print N Q M
9250 In addition to source files, macros can be defined on the compilation command
9251 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
9252 such a way, @value{GDBN} displays the location of their definition as line zero
9253 of the source file submitted to the compiler.
9256 (@value{GDBP}) info macro __STDC__
9257 Defined at /home/jimb/gdb/macros/play/sample.c:0
9264 @chapter Tracepoints
9265 @c This chapter is based on the documentation written by Michael
9266 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
9269 In some applications, it is not feasible for the debugger to interrupt
9270 the program's execution long enough for the developer to learn
9271 anything helpful about its behavior. If the program's correctness
9272 depends on its real-time behavior, delays introduced by a debugger
9273 might cause the program to change its behavior drastically, or perhaps
9274 fail, even when the code itself is correct. It is useful to be able
9275 to observe the program's behavior without interrupting it.
9277 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
9278 specify locations in the program, called @dfn{tracepoints}, and
9279 arbitrary expressions to evaluate when those tracepoints are reached.
9280 Later, using the @code{tfind} command, you can examine the values
9281 those expressions had when the program hit the tracepoints. The
9282 expressions may also denote objects in memory---structures or arrays,
9283 for example---whose values @value{GDBN} should record; while visiting
9284 a particular tracepoint, you may inspect those objects as if they were
9285 in memory at that moment. However, because @value{GDBN} records these
9286 values without interacting with you, it can do so quickly and
9287 unobtrusively, hopefully not disturbing the program's behavior.
9289 The tracepoint facility is currently available only for remote
9290 targets. @xref{Targets}. In addition, your remote target must know
9291 how to collect trace data. This functionality is implemented in the
9292 remote stub; however, none of the stubs distributed with @value{GDBN}
9293 support tracepoints as of this writing. The format of the remote
9294 packets used to implement tracepoints are described in @ref{Tracepoint
9297 It is also possible to get trace data from a file, in a manner reminiscent
9298 of corefiles; you specify the filename, and use @code{tfind} to search
9299 through the file. @xref{Trace Files}, for more details.
9301 This chapter describes the tracepoint commands and features.
9305 * Analyze Collected Data::
9306 * Tracepoint Variables::
9310 @node Set Tracepoints
9311 @section Commands to Set Tracepoints
9313 Before running such a @dfn{trace experiment}, an arbitrary number of
9314 tracepoints can be set. A tracepoint is actually a special type of
9315 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
9316 standard breakpoint commands. For instance, as with breakpoints,
9317 tracepoint numbers are successive integers starting from one, and many
9318 of the commands associated with tracepoints take the tracepoint number
9319 as their argument, to identify which tracepoint to work on.
9321 For each tracepoint, you can specify, in advance, some arbitrary set
9322 of data that you want the target to collect in the trace buffer when
9323 it hits that tracepoint. The collected data can include registers,
9324 local variables, or global data. Later, you can use @value{GDBN}
9325 commands to examine the values these data had at the time the
9328 Tracepoints do not support every breakpoint feature. Conditional
9329 expressions and ignore counts on tracepoints have no effect, and
9330 tracepoints cannot run @value{GDBN} commands when they are
9331 hit. Tracepoints may not be thread-specific either.
9333 @cindex fast tracepoints
9334 Some targets may support @dfn{fast tracepoints}, which are inserted in
9335 a different way (such as with a jump instead of a trap), that is
9336 faster but possibly restricted in where they may be installed.
9338 This section describes commands to set tracepoints and associated
9339 conditions and actions.
9342 * Create and Delete Tracepoints::
9343 * Enable and Disable Tracepoints::
9344 * Tracepoint Passcounts::
9345 * Tracepoint Conditions::
9346 * Trace State Variables::
9347 * Tracepoint Actions::
9348 * Listing Tracepoints::
9349 * Starting and Stopping Trace Experiments::
9352 @node Create and Delete Tracepoints
9353 @subsection Create and Delete Tracepoints
9356 @cindex set tracepoint
9358 @item trace @var{location}
9359 The @code{trace} command is very similar to the @code{break} command.
9360 Its argument @var{location} can be a source line, a function name, or
9361 an address in the target program. @xref{Specify Location}. The
9362 @code{trace} command defines a tracepoint, which is a point in the
9363 target program where the debugger will briefly stop, collect some
9364 data, and then allow the program to continue. Setting a tracepoint or
9365 changing its actions doesn't take effect until the next @code{tstart}
9366 command, and once a trace experiment is running, further changes will
9367 not have any effect until the next trace experiment starts.
9369 Here are some examples of using the @code{trace} command:
9372 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
9374 (@value{GDBP}) @b{trace +2} // 2 lines forward
9376 (@value{GDBP}) @b{trace my_function} // first source line of function
9378 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
9380 (@value{GDBP}) @b{trace *0x2117c4} // an address
9384 You can abbreviate @code{trace} as @code{tr}.
9386 @item trace @var{location} if @var{cond}
9387 Set a tracepoint with condition @var{cond}; evaluate the expression
9388 @var{cond} each time the tracepoint is reached, and collect data only
9389 if the value is nonzero---that is, if @var{cond} evaluates as true.
9390 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
9391 information on tracepoint conditions.
9393 @item ftrace @var{location} [ if @var{cond} ]
9394 @cindex set fast tracepoint
9396 The @code{ftrace} command sets a fast tracepoint. For targets that
9397 support them, fast tracepoints will use a more efficient but possibly
9398 less general technique to trigger data collection, such as a jump
9399 instruction instead of a trap, or some sort of hardware support. It
9400 may not be possible to create a fast tracepoint at the desired
9401 location, in which case the command will exit with an explanatory
9404 @value{GDBN} handles arguments to @code{ftrace} exactly as for
9408 @cindex last tracepoint number
9409 @cindex recent tracepoint number
9410 @cindex tracepoint number
9411 The convenience variable @code{$tpnum} records the tracepoint number
9412 of the most recently set tracepoint.
9414 @kindex delete tracepoint
9415 @cindex tracepoint deletion
9416 @item delete tracepoint @r{[}@var{num}@r{]}
9417 Permanently delete one or more tracepoints. With no argument, the
9418 default is to delete all tracepoints. Note that the regular
9419 @code{delete} command can remove tracepoints also.
9424 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
9426 (@value{GDBP}) @b{delete trace} // remove all tracepoints
9430 You can abbreviate this command as @code{del tr}.
9433 @node Enable and Disable Tracepoints
9434 @subsection Enable and Disable Tracepoints
9436 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
9439 @kindex disable tracepoint
9440 @item disable tracepoint @r{[}@var{num}@r{]}
9441 Disable tracepoint @var{num}, or all tracepoints if no argument
9442 @var{num} is given. A disabled tracepoint will have no effect during
9443 the next trace experiment, but it is not forgotten. You can re-enable
9444 a disabled tracepoint using the @code{enable tracepoint} command.
9446 @kindex enable tracepoint
9447 @item enable tracepoint @r{[}@var{num}@r{]}
9448 Enable tracepoint @var{num}, or all tracepoints. The enabled
9449 tracepoints will become effective the next time a trace experiment is
9453 @node Tracepoint Passcounts
9454 @subsection Tracepoint Passcounts
9458 @cindex tracepoint pass count
9459 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
9460 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
9461 automatically stop a trace experiment. If a tracepoint's passcount is
9462 @var{n}, then the trace experiment will be automatically stopped on
9463 the @var{n}'th time that tracepoint is hit. If the tracepoint number
9464 @var{num} is not specified, the @code{passcount} command sets the
9465 passcount of the most recently defined tracepoint. If no passcount is
9466 given, the trace experiment will run until stopped explicitly by the
9472 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
9473 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
9475 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
9476 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
9477 (@value{GDBP}) @b{trace foo}
9478 (@value{GDBP}) @b{pass 3}
9479 (@value{GDBP}) @b{trace bar}
9480 (@value{GDBP}) @b{pass 2}
9481 (@value{GDBP}) @b{trace baz}
9482 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
9483 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
9484 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
9485 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
9489 @node Tracepoint Conditions
9490 @subsection Tracepoint Conditions
9491 @cindex conditional tracepoints
9492 @cindex tracepoint conditions
9494 The simplest sort of tracepoint collects data every time your program
9495 reaches a specified place. You can also specify a @dfn{condition} for
9496 a tracepoint. A condition is just a Boolean expression in your
9497 programming language (@pxref{Expressions, ,Expressions}). A
9498 tracepoint with a condition evaluates the expression each time your
9499 program reaches it, and data collection happens only if the condition
9502 Tracepoint conditions can be specified when a tracepoint is set, by
9503 using @samp{if} in the arguments to the @code{trace} command.
9504 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
9505 also be set or changed at any time with the @code{condition} command,
9506 just as with breakpoints.
9508 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
9509 the conditional expression itself. Instead, @value{GDBN} encodes the
9510 expression into an agent expression (@pxref{Agent Expressions}
9511 suitable for execution on the target, independently of @value{GDBN}.
9512 Global variables become raw memory locations, locals become stack
9513 accesses, and so forth.
9515 For instance, suppose you have a function that is usually called
9516 frequently, but should not be called after an error has occurred. You
9517 could use the following tracepoint command to collect data about calls
9518 of that function that happen while the error code is propagating
9519 through the program; an unconditional tracepoint could end up
9520 collecting thousands of useless trace frames that you would have to
9524 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
9527 @node Trace State Variables
9528 @subsection Trace State Variables
9529 @cindex trace state variables
9531 A @dfn{trace state variable} is a special type of variable that is
9532 created and managed by target-side code. The syntax is the same as
9533 that for GDB's convenience variables (a string prefixed with ``$''),
9534 but they are stored on the target. They must be created explicitly,
9535 using a @code{tvariable} command. They are always 64-bit signed
9538 Trace state variables are remembered by @value{GDBN}, and downloaded
9539 to the target along with tracepoint information when the trace
9540 experiment starts. There are no intrinsic limits on the number of
9541 trace state variables, beyond memory limitations of the target.
9543 @cindex convenience variables, and trace state variables
9544 Although trace state variables are managed by the target, you can use
9545 them in print commands and expressions as if they were convenience
9546 variables; @value{GDBN} will get the current value from the target
9547 while the trace experiment is running. Trace state variables share
9548 the same namespace as other ``$'' variables, which means that you
9549 cannot have trace state variables with names like @code{$23} or
9550 @code{$pc}, nor can you have a trace state variable and a convenience
9551 variable with the same name.
9555 @item tvariable $@var{name} [ = @var{expression} ]
9557 The @code{tvariable} command creates a new trace state variable named
9558 @code{$@var{name}}, and optionally gives it an initial value of
9559 @var{expression}. @var{expression} is evaluated when this command is
9560 entered; the result will be converted to an integer if possible,
9561 otherwise @value{GDBN} will report an error. A subsequent
9562 @code{tvariable} command specifying the same name does not create a
9563 variable, but instead assigns the supplied initial value to the
9564 existing variable of that name, overwriting any previous initial
9565 value. The default initial value is 0.
9567 @item info tvariables
9568 @kindex info tvariables
9569 List all the trace state variables along with their initial values.
9570 Their current values may also be displayed, if the trace experiment is
9573 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
9574 @kindex delete tvariable
9575 Delete the given trace state variables, or all of them if no arguments
9580 @node Tracepoint Actions
9581 @subsection Tracepoint Action Lists
9585 @cindex tracepoint actions
9586 @item actions @r{[}@var{num}@r{]}
9587 This command will prompt for a list of actions to be taken when the
9588 tracepoint is hit. If the tracepoint number @var{num} is not
9589 specified, this command sets the actions for the one that was most
9590 recently defined (so that you can define a tracepoint and then say
9591 @code{actions} without bothering about its number). You specify the
9592 actions themselves on the following lines, one action at a time, and
9593 terminate the actions list with a line containing just @code{end}. So
9594 far, the only defined actions are @code{collect} and
9595 @code{while-stepping}.
9597 @cindex remove actions from a tracepoint
9598 To remove all actions from a tracepoint, type @samp{actions @var{num}}
9599 and follow it immediately with @samp{end}.
9602 (@value{GDBP}) @b{collect @var{data}} // collect some data
9604 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
9606 (@value{GDBP}) @b{end} // signals the end of actions.
9609 In the following example, the action list begins with @code{collect}
9610 commands indicating the things to be collected when the tracepoint is
9611 hit. Then, in order to single-step and collect additional data
9612 following the tracepoint, a @code{while-stepping} command is used,
9613 followed by the list of things to be collected while stepping. The
9614 @code{while-stepping} command is terminated by its own separate
9615 @code{end} command. Lastly, the action list is terminated by an
9619 (@value{GDBP}) @b{trace foo}
9620 (@value{GDBP}) @b{actions}
9621 Enter actions for tracepoint 1, one per line:
9630 @kindex collect @r{(tracepoints)}
9631 @item collect @var{expr1}, @var{expr2}, @dots{}
9632 Collect values of the given expressions when the tracepoint is hit.
9633 This command accepts a comma-separated list of any valid expressions.
9634 In addition to global, static, or local variables, the following
9635 special arguments are supported:
9639 collect all registers
9642 collect all function arguments
9645 collect all local variables.
9648 You can give several consecutive @code{collect} commands, each one
9649 with a single argument, or one @code{collect} command with several
9650 arguments separated by commas: the effect is the same.
9652 The command @code{info scope} (@pxref{Symbols, info scope}) is
9653 particularly useful for figuring out what data to collect.
9655 @kindex teval @r{(tracepoints)}
9656 @item teval @var{expr1}, @var{expr2}, @dots{}
9657 Evaluate the given expressions when the tracepoint is hit. This
9658 command accepts a comma-separated list of expressions. The results
9659 are discarded, so this is mainly useful for assigning values to trace
9660 state variables (@pxref{Trace State Variables}) without adding those
9661 values to the trace buffer, as would be the case if the @code{collect}
9664 @kindex while-stepping @r{(tracepoints)}
9665 @item while-stepping @var{n}
9666 Perform @var{n} single-step traces after the tracepoint, collecting
9667 new data at each step. The @code{while-stepping} command is
9668 followed by the list of what to collect while stepping (followed by
9669 its own @code{end} command):
9673 > collect $regs, myglobal
9679 You may abbreviate @code{while-stepping} as @code{ws} or
9682 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
9683 @kindex set default-collect
9684 @cindex default collection action
9685 This variable is a list of expressions to collect at each tracepoint
9686 hit. It is effectively an additional @code{collect} action prepended
9687 to every tracepoint action list. The expressions are parsed
9688 individually for each tracepoint, so for instance a variable named
9689 @code{xyz} may be interpreted as a global for one tracepoint, and a
9690 local for another, as appropriate to the tracepoint's location.
9692 @item show default-collect
9693 @kindex show default-collect
9694 Show the list of expressions that are collected by default at each
9699 @node Listing Tracepoints
9700 @subsection Listing Tracepoints
9703 @kindex info tracepoints
9705 @cindex information about tracepoints
9706 @item info tracepoints @r{[}@var{num}@r{]}
9707 Display information about the tracepoint @var{num}. If you don't
9708 specify a tracepoint number, displays information about all the
9709 tracepoints defined so far. The format is similar to that used for
9710 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
9711 command, simply restricting itself to tracepoints.
9713 A tracepoint's listing may include additional information specific to
9718 its passcount as given by the @code{passcount @var{n}} command
9720 its step count as given by the @code{while-stepping @var{n}} command
9722 its action list as given by the @code{actions} command. The actions
9723 are prefixed with an @samp{A} so as to distinguish them from commands.
9727 (@value{GDBP}) @b{info trace}
9728 Num Type Disp Enb Address What
9729 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
9733 A collect globfoo, $regs
9741 This command can be abbreviated @code{info tp}.
9744 @node Starting and Stopping Trace Experiments
9745 @subsection Starting and Stopping Trace Experiments
9749 @cindex start a new trace experiment
9750 @cindex collected data discarded
9752 This command takes no arguments. It starts the trace experiment, and
9753 begins collecting data. This has the side effect of discarding all
9754 the data collected in the trace buffer during the previous trace
9758 @cindex stop a running trace experiment
9760 This command takes no arguments. It ends the trace experiment, and
9761 stops collecting data.
9763 @strong{Note}: a trace experiment and data collection may stop
9764 automatically if any tracepoint's passcount is reached
9765 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
9768 @cindex status of trace data collection
9769 @cindex trace experiment, status of
9771 This command displays the status of the current trace data
9775 Here is an example of the commands we described so far:
9778 (@value{GDBP}) @b{trace gdb_c_test}
9779 (@value{GDBP}) @b{actions}
9780 Enter actions for tracepoint #1, one per line.
9781 > collect $regs,$locals,$args
9786 (@value{GDBP}) @b{tstart}
9787 [time passes @dots{}]
9788 (@value{GDBP}) @b{tstop}
9791 @cindex disconnected tracing
9792 You can choose to continue running the trace experiment even if
9793 @value{GDBN} disconnects from the target, voluntarily or
9794 involuntarily. For commands such as @code{detach}, the debugger will
9795 ask what you want to do with the trace. But for unexpected
9796 terminations (@value{GDBN} crash, network outage), it would be
9797 unfortunate to lose hard-won trace data, so the variable
9798 @code{disconnected-tracing} lets you decide whether the trace should
9799 continue running without @value{GDBN}.
9802 @item set disconnected-tracing on
9803 @itemx set disconnected-tracing off
9804 @kindex set disconnected-tracing
9805 Choose whether a tracing run should continue to run if @value{GDBN}
9806 has disconnected from the target. Note that @code{detach} or
9807 @code{quit} will ask you directly what to do about a running trace no
9808 matter what this variable's setting, so the variable is mainly useful
9809 for handling unexpected situations, such as loss of the network.
9811 @item show disconnected-tracing
9812 @kindex show disconnected-tracing
9813 Show the current choice for disconnected tracing.
9817 When you reconnect to the target, the trace experiment may or may not
9818 still be running; it might have filled the trace buffer in the
9819 meantime, or stopped for one of the other reasons. If it is running,
9820 it will continue after reconnection.
9822 Upon reconnection, the target will upload information about the
9823 tracepoints in effect. @value{GDBN} will then compare that
9824 information to the set of tracepoints currently defined, and attempt
9825 to match them up, allowing for the possibility that the numbers may
9826 have changed due to creation and deletion in the meantime. If one of
9827 the target's tracepoints does not match any in @value{GDBN}, the
9828 debugger will create a new tracepoint, so that you have a number with
9829 which to specify that tracepoint. This matching-up process is
9830 necessarily heuristic, and it may result in useless tracepoints being
9831 created; you may simply delete them if they are of no use.
9833 @node Analyze Collected Data
9834 @section Using the Collected Data
9836 After the tracepoint experiment ends, you use @value{GDBN} commands
9837 for examining the trace data. The basic idea is that each tracepoint
9838 collects a trace @dfn{snapshot} every time it is hit and another
9839 snapshot every time it single-steps. All these snapshots are
9840 consecutively numbered from zero and go into a buffer, and you can
9841 examine them later. The way you examine them is to @dfn{focus} on a
9842 specific trace snapshot. When the remote stub is focused on a trace
9843 snapshot, it will respond to all @value{GDBN} requests for memory and
9844 registers by reading from the buffer which belongs to that snapshot,
9845 rather than from @emph{real} memory or registers of the program being
9846 debugged. This means that @strong{all} @value{GDBN} commands
9847 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
9848 behave as if we were currently debugging the program state as it was
9849 when the tracepoint occurred. Any requests for data that are not in
9850 the buffer will fail.
9853 * tfind:: How to select a trace snapshot
9854 * tdump:: How to display all data for a snapshot
9855 * save-tracepoints:: How to save tracepoints for a future run
9859 @subsection @code{tfind @var{n}}
9862 @cindex select trace snapshot
9863 @cindex find trace snapshot
9864 The basic command for selecting a trace snapshot from the buffer is
9865 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
9866 counting from zero. If no argument @var{n} is given, the next
9867 snapshot is selected.
9869 Here are the various forms of using the @code{tfind} command.
9873 Find the first snapshot in the buffer. This is a synonym for
9874 @code{tfind 0} (since 0 is the number of the first snapshot).
9877 Stop debugging trace snapshots, resume @emph{live} debugging.
9880 Same as @samp{tfind none}.
9883 No argument means find the next trace snapshot.
9886 Find the previous trace snapshot before the current one. This permits
9887 retracing earlier steps.
9889 @item tfind tracepoint @var{num}
9890 Find the next snapshot associated with tracepoint @var{num}. Search
9891 proceeds forward from the last examined trace snapshot. If no
9892 argument @var{num} is given, it means find the next snapshot collected
9893 for the same tracepoint as the current snapshot.
9895 @item tfind pc @var{addr}
9896 Find the next snapshot associated with the value @var{addr} of the
9897 program counter. Search proceeds forward from the last examined trace
9898 snapshot. If no argument @var{addr} is given, it means find the next
9899 snapshot with the same value of PC as the current snapshot.
9901 @item tfind outside @var{addr1}, @var{addr2}
9902 Find the next snapshot whose PC is outside the given range of
9903 addresses (exclusive).
9905 @item tfind range @var{addr1}, @var{addr2}
9906 Find the next snapshot whose PC is between @var{addr1} and
9907 @var{addr2} (inclusive).
9909 @item tfind line @r{[}@var{file}:@r{]}@var{n}
9910 Find the next snapshot associated with the source line @var{n}. If
9911 the optional argument @var{file} is given, refer to line @var{n} in
9912 that source file. Search proceeds forward from the last examined
9913 trace snapshot. If no argument @var{n} is given, it means find the
9914 next line other than the one currently being examined; thus saying
9915 @code{tfind line} repeatedly can appear to have the same effect as
9916 stepping from line to line in a @emph{live} debugging session.
9919 The default arguments for the @code{tfind} commands are specifically
9920 designed to make it easy to scan through the trace buffer. For
9921 instance, @code{tfind} with no argument selects the next trace
9922 snapshot, and @code{tfind -} with no argument selects the previous
9923 trace snapshot. So, by giving one @code{tfind} command, and then
9924 simply hitting @key{RET} repeatedly you can examine all the trace
9925 snapshots in order. Or, by saying @code{tfind -} and then hitting
9926 @key{RET} repeatedly you can examine the snapshots in reverse order.
9927 The @code{tfind line} command with no argument selects the snapshot
9928 for the next source line executed. The @code{tfind pc} command with
9929 no argument selects the next snapshot with the same program counter
9930 (PC) as the current frame. The @code{tfind tracepoint} command with
9931 no argument selects the next trace snapshot collected by the same
9932 tracepoint as the current one.
9934 In addition to letting you scan through the trace buffer manually,
9935 these commands make it easy to construct @value{GDBN} scripts that
9936 scan through the trace buffer and print out whatever collected data
9937 you are interested in. Thus, if we want to examine the PC, FP, and SP
9938 registers from each trace frame in the buffer, we can say this:
9941 (@value{GDBP}) @b{tfind start}
9942 (@value{GDBP}) @b{while ($trace_frame != -1)}
9943 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
9944 $trace_frame, $pc, $sp, $fp
9948 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
9949 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
9950 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
9951 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
9952 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
9953 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
9954 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
9955 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
9956 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
9957 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
9958 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
9961 Or, if we want to examine the variable @code{X} at each source line in
9965 (@value{GDBP}) @b{tfind start}
9966 (@value{GDBP}) @b{while ($trace_frame != -1)}
9967 > printf "Frame %d, X == %d\n", $trace_frame, X
9977 @subsection @code{tdump}
9979 @cindex dump all data collected at tracepoint
9980 @cindex tracepoint data, display
9982 This command takes no arguments. It prints all the data collected at
9983 the current trace snapshot.
9986 (@value{GDBP}) @b{trace 444}
9987 (@value{GDBP}) @b{actions}
9988 Enter actions for tracepoint #2, one per line:
9989 > collect $regs, $locals, $args, gdb_long_test
9992 (@value{GDBP}) @b{tstart}
9994 (@value{GDBP}) @b{tfind line 444}
9995 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
9997 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
9999 (@value{GDBP}) @b{tdump}
10000 Data collected at tracepoint 2, trace frame 1:
10001 d0 0xc4aa0085 -995491707
10005 d4 0x71aea3d 119204413
10008 d7 0x380035 3670069
10009 a0 0x19e24a 1696330
10010 a1 0x3000668 50333288
10012 a3 0x322000 3284992
10013 a4 0x3000698 50333336
10014 a5 0x1ad3cc 1758156
10015 fp 0x30bf3c 0x30bf3c
10016 sp 0x30bf34 0x30bf34
10018 pc 0x20b2c8 0x20b2c8
10022 p = 0x20e5b4 "gdb-test"
10029 gdb_long_test = 17 '\021'
10034 @node save-tracepoints
10035 @subsection @code{save-tracepoints @var{filename}}
10036 @kindex save-tracepoints
10037 @cindex save tracepoints for future sessions
10039 This command saves all current tracepoint definitions together with
10040 their actions and passcounts, into a file @file{@var{filename}}
10041 suitable for use in a later debugging session. To read the saved
10042 tracepoint definitions, use the @code{source} command (@pxref{Command
10045 @node Tracepoint Variables
10046 @section Convenience Variables for Tracepoints
10047 @cindex tracepoint variables
10048 @cindex convenience variables for tracepoints
10051 @vindex $trace_frame
10052 @item (int) $trace_frame
10053 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
10054 snapshot is selected.
10056 @vindex $tracepoint
10057 @item (int) $tracepoint
10058 The tracepoint for the current trace snapshot.
10060 @vindex $trace_line
10061 @item (int) $trace_line
10062 The line number for the current trace snapshot.
10064 @vindex $trace_file
10065 @item (char []) $trace_file
10066 The source file for the current trace snapshot.
10068 @vindex $trace_func
10069 @item (char []) $trace_func
10070 The name of the function containing @code{$tracepoint}.
10073 Note: @code{$trace_file} is not suitable for use in @code{printf},
10074 use @code{output} instead.
10076 Here's a simple example of using these convenience variables for
10077 stepping through all the trace snapshots and printing some of their
10078 data. Note that these are not the same as trace state variables,
10079 which are managed by the target.
10082 (@value{GDBP}) @b{tfind start}
10084 (@value{GDBP}) @b{while $trace_frame != -1}
10085 > output $trace_file
10086 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
10092 @section Using Trace Files
10093 @cindex trace files
10095 In some situations, the target running a trace experiment may no
10096 longer be available; perhaps it crashed, or the hardware was needed
10097 for a different activity. To handle these cases, you can arrange to
10098 dump the trace data into a file, and later use that file as a source
10099 of trace data, via the @code{target tfile} command.
10104 @item tsave [ -r ] @var{filename}
10105 Save the trace data to @var{filename}. By default, this command
10106 assumes that @var{filename} refers to the host filesystem, so if
10107 necessary @value{GDBN} will copy raw trace data up from the target and
10108 then save it. If the target supports it, you can also supply the
10109 optional argument @code{-r} (``remote'') to direct the target to save
10110 the data directly into @var{filename} in its own filesystem, which may be
10111 more efficient if the trace buffer is very large. (Note, however, that
10112 @code{target tfile} can only read from files accessible to the host.)
10114 @kindex target tfile
10116 @item target tfile @var{filename}
10117 Use the file named @var{filename} as a source of trace data. Commands
10118 that examine data work as they do with a live target, but it is not
10119 possible to run any new trace experiments. @code{tstatus} will report
10120 the state of the trace run at the moment the data was saved, as well
10121 as the current trace frame you are examining. @var{filename} must be
10122 on a filesystem accessible to the host.
10127 @chapter Debugging Programs That Use Overlays
10130 If your program is too large to fit completely in your target system's
10131 memory, you can sometimes use @dfn{overlays} to work around this
10132 problem. @value{GDBN} provides some support for debugging programs that
10136 * How Overlays Work:: A general explanation of overlays.
10137 * Overlay Commands:: Managing overlays in @value{GDBN}.
10138 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
10139 mapped by asking the inferior.
10140 * Overlay Sample Program:: A sample program using overlays.
10143 @node How Overlays Work
10144 @section How Overlays Work
10145 @cindex mapped overlays
10146 @cindex unmapped overlays
10147 @cindex load address, overlay's
10148 @cindex mapped address
10149 @cindex overlay area
10151 Suppose you have a computer whose instruction address space is only 64
10152 kilobytes long, but which has much more memory which can be accessed by
10153 other means: special instructions, segment registers, or memory
10154 management hardware, for example. Suppose further that you want to
10155 adapt a program which is larger than 64 kilobytes to run on this system.
10157 One solution is to identify modules of your program which are relatively
10158 independent, and need not call each other directly; call these modules
10159 @dfn{overlays}. Separate the overlays from the main program, and place
10160 their machine code in the larger memory. Place your main program in
10161 instruction memory, but leave at least enough space there to hold the
10162 largest overlay as well.
10164 Now, to call a function located in an overlay, you must first copy that
10165 overlay's machine code from the large memory into the space set aside
10166 for it in the instruction memory, and then jump to its entry point
10169 @c NB: In the below the mapped area's size is greater or equal to the
10170 @c size of all overlays. This is intentional to remind the developer
10171 @c that overlays don't necessarily need to be the same size.
10175 Data Instruction Larger
10176 Address Space Address Space Address Space
10177 +-----------+ +-----------+ +-----------+
10179 +-----------+ +-----------+ +-----------+<-- overlay 1
10180 | program | | main | .----| overlay 1 | load address
10181 | variables | | program | | +-----------+
10182 | and heap | | | | | |
10183 +-----------+ | | | +-----------+<-- overlay 2
10184 | | +-----------+ | | | load address
10185 +-----------+ | | | .-| overlay 2 |
10187 mapped --->+-----------+ | | +-----------+
10188 address | | | | | |
10189 | overlay | <-' | | |
10190 | area | <---' +-----------+<-- overlay 3
10191 | | <---. | | load address
10192 +-----------+ `--| overlay 3 |
10199 @anchor{A code overlay}A code overlay
10203 The diagram (@pxref{A code overlay}) shows a system with separate data
10204 and instruction address spaces. To map an overlay, the program copies
10205 its code from the larger address space to the instruction address space.
10206 Since the overlays shown here all use the same mapped address, only one
10207 may be mapped at a time. For a system with a single address space for
10208 data and instructions, the diagram would be similar, except that the
10209 program variables and heap would share an address space with the main
10210 program and the overlay area.
10212 An overlay loaded into instruction memory and ready for use is called a
10213 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
10214 instruction memory. An overlay not present (or only partially present)
10215 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
10216 is its address in the larger memory. The mapped address is also called
10217 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
10218 called the @dfn{load memory address}, or @dfn{LMA}.
10220 Unfortunately, overlays are not a completely transparent way to adapt a
10221 program to limited instruction memory. They introduce a new set of
10222 global constraints you must keep in mind as you design your program:
10227 Before calling or returning to a function in an overlay, your program
10228 must make sure that overlay is actually mapped. Otherwise, the call or
10229 return will transfer control to the right address, but in the wrong
10230 overlay, and your program will probably crash.
10233 If the process of mapping an overlay is expensive on your system, you
10234 will need to choose your overlays carefully to minimize their effect on
10235 your program's performance.
10238 The executable file you load onto your system must contain each
10239 overlay's instructions, appearing at the overlay's load address, not its
10240 mapped address. However, each overlay's instructions must be relocated
10241 and its symbols defined as if the overlay were at its mapped address.
10242 You can use GNU linker scripts to specify different load and relocation
10243 addresses for pieces of your program; see @ref{Overlay Description,,,
10244 ld.info, Using ld: the GNU linker}.
10247 The procedure for loading executable files onto your system must be able
10248 to load their contents into the larger address space as well as the
10249 instruction and data spaces.
10253 The overlay system described above is rather simple, and could be
10254 improved in many ways:
10259 If your system has suitable bank switch registers or memory management
10260 hardware, you could use those facilities to make an overlay's load area
10261 contents simply appear at their mapped address in instruction space.
10262 This would probably be faster than copying the overlay to its mapped
10263 area in the usual way.
10266 If your overlays are small enough, you could set aside more than one
10267 overlay area, and have more than one overlay mapped at a time.
10270 You can use overlays to manage data, as well as instructions. In
10271 general, data overlays are even less transparent to your design than
10272 code overlays: whereas code overlays only require care when you call or
10273 return to functions, data overlays require care every time you access
10274 the data. Also, if you change the contents of a data overlay, you
10275 must copy its contents back out to its load address before you can copy a
10276 different data overlay into the same mapped area.
10281 @node Overlay Commands
10282 @section Overlay Commands
10284 To use @value{GDBN}'s overlay support, each overlay in your program must
10285 correspond to a separate section of the executable file. The section's
10286 virtual memory address and load memory address must be the overlay's
10287 mapped and load addresses. Identifying overlays with sections allows
10288 @value{GDBN} to determine the appropriate address of a function or
10289 variable, depending on whether the overlay is mapped or not.
10291 @value{GDBN}'s overlay commands all start with the word @code{overlay};
10292 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
10297 Disable @value{GDBN}'s overlay support. When overlay support is
10298 disabled, @value{GDBN} assumes that all functions and variables are
10299 always present at their mapped addresses. By default, @value{GDBN}'s
10300 overlay support is disabled.
10302 @item overlay manual
10303 @cindex manual overlay debugging
10304 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
10305 relies on you to tell it which overlays are mapped, and which are not,
10306 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
10307 commands described below.
10309 @item overlay map-overlay @var{overlay}
10310 @itemx overlay map @var{overlay}
10311 @cindex map an overlay
10312 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
10313 be the name of the object file section containing the overlay. When an
10314 overlay is mapped, @value{GDBN} assumes it can find the overlay's
10315 functions and variables at their mapped addresses. @value{GDBN} assumes
10316 that any other overlays whose mapped ranges overlap that of
10317 @var{overlay} are now unmapped.
10319 @item overlay unmap-overlay @var{overlay}
10320 @itemx overlay unmap @var{overlay}
10321 @cindex unmap an overlay
10322 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
10323 must be the name of the object file section containing the overlay.
10324 When an overlay is unmapped, @value{GDBN} assumes it can find the
10325 overlay's functions and variables at their load addresses.
10328 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
10329 consults a data structure the overlay manager maintains in the inferior
10330 to see which overlays are mapped. For details, see @ref{Automatic
10331 Overlay Debugging}.
10333 @item overlay load-target
10334 @itemx overlay load
10335 @cindex reloading the overlay table
10336 Re-read the overlay table from the inferior. Normally, @value{GDBN}
10337 re-reads the table @value{GDBN} automatically each time the inferior
10338 stops, so this command should only be necessary if you have changed the
10339 overlay mapping yourself using @value{GDBN}. This command is only
10340 useful when using automatic overlay debugging.
10342 @item overlay list-overlays
10343 @itemx overlay list
10344 @cindex listing mapped overlays
10345 Display a list of the overlays currently mapped, along with their mapped
10346 addresses, load addresses, and sizes.
10350 Normally, when @value{GDBN} prints a code address, it includes the name
10351 of the function the address falls in:
10354 (@value{GDBP}) print main
10355 $3 = @{int ()@} 0x11a0 <main>
10358 When overlay debugging is enabled, @value{GDBN} recognizes code in
10359 unmapped overlays, and prints the names of unmapped functions with
10360 asterisks around them. For example, if @code{foo} is a function in an
10361 unmapped overlay, @value{GDBN} prints it this way:
10364 (@value{GDBP}) overlay list
10365 No sections are mapped.
10366 (@value{GDBP}) print foo
10367 $5 = @{int (int)@} 0x100000 <*foo*>
10370 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
10374 (@value{GDBP}) overlay list
10375 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
10376 mapped at 0x1016 - 0x104a
10377 (@value{GDBP}) print foo
10378 $6 = @{int (int)@} 0x1016 <foo>
10381 When overlay debugging is enabled, @value{GDBN} can find the correct
10382 address for functions and variables in an overlay, whether or not the
10383 overlay is mapped. This allows most @value{GDBN} commands, like
10384 @code{break} and @code{disassemble}, to work normally, even on unmapped
10385 code. However, @value{GDBN}'s breakpoint support has some limitations:
10389 @cindex breakpoints in overlays
10390 @cindex overlays, setting breakpoints in
10391 You can set breakpoints in functions in unmapped overlays, as long as
10392 @value{GDBN} can write to the overlay at its load address.
10394 @value{GDBN} can not set hardware or simulator-based breakpoints in
10395 unmapped overlays. However, if you set a breakpoint at the end of your
10396 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
10397 you are using manual overlay management), @value{GDBN} will re-set its
10398 breakpoints properly.
10402 @node Automatic Overlay Debugging
10403 @section Automatic Overlay Debugging
10404 @cindex automatic overlay debugging
10406 @value{GDBN} can automatically track which overlays are mapped and which
10407 are not, given some simple co-operation from the overlay manager in the
10408 inferior. If you enable automatic overlay debugging with the
10409 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
10410 looks in the inferior's memory for certain variables describing the
10411 current state of the overlays.
10413 Here are the variables your overlay manager must define to support
10414 @value{GDBN}'s automatic overlay debugging:
10418 @item @code{_ovly_table}:
10419 This variable must be an array of the following structures:
10424 /* The overlay's mapped address. */
10427 /* The size of the overlay, in bytes. */
10428 unsigned long size;
10430 /* The overlay's load address. */
10433 /* Non-zero if the overlay is currently mapped;
10435 unsigned long mapped;
10439 @item @code{_novlys}:
10440 This variable must be a four-byte signed integer, holding the total
10441 number of elements in @code{_ovly_table}.
10445 To decide whether a particular overlay is mapped or not, @value{GDBN}
10446 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
10447 @code{lma} members equal the VMA and LMA of the overlay's section in the
10448 executable file. When @value{GDBN} finds a matching entry, it consults
10449 the entry's @code{mapped} member to determine whether the overlay is
10452 In addition, your overlay manager may define a function called
10453 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
10454 will silently set a breakpoint there. If the overlay manager then
10455 calls this function whenever it has changed the overlay table, this
10456 will enable @value{GDBN} to accurately keep track of which overlays
10457 are in program memory, and update any breakpoints that may be set
10458 in overlays. This will allow breakpoints to work even if the
10459 overlays are kept in ROM or other non-writable memory while they
10460 are not being executed.
10462 @node Overlay Sample Program
10463 @section Overlay Sample Program
10464 @cindex overlay example program
10466 When linking a program which uses overlays, you must place the overlays
10467 at their load addresses, while relocating them to run at their mapped
10468 addresses. To do this, you must write a linker script (@pxref{Overlay
10469 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
10470 since linker scripts are specific to a particular host system, target
10471 architecture, and target memory layout, this manual cannot provide
10472 portable sample code demonstrating @value{GDBN}'s overlay support.
10474 However, the @value{GDBN} source distribution does contain an overlaid
10475 program, with linker scripts for a few systems, as part of its test
10476 suite. The program consists of the following files from
10477 @file{gdb/testsuite/gdb.base}:
10481 The main program file.
10483 A simple overlay manager, used by @file{overlays.c}.
10488 Overlay modules, loaded and used by @file{overlays.c}.
10491 Linker scripts for linking the test program on the @code{d10v-elf}
10492 and @code{m32r-elf} targets.
10495 You can build the test program using the @code{d10v-elf} GCC
10496 cross-compiler like this:
10499 $ d10v-elf-gcc -g -c overlays.c
10500 $ d10v-elf-gcc -g -c ovlymgr.c
10501 $ d10v-elf-gcc -g -c foo.c
10502 $ d10v-elf-gcc -g -c bar.c
10503 $ d10v-elf-gcc -g -c baz.c
10504 $ d10v-elf-gcc -g -c grbx.c
10505 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
10506 baz.o grbx.o -Wl,-Td10v.ld -o overlays
10509 The build process is identical for any other architecture, except that
10510 you must substitute the appropriate compiler and linker script for the
10511 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
10515 @chapter Using @value{GDBN} with Different Languages
10518 Although programming languages generally have common aspects, they are
10519 rarely expressed in the same manner. For instance, in ANSI C,
10520 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
10521 Modula-2, it is accomplished by @code{p^}. Values can also be
10522 represented (and displayed) differently. Hex numbers in C appear as
10523 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
10525 @cindex working language
10526 Language-specific information is built into @value{GDBN} for some languages,
10527 allowing you to express operations like the above in your program's
10528 native language, and allowing @value{GDBN} to output values in a manner
10529 consistent with the syntax of your program's native language. The
10530 language you use to build expressions is called the @dfn{working
10534 * Setting:: Switching between source languages
10535 * Show:: Displaying the language
10536 * Checks:: Type and range checks
10537 * Supported Languages:: Supported languages
10538 * Unsupported Languages:: Unsupported languages
10542 @section Switching Between Source Languages
10544 There are two ways to control the working language---either have @value{GDBN}
10545 set it automatically, or select it manually yourself. You can use the
10546 @code{set language} command for either purpose. On startup, @value{GDBN}
10547 defaults to setting the language automatically. The working language is
10548 used to determine how expressions you type are interpreted, how values
10551 In addition to the working language, every source file that
10552 @value{GDBN} knows about has its own working language. For some object
10553 file formats, the compiler might indicate which language a particular
10554 source file is in. However, most of the time @value{GDBN} infers the
10555 language from the name of the file. The language of a source file
10556 controls whether C@t{++} names are demangled---this way @code{backtrace} can
10557 show each frame appropriately for its own language. There is no way to
10558 set the language of a source file from within @value{GDBN}, but you can
10559 set the language associated with a filename extension. @xref{Show, ,
10560 Displaying the Language}.
10562 This is most commonly a problem when you use a program, such
10563 as @code{cfront} or @code{f2c}, that generates C but is written in
10564 another language. In that case, make the
10565 program use @code{#line} directives in its C output; that way
10566 @value{GDBN} will know the correct language of the source code of the original
10567 program, and will display that source code, not the generated C code.
10570 * Filenames:: Filename extensions and languages.
10571 * Manually:: Setting the working language manually
10572 * Automatically:: Having @value{GDBN} infer the source language
10576 @subsection List of Filename Extensions and Languages
10578 If a source file name ends in one of the following extensions, then
10579 @value{GDBN} infers that its language is the one indicated.
10597 C@t{++} source file
10600 Objective-C source file
10604 Fortran source file
10607 Modula-2 source file
10611 Assembler source file. This actually behaves almost like C, but
10612 @value{GDBN} does not skip over function prologues when stepping.
10615 In addition, you may set the language associated with a filename
10616 extension. @xref{Show, , Displaying the Language}.
10619 @subsection Setting the Working Language
10621 If you allow @value{GDBN} to set the language automatically,
10622 expressions are interpreted the same way in your debugging session and
10625 @kindex set language
10626 If you wish, you may set the language manually. To do this, issue the
10627 command @samp{set language @var{lang}}, where @var{lang} is the name of
10628 a language, such as
10629 @code{c} or @code{modula-2}.
10630 For a list of the supported languages, type @samp{set language}.
10632 Setting the language manually prevents @value{GDBN} from updating the working
10633 language automatically. This can lead to confusion if you try
10634 to debug a program when the working language is not the same as the
10635 source language, when an expression is acceptable to both
10636 languages---but means different things. For instance, if the current
10637 source file were written in C, and @value{GDBN} was parsing Modula-2, a
10645 might not have the effect you intended. In C, this means to add
10646 @code{b} and @code{c} and place the result in @code{a}. The result
10647 printed would be the value of @code{a}. In Modula-2, this means to compare
10648 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
10650 @node Automatically
10651 @subsection Having @value{GDBN} Infer the Source Language
10653 To have @value{GDBN} set the working language automatically, use
10654 @samp{set language local} or @samp{set language auto}. @value{GDBN}
10655 then infers the working language. That is, when your program stops in a
10656 frame (usually by encountering a breakpoint), @value{GDBN} sets the
10657 working language to the language recorded for the function in that
10658 frame. If the language for a frame is unknown (that is, if the function
10659 or block corresponding to the frame was defined in a source file that
10660 does not have a recognized extension), the current working language is
10661 not changed, and @value{GDBN} issues a warning.
10663 This may not seem necessary for most programs, which are written
10664 entirely in one source language. However, program modules and libraries
10665 written in one source language can be used by a main program written in
10666 a different source language. Using @samp{set language auto} in this
10667 case frees you from having to set the working language manually.
10670 @section Displaying the Language
10672 The following commands help you find out which language is the
10673 working language, and also what language source files were written in.
10676 @item show language
10677 @kindex show language
10678 Display the current working language. This is the
10679 language you can use with commands such as @code{print} to
10680 build and compute expressions that may involve variables in your program.
10683 @kindex info frame@r{, show the source language}
10684 Display the source language for this frame. This language becomes the
10685 working language if you use an identifier from this frame.
10686 @xref{Frame Info, ,Information about a Frame}, to identify the other
10687 information listed here.
10690 @kindex info source@r{, show the source language}
10691 Display the source language of this source file.
10692 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
10693 information listed here.
10696 In unusual circumstances, you may have source files with extensions
10697 not in the standard list. You can then set the extension associated
10698 with a language explicitly:
10701 @item set extension-language @var{ext} @var{language}
10702 @kindex set extension-language
10703 Tell @value{GDBN} that source files with extension @var{ext} are to be
10704 assumed as written in the source language @var{language}.
10706 @item info extensions
10707 @kindex info extensions
10708 List all the filename extensions and the associated languages.
10712 @section Type and Range Checking
10715 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
10716 checking are included, but they do not yet have any effect. This
10717 section documents the intended facilities.
10719 @c FIXME remove warning when type/range code added
10721 Some languages are designed to guard you against making seemingly common
10722 errors through a series of compile- and run-time checks. These include
10723 checking the type of arguments to functions and operators, and making
10724 sure mathematical overflows are caught at run time. Checks such as
10725 these help to ensure a program's correctness once it has been compiled
10726 by eliminating type mismatches, and providing active checks for range
10727 errors when your program is running.
10729 @value{GDBN} can check for conditions like the above if you wish.
10730 Although @value{GDBN} does not check the statements in your program,
10731 it can check expressions entered directly into @value{GDBN} for
10732 evaluation via the @code{print} command, for example. As with the
10733 working language, @value{GDBN} can also decide whether or not to check
10734 automatically based on your program's source language.
10735 @xref{Supported Languages, ,Supported Languages}, for the default
10736 settings of supported languages.
10739 * Type Checking:: An overview of type checking
10740 * Range Checking:: An overview of range checking
10743 @cindex type checking
10744 @cindex checks, type
10745 @node Type Checking
10746 @subsection An Overview of Type Checking
10748 Some languages, such as Modula-2, are strongly typed, meaning that the
10749 arguments to operators and functions have to be of the correct type,
10750 otherwise an error occurs. These checks prevent type mismatch
10751 errors from ever causing any run-time problems. For example,
10759 The second example fails because the @code{CARDINAL} 1 is not
10760 type-compatible with the @code{REAL} 2.3.
10762 For the expressions you use in @value{GDBN} commands, you can tell the
10763 @value{GDBN} type checker to skip checking;
10764 to treat any mismatches as errors and abandon the expression;
10765 or to only issue warnings when type mismatches occur,
10766 but evaluate the expression anyway. When you choose the last of
10767 these, @value{GDBN} evaluates expressions like the second example above, but
10768 also issues a warning.
10770 Even if you turn type checking off, there may be other reasons
10771 related to type that prevent @value{GDBN} from evaluating an expression.
10772 For instance, @value{GDBN} does not know how to add an @code{int} and
10773 a @code{struct foo}. These particular type errors have nothing to do
10774 with the language in use, and usually arise from expressions, such as
10775 the one described above, which make little sense to evaluate anyway.
10777 Each language defines to what degree it is strict about type. For
10778 instance, both Modula-2 and C require the arguments to arithmetical
10779 operators to be numbers. In C, enumerated types and pointers can be
10780 represented as numbers, so that they are valid arguments to mathematical
10781 operators. @xref{Supported Languages, ,Supported Languages}, for further
10782 details on specific languages.
10784 @value{GDBN} provides some additional commands for controlling the type checker:
10786 @kindex set check type
10787 @kindex show check type
10789 @item set check type auto
10790 Set type checking on or off based on the current working language.
10791 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10794 @item set check type on
10795 @itemx set check type off
10796 Set type checking on or off, overriding the default setting for the
10797 current working language. Issue a warning if the setting does not
10798 match the language default. If any type mismatches occur in
10799 evaluating an expression while type checking is on, @value{GDBN} prints a
10800 message and aborts evaluation of the expression.
10802 @item set check type warn
10803 Cause the type checker to issue warnings, but to always attempt to
10804 evaluate the expression. Evaluating the expression may still
10805 be impossible for other reasons. For example, @value{GDBN} cannot add
10806 numbers and structures.
10809 Show the current setting of the type checker, and whether or not @value{GDBN}
10810 is setting it automatically.
10813 @cindex range checking
10814 @cindex checks, range
10815 @node Range Checking
10816 @subsection An Overview of Range Checking
10818 In some languages (such as Modula-2), it is an error to exceed the
10819 bounds of a type; this is enforced with run-time checks. Such range
10820 checking is meant to ensure program correctness by making sure
10821 computations do not overflow, or indices on an array element access do
10822 not exceed the bounds of the array.
10824 For expressions you use in @value{GDBN} commands, you can tell
10825 @value{GDBN} to treat range errors in one of three ways: ignore them,
10826 always treat them as errors and abandon the expression, or issue
10827 warnings but evaluate the expression anyway.
10829 A range error can result from numerical overflow, from exceeding an
10830 array index bound, or when you type a constant that is not a member
10831 of any type. Some languages, however, do not treat overflows as an
10832 error. In many implementations of C, mathematical overflow causes the
10833 result to ``wrap around'' to lower values---for example, if @var{m} is
10834 the largest integer value, and @var{s} is the smallest, then
10837 @var{m} + 1 @result{} @var{s}
10840 This, too, is specific to individual languages, and in some cases
10841 specific to individual compilers or machines. @xref{Supported Languages, ,
10842 Supported Languages}, for further details on specific languages.
10844 @value{GDBN} provides some additional commands for controlling the range checker:
10846 @kindex set check range
10847 @kindex show check range
10849 @item set check range auto
10850 Set range checking on or off based on the current working language.
10851 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10854 @item set check range on
10855 @itemx set check range off
10856 Set range checking on or off, overriding the default setting for the
10857 current working language. A warning is issued if the setting does not
10858 match the language default. If a range error occurs and range checking is on,
10859 then a message is printed and evaluation of the expression is aborted.
10861 @item set check range warn
10862 Output messages when the @value{GDBN} range checker detects a range error,
10863 but attempt to evaluate the expression anyway. Evaluating the
10864 expression may still be impossible for other reasons, such as accessing
10865 memory that the process does not own (a typical example from many Unix
10869 Show the current setting of the range checker, and whether or not it is
10870 being set automatically by @value{GDBN}.
10873 @node Supported Languages
10874 @section Supported Languages
10876 @value{GDBN} supports C, C@t{++}, Objective-C, Fortran, Java, Pascal,
10877 assembly, Modula-2, and Ada.
10878 @c This is false ...
10879 Some @value{GDBN} features may be used in expressions regardless of the
10880 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
10881 and the @samp{@{type@}addr} construct (@pxref{Expressions,
10882 ,Expressions}) can be used with the constructs of any supported
10885 The following sections detail to what degree each source language is
10886 supported by @value{GDBN}. These sections are not meant to be language
10887 tutorials or references, but serve only as a reference guide to what the
10888 @value{GDBN} expression parser accepts, and what input and output
10889 formats should look like for different languages. There are many good
10890 books written on each of these languages; please look to these for a
10891 language reference or tutorial.
10894 * C:: C and C@t{++}
10895 * Objective-C:: Objective-C
10896 * Fortran:: Fortran
10898 * Modula-2:: Modula-2
10903 @subsection C and C@t{++}
10905 @cindex C and C@t{++}
10906 @cindex expressions in C or C@t{++}
10908 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
10909 to both languages. Whenever this is the case, we discuss those languages
10913 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
10914 @cindex @sc{gnu} C@t{++}
10915 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
10916 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
10917 effectively, you must compile your C@t{++} programs with a supported
10918 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
10919 compiler (@code{aCC}).
10921 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
10922 format; if it doesn't work on your system, try the stabs+ debugging
10923 format. You can select those formats explicitly with the @code{g++}
10924 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
10925 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
10926 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
10929 * C Operators:: C and C@t{++} operators
10930 * C Constants:: C and C@t{++} constants
10931 * C Plus Plus Expressions:: C@t{++} expressions
10932 * C Defaults:: Default settings for C and C@t{++}
10933 * C Checks:: C and C@t{++} type and range checks
10934 * Debugging C:: @value{GDBN} and C
10935 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
10936 * Decimal Floating Point:: Numbers in Decimal Floating Point format
10940 @subsubsection C and C@t{++} Operators
10942 @cindex C and C@t{++} operators
10944 Operators must be defined on values of specific types. For instance,
10945 @code{+} is defined on numbers, but not on structures. Operators are
10946 often defined on groups of types.
10948 For the purposes of C and C@t{++}, the following definitions hold:
10953 @emph{Integral types} include @code{int} with any of its storage-class
10954 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
10957 @emph{Floating-point types} include @code{float}, @code{double}, and
10958 @code{long double} (if supported by the target platform).
10961 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
10964 @emph{Scalar types} include all of the above.
10969 The following operators are supported. They are listed here
10970 in order of increasing precedence:
10974 The comma or sequencing operator. Expressions in a comma-separated list
10975 are evaluated from left to right, with the result of the entire
10976 expression being the last expression evaluated.
10979 Assignment. The value of an assignment expression is the value
10980 assigned. Defined on scalar types.
10983 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
10984 and translated to @w{@code{@var{a} = @var{a op b}}}.
10985 @w{@code{@var{op}=}} and @code{=} have the same precedence.
10986 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
10987 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
10990 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
10991 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
10995 Logical @sc{or}. Defined on integral types.
10998 Logical @sc{and}. Defined on integral types.
11001 Bitwise @sc{or}. Defined on integral types.
11004 Bitwise exclusive-@sc{or}. Defined on integral types.
11007 Bitwise @sc{and}. Defined on integral types.
11010 Equality and inequality. Defined on scalar types. The value of these
11011 expressions is 0 for false and non-zero for true.
11013 @item <@r{, }>@r{, }<=@r{, }>=
11014 Less than, greater than, less than or equal, greater than or equal.
11015 Defined on scalar types. The value of these expressions is 0 for false
11016 and non-zero for true.
11019 left shift, and right shift. Defined on integral types.
11022 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11025 Addition and subtraction. Defined on integral types, floating-point types and
11028 @item *@r{, }/@r{, }%
11029 Multiplication, division, and modulus. Multiplication and division are
11030 defined on integral and floating-point types. Modulus is defined on
11034 Increment and decrement. When appearing before a variable, the
11035 operation is performed before the variable is used in an expression;
11036 when appearing after it, the variable's value is used before the
11037 operation takes place.
11040 Pointer dereferencing. Defined on pointer types. Same precedence as
11044 Address operator. Defined on variables. Same precedence as @code{++}.
11046 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
11047 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
11048 to examine the address
11049 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
11053 Negative. Defined on integral and floating-point types. Same
11054 precedence as @code{++}.
11057 Logical negation. Defined on integral types. Same precedence as
11061 Bitwise complement operator. Defined on integral types. Same precedence as
11066 Structure member, and pointer-to-structure member. For convenience,
11067 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
11068 pointer based on the stored type information.
11069 Defined on @code{struct} and @code{union} data.
11072 Dereferences of pointers to members.
11075 Array indexing. @code{@var{a}[@var{i}]} is defined as
11076 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
11079 Function parameter list. Same precedence as @code{->}.
11082 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
11083 and @code{class} types.
11086 Doubled colons also represent the @value{GDBN} scope operator
11087 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
11091 If an operator is redefined in the user code, @value{GDBN} usually
11092 attempts to invoke the redefined version instead of using the operator's
11093 predefined meaning.
11096 @subsubsection C and C@t{++} Constants
11098 @cindex C and C@t{++} constants
11100 @value{GDBN} allows you to express the constants of C and C@t{++} in the
11105 Integer constants are a sequence of digits. Octal constants are
11106 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
11107 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
11108 @samp{l}, specifying that the constant should be treated as a
11112 Floating point constants are a sequence of digits, followed by a decimal
11113 point, followed by a sequence of digits, and optionally followed by an
11114 exponent. An exponent is of the form:
11115 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
11116 sequence of digits. The @samp{+} is optional for positive exponents.
11117 A floating-point constant may also end with a letter @samp{f} or
11118 @samp{F}, specifying that the constant should be treated as being of
11119 the @code{float} (as opposed to the default @code{double}) type; or with
11120 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
11124 Enumerated constants consist of enumerated identifiers, or their
11125 integral equivalents.
11128 Character constants are a single character surrounded by single quotes
11129 (@code{'}), or a number---the ordinal value of the corresponding character
11130 (usually its @sc{ascii} value). Within quotes, the single character may
11131 be represented by a letter or by @dfn{escape sequences}, which are of
11132 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
11133 of the character's ordinal value; or of the form @samp{\@var{x}}, where
11134 @samp{@var{x}} is a predefined special character---for example,
11135 @samp{\n} for newline.
11138 String constants are a sequence of character constants surrounded by
11139 double quotes (@code{"}). Any valid character constant (as described
11140 above) may appear. Double quotes within the string must be preceded by
11141 a backslash, so for instance @samp{"a\"b'c"} is a string of five
11145 Pointer constants are an integral value. You can also write pointers
11146 to constants using the C operator @samp{&}.
11149 Array constants are comma-separated lists surrounded by braces @samp{@{}
11150 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
11151 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
11152 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
11155 @node C Plus Plus Expressions
11156 @subsubsection C@t{++} Expressions
11158 @cindex expressions in C@t{++}
11159 @value{GDBN} expression handling can interpret most C@t{++} expressions.
11161 @cindex debugging C@t{++} programs
11162 @cindex C@t{++} compilers
11163 @cindex debug formats and C@t{++}
11164 @cindex @value{NGCC} and C@t{++}
11166 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
11167 proper compiler and the proper debug format. Currently, @value{GDBN}
11168 works best when debugging C@t{++} code that is compiled with
11169 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
11170 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
11171 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
11172 stabs+ as their default debug format, so you usually don't need to
11173 specify a debug format explicitly. Other compilers and/or debug formats
11174 are likely to work badly or not at all when using @value{GDBN} to debug
11180 @cindex member functions
11182 Member function calls are allowed; you can use expressions like
11185 count = aml->GetOriginal(x, y)
11188 @vindex this@r{, inside C@t{++} member functions}
11189 @cindex namespace in C@t{++}
11191 While a member function is active (in the selected stack frame), your
11192 expressions have the same namespace available as the member function;
11193 that is, @value{GDBN} allows implicit references to the class instance
11194 pointer @code{this} following the same rules as C@t{++}.
11196 @cindex call overloaded functions
11197 @cindex overloaded functions, calling
11198 @cindex type conversions in C@t{++}
11200 You can call overloaded functions; @value{GDBN} resolves the function
11201 call to the right definition, with some restrictions. @value{GDBN} does not
11202 perform overload resolution involving user-defined type conversions,
11203 calls to constructors, or instantiations of templates that do not exist
11204 in the program. It also cannot handle ellipsis argument lists or
11207 It does perform integral conversions and promotions, floating-point
11208 promotions, arithmetic conversions, pointer conversions, conversions of
11209 class objects to base classes, and standard conversions such as those of
11210 functions or arrays to pointers; it requires an exact match on the
11211 number of function arguments.
11213 Overload resolution is always performed, unless you have specified
11214 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
11215 ,@value{GDBN} Features for C@t{++}}.
11217 You must specify @code{set overload-resolution off} in order to use an
11218 explicit function signature to call an overloaded function, as in
11220 p 'foo(char,int)'('x', 13)
11223 The @value{GDBN} command-completion facility can simplify this;
11224 see @ref{Completion, ,Command Completion}.
11226 @cindex reference declarations
11228 @value{GDBN} understands variables declared as C@t{++} references; you can use
11229 them in expressions just as you do in C@t{++} source---they are automatically
11232 In the parameter list shown when @value{GDBN} displays a frame, the values of
11233 reference variables are not displayed (unlike other variables); this
11234 avoids clutter, since references are often used for large structures.
11235 The @emph{address} of a reference variable is always shown, unless
11236 you have specified @samp{set print address off}.
11239 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
11240 expressions can use it just as expressions in your program do. Since
11241 one scope may be defined in another, you can use @code{::} repeatedly if
11242 necessary, for example in an expression like
11243 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
11244 resolving name scope by reference to source files, in both C and C@t{++}
11245 debugging (@pxref{Variables, ,Program Variables}).
11248 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
11249 calling virtual functions correctly, printing out virtual bases of
11250 objects, calling functions in a base subobject, casting objects, and
11251 invoking user-defined operators.
11254 @subsubsection C and C@t{++} Defaults
11256 @cindex C and C@t{++} defaults
11258 If you allow @value{GDBN} to set type and range checking automatically, they
11259 both default to @code{off} whenever the working language changes to
11260 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
11261 selects the working language.
11263 If you allow @value{GDBN} to set the language automatically, it
11264 recognizes source files whose names end with @file{.c}, @file{.C}, or
11265 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
11266 these files, it sets the working language to C or C@t{++}.
11267 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
11268 for further details.
11270 @c Type checking is (a) primarily motivated by Modula-2, and (b)
11271 @c unimplemented. If (b) changes, it might make sense to let this node
11272 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
11275 @subsubsection C and C@t{++} Type and Range Checks
11277 @cindex C and C@t{++} checks
11279 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
11280 is not used. However, if you turn type checking on, @value{GDBN}
11281 considers two variables type equivalent if:
11285 The two variables are structured and have the same structure, union, or
11289 The two variables have the same type name, or types that have been
11290 declared equivalent through @code{typedef}.
11293 @c leaving this out because neither J Gilmore nor R Pesch understand it.
11296 The two @code{struct}, @code{union}, or @code{enum} variables are
11297 declared in the same declaration. (Note: this may not be true for all C
11302 Range checking, if turned on, is done on mathematical operations. Array
11303 indices are not checked, since they are often used to index a pointer
11304 that is not itself an array.
11307 @subsubsection @value{GDBN} and C
11309 The @code{set print union} and @code{show print union} commands apply to
11310 the @code{union} type. When set to @samp{on}, any @code{union} that is
11311 inside a @code{struct} or @code{class} is also printed. Otherwise, it
11312 appears as @samp{@{...@}}.
11314 The @code{@@} operator aids in the debugging of dynamic arrays, formed
11315 with pointers and a memory allocation function. @xref{Expressions,
11318 @node Debugging C Plus Plus
11319 @subsubsection @value{GDBN} Features for C@t{++}
11321 @cindex commands for C@t{++}
11323 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
11324 designed specifically for use with C@t{++}. Here is a summary:
11327 @cindex break in overloaded functions
11328 @item @r{breakpoint menus}
11329 When you want a breakpoint in a function whose name is overloaded,
11330 @value{GDBN} has the capability to display a menu of possible breakpoint
11331 locations to help you specify which function definition you want.
11332 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
11334 @cindex overloading in C@t{++}
11335 @item rbreak @var{regex}
11336 Setting breakpoints using regular expressions is helpful for setting
11337 breakpoints on overloaded functions that are not members of any special
11339 @xref{Set Breaks, ,Setting Breakpoints}.
11341 @cindex C@t{++} exception handling
11344 Debug C@t{++} exception handling using these commands. @xref{Set
11345 Catchpoints, , Setting Catchpoints}.
11347 @cindex inheritance
11348 @item ptype @var{typename}
11349 Print inheritance relationships as well as other information for type
11351 @xref{Symbols, ,Examining the Symbol Table}.
11353 @cindex C@t{++} symbol display
11354 @item set print demangle
11355 @itemx show print demangle
11356 @itemx set print asm-demangle
11357 @itemx show print asm-demangle
11358 Control whether C@t{++} symbols display in their source form, both when
11359 displaying code as C@t{++} source and when displaying disassemblies.
11360 @xref{Print Settings, ,Print Settings}.
11362 @item set print object
11363 @itemx show print object
11364 Choose whether to print derived (actual) or declared types of objects.
11365 @xref{Print Settings, ,Print Settings}.
11367 @item set print vtbl
11368 @itemx show print vtbl
11369 Control the format for printing virtual function tables.
11370 @xref{Print Settings, ,Print Settings}.
11371 (The @code{vtbl} commands do not work on programs compiled with the HP
11372 ANSI C@t{++} compiler (@code{aCC}).)
11374 @kindex set overload-resolution
11375 @cindex overloaded functions, overload resolution
11376 @item set overload-resolution on
11377 Enable overload resolution for C@t{++} expression evaluation. The default
11378 is on. For overloaded functions, @value{GDBN} evaluates the arguments
11379 and searches for a function whose signature matches the argument types,
11380 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
11381 Expressions, ,C@t{++} Expressions}, for details).
11382 If it cannot find a match, it emits a message.
11384 @item set overload-resolution off
11385 Disable overload resolution for C@t{++} expression evaluation. For
11386 overloaded functions that are not class member functions, @value{GDBN}
11387 chooses the first function of the specified name that it finds in the
11388 symbol table, whether or not its arguments are of the correct type. For
11389 overloaded functions that are class member functions, @value{GDBN}
11390 searches for a function whose signature @emph{exactly} matches the
11393 @kindex show overload-resolution
11394 @item show overload-resolution
11395 Show the current setting of overload resolution.
11397 @item @r{Overloaded symbol names}
11398 You can specify a particular definition of an overloaded symbol, using
11399 the same notation that is used to declare such symbols in C@t{++}: type
11400 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
11401 also use the @value{GDBN} command-line word completion facilities to list the
11402 available choices, or to finish the type list for you.
11403 @xref{Completion,, Command Completion}, for details on how to do this.
11406 @node Decimal Floating Point
11407 @subsubsection Decimal Floating Point format
11408 @cindex decimal floating point format
11410 @value{GDBN} can examine, set and perform computations with numbers in
11411 decimal floating point format, which in the C language correspond to the
11412 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
11413 specified by the extension to support decimal floating-point arithmetic.
11415 There are two encodings in use, depending on the architecture: BID (Binary
11416 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
11417 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
11420 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
11421 to manipulate decimal floating point numbers, it is not possible to convert
11422 (using a cast, for example) integers wider than 32-bit to decimal float.
11424 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
11425 point computations, error checking in decimal float operations ignores
11426 underflow, overflow and divide by zero exceptions.
11428 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
11429 to inspect @code{_Decimal128} values stored in floating point registers.
11430 See @ref{PowerPC,,PowerPC} for more details.
11433 @subsection Objective-C
11435 @cindex Objective-C
11436 This section provides information about some commands and command
11437 options that are useful for debugging Objective-C code. See also
11438 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
11439 few more commands specific to Objective-C support.
11442 * Method Names in Commands::
11443 * The Print Command with Objective-C::
11446 @node Method Names in Commands
11447 @subsubsection Method Names in Commands
11449 The following commands have been extended to accept Objective-C method
11450 names as line specifications:
11452 @kindex clear@r{, and Objective-C}
11453 @kindex break@r{, and Objective-C}
11454 @kindex info line@r{, and Objective-C}
11455 @kindex jump@r{, and Objective-C}
11456 @kindex list@r{, and Objective-C}
11460 @item @code{info line}
11465 A fully qualified Objective-C method name is specified as
11468 -[@var{Class} @var{methodName}]
11471 where the minus sign is used to indicate an instance method and a
11472 plus sign (not shown) is used to indicate a class method. The class
11473 name @var{Class} and method name @var{methodName} are enclosed in
11474 brackets, similar to the way messages are specified in Objective-C
11475 source code. For example, to set a breakpoint at the @code{create}
11476 instance method of class @code{Fruit} in the program currently being
11480 break -[Fruit create]
11483 To list ten program lines around the @code{initialize} class method,
11487 list +[NSText initialize]
11490 In the current version of @value{GDBN}, the plus or minus sign is
11491 required. In future versions of @value{GDBN}, the plus or minus
11492 sign will be optional, but you can use it to narrow the search. It
11493 is also possible to specify just a method name:
11499 You must specify the complete method name, including any colons. If
11500 your program's source files contain more than one @code{create} method,
11501 you'll be presented with a numbered list of classes that implement that
11502 method. Indicate your choice by number, or type @samp{0} to exit if
11505 As another example, to clear a breakpoint established at the
11506 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
11509 clear -[NSWindow makeKeyAndOrderFront:]
11512 @node The Print Command with Objective-C
11513 @subsubsection The Print Command With Objective-C
11514 @cindex Objective-C, print objects
11515 @kindex print-object
11516 @kindex po @r{(@code{print-object})}
11518 The print command has also been extended to accept methods. For example:
11521 print -[@var{object} hash]
11524 @cindex print an Objective-C object description
11525 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
11527 will tell @value{GDBN} to send the @code{hash} message to @var{object}
11528 and print the result. Also, an additional command has been added,
11529 @code{print-object} or @code{po} for short, which is meant to print
11530 the description of an object. However, this command may only work
11531 with certain Objective-C libraries that have a particular hook
11532 function, @code{_NSPrintForDebugger}, defined.
11535 @subsection Fortran
11536 @cindex Fortran-specific support in @value{GDBN}
11538 @value{GDBN} can be used to debug programs written in Fortran, but it
11539 currently supports only the features of Fortran 77 language.
11541 @cindex trailing underscore, in Fortran symbols
11542 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
11543 among them) append an underscore to the names of variables and
11544 functions. When you debug programs compiled by those compilers, you
11545 will need to refer to variables and functions with a trailing
11549 * Fortran Operators:: Fortran operators and expressions
11550 * Fortran Defaults:: Default settings for Fortran
11551 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
11554 @node Fortran Operators
11555 @subsubsection Fortran Operators and Expressions
11557 @cindex Fortran operators and expressions
11559 Operators must be defined on values of specific types. For instance,
11560 @code{+} is defined on numbers, but not on characters or other non-
11561 arithmetic types. Operators are often defined on groups of types.
11565 The exponentiation operator. It raises the first operand to the power
11569 The range operator. Normally used in the form of array(low:high) to
11570 represent a section of array.
11573 The access component operator. Normally used to access elements in derived
11574 types. Also suitable for unions. As unions aren't part of regular Fortran,
11575 this can only happen when accessing a register that uses a gdbarch-defined
11579 @node Fortran Defaults
11580 @subsubsection Fortran Defaults
11582 @cindex Fortran Defaults
11584 Fortran symbols are usually case-insensitive, so @value{GDBN} by
11585 default uses case-insensitive matches for Fortran symbols. You can
11586 change that with the @samp{set case-insensitive} command, see
11587 @ref{Symbols}, for the details.
11589 @node Special Fortran Commands
11590 @subsubsection Special Fortran Commands
11592 @cindex Special Fortran commands
11594 @value{GDBN} has some commands to support Fortran-specific features,
11595 such as displaying common blocks.
11598 @cindex @code{COMMON} blocks, Fortran
11599 @kindex info common
11600 @item info common @r{[}@var{common-name}@r{]}
11601 This command prints the values contained in the Fortran @code{COMMON}
11602 block whose name is @var{common-name}. With no argument, the names of
11603 all @code{COMMON} blocks visible at the current program location are
11610 @cindex Pascal support in @value{GDBN}, limitations
11611 Debugging Pascal programs which use sets, subranges, file variables, or
11612 nested functions does not currently work. @value{GDBN} does not support
11613 entering expressions, printing values, or similar features using Pascal
11616 The Pascal-specific command @code{set print pascal_static-members}
11617 controls whether static members of Pascal objects are displayed.
11618 @xref{Print Settings, pascal_static-members}.
11621 @subsection Modula-2
11623 @cindex Modula-2, @value{GDBN} support
11625 The extensions made to @value{GDBN} to support Modula-2 only support
11626 output from the @sc{gnu} Modula-2 compiler (which is currently being
11627 developed). Other Modula-2 compilers are not currently supported, and
11628 attempting to debug executables produced by them is most likely
11629 to give an error as @value{GDBN} reads in the executable's symbol
11632 @cindex expressions in Modula-2
11634 * M2 Operators:: Built-in operators
11635 * Built-In Func/Proc:: Built-in functions and procedures
11636 * M2 Constants:: Modula-2 constants
11637 * M2 Types:: Modula-2 types
11638 * M2 Defaults:: Default settings for Modula-2
11639 * Deviations:: Deviations from standard Modula-2
11640 * M2 Checks:: Modula-2 type and range checks
11641 * M2 Scope:: The scope operators @code{::} and @code{.}
11642 * GDB/M2:: @value{GDBN} and Modula-2
11646 @subsubsection Operators
11647 @cindex Modula-2 operators
11649 Operators must be defined on values of specific types. For instance,
11650 @code{+} is defined on numbers, but not on structures. Operators are
11651 often defined on groups of types. For the purposes of Modula-2, the
11652 following definitions hold:
11657 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
11661 @emph{Character types} consist of @code{CHAR} and its subranges.
11664 @emph{Floating-point types} consist of @code{REAL}.
11667 @emph{Pointer types} consist of anything declared as @code{POINTER TO
11671 @emph{Scalar types} consist of all of the above.
11674 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
11677 @emph{Boolean types} consist of @code{BOOLEAN}.
11681 The following operators are supported, and appear in order of
11682 increasing precedence:
11686 Function argument or array index separator.
11689 Assignment. The value of @var{var} @code{:=} @var{value} is
11693 Less than, greater than on integral, floating-point, or enumerated
11697 Less than or equal to, greater than or equal to
11698 on integral, floating-point and enumerated types, or set inclusion on
11699 set types. Same precedence as @code{<}.
11701 @item =@r{, }<>@r{, }#
11702 Equality and two ways of expressing inequality, valid on scalar types.
11703 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
11704 available for inequality, since @code{#} conflicts with the script
11708 Set membership. Defined on set types and the types of their members.
11709 Same precedence as @code{<}.
11712 Boolean disjunction. Defined on boolean types.
11715 Boolean conjunction. Defined on boolean types.
11718 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11721 Addition and subtraction on integral and floating-point types, or union
11722 and difference on set types.
11725 Multiplication on integral and floating-point types, or set intersection
11729 Division on floating-point types, or symmetric set difference on set
11730 types. Same precedence as @code{*}.
11733 Integer division and remainder. Defined on integral types. Same
11734 precedence as @code{*}.
11737 Negative. Defined on @code{INTEGER} and @code{REAL} data.
11740 Pointer dereferencing. Defined on pointer types.
11743 Boolean negation. Defined on boolean types. Same precedence as
11747 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
11748 precedence as @code{^}.
11751 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
11754 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
11758 @value{GDBN} and Modula-2 scope operators.
11762 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
11763 treats the use of the operator @code{IN}, or the use of operators
11764 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
11765 @code{<=}, and @code{>=} on sets as an error.
11769 @node Built-In Func/Proc
11770 @subsubsection Built-in Functions and Procedures
11771 @cindex Modula-2 built-ins
11773 Modula-2 also makes available several built-in procedures and functions.
11774 In describing these, the following metavariables are used:
11779 represents an @code{ARRAY} variable.
11782 represents a @code{CHAR} constant or variable.
11785 represents a variable or constant of integral type.
11788 represents an identifier that belongs to a set. Generally used in the
11789 same function with the metavariable @var{s}. The type of @var{s} should
11790 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
11793 represents a variable or constant of integral or floating-point type.
11796 represents a variable or constant of floating-point type.
11802 represents a variable.
11805 represents a variable or constant of one of many types. See the
11806 explanation of the function for details.
11809 All Modula-2 built-in procedures also return a result, described below.
11813 Returns the absolute value of @var{n}.
11816 If @var{c} is a lower case letter, it returns its upper case
11817 equivalent, otherwise it returns its argument.
11820 Returns the character whose ordinal value is @var{i}.
11823 Decrements the value in the variable @var{v} by one. Returns the new value.
11825 @item DEC(@var{v},@var{i})
11826 Decrements the value in the variable @var{v} by @var{i}. Returns the
11829 @item EXCL(@var{m},@var{s})
11830 Removes the element @var{m} from the set @var{s}. Returns the new
11833 @item FLOAT(@var{i})
11834 Returns the floating point equivalent of the integer @var{i}.
11836 @item HIGH(@var{a})
11837 Returns the index of the last member of @var{a}.
11840 Increments the value in the variable @var{v} by one. Returns the new value.
11842 @item INC(@var{v},@var{i})
11843 Increments the value in the variable @var{v} by @var{i}. Returns the
11846 @item INCL(@var{m},@var{s})
11847 Adds the element @var{m} to the set @var{s} if it is not already
11848 there. Returns the new set.
11851 Returns the maximum value of the type @var{t}.
11854 Returns the minimum value of the type @var{t}.
11857 Returns boolean TRUE if @var{i} is an odd number.
11860 Returns the ordinal value of its argument. For example, the ordinal
11861 value of a character is its @sc{ascii} value (on machines supporting the
11862 @sc{ascii} character set). @var{x} must be of an ordered type, which include
11863 integral, character and enumerated types.
11865 @item SIZE(@var{x})
11866 Returns the size of its argument. @var{x} can be a variable or a type.
11868 @item TRUNC(@var{r})
11869 Returns the integral part of @var{r}.
11871 @item TSIZE(@var{x})
11872 Returns the size of its argument. @var{x} can be a variable or a type.
11874 @item VAL(@var{t},@var{i})
11875 Returns the member of the type @var{t} whose ordinal value is @var{i}.
11879 @emph{Warning:} Sets and their operations are not yet supported, so
11880 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
11884 @cindex Modula-2 constants
11886 @subsubsection Constants
11888 @value{GDBN} allows you to express the constants of Modula-2 in the following
11894 Integer constants are simply a sequence of digits. When used in an
11895 expression, a constant is interpreted to be type-compatible with the
11896 rest of the expression. Hexadecimal integers are specified by a
11897 trailing @samp{H}, and octal integers by a trailing @samp{B}.
11900 Floating point constants appear as a sequence of digits, followed by a
11901 decimal point and another sequence of digits. An optional exponent can
11902 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
11903 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
11904 digits of the floating point constant must be valid decimal (base 10)
11908 Character constants consist of a single character enclosed by a pair of
11909 like quotes, either single (@code{'}) or double (@code{"}). They may
11910 also be expressed by their ordinal value (their @sc{ascii} value, usually)
11911 followed by a @samp{C}.
11914 String constants consist of a sequence of characters enclosed by a
11915 pair of like quotes, either single (@code{'}) or double (@code{"}).
11916 Escape sequences in the style of C are also allowed. @xref{C
11917 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
11921 Enumerated constants consist of an enumerated identifier.
11924 Boolean constants consist of the identifiers @code{TRUE} and
11928 Pointer constants consist of integral values only.
11931 Set constants are not yet supported.
11935 @subsubsection Modula-2 Types
11936 @cindex Modula-2 types
11938 Currently @value{GDBN} can print the following data types in Modula-2
11939 syntax: array types, record types, set types, pointer types, procedure
11940 types, enumerated types, subrange types and base types. You can also
11941 print the contents of variables declared using these type.
11942 This section gives a number of simple source code examples together with
11943 sample @value{GDBN} sessions.
11945 The first example contains the following section of code:
11954 and you can request @value{GDBN} to interrogate the type and value of
11955 @code{r} and @code{s}.
11958 (@value{GDBP}) print s
11960 (@value{GDBP}) ptype s
11962 (@value{GDBP}) print r
11964 (@value{GDBP}) ptype r
11969 Likewise if your source code declares @code{s} as:
11973 s: SET ['A'..'Z'] ;
11977 then you may query the type of @code{s} by:
11980 (@value{GDBP}) ptype s
11981 type = SET ['A'..'Z']
11985 Note that at present you cannot interactively manipulate set
11986 expressions using the debugger.
11988 The following example shows how you might declare an array in Modula-2
11989 and how you can interact with @value{GDBN} to print its type and contents:
11993 s: ARRAY [-10..10] OF CHAR ;
11997 (@value{GDBP}) ptype s
11998 ARRAY [-10..10] OF CHAR
12001 Note that the array handling is not yet complete and although the type
12002 is printed correctly, expression handling still assumes that all
12003 arrays have a lower bound of zero and not @code{-10} as in the example
12006 Here are some more type related Modula-2 examples:
12010 colour = (blue, red, yellow, green) ;
12011 t = [blue..yellow] ;
12019 The @value{GDBN} interaction shows how you can query the data type
12020 and value of a variable.
12023 (@value{GDBP}) print s
12025 (@value{GDBP}) ptype t
12026 type = [blue..yellow]
12030 In this example a Modula-2 array is declared and its contents
12031 displayed. Observe that the contents are written in the same way as
12032 their @code{C} counterparts.
12036 s: ARRAY [1..5] OF CARDINAL ;
12042 (@value{GDBP}) print s
12043 $1 = @{1, 0, 0, 0, 0@}
12044 (@value{GDBP}) ptype s
12045 type = ARRAY [1..5] OF CARDINAL
12048 The Modula-2 language interface to @value{GDBN} also understands
12049 pointer types as shown in this example:
12053 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
12060 and you can request that @value{GDBN} describes the type of @code{s}.
12063 (@value{GDBP}) ptype s
12064 type = POINTER TO ARRAY [1..5] OF CARDINAL
12067 @value{GDBN} handles compound types as we can see in this example.
12068 Here we combine array types, record types, pointer types and subrange
12079 myarray = ARRAY myrange OF CARDINAL ;
12080 myrange = [-2..2] ;
12082 s: POINTER TO ARRAY myrange OF foo ;
12086 and you can ask @value{GDBN} to describe the type of @code{s} as shown
12090 (@value{GDBP}) ptype s
12091 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
12094 f3 : ARRAY [-2..2] OF CARDINAL;
12099 @subsubsection Modula-2 Defaults
12100 @cindex Modula-2 defaults
12102 If type and range checking are set automatically by @value{GDBN}, they
12103 both default to @code{on} whenever the working language changes to
12104 Modula-2. This happens regardless of whether you or @value{GDBN}
12105 selected the working language.
12107 If you allow @value{GDBN} to set the language automatically, then entering
12108 code compiled from a file whose name ends with @file{.mod} sets the
12109 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
12110 Infer the Source Language}, for further details.
12113 @subsubsection Deviations from Standard Modula-2
12114 @cindex Modula-2, deviations from
12116 A few changes have been made to make Modula-2 programs easier to debug.
12117 This is done primarily via loosening its type strictness:
12121 Unlike in standard Modula-2, pointer constants can be formed by
12122 integers. This allows you to modify pointer variables during
12123 debugging. (In standard Modula-2, the actual address contained in a
12124 pointer variable is hidden from you; it can only be modified
12125 through direct assignment to another pointer variable or expression that
12126 returned a pointer.)
12129 C escape sequences can be used in strings and characters to represent
12130 non-printable characters. @value{GDBN} prints out strings with these
12131 escape sequences embedded. Single non-printable characters are
12132 printed using the @samp{CHR(@var{nnn})} format.
12135 The assignment operator (@code{:=}) returns the value of its right-hand
12139 All built-in procedures both modify @emph{and} return their argument.
12143 @subsubsection Modula-2 Type and Range Checks
12144 @cindex Modula-2 checks
12147 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
12150 @c FIXME remove warning when type/range checks added
12152 @value{GDBN} considers two Modula-2 variables type equivalent if:
12156 They are of types that have been declared equivalent via a @code{TYPE
12157 @var{t1} = @var{t2}} statement
12160 They have been declared on the same line. (Note: This is true of the
12161 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
12164 As long as type checking is enabled, any attempt to combine variables
12165 whose types are not equivalent is an error.
12167 Range checking is done on all mathematical operations, assignment, array
12168 index bounds, and all built-in functions and procedures.
12171 @subsubsection The Scope Operators @code{::} and @code{.}
12173 @cindex @code{.}, Modula-2 scope operator
12174 @cindex colon, doubled as scope operator
12176 @vindex colon-colon@r{, in Modula-2}
12177 @c Info cannot handle :: but TeX can.
12180 @vindex ::@r{, in Modula-2}
12183 There are a few subtle differences between the Modula-2 scope operator
12184 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
12189 @var{module} . @var{id}
12190 @var{scope} :: @var{id}
12194 where @var{scope} is the name of a module or a procedure,
12195 @var{module} the name of a module, and @var{id} is any declared
12196 identifier within your program, except another module.
12198 Using the @code{::} operator makes @value{GDBN} search the scope
12199 specified by @var{scope} for the identifier @var{id}. If it is not
12200 found in the specified scope, then @value{GDBN} searches all scopes
12201 enclosing the one specified by @var{scope}.
12203 Using the @code{.} operator makes @value{GDBN} search the current scope for
12204 the identifier specified by @var{id} that was imported from the
12205 definition module specified by @var{module}. With this operator, it is
12206 an error if the identifier @var{id} was not imported from definition
12207 module @var{module}, or if @var{id} is not an identifier in
12211 @subsubsection @value{GDBN} and Modula-2
12213 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
12214 Five subcommands of @code{set print} and @code{show print} apply
12215 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
12216 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
12217 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
12218 analogue in Modula-2.
12220 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
12221 with any language, is not useful with Modula-2. Its
12222 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
12223 created in Modula-2 as they can in C or C@t{++}. However, because an
12224 address can be specified by an integral constant, the construct
12225 @samp{@{@var{type}@}@var{adrexp}} is still useful.
12227 @cindex @code{#} in Modula-2
12228 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
12229 interpreted as the beginning of a comment. Use @code{<>} instead.
12235 The extensions made to @value{GDBN} for Ada only support
12236 output from the @sc{gnu} Ada (GNAT) compiler.
12237 Other Ada compilers are not currently supported, and
12238 attempting to debug executables produced by them is most likely
12242 @cindex expressions in Ada
12244 * Ada Mode Intro:: General remarks on the Ada syntax
12245 and semantics supported by Ada mode
12247 * Omissions from Ada:: Restrictions on the Ada expression syntax.
12248 * Additions to Ada:: Extensions of the Ada expression syntax.
12249 * Stopping Before Main Program:: Debugging the program during elaboration.
12250 * Ada Tasks:: Listing and setting breakpoints in tasks.
12251 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
12252 * Ada Glitches:: Known peculiarities of Ada mode.
12255 @node Ada Mode Intro
12256 @subsubsection Introduction
12257 @cindex Ada mode, general
12259 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
12260 syntax, with some extensions.
12261 The philosophy behind the design of this subset is
12265 That @value{GDBN} should provide basic literals and access to operations for
12266 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
12267 leaving more sophisticated computations to subprograms written into the
12268 program (which therefore may be called from @value{GDBN}).
12271 That type safety and strict adherence to Ada language restrictions
12272 are not particularly important to the @value{GDBN} user.
12275 That brevity is important to the @value{GDBN} user.
12278 Thus, for brevity, the debugger acts as if all names declared in
12279 user-written packages are directly visible, even if they are not visible
12280 according to Ada rules, thus making it unnecessary to fully qualify most
12281 names with their packages, regardless of context. Where this causes
12282 ambiguity, @value{GDBN} asks the user's intent.
12284 The debugger will start in Ada mode if it detects an Ada main program.
12285 As for other languages, it will enter Ada mode when stopped in a program that
12286 was translated from an Ada source file.
12288 While in Ada mode, you may use `@t{--}' for comments. This is useful
12289 mostly for documenting command files. The standard @value{GDBN} comment
12290 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
12291 middle (to allow based literals).
12293 The debugger supports limited overloading. Given a subprogram call in which
12294 the function symbol has multiple definitions, it will use the number of
12295 actual parameters and some information about their types to attempt to narrow
12296 the set of definitions. It also makes very limited use of context, preferring
12297 procedures to functions in the context of the @code{call} command, and
12298 functions to procedures elsewhere.
12300 @node Omissions from Ada
12301 @subsubsection Omissions from Ada
12302 @cindex Ada, omissions from
12304 Here are the notable omissions from the subset:
12308 Only a subset of the attributes are supported:
12312 @t{'First}, @t{'Last}, and @t{'Length}
12313 on array objects (not on types and subtypes).
12316 @t{'Min} and @t{'Max}.
12319 @t{'Pos} and @t{'Val}.
12325 @t{'Range} on array objects (not subtypes), but only as the right
12326 operand of the membership (@code{in}) operator.
12329 @t{'Access}, @t{'Unchecked_Access}, and
12330 @t{'Unrestricted_Access} (a GNAT extension).
12338 @code{Characters.Latin_1} are not available and
12339 concatenation is not implemented. Thus, escape characters in strings are
12340 not currently available.
12343 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
12344 equality of representations. They will generally work correctly
12345 for strings and arrays whose elements have integer or enumeration types.
12346 They may not work correctly for arrays whose element
12347 types have user-defined equality, for arrays of real values
12348 (in particular, IEEE-conformant floating point, because of negative
12349 zeroes and NaNs), and for arrays whose elements contain unused bits with
12350 indeterminate values.
12353 The other component-by-component array operations (@code{and}, @code{or},
12354 @code{xor}, @code{not}, and relational tests other than equality)
12355 are not implemented.
12358 @cindex array aggregates (Ada)
12359 @cindex record aggregates (Ada)
12360 @cindex aggregates (Ada)
12361 There is limited support for array and record aggregates. They are
12362 permitted only on the right sides of assignments, as in these examples:
12365 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
12366 (@value{GDBP}) set An_Array := (1, others => 0)
12367 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
12368 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
12369 (@value{GDBP}) set A_Record := (1, "Peter", True);
12370 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
12374 discriminant's value by assigning an aggregate has an
12375 undefined effect if that discriminant is used within the record.
12376 However, you can first modify discriminants by directly assigning to
12377 them (which normally would not be allowed in Ada), and then performing an
12378 aggregate assignment. For example, given a variable @code{A_Rec}
12379 declared to have a type such as:
12382 type Rec (Len : Small_Integer := 0) is record
12384 Vals : IntArray (1 .. Len);
12388 you can assign a value with a different size of @code{Vals} with two
12392 (@value{GDBP}) set A_Rec.Len := 4
12393 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
12396 As this example also illustrates, @value{GDBN} is very loose about the usual
12397 rules concerning aggregates. You may leave out some of the
12398 components of an array or record aggregate (such as the @code{Len}
12399 component in the assignment to @code{A_Rec} above); they will retain their
12400 original values upon assignment. You may freely use dynamic values as
12401 indices in component associations. You may even use overlapping or
12402 redundant component associations, although which component values are
12403 assigned in such cases is not defined.
12406 Calls to dispatching subprograms are not implemented.
12409 The overloading algorithm is much more limited (i.e., less selective)
12410 than that of real Ada. It makes only limited use of the context in
12411 which a subexpression appears to resolve its meaning, and it is much
12412 looser in its rules for allowing type matches. As a result, some
12413 function calls will be ambiguous, and the user will be asked to choose
12414 the proper resolution.
12417 The @code{new} operator is not implemented.
12420 Entry calls are not implemented.
12423 Aside from printing, arithmetic operations on the native VAX floating-point
12424 formats are not supported.
12427 It is not possible to slice a packed array.
12430 The names @code{True} and @code{False}, when not part of a qualified name,
12431 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
12433 Should your program
12434 redefine these names in a package or procedure (at best a dubious practice),
12435 you will have to use fully qualified names to access their new definitions.
12438 @node Additions to Ada
12439 @subsubsection Additions to Ada
12440 @cindex Ada, deviations from
12442 As it does for other languages, @value{GDBN} makes certain generic
12443 extensions to Ada (@pxref{Expressions}):
12447 If the expression @var{E} is a variable residing in memory (typically
12448 a local variable or array element) and @var{N} is a positive integer,
12449 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
12450 @var{N}-1 adjacent variables following it in memory as an array. In
12451 Ada, this operator is generally not necessary, since its prime use is
12452 in displaying parts of an array, and slicing will usually do this in
12453 Ada. However, there are occasional uses when debugging programs in
12454 which certain debugging information has been optimized away.
12457 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
12458 appears in function or file @var{B}.'' When @var{B} is a file name,
12459 you must typically surround it in single quotes.
12462 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
12463 @var{type} that appears at address @var{addr}.''
12466 A name starting with @samp{$} is a convenience variable
12467 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
12470 In addition, @value{GDBN} provides a few other shortcuts and outright
12471 additions specific to Ada:
12475 The assignment statement is allowed as an expression, returning
12476 its right-hand operand as its value. Thus, you may enter
12479 (@value{GDBP}) set x := y + 3
12480 (@value{GDBP}) print A(tmp := y + 1)
12484 The semicolon is allowed as an ``operator,'' returning as its value
12485 the value of its right-hand operand.
12486 This allows, for example,
12487 complex conditional breaks:
12490 (@value{GDBP}) break f
12491 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
12495 Rather than use catenation and symbolic character names to introduce special
12496 characters into strings, one may instead use a special bracket notation,
12497 which is also used to print strings. A sequence of characters of the form
12498 @samp{["@var{XX}"]} within a string or character literal denotes the
12499 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
12500 sequence of characters @samp{["""]} also denotes a single quotation mark
12501 in strings. For example,
12503 "One line.["0a"]Next line.["0a"]"
12506 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
12510 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
12511 @t{'Max} is optional (and is ignored in any case). For example, it is valid
12515 (@value{GDBP}) print 'max(x, y)
12519 When printing arrays, @value{GDBN} uses positional notation when the
12520 array has a lower bound of 1, and uses a modified named notation otherwise.
12521 For example, a one-dimensional array of three integers with a lower bound
12522 of 3 might print as
12529 That is, in contrast to valid Ada, only the first component has a @code{=>}
12533 You may abbreviate attributes in expressions with any unique,
12534 multi-character subsequence of
12535 their names (an exact match gets preference).
12536 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
12537 in place of @t{a'length}.
12540 @cindex quoting Ada internal identifiers
12541 Since Ada is case-insensitive, the debugger normally maps identifiers you type
12542 to lower case. The GNAT compiler uses upper-case characters for
12543 some of its internal identifiers, which are normally of no interest to users.
12544 For the rare occasions when you actually have to look at them,
12545 enclose them in angle brackets to avoid the lower-case mapping.
12548 (@value{GDBP}) print <JMPBUF_SAVE>[0]
12552 Printing an object of class-wide type or dereferencing an
12553 access-to-class-wide value will display all the components of the object's
12554 specific type (as indicated by its run-time tag). Likewise, component
12555 selection on such a value will operate on the specific type of the
12560 @node Stopping Before Main Program
12561 @subsubsection Stopping at the Very Beginning
12563 @cindex breakpointing Ada elaboration code
12564 It is sometimes necessary to debug the program during elaboration, and
12565 before reaching the main procedure.
12566 As defined in the Ada Reference
12567 Manual, the elaboration code is invoked from a procedure called
12568 @code{adainit}. To run your program up to the beginning of
12569 elaboration, simply use the following two commands:
12570 @code{tbreak adainit} and @code{run}.
12573 @subsubsection Extensions for Ada Tasks
12574 @cindex Ada, tasking
12576 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
12577 @value{GDBN} provides the following task-related commands:
12582 This command shows a list of current Ada tasks, as in the following example:
12589 (@value{GDBP}) info tasks
12590 ID TID P-ID Pri State Name
12591 1 8088000 0 15 Child Activation Wait main_task
12592 2 80a4000 1 15 Accept Statement b
12593 3 809a800 1 15 Child Activation Wait a
12594 * 4 80ae800 3 15 Runnable c
12599 In this listing, the asterisk before the last task indicates it to be the
12600 task currently being inspected.
12604 Represents @value{GDBN}'s internal task number.
12610 The parent's task ID (@value{GDBN}'s internal task number).
12613 The base priority of the task.
12616 Current state of the task.
12620 The task has been created but has not been activated. It cannot be
12624 The task is not blocked for any reason known to Ada. (It may be waiting
12625 for a mutex, though.) It is conceptually "executing" in normal mode.
12628 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
12629 that were waiting on terminate alternatives have been awakened and have
12630 terminated themselves.
12632 @item Child Activation Wait
12633 The task is waiting for created tasks to complete activation.
12635 @item Accept Statement
12636 The task is waiting on an accept or selective wait statement.
12638 @item Waiting on entry call
12639 The task is waiting on an entry call.
12641 @item Async Select Wait
12642 The task is waiting to start the abortable part of an asynchronous
12646 The task is waiting on a select statement with only a delay
12649 @item Child Termination Wait
12650 The task is sleeping having completed a master within itself, and is
12651 waiting for the tasks dependent on that master to become terminated or
12652 waiting on a terminate Phase.
12654 @item Wait Child in Term Alt
12655 The task is sleeping waiting for tasks on terminate alternatives to
12656 finish terminating.
12658 @item Accepting RV with @var{taskno}
12659 The task is accepting a rendez-vous with the task @var{taskno}.
12663 Name of the task in the program.
12667 @kindex info task @var{taskno}
12668 @item info task @var{taskno}
12669 This command shows detailled informations on the specified task, as in
12670 the following example:
12675 (@value{GDBP}) info tasks
12676 ID TID P-ID Pri State Name
12677 1 8077880 0 15 Child Activation Wait main_task
12678 * 2 807c468 1 15 Runnable task_1
12679 (@value{GDBP}) info task 2
12680 Ada Task: 0x807c468
12683 Parent: 1 (main_task)
12689 @kindex task@r{ (Ada)}
12690 @cindex current Ada task ID
12691 This command prints the ID of the current task.
12697 (@value{GDBP}) info tasks
12698 ID TID P-ID Pri State Name
12699 1 8077870 0 15 Child Activation Wait main_task
12700 * 2 807c458 1 15 Runnable t
12701 (@value{GDBP}) task
12702 [Current task is 2]
12705 @item task @var{taskno}
12706 @cindex Ada task switching
12707 This command is like the @code{thread @var{threadno}}
12708 command (@pxref{Threads}). It switches the context of debugging
12709 from the current task to the given task.
12715 (@value{GDBP}) info tasks
12716 ID TID P-ID Pri State Name
12717 1 8077870 0 15 Child Activation Wait main_task
12718 * 2 807c458 1 15 Runnable t
12719 (@value{GDBP}) task 1
12720 [Switching to task 1]
12721 #0 0x8067726 in pthread_cond_wait ()
12723 #0 0x8067726 in pthread_cond_wait ()
12724 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
12725 #2 0x805cb63 in system.task_primitives.operations.sleep ()
12726 #3 0x806153e in system.tasking.stages.activate_tasks ()
12727 #4 0x804aacc in un () at un.adb:5
12730 @item break @var{linespec} task @var{taskno}
12731 @itemx break @var{linespec} task @var{taskno} if @dots{}
12732 @cindex breakpoints and tasks, in Ada
12733 @cindex task breakpoints, in Ada
12734 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
12735 These commands are like the @code{break @dots{} thread @dots{}}
12736 command (@pxref{Thread Stops}).
12737 @var{linespec} specifies source lines, as described
12738 in @ref{Specify Location}.
12740 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
12741 to specify that you only want @value{GDBN} to stop the program when a
12742 particular Ada task reaches this breakpoint. @var{taskno} is one of the
12743 numeric task identifiers assigned by @value{GDBN}, shown in the first
12744 column of the @samp{info tasks} display.
12746 If you do not specify @samp{task @var{taskno}} when you set a
12747 breakpoint, the breakpoint applies to @emph{all} tasks of your
12750 You can use the @code{task} qualifier on conditional breakpoints as
12751 well; in this case, place @samp{task @var{taskno}} before the
12752 breakpoint condition (before the @code{if}).
12760 (@value{GDBP}) info tasks
12761 ID TID P-ID Pri State Name
12762 1 140022020 0 15 Child Activation Wait main_task
12763 2 140045060 1 15 Accept/Select Wait t2
12764 3 140044840 1 15 Runnable t1
12765 * 4 140056040 1 15 Runnable t3
12766 (@value{GDBP}) b 15 task 2
12767 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
12768 (@value{GDBP}) cont
12773 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
12775 (@value{GDBP}) info tasks
12776 ID TID P-ID Pri State Name
12777 1 140022020 0 15 Child Activation Wait main_task
12778 * 2 140045060 1 15 Runnable t2
12779 3 140044840 1 15 Runnable t1
12780 4 140056040 1 15 Delay Sleep t3
12784 @node Ada Tasks and Core Files
12785 @subsubsection Tasking Support when Debugging Core Files
12786 @cindex Ada tasking and core file debugging
12788 When inspecting a core file, as opposed to debugging a live program,
12789 tasking support may be limited or even unavailable, depending on
12790 the platform being used.
12791 For instance, on x86-linux, the list of tasks is available, but task
12792 switching is not supported. On Tru64, however, task switching will work
12795 On certain platforms, including Tru64, the debugger needs to perform some
12796 memory writes in order to provide Ada tasking support. When inspecting
12797 a core file, this means that the core file must be opened with read-write
12798 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
12799 Under these circumstances, you should make a backup copy of the core
12800 file before inspecting it with @value{GDBN}.
12803 @subsubsection Known Peculiarities of Ada Mode
12804 @cindex Ada, problems
12806 Besides the omissions listed previously (@pxref{Omissions from Ada}),
12807 we know of several problems with and limitations of Ada mode in
12809 some of which will be fixed with planned future releases of the debugger
12810 and the GNU Ada compiler.
12814 Currently, the debugger
12815 has insufficient information to determine whether certain pointers represent
12816 pointers to objects or the objects themselves.
12817 Thus, the user may have to tack an extra @code{.all} after an expression
12818 to get it printed properly.
12821 Static constants that the compiler chooses not to materialize as objects in
12822 storage are invisible to the debugger.
12825 Named parameter associations in function argument lists are ignored (the
12826 argument lists are treated as positional).
12829 Many useful library packages are currently invisible to the debugger.
12832 Fixed-point arithmetic, conversions, input, and output is carried out using
12833 floating-point arithmetic, and may give results that only approximate those on
12837 The GNAT compiler never generates the prefix @code{Standard} for any of
12838 the standard symbols defined by the Ada language. @value{GDBN} knows about
12839 this: it will strip the prefix from names when you use it, and will never
12840 look for a name you have so qualified among local symbols, nor match against
12841 symbols in other packages or subprograms. If you have
12842 defined entities anywhere in your program other than parameters and
12843 local variables whose simple names match names in @code{Standard},
12844 GNAT's lack of qualification here can cause confusion. When this happens,
12845 you can usually resolve the confusion
12846 by qualifying the problematic names with package
12847 @code{Standard} explicitly.
12850 Older versions of the compiler sometimes generate erroneous debugging
12851 information, resulting in the debugger incorrectly printing the value
12852 of affected entities. In some cases, the debugger is able to work
12853 around an issue automatically. In other cases, the debugger is able
12854 to work around the issue, but the work-around has to be specifically
12857 @kindex set ada trust-PAD-over-XVS
12858 @kindex show ada trust-PAD-over-XVS
12861 @item set ada trust-PAD-over-XVS on
12862 Configure GDB to strictly follow the GNAT encoding when computing the
12863 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
12864 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
12865 a complete description of the encoding used by the GNAT compiler).
12866 This is the default.
12868 @item set ada trust-PAD-over-XVS off
12869 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
12870 sometimes prints the wrong value for certain entities, changing @code{ada
12871 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
12872 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
12873 @code{off}, but this incurs a slight performance penalty, so it is
12874 recommended to leave this setting to @code{on} unless necessary.
12878 @node Unsupported Languages
12879 @section Unsupported Languages
12881 @cindex unsupported languages
12882 @cindex minimal language
12883 In addition to the other fully-supported programming languages,
12884 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
12885 It does not represent a real programming language, but provides a set
12886 of capabilities close to what the C or assembly languages provide.
12887 This should allow most simple operations to be performed while debugging
12888 an application that uses a language currently not supported by @value{GDBN}.
12890 If the language is set to @code{auto}, @value{GDBN} will automatically
12891 select this language if the current frame corresponds to an unsupported
12895 @chapter Examining the Symbol Table
12897 The commands described in this chapter allow you to inquire about the
12898 symbols (names of variables, functions and types) defined in your
12899 program. This information is inherent in the text of your program and
12900 does not change as your program executes. @value{GDBN} finds it in your
12901 program's symbol table, in the file indicated when you started @value{GDBN}
12902 (@pxref{File Options, ,Choosing Files}), or by one of the
12903 file-management commands (@pxref{Files, ,Commands to Specify Files}).
12905 @cindex symbol names
12906 @cindex names of symbols
12907 @cindex quoting names
12908 Occasionally, you may need to refer to symbols that contain unusual
12909 characters, which @value{GDBN} ordinarily treats as word delimiters. The
12910 most frequent case is in referring to static variables in other
12911 source files (@pxref{Variables,,Program Variables}). File names
12912 are recorded in object files as debugging symbols, but @value{GDBN} would
12913 ordinarily parse a typical file name, like @file{foo.c}, as the three words
12914 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
12915 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
12922 looks up the value of @code{x} in the scope of the file @file{foo.c}.
12925 @cindex case-insensitive symbol names
12926 @cindex case sensitivity in symbol names
12927 @kindex set case-sensitive
12928 @item set case-sensitive on
12929 @itemx set case-sensitive off
12930 @itemx set case-sensitive auto
12931 Normally, when @value{GDBN} looks up symbols, it matches their names
12932 with case sensitivity determined by the current source language.
12933 Occasionally, you may wish to control that. The command @code{set
12934 case-sensitive} lets you do that by specifying @code{on} for
12935 case-sensitive matches or @code{off} for case-insensitive ones. If
12936 you specify @code{auto}, case sensitivity is reset to the default
12937 suitable for the source language. The default is case-sensitive
12938 matches for all languages except for Fortran, for which the default is
12939 case-insensitive matches.
12941 @kindex show case-sensitive
12942 @item show case-sensitive
12943 This command shows the current setting of case sensitivity for symbols
12946 @kindex info address
12947 @cindex address of a symbol
12948 @item info address @var{symbol}
12949 Describe where the data for @var{symbol} is stored. For a register
12950 variable, this says which register it is kept in. For a non-register
12951 local variable, this prints the stack-frame offset at which the variable
12954 Note the contrast with @samp{print &@var{symbol}}, which does not work
12955 at all for a register variable, and for a stack local variable prints
12956 the exact address of the current instantiation of the variable.
12958 @kindex info symbol
12959 @cindex symbol from address
12960 @cindex closest symbol and offset for an address
12961 @item info symbol @var{addr}
12962 Print the name of a symbol which is stored at the address @var{addr}.
12963 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
12964 nearest symbol and an offset from it:
12967 (@value{GDBP}) info symbol 0x54320
12968 _initialize_vx + 396 in section .text
12972 This is the opposite of the @code{info address} command. You can use
12973 it to find out the name of a variable or a function given its address.
12975 For dynamically linked executables, the name of executable or shared
12976 library containing the symbol is also printed:
12979 (@value{GDBP}) info symbol 0x400225
12980 _start + 5 in section .text of /tmp/a.out
12981 (@value{GDBP}) info symbol 0x2aaaac2811cf
12982 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
12986 @item whatis [@var{arg}]
12987 Print the data type of @var{arg}, which can be either an expression or
12988 a data type. With no argument, print the data type of @code{$}, the
12989 last value in the value history. If @var{arg} is an expression, it is
12990 not actually evaluated, and any side-effecting operations (such as
12991 assignments or function calls) inside it do not take place. If
12992 @var{arg} is a type name, it may be the name of a type or typedef, or
12993 for C code it may have the form @samp{class @var{class-name}},
12994 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
12995 @samp{enum @var{enum-tag}}.
12996 @xref{Expressions, ,Expressions}.
12999 @item ptype [@var{arg}]
13000 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
13001 detailed description of the type, instead of just the name of the type.
13002 @xref{Expressions, ,Expressions}.
13004 For example, for this variable declaration:
13007 struct complex @{double real; double imag;@} v;
13011 the two commands give this output:
13015 (@value{GDBP}) whatis v
13016 type = struct complex
13017 (@value{GDBP}) ptype v
13018 type = struct complex @{
13026 As with @code{whatis}, using @code{ptype} without an argument refers to
13027 the type of @code{$}, the last value in the value history.
13029 @cindex incomplete type
13030 Sometimes, programs use opaque data types or incomplete specifications
13031 of complex data structure. If the debug information included in the
13032 program does not allow @value{GDBN} to display a full declaration of
13033 the data type, it will say @samp{<incomplete type>}. For example,
13034 given these declarations:
13038 struct foo *fooptr;
13042 but no definition for @code{struct foo} itself, @value{GDBN} will say:
13045 (@value{GDBP}) ptype foo
13046 $1 = <incomplete type>
13050 ``Incomplete type'' is C terminology for data types that are not
13051 completely specified.
13054 @item info types @var{regexp}
13056 Print a brief description of all types whose names match the regular
13057 expression @var{regexp} (or all types in your program, if you supply
13058 no argument). Each complete typename is matched as though it were a
13059 complete line; thus, @samp{i type value} gives information on all
13060 types in your program whose names include the string @code{value}, but
13061 @samp{i type ^value$} gives information only on types whose complete
13062 name is @code{value}.
13064 This command differs from @code{ptype} in two ways: first, like
13065 @code{whatis}, it does not print a detailed description; second, it
13066 lists all source files where a type is defined.
13069 @cindex local variables
13070 @item info scope @var{location}
13071 List all the variables local to a particular scope. This command
13072 accepts a @var{location} argument---a function name, a source line, or
13073 an address preceded by a @samp{*}, and prints all the variables local
13074 to the scope defined by that location. (@xref{Specify Location}, for
13075 details about supported forms of @var{location}.) For example:
13078 (@value{GDBP}) @b{info scope command_line_handler}
13079 Scope for command_line_handler:
13080 Symbol rl is an argument at stack/frame offset 8, length 4.
13081 Symbol linebuffer is in static storage at address 0x150a18, length 4.
13082 Symbol linelength is in static storage at address 0x150a1c, length 4.
13083 Symbol p is a local variable in register $esi, length 4.
13084 Symbol p1 is a local variable in register $ebx, length 4.
13085 Symbol nline is a local variable in register $edx, length 4.
13086 Symbol repeat is a local variable at frame offset -8, length 4.
13090 This command is especially useful for determining what data to collect
13091 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
13094 @kindex info source
13096 Show information about the current source file---that is, the source file for
13097 the function containing the current point of execution:
13100 the name of the source file, and the directory containing it,
13102 the directory it was compiled in,
13104 its length, in lines,
13106 which programming language it is written in,
13108 whether the executable includes debugging information for that file, and
13109 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
13111 whether the debugging information includes information about
13112 preprocessor macros.
13116 @kindex info sources
13118 Print the names of all source files in your program for which there is
13119 debugging information, organized into two lists: files whose symbols
13120 have already been read, and files whose symbols will be read when needed.
13122 @kindex info functions
13123 @item info functions
13124 Print the names and data types of all defined functions.
13126 @item info functions @var{regexp}
13127 Print the names and data types of all defined functions
13128 whose names contain a match for regular expression @var{regexp}.
13129 Thus, @samp{info fun step} finds all functions whose names
13130 include @code{step}; @samp{info fun ^step} finds those whose names
13131 start with @code{step}. If a function name contains characters
13132 that conflict with the regular expression language (e.g.@:
13133 @samp{operator*()}), they may be quoted with a backslash.
13135 @kindex info variables
13136 @item info variables
13137 Print the names and data types of all variables that are defined
13138 outside of functions (i.e.@: excluding local variables).
13140 @item info variables @var{regexp}
13141 Print the names and data types of all variables (except for local
13142 variables) whose names contain a match for regular expression
13145 @kindex info classes
13146 @cindex Objective-C, classes and selectors
13148 @itemx info classes @var{regexp}
13149 Display all Objective-C classes in your program, or
13150 (with the @var{regexp} argument) all those matching a particular regular
13153 @kindex info selectors
13154 @item info selectors
13155 @itemx info selectors @var{regexp}
13156 Display all Objective-C selectors in your program, or
13157 (with the @var{regexp} argument) all those matching a particular regular
13161 This was never implemented.
13162 @kindex info methods
13164 @itemx info methods @var{regexp}
13165 The @code{info methods} command permits the user to examine all defined
13166 methods within C@t{++} program, or (with the @var{regexp} argument) a
13167 specific set of methods found in the various C@t{++} classes. Many
13168 C@t{++} classes provide a large number of methods. Thus, the output
13169 from the @code{ptype} command can be overwhelming and hard to use. The
13170 @code{info-methods} command filters the methods, printing only those
13171 which match the regular-expression @var{regexp}.
13174 @cindex reloading symbols
13175 Some systems allow individual object files that make up your program to
13176 be replaced without stopping and restarting your program. For example,
13177 in VxWorks you can simply recompile a defective object file and keep on
13178 running. If you are running on one of these systems, you can allow
13179 @value{GDBN} to reload the symbols for automatically relinked modules:
13182 @kindex set symbol-reloading
13183 @item set symbol-reloading on
13184 Replace symbol definitions for the corresponding source file when an
13185 object file with a particular name is seen again.
13187 @item set symbol-reloading off
13188 Do not replace symbol definitions when encountering object files of the
13189 same name more than once. This is the default state; if you are not
13190 running on a system that permits automatic relinking of modules, you
13191 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
13192 may discard symbols when linking large programs, that may contain
13193 several modules (from different directories or libraries) with the same
13196 @kindex show symbol-reloading
13197 @item show symbol-reloading
13198 Show the current @code{on} or @code{off} setting.
13201 @cindex opaque data types
13202 @kindex set opaque-type-resolution
13203 @item set opaque-type-resolution on
13204 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
13205 declared as a pointer to a @code{struct}, @code{class}, or
13206 @code{union}---for example, @code{struct MyType *}---that is used in one
13207 source file although the full declaration of @code{struct MyType} is in
13208 another source file. The default is on.
13210 A change in the setting of this subcommand will not take effect until
13211 the next time symbols for a file are loaded.
13213 @item set opaque-type-resolution off
13214 Tell @value{GDBN} not to resolve opaque types. In this case, the type
13215 is printed as follows:
13217 @{<no data fields>@}
13220 @kindex show opaque-type-resolution
13221 @item show opaque-type-resolution
13222 Show whether opaque types are resolved or not.
13224 @kindex maint print symbols
13225 @cindex symbol dump
13226 @kindex maint print psymbols
13227 @cindex partial symbol dump
13228 @item maint print symbols @var{filename}
13229 @itemx maint print psymbols @var{filename}
13230 @itemx maint print msymbols @var{filename}
13231 Write a dump of debugging symbol data into the file @var{filename}.
13232 These commands are used to debug the @value{GDBN} symbol-reading code. Only
13233 symbols with debugging data are included. If you use @samp{maint print
13234 symbols}, @value{GDBN} includes all the symbols for which it has already
13235 collected full details: that is, @var{filename} reflects symbols for
13236 only those files whose symbols @value{GDBN} has read. You can use the
13237 command @code{info sources} to find out which files these are. If you
13238 use @samp{maint print psymbols} instead, the dump shows information about
13239 symbols that @value{GDBN} only knows partially---that is, symbols defined in
13240 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
13241 @samp{maint print msymbols} dumps just the minimal symbol information
13242 required for each object file from which @value{GDBN} has read some symbols.
13243 @xref{Files, ,Commands to Specify Files}, for a discussion of how
13244 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
13246 @kindex maint info symtabs
13247 @kindex maint info psymtabs
13248 @cindex listing @value{GDBN}'s internal symbol tables
13249 @cindex symbol tables, listing @value{GDBN}'s internal
13250 @cindex full symbol tables, listing @value{GDBN}'s internal
13251 @cindex partial symbol tables, listing @value{GDBN}'s internal
13252 @item maint info symtabs @r{[} @var{regexp} @r{]}
13253 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
13255 List the @code{struct symtab} or @code{struct partial_symtab}
13256 structures whose names match @var{regexp}. If @var{regexp} is not
13257 given, list them all. The output includes expressions which you can
13258 copy into a @value{GDBN} debugging this one to examine a particular
13259 structure in more detail. For example:
13262 (@value{GDBP}) maint info psymtabs dwarf2read
13263 @{ objfile /home/gnu/build/gdb/gdb
13264 ((struct objfile *) 0x82e69d0)
13265 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
13266 ((struct partial_symtab *) 0x8474b10)
13269 text addresses 0x814d3c8 -- 0x8158074
13270 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
13271 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
13272 dependencies (none)
13275 (@value{GDBP}) maint info symtabs
13279 We see that there is one partial symbol table whose filename contains
13280 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
13281 and we see that @value{GDBN} has not read in any symtabs yet at all.
13282 If we set a breakpoint on a function, that will cause @value{GDBN} to
13283 read the symtab for the compilation unit containing that function:
13286 (@value{GDBP}) break dwarf2_psymtab_to_symtab
13287 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
13289 (@value{GDBP}) maint info symtabs
13290 @{ objfile /home/gnu/build/gdb/gdb
13291 ((struct objfile *) 0x82e69d0)
13292 @{ symtab /home/gnu/src/gdb/dwarf2read.c
13293 ((struct symtab *) 0x86c1f38)
13296 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
13297 linetable ((struct linetable *) 0x8370fa0)
13298 debugformat DWARF 2
13307 @chapter Altering Execution
13309 Once you think you have found an error in your program, you might want to
13310 find out for certain whether correcting the apparent error would lead to
13311 correct results in the rest of the run. You can find the answer by
13312 experiment, using the @value{GDBN} features for altering execution of the
13315 For example, you can store new values into variables or memory
13316 locations, give your program a signal, restart it at a different
13317 address, or even return prematurely from a function.
13320 * Assignment:: Assignment to variables
13321 * Jumping:: Continuing at a different address
13322 * Signaling:: Giving your program a signal
13323 * Returning:: Returning from a function
13324 * Calling:: Calling your program's functions
13325 * Patching:: Patching your program
13329 @section Assignment to Variables
13332 @cindex setting variables
13333 To alter the value of a variable, evaluate an assignment expression.
13334 @xref{Expressions, ,Expressions}. For example,
13341 stores the value 4 into the variable @code{x}, and then prints the
13342 value of the assignment expression (which is 4).
13343 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
13344 information on operators in supported languages.
13346 @kindex set variable
13347 @cindex variables, setting
13348 If you are not interested in seeing the value of the assignment, use the
13349 @code{set} command instead of the @code{print} command. @code{set} is
13350 really the same as @code{print} except that the expression's value is
13351 not printed and is not put in the value history (@pxref{Value History,
13352 ,Value History}). The expression is evaluated only for its effects.
13354 If the beginning of the argument string of the @code{set} command
13355 appears identical to a @code{set} subcommand, use the @code{set
13356 variable} command instead of just @code{set}. This command is identical
13357 to @code{set} except for its lack of subcommands. For example, if your
13358 program has a variable @code{width}, you get an error if you try to set
13359 a new value with just @samp{set width=13}, because @value{GDBN} has the
13360 command @code{set width}:
13363 (@value{GDBP}) whatis width
13365 (@value{GDBP}) p width
13367 (@value{GDBP}) set width=47
13368 Invalid syntax in expression.
13372 The invalid expression, of course, is @samp{=47}. In
13373 order to actually set the program's variable @code{width}, use
13376 (@value{GDBP}) set var width=47
13379 Because the @code{set} command has many subcommands that can conflict
13380 with the names of program variables, it is a good idea to use the
13381 @code{set variable} command instead of just @code{set}. For example, if
13382 your program has a variable @code{g}, you run into problems if you try
13383 to set a new value with just @samp{set g=4}, because @value{GDBN} has
13384 the command @code{set gnutarget}, abbreviated @code{set g}:
13388 (@value{GDBP}) whatis g
13392 (@value{GDBP}) set g=4
13396 The program being debugged has been started already.
13397 Start it from the beginning? (y or n) y
13398 Starting program: /home/smith/cc_progs/a.out
13399 "/home/smith/cc_progs/a.out": can't open to read symbols:
13400 Invalid bfd target.
13401 (@value{GDBP}) show g
13402 The current BFD target is "=4".
13407 The program variable @code{g} did not change, and you silently set the
13408 @code{gnutarget} to an invalid value. In order to set the variable
13412 (@value{GDBP}) set var g=4
13415 @value{GDBN} allows more implicit conversions in assignments than C; you can
13416 freely store an integer value into a pointer variable or vice versa,
13417 and you can convert any structure to any other structure that is the
13418 same length or shorter.
13419 @comment FIXME: how do structs align/pad in these conversions?
13420 @comment /doc@cygnus.com 18dec1990
13422 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
13423 construct to generate a value of specified type at a specified address
13424 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
13425 to memory location @code{0x83040} as an integer (which implies a certain size
13426 and representation in memory), and
13429 set @{int@}0x83040 = 4
13433 stores the value 4 into that memory location.
13436 @section Continuing at a Different Address
13438 Ordinarily, when you continue your program, you do so at the place where
13439 it stopped, with the @code{continue} command. You can instead continue at
13440 an address of your own choosing, with the following commands:
13444 @item jump @var{linespec}
13445 @itemx jump @var{location}
13446 Resume execution at line @var{linespec} or at address given by
13447 @var{location}. Execution stops again immediately if there is a
13448 breakpoint there. @xref{Specify Location}, for a description of the
13449 different forms of @var{linespec} and @var{location}. It is common
13450 practice to use the @code{tbreak} command in conjunction with
13451 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
13453 The @code{jump} command does not change the current stack frame, or
13454 the stack pointer, or the contents of any memory location or any
13455 register other than the program counter. If line @var{linespec} is in
13456 a different function from the one currently executing, the results may
13457 be bizarre if the two functions expect different patterns of arguments or
13458 of local variables. For this reason, the @code{jump} command requests
13459 confirmation if the specified line is not in the function currently
13460 executing. However, even bizarre results are predictable if you are
13461 well acquainted with the machine-language code of your program.
13464 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
13465 On many systems, you can get much the same effect as the @code{jump}
13466 command by storing a new value into the register @code{$pc}. The
13467 difference is that this does not start your program running; it only
13468 changes the address of where it @emph{will} run when you continue. For
13476 makes the next @code{continue} command or stepping command execute at
13477 address @code{0x485}, rather than at the address where your program stopped.
13478 @xref{Continuing and Stepping, ,Continuing and Stepping}.
13480 The most common occasion to use the @code{jump} command is to back
13481 up---perhaps with more breakpoints set---over a portion of a program
13482 that has already executed, in order to examine its execution in more
13487 @section Giving your Program a Signal
13488 @cindex deliver a signal to a program
13492 @item signal @var{signal}
13493 Resume execution where your program stopped, but immediately give it the
13494 signal @var{signal}. @var{signal} can be the name or the number of a
13495 signal. For example, on many systems @code{signal 2} and @code{signal
13496 SIGINT} are both ways of sending an interrupt signal.
13498 Alternatively, if @var{signal} is zero, continue execution without
13499 giving a signal. This is useful when your program stopped on account of
13500 a signal and would ordinary see the signal when resumed with the
13501 @code{continue} command; @samp{signal 0} causes it to resume without a
13504 @code{signal} does not repeat when you press @key{RET} a second time
13505 after executing the command.
13509 Invoking the @code{signal} command is not the same as invoking the
13510 @code{kill} utility from the shell. Sending a signal with @code{kill}
13511 causes @value{GDBN} to decide what to do with the signal depending on
13512 the signal handling tables (@pxref{Signals}). The @code{signal} command
13513 passes the signal directly to your program.
13517 @section Returning from a Function
13520 @cindex returning from a function
13523 @itemx return @var{expression}
13524 You can cancel execution of a function call with the @code{return}
13525 command. If you give an
13526 @var{expression} argument, its value is used as the function's return
13530 When you use @code{return}, @value{GDBN} discards the selected stack frame
13531 (and all frames within it). You can think of this as making the
13532 discarded frame return prematurely. If you wish to specify a value to
13533 be returned, give that value as the argument to @code{return}.
13535 This pops the selected stack frame (@pxref{Selection, ,Selecting a
13536 Frame}), and any other frames inside of it, leaving its caller as the
13537 innermost remaining frame. That frame becomes selected. The
13538 specified value is stored in the registers used for returning values
13541 The @code{return} command does not resume execution; it leaves the
13542 program stopped in the state that would exist if the function had just
13543 returned. In contrast, the @code{finish} command (@pxref{Continuing
13544 and Stepping, ,Continuing and Stepping}) resumes execution until the
13545 selected stack frame returns naturally.
13547 @value{GDBN} needs to know how the @var{expression} argument should be set for
13548 the inferior. The concrete registers assignment depends on the OS ABI and the
13549 type being returned by the selected stack frame. For example it is common for
13550 OS ABI to return floating point values in FPU registers while integer values in
13551 CPU registers. Still some ABIs return even floating point values in CPU
13552 registers. Larger integer widths (such as @code{long long int}) also have
13553 specific placement rules. @value{GDBN} already knows the OS ABI from its
13554 current target so it needs to find out also the type being returned to make the
13555 assignment into the right register(s).
13557 Normally, the selected stack frame has debug info. @value{GDBN} will always
13558 use the debug info instead of the implicit type of @var{expression} when the
13559 debug info is available. For example, if you type @kbd{return -1}, and the
13560 function in the current stack frame is declared to return a @code{long long
13561 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
13562 into a @code{long long int}:
13565 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
13567 (@value{GDBP}) return -1
13568 Make func return now? (y or n) y
13569 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
13570 43 printf ("result=%lld\n", func ());
13574 However, if the selected stack frame does not have a debug info, e.g., if the
13575 function was compiled without debug info, @value{GDBN} has to find out the type
13576 to return from user. Specifying a different type by mistake may set the value
13577 in different inferior registers than the caller code expects. For example,
13578 typing @kbd{return -1} with its implicit type @code{int} would set only a part
13579 of a @code{long long int} result for a debug info less function (on 32-bit
13580 architectures). Therefore the user is required to specify the return type by
13581 an appropriate cast explicitly:
13584 Breakpoint 2, 0x0040050b in func ()
13585 (@value{GDBP}) return -1
13586 Return value type not available for selected stack frame.
13587 Please use an explicit cast of the value to return.
13588 (@value{GDBP}) return (long long int) -1
13589 Make selected stack frame return now? (y or n) y
13590 #0 0x00400526 in main ()
13595 @section Calling Program Functions
13598 @cindex calling functions
13599 @cindex inferior functions, calling
13600 @item print @var{expr}
13601 Evaluate the expression @var{expr} and display the resulting value.
13602 @var{expr} may include calls to functions in the program being
13606 @item call @var{expr}
13607 Evaluate the expression @var{expr} without displaying @code{void}
13610 You can use this variant of the @code{print} command if you want to
13611 execute a function from your program that does not return anything
13612 (a.k.a.@: @dfn{a void function}), but without cluttering the output
13613 with @code{void} returned values that @value{GDBN} will otherwise
13614 print. If the result is not void, it is printed and saved in the
13618 It is possible for the function you call via the @code{print} or
13619 @code{call} command to generate a signal (e.g., if there's a bug in
13620 the function, or if you passed it incorrect arguments). What happens
13621 in that case is controlled by the @code{set unwindonsignal} command.
13623 Similarly, with a C@t{++} program it is possible for the function you
13624 call via the @code{print} or @code{call} command to generate an
13625 exception that is not handled due to the constraints of the dummy
13626 frame. In this case, any exception that is raised in the frame, but has
13627 an out-of-frame exception handler will not be found. GDB builds a
13628 dummy-frame for the inferior function call, and the unwinder cannot
13629 seek for exception handlers outside of this dummy-frame. What happens
13630 in that case is controlled by the
13631 @code{set unwind-on-terminating-exception} command.
13634 @item set unwindonsignal
13635 @kindex set unwindonsignal
13636 @cindex unwind stack in called functions
13637 @cindex call dummy stack unwinding
13638 Set unwinding of the stack if a signal is received while in a function
13639 that @value{GDBN} called in the program being debugged. If set to on,
13640 @value{GDBN} unwinds the stack it created for the call and restores
13641 the context to what it was before the call. If set to off (the
13642 default), @value{GDBN} stops in the frame where the signal was
13645 @item show unwindonsignal
13646 @kindex show unwindonsignal
13647 Show the current setting of stack unwinding in the functions called by
13650 @item set unwind-on-terminating-exception
13651 @kindex set unwind-on-terminating-exception
13652 @cindex unwind stack in called functions with unhandled exceptions
13653 @cindex call dummy stack unwinding on unhandled exception.
13654 Set unwinding of the stack if a C@t{++} exception is raised, but left
13655 unhandled while in a function that @value{GDBN} called in the program being
13656 debugged. If set to on (the default), @value{GDBN} unwinds the stack
13657 it created for the call and restores the context to what it was before
13658 the call. If set to off, @value{GDBN} the exception is delivered to
13659 the default C@t{++} exception handler and the inferior terminated.
13661 @item show unwind-on-terminating-exception
13662 @kindex show unwind-on-terminating-exception
13663 Show the current setting of stack unwinding in the functions called by
13668 @cindex weak alias functions
13669 Sometimes, a function you wish to call is actually a @dfn{weak alias}
13670 for another function. In such case, @value{GDBN} might not pick up
13671 the type information, including the types of the function arguments,
13672 which causes @value{GDBN} to call the inferior function incorrectly.
13673 As a result, the called function will function erroneously and may
13674 even crash. A solution to that is to use the name of the aliased
13678 @section Patching Programs
13680 @cindex patching binaries
13681 @cindex writing into executables
13682 @cindex writing into corefiles
13684 By default, @value{GDBN} opens the file containing your program's
13685 executable code (or the corefile) read-only. This prevents accidental
13686 alterations to machine code; but it also prevents you from intentionally
13687 patching your program's binary.
13689 If you'd like to be able to patch the binary, you can specify that
13690 explicitly with the @code{set write} command. For example, you might
13691 want to turn on internal debugging flags, or even to make emergency
13697 @itemx set write off
13698 If you specify @samp{set write on}, @value{GDBN} opens executable and
13699 core files for both reading and writing; if you specify @kbd{set write
13700 off} (the default), @value{GDBN} opens them read-only.
13702 If you have already loaded a file, you must load it again (using the
13703 @code{exec-file} or @code{core-file} command) after changing @code{set
13704 write}, for your new setting to take effect.
13708 Display whether executable files and core files are opened for writing
13709 as well as reading.
13713 @chapter @value{GDBN} Files
13715 @value{GDBN} needs to know the file name of the program to be debugged,
13716 both in order to read its symbol table and in order to start your
13717 program. To debug a core dump of a previous run, you must also tell
13718 @value{GDBN} the name of the core dump file.
13721 * Files:: Commands to specify files
13722 * Separate Debug Files:: Debugging information in separate files
13723 * Symbol Errors:: Errors reading symbol files
13724 * Data Files:: GDB data files
13728 @section Commands to Specify Files
13730 @cindex symbol table
13731 @cindex core dump file
13733 You may want to specify executable and core dump file names. The usual
13734 way to do this is at start-up time, using the arguments to
13735 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
13736 Out of @value{GDBN}}).
13738 Occasionally it is necessary to change to a different file during a
13739 @value{GDBN} session. Or you may run @value{GDBN} and forget to
13740 specify a file you want to use. Or you are debugging a remote target
13741 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
13742 Program}). In these situations the @value{GDBN} commands to specify
13743 new files are useful.
13746 @cindex executable file
13748 @item file @var{filename}
13749 Use @var{filename} as the program to be debugged. It is read for its
13750 symbols and for the contents of pure memory. It is also the program
13751 executed when you use the @code{run} command. If you do not specify a
13752 directory and the file is not found in the @value{GDBN} working directory,
13753 @value{GDBN} uses the environment variable @code{PATH} as a list of
13754 directories to search, just as the shell does when looking for a program
13755 to run. You can change the value of this variable, for both @value{GDBN}
13756 and your program, using the @code{path} command.
13758 @cindex unlinked object files
13759 @cindex patching object files
13760 You can load unlinked object @file{.o} files into @value{GDBN} using
13761 the @code{file} command. You will not be able to ``run'' an object
13762 file, but you can disassemble functions and inspect variables. Also,
13763 if the underlying BFD functionality supports it, you could use
13764 @kbd{gdb -write} to patch object files using this technique. Note
13765 that @value{GDBN} can neither interpret nor modify relocations in this
13766 case, so branches and some initialized variables will appear to go to
13767 the wrong place. But this feature is still handy from time to time.
13770 @code{file} with no argument makes @value{GDBN} discard any information it
13771 has on both executable file and the symbol table.
13774 @item exec-file @r{[} @var{filename} @r{]}
13775 Specify that the program to be run (but not the symbol table) is found
13776 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
13777 if necessary to locate your program. Omitting @var{filename} means to
13778 discard information on the executable file.
13780 @kindex symbol-file
13781 @item symbol-file @r{[} @var{filename} @r{]}
13782 Read symbol table information from file @var{filename}. @code{PATH} is
13783 searched when necessary. Use the @code{file} command to get both symbol
13784 table and program to run from the same file.
13786 @code{symbol-file} with no argument clears out @value{GDBN} information on your
13787 program's symbol table.
13789 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
13790 some breakpoints and auto-display expressions. This is because they may
13791 contain pointers to the internal data recording symbols and data types,
13792 which are part of the old symbol table data being discarded inside
13795 @code{symbol-file} does not repeat if you press @key{RET} again after
13798 When @value{GDBN} is configured for a particular environment, it
13799 understands debugging information in whatever format is the standard
13800 generated for that environment; you may use either a @sc{gnu} compiler, or
13801 other compilers that adhere to the local conventions.
13802 Best results are usually obtained from @sc{gnu} compilers; for example,
13803 using @code{@value{NGCC}} you can generate debugging information for
13806 For most kinds of object files, with the exception of old SVR3 systems
13807 using COFF, the @code{symbol-file} command does not normally read the
13808 symbol table in full right away. Instead, it scans the symbol table
13809 quickly to find which source files and which symbols are present. The
13810 details are read later, one source file at a time, as they are needed.
13812 The purpose of this two-stage reading strategy is to make @value{GDBN}
13813 start up faster. For the most part, it is invisible except for
13814 occasional pauses while the symbol table details for a particular source
13815 file are being read. (The @code{set verbose} command can turn these
13816 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
13817 Warnings and Messages}.)
13819 We have not implemented the two-stage strategy for COFF yet. When the
13820 symbol table is stored in COFF format, @code{symbol-file} reads the
13821 symbol table data in full right away. Note that ``stabs-in-COFF''
13822 still does the two-stage strategy, since the debug info is actually
13826 @cindex reading symbols immediately
13827 @cindex symbols, reading immediately
13828 @item symbol-file @r{[} -readnow @r{]} @var{filename}
13829 @itemx file @r{[} -readnow @r{]} @var{filename}
13830 You can override the @value{GDBN} two-stage strategy for reading symbol
13831 tables by using the @samp{-readnow} option with any of the commands that
13832 load symbol table information, if you want to be sure @value{GDBN} has the
13833 entire symbol table available.
13835 @c FIXME: for now no mention of directories, since this seems to be in
13836 @c flux. 13mar1992 status is that in theory GDB would look either in
13837 @c current dir or in same dir as myprog; but issues like competing
13838 @c GDB's, or clutter in system dirs, mean that in practice right now
13839 @c only current dir is used. FFish says maybe a special GDB hierarchy
13840 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
13844 @item core-file @r{[}@var{filename}@r{]}
13846 Specify the whereabouts of a core dump file to be used as the ``contents
13847 of memory''. Traditionally, core files contain only some parts of the
13848 address space of the process that generated them; @value{GDBN} can access the
13849 executable file itself for other parts.
13851 @code{core-file} with no argument specifies that no core file is
13854 Note that the core file is ignored when your program is actually running
13855 under @value{GDBN}. So, if you have been running your program and you
13856 wish to debug a core file instead, you must kill the subprocess in which
13857 the program is running. To do this, use the @code{kill} command
13858 (@pxref{Kill Process, ,Killing the Child Process}).
13860 @kindex add-symbol-file
13861 @cindex dynamic linking
13862 @item add-symbol-file @var{filename} @var{address}
13863 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
13864 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
13865 The @code{add-symbol-file} command reads additional symbol table
13866 information from the file @var{filename}. You would use this command
13867 when @var{filename} has been dynamically loaded (by some other means)
13868 into the program that is running. @var{address} should be the memory
13869 address at which the file has been loaded; @value{GDBN} cannot figure
13870 this out for itself. You can additionally specify an arbitrary number
13871 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
13872 section name and base address for that section. You can specify any
13873 @var{address} as an expression.
13875 The symbol table of the file @var{filename} is added to the symbol table
13876 originally read with the @code{symbol-file} command. You can use the
13877 @code{add-symbol-file} command any number of times; the new symbol data
13878 thus read keeps adding to the old. To discard all old symbol data
13879 instead, use the @code{symbol-file} command without any arguments.
13881 @cindex relocatable object files, reading symbols from
13882 @cindex object files, relocatable, reading symbols from
13883 @cindex reading symbols from relocatable object files
13884 @cindex symbols, reading from relocatable object files
13885 @cindex @file{.o} files, reading symbols from
13886 Although @var{filename} is typically a shared library file, an
13887 executable file, or some other object file which has been fully
13888 relocated for loading into a process, you can also load symbolic
13889 information from relocatable @file{.o} files, as long as:
13893 the file's symbolic information refers only to linker symbols defined in
13894 that file, not to symbols defined by other object files,
13896 every section the file's symbolic information refers to has actually
13897 been loaded into the inferior, as it appears in the file, and
13899 you can determine the address at which every section was loaded, and
13900 provide these to the @code{add-symbol-file} command.
13904 Some embedded operating systems, like Sun Chorus and VxWorks, can load
13905 relocatable files into an already running program; such systems
13906 typically make the requirements above easy to meet. However, it's
13907 important to recognize that many native systems use complex link
13908 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
13909 assembly, for example) that make the requirements difficult to meet. In
13910 general, one cannot assume that using @code{add-symbol-file} to read a
13911 relocatable object file's symbolic information will have the same effect
13912 as linking the relocatable object file into the program in the normal
13915 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
13917 @kindex add-symbol-file-from-memory
13918 @cindex @code{syscall DSO}
13919 @cindex load symbols from memory
13920 @item add-symbol-file-from-memory @var{address}
13921 Load symbols from the given @var{address} in a dynamically loaded
13922 object file whose image is mapped directly into the inferior's memory.
13923 For example, the Linux kernel maps a @code{syscall DSO} into each
13924 process's address space; this DSO provides kernel-specific code for
13925 some system calls. The argument can be any expression whose
13926 evaluation yields the address of the file's shared object file header.
13927 For this command to work, you must have used @code{symbol-file} or
13928 @code{exec-file} commands in advance.
13930 @kindex add-shared-symbol-files
13932 @item add-shared-symbol-files @var{library-file}
13933 @itemx assf @var{library-file}
13934 The @code{add-shared-symbol-files} command can currently be used only
13935 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
13936 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
13937 @value{GDBN} automatically looks for shared libraries, however if
13938 @value{GDBN} does not find yours, you can invoke
13939 @code{add-shared-symbol-files}. It takes one argument: the shared
13940 library's file name. @code{assf} is a shorthand alias for
13941 @code{add-shared-symbol-files}.
13944 @item section @var{section} @var{addr}
13945 The @code{section} command changes the base address of the named
13946 @var{section} of the exec file to @var{addr}. This can be used if the
13947 exec file does not contain section addresses, (such as in the
13948 @code{a.out} format), or when the addresses specified in the file
13949 itself are wrong. Each section must be changed separately. The
13950 @code{info files} command, described below, lists all the sections and
13954 @kindex info target
13957 @code{info files} and @code{info target} are synonymous; both print the
13958 current target (@pxref{Targets, ,Specifying a Debugging Target}),
13959 including the names of the executable and core dump files currently in
13960 use by @value{GDBN}, and the files from which symbols were loaded. The
13961 command @code{help target} lists all possible targets rather than
13964 @kindex maint info sections
13965 @item maint info sections
13966 Another command that can give you extra information about program sections
13967 is @code{maint info sections}. In addition to the section information
13968 displayed by @code{info files}, this command displays the flags and file
13969 offset of each section in the executable and core dump files. In addition,
13970 @code{maint info sections} provides the following command options (which
13971 may be arbitrarily combined):
13975 Display sections for all loaded object files, including shared libraries.
13976 @item @var{sections}
13977 Display info only for named @var{sections}.
13978 @item @var{section-flags}
13979 Display info only for sections for which @var{section-flags} are true.
13980 The section flags that @value{GDBN} currently knows about are:
13983 Section will have space allocated in the process when loaded.
13984 Set for all sections except those containing debug information.
13986 Section will be loaded from the file into the child process memory.
13987 Set for pre-initialized code and data, clear for @code{.bss} sections.
13989 Section needs to be relocated before loading.
13991 Section cannot be modified by the child process.
13993 Section contains executable code only.
13995 Section contains data only (no executable code).
13997 Section will reside in ROM.
13999 Section contains data for constructor/destructor lists.
14001 Section is not empty.
14003 An instruction to the linker to not output the section.
14004 @item COFF_SHARED_LIBRARY
14005 A notification to the linker that the section contains
14006 COFF shared library information.
14008 Section contains common symbols.
14011 @kindex set trust-readonly-sections
14012 @cindex read-only sections
14013 @item set trust-readonly-sections on
14014 Tell @value{GDBN} that readonly sections in your object file
14015 really are read-only (i.e.@: that their contents will not change).
14016 In that case, @value{GDBN} can fetch values from these sections
14017 out of the object file, rather than from the target program.
14018 For some targets (notably embedded ones), this can be a significant
14019 enhancement to debugging performance.
14021 The default is off.
14023 @item set trust-readonly-sections off
14024 Tell @value{GDBN} not to trust readonly sections. This means that
14025 the contents of the section might change while the program is running,
14026 and must therefore be fetched from the target when needed.
14028 @item show trust-readonly-sections
14029 Show the current setting of trusting readonly sections.
14032 All file-specifying commands allow both absolute and relative file names
14033 as arguments. @value{GDBN} always converts the file name to an absolute file
14034 name and remembers it that way.
14036 @cindex shared libraries
14037 @anchor{Shared Libraries}
14038 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
14039 and IBM RS/6000 AIX shared libraries.
14041 On MS-Windows @value{GDBN} must be linked with the Expat library to support
14042 shared libraries. @xref{Expat}.
14044 @value{GDBN} automatically loads symbol definitions from shared libraries
14045 when you use the @code{run} command, or when you examine a core file.
14046 (Before you issue the @code{run} command, @value{GDBN} does not understand
14047 references to a function in a shared library, however---unless you are
14048 debugging a core file).
14050 On HP-UX, if the program loads a library explicitly, @value{GDBN}
14051 automatically loads the symbols at the time of the @code{shl_load} call.
14053 @c FIXME: some @value{GDBN} release may permit some refs to undef
14054 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
14055 @c FIXME...lib; check this from time to time when updating manual
14057 There are times, however, when you may wish to not automatically load
14058 symbol definitions from shared libraries, such as when they are
14059 particularly large or there are many of them.
14061 To control the automatic loading of shared library symbols, use the
14065 @kindex set auto-solib-add
14066 @item set auto-solib-add @var{mode}
14067 If @var{mode} is @code{on}, symbols from all shared object libraries
14068 will be loaded automatically when the inferior begins execution, you
14069 attach to an independently started inferior, or when the dynamic linker
14070 informs @value{GDBN} that a new library has been loaded. If @var{mode}
14071 is @code{off}, symbols must be loaded manually, using the
14072 @code{sharedlibrary} command. The default value is @code{on}.
14074 @cindex memory used for symbol tables
14075 If your program uses lots of shared libraries with debug info that
14076 takes large amounts of memory, you can decrease the @value{GDBN}
14077 memory footprint by preventing it from automatically loading the
14078 symbols from shared libraries. To that end, type @kbd{set
14079 auto-solib-add off} before running the inferior, then load each
14080 library whose debug symbols you do need with @kbd{sharedlibrary
14081 @var{regexp}}, where @var{regexp} is a regular expression that matches
14082 the libraries whose symbols you want to be loaded.
14084 @kindex show auto-solib-add
14085 @item show auto-solib-add
14086 Display the current autoloading mode.
14089 @cindex load shared library
14090 To explicitly load shared library symbols, use the @code{sharedlibrary}
14094 @kindex info sharedlibrary
14096 @item info share @var{regex}
14097 @itemx info sharedlibrary @var{regex}
14098 Print the names of the shared libraries which are currently loaded
14099 that match @var{regex}. If @var{regex} is omitted then print
14100 all shared libraries that are loaded.
14102 @kindex sharedlibrary
14104 @item sharedlibrary @var{regex}
14105 @itemx share @var{regex}
14106 Load shared object library symbols for files matching a
14107 Unix regular expression.
14108 As with files loaded automatically, it only loads shared libraries
14109 required by your program for a core file or after typing @code{run}. If
14110 @var{regex} is omitted all shared libraries required by your program are
14113 @item nosharedlibrary
14114 @kindex nosharedlibrary
14115 @cindex unload symbols from shared libraries
14116 Unload all shared object library symbols. This discards all symbols
14117 that have been loaded from all shared libraries. Symbols from shared
14118 libraries that were loaded by explicit user requests are not
14122 Sometimes you may wish that @value{GDBN} stops and gives you control
14123 when any of shared library events happen. Use the @code{set
14124 stop-on-solib-events} command for this:
14127 @item set stop-on-solib-events
14128 @kindex set stop-on-solib-events
14129 This command controls whether @value{GDBN} should give you control
14130 when the dynamic linker notifies it about some shared library event.
14131 The most common event of interest is loading or unloading of a new
14134 @item show stop-on-solib-events
14135 @kindex show stop-on-solib-events
14136 Show whether @value{GDBN} stops and gives you control when shared
14137 library events happen.
14140 Shared libraries are also supported in many cross or remote debugging
14141 configurations. @value{GDBN} needs to have access to the target's libraries;
14142 this can be accomplished either by providing copies of the libraries
14143 on the host system, or by asking @value{GDBN} to automatically retrieve the
14144 libraries from the target. If copies of the target libraries are
14145 provided, they need to be the same as the target libraries, although the
14146 copies on the target can be stripped as long as the copies on the host are
14149 @cindex where to look for shared libraries
14150 For remote debugging, you need to tell @value{GDBN} where the target
14151 libraries are, so that it can load the correct copies---otherwise, it
14152 may try to load the host's libraries. @value{GDBN} has two variables
14153 to specify the search directories for target libraries.
14156 @cindex prefix for shared library file names
14157 @cindex system root, alternate
14158 @kindex set solib-absolute-prefix
14159 @kindex set sysroot
14160 @item set sysroot @var{path}
14161 Use @var{path} as the system root for the program being debugged. Any
14162 absolute shared library paths will be prefixed with @var{path}; many
14163 runtime loaders store the absolute paths to the shared library in the
14164 target program's memory. If you use @code{set sysroot} to find shared
14165 libraries, they need to be laid out in the same way that they are on
14166 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
14169 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
14170 retrieve the target libraries from the remote system. This is only
14171 supported when using a remote target that supports the @code{remote get}
14172 command (@pxref{File Transfer,,Sending files to a remote system}).
14173 The part of @var{path} following the initial @file{remote:}
14174 (if present) is used as system root prefix on the remote file system.
14175 @footnote{If you want to specify a local system root using a directory
14176 that happens to be named @file{remote:}, you need to use some equivalent
14177 variant of the name like @file{./remote:}.}
14179 The @code{set solib-absolute-prefix} command is an alias for @code{set
14182 @cindex default system root
14183 @cindex @samp{--with-sysroot}
14184 You can set the default system root by using the configure-time
14185 @samp{--with-sysroot} option. If the system root is inside
14186 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
14187 @samp{--exec-prefix}), then the default system root will be updated
14188 automatically if the installed @value{GDBN} is moved to a new
14191 @kindex show sysroot
14193 Display the current shared library prefix.
14195 @kindex set solib-search-path
14196 @item set solib-search-path @var{path}
14197 If this variable is set, @var{path} is a colon-separated list of
14198 directories to search for shared libraries. @samp{solib-search-path}
14199 is used after @samp{sysroot} fails to locate the library, or if the
14200 path to the library is relative instead of absolute. If you want to
14201 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
14202 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
14203 finding your host's libraries. @samp{sysroot} is preferred; setting
14204 it to a nonexistent directory may interfere with automatic loading
14205 of shared library symbols.
14207 @kindex show solib-search-path
14208 @item show solib-search-path
14209 Display the current shared library search path.
14213 @node Separate Debug Files
14214 @section Debugging Information in Separate Files
14215 @cindex separate debugging information files
14216 @cindex debugging information in separate files
14217 @cindex @file{.debug} subdirectories
14218 @cindex debugging information directory, global
14219 @cindex global debugging information directory
14220 @cindex build ID, and separate debugging files
14221 @cindex @file{.build-id} directory
14223 @value{GDBN} allows you to put a program's debugging information in a
14224 file separate from the executable itself, in a way that allows
14225 @value{GDBN} to find and load the debugging information automatically.
14226 Since debugging information can be very large---sometimes larger
14227 than the executable code itself---some systems distribute debugging
14228 information for their executables in separate files, which users can
14229 install only when they need to debug a problem.
14231 @value{GDBN} supports two ways of specifying the separate debug info
14236 The executable contains a @dfn{debug link} that specifies the name of
14237 the separate debug info file. The separate debug file's name is
14238 usually @file{@var{executable}.debug}, where @var{executable} is the
14239 name of the corresponding executable file without leading directories
14240 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
14241 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
14242 checksum for the debug file, which @value{GDBN} uses to validate that
14243 the executable and the debug file came from the same build.
14246 The executable contains a @dfn{build ID}, a unique bit string that is
14247 also present in the corresponding debug info file. (This is supported
14248 only on some operating systems, notably those which use the ELF format
14249 for binary files and the @sc{gnu} Binutils.) For more details about
14250 this feature, see the description of the @option{--build-id}
14251 command-line option in @ref{Options, , Command Line Options, ld.info,
14252 The GNU Linker}. The debug info file's name is not specified
14253 explicitly by the build ID, but can be computed from the build ID, see
14257 Depending on the way the debug info file is specified, @value{GDBN}
14258 uses two different methods of looking for the debug file:
14262 For the ``debug link'' method, @value{GDBN} looks up the named file in
14263 the directory of the executable file, then in a subdirectory of that
14264 directory named @file{.debug}, and finally under the global debug
14265 directory, in a subdirectory whose name is identical to the leading
14266 directories of the executable's absolute file name.
14269 For the ``build ID'' method, @value{GDBN} looks in the
14270 @file{.build-id} subdirectory of the global debug directory for a file
14271 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
14272 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
14273 are the rest of the bit string. (Real build ID strings are 32 or more
14274 hex characters, not 10.)
14277 So, for example, suppose you ask @value{GDBN} to debug
14278 @file{/usr/bin/ls}, which has a debug link that specifies the
14279 file @file{ls.debug}, and a build ID whose value in hex is
14280 @code{abcdef1234}. If the global debug directory is
14281 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
14282 debug information files, in the indicated order:
14286 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
14288 @file{/usr/bin/ls.debug}
14290 @file{/usr/bin/.debug/ls.debug}
14292 @file{/usr/lib/debug/usr/bin/ls.debug}.
14295 You can set the global debugging info directory's name, and view the
14296 name @value{GDBN} is currently using.
14300 @kindex set debug-file-directory
14301 @item set debug-file-directory @var{directories}
14302 Set the directories which @value{GDBN} searches for separate debugging
14303 information files to @var{directory}. Multiple directory components can be set
14304 concatenating them by a directory separator.
14306 @kindex show debug-file-directory
14307 @item show debug-file-directory
14308 Show the directories @value{GDBN} searches for separate debugging
14313 @cindex @code{.gnu_debuglink} sections
14314 @cindex debug link sections
14315 A debug link is a special section of the executable file named
14316 @code{.gnu_debuglink}. The section must contain:
14320 A filename, with any leading directory components removed, followed by
14323 zero to three bytes of padding, as needed to reach the next four-byte
14324 boundary within the section, and
14326 a four-byte CRC checksum, stored in the same endianness used for the
14327 executable file itself. The checksum is computed on the debugging
14328 information file's full contents by the function given below, passing
14329 zero as the @var{crc} argument.
14332 Any executable file format can carry a debug link, as long as it can
14333 contain a section named @code{.gnu_debuglink} with the contents
14336 @cindex @code{.note.gnu.build-id} sections
14337 @cindex build ID sections
14338 The build ID is a special section in the executable file (and in other
14339 ELF binary files that @value{GDBN} may consider). This section is
14340 often named @code{.note.gnu.build-id}, but that name is not mandatory.
14341 It contains unique identification for the built files---the ID remains
14342 the same across multiple builds of the same build tree. The default
14343 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
14344 content for the build ID string. The same section with an identical
14345 value is present in the original built binary with symbols, in its
14346 stripped variant, and in the separate debugging information file.
14348 The debugging information file itself should be an ordinary
14349 executable, containing a full set of linker symbols, sections, and
14350 debugging information. The sections of the debugging information file
14351 should have the same names, addresses, and sizes as the original file,
14352 but they need not contain any data---much like a @code{.bss} section
14353 in an ordinary executable.
14355 The @sc{gnu} binary utilities (Binutils) package includes the
14356 @samp{objcopy} utility that can produce
14357 the separated executable / debugging information file pairs using the
14358 following commands:
14361 @kbd{objcopy --only-keep-debug foo foo.debug}
14366 These commands remove the debugging
14367 information from the executable file @file{foo} and place it in the file
14368 @file{foo.debug}. You can use the first, second or both methods to link the
14373 The debug link method needs the following additional command to also leave
14374 behind a debug link in @file{foo}:
14377 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
14380 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
14381 a version of the @code{strip} command such that the command @kbd{strip foo -f
14382 foo.debug} has the same functionality as the two @code{objcopy} commands and
14383 the @code{ln -s} command above, together.
14386 Build ID gets embedded into the main executable using @code{ld --build-id} or
14387 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
14388 compatibility fixes for debug files separation are present in @sc{gnu} binary
14389 utilities (Binutils) package since version 2.18.
14394 @cindex CRC algorithm definition
14395 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
14396 IEEE 802.3 using the polynomial:
14398 @c TexInfo requires naked braces for multi-digit exponents for Tex
14399 @c output, but this causes HTML output to barf. HTML has to be set using
14400 @c raw commands. So we end up having to specify this equation in 2
14405 <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>
14406 + <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
14412 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
14413 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
14417 The function is computed byte at a time, taking the least
14418 significant bit of each byte first. The initial pattern
14419 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
14420 the final result is inverted to ensure trailing zeros also affect the
14423 @emph{Note:} This is the same CRC polynomial as used in handling the
14424 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
14425 , @value{GDBN} Remote Serial Protocol}). However in the
14426 case of the Remote Serial Protocol, the CRC is computed @emph{most}
14427 significant bit first, and the result is not inverted, so trailing
14428 zeros have no effect on the CRC value.
14430 To complete the description, we show below the code of the function
14431 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
14432 initially supplied @code{crc} argument means that an initial call to
14433 this function passing in zero will start computing the CRC using
14436 @kindex gnu_debuglink_crc32
14439 gnu_debuglink_crc32 (unsigned long crc,
14440 unsigned char *buf, size_t len)
14442 static const unsigned long crc32_table[256] =
14444 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
14445 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
14446 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
14447 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
14448 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
14449 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
14450 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
14451 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
14452 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
14453 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
14454 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
14455 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
14456 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
14457 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
14458 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
14459 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
14460 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
14461 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
14462 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
14463 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
14464 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
14465 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
14466 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
14467 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
14468 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
14469 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
14470 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
14471 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
14472 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
14473 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
14474 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
14475 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
14476 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
14477 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
14478 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
14479 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
14480 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
14481 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
14482 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
14483 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
14484 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
14485 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
14486 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
14487 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
14488 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
14489 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
14490 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
14491 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
14492 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
14493 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
14494 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
14497 unsigned char *end;
14499 crc = ~crc & 0xffffffff;
14500 for (end = buf + len; buf < end; ++buf)
14501 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
14502 return ~crc & 0xffffffff;
14507 This computation does not apply to the ``build ID'' method.
14510 @node Symbol Errors
14511 @section Errors Reading Symbol Files
14513 While reading a symbol file, @value{GDBN} occasionally encounters problems,
14514 such as symbol types it does not recognize, or known bugs in compiler
14515 output. By default, @value{GDBN} does not notify you of such problems, since
14516 they are relatively common and primarily of interest to people
14517 debugging compilers. If you are interested in seeing information
14518 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
14519 only one message about each such type of problem, no matter how many
14520 times the problem occurs; or you can ask @value{GDBN} to print more messages,
14521 to see how many times the problems occur, with the @code{set
14522 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
14525 The messages currently printed, and their meanings, include:
14528 @item inner block not inside outer block in @var{symbol}
14530 The symbol information shows where symbol scopes begin and end
14531 (such as at the start of a function or a block of statements). This
14532 error indicates that an inner scope block is not fully contained
14533 in its outer scope blocks.
14535 @value{GDBN} circumvents the problem by treating the inner block as if it had
14536 the same scope as the outer block. In the error message, @var{symbol}
14537 may be shown as ``@code{(don't know)}'' if the outer block is not a
14540 @item block at @var{address} out of order
14542 The symbol information for symbol scope blocks should occur in
14543 order of increasing addresses. This error indicates that it does not
14546 @value{GDBN} does not circumvent this problem, and has trouble
14547 locating symbols in the source file whose symbols it is reading. (You
14548 can often determine what source file is affected by specifying
14549 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
14552 @item bad block start address patched
14554 The symbol information for a symbol scope block has a start address
14555 smaller than the address of the preceding source line. This is known
14556 to occur in the SunOS 4.1.1 (and earlier) C compiler.
14558 @value{GDBN} circumvents the problem by treating the symbol scope block as
14559 starting on the previous source line.
14561 @item bad string table offset in symbol @var{n}
14564 Symbol number @var{n} contains a pointer into the string table which is
14565 larger than the size of the string table.
14567 @value{GDBN} circumvents the problem by considering the symbol to have the
14568 name @code{foo}, which may cause other problems if many symbols end up
14571 @item unknown symbol type @code{0x@var{nn}}
14573 The symbol information contains new data types that @value{GDBN} does
14574 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
14575 uncomprehended information, in hexadecimal.
14577 @value{GDBN} circumvents the error by ignoring this symbol information.
14578 This usually allows you to debug your program, though certain symbols
14579 are not accessible. If you encounter such a problem and feel like
14580 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
14581 on @code{complain}, then go up to the function @code{read_dbx_symtab}
14582 and examine @code{*bufp} to see the symbol.
14584 @item stub type has NULL name
14586 @value{GDBN} could not find the full definition for a struct or class.
14588 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
14589 The symbol information for a C@t{++} member function is missing some
14590 information that recent versions of the compiler should have output for
14593 @item info mismatch between compiler and debugger
14595 @value{GDBN} could not parse a type specification output by the compiler.
14600 @section GDB Data Files
14602 @cindex prefix for data files
14603 @value{GDBN} will sometimes read an auxiliary data file. These files
14604 are kept in a directory known as the @dfn{data directory}.
14606 You can set the data directory's name, and view the name @value{GDBN}
14607 is currently using.
14610 @kindex set data-directory
14611 @item set data-directory @var{directory}
14612 Set the directory which @value{GDBN} searches for auxiliary data files
14613 to @var{directory}.
14615 @kindex show data-directory
14616 @item show data-directory
14617 Show the directory @value{GDBN} searches for auxiliary data files.
14620 @cindex default data directory
14621 @cindex @samp{--with-gdb-datadir}
14622 You can set the default data directory by using the configure-time
14623 @samp{--with-gdb-datadir} option. If the data directory is inside
14624 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
14625 @samp{--exec-prefix}), then the default data directory will be updated
14626 automatically if the installed @value{GDBN} is moved to a new
14630 @chapter Specifying a Debugging Target
14632 @cindex debugging target
14633 A @dfn{target} is the execution environment occupied by your program.
14635 Often, @value{GDBN} runs in the same host environment as your program;
14636 in that case, the debugging target is specified as a side effect when
14637 you use the @code{file} or @code{core} commands. When you need more
14638 flexibility---for example, running @value{GDBN} on a physically separate
14639 host, or controlling a standalone system over a serial port or a
14640 realtime system over a TCP/IP connection---you can use the @code{target}
14641 command to specify one of the target types configured for @value{GDBN}
14642 (@pxref{Target Commands, ,Commands for Managing Targets}).
14644 @cindex target architecture
14645 It is possible to build @value{GDBN} for several different @dfn{target
14646 architectures}. When @value{GDBN} is built like that, you can choose
14647 one of the available architectures with the @kbd{set architecture}
14651 @kindex set architecture
14652 @kindex show architecture
14653 @item set architecture @var{arch}
14654 This command sets the current target architecture to @var{arch}. The
14655 value of @var{arch} can be @code{"auto"}, in addition to one of the
14656 supported architectures.
14658 @item show architecture
14659 Show the current target architecture.
14661 @item set processor
14663 @kindex set processor
14664 @kindex show processor
14665 These are alias commands for, respectively, @code{set architecture}
14666 and @code{show architecture}.
14670 * Active Targets:: Active targets
14671 * Target Commands:: Commands for managing targets
14672 * Byte Order:: Choosing target byte order
14675 @node Active Targets
14676 @section Active Targets
14678 @cindex stacking targets
14679 @cindex active targets
14680 @cindex multiple targets
14682 There are three classes of targets: processes, core files, and
14683 executable files. @value{GDBN} can work concurrently on up to three
14684 active targets, one in each class. This allows you to (for example)
14685 start a process and inspect its activity without abandoning your work on
14688 For example, if you execute @samp{gdb a.out}, then the executable file
14689 @code{a.out} is the only active target. If you designate a core file as
14690 well---presumably from a prior run that crashed and coredumped---then
14691 @value{GDBN} has two active targets and uses them in tandem, looking
14692 first in the corefile target, then in the executable file, to satisfy
14693 requests for memory addresses. (Typically, these two classes of target
14694 are complementary, since core files contain only a program's
14695 read-write memory---variables and so on---plus machine status, while
14696 executable files contain only the program text and initialized data.)
14698 When you type @code{run}, your executable file becomes an active process
14699 target as well. When a process target is active, all @value{GDBN}
14700 commands requesting memory addresses refer to that target; addresses in
14701 an active core file or executable file target are obscured while the
14702 process target is active.
14704 Use the @code{core-file} and @code{exec-file} commands to select a new
14705 core file or executable target (@pxref{Files, ,Commands to Specify
14706 Files}). To specify as a target a process that is already running, use
14707 the @code{attach} command (@pxref{Attach, ,Debugging an Already-running
14710 @node Target Commands
14711 @section Commands for Managing Targets
14714 @item target @var{type} @var{parameters}
14715 Connects the @value{GDBN} host environment to a target machine or
14716 process. A target is typically a protocol for talking to debugging
14717 facilities. You use the argument @var{type} to specify the type or
14718 protocol of the target machine.
14720 Further @var{parameters} are interpreted by the target protocol, but
14721 typically include things like device names or host names to connect
14722 with, process numbers, and baud rates.
14724 The @code{target} command does not repeat if you press @key{RET} again
14725 after executing the command.
14727 @kindex help target
14729 Displays the names of all targets available. To display targets
14730 currently selected, use either @code{info target} or @code{info files}
14731 (@pxref{Files, ,Commands to Specify Files}).
14733 @item help target @var{name}
14734 Describe a particular target, including any parameters necessary to
14737 @kindex set gnutarget
14738 @item set gnutarget @var{args}
14739 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
14740 knows whether it is reading an @dfn{executable},
14741 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
14742 with the @code{set gnutarget} command. Unlike most @code{target} commands,
14743 with @code{gnutarget} the @code{target} refers to a program, not a machine.
14746 @emph{Warning:} To specify a file format with @code{set gnutarget},
14747 you must know the actual BFD name.
14751 @xref{Files, , Commands to Specify Files}.
14753 @kindex show gnutarget
14754 @item show gnutarget
14755 Use the @code{show gnutarget} command to display what file format
14756 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
14757 @value{GDBN} will determine the file format for each file automatically,
14758 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
14761 @cindex common targets
14762 Here are some common targets (available, or not, depending on the GDB
14767 @item target exec @var{program}
14768 @cindex executable file target
14769 An executable file. @samp{target exec @var{program}} is the same as
14770 @samp{exec-file @var{program}}.
14772 @item target core @var{filename}
14773 @cindex core dump file target
14774 A core dump file. @samp{target core @var{filename}} is the same as
14775 @samp{core-file @var{filename}}.
14777 @item target remote @var{medium}
14778 @cindex remote target
14779 A remote system connected to @value{GDBN} via a serial line or network
14780 connection. This command tells @value{GDBN} to use its own remote
14781 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
14783 For example, if you have a board connected to @file{/dev/ttya} on the
14784 machine running @value{GDBN}, you could say:
14787 target remote /dev/ttya
14790 @code{target remote} supports the @code{load} command. This is only
14791 useful if you have some other way of getting the stub to the target
14792 system, and you can put it somewhere in memory where it won't get
14793 clobbered by the download.
14796 @cindex built-in simulator target
14797 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
14805 works; however, you cannot assume that a specific memory map, device
14806 drivers, or even basic I/O is available, although some simulators do
14807 provide these. For info about any processor-specific simulator details,
14808 see the appropriate section in @ref{Embedded Processors, ,Embedded
14813 Some configurations may include these targets as well:
14817 @item target nrom @var{dev}
14818 @cindex NetROM ROM emulator target
14819 NetROM ROM emulator. This target only supports downloading.
14823 Different targets are available on different configurations of @value{GDBN};
14824 your configuration may have more or fewer targets.
14826 Many remote targets require you to download the executable's code once
14827 you've successfully established a connection. You may wish to control
14828 various aspects of this process.
14833 @kindex set hash@r{, for remote monitors}
14834 @cindex hash mark while downloading
14835 This command controls whether a hash mark @samp{#} is displayed while
14836 downloading a file to the remote monitor. If on, a hash mark is
14837 displayed after each S-record is successfully downloaded to the
14841 @kindex show hash@r{, for remote monitors}
14842 Show the current status of displaying the hash mark.
14844 @item set debug monitor
14845 @kindex set debug monitor
14846 @cindex display remote monitor communications
14847 Enable or disable display of communications messages between
14848 @value{GDBN} and the remote monitor.
14850 @item show debug monitor
14851 @kindex show debug monitor
14852 Show the current status of displaying communications between
14853 @value{GDBN} and the remote monitor.
14858 @kindex load @var{filename}
14859 @item load @var{filename}
14861 Depending on what remote debugging facilities are configured into
14862 @value{GDBN}, the @code{load} command may be available. Where it exists, it
14863 is meant to make @var{filename} (an executable) available for debugging
14864 on the remote system---by downloading, or dynamic linking, for example.
14865 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
14866 the @code{add-symbol-file} command.
14868 If your @value{GDBN} does not have a @code{load} command, attempting to
14869 execute it gets the error message ``@code{You can't do that when your
14870 target is @dots{}}''
14872 The file is loaded at whatever address is specified in the executable.
14873 For some object file formats, you can specify the load address when you
14874 link the program; for other formats, like a.out, the object file format
14875 specifies a fixed address.
14876 @c FIXME! This would be a good place for an xref to the GNU linker doc.
14878 Depending on the remote side capabilities, @value{GDBN} may be able to
14879 load programs into flash memory.
14881 @code{load} does not repeat if you press @key{RET} again after using it.
14885 @section Choosing Target Byte Order
14887 @cindex choosing target byte order
14888 @cindex target byte order
14890 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
14891 offer the ability to run either big-endian or little-endian byte
14892 orders. Usually the executable or symbol will include a bit to
14893 designate the endian-ness, and you will not need to worry about
14894 which to use. However, you may still find it useful to adjust
14895 @value{GDBN}'s idea of processor endian-ness manually.
14899 @item set endian big
14900 Instruct @value{GDBN} to assume the target is big-endian.
14902 @item set endian little
14903 Instruct @value{GDBN} to assume the target is little-endian.
14905 @item set endian auto
14906 Instruct @value{GDBN} to use the byte order associated with the
14910 Display @value{GDBN}'s current idea of the target byte order.
14914 Note that these commands merely adjust interpretation of symbolic
14915 data on the host, and that they have absolutely no effect on the
14919 @node Remote Debugging
14920 @chapter Debugging Remote Programs
14921 @cindex remote debugging
14923 If you are trying to debug a program running on a machine that cannot run
14924 @value{GDBN} in the usual way, it is often useful to use remote debugging.
14925 For example, you might use remote debugging on an operating system kernel,
14926 or on a small system which does not have a general purpose operating system
14927 powerful enough to run a full-featured debugger.
14929 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
14930 to make this work with particular debugging targets. In addition,
14931 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
14932 but not specific to any particular target system) which you can use if you
14933 write the remote stubs---the code that runs on the remote system to
14934 communicate with @value{GDBN}.
14936 Other remote targets may be available in your
14937 configuration of @value{GDBN}; use @code{help target} to list them.
14940 * Connecting:: Connecting to a remote target
14941 * File Transfer:: Sending files to a remote system
14942 * Server:: Using the gdbserver program
14943 * Remote Configuration:: Remote configuration
14944 * Remote Stub:: Implementing a remote stub
14948 @section Connecting to a Remote Target
14950 On the @value{GDBN} host machine, you will need an unstripped copy of
14951 your program, since @value{GDBN} needs symbol and debugging information.
14952 Start up @value{GDBN} as usual, using the name of the local copy of your
14953 program as the first argument.
14955 @cindex @code{target remote}
14956 @value{GDBN} can communicate with the target over a serial line, or
14957 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
14958 each case, @value{GDBN} uses the same protocol for debugging your
14959 program; only the medium carrying the debugging packets varies. The
14960 @code{target remote} command establishes a connection to the target.
14961 Its arguments indicate which medium to use:
14965 @item target remote @var{serial-device}
14966 @cindex serial line, @code{target remote}
14967 Use @var{serial-device} to communicate with the target. For example,
14968 to use a serial line connected to the device named @file{/dev/ttyb}:
14971 target remote /dev/ttyb
14974 If you're using a serial line, you may want to give @value{GDBN} the
14975 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
14976 (@pxref{Remote Configuration, set remotebaud}) before the
14977 @code{target} command.
14979 @item target remote @code{@var{host}:@var{port}}
14980 @itemx target remote @code{tcp:@var{host}:@var{port}}
14981 @cindex @acronym{TCP} port, @code{target remote}
14982 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
14983 The @var{host} may be either a host name or a numeric @acronym{IP}
14984 address; @var{port} must be a decimal number. The @var{host} could be
14985 the target machine itself, if it is directly connected to the net, or
14986 it might be a terminal server which in turn has a serial line to the
14989 For example, to connect to port 2828 on a terminal server named
14993 target remote manyfarms:2828
14996 If your remote target is actually running on the same machine as your
14997 debugger session (e.g.@: a simulator for your target running on the
14998 same host), you can omit the hostname. For example, to connect to
14999 port 1234 on your local machine:
15002 target remote :1234
15006 Note that the colon is still required here.
15008 @item target remote @code{udp:@var{host}:@var{port}}
15009 @cindex @acronym{UDP} port, @code{target remote}
15010 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
15011 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
15014 target remote udp:manyfarms:2828
15017 When using a @acronym{UDP} connection for remote debugging, you should
15018 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
15019 can silently drop packets on busy or unreliable networks, which will
15020 cause havoc with your debugging session.
15022 @item target remote | @var{command}
15023 @cindex pipe, @code{target remote} to
15024 Run @var{command} in the background and communicate with it using a
15025 pipe. The @var{command} is a shell command, to be parsed and expanded
15026 by the system's command shell, @code{/bin/sh}; it should expect remote
15027 protocol packets on its standard input, and send replies on its
15028 standard output. You could use this to run a stand-alone simulator
15029 that speaks the remote debugging protocol, to make net connections
15030 using programs like @code{ssh}, or for other similar tricks.
15032 If @var{command} closes its standard output (perhaps by exiting),
15033 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
15034 program has already exited, this will have no effect.)
15038 Once the connection has been established, you can use all the usual
15039 commands to examine and change data. The remote program is already
15040 running; you can use @kbd{step} and @kbd{continue}, and you do not
15041 need to use @kbd{run}.
15043 @cindex interrupting remote programs
15044 @cindex remote programs, interrupting
15045 Whenever @value{GDBN} is waiting for the remote program, if you type the
15046 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
15047 program. This may or may not succeed, depending in part on the hardware
15048 and the serial drivers the remote system uses. If you type the
15049 interrupt character once again, @value{GDBN} displays this prompt:
15052 Interrupted while waiting for the program.
15053 Give up (and stop debugging it)? (y or n)
15056 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
15057 (If you decide you want to try again later, you can use @samp{target
15058 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
15059 goes back to waiting.
15062 @kindex detach (remote)
15064 When you have finished debugging the remote program, you can use the
15065 @code{detach} command to release it from @value{GDBN} control.
15066 Detaching from the target normally resumes its execution, but the results
15067 will depend on your particular remote stub. After the @code{detach}
15068 command, @value{GDBN} is free to connect to another target.
15072 The @code{disconnect} command behaves like @code{detach}, except that
15073 the target is generally not resumed. It will wait for @value{GDBN}
15074 (this instance or another one) to connect and continue debugging. After
15075 the @code{disconnect} command, @value{GDBN} is again free to connect to
15078 @cindex send command to remote monitor
15079 @cindex extend @value{GDBN} for remote targets
15080 @cindex add new commands for external monitor
15082 @item monitor @var{cmd}
15083 This command allows you to send arbitrary commands directly to the
15084 remote monitor. Since @value{GDBN} doesn't care about the commands it
15085 sends like this, this command is the way to extend @value{GDBN}---you
15086 can add new commands that only the external monitor will understand
15090 @node File Transfer
15091 @section Sending files to a remote system
15092 @cindex remote target, file transfer
15093 @cindex file transfer
15094 @cindex sending files to remote systems
15096 Some remote targets offer the ability to transfer files over the same
15097 connection used to communicate with @value{GDBN}. This is convenient
15098 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
15099 running @code{gdbserver} over a network interface. For other targets,
15100 e.g.@: embedded devices with only a single serial port, this may be
15101 the only way to upload or download files.
15103 Not all remote targets support these commands.
15107 @item remote put @var{hostfile} @var{targetfile}
15108 Copy file @var{hostfile} from the host system (the machine running
15109 @value{GDBN}) to @var{targetfile} on the target system.
15112 @item remote get @var{targetfile} @var{hostfile}
15113 Copy file @var{targetfile} from the target system to @var{hostfile}
15114 on the host system.
15116 @kindex remote delete
15117 @item remote delete @var{targetfile}
15118 Delete @var{targetfile} from the target system.
15123 @section Using the @code{gdbserver} Program
15126 @cindex remote connection without stubs
15127 @code{gdbserver} is a control program for Unix-like systems, which
15128 allows you to connect your program with a remote @value{GDBN} via
15129 @code{target remote}---but without linking in the usual debugging stub.
15131 @code{gdbserver} is not a complete replacement for the debugging stubs,
15132 because it requires essentially the same operating-system facilities
15133 that @value{GDBN} itself does. In fact, a system that can run
15134 @code{gdbserver} to connect to a remote @value{GDBN} could also run
15135 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
15136 because it is a much smaller program than @value{GDBN} itself. It is
15137 also easier to port than all of @value{GDBN}, so you may be able to get
15138 started more quickly on a new system by using @code{gdbserver}.
15139 Finally, if you develop code for real-time systems, you may find that
15140 the tradeoffs involved in real-time operation make it more convenient to
15141 do as much development work as possible on another system, for example
15142 by cross-compiling. You can use @code{gdbserver} to make a similar
15143 choice for debugging.
15145 @value{GDBN} and @code{gdbserver} communicate via either a serial line
15146 or a TCP connection, using the standard @value{GDBN} remote serial
15150 @emph{Warning:} @code{gdbserver} does not have any built-in security.
15151 Do not run @code{gdbserver} connected to any public network; a
15152 @value{GDBN} connection to @code{gdbserver} provides access to the
15153 target system with the same privileges as the user running
15157 @subsection Running @code{gdbserver}
15158 @cindex arguments, to @code{gdbserver}
15160 Run @code{gdbserver} on the target system. You need a copy of the
15161 program you want to debug, including any libraries it requires.
15162 @code{gdbserver} does not need your program's symbol table, so you can
15163 strip the program if necessary to save space. @value{GDBN} on the host
15164 system does all the symbol handling.
15166 To use the server, you must tell it how to communicate with @value{GDBN};
15167 the name of your program; and the arguments for your program. The usual
15171 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
15174 @var{comm} is either a device name (to use a serial line) or a TCP
15175 hostname and portnumber. For example, to debug Emacs with the argument
15176 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
15180 target> gdbserver /dev/com1 emacs foo.txt
15183 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
15186 To use a TCP connection instead of a serial line:
15189 target> gdbserver host:2345 emacs foo.txt
15192 The only difference from the previous example is the first argument,
15193 specifying that you are communicating with the host @value{GDBN} via
15194 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
15195 expect a TCP connection from machine @samp{host} to local TCP port 2345.
15196 (Currently, the @samp{host} part is ignored.) You can choose any number
15197 you want for the port number as long as it does not conflict with any
15198 TCP ports already in use on the target system (for example, @code{23} is
15199 reserved for @code{telnet}).@footnote{If you choose a port number that
15200 conflicts with another service, @code{gdbserver} prints an error message
15201 and exits.} You must use the same port number with the host @value{GDBN}
15202 @code{target remote} command.
15204 @subsubsection Attaching to a Running Program
15206 On some targets, @code{gdbserver} can also attach to running programs.
15207 This is accomplished via the @code{--attach} argument. The syntax is:
15210 target> gdbserver --attach @var{comm} @var{pid}
15213 @var{pid} is the process ID of a currently running process. It isn't necessary
15214 to point @code{gdbserver} at a binary for the running process.
15217 @cindex attach to a program by name
15218 You can debug processes by name instead of process ID if your target has the
15219 @code{pidof} utility:
15222 target> gdbserver --attach @var{comm} `pidof @var{program}`
15225 In case more than one copy of @var{program} is running, or @var{program}
15226 has multiple threads, most versions of @code{pidof} support the
15227 @code{-s} option to only return the first process ID.
15229 @subsubsection Multi-Process Mode for @code{gdbserver}
15230 @cindex gdbserver, multiple processes
15231 @cindex multiple processes with gdbserver
15233 When you connect to @code{gdbserver} using @code{target remote},
15234 @code{gdbserver} debugs the specified program only once. When the
15235 program exits, or you detach from it, @value{GDBN} closes the connection
15236 and @code{gdbserver} exits.
15238 If you connect using @kbd{target extended-remote}, @code{gdbserver}
15239 enters multi-process mode. When the debugged program exits, or you
15240 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
15241 though no program is running. The @code{run} and @code{attach}
15242 commands instruct @code{gdbserver} to run or attach to a new program.
15243 The @code{run} command uses @code{set remote exec-file} (@pxref{set
15244 remote exec-file}) to select the program to run. Command line
15245 arguments are supported, except for wildcard expansion and I/O
15246 redirection (@pxref{Arguments}).
15248 To start @code{gdbserver} without supplying an initial command to run
15249 or process ID to attach, use the @option{--multi} command line option.
15250 Then you can connect using @kbd{target extended-remote} and start
15251 the program you want to debug.
15253 @code{gdbserver} does not automatically exit in multi-process mode.
15254 You can terminate it by using @code{monitor exit}
15255 (@pxref{Monitor Commands for gdbserver}).
15257 @subsubsection Other Command-Line Arguments for @code{gdbserver}
15259 The @option{--debug} option tells @code{gdbserver} to display extra
15260 status information about the debugging process. The
15261 @option{--remote-debug} option tells @code{gdbserver} to display
15262 remote protocol debug output. These options are intended for
15263 @code{gdbserver} development and for bug reports to the developers.
15265 The @option{--wrapper} option specifies a wrapper to launch programs
15266 for debugging. The option should be followed by the name of the
15267 wrapper, then any command-line arguments to pass to the wrapper, then
15268 @kbd{--} indicating the end of the wrapper arguments.
15270 @code{gdbserver} runs the specified wrapper program with a combined
15271 command line including the wrapper arguments, then the name of the
15272 program to debug, then any arguments to the program. The wrapper
15273 runs until it executes your program, and then @value{GDBN} gains control.
15275 You can use any program that eventually calls @code{execve} with
15276 its arguments as a wrapper. Several standard Unix utilities do
15277 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
15278 with @code{exec "$@@"} will also work.
15280 For example, you can use @code{env} to pass an environment variable to
15281 the debugged program, without setting the variable in @code{gdbserver}'s
15285 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
15288 @subsection Connecting to @code{gdbserver}
15290 Run @value{GDBN} on the host system.
15292 First make sure you have the necessary symbol files. Load symbols for
15293 your application using the @code{file} command before you connect. Use
15294 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
15295 was compiled with the correct sysroot using @code{--with-sysroot}).
15297 The symbol file and target libraries must exactly match the executable
15298 and libraries on the target, with one exception: the files on the host
15299 system should not be stripped, even if the files on the target system
15300 are. Mismatched or missing files will lead to confusing results
15301 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
15302 files may also prevent @code{gdbserver} from debugging multi-threaded
15305 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
15306 For TCP connections, you must start up @code{gdbserver} prior to using
15307 the @code{target remote} command. Otherwise you may get an error whose
15308 text depends on the host system, but which usually looks something like
15309 @samp{Connection refused}. Don't use the @code{load}
15310 command in @value{GDBN} when using @code{gdbserver}, since the program is
15311 already on the target.
15313 @subsection Monitor Commands for @code{gdbserver}
15314 @cindex monitor commands, for @code{gdbserver}
15315 @anchor{Monitor Commands for gdbserver}
15317 During a @value{GDBN} session using @code{gdbserver}, you can use the
15318 @code{monitor} command to send special requests to @code{gdbserver}.
15319 Here are the available commands.
15323 List the available monitor commands.
15325 @item monitor set debug 0
15326 @itemx monitor set debug 1
15327 Disable or enable general debugging messages.
15329 @item monitor set remote-debug 0
15330 @itemx monitor set remote-debug 1
15331 Disable or enable specific debugging messages associated with the remote
15332 protocol (@pxref{Remote Protocol}).
15334 @item monitor set libthread-db-search-path [PATH]
15335 @cindex gdbserver, search path for @code{libthread_db}
15336 When this command is issued, @var{path} is a colon-separated list of
15337 directories to search for @code{libthread_db} (@pxref{Threads,,set
15338 libthread-db-search-path}). If you omit @var{path},
15339 @samp{libthread-db-search-path} will be reset to an empty list.
15342 Tell gdbserver to exit immediately. This command should be followed by
15343 @code{disconnect} to close the debugging session. @code{gdbserver} will
15344 detach from any attached processes and kill any processes it created.
15345 Use @code{monitor exit} to terminate @code{gdbserver} at the end
15346 of a multi-process mode debug session.
15350 @node Remote Configuration
15351 @section Remote Configuration
15354 @kindex show remote
15355 This section documents the configuration options available when
15356 debugging remote programs. For the options related to the File I/O
15357 extensions of the remote protocol, see @ref{system,
15358 system-call-allowed}.
15361 @item set remoteaddresssize @var{bits}
15362 @cindex address size for remote targets
15363 @cindex bits in remote address
15364 Set the maximum size of address in a memory packet to the specified
15365 number of bits. @value{GDBN} will mask off the address bits above
15366 that number, when it passes addresses to the remote target. The
15367 default value is the number of bits in the target's address.
15369 @item show remoteaddresssize
15370 Show the current value of remote address size in bits.
15372 @item set remotebaud @var{n}
15373 @cindex baud rate for remote targets
15374 Set the baud rate for the remote serial I/O to @var{n} baud. The
15375 value is used to set the speed of the serial port used for debugging
15378 @item show remotebaud
15379 Show the current speed of the remote connection.
15381 @item set remotebreak
15382 @cindex interrupt remote programs
15383 @cindex BREAK signal instead of Ctrl-C
15384 @anchor{set remotebreak}
15385 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
15386 when you type @kbd{Ctrl-c} to interrupt the program running
15387 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
15388 character instead. The default is off, since most remote systems
15389 expect to see @samp{Ctrl-C} as the interrupt signal.
15391 @item show remotebreak
15392 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
15393 interrupt the remote program.
15395 @item set remoteflow on
15396 @itemx set remoteflow off
15397 @kindex set remoteflow
15398 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
15399 on the serial port used to communicate to the remote target.
15401 @item show remoteflow
15402 @kindex show remoteflow
15403 Show the current setting of hardware flow control.
15405 @item set remotelogbase @var{base}
15406 Set the base (a.k.a.@: radix) of logging serial protocol
15407 communications to @var{base}. Supported values of @var{base} are:
15408 @code{ascii}, @code{octal}, and @code{hex}. The default is
15411 @item show remotelogbase
15412 Show the current setting of the radix for logging remote serial
15415 @item set remotelogfile @var{file}
15416 @cindex record serial communications on file
15417 Record remote serial communications on the named @var{file}. The
15418 default is not to record at all.
15420 @item show remotelogfile.
15421 Show the current setting of the file name on which to record the
15422 serial communications.
15424 @item set remotetimeout @var{num}
15425 @cindex timeout for serial communications
15426 @cindex remote timeout
15427 Set the timeout limit to wait for the remote target to respond to
15428 @var{num} seconds. The default is 2 seconds.
15430 @item show remotetimeout
15431 Show the current number of seconds to wait for the remote target
15434 @cindex limit hardware breakpoints and watchpoints
15435 @cindex remote target, limit break- and watchpoints
15436 @anchor{set remote hardware-watchpoint-limit}
15437 @anchor{set remote hardware-breakpoint-limit}
15438 @item set remote hardware-watchpoint-limit @var{limit}
15439 @itemx set remote hardware-breakpoint-limit @var{limit}
15440 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
15441 watchpoints. A limit of -1, the default, is treated as unlimited.
15443 @item set remote exec-file @var{filename}
15444 @itemx show remote exec-file
15445 @anchor{set remote exec-file}
15446 @cindex executable file, for remote target
15447 Select the file used for @code{run} with @code{target
15448 extended-remote}. This should be set to a filename valid on the
15449 target system. If it is not set, the target will use a default
15450 filename (e.g.@: the last program run).
15452 @item set remote interrupt-sequence
15453 @cindex interrupt remote programs
15454 @cindex select Ctrl-C, BREAK or BREAK-g
15455 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
15456 @samp{BREAK-g} as the
15457 sequence to the remote target in order to interrupt the execution.
15458 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
15459 is high level of serial line for some certain time.
15460 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
15461 It is @code{BREAK} signal followed by character @code{g}.
15463 @item show interrupt-sequence
15464 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
15465 is sent by @value{GDBN} to interrupt the remote program.
15466 @code{BREAK-g} is BREAK signal followed by @code{g} and
15467 also known as Magic SysRq g.
15469 @item set remote interrupt-on-connect
15470 @cindex send interrupt-sequence on start
15471 Specify whether interrupt-sequence is sent to remote target when
15472 @value{GDBN} connects to it. This is mostly needed when you debug
15473 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
15474 which is known as Magic SysRq g in order to connect @value{GDBN}.
15476 @item show interrupt-on-connect
15477 Show whether interrupt-sequence is sent
15478 to remote target when @value{GDBN} connects to it.
15482 @item set tcp auto-retry on
15483 @cindex auto-retry, for remote TCP target
15484 Enable auto-retry for remote TCP connections. This is useful if the remote
15485 debugging agent is launched in parallel with @value{GDBN}; there is a race
15486 condition because the agent may not become ready to accept the connection
15487 before @value{GDBN} attempts to connect. When auto-retry is
15488 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
15489 to establish the connection using the timeout specified by
15490 @code{set tcp connect-timeout}.
15492 @item set tcp auto-retry off
15493 Do not auto-retry failed TCP connections.
15495 @item show tcp auto-retry
15496 Show the current auto-retry setting.
15498 @item set tcp connect-timeout @var{seconds}
15499 @cindex connection timeout, for remote TCP target
15500 @cindex timeout, for remote target connection
15501 Set the timeout for establishing a TCP connection to the remote target to
15502 @var{seconds}. The timeout affects both polling to retry failed connections
15503 (enabled by @code{set tcp auto-retry on}) and waiting for connections
15504 that are merely slow to complete, and represents an approximate cumulative
15507 @item show tcp connect-timeout
15508 Show the current connection timeout setting.
15511 @cindex remote packets, enabling and disabling
15512 The @value{GDBN} remote protocol autodetects the packets supported by
15513 your debugging stub. If you need to override the autodetection, you
15514 can use these commands to enable or disable individual packets. Each
15515 packet can be set to @samp{on} (the remote target supports this
15516 packet), @samp{off} (the remote target does not support this packet),
15517 or @samp{auto} (detect remote target support for this packet). They
15518 all default to @samp{auto}. For more information about each packet,
15519 see @ref{Remote Protocol}.
15521 During normal use, you should not have to use any of these commands.
15522 If you do, that may be a bug in your remote debugging stub, or a bug
15523 in @value{GDBN}. You may want to report the problem to the
15524 @value{GDBN} developers.
15526 For each packet @var{name}, the command to enable or disable the
15527 packet is @code{set remote @var{name}-packet}. The available settings
15530 @multitable @columnfractions 0.28 0.32 0.25
15533 @tab Related Features
15535 @item @code{fetch-register}
15537 @tab @code{info registers}
15539 @item @code{set-register}
15543 @item @code{binary-download}
15545 @tab @code{load}, @code{set}
15547 @item @code{read-aux-vector}
15548 @tab @code{qXfer:auxv:read}
15549 @tab @code{info auxv}
15551 @item @code{symbol-lookup}
15552 @tab @code{qSymbol}
15553 @tab Detecting multiple threads
15555 @item @code{attach}
15556 @tab @code{vAttach}
15559 @item @code{verbose-resume}
15561 @tab Stepping or resuming multiple threads
15567 @item @code{software-breakpoint}
15571 @item @code{hardware-breakpoint}
15575 @item @code{write-watchpoint}
15579 @item @code{read-watchpoint}
15583 @item @code{access-watchpoint}
15587 @item @code{target-features}
15588 @tab @code{qXfer:features:read}
15589 @tab @code{set architecture}
15591 @item @code{library-info}
15592 @tab @code{qXfer:libraries:read}
15593 @tab @code{info sharedlibrary}
15595 @item @code{memory-map}
15596 @tab @code{qXfer:memory-map:read}
15597 @tab @code{info mem}
15599 @item @code{read-spu-object}
15600 @tab @code{qXfer:spu:read}
15601 @tab @code{info spu}
15603 @item @code{write-spu-object}
15604 @tab @code{qXfer:spu:write}
15605 @tab @code{info spu}
15607 @item @code{read-siginfo-object}
15608 @tab @code{qXfer:siginfo:read}
15609 @tab @code{print $_siginfo}
15611 @item @code{write-siginfo-object}
15612 @tab @code{qXfer:siginfo:write}
15613 @tab @code{set $_siginfo}
15615 @item @code{threads}
15616 @tab @code{qXfer:threads:read}
15617 @tab @code{info threads}
15619 @item @code{get-thread-local-@*storage-address}
15620 @tab @code{qGetTLSAddr}
15621 @tab Displaying @code{__thread} variables
15623 @item @code{search-memory}
15624 @tab @code{qSearch:memory}
15627 @item @code{supported-packets}
15628 @tab @code{qSupported}
15629 @tab Remote communications parameters
15631 @item @code{pass-signals}
15632 @tab @code{QPassSignals}
15633 @tab @code{handle @var{signal}}
15635 @item @code{hostio-close-packet}
15636 @tab @code{vFile:close}
15637 @tab @code{remote get}, @code{remote put}
15639 @item @code{hostio-open-packet}
15640 @tab @code{vFile:open}
15641 @tab @code{remote get}, @code{remote put}
15643 @item @code{hostio-pread-packet}
15644 @tab @code{vFile:pread}
15645 @tab @code{remote get}, @code{remote put}
15647 @item @code{hostio-pwrite-packet}
15648 @tab @code{vFile:pwrite}
15649 @tab @code{remote get}, @code{remote put}
15651 @item @code{hostio-unlink-packet}
15652 @tab @code{vFile:unlink}
15653 @tab @code{remote delete}
15655 @item @code{noack-packet}
15656 @tab @code{QStartNoAckMode}
15657 @tab Packet acknowledgment
15659 @item @code{osdata}
15660 @tab @code{qXfer:osdata:read}
15661 @tab @code{info os}
15663 @item @code{query-attached}
15664 @tab @code{qAttached}
15665 @tab Querying remote process attach state.
15669 @section Implementing a Remote Stub
15671 @cindex debugging stub, example
15672 @cindex remote stub, example
15673 @cindex stub example, remote debugging
15674 The stub files provided with @value{GDBN} implement the target side of the
15675 communication protocol, and the @value{GDBN} side is implemented in the
15676 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
15677 these subroutines to communicate, and ignore the details. (If you're
15678 implementing your own stub file, you can still ignore the details: start
15679 with one of the existing stub files. @file{sparc-stub.c} is the best
15680 organized, and therefore the easiest to read.)
15682 @cindex remote serial debugging, overview
15683 To debug a program running on another machine (the debugging
15684 @dfn{target} machine), you must first arrange for all the usual
15685 prerequisites for the program to run by itself. For example, for a C
15690 A startup routine to set up the C runtime environment; these usually
15691 have a name like @file{crt0}. The startup routine may be supplied by
15692 your hardware supplier, or you may have to write your own.
15695 A C subroutine library to support your program's
15696 subroutine calls, notably managing input and output.
15699 A way of getting your program to the other machine---for example, a
15700 download program. These are often supplied by the hardware
15701 manufacturer, but you may have to write your own from hardware
15705 The next step is to arrange for your program to use a serial port to
15706 communicate with the machine where @value{GDBN} is running (the @dfn{host}
15707 machine). In general terms, the scheme looks like this:
15711 @value{GDBN} already understands how to use this protocol; when everything
15712 else is set up, you can simply use the @samp{target remote} command
15713 (@pxref{Targets,,Specifying a Debugging Target}).
15715 @item On the target,
15716 you must link with your program a few special-purpose subroutines that
15717 implement the @value{GDBN} remote serial protocol. The file containing these
15718 subroutines is called a @dfn{debugging stub}.
15720 On certain remote targets, you can use an auxiliary program
15721 @code{gdbserver} instead of linking a stub into your program.
15722 @xref{Server,,Using the @code{gdbserver} Program}, for details.
15725 The debugging stub is specific to the architecture of the remote
15726 machine; for example, use @file{sparc-stub.c} to debug programs on
15729 @cindex remote serial stub list
15730 These working remote stubs are distributed with @value{GDBN}:
15735 @cindex @file{i386-stub.c}
15738 For Intel 386 and compatible architectures.
15741 @cindex @file{m68k-stub.c}
15742 @cindex Motorola 680x0
15744 For Motorola 680x0 architectures.
15747 @cindex @file{sh-stub.c}
15750 For Renesas SH architectures.
15753 @cindex @file{sparc-stub.c}
15755 For @sc{sparc} architectures.
15757 @item sparcl-stub.c
15758 @cindex @file{sparcl-stub.c}
15761 For Fujitsu @sc{sparclite} architectures.
15765 The @file{README} file in the @value{GDBN} distribution may list other
15766 recently added stubs.
15769 * Stub Contents:: What the stub can do for you
15770 * Bootstrapping:: What you must do for the stub
15771 * Debug Session:: Putting it all together
15774 @node Stub Contents
15775 @subsection What the Stub Can Do for You
15777 @cindex remote serial stub
15778 The debugging stub for your architecture supplies these three
15782 @item set_debug_traps
15783 @findex set_debug_traps
15784 @cindex remote serial stub, initialization
15785 This routine arranges for @code{handle_exception} to run when your
15786 program stops. You must call this subroutine explicitly near the
15787 beginning of your program.
15789 @item handle_exception
15790 @findex handle_exception
15791 @cindex remote serial stub, main routine
15792 This is the central workhorse, but your program never calls it
15793 explicitly---the setup code arranges for @code{handle_exception} to
15794 run when a trap is triggered.
15796 @code{handle_exception} takes control when your program stops during
15797 execution (for example, on a breakpoint), and mediates communications
15798 with @value{GDBN} on the host machine. This is where the communications
15799 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
15800 representative on the target machine. It begins by sending summary
15801 information on the state of your program, then continues to execute,
15802 retrieving and transmitting any information @value{GDBN} needs, until you
15803 execute a @value{GDBN} command that makes your program resume; at that point,
15804 @code{handle_exception} returns control to your own code on the target
15808 @cindex @code{breakpoint} subroutine, remote
15809 Use this auxiliary subroutine to make your program contain a
15810 breakpoint. Depending on the particular situation, this may be the only
15811 way for @value{GDBN} to get control. For instance, if your target
15812 machine has some sort of interrupt button, you won't need to call this;
15813 pressing the interrupt button transfers control to
15814 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
15815 simply receiving characters on the serial port may also trigger a trap;
15816 again, in that situation, you don't need to call @code{breakpoint} from
15817 your own program---simply running @samp{target remote} from the host
15818 @value{GDBN} session gets control.
15820 Call @code{breakpoint} if none of these is true, or if you simply want
15821 to make certain your program stops at a predetermined point for the
15822 start of your debugging session.
15825 @node Bootstrapping
15826 @subsection What You Must Do for the Stub
15828 @cindex remote stub, support routines
15829 The debugging stubs that come with @value{GDBN} are set up for a particular
15830 chip architecture, but they have no information about the rest of your
15831 debugging target machine.
15833 First of all you need to tell the stub how to communicate with the
15837 @item int getDebugChar()
15838 @findex getDebugChar
15839 Write this subroutine to read a single character from the serial port.
15840 It may be identical to @code{getchar} for your target system; a
15841 different name is used to allow you to distinguish the two if you wish.
15843 @item void putDebugChar(int)
15844 @findex putDebugChar
15845 Write this subroutine to write a single character to the serial port.
15846 It may be identical to @code{putchar} for your target system; a
15847 different name is used to allow you to distinguish the two if you wish.
15850 @cindex control C, and remote debugging
15851 @cindex interrupting remote targets
15852 If you want @value{GDBN} to be able to stop your program while it is
15853 running, you need to use an interrupt-driven serial driver, and arrange
15854 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
15855 character). That is the character which @value{GDBN} uses to tell the
15856 remote system to stop.
15858 Getting the debugging target to return the proper status to @value{GDBN}
15859 probably requires changes to the standard stub; one quick and dirty way
15860 is to just execute a breakpoint instruction (the ``dirty'' part is that
15861 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
15863 Other routines you need to supply are:
15866 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
15867 @findex exceptionHandler
15868 Write this function to install @var{exception_address} in the exception
15869 handling tables. You need to do this because the stub does not have any
15870 way of knowing what the exception handling tables on your target system
15871 are like (for example, the processor's table might be in @sc{rom},
15872 containing entries which point to a table in @sc{ram}).
15873 @var{exception_number} is the exception number which should be changed;
15874 its meaning is architecture-dependent (for example, different numbers
15875 might represent divide by zero, misaligned access, etc). When this
15876 exception occurs, control should be transferred directly to
15877 @var{exception_address}, and the processor state (stack, registers,
15878 and so on) should be just as it is when a processor exception occurs. So if
15879 you want to use a jump instruction to reach @var{exception_address}, it
15880 should be a simple jump, not a jump to subroutine.
15882 For the 386, @var{exception_address} should be installed as an interrupt
15883 gate so that interrupts are masked while the handler runs. The gate
15884 should be at privilege level 0 (the most privileged level). The
15885 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
15886 help from @code{exceptionHandler}.
15888 @item void flush_i_cache()
15889 @findex flush_i_cache
15890 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
15891 instruction cache, if any, on your target machine. If there is no
15892 instruction cache, this subroutine may be a no-op.
15894 On target machines that have instruction caches, @value{GDBN} requires this
15895 function to make certain that the state of your program is stable.
15899 You must also make sure this library routine is available:
15902 @item void *memset(void *, int, int)
15904 This is the standard library function @code{memset} that sets an area of
15905 memory to a known value. If you have one of the free versions of
15906 @code{libc.a}, @code{memset} can be found there; otherwise, you must
15907 either obtain it from your hardware manufacturer, or write your own.
15910 If you do not use the GNU C compiler, you may need other standard
15911 library subroutines as well; this varies from one stub to another,
15912 but in general the stubs are likely to use any of the common library
15913 subroutines which @code{@value{NGCC}} generates as inline code.
15916 @node Debug Session
15917 @subsection Putting it All Together
15919 @cindex remote serial debugging summary
15920 In summary, when your program is ready to debug, you must follow these
15925 Make sure you have defined the supporting low-level routines
15926 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
15928 @code{getDebugChar}, @code{putDebugChar},
15929 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
15933 Insert these lines near the top of your program:
15941 For the 680x0 stub only, you need to provide a variable called
15942 @code{exceptionHook}. Normally you just use:
15945 void (*exceptionHook)() = 0;
15949 but if before calling @code{set_debug_traps}, you set it to point to a
15950 function in your program, that function is called when
15951 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
15952 error). The function indicated by @code{exceptionHook} is called with
15953 one parameter: an @code{int} which is the exception number.
15956 Compile and link together: your program, the @value{GDBN} debugging stub for
15957 your target architecture, and the supporting subroutines.
15960 Make sure you have a serial connection between your target machine and
15961 the @value{GDBN} host, and identify the serial port on the host.
15964 @c The "remote" target now provides a `load' command, so we should
15965 @c document that. FIXME.
15966 Download your program to your target machine (or get it there by
15967 whatever means the manufacturer provides), and start it.
15970 Start @value{GDBN} on the host, and connect to the target
15971 (@pxref{Connecting,,Connecting to a Remote Target}).
15975 @node Configurations
15976 @chapter Configuration-Specific Information
15978 While nearly all @value{GDBN} commands are available for all native and
15979 cross versions of the debugger, there are some exceptions. This chapter
15980 describes things that are only available in certain configurations.
15982 There are three major categories of configurations: native
15983 configurations, where the host and target are the same, embedded
15984 operating system configurations, which are usually the same for several
15985 different processor architectures, and bare embedded processors, which
15986 are quite different from each other.
15991 * Embedded Processors::
15998 This section describes details specific to particular native
16003 * BSD libkvm Interface:: Debugging BSD kernel memory images
16004 * SVR4 Process Information:: SVR4 process information
16005 * DJGPP Native:: Features specific to the DJGPP port
16006 * Cygwin Native:: Features specific to the Cygwin port
16007 * Hurd Native:: Features specific to @sc{gnu} Hurd
16008 * Neutrino:: Features specific to QNX Neutrino
16009 * Darwin:: Features specific to Darwin
16015 On HP-UX systems, if you refer to a function or variable name that
16016 begins with a dollar sign, @value{GDBN} searches for a user or system
16017 name first, before it searches for a convenience variable.
16020 @node BSD libkvm Interface
16021 @subsection BSD libkvm Interface
16024 @cindex kernel memory image
16025 @cindex kernel crash dump
16027 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
16028 interface that provides a uniform interface for accessing kernel virtual
16029 memory images, including live systems and crash dumps. @value{GDBN}
16030 uses this interface to allow you to debug live kernels and kernel crash
16031 dumps on many native BSD configurations. This is implemented as a
16032 special @code{kvm} debugging target. For debugging a live system, load
16033 the currently running kernel into @value{GDBN} and connect to the
16037 (@value{GDBP}) @b{target kvm}
16040 For debugging crash dumps, provide the file name of the crash dump as an
16044 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
16047 Once connected to the @code{kvm} target, the following commands are
16053 Set current context from the @dfn{Process Control Block} (PCB) address.
16056 Set current context from proc address. This command isn't available on
16057 modern FreeBSD systems.
16060 @node SVR4 Process Information
16061 @subsection SVR4 Process Information
16063 @cindex examine process image
16064 @cindex process info via @file{/proc}
16066 Many versions of SVR4 and compatible systems provide a facility called
16067 @samp{/proc} that can be used to examine the image of a running
16068 process using file-system subroutines. If @value{GDBN} is configured
16069 for an operating system with this facility, the command @code{info
16070 proc} is available to report information about the process running
16071 your program, or about any process running on your system. @code{info
16072 proc} works only on SVR4 systems that include the @code{procfs} code.
16073 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
16074 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
16080 @itemx info proc @var{process-id}
16081 Summarize available information about any running process. If a
16082 process ID is specified by @var{process-id}, display information about
16083 that process; otherwise display information about the program being
16084 debugged. The summary includes the debugged process ID, the command
16085 line used to invoke it, its current working directory, and its
16086 executable file's absolute file name.
16088 On some systems, @var{process-id} can be of the form
16089 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
16090 within a process. If the optional @var{pid} part is missing, it means
16091 a thread from the process being debugged (the leading @samp{/} still
16092 needs to be present, or else @value{GDBN} will interpret the number as
16093 a process ID rather than a thread ID).
16095 @item info proc mappings
16096 @cindex memory address space mappings
16097 Report the memory address space ranges accessible in the program, with
16098 information on whether the process has read, write, or execute access
16099 rights to each range. On @sc{gnu}/Linux systems, each memory range
16100 includes the object file which is mapped to that range, instead of the
16101 memory access rights to that range.
16103 @item info proc stat
16104 @itemx info proc status
16105 @cindex process detailed status information
16106 These subcommands are specific to @sc{gnu}/Linux systems. They show
16107 the process-related information, including the user ID and group ID;
16108 how many threads are there in the process; its virtual memory usage;
16109 the signals that are pending, blocked, and ignored; its TTY; its
16110 consumption of system and user time; its stack size; its @samp{nice}
16111 value; etc. For more information, see the @samp{proc} man page
16112 (type @kbd{man 5 proc} from your shell prompt).
16114 @item info proc all
16115 Show all the information about the process described under all of the
16116 above @code{info proc} subcommands.
16119 @comment These sub-options of 'info proc' were not included when
16120 @comment procfs.c was re-written. Keep their descriptions around
16121 @comment against the day when someone finds the time to put them back in.
16122 @kindex info proc times
16123 @item info proc times
16124 Starting time, user CPU time, and system CPU time for your program and
16127 @kindex info proc id
16129 Report on the process IDs related to your program: its own process ID,
16130 the ID of its parent, the process group ID, and the session ID.
16133 @item set procfs-trace
16134 @kindex set procfs-trace
16135 @cindex @code{procfs} API calls
16136 This command enables and disables tracing of @code{procfs} API calls.
16138 @item show procfs-trace
16139 @kindex show procfs-trace
16140 Show the current state of @code{procfs} API call tracing.
16142 @item set procfs-file @var{file}
16143 @kindex set procfs-file
16144 Tell @value{GDBN} to write @code{procfs} API trace to the named
16145 @var{file}. @value{GDBN} appends the trace info to the previous
16146 contents of the file. The default is to display the trace on the
16149 @item show procfs-file
16150 @kindex show procfs-file
16151 Show the file to which @code{procfs} API trace is written.
16153 @item proc-trace-entry
16154 @itemx proc-trace-exit
16155 @itemx proc-untrace-entry
16156 @itemx proc-untrace-exit
16157 @kindex proc-trace-entry
16158 @kindex proc-trace-exit
16159 @kindex proc-untrace-entry
16160 @kindex proc-untrace-exit
16161 These commands enable and disable tracing of entries into and exits
16162 from the @code{syscall} interface.
16165 @kindex info pidlist
16166 @cindex process list, QNX Neutrino
16167 For QNX Neutrino only, this command displays the list of all the
16168 processes and all the threads within each process.
16171 @kindex info meminfo
16172 @cindex mapinfo list, QNX Neutrino
16173 For QNX Neutrino only, this command displays the list of all mapinfos.
16177 @subsection Features for Debugging @sc{djgpp} Programs
16178 @cindex @sc{djgpp} debugging
16179 @cindex native @sc{djgpp} debugging
16180 @cindex MS-DOS-specific commands
16183 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
16184 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
16185 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
16186 top of real-mode DOS systems and their emulations.
16188 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
16189 defines a few commands specific to the @sc{djgpp} port. This
16190 subsection describes those commands.
16195 This is a prefix of @sc{djgpp}-specific commands which print
16196 information about the target system and important OS structures.
16199 @cindex MS-DOS system info
16200 @cindex free memory information (MS-DOS)
16201 @item info dos sysinfo
16202 This command displays assorted information about the underlying
16203 platform: the CPU type and features, the OS version and flavor, the
16204 DPMI version, and the available conventional and DPMI memory.
16209 @cindex segment descriptor tables
16210 @cindex descriptor tables display
16212 @itemx info dos ldt
16213 @itemx info dos idt
16214 These 3 commands display entries from, respectively, Global, Local,
16215 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
16216 tables are data structures which store a descriptor for each segment
16217 that is currently in use. The segment's selector is an index into a
16218 descriptor table; the table entry for that index holds the
16219 descriptor's base address and limit, and its attributes and access
16222 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
16223 segment (used for both data and the stack), and a DOS segment (which
16224 allows access to DOS/BIOS data structures and absolute addresses in
16225 conventional memory). However, the DPMI host will usually define
16226 additional segments in order to support the DPMI environment.
16228 @cindex garbled pointers
16229 These commands allow to display entries from the descriptor tables.
16230 Without an argument, all entries from the specified table are
16231 displayed. An argument, which should be an integer expression, means
16232 display a single entry whose index is given by the argument. For
16233 example, here's a convenient way to display information about the
16234 debugged program's data segment:
16237 @exdent @code{(@value{GDBP}) info dos ldt $ds}
16238 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
16242 This comes in handy when you want to see whether a pointer is outside
16243 the data segment's limit (i.e.@: @dfn{garbled}).
16245 @cindex page tables display (MS-DOS)
16247 @itemx info dos pte
16248 These two commands display entries from, respectively, the Page
16249 Directory and the Page Tables. Page Directories and Page Tables are
16250 data structures which control how virtual memory addresses are mapped
16251 into physical addresses. A Page Table includes an entry for every
16252 page of memory that is mapped into the program's address space; there
16253 may be several Page Tables, each one holding up to 4096 entries. A
16254 Page Directory has up to 4096 entries, one each for every Page Table
16255 that is currently in use.
16257 Without an argument, @kbd{info dos pde} displays the entire Page
16258 Directory, and @kbd{info dos pte} displays all the entries in all of
16259 the Page Tables. An argument, an integer expression, given to the
16260 @kbd{info dos pde} command means display only that entry from the Page
16261 Directory table. An argument given to the @kbd{info dos pte} command
16262 means display entries from a single Page Table, the one pointed to by
16263 the specified entry in the Page Directory.
16265 @cindex direct memory access (DMA) on MS-DOS
16266 These commands are useful when your program uses @dfn{DMA} (Direct
16267 Memory Access), which needs physical addresses to program the DMA
16270 These commands are supported only with some DPMI servers.
16272 @cindex physical address from linear address
16273 @item info dos address-pte @var{addr}
16274 This command displays the Page Table entry for a specified linear
16275 address. The argument @var{addr} is a linear address which should
16276 already have the appropriate segment's base address added to it,
16277 because this command accepts addresses which may belong to @emph{any}
16278 segment. For example, here's how to display the Page Table entry for
16279 the page where a variable @code{i} is stored:
16282 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
16283 @exdent @code{Page Table entry for address 0x11a00d30:}
16284 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
16288 This says that @code{i} is stored at offset @code{0xd30} from the page
16289 whose physical base address is @code{0x02698000}, and shows all the
16290 attributes of that page.
16292 Note that you must cast the addresses of variables to a @code{char *},
16293 since otherwise the value of @code{__djgpp_base_address}, the base
16294 address of all variables and functions in a @sc{djgpp} program, will
16295 be added using the rules of C pointer arithmetics: if @code{i} is
16296 declared an @code{int}, @value{GDBN} will add 4 times the value of
16297 @code{__djgpp_base_address} to the address of @code{i}.
16299 Here's another example, it displays the Page Table entry for the
16303 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
16304 @exdent @code{Page Table entry for address 0x29110:}
16305 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
16309 (The @code{+ 3} offset is because the transfer buffer's address is the
16310 3rd member of the @code{_go32_info_block} structure.) The output
16311 clearly shows that this DPMI server maps the addresses in conventional
16312 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
16313 linear (@code{0x29110}) addresses are identical.
16315 This command is supported only with some DPMI servers.
16318 @cindex DOS serial data link, remote debugging
16319 In addition to native debugging, the DJGPP port supports remote
16320 debugging via a serial data link. The following commands are specific
16321 to remote serial debugging in the DJGPP port of @value{GDBN}.
16324 @kindex set com1base
16325 @kindex set com1irq
16326 @kindex set com2base
16327 @kindex set com2irq
16328 @kindex set com3base
16329 @kindex set com3irq
16330 @kindex set com4base
16331 @kindex set com4irq
16332 @item set com1base @var{addr}
16333 This command sets the base I/O port address of the @file{COM1} serial
16336 @item set com1irq @var{irq}
16337 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
16338 for the @file{COM1} serial port.
16340 There are similar commands @samp{set com2base}, @samp{set com3irq},
16341 etc.@: for setting the port address and the @code{IRQ} lines for the
16344 @kindex show com1base
16345 @kindex show com1irq
16346 @kindex show com2base
16347 @kindex show com2irq
16348 @kindex show com3base
16349 @kindex show com3irq
16350 @kindex show com4base
16351 @kindex show com4irq
16352 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
16353 display the current settings of the base address and the @code{IRQ}
16354 lines used by the COM ports.
16357 @kindex info serial
16358 @cindex DOS serial port status
16359 This command prints the status of the 4 DOS serial ports. For each
16360 port, it prints whether it's active or not, its I/O base address and
16361 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
16362 counts of various errors encountered so far.
16366 @node Cygwin Native
16367 @subsection Features for Debugging MS Windows PE Executables
16368 @cindex MS Windows debugging
16369 @cindex native Cygwin debugging
16370 @cindex Cygwin-specific commands
16372 @value{GDBN} supports native debugging of MS Windows programs, including
16373 DLLs with and without symbolic debugging information.
16375 @cindex Ctrl-BREAK, MS-Windows
16376 @cindex interrupt debuggee on MS-Windows
16377 MS-Windows programs that call @code{SetConsoleMode} to switch off the
16378 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
16379 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
16380 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
16381 sequence, which can be used to interrupt the debuggee even if it
16384 There are various additional Cygwin-specific commands, described in
16385 this section. Working with DLLs that have no debugging symbols is
16386 described in @ref{Non-debug DLL Symbols}.
16391 This is a prefix of MS Windows-specific commands which print
16392 information about the target system and important OS structures.
16394 @item info w32 selector
16395 This command displays information returned by
16396 the Win32 API @code{GetThreadSelectorEntry} function.
16397 It takes an optional argument that is evaluated to
16398 a long value to give the information about this given selector.
16399 Without argument, this command displays information
16400 about the six segment registers.
16404 This is a Cygwin-specific alias of @code{info shared}.
16406 @kindex dll-symbols
16408 This command loads symbols from a dll similarly to
16409 add-sym command but without the need to specify a base address.
16411 @kindex set cygwin-exceptions
16412 @cindex debugging the Cygwin DLL
16413 @cindex Cygwin DLL, debugging
16414 @item set cygwin-exceptions @var{mode}
16415 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
16416 happen inside the Cygwin DLL. If @var{mode} is @code{off},
16417 @value{GDBN} will delay recognition of exceptions, and may ignore some
16418 exceptions which seem to be caused by internal Cygwin DLL
16419 ``bookkeeping''. This option is meant primarily for debugging the
16420 Cygwin DLL itself; the default value is @code{off} to avoid annoying
16421 @value{GDBN} users with false @code{SIGSEGV} signals.
16423 @kindex show cygwin-exceptions
16424 @item show cygwin-exceptions
16425 Displays whether @value{GDBN} will break on exceptions that happen
16426 inside the Cygwin DLL itself.
16428 @kindex set new-console
16429 @item set new-console @var{mode}
16430 If @var{mode} is @code{on} the debuggee will
16431 be started in a new console on next start.
16432 If @var{mode} is @code{off}i, the debuggee will
16433 be started in the same console as the debugger.
16435 @kindex show new-console
16436 @item show new-console
16437 Displays whether a new console is used
16438 when the debuggee is started.
16440 @kindex set new-group
16441 @item set new-group @var{mode}
16442 This boolean value controls whether the debuggee should
16443 start a new group or stay in the same group as the debugger.
16444 This affects the way the Windows OS handles
16447 @kindex show new-group
16448 @item show new-group
16449 Displays current value of new-group boolean.
16451 @kindex set debugevents
16452 @item set debugevents
16453 This boolean value adds debug output concerning kernel events related
16454 to the debuggee seen by the debugger. This includes events that
16455 signal thread and process creation and exit, DLL loading and
16456 unloading, console interrupts, and debugging messages produced by the
16457 Windows @code{OutputDebugString} API call.
16459 @kindex set debugexec
16460 @item set debugexec
16461 This boolean value adds debug output concerning execute events
16462 (such as resume thread) seen by the debugger.
16464 @kindex set debugexceptions
16465 @item set debugexceptions
16466 This boolean value adds debug output concerning exceptions in the
16467 debuggee seen by the debugger.
16469 @kindex set debugmemory
16470 @item set debugmemory
16471 This boolean value adds debug output concerning debuggee memory reads
16472 and writes by the debugger.
16476 This boolean values specifies whether the debuggee is called
16477 via a shell or directly (default value is on).
16481 Displays if the debuggee will be started with a shell.
16486 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
16489 @node Non-debug DLL Symbols
16490 @subsubsection Support for DLLs without Debugging Symbols
16491 @cindex DLLs with no debugging symbols
16492 @cindex Minimal symbols and DLLs
16494 Very often on windows, some of the DLLs that your program relies on do
16495 not include symbolic debugging information (for example,
16496 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
16497 symbols in a DLL, it relies on the minimal amount of symbolic
16498 information contained in the DLL's export table. This section
16499 describes working with such symbols, known internally to @value{GDBN} as
16500 ``minimal symbols''.
16502 Note that before the debugged program has started execution, no DLLs
16503 will have been loaded. The easiest way around this problem is simply to
16504 start the program --- either by setting a breakpoint or letting the
16505 program run once to completion. It is also possible to force
16506 @value{GDBN} to load a particular DLL before starting the executable ---
16507 see the shared library information in @ref{Files}, or the
16508 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
16509 explicitly loading symbols from a DLL with no debugging information will
16510 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
16511 which may adversely affect symbol lookup performance.
16513 @subsubsection DLL Name Prefixes
16515 In keeping with the naming conventions used by the Microsoft debugging
16516 tools, DLL export symbols are made available with a prefix based on the
16517 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
16518 also entered into the symbol table, so @code{CreateFileA} is often
16519 sufficient. In some cases there will be name clashes within a program
16520 (particularly if the executable itself includes full debugging symbols)
16521 necessitating the use of the fully qualified name when referring to the
16522 contents of the DLL. Use single-quotes around the name to avoid the
16523 exclamation mark (``!'') being interpreted as a language operator.
16525 Note that the internal name of the DLL may be all upper-case, even
16526 though the file name of the DLL is lower-case, or vice-versa. Since
16527 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
16528 some confusion. If in doubt, try the @code{info functions} and
16529 @code{info variables} commands or even @code{maint print msymbols}
16530 (@pxref{Symbols}). Here's an example:
16533 (@value{GDBP}) info function CreateFileA
16534 All functions matching regular expression "CreateFileA":
16536 Non-debugging symbols:
16537 0x77e885f4 CreateFileA
16538 0x77e885f4 KERNEL32!CreateFileA
16542 (@value{GDBP}) info function !
16543 All functions matching regular expression "!":
16545 Non-debugging symbols:
16546 0x6100114c cygwin1!__assert
16547 0x61004034 cygwin1!_dll_crt0@@0
16548 0x61004240 cygwin1!dll_crt0(per_process *)
16552 @subsubsection Working with Minimal Symbols
16554 Symbols extracted from a DLL's export table do not contain very much
16555 type information. All that @value{GDBN} can do is guess whether a symbol
16556 refers to a function or variable depending on the linker section that
16557 contains the symbol. Also note that the actual contents of the memory
16558 contained in a DLL are not available unless the program is running. This
16559 means that you cannot examine the contents of a variable or disassemble
16560 a function within a DLL without a running program.
16562 Variables are generally treated as pointers and dereferenced
16563 automatically. For this reason, it is often necessary to prefix a
16564 variable name with the address-of operator (``&'') and provide explicit
16565 type information in the command. Here's an example of the type of
16569 (@value{GDBP}) print 'cygwin1!__argv'
16574 (@value{GDBP}) x 'cygwin1!__argv'
16575 0x10021610: "\230y\""
16578 And two possible solutions:
16581 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
16582 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
16586 (@value{GDBP}) x/2x &'cygwin1!__argv'
16587 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
16588 (@value{GDBP}) x/x 0x10021608
16589 0x10021608: 0x0022fd98
16590 (@value{GDBP}) x/s 0x0022fd98
16591 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
16594 Setting a break point within a DLL is possible even before the program
16595 starts execution. However, under these circumstances, @value{GDBN} can't
16596 examine the initial instructions of the function in order to skip the
16597 function's frame set-up code. You can work around this by using ``*&''
16598 to set the breakpoint at a raw memory address:
16601 (@value{GDBP}) break *&'python22!PyOS_Readline'
16602 Breakpoint 1 at 0x1e04eff0
16605 The author of these extensions is not entirely convinced that setting a
16606 break point within a shared DLL like @file{kernel32.dll} is completely
16610 @subsection Commands Specific to @sc{gnu} Hurd Systems
16611 @cindex @sc{gnu} Hurd debugging
16613 This subsection describes @value{GDBN} commands specific to the
16614 @sc{gnu} Hurd native debugging.
16619 @kindex set signals@r{, Hurd command}
16620 @kindex set sigs@r{, Hurd command}
16621 This command toggles the state of inferior signal interception by
16622 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
16623 affected by this command. @code{sigs} is a shorthand alias for
16628 @kindex show signals@r{, Hurd command}
16629 @kindex show sigs@r{, Hurd command}
16630 Show the current state of intercepting inferior's signals.
16632 @item set signal-thread
16633 @itemx set sigthread
16634 @kindex set signal-thread
16635 @kindex set sigthread
16636 This command tells @value{GDBN} which thread is the @code{libc} signal
16637 thread. That thread is run when a signal is delivered to a running
16638 process. @code{set sigthread} is the shorthand alias of @code{set
16641 @item show signal-thread
16642 @itemx show sigthread
16643 @kindex show signal-thread
16644 @kindex show sigthread
16645 These two commands show which thread will run when the inferior is
16646 delivered a signal.
16649 @kindex set stopped@r{, Hurd command}
16650 This commands tells @value{GDBN} that the inferior process is stopped,
16651 as with the @code{SIGSTOP} signal. The stopped process can be
16652 continued by delivering a signal to it.
16655 @kindex show stopped@r{, Hurd command}
16656 This command shows whether @value{GDBN} thinks the debuggee is
16659 @item set exceptions
16660 @kindex set exceptions@r{, Hurd command}
16661 Use this command to turn off trapping of exceptions in the inferior.
16662 When exception trapping is off, neither breakpoints nor
16663 single-stepping will work. To restore the default, set exception
16666 @item show exceptions
16667 @kindex show exceptions@r{, Hurd command}
16668 Show the current state of trapping exceptions in the inferior.
16670 @item set task pause
16671 @kindex set task@r{, Hurd commands}
16672 @cindex task attributes (@sc{gnu} Hurd)
16673 @cindex pause current task (@sc{gnu} Hurd)
16674 This command toggles task suspension when @value{GDBN} has control.
16675 Setting it to on takes effect immediately, and the task is suspended
16676 whenever @value{GDBN} gets control. Setting it to off will take
16677 effect the next time the inferior is continued. If this option is set
16678 to off, you can use @code{set thread default pause on} or @code{set
16679 thread pause on} (see below) to pause individual threads.
16681 @item show task pause
16682 @kindex show task@r{, Hurd commands}
16683 Show the current state of task suspension.
16685 @item set task detach-suspend-count
16686 @cindex task suspend count
16687 @cindex detach from task, @sc{gnu} Hurd
16688 This command sets the suspend count the task will be left with when
16689 @value{GDBN} detaches from it.
16691 @item show task detach-suspend-count
16692 Show the suspend count the task will be left with when detaching.
16694 @item set task exception-port
16695 @itemx set task excp
16696 @cindex task exception port, @sc{gnu} Hurd
16697 This command sets the task exception port to which @value{GDBN} will
16698 forward exceptions. The argument should be the value of the @dfn{send
16699 rights} of the task. @code{set task excp} is a shorthand alias.
16701 @item set noninvasive
16702 @cindex noninvasive task options
16703 This command switches @value{GDBN} to a mode that is the least
16704 invasive as far as interfering with the inferior is concerned. This
16705 is the same as using @code{set task pause}, @code{set exceptions}, and
16706 @code{set signals} to values opposite to the defaults.
16708 @item info send-rights
16709 @itemx info receive-rights
16710 @itemx info port-rights
16711 @itemx info port-sets
16712 @itemx info dead-names
16715 @cindex send rights, @sc{gnu} Hurd
16716 @cindex receive rights, @sc{gnu} Hurd
16717 @cindex port rights, @sc{gnu} Hurd
16718 @cindex port sets, @sc{gnu} Hurd
16719 @cindex dead names, @sc{gnu} Hurd
16720 These commands display information about, respectively, send rights,
16721 receive rights, port rights, port sets, and dead names of a task.
16722 There are also shorthand aliases: @code{info ports} for @code{info
16723 port-rights} and @code{info psets} for @code{info port-sets}.
16725 @item set thread pause
16726 @kindex set thread@r{, Hurd command}
16727 @cindex thread properties, @sc{gnu} Hurd
16728 @cindex pause current thread (@sc{gnu} Hurd)
16729 This command toggles current thread suspension when @value{GDBN} has
16730 control. Setting it to on takes effect immediately, and the current
16731 thread is suspended whenever @value{GDBN} gets control. Setting it to
16732 off will take effect the next time the inferior is continued.
16733 Normally, this command has no effect, since when @value{GDBN} has
16734 control, the whole task is suspended. However, if you used @code{set
16735 task pause off} (see above), this command comes in handy to suspend
16736 only the current thread.
16738 @item show thread pause
16739 @kindex show thread@r{, Hurd command}
16740 This command shows the state of current thread suspension.
16742 @item set thread run
16743 This command sets whether the current thread is allowed to run.
16745 @item show thread run
16746 Show whether the current thread is allowed to run.
16748 @item set thread detach-suspend-count
16749 @cindex thread suspend count, @sc{gnu} Hurd
16750 @cindex detach from thread, @sc{gnu} Hurd
16751 This command sets the suspend count @value{GDBN} will leave on a
16752 thread when detaching. This number is relative to the suspend count
16753 found by @value{GDBN} when it notices the thread; use @code{set thread
16754 takeover-suspend-count} to force it to an absolute value.
16756 @item show thread detach-suspend-count
16757 Show the suspend count @value{GDBN} will leave on the thread when
16760 @item set thread exception-port
16761 @itemx set thread excp
16762 Set the thread exception port to which to forward exceptions. This
16763 overrides the port set by @code{set task exception-port} (see above).
16764 @code{set thread excp} is the shorthand alias.
16766 @item set thread takeover-suspend-count
16767 Normally, @value{GDBN}'s thread suspend counts are relative to the
16768 value @value{GDBN} finds when it notices each thread. This command
16769 changes the suspend counts to be absolute instead.
16771 @item set thread default
16772 @itemx show thread default
16773 @cindex thread default settings, @sc{gnu} Hurd
16774 Each of the above @code{set thread} commands has a @code{set thread
16775 default} counterpart (e.g., @code{set thread default pause}, @code{set
16776 thread default exception-port}, etc.). The @code{thread default}
16777 variety of commands sets the default thread properties for all
16778 threads; you can then change the properties of individual threads with
16779 the non-default commands.
16784 @subsection QNX Neutrino
16785 @cindex QNX Neutrino
16787 @value{GDBN} provides the following commands specific to the QNX
16791 @item set debug nto-debug
16792 @kindex set debug nto-debug
16793 When set to on, enables debugging messages specific to the QNX
16796 @item show debug nto-debug
16797 @kindex show debug nto-debug
16798 Show the current state of QNX Neutrino messages.
16805 @value{GDBN} provides the following commands specific to the Darwin target:
16808 @item set debug darwin @var{num}
16809 @kindex set debug darwin
16810 When set to a non zero value, enables debugging messages specific to
16811 the Darwin support. Higher values produce more verbose output.
16813 @item show debug darwin
16814 @kindex show debug darwin
16815 Show the current state of Darwin messages.
16817 @item set debug mach-o @var{num}
16818 @kindex set debug mach-o
16819 When set to a non zero value, enables debugging messages while
16820 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
16821 file format used on Darwin for object and executable files.) Higher
16822 values produce more verbose output. This is a command to diagnose
16823 problems internal to @value{GDBN} and should not be needed in normal
16826 @item show debug mach-o
16827 @kindex show debug mach-o
16828 Show the current state of Mach-O file messages.
16830 @item set mach-exceptions on
16831 @itemx set mach-exceptions off
16832 @kindex set mach-exceptions
16833 On Darwin, faults are first reported as a Mach exception and are then
16834 mapped to a Posix signal. Use this command to turn on trapping of
16835 Mach exceptions in the inferior. This might be sometimes useful to
16836 better understand the cause of a fault. The default is off.
16838 @item show mach-exceptions
16839 @kindex show mach-exceptions
16840 Show the current state of exceptions trapping.
16845 @section Embedded Operating Systems
16847 This section describes configurations involving the debugging of
16848 embedded operating systems that are available for several different
16852 * VxWorks:: Using @value{GDBN} with VxWorks
16855 @value{GDBN} includes the ability to debug programs running on
16856 various real-time operating systems.
16859 @subsection Using @value{GDBN} with VxWorks
16865 @kindex target vxworks
16866 @item target vxworks @var{machinename}
16867 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
16868 is the target system's machine name or IP address.
16872 On VxWorks, @code{load} links @var{filename} dynamically on the
16873 current target system as well as adding its symbols in @value{GDBN}.
16875 @value{GDBN} enables developers to spawn and debug tasks running on networked
16876 VxWorks targets from a Unix host. Already-running tasks spawned from
16877 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
16878 both the Unix host and on the VxWorks target. The program
16879 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
16880 installed with the name @code{vxgdb}, to distinguish it from a
16881 @value{GDBN} for debugging programs on the host itself.)
16884 @item VxWorks-timeout @var{args}
16885 @kindex vxworks-timeout
16886 All VxWorks-based targets now support the option @code{vxworks-timeout}.
16887 This option is set by the user, and @var{args} represents the number of
16888 seconds @value{GDBN} waits for responses to rpc's. You might use this if
16889 your VxWorks target is a slow software simulator or is on the far side
16890 of a thin network line.
16893 The following information on connecting to VxWorks was current when
16894 this manual was produced; newer releases of VxWorks may use revised
16897 @findex INCLUDE_RDB
16898 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
16899 to include the remote debugging interface routines in the VxWorks
16900 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
16901 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
16902 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
16903 source debugging task @code{tRdbTask} when VxWorks is booted. For more
16904 information on configuring and remaking VxWorks, see the manufacturer's
16906 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
16908 Once you have included @file{rdb.a} in your VxWorks system image and set
16909 your Unix execution search path to find @value{GDBN}, you are ready to
16910 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
16911 @code{vxgdb}, depending on your installation).
16913 @value{GDBN} comes up showing the prompt:
16920 * VxWorks Connection:: Connecting to VxWorks
16921 * VxWorks Download:: VxWorks download
16922 * VxWorks Attach:: Running tasks
16925 @node VxWorks Connection
16926 @subsubsection Connecting to VxWorks
16928 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
16929 network. To connect to a target whose host name is ``@code{tt}'', type:
16932 (vxgdb) target vxworks tt
16936 @value{GDBN} displays messages like these:
16939 Attaching remote machine across net...
16944 @value{GDBN} then attempts to read the symbol tables of any object modules
16945 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
16946 these files by searching the directories listed in the command search
16947 path (@pxref{Environment, ,Your Program's Environment}); if it fails
16948 to find an object file, it displays a message such as:
16951 prog.o: No such file or directory.
16954 When this happens, add the appropriate directory to the search path with
16955 the @value{GDBN} command @code{path}, and execute the @code{target}
16958 @node VxWorks Download
16959 @subsubsection VxWorks Download
16961 @cindex download to VxWorks
16962 If you have connected to the VxWorks target and you want to debug an
16963 object that has not yet been loaded, you can use the @value{GDBN}
16964 @code{load} command to download a file from Unix to VxWorks
16965 incrementally. The object file given as an argument to the @code{load}
16966 command is actually opened twice: first by the VxWorks target in order
16967 to download the code, then by @value{GDBN} in order to read the symbol
16968 table. This can lead to problems if the current working directories on
16969 the two systems differ. If both systems have NFS mounted the same
16970 filesystems, you can avoid these problems by using absolute paths.
16971 Otherwise, it is simplest to set the working directory on both systems
16972 to the directory in which the object file resides, and then to reference
16973 the file by its name, without any path. For instance, a program
16974 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
16975 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
16976 program, type this on VxWorks:
16979 -> cd "@var{vxpath}/vw/demo/rdb"
16983 Then, in @value{GDBN}, type:
16986 (vxgdb) cd @var{hostpath}/vw/demo/rdb
16987 (vxgdb) load prog.o
16990 @value{GDBN} displays a response similar to this:
16993 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
16996 You can also use the @code{load} command to reload an object module
16997 after editing and recompiling the corresponding source file. Note that
16998 this makes @value{GDBN} delete all currently-defined breakpoints,
16999 auto-displays, and convenience variables, and to clear the value
17000 history. (This is necessary in order to preserve the integrity of
17001 debugger's data structures that reference the target system's symbol
17004 @node VxWorks Attach
17005 @subsubsection Running Tasks
17007 @cindex running VxWorks tasks
17008 You can also attach to an existing task using the @code{attach} command as
17012 (vxgdb) attach @var{task}
17016 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
17017 or suspended when you attach to it. Running tasks are suspended at
17018 the time of attachment.
17020 @node Embedded Processors
17021 @section Embedded Processors
17023 This section goes into details specific to particular embedded
17026 @cindex send command to simulator
17027 Whenever a specific embedded processor has a simulator, @value{GDBN}
17028 allows to send an arbitrary command to the simulator.
17031 @item sim @var{command}
17032 @kindex sim@r{, a command}
17033 Send an arbitrary @var{command} string to the simulator. Consult the
17034 documentation for the specific simulator in use for information about
17035 acceptable commands.
17041 * M32R/D:: Renesas M32R/D
17042 * M68K:: Motorola M68K
17043 * MicroBlaze:: Xilinx MicroBlaze
17044 * MIPS Embedded:: MIPS Embedded
17045 * OpenRISC 1000:: OpenRisc 1000
17046 * PA:: HP PA Embedded
17047 * PowerPC Embedded:: PowerPC Embedded
17048 * Sparclet:: Tsqware Sparclet
17049 * Sparclite:: Fujitsu Sparclite
17050 * Z8000:: Zilog Z8000
17053 * Super-H:: Renesas Super-H
17062 @item target rdi @var{dev}
17063 ARM Angel monitor, via RDI library interface to ADP protocol. You may
17064 use this target to communicate with both boards running the Angel
17065 monitor, or with the EmbeddedICE JTAG debug device.
17068 @item target rdp @var{dev}
17073 @value{GDBN} provides the following ARM-specific commands:
17076 @item set arm disassembler
17078 This commands selects from a list of disassembly styles. The
17079 @code{"std"} style is the standard style.
17081 @item show arm disassembler
17083 Show the current disassembly style.
17085 @item set arm apcs32
17086 @cindex ARM 32-bit mode
17087 This command toggles ARM operation mode between 32-bit and 26-bit.
17089 @item show arm apcs32
17090 Display the current usage of the ARM 32-bit mode.
17092 @item set arm fpu @var{fputype}
17093 This command sets the ARM floating-point unit (FPU) type. The
17094 argument @var{fputype} can be one of these:
17098 Determine the FPU type by querying the OS ABI.
17100 Software FPU, with mixed-endian doubles on little-endian ARM
17103 GCC-compiled FPA co-processor.
17105 Software FPU with pure-endian doubles.
17111 Show the current type of the FPU.
17114 This command forces @value{GDBN} to use the specified ABI.
17117 Show the currently used ABI.
17119 @item set arm fallback-mode (arm|thumb|auto)
17120 @value{GDBN} uses the symbol table, when available, to determine
17121 whether instructions are ARM or Thumb. This command controls
17122 @value{GDBN}'s default behavior when the symbol table is not
17123 available. The default is @samp{auto}, which causes @value{GDBN} to
17124 use the current execution mode (from the @code{T} bit in the @code{CPSR}
17127 @item show arm fallback-mode
17128 Show the current fallback instruction mode.
17130 @item set arm force-mode (arm|thumb|auto)
17131 This command overrides use of the symbol table to determine whether
17132 instructions are ARM or Thumb. The default is @samp{auto}, which
17133 causes @value{GDBN} to use the symbol table and then the setting
17134 of @samp{set arm fallback-mode}.
17136 @item show arm force-mode
17137 Show the current forced instruction mode.
17139 @item set debug arm
17140 Toggle whether to display ARM-specific debugging messages from the ARM
17141 target support subsystem.
17143 @item show debug arm
17144 Show whether ARM-specific debugging messages are enabled.
17147 The following commands are available when an ARM target is debugged
17148 using the RDI interface:
17151 @item rdilogfile @r{[}@var{file}@r{]}
17153 @cindex ADP (Angel Debugger Protocol) logging
17154 Set the filename for the ADP (Angel Debugger Protocol) packet log.
17155 With an argument, sets the log file to the specified @var{file}. With
17156 no argument, show the current log file name. The default log file is
17159 @item rdilogenable @r{[}@var{arg}@r{]}
17160 @kindex rdilogenable
17161 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
17162 enables logging, with an argument 0 or @code{"no"} disables it. With
17163 no arguments displays the current setting. When logging is enabled,
17164 ADP packets exchanged between @value{GDBN} and the RDI target device
17165 are logged to a file.
17167 @item set rdiromatzero
17168 @kindex set rdiromatzero
17169 @cindex ROM at zero address, RDI
17170 Tell @value{GDBN} whether the target has ROM at address 0. If on,
17171 vector catching is disabled, so that zero address can be used. If off
17172 (the default), vector catching is enabled. For this command to take
17173 effect, it needs to be invoked prior to the @code{target rdi} command.
17175 @item show rdiromatzero
17176 @kindex show rdiromatzero
17177 Show the current setting of ROM at zero address.
17179 @item set rdiheartbeat
17180 @kindex set rdiheartbeat
17181 @cindex RDI heartbeat
17182 Enable or disable RDI heartbeat packets. It is not recommended to
17183 turn on this option, since it confuses ARM and EPI JTAG interface, as
17184 well as the Angel monitor.
17186 @item show rdiheartbeat
17187 @kindex show rdiheartbeat
17188 Show the setting of RDI heartbeat packets.
17193 @subsection Renesas M32R/D and M32R/SDI
17196 @kindex target m32r
17197 @item target m32r @var{dev}
17198 Renesas M32R/D ROM monitor.
17200 @kindex target m32rsdi
17201 @item target m32rsdi @var{dev}
17202 Renesas M32R SDI server, connected via parallel port to the board.
17205 The following @value{GDBN} commands are specific to the M32R monitor:
17208 @item set download-path @var{path}
17209 @kindex set download-path
17210 @cindex find downloadable @sc{srec} files (M32R)
17211 Set the default path for finding downloadable @sc{srec} files.
17213 @item show download-path
17214 @kindex show download-path
17215 Show the default path for downloadable @sc{srec} files.
17217 @item set board-address @var{addr}
17218 @kindex set board-address
17219 @cindex M32-EVA target board address
17220 Set the IP address for the M32R-EVA target board.
17222 @item show board-address
17223 @kindex show board-address
17224 Show the current IP address of the target board.
17226 @item set server-address @var{addr}
17227 @kindex set server-address
17228 @cindex download server address (M32R)
17229 Set the IP address for the download server, which is the @value{GDBN}'s
17232 @item show server-address
17233 @kindex show server-address
17234 Display the IP address of the download server.
17236 @item upload @r{[}@var{file}@r{]}
17237 @kindex upload@r{, M32R}
17238 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
17239 upload capability. If no @var{file} argument is given, the current
17240 executable file is uploaded.
17242 @item tload @r{[}@var{file}@r{]}
17243 @kindex tload@r{, M32R}
17244 Test the @code{upload} command.
17247 The following commands are available for M32R/SDI:
17252 @cindex reset SDI connection, M32R
17253 This command resets the SDI connection.
17257 This command shows the SDI connection status.
17260 @kindex debug_chaos
17261 @cindex M32R/Chaos debugging
17262 Instructs the remote that M32R/Chaos debugging is to be used.
17264 @item use_debug_dma
17265 @kindex use_debug_dma
17266 Instructs the remote to use the DEBUG_DMA method of accessing memory.
17269 @kindex use_mon_code
17270 Instructs the remote to use the MON_CODE method of accessing memory.
17273 @kindex use_ib_break
17274 Instructs the remote to set breakpoints by IB break.
17276 @item use_dbt_break
17277 @kindex use_dbt_break
17278 Instructs the remote to set breakpoints by DBT.
17284 The Motorola m68k configuration includes ColdFire support, and a
17285 target command for the following ROM monitor.
17289 @kindex target dbug
17290 @item target dbug @var{dev}
17291 dBUG ROM monitor for Motorola ColdFire.
17296 @subsection MicroBlaze
17297 @cindex Xilinx MicroBlaze
17298 @cindex XMD, Xilinx Microprocessor Debugger
17300 The MicroBlaze is a soft-core processor supported on various Xilinx
17301 FPGAs, such as Spartan or Virtex series. Boards with these processors
17302 usually have JTAG ports which connect to a host system running the Xilinx
17303 Embedded Development Kit (EDK) or Software Development Kit (SDK).
17304 This host system is used to download the configuration bitstream to
17305 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
17306 communicates with the target board using the JTAG interface and
17307 presents a @code{gdbserver} interface to the board. By default
17308 @code{xmd} uses port @code{1234}. (While it is possible to change
17309 this default port, it requires the use of undocumented @code{xmd}
17310 commands. Contact Xilinx support if you need to do this.)
17312 Use these GDB commands to connect to the MicroBlaze target processor.
17315 @item target remote :1234
17316 Use this command to connect to the target if you are running @value{GDBN}
17317 on the same system as @code{xmd}.
17319 @item target remote @var{xmd-host}:1234
17320 Use this command to connect to the target if it is connected to @code{xmd}
17321 running on a different system named @var{xmd-host}.
17324 Use this command to download a program to the MicroBlaze target.
17326 @item set debug microblaze @var{n}
17327 Enable MicroBlaze-specific debugging messages if non-zero.
17329 @item show debug microblaze @var{n}
17330 Show MicroBlaze-specific debugging level.
17333 @node MIPS Embedded
17334 @subsection MIPS Embedded
17336 @cindex MIPS boards
17337 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
17338 MIPS board attached to a serial line. This is available when
17339 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
17342 Use these @value{GDBN} commands to specify the connection to your target board:
17345 @item target mips @var{port}
17346 @kindex target mips @var{port}
17347 To run a program on the board, start up @code{@value{GDBP}} with the
17348 name of your program as the argument. To connect to the board, use the
17349 command @samp{target mips @var{port}}, where @var{port} is the name of
17350 the serial port connected to the board. If the program has not already
17351 been downloaded to the board, you may use the @code{load} command to
17352 download it. You can then use all the usual @value{GDBN} commands.
17354 For example, this sequence connects to the target board through a serial
17355 port, and loads and runs a program called @var{prog} through the
17359 host$ @value{GDBP} @var{prog}
17360 @value{GDBN} is free software and @dots{}
17361 (@value{GDBP}) target mips /dev/ttyb
17362 (@value{GDBP}) load @var{prog}
17366 @item target mips @var{hostname}:@var{portnumber}
17367 On some @value{GDBN} host configurations, you can specify a TCP
17368 connection (for instance, to a serial line managed by a terminal
17369 concentrator) instead of a serial port, using the syntax
17370 @samp{@var{hostname}:@var{portnumber}}.
17372 @item target pmon @var{port}
17373 @kindex target pmon @var{port}
17376 @item target ddb @var{port}
17377 @kindex target ddb @var{port}
17378 NEC's DDB variant of PMON for Vr4300.
17380 @item target lsi @var{port}
17381 @kindex target lsi @var{port}
17382 LSI variant of PMON.
17384 @kindex target r3900
17385 @item target r3900 @var{dev}
17386 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
17388 @kindex target array
17389 @item target array @var{dev}
17390 Array Tech LSI33K RAID controller board.
17396 @value{GDBN} also supports these special commands for MIPS targets:
17399 @item set mipsfpu double
17400 @itemx set mipsfpu single
17401 @itemx set mipsfpu none
17402 @itemx set mipsfpu auto
17403 @itemx show mipsfpu
17404 @kindex set mipsfpu
17405 @kindex show mipsfpu
17406 @cindex MIPS remote floating point
17407 @cindex floating point, MIPS remote
17408 If your target board does not support the MIPS floating point
17409 coprocessor, you should use the command @samp{set mipsfpu none} (if you
17410 need this, you may wish to put the command in your @value{GDBN} init
17411 file). This tells @value{GDBN} how to find the return value of
17412 functions which return floating point values. It also allows
17413 @value{GDBN} to avoid saving the floating point registers when calling
17414 functions on the board. If you are using a floating point coprocessor
17415 with only single precision floating point support, as on the @sc{r4650}
17416 processor, use the command @samp{set mipsfpu single}. The default
17417 double precision floating point coprocessor may be selected using
17418 @samp{set mipsfpu double}.
17420 In previous versions the only choices were double precision or no
17421 floating point, so @samp{set mipsfpu on} will select double precision
17422 and @samp{set mipsfpu off} will select no floating point.
17424 As usual, you can inquire about the @code{mipsfpu} variable with
17425 @samp{show mipsfpu}.
17427 @item set timeout @var{seconds}
17428 @itemx set retransmit-timeout @var{seconds}
17429 @itemx show timeout
17430 @itemx show retransmit-timeout
17431 @cindex @code{timeout}, MIPS protocol
17432 @cindex @code{retransmit-timeout}, MIPS protocol
17433 @kindex set timeout
17434 @kindex show timeout
17435 @kindex set retransmit-timeout
17436 @kindex show retransmit-timeout
17437 You can control the timeout used while waiting for a packet, in the MIPS
17438 remote protocol, with the @code{set timeout @var{seconds}} command. The
17439 default is 5 seconds. Similarly, you can control the timeout used while
17440 waiting for an acknowledgment of a packet with the @code{set
17441 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
17442 You can inspect both values with @code{show timeout} and @code{show
17443 retransmit-timeout}. (These commands are @emph{only} available when
17444 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
17446 The timeout set by @code{set timeout} does not apply when @value{GDBN}
17447 is waiting for your program to stop. In that case, @value{GDBN} waits
17448 forever because it has no way of knowing how long the program is going
17449 to run before stopping.
17451 @item set syn-garbage-limit @var{num}
17452 @kindex set syn-garbage-limit@r{, MIPS remote}
17453 @cindex synchronize with remote MIPS target
17454 Limit the maximum number of characters @value{GDBN} should ignore when
17455 it tries to synchronize with the remote target. The default is 10
17456 characters. Setting the limit to -1 means there's no limit.
17458 @item show syn-garbage-limit
17459 @kindex show syn-garbage-limit@r{, MIPS remote}
17460 Show the current limit on the number of characters to ignore when
17461 trying to synchronize with the remote system.
17463 @item set monitor-prompt @var{prompt}
17464 @kindex set monitor-prompt@r{, MIPS remote}
17465 @cindex remote monitor prompt
17466 Tell @value{GDBN} to expect the specified @var{prompt} string from the
17467 remote monitor. The default depends on the target:
17477 @item show monitor-prompt
17478 @kindex show monitor-prompt@r{, MIPS remote}
17479 Show the current strings @value{GDBN} expects as the prompt from the
17482 @item set monitor-warnings
17483 @kindex set monitor-warnings@r{, MIPS remote}
17484 Enable or disable monitor warnings about hardware breakpoints. This
17485 has effect only for the @code{lsi} target. When on, @value{GDBN} will
17486 display warning messages whose codes are returned by the @code{lsi}
17487 PMON monitor for breakpoint commands.
17489 @item show monitor-warnings
17490 @kindex show monitor-warnings@r{, MIPS remote}
17491 Show the current setting of printing monitor warnings.
17493 @item pmon @var{command}
17494 @kindex pmon@r{, MIPS remote}
17495 @cindex send PMON command
17496 This command allows sending an arbitrary @var{command} string to the
17497 monitor. The monitor must be in debug mode for this to work.
17500 @node OpenRISC 1000
17501 @subsection OpenRISC 1000
17502 @cindex OpenRISC 1000
17504 @cindex or1k boards
17505 See OR1k Architecture document (@uref{www.opencores.org}) for more information
17506 about platform and commands.
17510 @kindex target jtag
17511 @item target jtag jtag://@var{host}:@var{port}
17513 Connects to remote JTAG server.
17514 JTAG remote server can be either an or1ksim or JTAG server,
17515 connected via parallel port to the board.
17517 Example: @code{target jtag jtag://localhost:9999}
17520 @item or1ksim @var{command}
17521 If connected to @code{or1ksim} OpenRISC 1000 Architectural
17522 Simulator, proprietary commands can be executed.
17524 @kindex info or1k spr
17525 @item info or1k spr
17526 Displays spr groups.
17528 @item info or1k spr @var{group}
17529 @itemx info or1k spr @var{groupno}
17530 Displays register names in selected group.
17532 @item info or1k spr @var{group} @var{register}
17533 @itemx info or1k spr @var{register}
17534 @itemx info or1k spr @var{groupno} @var{registerno}
17535 @itemx info or1k spr @var{registerno}
17536 Shows information about specified spr register.
17539 @item spr @var{group} @var{register} @var{value}
17540 @itemx spr @var{register @var{value}}
17541 @itemx spr @var{groupno} @var{registerno @var{value}}
17542 @itemx spr @var{registerno @var{value}}
17543 Writes @var{value} to specified spr register.
17546 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
17547 It is very similar to @value{GDBN} trace, except it does not interfere with normal
17548 program execution and is thus much faster. Hardware breakpoints/watchpoint
17549 triggers can be set using:
17552 Load effective address/data
17554 Store effective address/data
17556 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
17561 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
17562 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
17564 @code{htrace} commands:
17565 @cindex OpenRISC 1000 htrace
17568 @item hwatch @var{conditional}
17569 Set hardware watchpoint on combination of Load/Store Effective Address(es)
17570 or Data. For example:
17572 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
17574 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
17578 Display information about current HW trace configuration.
17580 @item htrace trigger @var{conditional}
17581 Set starting criteria for HW trace.
17583 @item htrace qualifier @var{conditional}
17584 Set acquisition qualifier for HW trace.
17586 @item htrace stop @var{conditional}
17587 Set HW trace stopping criteria.
17589 @item htrace record [@var{data}]*
17590 Selects the data to be recorded, when qualifier is met and HW trace was
17593 @item htrace enable
17594 @itemx htrace disable
17595 Enables/disables the HW trace.
17597 @item htrace rewind [@var{filename}]
17598 Clears currently recorded trace data.
17600 If filename is specified, new trace file is made and any newly collected data
17601 will be written there.
17603 @item htrace print [@var{start} [@var{len}]]
17604 Prints trace buffer, using current record configuration.
17606 @item htrace mode continuous
17607 Set continuous trace mode.
17609 @item htrace mode suspend
17610 Set suspend trace mode.
17614 @node PowerPC Embedded
17615 @subsection PowerPC Embedded
17617 @value{GDBN} provides the following PowerPC-specific commands:
17620 @kindex set powerpc
17621 @item set powerpc soft-float
17622 @itemx show powerpc soft-float
17623 Force @value{GDBN} to use (or not use) a software floating point calling
17624 convention. By default, @value{GDBN} selects the calling convention based
17625 on the selected architecture and the provided executable file.
17627 @item set powerpc vector-abi
17628 @itemx show powerpc vector-abi
17629 Force @value{GDBN} to use the specified calling convention for vector
17630 arguments and return values. The valid options are @samp{auto};
17631 @samp{generic}, to avoid vector registers even if they are present;
17632 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
17633 registers. By default, @value{GDBN} selects the calling convention
17634 based on the selected architecture and the provided executable file.
17636 @kindex target dink32
17637 @item target dink32 @var{dev}
17638 DINK32 ROM monitor.
17640 @kindex target ppcbug
17641 @item target ppcbug @var{dev}
17642 @kindex target ppcbug1
17643 @item target ppcbug1 @var{dev}
17644 PPCBUG ROM monitor for PowerPC.
17647 @item target sds @var{dev}
17648 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
17651 @cindex SDS protocol
17652 The following commands specific to the SDS protocol are supported
17656 @item set sdstimeout @var{nsec}
17657 @kindex set sdstimeout
17658 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
17659 default is 2 seconds.
17661 @item show sdstimeout
17662 @kindex show sdstimeout
17663 Show the current value of the SDS timeout.
17665 @item sds @var{command}
17666 @kindex sds@r{, a command}
17667 Send the specified @var{command} string to the SDS monitor.
17672 @subsection HP PA Embedded
17676 @kindex target op50n
17677 @item target op50n @var{dev}
17678 OP50N monitor, running on an OKI HPPA board.
17680 @kindex target w89k
17681 @item target w89k @var{dev}
17682 W89K monitor, running on a Winbond HPPA board.
17687 @subsection Tsqware Sparclet
17691 @value{GDBN} enables developers to debug tasks running on
17692 Sparclet targets from a Unix host.
17693 @value{GDBN} uses code that runs on
17694 both the Unix host and on the Sparclet target. The program
17695 @code{@value{GDBP}} is installed and executed on the Unix host.
17698 @item remotetimeout @var{args}
17699 @kindex remotetimeout
17700 @value{GDBN} supports the option @code{remotetimeout}.
17701 This option is set by the user, and @var{args} represents the number of
17702 seconds @value{GDBN} waits for responses.
17705 @cindex compiling, on Sparclet
17706 When compiling for debugging, include the options @samp{-g} to get debug
17707 information and @samp{-Ttext} to relocate the program to where you wish to
17708 load it on the target. You may also want to add the options @samp{-n} or
17709 @samp{-N} in order to reduce the size of the sections. Example:
17712 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
17715 You can use @code{objdump} to verify that the addresses are what you intended:
17718 sparclet-aout-objdump --headers --syms prog
17721 @cindex running, on Sparclet
17723 your Unix execution search path to find @value{GDBN}, you are ready to
17724 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
17725 (or @code{sparclet-aout-gdb}, depending on your installation).
17727 @value{GDBN} comes up showing the prompt:
17734 * Sparclet File:: Setting the file to debug
17735 * Sparclet Connection:: Connecting to Sparclet
17736 * Sparclet Download:: Sparclet download
17737 * Sparclet Execution:: Running and debugging
17740 @node Sparclet File
17741 @subsubsection Setting File to Debug
17743 The @value{GDBN} command @code{file} lets you choose with program to debug.
17746 (gdbslet) file prog
17750 @value{GDBN} then attempts to read the symbol table of @file{prog}.
17751 @value{GDBN} locates
17752 the file by searching the directories listed in the command search
17754 If the file was compiled with debug information (option @samp{-g}), source
17755 files will be searched as well.
17756 @value{GDBN} locates
17757 the source files by searching the directories listed in the directory search
17758 path (@pxref{Environment, ,Your Program's Environment}).
17760 to find a file, it displays a message such as:
17763 prog: No such file or directory.
17766 When this happens, add the appropriate directories to the search paths with
17767 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
17768 @code{target} command again.
17770 @node Sparclet Connection
17771 @subsubsection Connecting to Sparclet
17773 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
17774 To connect to a target on serial port ``@code{ttya}'', type:
17777 (gdbslet) target sparclet /dev/ttya
17778 Remote target sparclet connected to /dev/ttya
17779 main () at ../prog.c:3
17783 @value{GDBN} displays messages like these:
17789 @node Sparclet Download
17790 @subsubsection Sparclet Download
17792 @cindex download to Sparclet
17793 Once connected to the Sparclet target,
17794 you can use the @value{GDBN}
17795 @code{load} command to download the file from the host to the target.
17796 The file name and load offset should be given as arguments to the @code{load}
17798 Since the file format is aout, the program must be loaded to the starting
17799 address. You can use @code{objdump} to find out what this value is. The load
17800 offset is an offset which is added to the VMA (virtual memory address)
17801 of each of the file's sections.
17802 For instance, if the program
17803 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
17804 and bss at 0x12010170, in @value{GDBN}, type:
17807 (gdbslet) load prog 0x12010000
17808 Loading section .text, size 0xdb0 vma 0x12010000
17811 If the code is loaded at a different address then what the program was linked
17812 to, you may need to use the @code{section} and @code{add-symbol-file} commands
17813 to tell @value{GDBN} where to map the symbol table.
17815 @node Sparclet Execution
17816 @subsubsection Running and Debugging
17818 @cindex running and debugging Sparclet programs
17819 You can now begin debugging the task using @value{GDBN}'s execution control
17820 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
17821 manual for the list of commands.
17825 Breakpoint 1 at 0x12010000: file prog.c, line 3.
17827 Starting program: prog
17828 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
17829 3 char *symarg = 0;
17831 4 char *execarg = "hello!";
17836 @subsection Fujitsu Sparclite
17840 @kindex target sparclite
17841 @item target sparclite @var{dev}
17842 Fujitsu sparclite boards, used only for the purpose of loading.
17843 You must use an additional command to debug the program.
17844 For example: target remote @var{dev} using @value{GDBN} standard
17850 @subsection Zilog Z8000
17853 @cindex simulator, Z8000
17854 @cindex Zilog Z8000 simulator
17856 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
17859 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
17860 unsegmented variant of the Z8000 architecture) or the Z8001 (the
17861 segmented variant). The simulator recognizes which architecture is
17862 appropriate by inspecting the object code.
17865 @item target sim @var{args}
17867 @kindex target sim@r{, with Z8000}
17868 Debug programs on a simulated CPU. If the simulator supports setup
17869 options, specify them via @var{args}.
17873 After specifying this target, you can debug programs for the simulated
17874 CPU in the same style as programs for your host computer; use the
17875 @code{file} command to load a new program image, the @code{run} command
17876 to run your program, and so on.
17878 As well as making available all the usual machine registers
17879 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
17880 additional items of information as specially named registers:
17885 Counts clock-ticks in the simulator.
17888 Counts instructions run in the simulator.
17891 Execution time in 60ths of a second.
17895 You can refer to these values in @value{GDBN} expressions with the usual
17896 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
17897 conditional breakpoint that suspends only after at least 5000
17898 simulated clock ticks.
17901 @subsection Atmel AVR
17904 When configured for debugging the Atmel AVR, @value{GDBN} supports the
17905 following AVR-specific commands:
17908 @item info io_registers
17909 @kindex info io_registers@r{, AVR}
17910 @cindex I/O registers (Atmel AVR)
17911 This command displays information about the AVR I/O registers. For
17912 each register, @value{GDBN} prints its number and value.
17919 When configured for debugging CRIS, @value{GDBN} provides the
17920 following CRIS-specific commands:
17923 @item set cris-version @var{ver}
17924 @cindex CRIS version
17925 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
17926 The CRIS version affects register names and sizes. This command is useful in
17927 case autodetection of the CRIS version fails.
17929 @item show cris-version
17930 Show the current CRIS version.
17932 @item set cris-dwarf2-cfi
17933 @cindex DWARF-2 CFI and CRIS
17934 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
17935 Change to @samp{off} when using @code{gcc-cris} whose version is below
17938 @item show cris-dwarf2-cfi
17939 Show the current state of using DWARF-2 CFI.
17941 @item set cris-mode @var{mode}
17943 Set the current CRIS mode to @var{mode}. It should only be changed when
17944 debugging in guru mode, in which case it should be set to
17945 @samp{guru} (the default is @samp{normal}).
17947 @item show cris-mode
17948 Show the current CRIS mode.
17952 @subsection Renesas Super-H
17955 For the Renesas Super-H processor, @value{GDBN} provides these
17960 @kindex regs@r{, Super-H}
17961 Show the values of all Super-H registers.
17963 @item set sh calling-convention @var{convention}
17964 @kindex set sh calling-convention
17965 Set the calling-convention used when calling functions from @value{GDBN}.
17966 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
17967 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
17968 convention. If the DWARF-2 information of the called function specifies
17969 that the function follows the Renesas calling convention, the function
17970 is called using the Renesas calling convention. If the calling convention
17971 is set to @samp{renesas}, the Renesas calling convention is always used,
17972 regardless of the DWARF-2 information. This can be used to override the
17973 default of @samp{gcc} if debug information is missing, or the compiler
17974 does not emit the DWARF-2 calling convention entry for a function.
17976 @item show sh calling-convention
17977 @kindex show sh calling-convention
17978 Show the current calling convention setting.
17983 @node Architectures
17984 @section Architectures
17986 This section describes characteristics of architectures that affect
17987 all uses of @value{GDBN} with the architecture, both native and cross.
17994 * HPPA:: HP PA architecture
17995 * SPU:: Cell Broadband Engine SPU architecture
18000 @subsection x86 Architecture-specific Issues
18003 @item set struct-convention @var{mode}
18004 @kindex set struct-convention
18005 @cindex struct return convention
18006 @cindex struct/union returned in registers
18007 Set the convention used by the inferior to return @code{struct}s and
18008 @code{union}s from functions to @var{mode}. Possible values of
18009 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
18010 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
18011 are returned on the stack, while @code{"reg"} means that a
18012 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
18013 be returned in a register.
18015 @item show struct-convention
18016 @kindex show struct-convention
18017 Show the current setting of the convention to return @code{struct}s
18026 @kindex set rstack_high_address
18027 @cindex AMD 29K register stack
18028 @cindex register stack, AMD29K
18029 @item set rstack_high_address @var{address}
18030 On AMD 29000 family processors, registers are saved in a separate
18031 @dfn{register stack}. There is no way for @value{GDBN} to determine the
18032 extent of this stack. Normally, @value{GDBN} just assumes that the
18033 stack is ``large enough''. This may result in @value{GDBN} referencing
18034 memory locations that do not exist. If necessary, you can get around
18035 this problem by specifying the ending address of the register stack with
18036 the @code{set rstack_high_address} command. The argument should be an
18037 address, which you probably want to precede with @samp{0x} to specify in
18040 @kindex show rstack_high_address
18041 @item show rstack_high_address
18042 Display the current limit of the register stack, on AMD 29000 family
18050 See the following section.
18055 @cindex stack on Alpha
18056 @cindex stack on MIPS
18057 @cindex Alpha stack
18059 Alpha- and MIPS-based computers use an unusual stack frame, which
18060 sometimes requires @value{GDBN} to search backward in the object code to
18061 find the beginning of a function.
18063 @cindex response time, MIPS debugging
18064 To improve response time (especially for embedded applications, where
18065 @value{GDBN} may be restricted to a slow serial line for this search)
18066 you may want to limit the size of this search, using one of these
18070 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
18071 @item set heuristic-fence-post @var{limit}
18072 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
18073 search for the beginning of a function. A value of @var{0} (the
18074 default) means there is no limit. However, except for @var{0}, the
18075 larger the limit the more bytes @code{heuristic-fence-post} must search
18076 and therefore the longer it takes to run. You should only need to use
18077 this command when debugging a stripped executable.
18079 @item show heuristic-fence-post
18080 Display the current limit.
18084 These commands are available @emph{only} when @value{GDBN} is configured
18085 for debugging programs on Alpha or MIPS processors.
18087 Several MIPS-specific commands are available when debugging MIPS
18091 @item set mips abi @var{arg}
18092 @kindex set mips abi
18093 @cindex set ABI for MIPS
18094 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
18095 values of @var{arg} are:
18099 The default ABI associated with the current binary (this is the
18110 @item show mips abi
18111 @kindex show mips abi
18112 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
18115 @itemx show mipsfpu
18116 @xref{MIPS Embedded, set mipsfpu}.
18118 @item set mips mask-address @var{arg}
18119 @kindex set mips mask-address
18120 @cindex MIPS addresses, masking
18121 This command determines whether the most-significant 32 bits of 64-bit
18122 MIPS addresses are masked off. The argument @var{arg} can be
18123 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
18124 setting, which lets @value{GDBN} determine the correct value.
18126 @item show mips mask-address
18127 @kindex show mips mask-address
18128 Show whether the upper 32 bits of MIPS addresses are masked off or
18131 @item set remote-mips64-transfers-32bit-regs
18132 @kindex set remote-mips64-transfers-32bit-regs
18133 This command controls compatibility with 64-bit MIPS targets that
18134 transfer data in 32-bit quantities. If you have an old MIPS 64 target
18135 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
18136 and 64 bits for other registers, set this option to @samp{on}.
18138 @item show remote-mips64-transfers-32bit-regs
18139 @kindex show remote-mips64-transfers-32bit-regs
18140 Show the current setting of compatibility with older MIPS 64 targets.
18142 @item set debug mips
18143 @kindex set debug mips
18144 This command turns on and off debugging messages for the MIPS-specific
18145 target code in @value{GDBN}.
18147 @item show debug mips
18148 @kindex show debug mips
18149 Show the current setting of MIPS debugging messages.
18155 @cindex HPPA support
18157 When @value{GDBN} is debugging the HP PA architecture, it provides the
18158 following special commands:
18161 @item set debug hppa
18162 @kindex set debug hppa
18163 This command determines whether HPPA architecture-specific debugging
18164 messages are to be displayed.
18166 @item show debug hppa
18167 Show whether HPPA debugging messages are displayed.
18169 @item maint print unwind @var{address}
18170 @kindex maint print unwind@r{, HPPA}
18171 This command displays the contents of the unwind table entry at the
18172 given @var{address}.
18178 @subsection Cell Broadband Engine SPU architecture
18179 @cindex Cell Broadband Engine
18182 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
18183 it provides the following special commands:
18186 @item info spu event
18188 Display SPU event facility status. Shows current event mask
18189 and pending event status.
18191 @item info spu signal
18192 Display SPU signal notification facility status. Shows pending
18193 signal-control word and signal notification mode of both signal
18194 notification channels.
18196 @item info spu mailbox
18197 Display SPU mailbox facility status. Shows all pending entries,
18198 in order of processing, in each of the SPU Write Outbound,
18199 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
18202 Display MFC DMA status. Shows all pending commands in the MFC
18203 DMA queue. For each entry, opcode, tag, class IDs, effective
18204 and local store addresses and transfer size are shown.
18206 @item info spu proxydma
18207 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
18208 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
18209 and local store addresses and transfer size are shown.
18213 When @value{GDBN} is debugging a combined PowerPC/SPU application
18214 on the Cell Broadband Engine, it provides in addition the following
18218 @item set spu stop-on-load @var{arg}
18220 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
18221 will give control to the user when a new SPE thread enters its @code{main}
18222 function. The default is @code{off}.
18224 @item show spu stop-on-load
18226 Show whether to stop for new SPE threads.
18228 @item set spu auto-flush-cache @var{arg}
18229 Set whether to automatically flush the software-managed cache. When set to
18230 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
18231 cache to be flushed whenever SPE execution stops. This provides a consistent
18232 view of PowerPC memory that is accessed via the cache. If an application
18233 does not use the software-managed cache, this option has no effect.
18235 @item show spu auto-flush-cache
18236 Show whether to automatically flush the software-managed cache.
18241 @subsection PowerPC
18242 @cindex PowerPC architecture
18244 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
18245 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
18246 numbers stored in the floating point registers. These values must be stored
18247 in two consecutive registers, always starting at an even register like
18248 @code{f0} or @code{f2}.
18250 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
18251 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
18252 @code{f2} and @code{f3} for @code{$dl1} and so on.
18254 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
18255 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
18258 @node Controlling GDB
18259 @chapter Controlling @value{GDBN}
18261 You can alter the way @value{GDBN} interacts with you by using the
18262 @code{set} command. For commands controlling how @value{GDBN} displays
18263 data, see @ref{Print Settings, ,Print Settings}. Other settings are
18268 * Editing:: Command editing
18269 * Command History:: Command history
18270 * Screen Size:: Screen size
18271 * Numbers:: Numbers
18272 * ABI:: Configuring the current ABI
18273 * Messages/Warnings:: Optional warnings and messages
18274 * Debugging Output:: Optional messages about internal happenings
18275 * Other Misc Settings:: Other Miscellaneous Settings
18283 @value{GDBN} indicates its readiness to read a command by printing a string
18284 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
18285 can change the prompt string with the @code{set prompt} command. For
18286 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
18287 the prompt in one of the @value{GDBN} sessions so that you can always tell
18288 which one you are talking to.
18290 @emph{Note:} @code{set prompt} does not add a space for you after the
18291 prompt you set. This allows you to set a prompt which ends in a space
18292 or a prompt that does not.
18296 @item set prompt @var{newprompt}
18297 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
18299 @kindex show prompt
18301 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
18305 @section Command Editing
18307 @cindex command line editing
18309 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
18310 @sc{gnu} library provides consistent behavior for programs which provide a
18311 command line interface to the user. Advantages are @sc{gnu} Emacs-style
18312 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
18313 substitution, and a storage and recall of command history across
18314 debugging sessions.
18316 You may control the behavior of command line editing in @value{GDBN} with the
18317 command @code{set}.
18320 @kindex set editing
18323 @itemx set editing on
18324 Enable command line editing (enabled by default).
18326 @item set editing off
18327 Disable command line editing.
18329 @kindex show editing
18331 Show whether command line editing is enabled.
18334 @xref{Command Line Editing}, for more details about the Readline
18335 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
18336 encouraged to read that chapter.
18338 @node Command History
18339 @section Command History
18340 @cindex command history
18342 @value{GDBN} can keep track of the commands you type during your
18343 debugging sessions, so that you can be certain of precisely what
18344 happened. Use these commands to manage the @value{GDBN} command
18347 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
18348 package, to provide the history facility. @xref{Using History
18349 Interactively}, for the detailed description of the History library.
18351 To issue a command to @value{GDBN} without affecting certain aspects of
18352 the state which is seen by users, prefix it with @samp{server }
18353 (@pxref{Server Prefix}). This
18354 means that this command will not affect the command history, nor will it
18355 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
18356 pressed on a line by itself.
18358 @cindex @code{server}, command prefix
18359 The server prefix does not affect the recording of values into the value
18360 history; to print a value without recording it into the value history,
18361 use the @code{output} command instead of the @code{print} command.
18363 Here is the description of @value{GDBN} commands related to command
18367 @cindex history substitution
18368 @cindex history file
18369 @kindex set history filename
18370 @cindex @env{GDBHISTFILE}, environment variable
18371 @item set history filename @var{fname}
18372 Set the name of the @value{GDBN} command history file to @var{fname}.
18373 This is the file where @value{GDBN} reads an initial command history
18374 list, and where it writes the command history from this session when it
18375 exits. You can access this list through history expansion or through
18376 the history command editing characters listed below. This file defaults
18377 to the value of the environment variable @code{GDBHISTFILE}, or to
18378 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
18381 @cindex save command history
18382 @kindex set history save
18383 @item set history save
18384 @itemx set history save on
18385 Record command history in a file, whose name may be specified with the
18386 @code{set history filename} command. By default, this option is disabled.
18388 @item set history save off
18389 Stop recording command history in a file.
18391 @cindex history size
18392 @kindex set history size
18393 @cindex @env{HISTSIZE}, environment variable
18394 @item set history size @var{size}
18395 Set the number of commands which @value{GDBN} keeps in its history list.
18396 This defaults to the value of the environment variable
18397 @code{HISTSIZE}, or to 256 if this variable is not set.
18400 History expansion assigns special meaning to the character @kbd{!}.
18401 @xref{Event Designators}, for more details.
18403 @cindex history expansion, turn on/off
18404 Since @kbd{!} is also the logical not operator in C, history expansion
18405 is off by default. If you decide to enable history expansion with the
18406 @code{set history expansion on} command, you may sometimes need to
18407 follow @kbd{!} (when it is used as logical not, in an expression) with
18408 a space or a tab to prevent it from being expanded. The readline
18409 history facilities do not attempt substitution on the strings
18410 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
18412 The commands to control history expansion are:
18415 @item set history expansion on
18416 @itemx set history expansion
18417 @kindex set history expansion
18418 Enable history expansion. History expansion is off by default.
18420 @item set history expansion off
18421 Disable history expansion.
18424 @kindex show history
18426 @itemx show history filename
18427 @itemx show history save
18428 @itemx show history size
18429 @itemx show history expansion
18430 These commands display the state of the @value{GDBN} history parameters.
18431 @code{show history} by itself displays all four states.
18436 @kindex show commands
18437 @cindex show last commands
18438 @cindex display command history
18439 @item show commands
18440 Display the last ten commands in the command history.
18442 @item show commands @var{n}
18443 Print ten commands centered on command number @var{n}.
18445 @item show commands +
18446 Print ten commands just after the commands last printed.
18450 @section Screen Size
18451 @cindex size of screen
18452 @cindex pauses in output
18454 Certain commands to @value{GDBN} may produce large amounts of
18455 information output to the screen. To help you read all of it,
18456 @value{GDBN} pauses and asks you for input at the end of each page of
18457 output. Type @key{RET} when you want to continue the output, or @kbd{q}
18458 to discard the remaining output. Also, the screen width setting
18459 determines when to wrap lines of output. Depending on what is being
18460 printed, @value{GDBN} tries to break the line at a readable place,
18461 rather than simply letting it overflow onto the following line.
18463 Normally @value{GDBN} knows the size of the screen from the terminal
18464 driver software. For example, on Unix @value{GDBN} uses the termcap data base
18465 together with the value of the @code{TERM} environment variable and the
18466 @code{stty rows} and @code{stty cols} settings. If this is not correct,
18467 you can override it with the @code{set height} and @code{set
18474 @kindex show height
18475 @item set height @var{lpp}
18477 @itemx set width @var{cpl}
18479 These @code{set} commands specify a screen height of @var{lpp} lines and
18480 a screen width of @var{cpl} characters. The associated @code{show}
18481 commands display the current settings.
18483 If you specify a height of zero lines, @value{GDBN} does not pause during
18484 output no matter how long the output is. This is useful if output is to a
18485 file or to an editor buffer.
18487 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
18488 from wrapping its output.
18490 @item set pagination on
18491 @itemx set pagination off
18492 @kindex set pagination
18493 Turn the output pagination on or off; the default is on. Turning
18494 pagination off is the alternative to @code{set height 0}.
18496 @item show pagination
18497 @kindex show pagination
18498 Show the current pagination mode.
18503 @cindex number representation
18504 @cindex entering numbers
18506 You can always enter numbers in octal, decimal, or hexadecimal in
18507 @value{GDBN} by the usual conventions: octal numbers begin with
18508 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
18509 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
18510 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
18511 10; likewise, the default display for numbers---when no particular
18512 format is specified---is base 10. You can change the default base for
18513 both input and output with the commands described below.
18516 @kindex set input-radix
18517 @item set input-radix @var{base}
18518 Set the default base for numeric input. Supported choices
18519 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
18520 specified either unambiguously or using the current input radix; for
18524 set input-radix 012
18525 set input-radix 10.
18526 set input-radix 0xa
18530 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
18531 leaves the input radix unchanged, no matter what it was, since
18532 @samp{10}, being without any leading or trailing signs of its base, is
18533 interpreted in the current radix. Thus, if the current radix is 16,
18534 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
18537 @kindex set output-radix
18538 @item set output-radix @var{base}
18539 Set the default base for numeric display. Supported choices
18540 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
18541 specified either unambiguously or using the current input radix.
18543 @kindex show input-radix
18544 @item show input-radix
18545 Display the current default base for numeric input.
18547 @kindex show output-radix
18548 @item show output-radix
18549 Display the current default base for numeric display.
18551 @item set radix @r{[}@var{base}@r{]}
18555 These commands set and show the default base for both input and output
18556 of numbers. @code{set radix} sets the radix of input and output to
18557 the same base; without an argument, it resets the radix back to its
18558 default value of 10.
18563 @section Configuring the Current ABI
18565 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
18566 application automatically. However, sometimes you need to override its
18567 conclusions. Use these commands to manage @value{GDBN}'s view of the
18574 One @value{GDBN} configuration can debug binaries for multiple operating
18575 system targets, either via remote debugging or native emulation.
18576 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
18577 but you can override its conclusion using the @code{set osabi} command.
18578 One example where this is useful is in debugging of binaries which use
18579 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
18580 not have the same identifying marks that the standard C library for your
18585 Show the OS ABI currently in use.
18588 With no argument, show the list of registered available OS ABI's.
18590 @item set osabi @var{abi}
18591 Set the current OS ABI to @var{abi}.
18594 @cindex float promotion
18596 Generally, the way that an argument of type @code{float} is passed to a
18597 function depends on whether the function is prototyped. For a prototyped
18598 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
18599 according to the architecture's convention for @code{float}. For unprototyped
18600 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
18601 @code{double} and then passed.
18603 Unfortunately, some forms of debug information do not reliably indicate whether
18604 a function is prototyped. If @value{GDBN} calls a function that is not marked
18605 as prototyped, it consults @kbd{set coerce-float-to-double}.
18608 @kindex set coerce-float-to-double
18609 @item set coerce-float-to-double
18610 @itemx set coerce-float-to-double on
18611 Arguments of type @code{float} will be promoted to @code{double} when passed
18612 to an unprototyped function. This is the default setting.
18614 @item set coerce-float-to-double off
18615 Arguments of type @code{float} will be passed directly to unprototyped
18618 @kindex show coerce-float-to-double
18619 @item show coerce-float-to-double
18620 Show the current setting of promoting @code{float} to @code{double}.
18624 @kindex show cp-abi
18625 @value{GDBN} needs to know the ABI used for your program's C@t{++}
18626 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
18627 used to build your application. @value{GDBN} only fully supports
18628 programs with a single C@t{++} ABI; if your program contains code using
18629 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
18630 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
18631 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
18632 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
18633 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
18634 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
18639 Show the C@t{++} ABI currently in use.
18642 With no argument, show the list of supported C@t{++} ABI's.
18644 @item set cp-abi @var{abi}
18645 @itemx set cp-abi auto
18646 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
18649 @node Messages/Warnings
18650 @section Optional Warnings and Messages
18652 @cindex verbose operation
18653 @cindex optional warnings
18654 By default, @value{GDBN} is silent about its inner workings. If you are
18655 running on a slow machine, you may want to use the @code{set verbose}
18656 command. This makes @value{GDBN} tell you when it does a lengthy
18657 internal operation, so you will not think it has crashed.
18659 Currently, the messages controlled by @code{set verbose} are those
18660 which announce that the symbol table for a source file is being read;
18661 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
18664 @kindex set verbose
18665 @item set verbose on
18666 Enables @value{GDBN} output of certain informational messages.
18668 @item set verbose off
18669 Disables @value{GDBN} output of certain informational messages.
18671 @kindex show verbose
18673 Displays whether @code{set verbose} is on or off.
18676 By default, if @value{GDBN} encounters bugs in the symbol table of an
18677 object file, it is silent; but if you are debugging a compiler, you may
18678 find this information useful (@pxref{Symbol Errors, ,Errors Reading
18683 @kindex set complaints
18684 @item set complaints @var{limit}
18685 Permits @value{GDBN} to output @var{limit} complaints about each type of
18686 unusual symbols before becoming silent about the problem. Set
18687 @var{limit} to zero to suppress all complaints; set it to a large number
18688 to prevent complaints from being suppressed.
18690 @kindex show complaints
18691 @item show complaints
18692 Displays how many symbol complaints @value{GDBN} is permitted to produce.
18696 @anchor{confirmation requests}
18697 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
18698 lot of stupid questions to confirm certain commands. For example, if
18699 you try to run a program which is already running:
18703 The program being debugged has been started already.
18704 Start it from the beginning? (y or n)
18707 If you are willing to unflinchingly face the consequences of your own
18708 commands, you can disable this ``feature'':
18712 @kindex set confirm
18714 @cindex confirmation
18715 @cindex stupid questions
18716 @item set confirm off
18717 Disables confirmation requests.
18719 @item set confirm on
18720 Enables confirmation requests (the default).
18722 @kindex show confirm
18724 Displays state of confirmation requests.
18728 @cindex command tracing
18729 If you need to debug user-defined commands or sourced files you may find it
18730 useful to enable @dfn{command tracing}. In this mode each command will be
18731 printed as it is executed, prefixed with one or more @samp{+} symbols, the
18732 quantity denoting the call depth of each command.
18735 @kindex set trace-commands
18736 @cindex command scripts, debugging
18737 @item set trace-commands on
18738 Enable command tracing.
18739 @item set trace-commands off
18740 Disable command tracing.
18741 @item show trace-commands
18742 Display the current state of command tracing.
18745 @node Debugging Output
18746 @section Optional Messages about Internal Happenings
18747 @cindex optional debugging messages
18749 @value{GDBN} has commands that enable optional debugging messages from
18750 various @value{GDBN} subsystems; normally these commands are of
18751 interest to @value{GDBN} maintainers, or when reporting a bug. This
18752 section documents those commands.
18755 @kindex set exec-done-display
18756 @item set exec-done-display
18757 Turns on or off the notification of asynchronous commands'
18758 completion. When on, @value{GDBN} will print a message when an
18759 asynchronous command finishes its execution. The default is off.
18760 @kindex show exec-done-display
18761 @item show exec-done-display
18762 Displays the current setting of asynchronous command completion
18765 @cindex gdbarch debugging info
18766 @cindex architecture debugging info
18767 @item set debug arch
18768 Turns on or off display of gdbarch debugging info. The default is off
18770 @item show debug arch
18771 Displays the current state of displaying gdbarch debugging info.
18772 @item set debug aix-thread
18773 @cindex AIX threads
18774 Display debugging messages about inner workings of the AIX thread
18776 @item show debug aix-thread
18777 Show the current state of AIX thread debugging info display.
18778 @item set debug dwarf2-die
18779 @cindex DWARF2 DIEs
18780 Dump DWARF2 DIEs after they are read in.
18781 The value is the number of nesting levels to print.
18782 A value of zero turns off the display.
18783 @item show debug dwarf2-die
18784 Show the current state of DWARF2 DIE debugging.
18785 @item set debug displaced
18786 @cindex displaced stepping debugging info
18787 Turns on or off display of @value{GDBN} debugging info for the
18788 displaced stepping support. The default is off.
18789 @item show debug displaced
18790 Displays the current state of displaying @value{GDBN} debugging info
18791 related to displaced stepping.
18792 @item set debug event
18793 @cindex event debugging info
18794 Turns on or off display of @value{GDBN} event debugging info. The
18796 @item show debug event
18797 Displays the current state of displaying @value{GDBN} event debugging
18799 @item set debug expression
18800 @cindex expression debugging info
18801 Turns on or off display of debugging info about @value{GDBN}
18802 expression parsing. The default is off.
18803 @item show debug expression
18804 Displays the current state of displaying debugging info about
18805 @value{GDBN} expression parsing.
18806 @item set debug frame
18807 @cindex frame debugging info
18808 Turns on or off display of @value{GDBN} frame debugging info. The
18810 @item show debug frame
18811 Displays the current state of displaying @value{GDBN} frame debugging
18813 @item set debug gnu-nat
18814 @cindex @sc{gnu}/Hurd debug messages
18815 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
18816 @item show debug gnu-nat
18817 Show the current state of @sc{gnu}/Hurd debugging messages.
18818 @item set debug infrun
18819 @cindex inferior debugging info
18820 Turns on or off display of @value{GDBN} debugging info for running the inferior.
18821 The default is off. @file{infrun.c} contains GDB's runtime state machine used
18822 for implementing operations such as single-stepping the inferior.
18823 @item show debug infrun
18824 Displays the current state of @value{GDBN} inferior debugging.
18825 @item set debug lin-lwp
18826 @cindex @sc{gnu}/Linux LWP debug messages
18827 @cindex Linux lightweight processes
18828 Turns on or off debugging messages from the Linux LWP debug support.
18829 @item show debug lin-lwp
18830 Show the current state of Linux LWP debugging messages.
18831 @item set debug lin-lwp-async
18832 @cindex @sc{gnu}/Linux LWP async debug messages
18833 @cindex Linux lightweight processes
18834 Turns on or off debugging messages from the Linux LWP async debug support.
18835 @item show debug lin-lwp-async
18836 Show the current state of Linux LWP async debugging messages.
18837 @item set debug observer
18838 @cindex observer debugging info
18839 Turns on or off display of @value{GDBN} observer debugging. This
18840 includes info such as the notification of observable events.
18841 @item show debug observer
18842 Displays the current state of observer debugging.
18843 @item set debug overload
18844 @cindex C@t{++} overload debugging info
18845 Turns on or off display of @value{GDBN} C@t{++} overload debugging
18846 info. This includes info such as ranking of functions, etc. The default
18848 @item show debug overload
18849 Displays the current state of displaying @value{GDBN} C@t{++} overload
18851 @cindex expression parser, debugging info
18852 @cindex debug expression parser
18853 @item set debug parser
18854 Turns on or off the display of expression parser debugging output.
18855 Internally, this sets the @code{yydebug} variable in the expression
18856 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
18857 details. The default is off.
18858 @item show debug parser
18859 Show the current state of expression parser debugging.
18860 @cindex packets, reporting on stdout
18861 @cindex serial connections, debugging
18862 @cindex debug remote protocol
18863 @cindex remote protocol debugging
18864 @cindex display remote packets
18865 @item set debug remote
18866 Turns on or off display of reports on all packets sent back and forth across
18867 the serial line to the remote machine. The info is printed on the
18868 @value{GDBN} standard output stream. The default is off.
18869 @item show debug remote
18870 Displays the state of display of remote packets.
18871 @item set debug serial
18872 Turns on or off display of @value{GDBN} serial debugging info. The
18874 @item show debug serial
18875 Displays the current state of displaying @value{GDBN} serial debugging
18877 @item set debug solib-frv
18878 @cindex FR-V shared-library debugging
18879 Turns on or off debugging messages for FR-V shared-library code.
18880 @item show debug solib-frv
18881 Display the current state of FR-V shared-library code debugging
18883 @item set debug target
18884 @cindex target debugging info
18885 Turns on or off display of @value{GDBN} target debugging info. This info
18886 includes what is going on at the target level of GDB, as it happens. The
18887 default is 0. Set it to 1 to track events, and to 2 to also track the
18888 value of large memory transfers. Changes to this flag do not take effect
18889 until the next time you connect to a target or use the @code{run} command.
18890 @item show debug target
18891 Displays the current state of displaying @value{GDBN} target debugging
18893 @item set debug timestamp
18894 @cindex timestampping debugging info
18895 Turns on or off display of timestamps with @value{GDBN} debugging info.
18896 When enabled, seconds and microseconds are displayed before each debugging
18898 @item show debug timestamp
18899 Displays the current state of displaying timestamps with @value{GDBN}
18901 @item set debugvarobj
18902 @cindex variable object debugging info
18903 Turns on or off display of @value{GDBN} variable object debugging
18904 info. The default is off.
18905 @item show debugvarobj
18906 Displays the current state of displaying @value{GDBN} variable object
18908 @item set debug xml
18909 @cindex XML parser debugging
18910 Turns on or off debugging messages for built-in XML parsers.
18911 @item show debug xml
18912 Displays the current state of XML debugging messages.
18915 @node Other Misc Settings
18916 @section Other Miscellaneous Settings
18917 @cindex miscellaneous settings
18920 @kindex set interactive-mode
18921 @item set interactive-mode
18922 If @code{on}, forces @value{GDBN} to operate interactively.
18923 If @code{off}, forces @value{GDBN} to operate non-interactively,
18924 If @code{auto} (the default), @value{GDBN} guesses which mode to use,
18925 based on whether the debugger was started in a terminal or not.
18927 In the vast majority of cases, the debugger should be able to guess
18928 correctly which mode should be used. But this setting can be useful
18929 in certain specific cases, such as running a MinGW @value{GDBN}
18930 inside a cygwin window.
18932 @kindex show interactive-mode
18933 @item show interactive-mode
18934 Displays whether the debugger is operating in interactive mode or not.
18937 @node Extending GDB
18938 @chapter Extending @value{GDBN}
18939 @cindex extending GDB
18941 @value{GDBN} provides two mechanisms for extension. The first is based
18942 on composition of @value{GDBN} commands, and the second is based on the
18943 Python scripting language.
18945 To facilitate the use of these extensions, @value{GDBN} is capable
18946 of evaluating the contents of a file. When doing so, @value{GDBN}
18947 can recognize which scripting language is being used by looking at
18948 the filename extension. Files with an unrecognized filename extension
18949 are always treated as a @value{GDBN} Command Files.
18950 @xref{Command Files,, Command files}.
18952 You can control how @value{GDBN} evaluates these files with the following
18956 @kindex set script-extension
18957 @kindex show script-extension
18958 @item set script-extension off
18959 All scripts are always evaluated as @value{GDBN} Command Files.
18961 @item set script-extension soft
18962 The debugger determines the scripting language based on filename
18963 extension. If this scripting language is supported, @value{GDBN}
18964 evaluates the script using that language. Otherwise, it evaluates
18965 the file as a @value{GDBN} Command File.
18967 @item set script-extension strict
18968 The debugger determines the scripting language based on filename
18969 extension, and evaluates the script using that language. If the
18970 language is not supported, then the evaluation fails.
18972 @item show script-extension
18973 Display the current value of the @code{script-extension} option.
18978 * Sequences:: Canned Sequences of Commands
18979 * Python:: Scripting @value{GDBN} using Python
18983 @section Canned Sequences of Commands
18985 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
18986 Command Lists}), @value{GDBN} provides two ways to store sequences of
18987 commands for execution as a unit: user-defined commands and command
18991 * Define:: How to define your own commands
18992 * Hooks:: Hooks for user-defined commands
18993 * Command Files:: How to write scripts of commands to be stored in a file
18994 * Output:: Commands for controlled output
18998 @subsection User-defined Commands
19000 @cindex user-defined command
19001 @cindex arguments, to user-defined commands
19002 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
19003 which you assign a new name as a command. This is done with the
19004 @code{define} command. User commands may accept up to 10 arguments
19005 separated by whitespace. Arguments are accessed within the user command
19006 via @code{$arg0@dots{}$arg9}. A trivial example:
19010 print $arg0 + $arg1 + $arg2
19015 To execute the command use:
19022 This defines the command @code{adder}, which prints the sum of
19023 its three arguments. Note the arguments are text substitutions, so they may
19024 reference variables, use complex expressions, or even perform inferior
19027 @cindex argument count in user-defined commands
19028 @cindex how many arguments (user-defined commands)
19029 In addition, @code{$argc} may be used to find out how many arguments have
19030 been passed. This expands to a number in the range 0@dots{}10.
19035 print $arg0 + $arg1
19038 print $arg0 + $arg1 + $arg2
19046 @item define @var{commandname}
19047 Define a command named @var{commandname}. If there is already a command
19048 by that name, you are asked to confirm that you want to redefine it.
19049 @var{commandname} may be a bare command name consisting of letters,
19050 numbers, dashes, and underscores. It may also start with any predefined
19051 prefix command. For example, @samp{define target my-target} creates
19052 a user-defined @samp{target my-target} command.
19054 The definition of the command is made up of other @value{GDBN} command lines,
19055 which are given following the @code{define} command. The end of these
19056 commands is marked by a line containing @code{end}.
19059 @kindex end@r{ (user-defined commands)}
19060 @item document @var{commandname}
19061 Document the user-defined command @var{commandname}, so that it can be
19062 accessed by @code{help}. The command @var{commandname} must already be
19063 defined. This command reads lines of documentation just as @code{define}
19064 reads the lines of the command definition, ending with @code{end}.
19065 After the @code{document} command is finished, @code{help} on command
19066 @var{commandname} displays the documentation you have written.
19068 You may use the @code{document} command again to change the
19069 documentation of a command. Redefining the command with @code{define}
19070 does not change the documentation.
19072 @kindex dont-repeat
19073 @cindex don't repeat command
19075 Used inside a user-defined command, this tells @value{GDBN} that this
19076 command should not be repeated when the user hits @key{RET}
19077 (@pxref{Command Syntax, repeat last command}).
19079 @kindex help user-defined
19080 @item help user-defined
19081 List all user-defined commands, with the first line of the documentation
19086 @itemx show user @var{commandname}
19087 Display the @value{GDBN} commands used to define @var{commandname} (but
19088 not its documentation). If no @var{commandname} is given, display the
19089 definitions for all user-defined commands.
19091 @cindex infinite recursion in user-defined commands
19092 @kindex show max-user-call-depth
19093 @kindex set max-user-call-depth
19094 @item show max-user-call-depth
19095 @itemx set max-user-call-depth
19096 The value of @code{max-user-call-depth} controls how many recursion
19097 levels are allowed in user-defined commands before @value{GDBN} suspects an
19098 infinite recursion and aborts the command.
19101 In addition to the above commands, user-defined commands frequently
19102 use control flow commands, described in @ref{Command Files}.
19104 When user-defined commands are executed, the
19105 commands of the definition are not printed. An error in any command
19106 stops execution of the user-defined command.
19108 If used interactively, commands that would ask for confirmation proceed
19109 without asking when used inside a user-defined command. Many @value{GDBN}
19110 commands that normally print messages to say what they are doing omit the
19111 messages when used in a user-defined command.
19114 @subsection User-defined Command Hooks
19115 @cindex command hooks
19116 @cindex hooks, for commands
19117 @cindex hooks, pre-command
19120 You may define @dfn{hooks}, which are a special kind of user-defined
19121 command. Whenever you run the command @samp{foo}, if the user-defined
19122 command @samp{hook-foo} exists, it is executed (with no arguments)
19123 before that command.
19125 @cindex hooks, post-command
19127 A hook may also be defined which is run after the command you executed.
19128 Whenever you run the command @samp{foo}, if the user-defined command
19129 @samp{hookpost-foo} exists, it is executed (with no arguments) after
19130 that command. Post-execution hooks may exist simultaneously with
19131 pre-execution hooks, for the same command.
19133 It is valid for a hook to call the command which it hooks. If this
19134 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
19136 @c It would be nice if hookpost could be passed a parameter indicating
19137 @c if the command it hooks executed properly or not. FIXME!
19139 @kindex stop@r{, a pseudo-command}
19140 In addition, a pseudo-command, @samp{stop} exists. Defining
19141 (@samp{hook-stop}) makes the associated commands execute every time
19142 execution stops in your program: before breakpoint commands are run,
19143 displays are printed, or the stack frame is printed.
19145 For example, to ignore @code{SIGALRM} signals while
19146 single-stepping, but treat them normally during normal execution,
19151 handle SIGALRM nopass
19155 handle SIGALRM pass
19158 define hook-continue
19159 handle SIGALRM pass
19163 As a further example, to hook at the beginning and end of the @code{echo}
19164 command, and to add extra text to the beginning and end of the message,
19172 define hookpost-echo
19176 (@value{GDBP}) echo Hello World
19177 <<<---Hello World--->>>
19182 You can define a hook for any single-word command in @value{GDBN}, but
19183 not for command aliases; you should define a hook for the basic command
19184 name, e.g.@: @code{backtrace} rather than @code{bt}.
19185 @c FIXME! So how does Joe User discover whether a command is an alias
19187 You can hook a multi-word command by adding @code{hook-} or
19188 @code{hookpost-} to the last word of the command, e.g.@:
19189 @samp{define target hook-remote} to add a hook to @samp{target remote}.
19191 If an error occurs during the execution of your hook, execution of
19192 @value{GDBN} commands stops and @value{GDBN} issues a prompt
19193 (before the command that you actually typed had a chance to run).
19195 If you try to define a hook which does not match any known command, you
19196 get a warning from the @code{define} command.
19198 @node Command Files
19199 @subsection Command Files
19201 @cindex command files
19202 @cindex scripting commands
19203 A command file for @value{GDBN} is a text file made of lines that are
19204 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
19205 also be included. An empty line in a command file does nothing; it
19206 does not mean to repeat the last command, as it would from the
19209 You can request the execution of a command file with the @code{source}
19210 command. Note that the @code{source} command is also used to evaluate
19211 scripts that are not Command Files. The exact behavior can be configured
19212 using the @code{script-extension} setting.
19213 @xref{Extending GDB,, Extending GDB}.
19217 @cindex execute commands from a file
19218 @item source [@code{-v}] @var{filename}
19219 Execute the command file @var{filename}.
19222 The lines in a command file are generally executed sequentially,
19223 unless the order of execution is changed by one of the
19224 @emph{flow-control commands} described below. The commands are not
19225 printed as they are executed. An error in any command terminates
19226 execution of the command file and control is returned to the console.
19228 @value{GDBN} searches for @var{filename} in the current directory and then
19229 on the search path (specified with the @samp{directory} command).
19231 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
19232 each command as it is executed. The option must be given before
19233 @var{filename}, and is interpreted as part of the filename anywhere else.
19235 Commands that would ask for confirmation if used interactively proceed
19236 without asking when used in a command file. Many @value{GDBN} commands that
19237 normally print messages to say what they are doing omit the messages
19238 when called from command files.
19240 @value{GDBN} also accepts command input from standard input. In this
19241 mode, normal output goes to standard output and error output goes to
19242 standard error. Errors in a command file supplied on standard input do
19243 not terminate execution of the command file---execution continues with
19247 gdb < cmds > log 2>&1
19250 (The syntax above will vary depending on the shell used.) This example
19251 will execute commands from the file @file{cmds}. All output and errors
19252 would be directed to @file{log}.
19254 Since commands stored on command files tend to be more general than
19255 commands typed interactively, they frequently need to deal with
19256 complicated situations, such as different or unexpected values of
19257 variables and symbols, changes in how the program being debugged is
19258 built, etc. @value{GDBN} provides a set of flow-control commands to
19259 deal with these complexities. Using these commands, you can write
19260 complex scripts that loop over data structures, execute commands
19261 conditionally, etc.
19268 This command allows to include in your script conditionally executed
19269 commands. The @code{if} command takes a single argument, which is an
19270 expression to evaluate. It is followed by a series of commands that
19271 are executed only if the expression is true (its value is nonzero).
19272 There can then optionally be an @code{else} line, followed by a series
19273 of commands that are only executed if the expression was false. The
19274 end of the list is marked by a line containing @code{end}.
19278 This command allows to write loops. Its syntax is similar to
19279 @code{if}: the command takes a single argument, which is an expression
19280 to evaluate, and must be followed by the commands to execute, one per
19281 line, terminated by an @code{end}. These commands are called the
19282 @dfn{body} of the loop. The commands in the body of @code{while} are
19283 executed repeatedly as long as the expression evaluates to true.
19287 This command exits the @code{while} loop in whose body it is included.
19288 Execution of the script continues after that @code{while}s @code{end}
19291 @kindex loop_continue
19292 @item loop_continue
19293 This command skips the execution of the rest of the body of commands
19294 in the @code{while} loop in whose body it is included. Execution
19295 branches to the beginning of the @code{while} loop, where it evaluates
19296 the controlling expression.
19298 @kindex end@r{ (if/else/while commands)}
19300 Terminate the block of commands that are the body of @code{if},
19301 @code{else}, or @code{while} flow-control commands.
19306 @subsection Commands for Controlled Output
19308 During the execution of a command file or a user-defined command, normal
19309 @value{GDBN} output is suppressed; the only output that appears is what is
19310 explicitly printed by the commands in the definition. This section
19311 describes three commands useful for generating exactly the output you
19316 @item echo @var{text}
19317 @c I do not consider backslash-space a standard C escape sequence
19318 @c because it is not in ANSI.
19319 Print @var{text}. Nonprinting characters can be included in
19320 @var{text} using C escape sequences, such as @samp{\n} to print a
19321 newline. @strong{No newline is printed unless you specify one.}
19322 In addition to the standard C escape sequences, a backslash followed
19323 by a space stands for a space. This is useful for displaying a
19324 string with spaces at the beginning or the end, since leading and
19325 trailing spaces are otherwise trimmed from all arguments.
19326 To print @samp{@w{ }and foo =@w{ }}, use the command
19327 @samp{echo \@w{ }and foo = \@w{ }}.
19329 A backslash at the end of @var{text} can be used, as in C, to continue
19330 the command onto subsequent lines. For example,
19333 echo This is some text\n\
19334 which is continued\n\
19335 onto several lines.\n
19338 produces the same output as
19341 echo This is some text\n
19342 echo which is continued\n
19343 echo onto several lines.\n
19347 @item output @var{expression}
19348 Print the value of @var{expression} and nothing but that value: no
19349 newlines, no @samp{$@var{nn} = }. The value is not entered in the
19350 value history either. @xref{Expressions, ,Expressions}, for more information
19353 @item output/@var{fmt} @var{expression}
19354 Print the value of @var{expression} in format @var{fmt}. You can use
19355 the same formats as for @code{print}. @xref{Output Formats,,Output
19356 Formats}, for more information.
19359 @item printf @var{template}, @var{expressions}@dots{}
19360 Print the values of one or more @var{expressions} under the control of
19361 the string @var{template}. To print several values, make
19362 @var{expressions} be a comma-separated list of individual expressions,
19363 which may be either numbers or pointers. Their values are printed as
19364 specified by @var{template}, exactly as a C program would do by
19365 executing the code below:
19368 printf (@var{template}, @var{expressions}@dots{});
19371 As in @code{C} @code{printf}, ordinary characters in @var{template}
19372 are printed verbatim, while @dfn{conversion specification} introduced
19373 by the @samp{%} character cause subsequent @var{expressions} to be
19374 evaluated, their values converted and formatted according to type and
19375 style information encoded in the conversion specifications, and then
19378 For example, you can print two values in hex like this:
19381 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
19384 @code{printf} supports all the standard @code{C} conversion
19385 specifications, including the flags and modifiers between the @samp{%}
19386 character and the conversion letter, with the following exceptions:
19390 The argument-ordering modifiers, such as @samp{2$}, are not supported.
19393 The modifier @samp{*} is not supported for specifying precision or
19397 The @samp{'} flag (for separation of digits into groups according to
19398 @code{LC_NUMERIC'}) is not supported.
19401 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
19405 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
19408 The conversion letters @samp{a} and @samp{A} are not supported.
19412 Note that the @samp{ll} type modifier is supported only if the
19413 underlying @code{C} implementation used to build @value{GDBN} supports
19414 the @code{long long int} type, and the @samp{L} type modifier is
19415 supported only if @code{long double} type is available.
19417 As in @code{C}, @code{printf} supports simple backslash-escape
19418 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
19419 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
19420 single character. Octal and hexadecimal escape sequences are not
19423 Additionally, @code{printf} supports conversion specifications for DFP
19424 (@dfn{Decimal Floating Point}) types using the following length modifiers
19425 together with a floating point specifier.
19430 @samp{H} for printing @code{Decimal32} types.
19433 @samp{D} for printing @code{Decimal64} types.
19436 @samp{DD} for printing @code{Decimal128} types.
19439 If the underlying @code{C} implementation used to build @value{GDBN} has
19440 support for the three length modifiers for DFP types, other modifiers
19441 such as width and precision will also be available for @value{GDBN} to use.
19443 In case there is no such @code{C} support, no additional modifiers will be
19444 available and the value will be printed in the standard way.
19446 Here's an example of printing DFP types using the above conversion letters:
19448 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
19454 @section Scripting @value{GDBN} using Python
19455 @cindex python scripting
19456 @cindex scripting with python
19458 You can script @value{GDBN} using the @uref{http://www.python.org/,
19459 Python programming language}. This feature is available only if
19460 @value{GDBN} was configured using @option{--with-python}.
19463 * Python Commands:: Accessing Python from @value{GDBN}.
19464 * Python API:: Accessing @value{GDBN} from Python.
19467 @node Python Commands
19468 @subsection Python Commands
19469 @cindex python commands
19470 @cindex commands to access python
19472 @value{GDBN} provides one command for accessing the Python interpreter,
19473 and one related setting:
19477 @item python @r{[}@var{code}@r{]}
19478 The @code{python} command can be used to evaluate Python code.
19480 If given an argument, the @code{python} command will evaluate the
19481 argument as a Python command. For example:
19484 (@value{GDBP}) python print 23
19488 If you do not provide an argument to @code{python}, it will act as a
19489 multi-line command, like @code{define}. In this case, the Python
19490 script is made up of subsequent command lines, given after the
19491 @code{python} command. This command list is terminated using a line
19492 containing @code{end}. For example:
19495 (@value{GDBP}) python
19497 End with a line saying just "end".
19503 @kindex maint set python print-stack
19504 @item maint set python print-stack
19505 By default, @value{GDBN} will print a stack trace when an error occurs
19506 in a Python script. This can be controlled using @code{maint set
19507 python print-stack}: if @code{on}, the default, then Python stack
19508 printing is enabled; if @code{off}, then Python stack printing is
19512 It is also possible to execute a Python script from the @value{GDBN}
19516 @item source @file{script-name}
19517 The script name must end with @samp{.py} and @value{GDBN} must be configured
19518 to recognize the script language based on filename extension using
19519 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
19521 @item python execfile ("script-name")
19522 This method is based on the @code{execfile} Python built-in function,
19523 and thus is always available.
19527 @subsection Python API
19529 @cindex programming in python
19531 @cindex python stdout
19532 @cindex python pagination
19533 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
19534 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
19535 A Python program which outputs to one of these streams may have its
19536 output interrupted by the user (@pxref{Screen Size}). In this
19537 situation, a Python @code{KeyboardInterrupt} exception is thrown.
19540 * Basic Python:: Basic Python Functions.
19541 * Exception Handling::
19542 * Auto-loading:: Automatically loading Python code.
19543 * Values From Inferior::
19544 * Types In Python:: Python representation of types.
19545 * Pretty Printing:: Pretty-printing values.
19546 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
19547 * Commands In Python:: Implementing new commands in Python.
19548 * Functions In Python:: Writing new convenience functions.
19549 * Objfiles In Python:: Object files.
19550 * Frames In Python:: Acessing inferior stack frames from Python.
19551 * Lazy Strings In Python:: Python representation of lazy strings.
19555 @subsubsection Basic Python
19557 @cindex python functions
19558 @cindex python module
19560 @value{GDBN} introduces a new Python module, named @code{gdb}. All
19561 methods and classes added by @value{GDBN} are placed in this module.
19562 @value{GDBN} automatically @code{import}s the @code{gdb} module for
19563 use in all scripts evaluated by the @code{python} command.
19565 @findex gdb.execute
19566 @defun execute command [from_tty]
19567 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
19568 If a GDB exception happens while @var{command} runs, it is
19569 translated as described in @ref{Exception Handling,,Exception Handling}.
19570 If no exceptions occur, this function returns @code{None}.
19572 @var{from_tty} specifies whether @value{GDBN} ought to consider this
19573 command as having originated from the user invoking it interactively.
19574 It must be a boolean value. If omitted, it defaults to @code{False}.
19577 @findex gdb.parameter
19578 @defun parameter parameter
19579 Return the value of a @value{GDBN} parameter. @var{parameter} is a
19580 string naming the parameter to look up; @var{parameter} may contain
19581 spaces if the parameter has a multi-part name. For example,
19582 @samp{print object} is a valid parameter name.
19584 If the named parameter does not exist, this function throws a
19585 @code{RuntimeError}. Otherwise, the parameter's value is converted to
19586 a Python value of the appropriate type, and returned.
19589 @findex gdb.history
19590 @defun history number
19591 Return a value from @value{GDBN}'s value history (@pxref{Value
19592 History}). @var{number} indicates which history element to return.
19593 If @var{number} is negative, then @value{GDBN} will take its absolute value
19594 and count backward from the last element (i.e., the most recent element) to
19595 find the value to return. If @var{number} is zero, then @value{GDBN} will
19596 return the most recent element. If the element specified by @var{number}
19597 doesn't exist in the value history, a @code{RuntimeError} exception will be
19600 If no exception is raised, the return value is always an instance of
19601 @code{gdb.Value} (@pxref{Values From Inferior}).
19604 @findex gdb.parse_and_eval
19605 @defun parse_and_eval expression
19606 Parse @var{expression} as an expression in the current language,
19607 evaluate it, and return the result as a @code{gdb.Value}.
19608 @var{expression} must be a string.
19610 This function can be useful when implementing a new command
19611 (@pxref{Commands In Python}), as it provides a way to parse the
19612 command's argument as an expression. It is also useful simply to
19613 compute values, for example, it is the only way to get the value of a
19614 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
19618 @defun write string
19619 Print a string to @value{GDBN}'s paginated standard output stream.
19620 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
19621 call this function.
19626 Flush @value{GDBN}'s paginated standard output stream. Flushing
19627 @code{sys.stdout} or @code{sys.stderr} will automatically call this
19631 @node Exception Handling
19632 @subsubsection Exception Handling
19633 @cindex python exceptions
19634 @cindex exceptions, python
19636 When executing the @code{python} command, Python exceptions
19637 uncaught within the Python code are translated to calls to
19638 @value{GDBN} error-reporting mechanism. If the command that called
19639 @code{python} does not handle the error, @value{GDBN} will
19640 terminate it and print an error message containing the Python
19641 exception name, the associated value, and the Python call stack
19642 backtrace at the point where the exception was raised. Example:
19645 (@value{GDBP}) python print foo
19646 Traceback (most recent call last):
19647 File "<string>", line 1, in <module>
19648 NameError: name 'foo' is not defined
19651 @value{GDBN} errors that happen in @value{GDBN} commands invoked by Python
19652 code are converted to Python @code{RuntimeError} exceptions. User
19653 interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
19654 prompt) is translated to a Python @code{KeyboardInterrupt}
19655 exception. If you catch these exceptions in your Python code, your
19656 exception handler will see @code{RuntimeError} or
19657 @code{KeyboardInterrupt} as the exception type, the @value{GDBN} error
19658 message as its value, and the Python call stack backtrace at the
19659 Python statement closest to where the @value{GDBN} error occured as the
19663 @subsubsection Auto-loading
19664 @cindex auto-loading, Python
19666 When a new object file is read (for example, due to the @code{file}
19667 command, or because the inferior has loaded a shared library),
19668 @value{GDBN} will look for a file named @file{@var{objfile}-gdb.py},
19669 where @var{objfile} is the object file's real name, formed by ensuring
19670 that the file name is absolute, following all symlinks, and resolving
19671 @code{.} and @code{..} components. If this file exists and is
19672 readable, @value{GDBN} will evaluate it as a Python script.
19674 If this file does not exist, and if the parameter
19675 @code{debug-file-directory} is set (@pxref{Separate Debug Files}),
19676 then @value{GDBN} will use for its each separated directory component
19677 @code{component} the file named @file{@code{component}/@var{real-name}}, where
19678 @var{real-name} is the object file's real name, as described above.
19680 Finally, if this file does not exist, then @value{GDBN} will look for
19681 a file named @file{@var{data-directory}/python/auto-load/@var{real-name}}, where
19682 @var{data-directory} is @value{GDBN}'s data directory (available via
19683 @code{show data-directory}, @pxref{Data Files}), and @var{real-name}
19684 is the object file's real name, as described above.
19686 When reading an auto-loaded file, @value{GDBN} sets the ``current
19687 objfile''. This is available via the @code{gdb.current_objfile}
19688 function (@pxref{Objfiles In Python}). This can be useful for
19689 registering objfile-specific pretty-printers.
19691 The auto-loading feature is useful for supplying application-specific
19692 debugging commands and scripts. You can enable or disable this
19693 feature, and view its current state.
19696 @kindex maint set python auto-load
19697 @item maint set python auto-load [yes|no]
19698 Enable or disable the Python auto-loading feature.
19700 @kindex show python auto-load
19701 @item show python auto-load
19702 Show whether Python auto-loading is enabled or disabled.
19705 @value{GDBN} does not track which files it has already auto-loaded.
19706 So, your @samp{-gdb.py} file should take care to ensure that it may be
19707 evaluated multiple times without error.
19709 @node Values From Inferior
19710 @subsubsection Values From Inferior
19711 @cindex values from inferior, with Python
19712 @cindex python, working with values from inferior
19714 @cindex @code{gdb.Value}
19715 @value{GDBN} provides values it obtains from the inferior program in
19716 an object of type @code{gdb.Value}. @value{GDBN} uses this object
19717 for its internal bookkeeping of the inferior's values, and for
19718 fetching values when necessary.
19720 Inferior values that are simple scalars can be used directly in
19721 Python expressions that are valid for the value's data type. Here's
19722 an example for an integer or floating-point value @code{some_val}:
19729 As result of this, @code{bar} will also be a @code{gdb.Value} object
19730 whose values are of the same type as those of @code{some_val}.
19732 Inferior values that are structures or instances of some class can
19733 be accessed using the Python @dfn{dictionary syntax}. For example, if
19734 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
19735 can access its @code{foo} element with:
19738 bar = some_val['foo']
19741 Again, @code{bar} will also be a @code{gdb.Value} object.
19743 The following attributes are provided:
19746 @defivar Value address
19747 If this object is addressable, this read-only attribute holds a
19748 @code{gdb.Value} object representing the address. Otherwise,
19749 this attribute holds @code{None}.
19752 @cindex optimized out value in Python
19753 @defivar Value is_optimized_out
19754 This read-only boolean attribute is true if the compiler optimized out
19755 this value, thus it is not available for fetching from the inferior.
19758 @defivar Value type
19759 The type of this @code{gdb.Value}. The value of this attribute is a
19760 @code{gdb.Type} object.
19764 The following methods are provided:
19767 @defmethod Value cast type
19768 Return a new instance of @code{gdb.Value} that is the result of
19769 casting this instance to the type described by @var{type}, which must
19770 be a @code{gdb.Type} object. If the cast cannot be performed for some
19771 reason, this method throws an exception.
19774 @defmethod Value dereference
19775 For pointer data types, this method returns a new @code{gdb.Value} object
19776 whose contents is the object pointed to by the pointer. For example, if
19777 @code{foo} is a C pointer to an @code{int}, declared in your C program as
19784 then you can use the corresponding @code{gdb.Value} to access what
19785 @code{foo} points to like this:
19788 bar = foo.dereference ()
19791 The result @code{bar} will be a @code{gdb.Value} object holding the
19792 value pointed to by @code{foo}.
19795 @defmethod Value string @r{[}encoding@r{]} @r{[}errors@r{]} @r{[}length@r{]}
19796 If this @code{gdb.Value} represents a string, then this method
19797 converts the contents to a Python string. Otherwise, this method will
19798 throw an exception.
19800 Strings are recognized in a language-specific way; whether a given
19801 @code{gdb.Value} represents a string is determined by the current
19804 For C-like languages, a value is a string if it is a pointer to or an
19805 array of characters or ints. The string is assumed to be terminated
19806 by a zero of the appropriate width. However if the optional length
19807 argument is given, the string will be converted to that given length,
19808 ignoring any embedded zeros that the string may contain.
19810 If the optional @var{encoding} argument is given, it must be a string
19811 naming the encoding of the string in the @code{gdb.Value}, such as
19812 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
19813 the same encodings as the corresponding argument to Python's
19814 @code{string.decode} method, and the Python codec machinery will be used
19815 to convert the string. If @var{encoding} is not given, or if
19816 @var{encoding} is the empty string, then either the @code{target-charset}
19817 (@pxref{Character Sets}) will be used, or a language-specific encoding
19818 will be used, if the current language is able to supply one.
19820 The optional @var{errors} argument is the same as the corresponding
19821 argument to Python's @code{string.decode} method.
19823 If the optional @var{length} argument is given, the string will be
19824 fetched and converted to the given length.
19827 @defmethod Value lazy_string @r{[}encoding@r{]} @r{[}length@r{]}
19828 If this @code{gdb.Value} represents a string, then this method
19829 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
19830 In Python}). Otherwise, this method will throw an exception.
19832 If the optional @var{encoding} argument is given, it must be a string
19833 naming the encoding of the @code{gdb.LazyString}. Some examples are:
19834 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
19835 @var{encoding} argument is an encoding that @value{GDBN} does
19836 recognize, @value{GDBN} will raise an error.
19838 When a lazy string is printed, the @value{GDBN} encoding machinery is
19839 used to convert the string during printing. If the optional
19840 @var{encoding} argument is not provided, or is an empty string,
19841 @value{GDBN} will automatically select the encoding most suitable for
19842 the string type. For further information on encoding in @value{GDBN}
19843 please see @ref{Character Sets}.
19845 If the optional @var{length} argument is given, the string will be
19846 fetched and encoded to the length of characters specified. If
19847 the @var{length} argument is not provided, the string will be fetched
19848 and encoded until a null of appropriate width is found.
19852 @node Types In Python
19853 @subsubsection Types In Python
19854 @cindex types in Python
19855 @cindex Python, working with types
19858 @value{GDBN} represents types from the inferior using the class
19861 The following type-related functions are available in the @code{gdb}
19864 @findex gdb.lookup_type
19865 @defun lookup_type name [block]
19866 This function looks up a type by name. @var{name} is the name of the
19867 type to look up. It must be a string.
19869 Ordinarily, this function will return an instance of @code{gdb.Type}.
19870 If the named type cannot be found, it will throw an exception.
19873 An instance of @code{Type} has the following attributes:
19877 The type code for this type. The type code will be one of the
19878 @code{TYPE_CODE_} constants defined below.
19881 @defivar Type sizeof
19882 The size of this type, in target @code{char} units. Usually, a
19883 target's @code{char} type will be an 8-bit byte. However, on some
19884 unusual platforms, this type may have a different size.
19888 The tag name for this type. The tag name is the name after
19889 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
19890 languages have this concept. If this type has no tag name, then
19891 @code{None} is returned.
19895 The following methods are provided:
19898 @defmethod Type fields
19899 For structure and union types, this method returns the fields. Range
19900 types have two fields, the minimum and maximum values. Enum types
19901 have one field per enum constant. Function and method types have one
19902 field per parameter. The base types of C@t{++} classes are also
19903 represented as fields. If the type has no fields, or does not fit
19904 into one of these categories, an empty sequence will be returned.
19906 Each field is an object, with some pre-defined attributes:
19909 This attribute is not available for @code{static} fields (as in
19910 C@t{++} or Java). For non-@code{static} fields, the value is the bit
19911 position of the field.
19914 The name of the field, or @code{None} for anonymous fields.
19917 This is @code{True} if the field is artificial, usually meaning that
19918 it was provided by the compiler and not the user. This attribute is
19919 always provided, and is @code{False} if the field is not artificial.
19921 @item is_base_class
19922 This is @code{True} if the field represents a base class of a C@t{++}
19923 structure. This attribute is always provided, and is @code{False}
19924 if the field is not a base class of the type that is the argument of
19925 @code{fields}, or if that type was not a C@t{++} class.
19928 If the field is packed, or is a bitfield, then this will have a
19929 non-zero value, which is the size of the field in bits. Otherwise,
19930 this will be zero; in this case the field's size is given by its type.
19933 The type of the field. This is usually an instance of @code{Type},
19934 but it can be @code{None} in some situations.
19938 @defmethod Type const
19939 Return a new @code{gdb.Type} object which represents a
19940 @code{const}-qualified variant of this type.
19943 @defmethod Type volatile
19944 Return a new @code{gdb.Type} object which represents a
19945 @code{volatile}-qualified variant of this type.
19948 @defmethod Type unqualified
19949 Return a new @code{gdb.Type} object which represents an unqualified
19950 variant of this type. That is, the result is neither @code{const} nor
19954 @defmethod Type range
19955 Return a Python @code{Tuple} object that contains two elements: the
19956 low bound of the argument type and the high bound of that type. If
19957 the type does not have a range, @value{GDBN} will raise a
19958 @code{RuntimeError} exception.
19961 @defmethod Type reference
19962 Return a new @code{gdb.Type} object which represents a reference to this
19966 @defmethod Type pointer
19967 Return a new @code{gdb.Type} object which represents a pointer to this
19971 @defmethod Type strip_typedefs
19972 Return a new @code{gdb.Type} that represents the real type,
19973 after removing all layers of typedefs.
19976 @defmethod Type target
19977 Return a new @code{gdb.Type} object which represents the target type
19980 For a pointer type, the target type is the type of the pointed-to
19981 object. For an array type (meaning C-like arrays), the target type is
19982 the type of the elements of the array. For a function or method type,
19983 the target type is the type of the return value. For a complex type,
19984 the target type is the type of the elements. For a typedef, the
19985 target type is the aliased type.
19987 If the type does not have a target, this method will throw an
19991 @defmethod Type template_argument n
19992 If this @code{gdb.Type} is an instantiation of a template, this will
19993 return a new @code{gdb.Type} which represents the type of the
19994 @var{n}th template argument.
19996 If this @code{gdb.Type} is not a template type, this will throw an
19997 exception. Ordinarily, only C@t{++} code will have template types.
19999 @var{name} is searched for globally.
20004 Each type has a code, which indicates what category this type falls
20005 into. The available type categories are represented by constants
20006 defined in the @code{gdb} module:
20009 @findex TYPE_CODE_PTR
20010 @findex gdb.TYPE_CODE_PTR
20011 @item TYPE_CODE_PTR
20012 The type is a pointer.
20014 @findex TYPE_CODE_ARRAY
20015 @findex gdb.TYPE_CODE_ARRAY
20016 @item TYPE_CODE_ARRAY
20017 The type is an array.
20019 @findex TYPE_CODE_STRUCT
20020 @findex gdb.TYPE_CODE_STRUCT
20021 @item TYPE_CODE_STRUCT
20022 The type is a structure.
20024 @findex TYPE_CODE_UNION
20025 @findex gdb.TYPE_CODE_UNION
20026 @item TYPE_CODE_UNION
20027 The type is a union.
20029 @findex TYPE_CODE_ENUM
20030 @findex gdb.TYPE_CODE_ENUM
20031 @item TYPE_CODE_ENUM
20032 The type is an enum.
20034 @findex TYPE_CODE_FLAGS
20035 @findex gdb.TYPE_CODE_FLAGS
20036 @item TYPE_CODE_FLAGS
20037 A bit flags type, used for things such as status registers.
20039 @findex TYPE_CODE_FUNC
20040 @findex gdb.TYPE_CODE_FUNC
20041 @item TYPE_CODE_FUNC
20042 The type is a function.
20044 @findex TYPE_CODE_INT
20045 @findex gdb.TYPE_CODE_INT
20046 @item TYPE_CODE_INT
20047 The type is an integer type.
20049 @findex TYPE_CODE_FLT
20050 @findex gdb.TYPE_CODE_FLT
20051 @item TYPE_CODE_FLT
20052 A floating point type.
20054 @findex TYPE_CODE_VOID
20055 @findex gdb.TYPE_CODE_VOID
20056 @item TYPE_CODE_VOID
20057 The special type @code{void}.
20059 @findex TYPE_CODE_SET
20060 @findex gdb.TYPE_CODE_SET
20061 @item TYPE_CODE_SET
20064 @findex TYPE_CODE_RANGE
20065 @findex gdb.TYPE_CODE_RANGE
20066 @item TYPE_CODE_RANGE
20067 A range type, that is, an integer type with bounds.
20069 @findex TYPE_CODE_STRING
20070 @findex gdb.TYPE_CODE_STRING
20071 @item TYPE_CODE_STRING
20072 A string type. Note that this is only used for certain languages with
20073 language-defined string types; C strings are not represented this way.
20075 @findex TYPE_CODE_BITSTRING
20076 @findex gdb.TYPE_CODE_BITSTRING
20077 @item TYPE_CODE_BITSTRING
20080 @findex TYPE_CODE_ERROR
20081 @findex gdb.TYPE_CODE_ERROR
20082 @item TYPE_CODE_ERROR
20083 An unknown or erroneous type.
20085 @findex TYPE_CODE_METHOD
20086 @findex gdb.TYPE_CODE_METHOD
20087 @item TYPE_CODE_METHOD
20088 A method type, as found in C@t{++} or Java.
20090 @findex TYPE_CODE_METHODPTR
20091 @findex gdb.TYPE_CODE_METHODPTR
20092 @item TYPE_CODE_METHODPTR
20093 A pointer-to-member-function.
20095 @findex TYPE_CODE_MEMBERPTR
20096 @findex gdb.TYPE_CODE_MEMBERPTR
20097 @item TYPE_CODE_MEMBERPTR
20098 A pointer-to-member.
20100 @findex TYPE_CODE_REF
20101 @findex gdb.TYPE_CODE_REF
20102 @item TYPE_CODE_REF
20105 @findex TYPE_CODE_CHAR
20106 @findex gdb.TYPE_CODE_CHAR
20107 @item TYPE_CODE_CHAR
20110 @findex TYPE_CODE_BOOL
20111 @findex gdb.TYPE_CODE_BOOL
20112 @item TYPE_CODE_BOOL
20115 @findex TYPE_CODE_COMPLEX
20116 @findex gdb.TYPE_CODE_COMPLEX
20117 @item TYPE_CODE_COMPLEX
20118 A complex float type.
20120 @findex TYPE_CODE_TYPEDEF
20121 @findex gdb.TYPE_CODE_TYPEDEF
20122 @item TYPE_CODE_TYPEDEF
20123 A typedef to some other type.
20125 @findex TYPE_CODE_NAMESPACE
20126 @findex gdb.TYPE_CODE_NAMESPACE
20127 @item TYPE_CODE_NAMESPACE
20128 A C@t{++} namespace.
20130 @findex TYPE_CODE_DECFLOAT
20131 @findex gdb.TYPE_CODE_DECFLOAT
20132 @item TYPE_CODE_DECFLOAT
20133 A decimal floating point type.
20135 @findex TYPE_CODE_INTERNAL_FUNCTION
20136 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
20137 @item TYPE_CODE_INTERNAL_FUNCTION
20138 A function internal to @value{GDBN}. This is the type used to represent
20139 convenience functions.
20142 @node Pretty Printing
20143 @subsubsection Pretty Printing
20145 @value{GDBN} provides a mechanism to allow pretty-printing of values
20146 using Python code. The pretty-printer API allows application-specific
20147 code to greatly simplify the display of complex objects. This
20148 mechanism works for both MI and the CLI.
20150 For example, here is how a C@t{++} @code{std::string} looks without a
20154 (@value{GDBP}) print s
20156 static npos = 4294967295,
20158 <std::allocator<char>> = @{
20159 <__gnu_cxx::new_allocator<char>> = @{<No data fields>@}, <No data fields>@},
20160 members of std::basic_string<char, std::char_traits<char>, std::allocator<char> >::_Alloc_hider:
20161 _M_p = 0x804a014 "abcd"
20166 After a pretty-printer for @code{std::string} has been installed, only
20167 the contents are printed:
20170 (@value{GDBP}) print s
20174 A pretty-printer is just an object that holds a value and implements a
20175 specific interface, defined here.
20177 @defop Operation {pretty printer} children (self)
20178 @value{GDBN} will call this method on a pretty-printer to compute the
20179 children of the pretty-printer's value.
20181 This method must return an object conforming to the Python iterator
20182 protocol. Each item returned by the iterator must be a tuple holding
20183 two elements. The first element is the ``name'' of the child; the
20184 second element is the child's value. The value can be any Python
20185 object which is convertible to a @value{GDBN} value.
20187 This method is optional. If it does not exist, @value{GDBN} will act
20188 as though the value has no children.
20191 @defop Operation {pretty printer} display_hint (self)
20192 The CLI may call this method and use its result to change the
20193 formatting of a value. The result will also be supplied to an MI
20194 consumer as a @samp{displayhint} attribute of the variable being
20197 This method is optional. If it does exist, this method must return a
20200 Some display hints are predefined by @value{GDBN}:
20204 Indicate that the object being printed is ``array-like''. The CLI
20205 uses this to respect parameters such as @code{set print elements} and
20206 @code{set print array}.
20209 Indicate that the object being printed is ``map-like'', and that the
20210 children of this value can be assumed to alternate between keys and
20214 Indicate that the object being printed is ``string-like''. If the
20215 printer's @code{to_string} method returns a Python string of some
20216 kind, then @value{GDBN} will call its internal language-specific
20217 string-printing function to format the string. For the CLI this means
20218 adding quotation marks, possibly escaping some characters, respecting
20219 @code{set print elements}, and the like.
20223 @defop Operation {pretty printer} to_string (self)
20224 @value{GDBN} will call this method to display the string
20225 representation of the value passed to the object's constructor.
20227 When printing from the CLI, if the @code{to_string} method exists,
20228 then @value{GDBN} will prepend its result to the values returned by
20229 @code{children}. Exactly how this formatting is done is dependent on
20230 the display hint, and may change as more hints are added. Also,
20231 depending on the print settings (@pxref{Print Settings}), the CLI may
20232 print just the result of @code{to_string} in a stack trace, omitting
20233 the result of @code{children}.
20235 If this method returns a string, it is printed verbatim.
20237 Otherwise, if this method returns an instance of @code{gdb.Value},
20238 then @value{GDBN} prints this value. This may result in a call to
20239 another pretty-printer.
20241 If instead the method returns a Python value which is convertible to a
20242 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
20243 the resulting value. Again, this may result in a call to another
20244 pretty-printer. Python scalars (integers, floats, and booleans) and
20245 strings are convertible to @code{gdb.Value}; other types are not.
20247 If the result is not one of these types, an exception is raised.
20250 @node Selecting Pretty-Printers
20251 @subsubsection Selecting Pretty-Printers
20253 The Python list @code{gdb.pretty_printers} contains an array of
20254 functions that have been registered via addition as a pretty-printer.
20255 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
20258 A function on one of these lists is passed a single @code{gdb.Value}
20259 argument and should return a pretty-printer object conforming to the
20260 interface definition above (@pxref{Pretty Printing}). If a function
20261 cannot create a pretty-printer for the value, it should return
20264 @value{GDBN} first checks the @code{pretty_printers} attribute of each
20265 @code{gdb.Objfile} and iteratively calls each function in the list for
20266 that @code{gdb.Objfile} until it receives a pretty-printer object.
20267 After these lists have been exhausted, it tries the global
20268 @code{gdb.pretty-printers} list, again calling each function until an
20269 object is returned.
20271 The order in which the objfiles are searched is not specified. For a
20272 given list, functions are always invoked from the head of the list,
20273 and iterated over sequentially until the end of the list, or a printer
20274 object is returned.
20276 Here is an example showing how a @code{std::string} printer might be
20280 class StdStringPrinter:
20281 "Print a std::string"
20283 def __init__ (self, val):
20286 def to_string (self):
20287 return self.val['_M_dataplus']['_M_p']
20289 def display_hint (self):
20293 And here is an example showing how a lookup function for the printer
20294 example above might be written.
20297 def str_lookup_function (val):
20299 lookup_tag = val.type.tag
20300 regex = re.compile ("^std::basic_string<char,.*>$")
20301 if lookup_tag == None:
20303 if regex.match (lookup_tag):
20304 return StdStringPrinter (val)
20309 The example lookup function extracts the value's type, and attempts to
20310 match it to a type that it can pretty-print. If it is a type the
20311 printer can pretty-print, it will return a printer object. If not, it
20312 returns @code{None}.
20314 We recommend that you put your core pretty-printers into a Python
20315 package. If your pretty-printers are for use with a library, we
20316 further recommend embedding a version number into the package name.
20317 This practice will enable @value{GDBN} to load multiple versions of
20318 your pretty-printers at the same time, because they will have
20321 You should write auto-loaded code (@pxref{Auto-loading}) such that it
20322 can be evaluated multiple times without changing its meaning. An
20323 ideal auto-load file will consist solely of @code{import}s of your
20324 printer modules, followed by a call to a register pretty-printers with
20325 the current objfile.
20327 Taken as a whole, this approach will scale nicely to multiple
20328 inferiors, each potentially using a different library version.
20329 Embedding a version number in the Python package name will ensure that
20330 @value{GDBN} is able to load both sets of printers simultaneously.
20331 Then, because the search for pretty-printers is done by objfile, and
20332 because your auto-loaded code took care to register your library's
20333 printers with a specific objfile, @value{GDBN} will find the correct
20334 printers for the specific version of the library used by each
20337 To continue the @code{std::string} example (@pxref{Pretty Printing}),
20338 this code might appear in @code{gdb.libstdcxx.v6}:
20341 def register_printers (objfile):
20342 objfile.pretty_printers.add (str_lookup_function)
20346 And then the corresponding contents of the auto-load file would be:
20349 import gdb.libstdcxx.v6
20350 gdb.libstdcxx.v6.register_printers (gdb.current_objfile ())
20353 @node Commands In Python
20354 @subsubsection Commands In Python
20356 @cindex commands in python
20357 @cindex python commands
20358 You can implement new @value{GDBN} CLI commands in Python. A CLI
20359 command is implemented using an instance of the @code{gdb.Command}
20360 class, most commonly using a subclass.
20362 @defmethod Command __init__ name @var{command_class} @r{[}@var{completer_class}@r{]} @r{[}@var{prefix}@r{]}
20363 The object initializer for @code{Command} registers the new command
20364 with @value{GDBN}. This initializer is normally invoked from the
20365 subclass' own @code{__init__} method.
20367 @var{name} is the name of the command. If @var{name} consists of
20368 multiple words, then the initial words are looked for as prefix
20369 commands. In this case, if one of the prefix commands does not exist,
20370 an exception is raised.
20372 There is no support for multi-line commands.
20374 @var{command_class} should be one of the @samp{COMMAND_} constants
20375 defined below. This argument tells @value{GDBN} how to categorize the
20376 new command in the help system.
20378 @var{completer_class} is an optional argument. If given, it should be
20379 one of the @samp{COMPLETE_} constants defined below. This argument
20380 tells @value{GDBN} how to perform completion for this command. If not
20381 given, @value{GDBN} will attempt to complete using the object's
20382 @code{complete} method (see below); if no such method is found, an
20383 error will occur when completion is attempted.
20385 @var{prefix} is an optional argument. If @code{True}, then the new
20386 command is a prefix command; sub-commands of this command may be
20389 The help text for the new command is taken from the Python
20390 documentation string for the command's class, if there is one. If no
20391 documentation string is provided, the default value ``This command is
20392 not documented.'' is used.
20395 @cindex don't repeat Python command
20396 @defmethod Command dont_repeat
20397 By default, a @value{GDBN} command is repeated when the user enters a
20398 blank line at the command prompt. A command can suppress this
20399 behavior by invoking the @code{dont_repeat} method. This is similar
20400 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
20403 @defmethod Command invoke argument from_tty
20404 This method is called by @value{GDBN} when this command is invoked.
20406 @var{argument} is a string. It is the argument to the command, after
20407 leading and trailing whitespace has been stripped.
20409 @var{from_tty} is a boolean argument. When true, this means that the
20410 command was entered by the user at the terminal; when false it means
20411 that the command came from elsewhere.
20413 If this method throws an exception, it is turned into a @value{GDBN}
20414 @code{error} call. Otherwise, the return value is ignored.
20417 @cindex completion of Python commands
20418 @defmethod Command complete text word
20419 This method is called by @value{GDBN} when the user attempts
20420 completion on this command. All forms of completion are handled by
20421 this method, that is, the @key{TAB} and @key{M-?} key bindings
20422 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
20425 The arguments @var{text} and @var{word} are both strings. @var{text}
20426 holds the complete command line up to the cursor's location.
20427 @var{word} holds the last word of the command line; this is computed
20428 using a word-breaking heuristic.
20430 The @code{complete} method can return several values:
20433 If the return value is a sequence, the contents of the sequence are
20434 used as the completions. It is up to @code{complete} to ensure that the
20435 contents actually do complete the word. A zero-length sequence is
20436 allowed, it means that there were no completions available. Only
20437 string elements of the sequence are used; other elements in the
20438 sequence are ignored.
20441 If the return value is one of the @samp{COMPLETE_} constants defined
20442 below, then the corresponding @value{GDBN}-internal completion
20443 function is invoked, and its result is used.
20446 All other results are treated as though there were no available
20451 When a new command is registered, it must be declared as a member of
20452 some general class of commands. This is used to classify top-level
20453 commands in the on-line help system; note that prefix commands are not
20454 listed under their own category but rather that of their top-level
20455 command. The available classifications are represented by constants
20456 defined in the @code{gdb} module:
20459 @findex COMMAND_NONE
20460 @findex gdb.COMMAND_NONE
20462 The command does not belong to any particular class. A command in
20463 this category will not be displayed in any of the help categories.
20465 @findex COMMAND_RUNNING
20466 @findex gdb.COMMAND_RUNNING
20467 @item COMMAND_RUNNING
20468 The command is related to running the inferior. For example,
20469 @code{start}, @code{step}, and @code{continue} are in this category.
20470 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
20471 commands in this category.
20473 @findex COMMAND_DATA
20474 @findex gdb.COMMAND_DATA
20476 The command is related to data or variables. For example,
20477 @code{call}, @code{find}, and @code{print} are in this category. Type
20478 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
20481 @findex COMMAND_STACK
20482 @findex gdb.COMMAND_STACK
20483 @item COMMAND_STACK
20484 The command has to do with manipulation of the stack. For example,
20485 @code{backtrace}, @code{frame}, and @code{return} are in this
20486 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
20487 list of commands in this category.
20489 @findex COMMAND_FILES
20490 @findex gdb.COMMAND_FILES
20491 @item COMMAND_FILES
20492 This class is used for file-related commands. For example,
20493 @code{file}, @code{list} and @code{section} are in this category.
20494 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
20495 commands in this category.
20497 @findex COMMAND_SUPPORT
20498 @findex gdb.COMMAND_SUPPORT
20499 @item COMMAND_SUPPORT
20500 This should be used for ``support facilities'', generally meaning
20501 things that are useful to the user when interacting with @value{GDBN},
20502 but not related to the state of the inferior. For example,
20503 @code{help}, @code{make}, and @code{shell} are in this category. Type
20504 @kbd{help support} at the @value{GDBN} prompt to see a list of
20505 commands in this category.
20507 @findex COMMAND_STATUS
20508 @findex gdb.COMMAND_STATUS
20509 @item COMMAND_STATUS
20510 The command is an @samp{info}-related command, that is, related to the
20511 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
20512 and @code{show} are in this category. Type @kbd{help status} at the
20513 @value{GDBN} prompt to see a list of commands in this category.
20515 @findex COMMAND_BREAKPOINTS
20516 @findex gdb.COMMAND_BREAKPOINTS
20517 @item COMMAND_BREAKPOINTS
20518 The command has to do with breakpoints. For example, @code{break},
20519 @code{clear}, and @code{delete} are in this category. Type @kbd{help
20520 breakpoints} at the @value{GDBN} prompt to see a list of commands in
20523 @findex COMMAND_TRACEPOINTS
20524 @findex gdb.COMMAND_TRACEPOINTS
20525 @item COMMAND_TRACEPOINTS
20526 The command has to do with tracepoints. For example, @code{trace},
20527 @code{actions}, and @code{tfind} are in this category. Type
20528 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
20529 commands in this category.
20531 @findex COMMAND_OBSCURE
20532 @findex gdb.COMMAND_OBSCURE
20533 @item COMMAND_OBSCURE
20534 The command is only used in unusual circumstances, or is not of
20535 general interest to users. For example, @code{checkpoint},
20536 @code{fork}, and @code{stop} are in this category. Type @kbd{help
20537 obscure} at the @value{GDBN} prompt to see a list of commands in this
20540 @findex COMMAND_MAINTENANCE
20541 @findex gdb.COMMAND_MAINTENANCE
20542 @item COMMAND_MAINTENANCE
20543 The command is only useful to @value{GDBN} maintainers. The
20544 @code{maintenance} and @code{flushregs} commands are in this category.
20545 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
20546 commands in this category.
20549 A new command can use a predefined completion function, either by
20550 specifying it via an argument at initialization, or by returning it
20551 from the @code{complete} method. These predefined completion
20552 constants are all defined in the @code{gdb} module:
20555 @findex COMPLETE_NONE
20556 @findex gdb.COMPLETE_NONE
20557 @item COMPLETE_NONE
20558 This constant means that no completion should be done.
20560 @findex COMPLETE_FILENAME
20561 @findex gdb.COMPLETE_FILENAME
20562 @item COMPLETE_FILENAME
20563 This constant means that filename completion should be performed.
20565 @findex COMPLETE_LOCATION
20566 @findex gdb.COMPLETE_LOCATION
20567 @item COMPLETE_LOCATION
20568 This constant means that location completion should be done.
20569 @xref{Specify Location}.
20571 @findex COMPLETE_COMMAND
20572 @findex gdb.COMPLETE_COMMAND
20573 @item COMPLETE_COMMAND
20574 This constant means that completion should examine @value{GDBN}
20577 @findex COMPLETE_SYMBOL
20578 @findex gdb.COMPLETE_SYMBOL
20579 @item COMPLETE_SYMBOL
20580 This constant means that completion should be done using symbol names
20584 The following code snippet shows how a trivial CLI command can be
20585 implemented in Python:
20588 class HelloWorld (gdb.Command):
20589 """Greet the whole world."""
20591 def __init__ (self):
20592 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
20594 def invoke (self, arg, from_tty):
20595 print "Hello, World!"
20600 The last line instantiates the class, and is necessary to trigger the
20601 registration of the command with @value{GDBN}. Depending on how the
20602 Python code is read into @value{GDBN}, you may need to import the
20603 @code{gdb} module explicitly.
20605 @node Functions In Python
20606 @subsubsection Writing new convenience functions
20608 @cindex writing convenience functions
20609 @cindex convenience functions in python
20610 @cindex python convenience functions
20611 @tindex gdb.Function
20613 You can implement new convenience functions (@pxref{Convenience Vars})
20614 in Python. A convenience function is an instance of a subclass of the
20615 class @code{gdb.Function}.
20617 @defmethod Function __init__ name
20618 The initializer for @code{Function} registers the new function with
20619 @value{GDBN}. The argument @var{name} is the name of the function,
20620 a string. The function will be visible to the user as a convenience
20621 variable of type @code{internal function}, whose name is the same as
20622 the given @var{name}.
20624 The documentation for the new function is taken from the documentation
20625 string for the new class.
20628 @defmethod Function invoke @var{*args}
20629 When a convenience function is evaluated, its arguments are converted
20630 to instances of @code{gdb.Value}, and then the function's
20631 @code{invoke} method is called. Note that @value{GDBN} does not
20632 predetermine the arity of convenience functions. Instead, all
20633 available arguments are passed to @code{invoke}, following the
20634 standard Python calling convention. In particular, a convenience
20635 function can have default values for parameters without ill effect.
20637 The return value of this method is used as its value in the enclosing
20638 expression. If an ordinary Python value is returned, it is converted
20639 to a @code{gdb.Value} following the usual rules.
20642 The following code snippet shows how a trivial convenience function can
20643 be implemented in Python:
20646 class Greet (gdb.Function):
20647 """Return string to greet someone.
20648 Takes a name as argument."""
20650 def __init__ (self):
20651 super (Greet, self).__init__ ("greet")
20653 def invoke (self, name):
20654 return "Hello, %s!" % name.string ()
20659 The last line instantiates the class, and is necessary to trigger the
20660 registration of the function with @value{GDBN}. Depending on how the
20661 Python code is read into @value{GDBN}, you may need to import the
20662 @code{gdb} module explicitly.
20664 @node Objfiles In Python
20665 @subsubsection Objfiles In Python
20667 @cindex objfiles in python
20668 @tindex gdb.Objfile
20670 @value{GDBN} loads symbols for an inferior from various
20671 symbol-containing files (@pxref{Files}). These include the primary
20672 executable file, any shared libraries used by the inferior, and any
20673 separate debug info files (@pxref{Separate Debug Files}).
20674 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
20676 The following objfile-related functions are available in the
20679 @findex gdb.current_objfile
20680 @defun current_objfile
20681 When auto-loading a Python script (@pxref{Auto-loading}), @value{GDBN}
20682 sets the ``current objfile'' to the corresponding objfile. This
20683 function returns the current objfile. If there is no current objfile,
20684 this function returns @code{None}.
20687 @findex gdb.objfiles
20689 Return a sequence of all the objfiles current known to @value{GDBN}.
20690 @xref{Objfiles In Python}.
20693 Each objfile is represented by an instance of the @code{gdb.Objfile}
20696 @defivar Objfile filename
20697 The file name of the objfile as a string.
20700 @defivar Objfile pretty_printers
20701 The @code{pretty_printers} attribute is a list of functions. It is
20702 used to look up pretty-printers. A @code{Value} is passed to each
20703 function in order; if the function returns @code{None}, then the
20704 search continues. Otherwise, the return value should be an object
20705 which is used to format the value. @xref{Pretty Printing}, for more
20709 @node Frames In Python
20710 @subsubsection Acessing inferior stack frames from Python.
20712 @cindex frames in python
20713 When the debugged program stops, @value{GDBN} is able to analyze its call
20714 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
20715 represents a frame in the stack. A @code{gdb.Frame} object is only valid
20716 while its corresponding frame exists in the inferior's stack. If you try
20717 to use an invalid frame object, @value{GDBN} will throw a @code{RuntimeError}
20720 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
20724 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
20728 The following frame-related functions are available in the @code{gdb} module:
20730 @findex gdb.selected_frame
20731 @defun selected_frame
20732 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
20735 @defun frame_stop_reason_string reason
20736 Return a string explaining the reason why @value{GDBN} stopped unwinding
20737 frames, as expressed by the given @var{reason} code (an integer, see the
20738 @code{unwind_stop_reason} method further down in this section).
20741 A @code{gdb.Frame} object has the following methods:
20744 @defmethod Frame is_valid
20745 Returns true if the @code{gdb.Frame} object is valid, false if not.
20746 A frame object can become invalid if the frame it refers to doesn't
20747 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
20748 an exception if it is invalid at the time the method is called.
20751 @defmethod Frame name
20752 Returns the function name of the frame, or @code{None} if it can't be
20756 @defmethod Frame type
20757 Returns the type of the frame. The value can be one of
20758 @code{gdb.NORMAL_FRAME}, @code{gdb.DUMMY_FRAME}, @code{gdb.SIGTRAMP_FRAME}
20759 or @code{gdb.SENTINEL_FRAME}.
20762 @defmethod Frame unwind_stop_reason
20763 Return an integer representing the reason why it's not possible to find
20764 more frames toward the outermost frame. Use
20765 @code{gdb.frame_stop_reason_string} to convert the value returned by this
20766 function to a string.
20769 @defmethod Frame pc
20770 Returns the frame's resume address.
20773 @defmethod Frame older
20774 Return the frame that called this frame.
20777 @defmethod Frame newer
20778 Return the frame called by this frame.
20781 @defmethod Frame read_var variable
20782 Return the value of the given variable in this frame. @var{variable} must
20787 @node Lazy Strings In Python
20788 @subsubsection Python representation of lazy strings.
20790 @cindex lazy strings in python
20791 @tindex gdb.LazyString
20793 A @dfn{lazy string} is a string whose contents is not retrieved or
20794 encoded until it is needed.
20796 A @code{gdb.LazyString} is represented in @value{GDBN} as an
20797 @code{address} that points to a region of memory, an @code{encoding}
20798 that will be used to encode that region of memory, and a @code{length}
20799 to delimit the region of memory that represents the string. The
20800 difference between a @code{gdb.LazyString} and a string wrapped within
20801 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
20802 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
20803 retrieved and encoded during printing, while a @code{gdb.Value}
20804 wrapping a string is immediately retrieved and encoded on creation.
20806 A @code{gdb.LazyString} object has the following functions:
20808 @defmethod LazyString value
20809 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
20810 will point to the string in memory, but will lose all the delayed
20811 retrieval, encoding and handling that @value{GDBN} applies to a
20812 @code{gdb.LazyString}.
20815 @defivar LazyString address
20816 This attribute holds the address of the string. This attribute is not
20820 @defivar LazyString length
20821 This attribute holds the length of the string in characters. If the
20822 length is -1, then the string will be fetched and encoded up to the
20823 first null of appropriate width. This attribute is not writable.
20826 @defivar LazyString encoding
20827 This attribute holds the encoding that will be applied to the string
20828 when the string is printed by @value{GDBN}. If the encoding is not
20829 set, or contains an empty string, then @value{GDBN} will select the
20830 most appropriate encoding when the string is printed. This attribute
20834 @defivar LazyString type
20835 This attribute holds the type that is represented by the lazy string's
20836 type. For a lazy string this will always be a pointer type. To
20837 resolve this to the lazy string's character type, use the type's
20838 @code{target} method. @xref{Types In Python}. This attribute is not
20843 @chapter Command Interpreters
20844 @cindex command interpreters
20846 @value{GDBN} supports multiple command interpreters, and some command
20847 infrastructure to allow users or user interface writers to switch
20848 between interpreters or run commands in other interpreters.
20850 @value{GDBN} currently supports two command interpreters, the console
20851 interpreter (sometimes called the command-line interpreter or @sc{cli})
20852 and the machine interface interpreter (or @sc{gdb/mi}). This manual
20853 describes both of these interfaces in great detail.
20855 By default, @value{GDBN} will start with the console interpreter.
20856 However, the user may choose to start @value{GDBN} with another
20857 interpreter by specifying the @option{-i} or @option{--interpreter}
20858 startup options. Defined interpreters include:
20862 @cindex console interpreter
20863 The traditional console or command-line interpreter. This is the most often
20864 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
20865 @value{GDBN} will use this interpreter.
20868 @cindex mi interpreter
20869 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
20870 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
20871 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
20875 @cindex mi2 interpreter
20876 The current @sc{gdb/mi} interface.
20879 @cindex mi1 interpreter
20880 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
20884 @cindex invoke another interpreter
20885 The interpreter being used by @value{GDBN} may not be dynamically
20886 switched at runtime. Although possible, this could lead to a very
20887 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
20888 enters the command "interpreter-set console" in a console view,
20889 @value{GDBN} would switch to using the console interpreter, rendering
20890 the IDE inoperable!
20892 @kindex interpreter-exec
20893 Although you may only choose a single interpreter at startup, you may execute
20894 commands in any interpreter from the current interpreter using the appropriate
20895 command. If you are running the console interpreter, simply use the
20896 @code{interpreter-exec} command:
20899 interpreter-exec mi "-data-list-register-names"
20902 @sc{gdb/mi} has a similar command, although it is only available in versions of
20903 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
20906 @chapter @value{GDBN} Text User Interface
20908 @cindex Text User Interface
20911 * TUI Overview:: TUI overview
20912 * TUI Keys:: TUI key bindings
20913 * TUI Single Key Mode:: TUI single key mode
20914 * TUI Commands:: TUI-specific commands
20915 * TUI Configuration:: TUI configuration variables
20918 The @value{GDBN} Text User Interface (TUI) is a terminal
20919 interface which uses the @code{curses} library to show the source
20920 file, the assembly output, the program registers and @value{GDBN}
20921 commands in separate text windows. The TUI mode is supported only
20922 on platforms where a suitable version of the @code{curses} library
20925 @pindex @value{GDBTUI}
20926 The TUI mode is enabled by default when you invoke @value{GDBN} as
20927 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
20928 You can also switch in and out of TUI mode while @value{GDBN} runs by
20929 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
20930 @xref{TUI Keys, ,TUI Key Bindings}.
20933 @section TUI Overview
20935 In TUI mode, @value{GDBN} can display several text windows:
20939 This window is the @value{GDBN} command window with the @value{GDBN}
20940 prompt and the @value{GDBN} output. The @value{GDBN} input is still
20941 managed using readline.
20944 The source window shows the source file of the program. The current
20945 line and active breakpoints are displayed in this window.
20948 The assembly window shows the disassembly output of the program.
20951 This window shows the processor registers. Registers are highlighted
20952 when their values change.
20955 The source and assembly windows show the current program position
20956 by highlighting the current line and marking it with a @samp{>} marker.
20957 Breakpoints are indicated with two markers. The first marker
20958 indicates the breakpoint type:
20962 Breakpoint which was hit at least once.
20965 Breakpoint which was never hit.
20968 Hardware breakpoint which was hit at least once.
20971 Hardware breakpoint which was never hit.
20974 The second marker indicates whether the breakpoint is enabled or not:
20978 Breakpoint is enabled.
20981 Breakpoint is disabled.
20984 The source, assembly and register windows are updated when the current
20985 thread changes, when the frame changes, or when the program counter
20988 These windows are not all visible at the same time. The command
20989 window is always visible. The others can be arranged in several
21000 source and assembly,
21003 source and registers, or
21006 assembly and registers.
21009 A status line above the command window shows the following information:
21013 Indicates the current @value{GDBN} target.
21014 (@pxref{Targets, ,Specifying a Debugging Target}).
21017 Gives the current process or thread number.
21018 When no process is being debugged, this field is set to @code{No process}.
21021 Gives the current function name for the selected frame.
21022 The name is demangled if demangling is turned on (@pxref{Print Settings}).
21023 When there is no symbol corresponding to the current program counter,
21024 the string @code{??} is displayed.
21027 Indicates the current line number for the selected frame.
21028 When the current line number is not known, the string @code{??} is displayed.
21031 Indicates the current program counter address.
21035 @section TUI Key Bindings
21036 @cindex TUI key bindings
21038 The TUI installs several key bindings in the readline keymaps
21039 (@pxref{Command Line Editing}). The following key bindings
21040 are installed for both TUI mode and the @value{GDBN} standard mode.
21049 Enter or leave the TUI mode. When leaving the TUI mode,
21050 the curses window management stops and @value{GDBN} operates using
21051 its standard mode, writing on the terminal directly. When reentering
21052 the TUI mode, control is given back to the curses windows.
21053 The screen is then refreshed.
21057 Use a TUI layout with only one window. The layout will
21058 either be @samp{source} or @samp{assembly}. When the TUI mode
21059 is not active, it will switch to the TUI mode.
21061 Think of this key binding as the Emacs @kbd{C-x 1} binding.
21065 Use a TUI layout with at least two windows. When the current
21066 layout already has two windows, the next layout with two windows is used.
21067 When a new layout is chosen, one window will always be common to the
21068 previous layout and the new one.
21070 Think of it as the Emacs @kbd{C-x 2} binding.
21074 Change the active window. The TUI associates several key bindings
21075 (like scrolling and arrow keys) with the active window. This command
21076 gives the focus to the next TUI window.
21078 Think of it as the Emacs @kbd{C-x o} binding.
21082 Switch in and out of the TUI SingleKey mode that binds single
21083 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
21086 The following key bindings only work in the TUI mode:
21091 Scroll the active window one page up.
21095 Scroll the active window one page down.
21099 Scroll the active window one line up.
21103 Scroll the active window one line down.
21107 Scroll the active window one column left.
21111 Scroll the active window one column right.
21115 Refresh the screen.
21118 Because the arrow keys scroll the active window in the TUI mode, they
21119 are not available for their normal use by readline unless the command
21120 window has the focus. When another window is active, you must use
21121 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
21122 and @kbd{C-f} to control the command window.
21124 @node TUI Single Key Mode
21125 @section TUI Single Key Mode
21126 @cindex TUI single key mode
21128 The TUI also provides a @dfn{SingleKey} mode, which binds several
21129 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
21130 switch into this mode, where the following key bindings are used:
21133 @kindex c @r{(SingleKey TUI key)}
21137 @kindex d @r{(SingleKey TUI key)}
21141 @kindex f @r{(SingleKey TUI key)}
21145 @kindex n @r{(SingleKey TUI key)}
21149 @kindex q @r{(SingleKey TUI key)}
21151 exit the SingleKey mode.
21153 @kindex r @r{(SingleKey TUI key)}
21157 @kindex s @r{(SingleKey TUI key)}
21161 @kindex u @r{(SingleKey TUI key)}
21165 @kindex v @r{(SingleKey TUI key)}
21169 @kindex w @r{(SingleKey TUI key)}
21174 Other keys temporarily switch to the @value{GDBN} command prompt.
21175 The key that was pressed is inserted in the editing buffer so that
21176 it is possible to type most @value{GDBN} commands without interaction
21177 with the TUI SingleKey mode. Once the command is entered the TUI
21178 SingleKey mode is restored. The only way to permanently leave
21179 this mode is by typing @kbd{q} or @kbd{C-x s}.
21183 @section TUI-specific Commands
21184 @cindex TUI commands
21186 The TUI has specific commands to control the text windows.
21187 These commands are always available, even when @value{GDBN} is not in
21188 the TUI mode. When @value{GDBN} is in the standard mode, most
21189 of these commands will automatically switch to the TUI mode.
21194 List and give the size of all displayed windows.
21198 Display the next layout.
21201 Display the previous layout.
21204 Display the source window only.
21207 Display the assembly window only.
21210 Display the source and assembly window.
21213 Display the register window together with the source or assembly window.
21217 Make the next window active for scrolling.
21220 Make the previous window active for scrolling.
21223 Make the source window active for scrolling.
21226 Make the assembly window active for scrolling.
21229 Make the register window active for scrolling.
21232 Make the command window active for scrolling.
21236 Refresh the screen. This is similar to typing @kbd{C-L}.
21238 @item tui reg float
21240 Show the floating point registers in the register window.
21242 @item tui reg general
21243 Show the general registers in the register window.
21246 Show the next register group. The list of register groups as well as
21247 their order is target specific. The predefined register groups are the
21248 following: @code{general}, @code{float}, @code{system}, @code{vector},
21249 @code{all}, @code{save}, @code{restore}.
21251 @item tui reg system
21252 Show the system registers in the register window.
21256 Update the source window and the current execution point.
21258 @item winheight @var{name} +@var{count}
21259 @itemx winheight @var{name} -@var{count}
21261 Change the height of the window @var{name} by @var{count}
21262 lines. Positive counts increase the height, while negative counts
21265 @item tabset @var{nchars}
21267 Set the width of tab stops to be @var{nchars} characters.
21270 @node TUI Configuration
21271 @section TUI Configuration Variables
21272 @cindex TUI configuration variables
21274 Several configuration variables control the appearance of TUI windows.
21277 @item set tui border-kind @var{kind}
21278 @kindex set tui border-kind
21279 Select the border appearance for the source, assembly and register windows.
21280 The possible values are the following:
21283 Use a space character to draw the border.
21286 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
21289 Use the Alternate Character Set to draw the border. The border is
21290 drawn using character line graphics if the terminal supports them.
21293 @item set tui border-mode @var{mode}
21294 @kindex set tui border-mode
21295 @itemx set tui active-border-mode @var{mode}
21296 @kindex set tui active-border-mode
21297 Select the display attributes for the borders of the inactive windows
21298 or the active window. The @var{mode} can be one of the following:
21301 Use normal attributes to display the border.
21307 Use reverse video mode.
21310 Use half bright mode.
21312 @item half-standout
21313 Use half bright and standout mode.
21316 Use extra bright or bold mode.
21318 @item bold-standout
21319 Use extra bright or bold and standout mode.
21324 @chapter Using @value{GDBN} under @sc{gnu} Emacs
21327 @cindex @sc{gnu} Emacs
21328 A special interface allows you to use @sc{gnu} Emacs to view (and
21329 edit) the source files for the program you are debugging with
21332 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
21333 executable file you want to debug as an argument. This command starts
21334 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
21335 created Emacs buffer.
21336 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
21338 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
21343 All ``terminal'' input and output goes through an Emacs buffer, called
21346 This applies both to @value{GDBN} commands and their output, and to the input
21347 and output done by the program you are debugging.
21349 This is useful because it means that you can copy the text of previous
21350 commands and input them again; you can even use parts of the output
21353 All the facilities of Emacs' Shell mode are available for interacting
21354 with your program. In particular, you can send signals the usual
21355 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
21359 @value{GDBN} displays source code through Emacs.
21361 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
21362 source file for that frame and puts an arrow (@samp{=>}) at the
21363 left margin of the current line. Emacs uses a separate buffer for
21364 source display, and splits the screen to show both your @value{GDBN} session
21367 Explicit @value{GDBN} @code{list} or search commands still produce output as
21368 usual, but you probably have no reason to use them from Emacs.
21371 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
21372 a graphical mode, enabled by default, which provides further buffers
21373 that can control the execution and describe the state of your program.
21374 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
21376 If you specify an absolute file name when prompted for the @kbd{M-x
21377 gdb} argument, then Emacs sets your current working directory to where
21378 your program resides. If you only specify the file name, then Emacs
21379 sets your current working directory to to the directory associated
21380 with the previous buffer. In this case, @value{GDBN} may find your
21381 program by searching your environment's @code{PATH} variable, but on
21382 some operating systems it might not find the source. So, although the
21383 @value{GDBN} input and output session proceeds normally, the auxiliary
21384 buffer does not display the current source and line of execution.
21386 The initial working directory of @value{GDBN} is printed on the top
21387 line of the GUD buffer and this serves as a default for the commands
21388 that specify files for @value{GDBN} to operate on. @xref{Files,
21389 ,Commands to Specify Files}.
21391 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
21392 need to call @value{GDBN} by a different name (for example, if you
21393 keep several configurations around, with different names) you can
21394 customize the Emacs variable @code{gud-gdb-command-name} to run the
21397 In the GUD buffer, you can use these special Emacs commands in
21398 addition to the standard Shell mode commands:
21402 Describe the features of Emacs' GUD Mode.
21405 Execute to another source line, like the @value{GDBN} @code{step} command; also
21406 update the display window to show the current file and location.
21409 Execute to next source line in this function, skipping all function
21410 calls, like the @value{GDBN} @code{next} command. Then update the display window
21411 to show the current file and location.
21414 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
21415 display window accordingly.
21418 Execute until exit from the selected stack frame, like the @value{GDBN}
21419 @code{finish} command.
21422 Continue execution of your program, like the @value{GDBN} @code{continue}
21426 Go up the number of frames indicated by the numeric argument
21427 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
21428 like the @value{GDBN} @code{up} command.
21431 Go down the number of frames indicated by the numeric argument, like the
21432 @value{GDBN} @code{down} command.
21435 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
21436 tells @value{GDBN} to set a breakpoint on the source line point is on.
21438 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
21439 separate frame which shows a backtrace when the GUD buffer is current.
21440 Move point to any frame in the stack and type @key{RET} to make it
21441 become the current frame and display the associated source in the
21442 source buffer. Alternatively, click @kbd{Mouse-2} to make the
21443 selected frame become the current one. In graphical mode, the
21444 speedbar displays watch expressions.
21446 If you accidentally delete the source-display buffer, an easy way to get
21447 it back is to type the command @code{f} in the @value{GDBN} buffer, to
21448 request a frame display; when you run under Emacs, this recreates
21449 the source buffer if necessary to show you the context of the current
21452 The source files displayed in Emacs are in ordinary Emacs buffers
21453 which are visiting the source files in the usual way. You can edit
21454 the files with these buffers if you wish; but keep in mind that @value{GDBN}
21455 communicates with Emacs in terms of line numbers. If you add or
21456 delete lines from the text, the line numbers that @value{GDBN} knows cease
21457 to correspond properly with the code.
21459 A more detailed description of Emacs' interaction with @value{GDBN} is
21460 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
21463 @c The following dropped because Epoch is nonstandard. Reactivate
21464 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
21466 @kindex Emacs Epoch environment
21470 Version 18 of @sc{gnu} Emacs has a built-in window system
21471 called the @code{epoch}
21472 environment. Users of this environment can use a new command,
21473 @code{inspect} which performs identically to @code{print} except that
21474 each value is printed in its own window.
21479 @chapter The @sc{gdb/mi} Interface
21481 @unnumberedsec Function and Purpose
21483 @cindex @sc{gdb/mi}, its purpose
21484 @sc{gdb/mi} is a line based machine oriented text interface to
21485 @value{GDBN} and is activated by specifying using the
21486 @option{--interpreter} command line option (@pxref{Mode Options}). It
21487 is specifically intended to support the development of systems which
21488 use the debugger as just one small component of a larger system.
21490 This chapter is a specification of the @sc{gdb/mi} interface. It is written
21491 in the form of a reference manual.
21493 Note that @sc{gdb/mi} is still under construction, so some of the
21494 features described below are incomplete and subject to change
21495 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
21497 @unnumberedsec Notation and Terminology
21499 @cindex notational conventions, for @sc{gdb/mi}
21500 This chapter uses the following notation:
21504 @code{|} separates two alternatives.
21507 @code{[ @var{something} ]} indicates that @var{something} is optional:
21508 it may or may not be given.
21511 @code{( @var{group} )*} means that @var{group} inside the parentheses
21512 may repeat zero or more times.
21515 @code{( @var{group} )+} means that @var{group} inside the parentheses
21516 may repeat one or more times.
21519 @code{"@var{string}"} means a literal @var{string}.
21523 @heading Dependencies
21527 * GDB/MI General Design::
21528 * GDB/MI Command Syntax::
21529 * GDB/MI Compatibility with CLI::
21530 * GDB/MI Development and Front Ends::
21531 * GDB/MI Output Records::
21532 * GDB/MI Simple Examples::
21533 * GDB/MI Command Description Format::
21534 * GDB/MI Breakpoint Commands::
21535 * GDB/MI Program Context::
21536 * GDB/MI Thread Commands::
21537 * GDB/MI Program Execution::
21538 * GDB/MI Stack Manipulation::
21539 * GDB/MI Variable Objects::
21540 * GDB/MI Data Manipulation::
21541 * GDB/MI Tracepoint Commands::
21542 * GDB/MI Symbol Query::
21543 * GDB/MI File Commands::
21545 * GDB/MI Kod Commands::
21546 * GDB/MI Memory Overlay Commands::
21547 * GDB/MI Signal Handling Commands::
21549 * GDB/MI Target Manipulation::
21550 * GDB/MI File Transfer Commands::
21551 * GDB/MI Miscellaneous Commands::
21554 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21555 @node GDB/MI General Design
21556 @section @sc{gdb/mi} General Design
21557 @cindex GDB/MI General Design
21559 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
21560 parts---commands sent to @value{GDBN}, responses to those commands
21561 and notifications. Each command results in exactly one response,
21562 indicating either successful completion of the command, or an error.
21563 For the commands that do not resume the target, the response contains the
21564 requested information. For the commands that resume the target, the
21565 response only indicates whether the target was successfully resumed.
21566 Notifications is the mechanism for reporting changes in the state of the
21567 target, or in @value{GDBN} state, that cannot conveniently be associated with
21568 a command and reported as part of that command response.
21570 The important examples of notifications are:
21574 Exec notifications. These are used to report changes in
21575 target state---when a target is resumed, or stopped. It would not
21576 be feasible to include this information in response of resuming
21577 commands, because one resume commands can result in multiple events in
21578 different threads. Also, quite some time may pass before any event
21579 happens in the target, while a frontend needs to know whether the resuming
21580 command itself was successfully executed.
21583 Console output, and status notifications. Console output
21584 notifications are used to report output of CLI commands, as well as
21585 diagnostics for other commands. Status notifications are used to
21586 report the progress of a long-running operation. Naturally, including
21587 this information in command response would mean no output is produced
21588 until the command is finished, which is undesirable.
21591 General notifications. Commands may have various side effects on
21592 the @value{GDBN} or target state beyond their official purpose. For example,
21593 a command may change the selected thread. Although such changes can
21594 be included in command response, using notification allows for more
21595 orthogonal frontend design.
21599 There's no guarantee that whenever an MI command reports an error,
21600 @value{GDBN} or the target are in any specific state, and especially,
21601 the state is not reverted to the state before the MI command was
21602 processed. Therefore, whenever an MI command results in an error,
21603 we recommend that the frontend refreshes all the information shown in
21604 the user interface.
21608 * Context management::
21609 * Asynchronous and non-stop modes::
21613 @node Context management
21614 @subsection Context management
21616 In most cases when @value{GDBN} accesses the target, this access is
21617 done in context of a specific thread and frame (@pxref{Frames}).
21618 Often, even when accessing global data, the target requires that a thread
21619 be specified. The CLI interface maintains the selected thread and frame,
21620 and supplies them to target on each command. This is convenient,
21621 because a command line user would not want to specify that information
21622 explicitly on each command, and because user interacts with
21623 @value{GDBN} via a single terminal, so no confusion is possible as
21624 to what thread and frame are the current ones.
21626 In the case of MI, the concept of selected thread and frame is less
21627 useful. First, a frontend can easily remember this information
21628 itself. Second, a graphical frontend can have more than one window,
21629 each one used for debugging a different thread, and the frontend might
21630 want to access additional threads for internal purposes. This
21631 increases the risk that by relying on implicitly selected thread, the
21632 frontend may be operating on a wrong one. Therefore, each MI command
21633 should explicitly specify which thread and frame to operate on. To
21634 make it possible, each MI command accepts the @samp{--thread} and
21635 @samp{--frame} options, the value to each is @value{GDBN} identifier
21636 for thread and frame to operate on.
21638 Usually, each top-level window in a frontend allows the user to select
21639 a thread and a frame, and remembers the user selection for further
21640 operations. However, in some cases @value{GDBN} may suggest that the
21641 current thread be changed. For example, when stopping on a breakpoint
21642 it is reasonable to switch to the thread where breakpoint is hit. For
21643 another example, if the user issues the CLI @samp{thread} command via
21644 the frontend, it is desirable to change the frontend's selected thread to the
21645 one specified by user. @value{GDBN} communicates the suggestion to
21646 change current thread using the @samp{=thread-selected} notification.
21647 No such notification is available for the selected frame at the moment.
21649 Note that historically, MI shares the selected thread with CLI, so
21650 frontends used the @code{-thread-select} to execute commands in the
21651 right context. However, getting this to work right is cumbersome. The
21652 simplest way is for frontend to emit @code{-thread-select} command
21653 before every command. This doubles the number of commands that need
21654 to be sent. The alternative approach is to suppress @code{-thread-select}
21655 if the selected thread in @value{GDBN} is supposed to be identical to the
21656 thread the frontend wants to operate on. However, getting this
21657 optimization right can be tricky. In particular, if the frontend
21658 sends several commands to @value{GDBN}, and one of the commands changes the
21659 selected thread, then the behaviour of subsequent commands will
21660 change. So, a frontend should either wait for response from such
21661 problematic commands, or explicitly add @code{-thread-select} for
21662 all subsequent commands. No frontend is known to do this exactly
21663 right, so it is suggested to just always pass the @samp{--thread} and
21664 @samp{--frame} options.
21666 @node Asynchronous and non-stop modes
21667 @subsection Asynchronous command execution and non-stop mode
21669 On some targets, @value{GDBN} is capable of processing MI commands
21670 even while the target is running. This is called @dfn{asynchronous
21671 command execution} (@pxref{Background Execution}). The frontend may
21672 specify a preferrence for asynchronous execution using the
21673 @code{-gdb-set target-async 1} command, which should be emitted before
21674 either running the executable or attaching to the target. After the
21675 frontend has started the executable or attached to the target, it can
21676 find if asynchronous execution is enabled using the
21677 @code{-list-target-features} command.
21679 Even if @value{GDBN} can accept a command while target is running,
21680 many commands that access the target do not work when the target is
21681 running. Therefore, asynchronous command execution is most useful
21682 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
21683 it is possible to examine the state of one thread, while other threads
21686 When a given thread is running, MI commands that try to access the
21687 target in the context of that thread may not work, or may work only on
21688 some targets. In particular, commands that try to operate on thread's
21689 stack will not work, on any target. Commands that read memory, or
21690 modify breakpoints, may work or not work, depending on the target. Note
21691 that even commands that operate on global state, such as @code{print},
21692 @code{set}, and breakpoint commands, still access the target in the
21693 context of a specific thread, so frontend should try to find a
21694 stopped thread and perform the operation on that thread (using the
21695 @samp{--thread} option).
21697 Which commands will work in the context of a running thread is
21698 highly target dependent. However, the two commands
21699 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
21700 to find the state of a thread, will always work.
21702 @node Thread groups
21703 @subsection Thread groups
21704 @value{GDBN} may be used to debug several processes at the same time.
21705 On some platfroms, @value{GDBN} may support debugging of several
21706 hardware systems, each one having several cores with several different
21707 processes running on each core. This section describes the MI
21708 mechanism to support such debugging scenarios.
21710 The key observation is that regardless of the structure of the
21711 target, MI can have a global list of threads, because most commands that
21712 accept the @samp{--thread} option do not need to know what process that
21713 thread belongs to. Therefore, it is not necessary to introduce
21714 neither additional @samp{--process} option, nor an notion of the
21715 current process in the MI interface. The only strictly new feature
21716 that is required is the ability to find how the threads are grouped
21719 To allow the user to discover such grouping, and to support arbitrary
21720 hierarchy of machines/cores/processes, MI introduces the concept of a
21721 @dfn{thread group}. Thread group is a collection of threads and other
21722 thread groups. A thread group always has a string identifier, a type,
21723 and may have additional attributes specific to the type. A new
21724 command, @code{-list-thread-groups}, returns the list of top-level
21725 thread groups, which correspond to processes that @value{GDBN} is
21726 debugging at the moment. By passing an identifier of a thread group
21727 to the @code{-list-thread-groups} command, it is possible to obtain
21728 the members of specific thread group.
21730 To allow the user to easily discover processes, and other objects, he
21731 wishes to debug, a concept of @dfn{available thread group} is
21732 introduced. Available thread group is an thread group that
21733 @value{GDBN} is not debugging, but that can be attached to, using the
21734 @code{-target-attach} command. The list of available top-level thread
21735 groups can be obtained using @samp{-list-thread-groups --available}.
21736 In general, the content of a thread group may be only retrieved only
21737 after attaching to that thread group.
21739 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21740 @node GDB/MI Command Syntax
21741 @section @sc{gdb/mi} Command Syntax
21744 * GDB/MI Input Syntax::
21745 * GDB/MI Output Syntax::
21748 @node GDB/MI Input Syntax
21749 @subsection @sc{gdb/mi} Input Syntax
21751 @cindex input syntax for @sc{gdb/mi}
21752 @cindex @sc{gdb/mi}, input syntax
21754 @item @var{command} @expansion{}
21755 @code{@var{cli-command} | @var{mi-command}}
21757 @item @var{cli-command} @expansion{}
21758 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
21759 @var{cli-command} is any existing @value{GDBN} CLI command.
21761 @item @var{mi-command} @expansion{}
21762 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
21763 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
21765 @item @var{token} @expansion{}
21766 "any sequence of digits"
21768 @item @var{option} @expansion{}
21769 @code{"-" @var{parameter} [ " " @var{parameter} ]}
21771 @item @var{parameter} @expansion{}
21772 @code{@var{non-blank-sequence} | @var{c-string}}
21774 @item @var{operation} @expansion{}
21775 @emph{any of the operations described in this chapter}
21777 @item @var{non-blank-sequence} @expansion{}
21778 @emph{anything, provided it doesn't contain special characters such as
21779 "-", @var{nl}, """ and of course " "}
21781 @item @var{c-string} @expansion{}
21782 @code{""" @var{seven-bit-iso-c-string-content} """}
21784 @item @var{nl} @expansion{}
21793 The CLI commands are still handled by the @sc{mi} interpreter; their
21794 output is described below.
21797 The @code{@var{token}}, when present, is passed back when the command
21801 Some @sc{mi} commands accept optional arguments as part of the parameter
21802 list. Each option is identified by a leading @samp{-} (dash) and may be
21803 followed by an optional argument parameter. Options occur first in the
21804 parameter list and can be delimited from normal parameters using
21805 @samp{--} (this is useful when some parameters begin with a dash).
21812 We want easy access to the existing CLI syntax (for debugging).
21815 We want it to be easy to spot a @sc{mi} operation.
21818 @node GDB/MI Output Syntax
21819 @subsection @sc{gdb/mi} Output Syntax
21821 @cindex output syntax of @sc{gdb/mi}
21822 @cindex @sc{gdb/mi}, output syntax
21823 The output from @sc{gdb/mi} consists of zero or more out-of-band records
21824 followed, optionally, by a single result record. This result record
21825 is for the most recent command. The sequence of output records is
21826 terminated by @samp{(gdb)}.
21828 If an input command was prefixed with a @code{@var{token}} then the
21829 corresponding output for that command will also be prefixed by that same
21833 @item @var{output} @expansion{}
21834 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
21836 @item @var{result-record} @expansion{}
21837 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
21839 @item @var{out-of-band-record} @expansion{}
21840 @code{@var{async-record} | @var{stream-record}}
21842 @item @var{async-record} @expansion{}
21843 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
21845 @item @var{exec-async-output} @expansion{}
21846 @code{[ @var{token} ] "*" @var{async-output}}
21848 @item @var{status-async-output} @expansion{}
21849 @code{[ @var{token} ] "+" @var{async-output}}
21851 @item @var{notify-async-output} @expansion{}
21852 @code{[ @var{token} ] "=" @var{async-output}}
21854 @item @var{async-output} @expansion{}
21855 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
21857 @item @var{result-class} @expansion{}
21858 @code{"done" | "running" | "connected" | "error" | "exit"}
21860 @item @var{async-class} @expansion{}
21861 @code{"stopped" | @var{others}} (where @var{others} will be added
21862 depending on the needs---this is still in development).
21864 @item @var{result} @expansion{}
21865 @code{ @var{variable} "=" @var{value}}
21867 @item @var{variable} @expansion{}
21868 @code{ @var{string} }
21870 @item @var{value} @expansion{}
21871 @code{ @var{const} | @var{tuple} | @var{list} }
21873 @item @var{const} @expansion{}
21874 @code{@var{c-string}}
21876 @item @var{tuple} @expansion{}
21877 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
21879 @item @var{list} @expansion{}
21880 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
21881 @var{result} ( "," @var{result} )* "]" }
21883 @item @var{stream-record} @expansion{}
21884 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
21886 @item @var{console-stream-output} @expansion{}
21887 @code{"~" @var{c-string}}
21889 @item @var{target-stream-output} @expansion{}
21890 @code{"@@" @var{c-string}}
21892 @item @var{log-stream-output} @expansion{}
21893 @code{"&" @var{c-string}}
21895 @item @var{nl} @expansion{}
21898 @item @var{token} @expansion{}
21899 @emph{any sequence of digits}.
21907 All output sequences end in a single line containing a period.
21910 The @code{@var{token}} is from the corresponding request. Note that
21911 for all async output, while the token is allowed by the grammar and
21912 may be output by future versions of @value{GDBN} for select async
21913 output messages, it is generally omitted. Frontends should treat
21914 all async output as reporting general changes in the state of the
21915 target and there should be no need to associate async output to any
21919 @cindex status output in @sc{gdb/mi}
21920 @var{status-async-output} contains on-going status information about the
21921 progress of a slow operation. It can be discarded. All status output is
21922 prefixed by @samp{+}.
21925 @cindex async output in @sc{gdb/mi}
21926 @var{exec-async-output} contains asynchronous state change on the target
21927 (stopped, started, disappeared). All async output is prefixed by
21931 @cindex notify output in @sc{gdb/mi}
21932 @var{notify-async-output} contains supplementary information that the
21933 client should handle (e.g., a new breakpoint information). All notify
21934 output is prefixed by @samp{=}.
21937 @cindex console output in @sc{gdb/mi}
21938 @var{console-stream-output} is output that should be displayed as is in the
21939 console. It is the textual response to a CLI command. All the console
21940 output is prefixed by @samp{~}.
21943 @cindex target output in @sc{gdb/mi}
21944 @var{target-stream-output} is the output produced by the target program.
21945 All the target output is prefixed by @samp{@@}.
21948 @cindex log output in @sc{gdb/mi}
21949 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
21950 instance messages that should be displayed as part of an error log. All
21951 the log output is prefixed by @samp{&}.
21954 @cindex list output in @sc{gdb/mi}
21955 New @sc{gdb/mi} commands should only output @var{lists} containing
21961 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
21962 details about the various output records.
21964 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21965 @node GDB/MI Compatibility with CLI
21966 @section @sc{gdb/mi} Compatibility with CLI
21968 @cindex compatibility, @sc{gdb/mi} and CLI
21969 @cindex @sc{gdb/mi}, compatibility with CLI
21971 For the developers convenience CLI commands can be entered directly,
21972 but there may be some unexpected behaviour. For example, commands
21973 that query the user will behave as if the user replied yes, breakpoint
21974 command lists are not executed and some CLI commands, such as
21975 @code{if}, @code{when} and @code{define}, prompt for further input with
21976 @samp{>}, which is not valid MI output.
21978 This feature may be removed at some stage in the future and it is
21979 recommended that front ends use the @code{-interpreter-exec} command
21980 (@pxref{-interpreter-exec}).
21982 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21983 @node GDB/MI Development and Front Ends
21984 @section @sc{gdb/mi} Development and Front Ends
21985 @cindex @sc{gdb/mi} development
21987 The application which takes the MI output and presents the state of the
21988 program being debugged to the user is called a @dfn{front end}.
21990 Although @sc{gdb/mi} is still incomplete, it is currently being used
21991 by a variety of front ends to @value{GDBN}. This makes it difficult
21992 to introduce new functionality without breaking existing usage. This
21993 section tries to minimize the problems by describing how the protocol
21996 Some changes in MI need not break a carefully designed front end, and
21997 for these the MI version will remain unchanged. The following is a
21998 list of changes that may occur within one level, so front ends should
21999 parse MI output in a way that can handle them:
22003 New MI commands may be added.
22006 New fields may be added to the output of any MI command.
22009 The range of values for fields with specified values, e.g.,
22010 @code{in_scope} (@pxref{-var-update}) may be extended.
22012 @c The format of field's content e.g type prefix, may change so parse it
22013 @c at your own risk. Yes, in general?
22015 @c The order of fields may change? Shouldn't really matter but it might
22016 @c resolve inconsistencies.
22019 If the changes are likely to break front ends, the MI version level
22020 will be increased by one. This will allow the front end to parse the
22021 output according to the MI version. Apart from mi0, new versions of
22022 @value{GDBN} will not support old versions of MI and it will be the
22023 responsibility of the front end to work with the new one.
22025 @c Starting with mi3, add a new command -mi-version that prints the MI
22028 The best way to avoid unexpected changes in MI that might break your front
22029 end is to make your project known to @value{GDBN} developers and
22030 follow development on @email{gdb@@sourceware.org} and
22031 @email{gdb-patches@@sourceware.org}.
22032 @cindex mailing lists
22034 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22035 @node GDB/MI Output Records
22036 @section @sc{gdb/mi} Output Records
22039 * GDB/MI Result Records::
22040 * GDB/MI Stream Records::
22041 * GDB/MI Async Records::
22042 * GDB/MI Frame Information::
22043 * GDB/MI Thread Information::
22046 @node GDB/MI Result Records
22047 @subsection @sc{gdb/mi} Result Records
22049 @cindex result records in @sc{gdb/mi}
22050 @cindex @sc{gdb/mi}, result records
22051 In addition to a number of out-of-band notifications, the response to a
22052 @sc{gdb/mi} command includes one of the following result indications:
22056 @item "^done" [ "," @var{results} ]
22057 The synchronous operation was successful, @code{@var{results}} are the return
22062 This result record is equivalent to @samp{^done}. Historically, it
22063 was output instead of @samp{^done} if the command has resumed the
22064 target. This behaviour is maintained for backward compatibility, but
22065 all frontends should treat @samp{^done} and @samp{^running}
22066 identically and rely on the @samp{*running} output record to determine
22067 which threads are resumed.
22071 @value{GDBN} has connected to a remote target.
22073 @item "^error" "," @var{c-string}
22075 The operation failed. The @code{@var{c-string}} contains the corresponding
22080 @value{GDBN} has terminated.
22084 @node GDB/MI Stream Records
22085 @subsection @sc{gdb/mi} Stream Records
22087 @cindex @sc{gdb/mi}, stream records
22088 @cindex stream records in @sc{gdb/mi}
22089 @value{GDBN} internally maintains a number of output streams: the console, the
22090 target, and the log. The output intended for each of these streams is
22091 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
22093 Each stream record begins with a unique @dfn{prefix character} which
22094 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
22095 Syntax}). In addition to the prefix, each stream record contains a
22096 @code{@var{string-output}}. This is either raw text (with an implicit new
22097 line) or a quoted C string (which does not contain an implicit newline).
22100 @item "~" @var{string-output}
22101 The console output stream contains text that should be displayed in the
22102 CLI console window. It contains the textual responses to CLI commands.
22104 @item "@@" @var{string-output}
22105 The target output stream contains any textual output from the running
22106 target. This is only present when GDB's event loop is truly
22107 asynchronous, which is currently only the case for remote targets.
22109 @item "&" @var{string-output}
22110 The log stream contains debugging messages being produced by @value{GDBN}'s
22114 @node GDB/MI Async Records
22115 @subsection @sc{gdb/mi} Async Records
22117 @cindex async records in @sc{gdb/mi}
22118 @cindex @sc{gdb/mi}, async records
22119 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
22120 additional changes that have occurred. Those changes can either be a
22121 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
22122 target activity (e.g., target stopped).
22124 The following is the list of possible async records:
22128 @item *running,thread-id="@var{thread}"
22129 The target is now running. The @var{thread} field tells which
22130 specific thread is now running, and can be @samp{all} if all threads
22131 are running. The frontend should assume that no interaction with a
22132 running thread is possible after this notification is produced.
22133 The frontend should not assume that this notification is output
22134 only once for any command. @value{GDBN} may emit this notification
22135 several times, either for different threads, because it cannot resume
22136 all threads together, or even for a single thread, if the thread must
22137 be stepped though some code before letting it run freely.
22139 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
22140 The target has stopped. The @var{reason} field can have one of the
22144 @item breakpoint-hit
22145 A breakpoint was reached.
22146 @item watchpoint-trigger
22147 A watchpoint was triggered.
22148 @item read-watchpoint-trigger
22149 A read watchpoint was triggered.
22150 @item access-watchpoint-trigger
22151 An access watchpoint was triggered.
22152 @item function-finished
22153 An -exec-finish or similar CLI command was accomplished.
22154 @item location-reached
22155 An -exec-until or similar CLI command was accomplished.
22156 @item watchpoint-scope
22157 A watchpoint has gone out of scope.
22158 @item end-stepping-range
22159 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
22160 similar CLI command was accomplished.
22161 @item exited-signalled
22162 The inferior exited because of a signal.
22164 The inferior exited.
22165 @item exited-normally
22166 The inferior exited normally.
22167 @item signal-received
22168 A signal was received by the inferior.
22171 The @var{id} field identifies the thread that directly caused the stop
22172 -- for example by hitting a breakpoint. Depending on whether all-stop
22173 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
22174 stop all threads, or only the thread that directly triggered the stop.
22175 If all threads are stopped, the @var{stopped} field will have the
22176 value of @code{"all"}. Otherwise, the value of the @var{stopped}
22177 field will be a list of thread identifiers. Presently, this list will
22178 always include a single thread, but frontend should be prepared to see
22179 several threads in the list. The @var{core} field reports the
22180 processor core on which the stop event has happened. This field may be absent
22181 if such information is not available.
22183 @item =thread-group-created,id="@var{id}"
22184 @itemx =thread-group-exited,id="@var{id}"
22185 A thread thread group either was attached to, or has exited/detached
22186 from. The @var{id} field contains the @value{GDBN} identifier of the
22189 @item =thread-created,id="@var{id}",group-id="@var{gid}"
22190 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
22191 A thread either was created, or has exited. The @var{id} field
22192 contains the @value{GDBN} identifier of the thread. The @var{gid}
22193 field identifies the thread group this thread belongs to.
22195 @item =thread-selected,id="@var{id}"
22196 Informs that the selected thread was changed as result of the last
22197 command. This notification is not emitted as result of @code{-thread-select}
22198 command but is emitted whenever an MI command that is not documented
22199 to change the selected thread actually changes it. In particular,
22200 invoking, directly or indirectly (via user-defined command), the CLI
22201 @code{thread} command, will generate this notification.
22203 We suggest that in response to this notification, front ends
22204 highlight the selected thread and cause subsequent commands to apply to
22207 @item =library-loaded,...
22208 Reports that a new library file was loaded by the program. This
22209 notification has 4 fields---@var{id}, @var{target-name},
22210 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
22211 opaque identifier of the library. For remote debugging case,
22212 @var{target-name} and @var{host-name} fields give the name of the
22213 library file on the target, and on the host respectively. For native
22214 debugging, both those fields have the same value. The
22215 @var{symbols-loaded} field reports if the debug symbols for this
22216 library are loaded.
22218 @item =library-unloaded,...
22219 Reports that a library was unloaded by the program. This notification
22220 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
22221 the same meaning as for the @code{=library-loaded} notification
22225 @node GDB/MI Frame Information
22226 @subsection @sc{gdb/mi} Frame Information
22228 Response from many MI commands includes an information about stack
22229 frame. This information is a tuple that may have the following
22234 The level of the stack frame. The innermost frame has the level of
22235 zero. This field is always present.
22238 The name of the function corresponding to the frame. This field may
22239 be absent if @value{GDBN} is unable to determine the function name.
22242 The code address for the frame. This field is always present.
22245 The name of the source files that correspond to the frame's code
22246 address. This field may be absent.
22249 The source line corresponding to the frames' code address. This field
22253 The name of the binary file (either executable or shared library) the
22254 corresponds to the frame's code address. This field may be absent.
22258 @node GDB/MI Thread Information
22259 @subsection @sc{gdb/mi} Thread Information
22261 Whenever @value{GDBN} has to report an information about a thread, it
22262 uses a tuple with the following fields:
22266 The numeric id assigned to the thread by @value{GDBN}. This field is
22270 Target-specific string identifying the thread. This field is always present.
22273 Additional information about the thread provided by the target.
22274 It is supposed to be human-readable and not interpreted by the
22275 frontend. This field is optional.
22278 Either @samp{stopped} or @samp{running}, depending on whether the
22279 thread is presently running. This field is always present.
22282 The value of this field is an integer number of the processor core the
22283 thread was last seen on. This field is optional.
22287 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22288 @node GDB/MI Simple Examples
22289 @section Simple Examples of @sc{gdb/mi} Interaction
22290 @cindex @sc{gdb/mi}, simple examples
22292 This subsection presents several simple examples of interaction using
22293 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
22294 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
22295 the output received from @sc{gdb/mi}.
22297 Note the line breaks shown in the examples are here only for
22298 readability, they don't appear in the real output.
22300 @subheading Setting a Breakpoint
22302 Setting a breakpoint generates synchronous output which contains detailed
22303 information of the breakpoint.
22306 -> -break-insert main
22307 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
22308 enabled="y",addr="0x08048564",func="main",file="myprog.c",
22309 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
22313 @subheading Program Execution
22315 Program execution generates asynchronous records and MI gives the
22316 reason that execution stopped.
22322 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
22323 frame=@{addr="0x08048564",func="main",
22324 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
22325 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
22330 <- *stopped,reason="exited-normally"
22334 @subheading Quitting @value{GDBN}
22336 Quitting @value{GDBN} just prints the result class @samp{^exit}.
22344 Please note that @samp{^exit} is printed immediately, but it might
22345 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
22346 performs necessary cleanups, including killing programs being debugged
22347 or disconnecting from debug hardware, so the frontend should wait till
22348 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
22349 fails to exit in reasonable time.
22351 @subheading A Bad Command
22353 Here's what happens if you pass a non-existent command:
22357 <- ^error,msg="Undefined MI command: rubbish"
22362 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22363 @node GDB/MI Command Description Format
22364 @section @sc{gdb/mi} Command Description Format
22366 The remaining sections describe blocks of commands. Each block of
22367 commands is laid out in a fashion similar to this section.
22369 @subheading Motivation
22371 The motivation for this collection of commands.
22373 @subheading Introduction
22375 A brief introduction to this collection of commands as a whole.
22377 @subheading Commands
22379 For each command in the block, the following is described:
22381 @subsubheading Synopsis
22384 -command @var{args}@dots{}
22387 @subsubheading Result
22389 @subsubheading @value{GDBN} Command
22391 The corresponding @value{GDBN} CLI command(s), if any.
22393 @subsubheading Example
22395 Example(s) formatted for readability. Some of the described commands have
22396 not been implemented yet and these are labeled N.A.@: (not available).
22399 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22400 @node GDB/MI Breakpoint Commands
22401 @section @sc{gdb/mi} Breakpoint Commands
22403 @cindex breakpoint commands for @sc{gdb/mi}
22404 @cindex @sc{gdb/mi}, breakpoint commands
22405 This section documents @sc{gdb/mi} commands for manipulating
22408 @subheading The @code{-break-after} Command
22409 @findex -break-after
22411 @subsubheading Synopsis
22414 -break-after @var{number} @var{count}
22417 The breakpoint number @var{number} is not in effect until it has been
22418 hit @var{count} times. To see how this is reflected in the output of
22419 the @samp{-break-list} command, see the description of the
22420 @samp{-break-list} command below.
22422 @subsubheading @value{GDBN} Command
22424 The corresponding @value{GDBN} command is @samp{ignore}.
22426 @subsubheading Example
22431 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
22432 enabled="y",addr="0x000100d0",func="main",file="hello.c",
22433 fullname="/home/foo/hello.c",line="5",times="0"@}
22440 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22441 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22442 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22443 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22444 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22445 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22446 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22447 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22448 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
22449 line="5",times="0",ignore="3"@}]@}
22454 @subheading The @code{-break-catch} Command
22455 @findex -break-catch
22458 @subheading The @code{-break-commands} Command
22459 @findex -break-commands
22461 @subsubheading Synopsis
22464 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
22467 Specifies the CLI commands that should be executed when breakpoint
22468 @var{number} is hit. The parameters @var{command1} to @var{commandN}
22469 are the commands. If no command is specified, any previously-set
22470 commands are cleared. @xref{Break Commands}. Typical use of this
22471 functionality is tracing a program, that is, printing of values of
22472 some variables whenever breakpoint is hit and then continuing.
22474 @subsubheading @value{GDBN} Command
22476 The corresponding @value{GDBN} command is @samp{commands}.
22478 @subsubheading Example
22483 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
22484 enabled="y",addr="0x000100d0",func="main",file="hello.c",
22485 fullname="/home/foo/hello.c",line="5",times="0"@}
22487 -break-commands 1 "print v" "continue"
22492 @subheading The @code{-break-condition} Command
22493 @findex -break-condition
22495 @subsubheading Synopsis
22498 -break-condition @var{number} @var{expr}
22501 Breakpoint @var{number} will stop the program only if the condition in
22502 @var{expr} is true. The condition becomes part of the
22503 @samp{-break-list} output (see the description of the @samp{-break-list}
22506 @subsubheading @value{GDBN} Command
22508 The corresponding @value{GDBN} command is @samp{condition}.
22510 @subsubheading Example
22514 -break-condition 1 1
22518 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22519 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22520 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22521 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22522 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22523 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22524 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22525 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22526 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
22527 line="5",cond="1",times="0",ignore="3"@}]@}
22531 @subheading The @code{-break-delete} Command
22532 @findex -break-delete
22534 @subsubheading Synopsis
22537 -break-delete ( @var{breakpoint} )+
22540 Delete the breakpoint(s) whose number(s) are specified in the argument
22541 list. This is obviously reflected in the breakpoint list.
22543 @subsubheading @value{GDBN} Command
22545 The corresponding @value{GDBN} command is @samp{delete}.
22547 @subsubheading Example
22555 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
22556 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22557 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22558 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22559 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22560 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22561 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22566 @subheading The @code{-break-disable} Command
22567 @findex -break-disable
22569 @subsubheading Synopsis
22572 -break-disable ( @var{breakpoint} )+
22575 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
22576 break list is now set to @samp{n} for the named @var{breakpoint}(s).
22578 @subsubheading @value{GDBN} Command
22580 The corresponding @value{GDBN} command is @samp{disable}.
22582 @subsubheading Example
22590 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22591 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22592 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22593 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22594 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22595 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22596 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22597 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
22598 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
22599 line="5",times="0"@}]@}
22603 @subheading The @code{-break-enable} Command
22604 @findex -break-enable
22606 @subsubheading Synopsis
22609 -break-enable ( @var{breakpoint} )+
22612 Enable (previously disabled) @var{breakpoint}(s).
22614 @subsubheading @value{GDBN} Command
22616 The corresponding @value{GDBN} command is @samp{enable}.
22618 @subsubheading Example
22626 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22627 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22628 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22629 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22630 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22631 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22632 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22633 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
22634 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
22635 line="5",times="0"@}]@}
22639 @subheading The @code{-break-info} Command
22640 @findex -break-info
22642 @subsubheading Synopsis
22645 -break-info @var{breakpoint}
22649 Get information about a single breakpoint.
22651 @subsubheading @value{GDBN} Command
22653 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
22655 @subsubheading Example
22658 @subheading The @code{-break-insert} Command
22659 @findex -break-insert
22661 @subsubheading Synopsis
22664 -break-insert [ -t ] [ -h ] [ -f ] [ -d ]
22665 [ -c @var{condition} ] [ -i @var{ignore-count} ]
22666 [ -p @var{thread} ] [ @var{location} ]
22670 If specified, @var{location}, can be one of:
22677 @item filename:linenum
22678 @item filename:function
22682 The possible optional parameters of this command are:
22686 Insert a temporary breakpoint.
22688 Insert a hardware breakpoint.
22689 @item -c @var{condition}
22690 Make the breakpoint conditional on @var{condition}.
22691 @item -i @var{ignore-count}
22692 Initialize the @var{ignore-count}.
22694 If @var{location} cannot be parsed (for example if it
22695 refers to unknown files or functions), create a pending
22696 breakpoint. Without this flag, @value{GDBN} will report
22697 an error, and won't create a breakpoint, if @var{location}
22700 Create a disabled breakpoint.
22703 @subsubheading Result
22705 The result is in the form:
22708 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
22709 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
22710 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
22711 times="@var{times}"@}
22715 where @var{number} is the @value{GDBN} number for this breakpoint,
22716 @var{funcname} is the name of the function where the breakpoint was
22717 inserted, @var{filename} is the name of the source file which contains
22718 this function, @var{lineno} is the source line number within that file
22719 and @var{times} the number of times that the breakpoint has been hit
22720 (always 0 for -break-insert but may be greater for -break-info or -break-list
22721 which use the same output).
22723 Note: this format is open to change.
22724 @c An out-of-band breakpoint instead of part of the result?
22726 @subsubheading @value{GDBN} Command
22728 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
22729 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
22731 @subsubheading Example
22736 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
22737 fullname="/home/foo/recursive2.c,line="4",times="0"@}
22739 -break-insert -t foo
22740 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
22741 fullname="/home/foo/recursive2.c,line="11",times="0"@}
22744 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
22745 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22746 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22747 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22748 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22749 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22750 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22751 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22752 addr="0x0001072c", func="main",file="recursive2.c",
22753 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
22754 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
22755 addr="0x00010774",func="foo",file="recursive2.c",
22756 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
22758 -break-insert -r foo.*
22759 ~int foo(int, int);
22760 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
22761 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
22765 @subheading The @code{-break-list} Command
22766 @findex -break-list
22768 @subsubheading Synopsis
22774 Displays the list of inserted breakpoints, showing the following fields:
22778 number of the breakpoint
22780 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
22782 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
22785 is the breakpoint enabled or no: @samp{y} or @samp{n}
22787 memory location at which the breakpoint is set
22789 logical location of the breakpoint, expressed by function name, file
22792 number of times the breakpoint has been hit
22795 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
22796 @code{body} field is an empty list.
22798 @subsubheading @value{GDBN} Command
22800 The corresponding @value{GDBN} command is @samp{info break}.
22802 @subsubheading Example
22807 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
22808 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22809 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22810 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22811 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22812 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22813 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22814 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22815 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
22816 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
22817 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
22818 line="13",times="0"@}]@}
22822 Here's an example of the result when there are no breakpoints:
22827 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
22828 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22829 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22830 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22831 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22832 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22833 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22838 @subheading The @code{-break-watch} Command
22839 @findex -break-watch
22841 @subsubheading Synopsis
22844 -break-watch [ -a | -r ]
22847 Create a watchpoint. With the @samp{-a} option it will create an
22848 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
22849 read from or on a write to the memory location. With the @samp{-r}
22850 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
22851 trigger only when the memory location is accessed for reading. Without
22852 either of the options, the watchpoint created is a regular watchpoint,
22853 i.e., it will trigger when the memory location is accessed for writing.
22854 @xref{Set Watchpoints, , Setting Watchpoints}.
22856 Note that @samp{-break-list} will report a single list of watchpoints and
22857 breakpoints inserted.
22859 @subsubheading @value{GDBN} Command
22861 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
22864 @subsubheading Example
22866 Setting a watchpoint on a variable in the @code{main} function:
22871 ^done,wpt=@{number="2",exp="x"@}
22876 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
22877 value=@{old="-268439212",new="55"@},
22878 frame=@{func="main",args=[],file="recursive2.c",
22879 fullname="/home/foo/bar/recursive2.c",line="5"@}
22883 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
22884 the program execution twice: first for the variable changing value, then
22885 for the watchpoint going out of scope.
22890 ^done,wpt=@{number="5",exp="C"@}
22895 *stopped,reason="watchpoint-trigger",
22896 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
22897 frame=@{func="callee4",args=[],
22898 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22899 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
22904 *stopped,reason="watchpoint-scope",wpnum="5",
22905 frame=@{func="callee3",args=[@{name="strarg",
22906 value="0x11940 \"A string argument.\""@}],
22907 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22908 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
22912 Listing breakpoints and watchpoints, at different points in the program
22913 execution. Note that once the watchpoint goes out of scope, it is
22919 ^done,wpt=@{number="2",exp="C"@}
22922 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
22923 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22924 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22925 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22926 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22927 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22928 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22929 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22930 addr="0x00010734",func="callee4",
22931 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22932 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
22933 bkpt=@{number="2",type="watchpoint",disp="keep",
22934 enabled="y",addr="",what="C",times="0"@}]@}
22939 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
22940 value=@{old="-276895068",new="3"@},
22941 frame=@{func="callee4",args=[],
22942 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22943 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
22946 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
22947 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22948 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22949 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22950 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22951 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22952 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22953 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22954 addr="0x00010734",func="callee4",
22955 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22956 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
22957 bkpt=@{number="2",type="watchpoint",disp="keep",
22958 enabled="y",addr="",what="C",times="-5"@}]@}
22962 ^done,reason="watchpoint-scope",wpnum="2",
22963 frame=@{func="callee3",args=[@{name="strarg",
22964 value="0x11940 \"A string argument.\""@}],
22965 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22966 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
22969 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22970 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22971 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22972 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22973 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22974 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22975 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22976 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22977 addr="0x00010734",func="callee4",
22978 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22979 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
22984 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22985 @node GDB/MI Program Context
22986 @section @sc{gdb/mi} Program Context
22988 @subheading The @code{-exec-arguments} Command
22989 @findex -exec-arguments
22992 @subsubheading Synopsis
22995 -exec-arguments @var{args}
22998 Set the inferior program arguments, to be used in the next
23001 @subsubheading @value{GDBN} Command
23003 The corresponding @value{GDBN} command is @samp{set args}.
23005 @subsubheading Example
23009 -exec-arguments -v word
23016 @subheading The @code{-exec-show-arguments} Command
23017 @findex -exec-show-arguments
23019 @subsubheading Synopsis
23022 -exec-show-arguments
23025 Print the arguments of the program.
23027 @subsubheading @value{GDBN} Command
23029 The corresponding @value{GDBN} command is @samp{show args}.
23031 @subsubheading Example
23036 @subheading The @code{-environment-cd} Command
23037 @findex -environment-cd
23039 @subsubheading Synopsis
23042 -environment-cd @var{pathdir}
23045 Set @value{GDBN}'s working directory.
23047 @subsubheading @value{GDBN} Command
23049 The corresponding @value{GDBN} command is @samp{cd}.
23051 @subsubheading Example
23055 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
23061 @subheading The @code{-environment-directory} Command
23062 @findex -environment-directory
23064 @subsubheading Synopsis
23067 -environment-directory [ -r ] [ @var{pathdir} ]+
23070 Add directories @var{pathdir} to beginning of search path for source files.
23071 If the @samp{-r} option is used, the search path is reset to the default
23072 search path. If directories @var{pathdir} are supplied in addition to the
23073 @samp{-r} option, the search path is first reset and then addition
23075 Multiple directories may be specified, separated by blanks. Specifying
23076 multiple directories in a single command
23077 results in the directories added to the beginning of the
23078 search path in the same order they were presented in the command.
23079 If blanks are needed as
23080 part of a directory name, double-quotes should be used around
23081 the name. In the command output, the path will show up separated
23082 by the system directory-separator character. The directory-separator
23083 character must not be used
23084 in any directory name.
23085 If no directories are specified, the current search path is displayed.
23087 @subsubheading @value{GDBN} Command
23089 The corresponding @value{GDBN} command is @samp{dir}.
23091 @subsubheading Example
23095 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
23096 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
23098 -environment-directory ""
23099 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
23101 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
23102 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
23104 -environment-directory -r
23105 ^done,source-path="$cdir:$cwd"
23110 @subheading The @code{-environment-path} Command
23111 @findex -environment-path
23113 @subsubheading Synopsis
23116 -environment-path [ -r ] [ @var{pathdir} ]+
23119 Add directories @var{pathdir} to beginning of search path for object files.
23120 If the @samp{-r} option is used, the search path is reset to the original
23121 search path that existed at gdb start-up. If directories @var{pathdir} are
23122 supplied in addition to the
23123 @samp{-r} option, the search path is first reset and then addition
23125 Multiple directories may be specified, separated by blanks. Specifying
23126 multiple directories in a single command
23127 results in the directories added to the beginning of the
23128 search path in the same order they were presented in the command.
23129 If blanks are needed as
23130 part of a directory name, double-quotes should be used around
23131 the name. In the command output, the path will show up separated
23132 by the system directory-separator character. The directory-separator
23133 character must not be used
23134 in any directory name.
23135 If no directories are specified, the current path is displayed.
23138 @subsubheading @value{GDBN} Command
23140 The corresponding @value{GDBN} command is @samp{path}.
23142 @subsubheading Example
23147 ^done,path="/usr/bin"
23149 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
23150 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
23152 -environment-path -r /usr/local/bin
23153 ^done,path="/usr/local/bin:/usr/bin"
23158 @subheading The @code{-environment-pwd} Command
23159 @findex -environment-pwd
23161 @subsubheading Synopsis
23167 Show the current working directory.
23169 @subsubheading @value{GDBN} Command
23171 The corresponding @value{GDBN} command is @samp{pwd}.
23173 @subsubheading Example
23178 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
23182 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23183 @node GDB/MI Thread Commands
23184 @section @sc{gdb/mi} Thread Commands
23187 @subheading The @code{-thread-info} Command
23188 @findex -thread-info
23190 @subsubheading Synopsis
23193 -thread-info [ @var{thread-id} ]
23196 Reports information about either a specific thread, if
23197 the @var{thread-id} parameter is present, or about all
23198 threads. When printing information about all threads,
23199 also reports the current thread.
23201 @subsubheading @value{GDBN} Command
23203 The @samp{info thread} command prints the same information
23206 @subsubheading Example
23211 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
23212 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
23213 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
23214 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
23215 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}],
23216 current-thread-id="1"
23220 The @samp{state} field may have the following values:
23224 The thread is stopped. Frame information is available for stopped
23228 The thread is running. There's no frame information for running
23233 @subheading The @code{-thread-list-ids} Command
23234 @findex -thread-list-ids
23236 @subsubheading Synopsis
23242 Produces a list of the currently known @value{GDBN} thread ids. At the
23243 end of the list it also prints the total number of such threads.
23245 This command is retained for historical reasons, the
23246 @code{-thread-info} command should be used instead.
23248 @subsubheading @value{GDBN} Command
23250 Part of @samp{info threads} supplies the same information.
23252 @subsubheading Example
23257 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
23258 current-thread-id="1",number-of-threads="3"
23263 @subheading The @code{-thread-select} Command
23264 @findex -thread-select
23266 @subsubheading Synopsis
23269 -thread-select @var{threadnum}
23272 Make @var{threadnum} the current thread. It prints the number of the new
23273 current thread, and the topmost frame for that thread.
23275 This command is deprecated in favor of explicitly using the
23276 @samp{--thread} option to each command.
23278 @subsubheading @value{GDBN} Command
23280 The corresponding @value{GDBN} command is @samp{thread}.
23282 @subsubheading Example
23289 *stopped,reason="end-stepping-range",thread-id="2",line="187",
23290 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
23294 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
23295 number-of-threads="3"
23298 ^done,new-thread-id="3",
23299 frame=@{level="0",func="vprintf",
23300 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
23301 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
23305 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23306 @node GDB/MI Program Execution
23307 @section @sc{gdb/mi} Program Execution
23309 These are the asynchronous commands which generate the out-of-band
23310 record @samp{*stopped}. Currently @value{GDBN} only really executes
23311 asynchronously with remote targets and this interaction is mimicked in
23314 @subheading The @code{-exec-continue} Command
23315 @findex -exec-continue
23317 @subsubheading Synopsis
23320 -exec-continue [--all|--thread-group N]
23323 Resumes the execution of the inferior program until a breakpoint is
23324 encountered, or until the inferior exits. In all-stop mode
23325 (@pxref{All-Stop Mode}), may resume only one thread, or all threads,
23326 depending on the value of the @samp{scheduler-locking} variable. In
23327 non-stop mode (@pxref{Non-Stop Mode}), if the @samp{--all} is not
23328 specified, only the thread specified with the @samp{--thread} option
23329 (or current thread, if no @samp{--thread} is provided) is resumed. If
23330 @samp{--all} is specified, all threads will be resumed. The
23331 @samp{--all} option is ignored in all-stop mode. If the
23332 @samp{--thread-group} options is specified, then all threads in that
23333 thread group are resumed.
23335 @subsubheading @value{GDBN} Command
23337 The corresponding @value{GDBN} corresponding is @samp{continue}.
23339 @subsubheading Example
23346 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
23347 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
23353 @subheading The @code{-exec-finish} Command
23354 @findex -exec-finish
23356 @subsubheading Synopsis
23362 Resumes the execution of the inferior program until the current
23363 function is exited. Displays the results returned by the function.
23365 @subsubheading @value{GDBN} Command
23367 The corresponding @value{GDBN} command is @samp{finish}.
23369 @subsubheading Example
23371 Function returning @code{void}.
23378 *stopped,reason="function-finished",frame=@{func="main",args=[],
23379 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
23383 Function returning other than @code{void}. The name of the internal
23384 @value{GDBN} variable storing the result is printed, together with the
23391 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
23392 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
23393 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23394 gdb-result-var="$1",return-value="0"
23399 @subheading The @code{-exec-interrupt} Command
23400 @findex -exec-interrupt
23402 @subsubheading Synopsis
23405 -exec-interrupt [--all|--thread-group N]
23408 Interrupts the background execution of the target. Note how the token
23409 associated with the stop message is the one for the execution command
23410 that has been interrupted. The token for the interrupt itself only
23411 appears in the @samp{^done} output. If the user is trying to
23412 interrupt a non-running program, an error message will be printed.
23414 Note that when asynchronous execution is enabled, this command is
23415 asynchronous just like other execution commands. That is, first the
23416 @samp{^done} response will be printed, and the target stop will be
23417 reported after that using the @samp{*stopped} notification.
23419 In non-stop mode, only the context thread is interrupted by default.
23420 All threads will be interrupted if the @samp{--all} option is
23421 specified. If the @samp{--thread-group} option is specified, all
23422 threads in that group will be interrupted.
23424 @subsubheading @value{GDBN} Command
23426 The corresponding @value{GDBN} command is @samp{interrupt}.
23428 @subsubheading Example
23439 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
23440 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
23441 fullname="/home/foo/bar/try.c",line="13"@}
23446 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
23450 @subheading The @code{-exec-jump} Command
23453 @subsubheading Synopsis
23456 -exec-jump @var{location}
23459 Resumes execution of the inferior program at the location specified by
23460 parameter. @xref{Specify Location}, for a description of the
23461 different forms of @var{location}.
23463 @subsubheading @value{GDBN} Command
23465 The corresponding @value{GDBN} command is @samp{jump}.
23467 @subsubheading Example
23470 -exec-jump foo.c:10
23471 *running,thread-id="all"
23476 @subheading The @code{-exec-next} Command
23479 @subsubheading Synopsis
23485 Resumes execution of the inferior program, stopping when the beginning
23486 of the next source line is reached.
23488 @subsubheading @value{GDBN} Command
23490 The corresponding @value{GDBN} command is @samp{next}.
23492 @subsubheading Example
23498 *stopped,reason="end-stepping-range",line="8",file="hello.c"
23503 @subheading The @code{-exec-next-instruction} Command
23504 @findex -exec-next-instruction
23506 @subsubheading Synopsis
23509 -exec-next-instruction
23512 Executes one machine instruction. If the instruction is a function
23513 call, continues until the function returns. If the program stops at an
23514 instruction in the middle of a source line, the address will be
23517 @subsubheading @value{GDBN} Command
23519 The corresponding @value{GDBN} command is @samp{nexti}.
23521 @subsubheading Example
23525 -exec-next-instruction
23529 *stopped,reason="end-stepping-range",
23530 addr="0x000100d4",line="5",file="hello.c"
23535 @subheading The @code{-exec-return} Command
23536 @findex -exec-return
23538 @subsubheading Synopsis
23544 Makes current function return immediately. Doesn't execute the inferior.
23545 Displays the new current frame.
23547 @subsubheading @value{GDBN} Command
23549 The corresponding @value{GDBN} command is @samp{return}.
23551 @subsubheading Example
23555 200-break-insert callee4
23556 200^done,bkpt=@{number="1",addr="0x00010734",
23557 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
23562 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
23563 frame=@{func="callee4",args=[],
23564 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23565 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
23571 111^done,frame=@{level="0",func="callee3",
23572 args=[@{name="strarg",
23573 value="0x11940 \"A string argument.\""@}],
23574 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23575 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
23580 @subheading The @code{-exec-run} Command
23583 @subsubheading Synopsis
23589 Starts execution of the inferior from the beginning. The inferior
23590 executes until either a breakpoint is encountered or the program
23591 exits. In the latter case the output will include an exit code, if
23592 the program has exited exceptionally.
23594 @subsubheading @value{GDBN} Command
23596 The corresponding @value{GDBN} command is @samp{run}.
23598 @subsubheading Examples
23603 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
23608 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
23609 frame=@{func="main",args=[],file="recursive2.c",
23610 fullname="/home/foo/bar/recursive2.c",line="4"@}
23615 Program exited normally:
23623 *stopped,reason="exited-normally"
23628 Program exited exceptionally:
23636 *stopped,reason="exited",exit-code="01"
23640 Another way the program can terminate is if it receives a signal such as
23641 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
23645 *stopped,reason="exited-signalled",signal-name="SIGINT",
23646 signal-meaning="Interrupt"
23650 @c @subheading -exec-signal
23653 @subheading The @code{-exec-step} Command
23656 @subsubheading Synopsis
23662 Resumes execution of the inferior program, stopping when the beginning
23663 of the next source line is reached, if the next source line is not a
23664 function call. If it is, stop at the first instruction of the called
23667 @subsubheading @value{GDBN} Command
23669 The corresponding @value{GDBN} command is @samp{step}.
23671 @subsubheading Example
23673 Stepping into a function:
23679 *stopped,reason="end-stepping-range",
23680 frame=@{func="foo",args=[@{name="a",value="10"@},
23681 @{name="b",value="0"@}],file="recursive2.c",
23682 fullname="/home/foo/bar/recursive2.c",line="11"@}
23692 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
23697 @subheading The @code{-exec-step-instruction} Command
23698 @findex -exec-step-instruction
23700 @subsubheading Synopsis
23703 -exec-step-instruction
23706 Resumes the inferior which executes one machine instruction. The
23707 output, once @value{GDBN} has stopped, will vary depending on whether
23708 we have stopped in the middle of a source line or not. In the former
23709 case, the address at which the program stopped will be printed as
23712 @subsubheading @value{GDBN} Command
23714 The corresponding @value{GDBN} command is @samp{stepi}.
23716 @subsubheading Example
23720 -exec-step-instruction
23724 *stopped,reason="end-stepping-range",
23725 frame=@{func="foo",args=[],file="try.c",
23726 fullname="/home/foo/bar/try.c",line="10"@}
23728 -exec-step-instruction
23732 *stopped,reason="end-stepping-range",
23733 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
23734 fullname="/home/foo/bar/try.c",line="10"@}
23739 @subheading The @code{-exec-until} Command
23740 @findex -exec-until
23742 @subsubheading Synopsis
23745 -exec-until [ @var{location} ]
23748 Executes the inferior until the @var{location} specified in the
23749 argument is reached. If there is no argument, the inferior executes
23750 until a source line greater than the current one is reached. The
23751 reason for stopping in this case will be @samp{location-reached}.
23753 @subsubheading @value{GDBN} Command
23755 The corresponding @value{GDBN} command is @samp{until}.
23757 @subsubheading Example
23761 -exec-until recursive2.c:6
23765 *stopped,reason="location-reached",frame=@{func="main",args=[],
23766 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
23771 @subheading -file-clear
23772 Is this going away????
23775 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23776 @node GDB/MI Stack Manipulation
23777 @section @sc{gdb/mi} Stack Manipulation Commands
23780 @subheading The @code{-stack-info-frame} Command
23781 @findex -stack-info-frame
23783 @subsubheading Synopsis
23789 Get info on the selected frame.
23791 @subsubheading @value{GDBN} Command
23793 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
23794 (without arguments).
23796 @subsubheading Example
23801 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
23802 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23803 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
23807 @subheading The @code{-stack-info-depth} Command
23808 @findex -stack-info-depth
23810 @subsubheading Synopsis
23813 -stack-info-depth [ @var{max-depth} ]
23816 Return the depth of the stack. If the integer argument @var{max-depth}
23817 is specified, do not count beyond @var{max-depth} frames.
23819 @subsubheading @value{GDBN} Command
23821 There's no equivalent @value{GDBN} command.
23823 @subsubheading Example
23825 For a stack with frame levels 0 through 11:
23832 -stack-info-depth 4
23835 -stack-info-depth 12
23838 -stack-info-depth 11
23841 -stack-info-depth 13
23846 @subheading The @code{-stack-list-arguments} Command
23847 @findex -stack-list-arguments
23849 @subsubheading Synopsis
23852 -stack-list-arguments @var{print-values}
23853 [ @var{low-frame} @var{high-frame} ]
23856 Display a list of the arguments for the frames between @var{low-frame}
23857 and @var{high-frame} (inclusive). If @var{low-frame} and
23858 @var{high-frame} are not provided, list the arguments for the whole
23859 call stack. If the two arguments are equal, show the single frame
23860 at the corresponding level. It is an error if @var{low-frame} is
23861 larger than the actual number of frames. On the other hand,
23862 @var{high-frame} may be larger than the actual number of frames, in
23863 which case only existing frames will be returned.
23865 If @var{print-values} is 0 or @code{--no-values}, print only the names of
23866 the variables; if it is 1 or @code{--all-values}, print also their
23867 values; and if it is 2 or @code{--simple-values}, print the name,
23868 type and value for simple data types, and the name and type for arrays,
23869 structures and unions.
23871 Use of this command to obtain arguments in a single frame is
23872 deprecated in favor of the @samp{-stack-list-variables} command.
23874 @subsubheading @value{GDBN} Command
23876 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
23877 @samp{gdb_get_args} command which partially overlaps with the
23878 functionality of @samp{-stack-list-arguments}.
23880 @subsubheading Example
23887 frame=@{level="0",addr="0x00010734",func="callee4",
23888 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23889 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
23890 frame=@{level="1",addr="0x0001076c",func="callee3",
23891 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23892 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
23893 frame=@{level="2",addr="0x0001078c",func="callee2",
23894 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23895 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
23896 frame=@{level="3",addr="0x000107b4",func="callee1",
23897 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23898 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
23899 frame=@{level="4",addr="0x000107e0",func="main",
23900 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23901 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
23903 -stack-list-arguments 0
23906 frame=@{level="0",args=[]@},
23907 frame=@{level="1",args=[name="strarg"]@},
23908 frame=@{level="2",args=[name="intarg",name="strarg"]@},
23909 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
23910 frame=@{level="4",args=[]@}]
23912 -stack-list-arguments 1
23915 frame=@{level="0",args=[]@},
23917 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
23918 frame=@{level="2",args=[
23919 @{name="intarg",value="2"@},
23920 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
23921 @{frame=@{level="3",args=[
23922 @{name="intarg",value="2"@},
23923 @{name="strarg",value="0x11940 \"A string argument.\""@},
23924 @{name="fltarg",value="3.5"@}]@},
23925 frame=@{level="4",args=[]@}]
23927 -stack-list-arguments 0 2 2
23928 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
23930 -stack-list-arguments 1 2 2
23931 ^done,stack-args=[frame=@{level="2",
23932 args=[@{name="intarg",value="2"@},
23933 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
23937 @c @subheading -stack-list-exception-handlers
23940 @subheading The @code{-stack-list-frames} Command
23941 @findex -stack-list-frames
23943 @subsubheading Synopsis
23946 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
23949 List the frames currently on the stack. For each frame it displays the
23954 The frame number, 0 being the topmost frame, i.e., the innermost function.
23956 The @code{$pc} value for that frame.
23960 File name of the source file where the function lives.
23962 Line number corresponding to the @code{$pc}.
23965 If invoked without arguments, this command prints a backtrace for the
23966 whole stack. If given two integer arguments, it shows the frames whose
23967 levels are between the two arguments (inclusive). If the two arguments
23968 are equal, it shows the single frame at the corresponding level. It is
23969 an error if @var{low-frame} is larger than the actual number of
23970 frames. On the other hand, @var{high-frame} may be larger than the
23971 actual number of frames, in which case only existing frames will be returned.
23973 @subsubheading @value{GDBN} Command
23975 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
23977 @subsubheading Example
23979 Full stack backtrace:
23985 [frame=@{level="0",addr="0x0001076c",func="foo",
23986 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
23987 frame=@{level="1",addr="0x000107a4",func="foo",
23988 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23989 frame=@{level="2",addr="0x000107a4",func="foo",
23990 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23991 frame=@{level="3",addr="0x000107a4",func="foo",
23992 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23993 frame=@{level="4",addr="0x000107a4",func="foo",
23994 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23995 frame=@{level="5",addr="0x000107a4",func="foo",
23996 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23997 frame=@{level="6",addr="0x000107a4",func="foo",
23998 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23999 frame=@{level="7",addr="0x000107a4",func="foo",
24000 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24001 frame=@{level="8",addr="0x000107a4",func="foo",
24002 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24003 frame=@{level="9",addr="0x000107a4",func="foo",
24004 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24005 frame=@{level="10",addr="0x000107a4",func="foo",
24006 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24007 frame=@{level="11",addr="0x00010738",func="main",
24008 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
24012 Show frames between @var{low_frame} and @var{high_frame}:
24016 -stack-list-frames 3 5
24018 [frame=@{level="3",addr="0x000107a4",func="foo",
24019 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24020 frame=@{level="4",addr="0x000107a4",func="foo",
24021 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
24022 frame=@{level="5",addr="0x000107a4",func="foo",
24023 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
24027 Show a single frame:
24031 -stack-list-frames 3 3
24033 [frame=@{level="3",addr="0x000107a4",func="foo",
24034 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
24039 @subheading The @code{-stack-list-locals} Command
24040 @findex -stack-list-locals
24042 @subsubheading Synopsis
24045 -stack-list-locals @var{print-values}
24048 Display the local variable names for the selected frame. If
24049 @var{print-values} is 0 or @code{--no-values}, print only the names of
24050 the variables; if it is 1 or @code{--all-values}, print also their
24051 values; and if it is 2 or @code{--simple-values}, print the name,
24052 type and value for simple data types, and the name and type for arrays,
24053 structures and unions. In this last case, a frontend can immediately
24054 display the value of simple data types and create variable objects for
24055 other data types when the user wishes to explore their values in
24058 This command is deprecated in favor of the
24059 @samp{-stack-list-variables} command.
24061 @subsubheading @value{GDBN} Command
24063 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
24065 @subsubheading Example
24069 -stack-list-locals 0
24070 ^done,locals=[name="A",name="B",name="C"]
24072 -stack-list-locals --all-values
24073 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
24074 @{name="C",value="@{1, 2, 3@}"@}]
24075 -stack-list-locals --simple-values
24076 ^done,locals=[@{name="A",type="int",value="1"@},
24077 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
24081 @subheading The @code{-stack-list-variables} Command
24082 @findex -stack-list-variables
24084 @subsubheading Synopsis
24087 -stack-list-variables @var{print-values}
24090 Display the names of local variables and function arguments for the selected frame. If
24091 @var{print-values} is 0 or @code{--no-values}, print only the names of
24092 the variables; if it is 1 or @code{--all-values}, print also their
24093 values; and if it is 2 or @code{--simple-values}, print the name,
24094 type and value for simple data types, and the name and type for arrays,
24095 structures and unions.
24097 @subsubheading Example
24101 -stack-list-variables --thread 1 --frame 0 --all-values
24102 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
24107 @subheading The @code{-stack-select-frame} Command
24108 @findex -stack-select-frame
24110 @subsubheading Synopsis
24113 -stack-select-frame @var{framenum}
24116 Change the selected frame. Select a different frame @var{framenum} on
24119 This command in deprecated in favor of passing the @samp{--frame}
24120 option to every command.
24122 @subsubheading @value{GDBN} Command
24124 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
24125 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
24127 @subsubheading Example
24131 -stack-select-frame 2
24136 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24137 @node GDB/MI Variable Objects
24138 @section @sc{gdb/mi} Variable Objects
24142 @subheading Motivation for Variable Objects in @sc{gdb/mi}
24144 For the implementation of a variable debugger window (locals, watched
24145 expressions, etc.), we are proposing the adaptation of the existing code
24146 used by @code{Insight}.
24148 The two main reasons for that are:
24152 It has been proven in practice (it is already on its second generation).
24155 It will shorten development time (needless to say how important it is
24159 The original interface was designed to be used by Tcl code, so it was
24160 slightly changed so it could be used through @sc{gdb/mi}. This section
24161 describes the @sc{gdb/mi} operations that will be available and gives some
24162 hints about their use.
24164 @emph{Note}: In addition to the set of operations described here, we
24165 expect the @sc{gui} implementation of a variable window to require, at
24166 least, the following operations:
24169 @item @code{-gdb-show} @code{output-radix}
24170 @item @code{-stack-list-arguments}
24171 @item @code{-stack-list-locals}
24172 @item @code{-stack-select-frame}
24177 @subheading Introduction to Variable Objects
24179 @cindex variable objects in @sc{gdb/mi}
24181 Variable objects are "object-oriented" MI interface for examining and
24182 changing values of expressions. Unlike some other MI interfaces that
24183 work with expressions, variable objects are specifically designed for
24184 simple and efficient presentation in the frontend. A variable object
24185 is identified by string name. When a variable object is created, the
24186 frontend specifies the expression for that variable object. The
24187 expression can be a simple variable, or it can be an arbitrary complex
24188 expression, and can even involve CPU registers. After creating a
24189 variable object, the frontend can invoke other variable object
24190 operations---for example to obtain or change the value of a variable
24191 object, or to change display format.
24193 Variable objects have hierarchical tree structure. Any variable object
24194 that corresponds to a composite type, such as structure in C, has
24195 a number of child variable objects, for example corresponding to each
24196 element of a structure. A child variable object can itself have
24197 children, recursively. Recursion ends when we reach
24198 leaf variable objects, which always have built-in types. Child variable
24199 objects are created only by explicit request, so if a frontend
24200 is not interested in the children of a particular variable object, no
24201 child will be created.
24203 For a leaf variable object it is possible to obtain its value as a
24204 string, or set the value from a string. String value can be also
24205 obtained for a non-leaf variable object, but it's generally a string
24206 that only indicates the type of the object, and does not list its
24207 contents. Assignment to a non-leaf variable object is not allowed.
24209 A frontend does not need to read the values of all variable objects each time
24210 the program stops. Instead, MI provides an update command that lists all
24211 variable objects whose values has changed since the last update
24212 operation. This considerably reduces the amount of data that must
24213 be transferred to the frontend. As noted above, children variable
24214 objects are created on demand, and only leaf variable objects have a
24215 real value. As result, gdb will read target memory only for leaf
24216 variables that frontend has created.
24218 The automatic update is not always desirable. For example, a frontend
24219 might want to keep a value of some expression for future reference,
24220 and never update it. For another example, fetching memory is
24221 relatively slow for embedded targets, so a frontend might want
24222 to disable automatic update for the variables that are either not
24223 visible on the screen, or ``closed''. This is possible using so
24224 called ``frozen variable objects''. Such variable objects are never
24225 implicitly updated.
24227 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
24228 fixed variable object, the expression is parsed when the variable
24229 object is created, including associating identifiers to specific
24230 variables. The meaning of expression never changes. For a floating
24231 variable object the values of variables whose names appear in the
24232 expressions are re-evaluated every time in the context of the current
24233 frame. Consider this example:
24238 struct work_state state;
24245 If a fixed variable object for the @code{state} variable is created in
24246 this function, and we enter the recursive call, the the variable
24247 object will report the value of @code{state} in the top-level
24248 @code{do_work} invocation. On the other hand, a floating variable
24249 object will report the value of @code{state} in the current frame.
24251 If an expression specified when creating a fixed variable object
24252 refers to a local variable, the variable object becomes bound to the
24253 thread and frame in which the variable object is created. When such
24254 variable object is updated, @value{GDBN} makes sure that the
24255 thread/frame combination the variable object is bound to still exists,
24256 and re-evaluates the variable object in context of that thread/frame.
24258 The following is the complete set of @sc{gdb/mi} operations defined to
24259 access this functionality:
24261 @multitable @columnfractions .4 .6
24262 @item @strong{Operation}
24263 @tab @strong{Description}
24265 @item @code{-enable-pretty-printing}
24266 @tab enable Python-based pretty-printing
24267 @item @code{-var-create}
24268 @tab create a variable object
24269 @item @code{-var-delete}
24270 @tab delete the variable object and/or its children
24271 @item @code{-var-set-format}
24272 @tab set the display format of this variable
24273 @item @code{-var-show-format}
24274 @tab show the display format of this variable
24275 @item @code{-var-info-num-children}
24276 @tab tells how many children this object has
24277 @item @code{-var-list-children}
24278 @tab return a list of the object's children
24279 @item @code{-var-info-type}
24280 @tab show the type of this variable object
24281 @item @code{-var-info-expression}
24282 @tab print parent-relative expression that this variable object represents
24283 @item @code{-var-info-path-expression}
24284 @tab print full expression that this variable object represents
24285 @item @code{-var-show-attributes}
24286 @tab is this variable editable? does it exist here?
24287 @item @code{-var-evaluate-expression}
24288 @tab get the value of this variable
24289 @item @code{-var-assign}
24290 @tab set the value of this variable
24291 @item @code{-var-update}
24292 @tab update the variable and its children
24293 @item @code{-var-set-frozen}
24294 @tab set frozeness attribute
24295 @item @code{-var-set-update-range}
24296 @tab set range of children to display on update
24299 In the next subsection we describe each operation in detail and suggest
24300 how it can be used.
24302 @subheading Description And Use of Operations on Variable Objects
24304 @subheading The @code{-enable-pretty-printing} Command
24305 @findex -enable-pretty-printing
24308 -enable-pretty-printing
24311 @value{GDBN} allows Python-based visualizers to affect the output of the
24312 MI variable object commands. However, because there was no way to
24313 implement this in a fully backward-compatible way, a front end must
24314 request that this functionality be enabled.
24316 Once enabled, this feature cannot be disabled.
24318 Note that if Python support has not been compiled into @value{GDBN},
24319 this command will still succeed (and do nothing).
24321 This feature is currently (as of @value{GDBN} 7.0) experimental, and
24322 may work differently in future versions of @value{GDBN}.
24324 @subheading The @code{-var-create} Command
24325 @findex -var-create
24327 @subsubheading Synopsis
24330 -var-create @{@var{name} | "-"@}
24331 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
24334 This operation creates a variable object, which allows the monitoring of
24335 a variable, the result of an expression, a memory cell or a CPU
24338 The @var{name} parameter is the string by which the object can be
24339 referenced. It must be unique. If @samp{-} is specified, the varobj
24340 system will generate a string ``varNNNNNN'' automatically. It will be
24341 unique provided that one does not specify @var{name} of that format.
24342 The command fails if a duplicate name is found.
24344 The frame under which the expression should be evaluated can be
24345 specified by @var{frame-addr}. A @samp{*} indicates that the current
24346 frame should be used. A @samp{@@} indicates that a floating variable
24347 object must be created.
24349 @var{expression} is any expression valid on the current language set (must not
24350 begin with a @samp{*}), or one of the following:
24354 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
24357 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
24360 @samp{$@var{regname}} --- a CPU register name
24363 @cindex dynamic varobj
24364 A varobj's contents may be provided by a Python-based pretty-printer. In this
24365 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
24366 have slightly different semantics in some cases. If the
24367 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
24368 will never create a dynamic varobj. This ensures backward
24369 compatibility for existing clients.
24371 @subsubheading Result
24373 This operation returns attributes of the newly-created varobj. These
24378 The name of the varobj.
24381 The number of children of the varobj. This number is not necessarily
24382 reliable for a dynamic varobj. Instead, you must examine the
24383 @samp{has_more} attribute.
24386 The varobj's scalar value. For a varobj whose type is some sort of
24387 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
24388 will not be interesting.
24391 The varobj's type. This is a string representation of the type, as
24392 would be printed by the @value{GDBN} CLI.
24395 If a variable object is bound to a specific thread, then this is the
24396 thread's identifier.
24399 For a dynamic varobj, this indicates whether there appear to be any
24400 children available. For a non-dynamic varobj, this will be 0.
24403 This attribute will be present and have the value @samp{1} if the
24404 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
24405 then this attribute will not be present.
24408 A dynamic varobj can supply a display hint to the front end. The
24409 value comes directly from the Python pretty-printer object's
24410 @code{display_hint} method. @xref{Pretty Printing}.
24413 Typical output will look like this:
24416 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
24417 has_more="@var{has_more}"
24421 @subheading The @code{-var-delete} Command
24422 @findex -var-delete
24424 @subsubheading Synopsis
24427 -var-delete [ -c ] @var{name}
24430 Deletes a previously created variable object and all of its children.
24431 With the @samp{-c} option, just deletes the children.
24433 Returns an error if the object @var{name} is not found.
24436 @subheading The @code{-var-set-format} Command
24437 @findex -var-set-format
24439 @subsubheading Synopsis
24442 -var-set-format @var{name} @var{format-spec}
24445 Sets the output format for the value of the object @var{name} to be
24448 @anchor{-var-set-format}
24449 The syntax for the @var{format-spec} is as follows:
24452 @var{format-spec} @expansion{}
24453 @{binary | decimal | hexadecimal | octal | natural@}
24456 The natural format is the default format choosen automatically
24457 based on the variable type (like decimal for an @code{int}, hex
24458 for pointers, etc.).
24460 For a variable with children, the format is set only on the
24461 variable itself, and the children are not affected.
24463 @subheading The @code{-var-show-format} Command
24464 @findex -var-show-format
24466 @subsubheading Synopsis
24469 -var-show-format @var{name}
24472 Returns the format used to display the value of the object @var{name}.
24475 @var{format} @expansion{}
24480 @subheading The @code{-var-info-num-children} Command
24481 @findex -var-info-num-children
24483 @subsubheading Synopsis
24486 -var-info-num-children @var{name}
24489 Returns the number of children of a variable object @var{name}:
24495 Note that this number is not completely reliable for a dynamic varobj.
24496 It will return the current number of children, but more children may
24500 @subheading The @code{-var-list-children} Command
24501 @findex -var-list-children
24503 @subsubheading Synopsis
24506 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
24508 @anchor{-var-list-children}
24510 Return a list of the children of the specified variable object and
24511 create variable objects for them, if they do not already exist. With
24512 a single argument or if @var{print-values} has a value for of 0 or
24513 @code{--no-values}, print only the names of the variables; if
24514 @var{print-values} is 1 or @code{--all-values}, also print their
24515 values; and if it is 2 or @code{--simple-values} print the name and
24516 value for simple data types and just the name for arrays, structures
24519 @var{from} and @var{to}, if specified, indicate the range of children
24520 to report. If @var{from} or @var{to} is less than zero, the range is
24521 reset and all children will be reported. Otherwise, children starting
24522 at @var{from} (zero-based) and up to and excluding @var{to} will be
24525 If a child range is requested, it will only affect the current call to
24526 @code{-var-list-children}, but not future calls to @code{-var-update}.
24527 For this, you must instead use @code{-var-set-update-range}. The
24528 intent of this approach is to enable a front end to implement any
24529 update approach it likes; for example, scrolling a view may cause the
24530 front end to request more children with @code{-var-list-children}, and
24531 then the front end could call @code{-var-set-update-range} with a
24532 different range to ensure that future updates are restricted to just
24535 For each child the following results are returned:
24540 Name of the variable object created for this child.
24543 The expression to be shown to the user by the front end to designate this child.
24544 For example this may be the name of a structure member.
24546 For a dynamic varobj, this value cannot be used to form an
24547 expression. There is no way to do this at all with a dynamic varobj.
24549 For C/C@t{++} structures there are several pseudo children returned to
24550 designate access qualifiers. For these pseudo children @var{exp} is
24551 @samp{public}, @samp{private}, or @samp{protected}. In this case the
24552 type and value are not present.
24554 A dynamic varobj will not report the access qualifying
24555 pseudo-children, regardless of the language. This information is not
24556 available at all with a dynamic varobj.
24559 Number of children this child has. For a dynamic varobj, this will be
24563 The type of the child.
24566 If values were requested, this is the value.
24569 If this variable object is associated with a thread, this is the thread id.
24570 Otherwise this result is not present.
24573 If the variable object is frozen, this variable will be present with a value of 1.
24576 The result may have its own attributes:
24580 A dynamic varobj can supply a display hint to the front end. The
24581 value comes directly from the Python pretty-printer object's
24582 @code{display_hint} method. @xref{Pretty Printing}.
24585 This is an integer attribute which is nonzero if there are children
24586 remaining after the end of the selected range.
24589 @subsubheading Example
24593 -var-list-children n
24594 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
24595 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
24597 -var-list-children --all-values n
24598 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
24599 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
24603 @subheading The @code{-var-info-type} Command
24604 @findex -var-info-type
24606 @subsubheading Synopsis
24609 -var-info-type @var{name}
24612 Returns the type of the specified variable @var{name}. The type is
24613 returned as a string in the same format as it is output by the
24617 type=@var{typename}
24621 @subheading The @code{-var-info-expression} Command
24622 @findex -var-info-expression
24624 @subsubheading Synopsis
24627 -var-info-expression @var{name}
24630 Returns a string that is suitable for presenting this
24631 variable object in user interface. The string is generally
24632 not valid expression in the current language, and cannot be evaluated.
24634 For example, if @code{a} is an array, and variable object
24635 @code{A} was created for @code{a}, then we'll get this output:
24638 (gdb) -var-info-expression A.1
24639 ^done,lang="C",exp="1"
24643 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
24645 Note that the output of the @code{-var-list-children} command also
24646 includes those expressions, so the @code{-var-info-expression} command
24649 @subheading The @code{-var-info-path-expression} Command
24650 @findex -var-info-path-expression
24652 @subsubheading Synopsis
24655 -var-info-path-expression @var{name}
24658 Returns an expression that can be evaluated in the current
24659 context and will yield the same value that a variable object has.
24660 Compare this with the @code{-var-info-expression} command, which
24661 result can be used only for UI presentation. Typical use of
24662 the @code{-var-info-path-expression} command is creating a
24663 watchpoint from a variable object.
24665 This command is currently not valid for children of a dynamic varobj,
24666 and will give an error when invoked on one.
24668 For example, suppose @code{C} is a C@t{++} class, derived from class
24669 @code{Base}, and that the @code{Base} class has a member called
24670 @code{m_size}. Assume a variable @code{c} is has the type of
24671 @code{C} and a variable object @code{C} was created for variable
24672 @code{c}. Then, we'll get this output:
24674 (gdb) -var-info-path-expression C.Base.public.m_size
24675 ^done,path_expr=((Base)c).m_size)
24678 @subheading The @code{-var-show-attributes} Command
24679 @findex -var-show-attributes
24681 @subsubheading Synopsis
24684 -var-show-attributes @var{name}
24687 List attributes of the specified variable object @var{name}:
24690 status=@var{attr} [ ( ,@var{attr} )* ]
24694 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
24696 @subheading The @code{-var-evaluate-expression} Command
24697 @findex -var-evaluate-expression
24699 @subsubheading Synopsis
24702 -var-evaluate-expression [-f @var{format-spec}] @var{name}
24705 Evaluates the expression that is represented by the specified variable
24706 object and returns its value as a string. The format of the string
24707 can be specified with the @samp{-f} option. The possible values of
24708 this option are the same as for @code{-var-set-format}
24709 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
24710 the current display format will be used. The current display format
24711 can be changed using the @code{-var-set-format} command.
24717 Note that one must invoke @code{-var-list-children} for a variable
24718 before the value of a child variable can be evaluated.
24720 @subheading The @code{-var-assign} Command
24721 @findex -var-assign
24723 @subsubheading Synopsis
24726 -var-assign @var{name} @var{expression}
24729 Assigns the value of @var{expression} to the variable object specified
24730 by @var{name}. The object must be @samp{editable}. If the variable's
24731 value is altered by the assign, the variable will show up in any
24732 subsequent @code{-var-update} list.
24734 @subsubheading Example
24742 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
24746 @subheading The @code{-var-update} Command
24747 @findex -var-update
24749 @subsubheading Synopsis
24752 -var-update [@var{print-values}] @{@var{name} | "*"@}
24755 Reevaluate the expressions corresponding to the variable object
24756 @var{name} and all its direct and indirect children, and return the
24757 list of variable objects whose values have changed; @var{name} must
24758 be a root variable object. Here, ``changed'' means that the result of
24759 @code{-var-evaluate-expression} before and after the
24760 @code{-var-update} is different. If @samp{*} is used as the variable
24761 object names, all existing variable objects are updated, except
24762 for frozen ones (@pxref{-var-set-frozen}). The option
24763 @var{print-values} determines whether both names and values, or just
24764 names are printed. The possible values of this option are the same
24765 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
24766 recommended to use the @samp{--all-values} option, to reduce the
24767 number of MI commands needed on each program stop.
24769 With the @samp{*} parameter, if a variable object is bound to a
24770 currently running thread, it will not be updated, without any
24773 If @code{-var-set-update-range} was previously used on a varobj, then
24774 only the selected range of children will be reported.
24776 @code{-var-update} reports all the changed varobjs in a tuple named
24779 Each item in the change list is itself a tuple holding:
24783 The name of the varobj.
24786 If values were requested for this update, then this field will be
24787 present and will hold the value of the varobj.
24790 @anchor{-var-update}
24791 This field is a string which may take one of three values:
24795 The variable object's current value is valid.
24798 The variable object does not currently hold a valid value but it may
24799 hold one in the future if its associated expression comes back into
24803 The variable object no longer holds a valid value.
24804 This can occur when the executable file being debugged has changed,
24805 either through recompilation or by using the @value{GDBN} @code{file}
24806 command. The front end should normally choose to delete these variable
24810 In the future new values may be added to this list so the front should
24811 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
24814 This is only present if the varobj is still valid. If the type
24815 changed, then this will be the string @samp{true}; otherwise it will
24819 If the varobj's type changed, then this field will be present and will
24822 @item new_num_children
24823 For a dynamic varobj, if the number of children changed, or if the
24824 type changed, this will be the new number of children.
24826 The @samp{numchild} field in other varobj responses is generally not
24827 valid for a dynamic varobj -- it will show the number of children that
24828 @value{GDBN} knows about, but because dynamic varobjs lazily
24829 instantiate their children, this will not reflect the number of
24830 children which may be available.
24832 The @samp{new_num_children} attribute only reports changes to the
24833 number of children known by @value{GDBN}. This is the only way to
24834 detect whether an update has removed children (which necessarily can
24835 only happen at the end of the update range).
24838 The display hint, if any.
24841 This is an integer value, which will be 1 if there are more children
24842 available outside the varobj's update range.
24845 This attribute will be present and have the value @samp{1} if the
24846 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
24847 then this attribute will not be present.
24850 If new children were added to a dynamic varobj within the selected
24851 update range (as set by @code{-var-set-update-range}), then they will
24852 be listed in this attribute.
24855 @subsubheading Example
24862 -var-update --all-values var1
24863 ^done,changelist=[@{name="var1",value="3",in_scope="true",
24864 type_changed="false"@}]
24868 @subheading The @code{-var-set-frozen} Command
24869 @findex -var-set-frozen
24870 @anchor{-var-set-frozen}
24872 @subsubheading Synopsis
24875 -var-set-frozen @var{name} @var{flag}
24878 Set the frozenness flag on the variable object @var{name}. The
24879 @var{flag} parameter should be either @samp{1} to make the variable
24880 frozen or @samp{0} to make it unfrozen. If a variable object is
24881 frozen, then neither itself, nor any of its children, are
24882 implicitly updated by @code{-var-update} of
24883 a parent variable or by @code{-var-update *}. Only
24884 @code{-var-update} of the variable itself will update its value and
24885 values of its children. After a variable object is unfrozen, it is
24886 implicitly updated by all subsequent @code{-var-update} operations.
24887 Unfreezing a variable does not update it, only subsequent
24888 @code{-var-update} does.
24890 @subsubheading Example
24894 -var-set-frozen V 1
24899 @subheading The @code{-var-set-update-range} command
24900 @findex -var-set-update-range
24901 @anchor{-var-set-update-range}
24903 @subsubheading Synopsis
24906 -var-set-update-range @var{name} @var{from} @var{to}
24909 Set the range of children to be returned by future invocations of
24910 @code{-var-update}.
24912 @var{from} and @var{to} indicate the range of children to report. If
24913 @var{from} or @var{to} is less than zero, the range is reset and all
24914 children will be reported. Otherwise, children starting at @var{from}
24915 (zero-based) and up to and excluding @var{to} will be reported.
24917 @subsubheading Example
24921 -var-set-update-range V 1 2
24925 @subheading The @code{-var-set-visualizer} command
24926 @findex -var-set-visualizer
24927 @anchor{-var-set-visualizer}
24929 @subsubheading Synopsis
24932 -var-set-visualizer @var{name} @var{visualizer}
24935 Set a visualizer for the variable object @var{name}.
24937 @var{visualizer} is the visualizer to use. The special value
24938 @samp{None} means to disable any visualizer in use.
24940 If not @samp{None}, @var{visualizer} must be a Python expression.
24941 This expression must evaluate to a callable object which accepts a
24942 single argument. @value{GDBN} will call this object with the value of
24943 the varobj @var{name} as an argument (this is done so that the same
24944 Python pretty-printing code can be used for both the CLI and MI).
24945 When called, this object must return an object which conforms to the
24946 pretty-printing interface (@pxref{Pretty Printing}).
24948 The pre-defined function @code{gdb.default_visualizer} may be used to
24949 select a visualizer by following the built-in process
24950 (@pxref{Selecting Pretty-Printers}). This is done automatically when
24951 a varobj is created, and so ordinarily is not needed.
24953 This feature is only available if Python support is enabled. The MI
24954 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
24955 can be used to check this.
24957 @subsubheading Example
24959 Resetting the visualizer:
24963 -var-set-visualizer V None
24967 Reselecting the default (type-based) visualizer:
24971 -var-set-visualizer V gdb.default_visualizer
24975 Suppose @code{SomeClass} is a visualizer class. A lambda expression
24976 can be used to instantiate this class for a varobj:
24980 -var-set-visualizer V "lambda val: SomeClass()"
24984 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24985 @node GDB/MI Data Manipulation
24986 @section @sc{gdb/mi} Data Manipulation
24988 @cindex data manipulation, in @sc{gdb/mi}
24989 @cindex @sc{gdb/mi}, data manipulation
24990 This section describes the @sc{gdb/mi} commands that manipulate data:
24991 examine memory and registers, evaluate expressions, etc.
24993 @c REMOVED FROM THE INTERFACE.
24994 @c @subheading -data-assign
24995 @c Change the value of a program variable. Plenty of side effects.
24996 @c @subsubheading GDB Command
24998 @c @subsubheading Example
25001 @subheading The @code{-data-disassemble} Command
25002 @findex -data-disassemble
25004 @subsubheading Synopsis
25008 [ -s @var{start-addr} -e @var{end-addr} ]
25009 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
25017 @item @var{start-addr}
25018 is the beginning address (or @code{$pc})
25019 @item @var{end-addr}
25021 @item @var{filename}
25022 is the name of the file to disassemble
25023 @item @var{linenum}
25024 is the line number to disassemble around
25026 is the number of disassembly lines to be produced. If it is -1,
25027 the whole function will be disassembled, in case no @var{end-addr} is
25028 specified. If @var{end-addr} is specified as a non-zero value, and
25029 @var{lines} is lower than the number of disassembly lines between
25030 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
25031 displayed; if @var{lines} is higher than the number of lines between
25032 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
25035 is either 0 (meaning only disassembly) or 1 (meaning mixed source and
25039 @subsubheading Result
25041 The output for each instruction is composed of four fields:
25050 Note that whatever included in the instruction field, is not manipulated
25051 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
25053 @subsubheading @value{GDBN} Command
25055 There's no direct mapping from this command to the CLI.
25057 @subsubheading Example
25059 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
25063 -data-disassemble -s $pc -e "$pc + 20" -- 0
25066 @{address="0x000107c0",func-name="main",offset="4",
25067 inst="mov 2, %o0"@},
25068 @{address="0x000107c4",func-name="main",offset="8",
25069 inst="sethi %hi(0x11800), %o2"@},
25070 @{address="0x000107c8",func-name="main",offset="12",
25071 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
25072 @{address="0x000107cc",func-name="main",offset="16",
25073 inst="sethi %hi(0x11800), %o2"@},
25074 @{address="0x000107d0",func-name="main",offset="20",
25075 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
25079 Disassemble the whole @code{main} function. Line 32 is part of
25083 -data-disassemble -f basics.c -l 32 -- 0
25085 @{address="0x000107bc",func-name="main",offset="0",
25086 inst="save %sp, -112, %sp"@},
25087 @{address="0x000107c0",func-name="main",offset="4",
25088 inst="mov 2, %o0"@},
25089 @{address="0x000107c4",func-name="main",offset="8",
25090 inst="sethi %hi(0x11800), %o2"@},
25092 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
25093 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
25097 Disassemble 3 instructions from the start of @code{main}:
25101 -data-disassemble -f basics.c -l 32 -n 3 -- 0
25103 @{address="0x000107bc",func-name="main",offset="0",
25104 inst="save %sp, -112, %sp"@},
25105 @{address="0x000107c0",func-name="main",offset="4",
25106 inst="mov 2, %o0"@},
25107 @{address="0x000107c4",func-name="main",offset="8",
25108 inst="sethi %hi(0x11800), %o2"@}]
25112 Disassemble 3 instructions from the start of @code{main} in mixed mode:
25116 -data-disassemble -f basics.c -l 32 -n 3 -- 1
25118 src_and_asm_line=@{line="31",
25119 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
25120 testsuite/gdb.mi/basics.c",line_asm_insn=[
25121 @{address="0x000107bc",func-name="main",offset="0",
25122 inst="save %sp, -112, %sp"@}]@},
25123 src_and_asm_line=@{line="32",
25124 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
25125 testsuite/gdb.mi/basics.c",line_asm_insn=[
25126 @{address="0x000107c0",func-name="main",offset="4",
25127 inst="mov 2, %o0"@},
25128 @{address="0x000107c4",func-name="main",offset="8",
25129 inst="sethi %hi(0x11800), %o2"@}]@}]
25134 @subheading The @code{-data-evaluate-expression} Command
25135 @findex -data-evaluate-expression
25137 @subsubheading Synopsis
25140 -data-evaluate-expression @var{expr}
25143 Evaluate @var{expr} as an expression. The expression could contain an
25144 inferior function call. The function call will execute synchronously.
25145 If the expression contains spaces, it must be enclosed in double quotes.
25147 @subsubheading @value{GDBN} Command
25149 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
25150 @samp{call}. In @code{gdbtk} only, there's a corresponding
25151 @samp{gdb_eval} command.
25153 @subsubheading Example
25155 In the following example, the numbers that precede the commands are the
25156 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
25157 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
25161 211-data-evaluate-expression A
25164 311-data-evaluate-expression &A
25165 311^done,value="0xefffeb7c"
25167 411-data-evaluate-expression A+3
25170 511-data-evaluate-expression "A + 3"
25176 @subheading The @code{-data-list-changed-registers} Command
25177 @findex -data-list-changed-registers
25179 @subsubheading Synopsis
25182 -data-list-changed-registers
25185 Display a list of the registers that have changed.
25187 @subsubheading @value{GDBN} Command
25189 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
25190 has the corresponding command @samp{gdb_changed_register_list}.
25192 @subsubheading Example
25194 On a PPC MBX board:
25202 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
25203 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
25206 -data-list-changed-registers
25207 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
25208 "10","11","13","14","15","16","17","18","19","20","21","22","23",
25209 "24","25","26","27","28","30","31","64","65","66","67","69"]
25214 @subheading The @code{-data-list-register-names} Command
25215 @findex -data-list-register-names
25217 @subsubheading Synopsis
25220 -data-list-register-names [ ( @var{regno} )+ ]
25223 Show a list of register names for the current target. If no arguments
25224 are given, it shows a list of the names of all the registers. If
25225 integer numbers are given as arguments, it will print a list of the
25226 names of the registers corresponding to the arguments. To ensure
25227 consistency between a register name and its number, the output list may
25228 include empty register names.
25230 @subsubheading @value{GDBN} Command
25232 @value{GDBN} does not have a command which corresponds to
25233 @samp{-data-list-register-names}. In @code{gdbtk} there is a
25234 corresponding command @samp{gdb_regnames}.
25236 @subsubheading Example
25238 For the PPC MBX board:
25241 -data-list-register-names
25242 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
25243 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
25244 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
25245 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
25246 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
25247 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
25248 "", "pc","ps","cr","lr","ctr","xer"]
25250 -data-list-register-names 1 2 3
25251 ^done,register-names=["r1","r2","r3"]
25255 @subheading The @code{-data-list-register-values} Command
25256 @findex -data-list-register-values
25258 @subsubheading Synopsis
25261 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
25264 Display the registers' contents. @var{fmt} is the format according to
25265 which the registers' contents are to be returned, followed by an optional
25266 list of numbers specifying the registers to display. A missing list of
25267 numbers indicates that the contents of all the registers must be returned.
25269 Allowed formats for @var{fmt} are:
25286 @subsubheading @value{GDBN} Command
25288 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
25289 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
25291 @subsubheading Example
25293 For a PPC MBX board (note: line breaks are for readability only, they
25294 don't appear in the actual output):
25298 -data-list-register-values r 64 65
25299 ^done,register-values=[@{number="64",value="0xfe00a300"@},
25300 @{number="65",value="0x00029002"@}]
25302 -data-list-register-values x
25303 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
25304 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
25305 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
25306 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
25307 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
25308 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
25309 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
25310 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
25311 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
25312 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
25313 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
25314 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
25315 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
25316 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
25317 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
25318 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
25319 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
25320 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
25321 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
25322 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
25323 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
25324 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
25325 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
25326 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
25327 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
25328 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
25329 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
25330 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
25331 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
25332 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
25333 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
25334 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
25335 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
25336 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
25337 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
25338 @{number="69",value="0x20002b03"@}]
25343 @subheading The @code{-data-read-memory} Command
25344 @findex -data-read-memory
25346 @subsubheading Synopsis
25349 -data-read-memory [ -o @var{byte-offset} ]
25350 @var{address} @var{word-format} @var{word-size}
25351 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
25358 @item @var{address}
25359 An expression specifying the address of the first memory word to be
25360 read. Complex expressions containing embedded white space should be
25361 quoted using the C convention.
25363 @item @var{word-format}
25364 The format to be used to print the memory words. The notation is the
25365 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
25368 @item @var{word-size}
25369 The size of each memory word in bytes.
25371 @item @var{nr-rows}
25372 The number of rows in the output table.
25374 @item @var{nr-cols}
25375 The number of columns in the output table.
25378 If present, indicates that each row should include an @sc{ascii} dump. The
25379 value of @var{aschar} is used as a padding character when a byte is not a
25380 member of the printable @sc{ascii} character set (printable @sc{ascii}
25381 characters are those whose code is between 32 and 126, inclusively).
25383 @item @var{byte-offset}
25384 An offset to add to the @var{address} before fetching memory.
25387 This command displays memory contents as a table of @var{nr-rows} by
25388 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
25389 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
25390 (returned as @samp{total-bytes}). Should less than the requested number
25391 of bytes be returned by the target, the missing words are identified
25392 using @samp{N/A}. The number of bytes read from the target is returned
25393 in @samp{nr-bytes} and the starting address used to read memory in
25396 The address of the next/previous row or page is available in
25397 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
25400 @subsubheading @value{GDBN} Command
25402 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
25403 @samp{gdb_get_mem} memory read command.
25405 @subsubheading Example
25407 Read six bytes of memory starting at @code{bytes+6} but then offset by
25408 @code{-6} bytes. Format as three rows of two columns. One byte per
25409 word. Display each word in hex.
25413 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
25414 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
25415 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
25416 prev-page="0x0000138a",memory=[
25417 @{addr="0x00001390",data=["0x00","0x01"]@},
25418 @{addr="0x00001392",data=["0x02","0x03"]@},
25419 @{addr="0x00001394",data=["0x04","0x05"]@}]
25423 Read two bytes of memory starting at address @code{shorts + 64} and
25424 display as a single word formatted in decimal.
25428 5-data-read-memory shorts+64 d 2 1 1
25429 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
25430 next-row="0x00001512",prev-row="0x0000150e",
25431 next-page="0x00001512",prev-page="0x0000150e",memory=[
25432 @{addr="0x00001510",data=["128"]@}]
25436 Read thirty two bytes of memory starting at @code{bytes+16} and format
25437 as eight rows of four columns. Include a string encoding with @samp{x}
25438 used as the non-printable character.
25442 4-data-read-memory bytes+16 x 1 8 4 x
25443 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
25444 next-row="0x000013c0",prev-row="0x0000139c",
25445 next-page="0x000013c0",prev-page="0x00001380",memory=[
25446 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
25447 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
25448 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
25449 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
25450 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
25451 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
25452 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
25453 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
25457 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25458 @node GDB/MI Tracepoint Commands
25459 @section @sc{gdb/mi} Tracepoint Commands
25461 The tracepoint commands are not yet implemented.
25463 @c @subheading -trace-actions
25465 @c @subheading -trace-delete
25467 @c @subheading -trace-disable
25469 @c @subheading -trace-dump
25471 @c @subheading -trace-enable
25473 @c @subheading -trace-exists
25475 @c @subheading -trace-find
25477 @c @subheading -trace-frame-number
25479 @c @subheading -trace-info
25481 @c @subheading -trace-insert
25483 @c @subheading -trace-list
25485 @c @subheading -trace-pass-count
25487 @c @subheading -trace-save
25489 @c @subheading -trace-start
25491 @c @subheading -trace-stop
25494 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25495 @node GDB/MI Symbol Query
25496 @section @sc{gdb/mi} Symbol Query Commands
25500 @subheading The @code{-symbol-info-address} Command
25501 @findex -symbol-info-address
25503 @subsubheading Synopsis
25506 -symbol-info-address @var{symbol}
25509 Describe where @var{symbol} is stored.
25511 @subsubheading @value{GDBN} Command
25513 The corresponding @value{GDBN} command is @samp{info address}.
25515 @subsubheading Example
25519 @subheading The @code{-symbol-info-file} Command
25520 @findex -symbol-info-file
25522 @subsubheading Synopsis
25528 Show the file for the symbol.
25530 @subsubheading @value{GDBN} Command
25532 There's no equivalent @value{GDBN} command. @code{gdbtk} has
25533 @samp{gdb_find_file}.
25535 @subsubheading Example
25539 @subheading The @code{-symbol-info-function} Command
25540 @findex -symbol-info-function
25542 @subsubheading Synopsis
25545 -symbol-info-function
25548 Show which function the symbol lives in.
25550 @subsubheading @value{GDBN} Command
25552 @samp{gdb_get_function} in @code{gdbtk}.
25554 @subsubheading Example
25558 @subheading The @code{-symbol-info-line} Command
25559 @findex -symbol-info-line
25561 @subsubheading Synopsis
25567 Show the core addresses of the code for a source line.
25569 @subsubheading @value{GDBN} Command
25571 The corresponding @value{GDBN} command is @samp{info line}.
25572 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
25574 @subsubheading Example
25578 @subheading The @code{-symbol-info-symbol} Command
25579 @findex -symbol-info-symbol
25581 @subsubheading Synopsis
25584 -symbol-info-symbol @var{addr}
25587 Describe what symbol is at location @var{addr}.
25589 @subsubheading @value{GDBN} Command
25591 The corresponding @value{GDBN} command is @samp{info symbol}.
25593 @subsubheading Example
25597 @subheading The @code{-symbol-list-functions} Command
25598 @findex -symbol-list-functions
25600 @subsubheading Synopsis
25603 -symbol-list-functions
25606 List the functions in the executable.
25608 @subsubheading @value{GDBN} Command
25610 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
25611 @samp{gdb_search} in @code{gdbtk}.
25613 @subsubheading Example
25618 @subheading The @code{-symbol-list-lines} Command
25619 @findex -symbol-list-lines
25621 @subsubheading Synopsis
25624 -symbol-list-lines @var{filename}
25627 Print the list of lines that contain code and their associated program
25628 addresses for the given source filename. The entries are sorted in
25629 ascending PC order.
25631 @subsubheading @value{GDBN} Command
25633 There is no corresponding @value{GDBN} command.
25635 @subsubheading Example
25638 -symbol-list-lines basics.c
25639 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
25645 @subheading The @code{-symbol-list-types} Command
25646 @findex -symbol-list-types
25648 @subsubheading Synopsis
25654 List all the type names.
25656 @subsubheading @value{GDBN} Command
25658 The corresponding commands are @samp{info types} in @value{GDBN},
25659 @samp{gdb_search} in @code{gdbtk}.
25661 @subsubheading Example
25665 @subheading The @code{-symbol-list-variables} Command
25666 @findex -symbol-list-variables
25668 @subsubheading Synopsis
25671 -symbol-list-variables
25674 List all the global and static variable names.
25676 @subsubheading @value{GDBN} Command
25678 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
25680 @subsubheading Example
25684 @subheading The @code{-symbol-locate} Command
25685 @findex -symbol-locate
25687 @subsubheading Synopsis
25693 @subsubheading @value{GDBN} Command
25695 @samp{gdb_loc} in @code{gdbtk}.
25697 @subsubheading Example
25701 @subheading The @code{-symbol-type} Command
25702 @findex -symbol-type
25704 @subsubheading Synopsis
25707 -symbol-type @var{variable}
25710 Show type of @var{variable}.
25712 @subsubheading @value{GDBN} Command
25714 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
25715 @samp{gdb_obj_variable}.
25717 @subsubheading Example
25722 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25723 @node GDB/MI File Commands
25724 @section @sc{gdb/mi} File Commands
25726 This section describes the GDB/MI commands to specify executable file names
25727 and to read in and obtain symbol table information.
25729 @subheading The @code{-file-exec-and-symbols} Command
25730 @findex -file-exec-and-symbols
25732 @subsubheading Synopsis
25735 -file-exec-and-symbols @var{file}
25738 Specify the executable file to be debugged. This file is the one from
25739 which the symbol table is also read. If no file is specified, the
25740 command clears the executable and symbol information. If breakpoints
25741 are set when using this command with no arguments, @value{GDBN} will produce
25742 error messages. Otherwise, no output is produced, except a completion
25745 @subsubheading @value{GDBN} Command
25747 The corresponding @value{GDBN} command is @samp{file}.
25749 @subsubheading Example
25753 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
25759 @subheading The @code{-file-exec-file} Command
25760 @findex -file-exec-file
25762 @subsubheading Synopsis
25765 -file-exec-file @var{file}
25768 Specify the executable file to be debugged. Unlike
25769 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
25770 from this file. If used without argument, @value{GDBN} clears the information
25771 about the executable file. No output is produced, except a completion
25774 @subsubheading @value{GDBN} Command
25776 The corresponding @value{GDBN} command is @samp{exec-file}.
25778 @subsubheading Example
25782 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
25789 @subheading The @code{-file-list-exec-sections} Command
25790 @findex -file-list-exec-sections
25792 @subsubheading Synopsis
25795 -file-list-exec-sections
25798 List the sections of the current executable file.
25800 @subsubheading @value{GDBN} Command
25802 The @value{GDBN} command @samp{info file} shows, among the rest, the same
25803 information as this command. @code{gdbtk} has a corresponding command
25804 @samp{gdb_load_info}.
25806 @subsubheading Example
25811 @subheading The @code{-file-list-exec-source-file} Command
25812 @findex -file-list-exec-source-file
25814 @subsubheading Synopsis
25817 -file-list-exec-source-file
25820 List the line number, the current source file, and the absolute path
25821 to the current source file for the current executable. The macro
25822 information field has a value of @samp{1} or @samp{0} depending on
25823 whether or not the file includes preprocessor macro information.
25825 @subsubheading @value{GDBN} Command
25827 The @value{GDBN} equivalent is @samp{info source}
25829 @subsubheading Example
25833 123-file-list-exec-source-file
25834 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
25839 @subheading The @code{-file-list-exec-source-files} Command
25840 @findex -file-list-exec-source-files
25842 @subsubheading Synopsis
25845 -file-list-exec-source-files
25848 List the source files for the current executable.
25850 It will always output the filename, but only when @value{GDBN} can find
25851 the absolute file name of a source file, will it output the fullname.
25853 @subsubheading @value{GDBN} Command
25855 The @value{GDBN} equivalent is @samp{info sources}.
25856 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
25858 @subsubheading Example
25861 -file-list-exec-source-files
25863 @{file=foo.c,fullname=/home/foo.c@},
25864 @{file=/home/bar.c,fullname=/home/bar.c@},
25865 @{file=gdb_could_not_find_fullpath.c@}]
25870 @subheading The @code{-file-list-shared-libraries} Command
25871 @findex -file-list-shared-libraries
25873 @subsubheading Synopsis
25876 -file-list-shared-libraries
25879 List the shared libraries in the program.
25881 @subsubheading @value{GDBN} Command
25883 The corresponding @value{GDBN} command is @samp{info shared}.
25885 @subsubheading Example
25889 @subheading The @code{-file-list-symbol-files} Command
25890 @findex -file-list-symbol-files
25892 @subsubheading Synopsis
25895 -file-list-symbol-files
25900 @subsubheading @value{GDBN} Command
25902 The corresponding @value{GDBN} command is @samp{info file} (part of it).
25904 @subsubheading Example
25909 @subheading The @code{-file-symbol-file} Command
25910 @findex -file-symbol-file
25912 @subsubheading Synopsis
25915 -file-symbol-file @var{file}
25918 Read symbol table info from the specified @var{file} argument. When
25919 used without arguments, clears @value{GDBN}'s symbol table info. No output is
25920 produced, except for a completion notification.
25922 @subsubheading @value{GDBN} Command
25924 The corresponding @value{GDBN} command is @samp{symbol-file}.
25926 @subsubheading Example
25930 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
25936 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25937 @node GDB/MI Memory Overlay Commands
25938 @section @sc{gdb/mi} Memory Overlay Commands
25940 The memory overlay commands are not implemented.
25942 @c @subheading -overlay-auto
25944 @c @subheading -overlay-list-mapping-state
25946 @c @subheading -overlay-list-overlays
25948 @c @subheading -overlay-map
25950 @c @subheading -overlay-off
25952 @c @subheading -overlay-on
25954 @c @subheading -overlay-unmap
25956 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25957 @node GDB/MI Signal Handling Commands
25958 @section @sc{gdb/mi} Signal Handling Commands
25960 Signal handling commands are not implemented.
25962 @c @subheading -signal-handle
25964 @c @subheading -signal-list-handle-actions
25966 @c @subheading -signal-list-signal-types
25970 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25971 @node GDB/MI Target Manipulation
25972 @section @sc{gdb/mi} Target Manipulation Commands
25975 @subheading The @code{-target-attach} Command
25976 @findex -target-attach
25978 @subsubheading Synopsis
25981 -target-attach @var{pid} | @var{gid} | @var{file}
25984 Attach to a process @var{pid} or a file @var{file} outside of
25985 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
25986 group, the id previously returned by
25987 @samp{-list-thread-groups --available} must be used.
25989 @subsubheading @value{GDBN} Command
25991 The corresponding @value{GDBN} command is @samp{attach}.
25993 @subsubheading Example
25997 =thread-created,id="1"
25998 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
26004 @subheading The @code{-target-compare-sections} Command
26005 @findex -target-compare-sections
26007 @subsubheading Synopsis
26010 -target-compare-sections [ @var{section} ]
26013 Compare data of section @var{section} on target to the exec file.
26014 Without the argument, all sections are compared.
26016 @subsubheading @value{GDBN} Command
26018 The @value{GDBN} equivalent is @samp{compare-sections}.
26020 @subsubheading Example
26025 @subheading The @code{-target-detach} Command
26026 @findex -target-detach
26028 @subsubheading Synopsis
26031 -target-detach [ @var{pid} | @var{gid} ]
26034 Detach from the remote target which normally resumes its execution.
26035 If either @var{pid} or @var{gid} is specified, detaches from either
26036 the specified process, or specified thread group. There's no output.
26038 @subsubheading @value{GDBN} Command
26040 The corresponding @value{GDBN} command is @samp{detach}.
26042 @subsubheading Example
26052 @subheading The @code{-target-disconnect} Command
26053 @findex -target-disconnect
26055 @subsubheading Synopsis
26061 Disconnect from the remote target. There's no output and the target is
26062 generally not resumed.
26064 @subsubheading @value{GDBN} Command
26066 The corresponding @value{GDBN} command is @samp{disconnect}.
26068 @subsubheading Example
26078 @subheading The @code{-target-download} Command
26079 @findex -target-download
26081 @subsubheading Synopsis
26087 Loads the executable onto the remote target.
26088 It prints out an update message every half second, which includes the fields:
26092 The name of the section.
26094 The size of what has been sent so far for that section.
26096 The size of the section.
26098 The total size of what was sent so far (the current and the previous sections).
26100 The size of the overall executable to download.
26104 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
26105 @sc{gdb/mi} Output Syntax}).
26107 In addition, it prints the name and size of the sections, as they are
26108 downloaded. These messages include the following fields:
26112 The name of the section.
26114 The size of the section.
26116 The size of the overall executable to download.
26120 At the end, a summary is printed.
26122 @subsubheading @value{GDBN} Command
26124 The corresponding @value{GDBN} command is @samp{load}.
26126 @subsubheading Example
26128 Note: each status message appears on a single line. Here the messages
26129 have been broken down so that they can fit onto a page.
26134 +download,@{section=".text",section-size="6668",total-size="9880"@}
26135 +download,@{section=".text",section-sent="512",section-size="6668",
26136 total-sent="512",total-size="9880"@}
26137 +download,@{section=".text",section-sent="1024",section-size="6668",
26138 total-sent="1024",total-size="9880"@}
26139 +download,@{section=".text",section-sent="1536",section-size="6668",
26140 total-sent="1536",total-size="9880"@}
26141 +download,@{section=".text",section-sent="2048",section-size="6668",
26142 total-sent="2048",total-size="9880"@}
26143 +download,@{section=".text",section-sent="2560",section-size="6668",
26144 total-sent="2560",total-size="9880"@}
26145 +download,@{section=".text",section-sent="3072",section-size="6668",
26146 total-sent="3072",total-size="9880"@}
26147 +download,@{section=".text",section-sent="3584",section-size="6668",
26148 total-sent="3584",total-size="9880"@}
26149 +download,@{section=".text",section-sent="4096",section-size="6668",
26150 total-sent="4096",total-size="9880"@}
26151 +download,@{section=".text",section-sent="4608",section-size="6668",
26152 total-sent="4608",total-size="9880"@}
26153 +download,@{section=".text",section-sent="5120",section-size="6668",
26154 total-sent="5120",total-size="9880"@}
26155 +download,@{section=".text",section-sent="5632",section-size="6668",
26156 total-sent="5632",total-size="9880"@}
26157 +download,@{section=".text",section-sent="6144",section-size="6668",
26158 total-sent="6144",total-size="9880"@}
26159 +download,@{section=".text",section-sent="6656",section-size="6668",
26160 total-sent="6656",total-size="9880"@}
26161 +download,@{section=".init",section-size="28",total-size="9880"@}
26162 +download,@{section=".fini",section-size="28",total-size="9880"@}
26163 +download,@{section=".data",section-size="3156",total-size="9880"@}
26164 +download,@{section=".data",section-sent="512",section-size="3156",
26165 total-sent="7236",total-size="9880"@}
26166 +download,@{section=".data",section-sent="1024",section-size="3156",
26167 total-sent="7748",total-size="9880"@}
26168 +download,@{section=".data",section-sent="1536",section-size="3156",
26169 total-sent="8260",total-size="9880"@}
26170 +download,@{section=".data",section-sent="2048",section-size="3156",
26171 total-sent="8772",total-size="9880"@}
26172 +download,@{section=".data",section-sent="2560",section-size="3156",
26173 total-sent="9284",total-size="9880"@}
26174 +download,@{section=".data",section-sent="3072",section-size="3156",
26175 total-sent="9796",total-size="9880"@}
26176 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
26183 @subheading The @code{-target-exec-status} Command
26184 @findex -target-exec-status
26186 @subsubheading Synopsis
26189 -target-exec-status
26192 Provide information on the state of the target (whether it is running or
26193 not, for instance).
26195 @subsubheading @value{GDBN} Command
26197 There's no equivalent @value{GDBN} command.
26199 @subsubheading Example
26203 @subheading The @code{-target-list-available-targets} Command
26204 @findex -target-list-available-targets
26206 @subsubheading Synopsis
26209 -target-list-available-targets
26212 List the possible targets to connect to.
26214 @subsubheading @value{GDBN} Command
26216 The corresponding @value{GDBN} command is @samp{help target}.
26218 @subsubheading Example
26222 @subheading The @code{-target-list-current-targets} Command
26223 @findex -target-list-current-targets
26225 @subsubheading Synopsis
26228 -target-list-current-targets
26231 Describe the current target.
26233 @subsubheading @value{GDBN} Command
26235 The corresponding information is printed by @samp{info file} (among
26238 @subsubheading Example
26242 @subheading The @code{-target-list-parameters} Command
26243 @findex -target-list-parameters
26245 @subsubheading Synopsis
26248 -target-list-parameters
26254 @subsubheading @value{GDBN} Command
26258 @subsubheading Example
26262 @subheading The @code{-target-select} Command
26263 @findex -target-select
26265 @subsubheading Synopsis
26268 -target-select @var{type} @var{parameters @dots{}}
26271 Connect @value{GDBN} to the remote target. This command takes two args:
26275 The type of target, for instance @samp{remote}, etc.
26276 @item @var{parameters}
26277 Device names, host names and the like. @xref{Target Commands, ,
26278 Commands for Managing Targets}, for more details.
26281 The output is a connection notification, followed by the address at
26282 which the target program is, in the following form:
26285 ^connected,addr="@var{address}",func="@var{function name}",
26286 args=[@var{arg list}]
26289 @subsubheading @value{GDBN} Command
26291 The corresponding @value{GDBN} command is @samp{target}.
26293 @subsubheading Example
26297 -target-select remote /dev/ttya
26298 ^connected,addr="0xfe00a300",func="??",args=[]
26302 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26303 @node GDB/MI File Transfer Commands
26304 @section @sc{gdb/mi} File Transfer Commands
26307 @subheading The @code{-target-file-put} Command
26308 @findex -target-file-put
26310 @subsubheading Synopsis
26313 -target-file-put @var{hostfile} @var{targetfile}
26316 Copy file @var{hostfile} from the host system (the machine running
26317 @value{GDBN}) to @var{targetfile} on the target system.
26319 @subsubheading @value{GDBN} Command
26321 The corresponding @value{GDBN} command is @samp{remote put}.
26323 @subsubheading Example
26327 -target-file-put localfile remotefile
26333 @subheading The @code{-target-file-get} Command
26334 @findex -target-file-get
26336 @subsubheading Synopsis
26339 -target-file-get @var{targetfile} @var{hostfile}
26342 Copy file @var{targetfile} from the target system to @var{hostfile}
26343 on the host system.
26345 @subsubheading @value{GDBN} Command
26347 The corresponding @value{GDBN} command is @samp{remote get}.
26349 @subsubheading Example
26353 -target-file-get remotefile localfile
26359 @subheading The @code{-target-file-delete} Command
26360 @findex -target-file-delete
26362 @subsubheading Synopsis
26365 -target-file-delete @var{targetfile}
26368 Delete @var{targetfile} from the target system.
26370 @subsubheading @value{GDBN} Command
26372 The corresponding @value{GDBN} command is @samp{remote delete}.
26374 @subsubheading Example
26378 -target-file-delete remotefile
26384 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26385 @node GDB/MI Miscellaneous Commands
26386 @section Miscellaneous @sc{gdb/mi} Commands
26388 @c @subheading -gdb-complete
26390 @subheading The @code{-gdb-exit} Command
26393 @subsubheading Synopsis
26399 Exit @value{GDBN} immediately.
26401 @subsubheading @value{GDBN} Command
26403 Approximately corresponds to @samp{quit}.
26405 @subsubheading Example
26415 @subheading The @code{-exec-abort} Command
26416 @findex -exec-abort
26418 @subsubheading Synopsis
26424 Kill the inferior running program.
26426 @subsubheading @value{GDBN} Command
26428 The corresponding @value{GDBN} command is @samp{kill}.
26430 @subsubheading Example
26435 @subheading The @code{-gdb-set} Command
26438 @subsubheading Synopsis
26444 Set an internal @value{GDBN} variable.
26445 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
26447 @subsubheading @value{GDBN} Command
26449 The corresponding @value{GDBN} command is @samp{set}.
26451 @subsubheading Example
26461 @subheading The @code{-gdb-show} Command
26464 @subsubheading Synopsis
26470 Show the current value of a @value{GDBN} variable.
26472 @subsubheading @value{GDBN} Command
26474 The corresponding @value{GDBN} command is @samp{show}.
26476 @subsubheading Example
26485 @c @subheading -gdb-source
26488 @subheading The @code{-gdb-version} Command
26489 @findex -gdb-version
26491 @subsubheading Synopsis
26497 Show version information for @value{GDBN}. Used mostly in testing.
26499 @subsubheading @value{GDBN} Command
26501 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
26502 default shows this information when you start an interactive session.
26504 @subsubheading Example
26506 @c This example modifies the actual output from GDB to avoid overfull
26512 ~Copyright 2000 Free Software Foundation, Inc.
26513 ~GDB is free software, covered by the GNU General Public License, and
26514 ~you are welcome to change it and/or distribute copies of it under
26515 ~ certain conditions.
26516 ~Type "show copying" to see the conditions.
26517 ~There is absolutely no warranty for GDB. Type "show warranty" for
26519 ~This GDB was configured as
26520 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
26525 @subheading The @code{-list-features} Command
26526 @findex -list-features
26528 Returns a list of particular features of the MI protocol that
26529 this version of gdb implements. A feature can be a command,
26530 or a new field in an output of some command, or even an
26531 important bugfix. While a frontend can sometimes detect presence
26532 of a feature at runtime, it is easier to perform detection at debugger
26535 The command returns a list of strings, with each string naming an
26536 available feature. Each returned string is just a name, it does not
26537 have any internal structure. The list of possible feature names
26543 (gdb) -list-features
26544 ^done,result=["feature1","feature2"]
26547 The current list of features is:
26550 @item frozen-varobjs
26551 Indicates presence of the @code{-var-set-frozen} command, as well
26552 as possible presense of the @code{frozen} field in the output
26553 of @code{-varobj-create}.
26554 @item pending-breakpoints
26555 Indicates presence of the @option{-f} option to the @code{-break-insert} command.
26557 Indicates presence of Python scripting support, Python-based
26558 pretty-printing commands, and possible presence of the
26559 @samp{display_hint} field in the output of @code{-var-list-children}
26561 Indicates presence of the @code{-thread-info} command.
26565 @subheading The @code{-list-target-features} Command
26566 @findex -list-target-features
26568 Returns a list of particular features that are supported by the
26569 target. Those features affect the permitted MI commands, but
26570 unlike the features reported by the @code{-list-features} command, the
26571 features depend on which target GDB is using at the moment. Whenever
26572 a target can change, due to commands such as @code{-target-select},
26573 @code{-target-attach} or @code{-exec-run}, the list of target features
26574 may change, and the frontend should obtain it again.
26578 (gdb) -list-features
26579 ^done,result=["async"]
26582 The current list of features is:
26586 Indicates that the target is capable of asynchronous command
26587 execution, which means that @value{GDBN} will accept further commands
26588 while the target is running.
26592 @subheading The @code{-list-thread-groups} Command
26593 @findex -list-thread-groups
26595 @subheading Synopsis
26598 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
26601 Lists thread groups (@pxref{Thread groups}). When a single thread
26602 group is passed as the argument, lists the children of that group.
26603 When several thread group are passed, lists information about those
26604 thread groups. Without any parameters, lists information about all
26605 top-level thread groups.
26607 Normally, thread groups that are being debugged are reported.
26608 With the @samp{--available} option, @value{GDBN} reports thread groups
26609 available on the target.
26611 The output of this command may have either a @samp{threads} result or
26612 a @samp{groups} result. The @samp{thread} result has a list of tuples
26613 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
26614 Information}). The @samp{groups} result has a list of tuples as value,
26615 each tuple describing a thread group. If top-level groups are
26616 requested (that is, no parameter is passed), or when several groups
26617 are passed, the output always has a @samp{groups} result. The format
26618 of the @samp{group} result is described below.
26620 To reduce the number of roundtrips it's possible to list thread groups
26621 together with their children, by passing the @samp{--recurse} option
26622 and the recursion depth. Presently, only recursion depth of 1 is
26623 permitted. If this option is present, then every reported thread group
26624 will also include its children, either as @samp{group} or
26625 @samp{threads} field.
26627 In general, any combination of option and parameters is permitted, with
26628 the following caveats:
26632 When a single thread group is passed, the output will typically
26633 be the @samp{threads} result. Because threads may not contain
26634 anything, the @samp{recurse} option will be ignored.
26637 When the @samp{--available} option is passed, limited information may
26638 be available. In particular, the list of threads of a process might
26639 be inaccessible. Further, specifying specific thread groups might
26640 not give any performance advantage over listing all thread groups.
26641 The frontend should assume that @samp{-list-thread-groups --available}
26642 is always an expensive operation and cache the results.
26646 The @samp{groups} result is a list of tuples, where each tuple may
26647 have the following fields:
26651 Identifier of the thread group. This field is always present.
26654 The type of the thread group. At present, only @samp{process} is a
26658 The target-specific process identifier. This field is only present
26659 for thread groups of type @samp{process}.
26662 The number of children this thread group has. This field may be
26663 absent for an available thread group.
26666 This field has a list of tuples as value, each tuple describing a
26667 thread. It may be present if the @samp{--recurse} option is
26668 specified, and it's actually possible to obtain the threads.
26671 This field is a list of integers, each identifying a core that one
26672 thread of the group is running on. This field may be absent if
26673 such information is not available.
26677 @subheading Example
26681 -list-thread-groups
26682 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
26683 -list-thread-groups 17
26684 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
26685 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
26686 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
26687 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
26688 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
26689 -list-thread-groups --available
26690 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
26691 -list-thread-groups --available --recurse 1
26692 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
26693 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
26694 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
26695 -list-thread-groups --available --recurse 1 17 18
26696 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
26697 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
26698 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
26701 @subheading The @code{-interpreter-exec} Command
26702 @findex -interpreter-exec
26704 @subheading Synopsis
26707 -interpreter-exec @var{interpreter} @var{command}
26709 @anchor{-interpreter-exec}
26711 Execute the specified @var{command} in the given @var{interpreter}.
26713 @subheading @value{GDBN} Command
26715 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
26717 @subheading Example
26721 -interpreter-exec console "break main"
26722 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
26723 &"During symbol reading, bad structure-type format.\n"
26724 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
26729 @subheading The @code{-inferior-tty-set} Command
26730 @findex -inferior-tty-set
26732 @subheading Synopsis
26735 -inferior-tty-set /dev/pts/1
26738 Set terminal for future runs of the program being debugged.
26740 @subheading @value{GDBN} Command
26742 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
26744 @subheading Example
26748 -inferior-tty-set /dev/pts/1
26753 @subheading The @code{-inferior-tty-show} Command
26754 @findex -inferior-tty-show
26756 @subheading Synopsis
26762 Show terminal for future runs of program being debugged.
26764 @subheading @value{GDBN} Command
26766 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
26768 @subheading Example
26772 -inferior-tty-set /dev/pts/1
26776 ^done,inferior_tty_terminal="/dev/pts/1"
26780 @subheading The @code{-enable-timings} Command
26781 @findex -enable-timings
26783 @subheading Synopsis
26786 -enable-timings [yes | no]
26789 Toggle the printing of the wallclock, user and system times for an MI
26790 command as a field in its output. This command is to help frontend
26791 developers optimize the performance of their code. No argument is
26792 equivalent to @samp{yes}.
26794 @subheading @value{GDBN} Command
26798 @subheading Example
26806 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26807 addr="0x080484ed",func="main",file="myprog.c",
26808 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
26809 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
26817 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
26818 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
26819 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
26820 fullname="/home/nickrob/myprog.c",line="73"@}
26825 @chapter @value{GDBN} Annotations
26827 This chapter describes annotations in @value{GDBN}. Annotations were
26828 designed to interface @value{GDBN} to graphical user interfaces or other
26829 similar programs which want to interact with @value{GDBN} at a
26830 relatively high level.
26832 The annotation mechanism has largely been superseded by @sc{gdb/mi}
26836 This is Edition @value{EDITION}, @value{DATE}.
26840 * Annotations Overview:: What annotations are; the general syntax.
26841 * Server Prefix:: Issuing a command without affecting user state.
26842 * Prompting:: Annotations marking @value{GDBN}'s need for input.
26843 * Errors:: Annotations for error messages.
26844 * Invalidation:: Some annotations describe things now invalid.
26845 * Annotations for Running::
26846 Whether the program is running, how it stopped, etc.
26847 * Source Annotations:: Annotations describing source code.
26850 @node Annotations Overview
26851 @section What is an Annotation?
26852 @cindex annotations
26854 Annotations start with a newline character, two @samp{control-z}
26855 characters, and the name of the annotation. If there is no additional
26856 information associated with this annotation, the name of the annotation
26857 is followed immediately by a newline. If there is additional
26858 information, the name of the annotation is followed by a space, the
26859 additional information, and a newline. The additional information
26860 cannot contain newline characters.
26862 Any output not beginning with a newline and two @samp{control-z}
26863 characters denotes literal output from @value{GDBN}. Currently there is
26864 no need for @value{GDBN} to output a newline followed by two
26865 @samp{control-z} characters, but if there was such a need, the
26866 annotations could be extended with an @samp{escape} annotation which
26867 means those three characters as output.
26869 The annotation @var{level}, which is specified using the
26870 @option{--annotate} command line option (@pxref{Mode Options}), controls
26871 how much information @value{GDBN} prints together with its prompt,
26872 values of expressions, source lines, and other types of output. Level 0
26873 is for no annotations, level 1 is for use when @value{GDBN} is run as a
26874 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
26875 for programs that control @value{GDBN}, and level 2 annotations have
26876 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
26877 Interface, annotate, GDB's Obsolete Annotations}).
26880 @kindex set annotate
26881 @item set annotate @var{level}
26882 The @value{GDBN} command @code{set annotate} sets the level of
26883 annotations to the specified @var{level}.
26885 @item show annotate
26886 @kindex show annotate
26887 Show the current annotation level.
26890 This chapter describes level 3 annotations.
26892 A simple example of starting up @value{GDBN} with annotations is:
26895 $ @kbd{gdb --annotate=3}
26897 Copyright 2003 Free Software Foundation, Inc.
26898 GDB is free software, covered by the GNU General Public License,
26899 and you are welcome to change it and/or distribute copies of it
26900 under certain conditions.
26901 Type "show copying" to see the conditions.
26902 There is absolutely no warranty for GDB. Type "show warranty"
26904 This GDB was configured as "i386-pc-linux-gnu"
26915 Here @samp{quit} is input to @value{GDBN}; the rest is output from
26916 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
26917 denotes a @samp{control-z} character) are annotations; the rest is
26918 output from @value{GDBN}.
26920 @node Server Prefix
26921 @section The Server Prefix
26922 @cindex server prefix
26924 If you prefix a command with @samp{server } then it will not affect
26925 the command history, nor will it affect @value{GDBN}'s notion of which
26926 command to repeat if @key{RET} is pressed on a line by itself. This
26927 means that commands can be run behind a user's back by a front-end in
26928 a transparent manner.
26930 The @code{server } prefix does not affect the recording of values into
26931 the value history; to print a value without recording it into the
26932 value history, use the @code{output} command instead of the
26933 @code{print} command.
26935 Using this prefix also disables confirmation requests
26936 (@pxref{confirmation requests}).
26939 @section Annotation for @value{GDBN} Input
26941 @cindex annotations for prompts
26942 When @value{GDBN} prompts for input, it annotates this fact so it is possible
26943 to know when to send output, when the output from a given command is
26946 Different kinds of input each have a different @dfn{input type}. Each
26947 input type has three annotations: a @code{pre-} annotation, which
26948 denotes the beginning of any prompt which is being output, a plain
26949 annotation, which denotes the end of the prompt, and then a @code{post-}
26950 annotation which denotes the end of any echo which may (or may not) be
26951 associated with the input. For example, the @code{prompt} input type
26952 features the following annotations:
26960 The input types are
26963 @findex pre-prompt annotation
26964 @findex prompt annotation
26965 @findex post-prompt annotation
26967 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
26969 @findex pre-commands annotation
26970 @findex commands annotation
26971 @findex post-commands annotation
26973 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
26974 command. The annotations are repeated for each command which is input.
26976 @findex pre-overload-choice annotation
26977 @findex overload-choice annotation
26978 @findex post-overload-choice annotation
26979 @item overload-choice
26980 When @value{GDBN} wants the user to select between various overloaded functions.
26982 @findex pre-query annotation
26983 @findex query annotation
26984 @findex post-query annotation
26986 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
26988 @findex pre-prompt-for-continue annotation
26989 @findex prompt-for-continue annotation
26990 @findex post-prompt-for-continue annotation
26991 @item prompt-for-continue
26992 When @value{GDBN} is asking the user to press return to continue. Note: Don't
26993 expect this to work well; instead use @code{set height 0} to disable
26994 prompting. This is because the counting of lines is buggy in the
26995 presence of annotations.
27000 @cindex annotations for errors, warnings and interrupts
27002 @findex quit annotation
27007 This annotation occurs right before @value{GDBN} responds to an interrupt.
27009 @findex error annotation
27014 This annotation occurs right before @value{GDBN} responds to an error.
27016 Quit and error annotations indicate that any annotations which @value{GDBN} was
27017 in the middle of may end abruptly. For example, if a
27018 @code{value-history-begin} annotation is followed by a @code{error}, one
27019 cannot expect to receive the matching @code{value-history-end}. One
27020 cannot expect not to receive it either, however; an error annotation
27021 does not necessarily mean that @value{GDBN} is immediately returning all the way
27024 @findex error-begin annotation
27025 A quit or error annotation may be preceded by
27031 Any output between that and the quit or error annotation is the error
27034 Warning messages are not yet annotated.
27035 @c If we want to change that, need to fix warning(), type_error(),
27036 @c range_error(), and possibly other places.
27039 @section Invalidation Notices
27041 @cindex annotations for invalidation messages
27042 The following annotations say that certain pieces of state may have
27046 @findex frames-invalid annotation
27047 @item ^Z^Zframes-invalid
27049 The frames (for example, output from the @code{backtrace} command) may
27052 @findex breakpoints-invalid annotation
27053 @item ^Z^Zbreakpoints-invalid
27055 The breakpoints may have changed. For example, the user just added or
27056 deleted a breakpoint.
27059 @node Annotations for Running
27060 @section Running the Program
27061 @cindex annotations for running programs
27063 @findex starting annotation
27064 @findex stopping annotation
27065 When the program starts executing due to a @value{GDBN} command such as
27066 @code{step} or @code{continue},
27072 is output. When the program stops,
27078 is output. Before the @code{stopped} annotation, a variety of
27079 annotations describe how the program stopped.
27082 @findex exited annotation
27083 @item ^Z^Zexited @var{exit-status}
27084 The program exited, and @var{exit-status} is the exit status (zero for
27085 successful exit, otherwise nonzero).
27087 @findex signalled annotation
27088 @findex signal-name annotation
27089 @findex signal-name-end annotation
27090 @findex signal-string annotation
27091 @findex signal-string-end annotation
27092 @item ^Z^Zsignalled
27093 The program exited with a signal. After the @code{^Z^Zsignalled}, the
27094 annotation continues:
27100 ^Z^Zsignal-name-end
27104 ^Z^Zsignal-string-end
27109 where @var{name} is the name of the signal, such as @code{SIGILL} or
27110 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
27111 as @code{Illegal Instruction} or @code{Segmentation fault}.
27112 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
27113 user's benefit and have no particular format.
27115 @findex signal annotation
27117 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
27118 just saying that the program received the signal, not that it was
27119 terminated with it.
27121 @findex breakpoint annotation
27122 @item ^Z^Zbreakpoint @var{number}
27123 The program hit breakpoint number @var{number}.
27125 @findex watchpoint annotation
27126 @item ^Z^Zwatchpoint @var{number}
27127 The program hit watchpoint number @var{number}.
27130 @node Source Annotations
27131 @section Displaying Source
27132 @cindex annotations for source display
27134 @findex source annotation
27135 The following annotation is used instead of displaying source code:
27138 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
27141 where @var{filename} is an absolute file name indicating which source
27142 file, @var{line} is the line number within that file (where 1 is the
27143 first line in the file), @var{character} is the character position
27144 within the file (where 0 is the first character in the file) (for most
27145 debug formats this will necessarily point to the beginning of a line),
27146 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
27147 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
27148 @var{addr} is the address in the target program associated with the
27149 source which is being displayed. @var{addr} is in the form @samp{0x}
27150 followed by one or more lowercase hex digits (note that this does not
27151 depend on the language).
27153 @node JIT Interface
27154 @chapter JIT Compilation Interface
27155 @cindex just-in-time compilation
27156 @cindex JIT compilation interface
27158 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
27159 interface. A JIT compiler is a program or library that generates native
27160 executable code at runtime and executes it, usually in order to achieve good
27161 performance while maintaining platform independence.
27163 Programs that use JIT compilation are normally difficult to debug because
27164 portions of their code are generated at runtime, instead of being loaded from
27165 object files, which is where @value{GDBN} normally finds the program's symbols
27166 and debug information. In order to debug programs that use JIT compilation,
27167 @value{GDBN} has an interface that allows the program to register in-memory
27168 symbol files with @value{GDBN} at runtime.
27170 If you are using @value{GDBN} to debug a program that uses this interface, then
27171 it should work transparently so long as you have not stripped the binary. If
27172 you are developing a JIT compiler, then the interface is documented in the rest
27173 of this chapter. At this time, the only known client of this interface is the
27176 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
27177 JIT compiler communicates with @value{GDBN} by writing data into a global
27178 variable and calling a fuction at a well-known symbol. When @value{GDBN}
27179 attaches, it reads a linked list of symbol files from the global variable to
27180 find existing code, and puts a breakpoint in the function so that it can find
27181 out about additional code.
27184 * Declarations:: Relevant C struct declarations
27185 * Registering Code:: Steps to register code
27186 * Unregistering Code:: Steps to unregister code
27190 @section JIT Declarations
27192 These are the relevant struct declarations that a C program should include to
27193 implement the interface:
27203 struct jit_code_entry
27205 struct jit_code_entry *next_entry;
27206 struct jit_code_entry *prev_entry;
27207 const char *symfile_addr;
27208 uint64_t symfile_size;
27211 struct jit_descriptor
27214 /* This type should be jit_actions_t, but we use uint32_t
27215 to be explicit about the bitwidth. */
27216 uint32_t action_flag;
27217 struct jit_code_entry *relevant_entry;
27218 struct jit_code_entry *first_entry;
27221 /* GDB puts a breakpoint in this function. */
27222 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
27224 /* Make sure to specify the version statically, because the
27225 debugger may check the version before we can set it. */
27226 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
27229 If the JIT is multi-threaded, then it is important that the JIT synchronize any
27230 modifications to this global data properly, which can easily be done by putting
27231 a global mutex around modifications to these structures.
27233 @node Registering Code
27234 @section Registering Code
27236 To register code with @value{GDBN}, the JIT should follow this protocol:
27240 Generate an object file in memory with symbols and other desired debug
27241 information. The file must include the virtual addresses of the sections.
27244 Create a code entry for the file, which gives the start and size of the symbol
27248 Add it to the linked list in the JIT descriptor.
27251 Point the relevant_entry field of the descriptor at the entry.
27254 Set @code{action_flag} to @code{JIT_REGISTER} and call
27255 @code{__jit_debug_register_code}.
27258 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
27259 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
27260 new code. However, the linked list must still be maintained in order to allow
27261 @value{GDBN} to attach to a running process and still find the symbol files.
27263 @node Unregistering Code
27264 @section Unregistering Code
27266 If code is freed, then the JIT should use the following protocol:
27270 Remove the code entry corresponding to the code from the linked list.
27273 Point the @code{relevant_entry} field of the descriptor at the code entry.
27276 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
27277 @code{__jit_debug_register_code}.
27280 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
27281 and the JIT will leak the memory used for the associated symbol files.
27284 @chapter Reporting Bugs in @value{GDBN}
27285 @cindex bugs in @value{GDBN}
27286 @cindex reporting bugs in @value{GDBN}
27288 Your bug reports play an essential role in making @value{GDBN} reliable.
27290 Reporting a bug may help you by bringing a solution to your problem, or it
27291 may not. But in any case the principal function of a bug report is to help
27292 the entire community by making the next version of @value{GDBN} work better. Bug
27293 reports are your contribution to the maintenance of @value{GDBN}.
27295 In order for a bug report to serve its purpose, you must include the
27296 information that enables us to fix the bug.
27299 * Bug Criteria:: Have you found a bug?
27300 * Bug Reporting:: How to report bugs
27304 @section Have You Found a Bug?
27305 @cindex bug criteria
27307 If you are not sure whether you have found a bug, here are some guidelines:
27310 @cindex fatal signal
27311 @cindex debugger crash
27312 @cindex crash of debugger
27314 If the debugger gets a fatal signal, for any input whatever, that is a
27315 @value{GDBN} bug. Reliable debuggers never crash.
27317 @cindex error on valid input
27319 If @value{GDBN} produces an error message for valid input, that is a
27320 bug. (Note that if you're cross debugging, the problem may also be
27321 somewhere in the connection to the target.)
27323 @cindex invalid input
27325 If @value{GDBN} does not produce an error message for invalid input,
27326 that is a bug. However, you should note that your idea of
27327 ``invalid input'' might be our idea of ``an extension'' or ``support
27328 for traditional practice''.
27331 If you are an experienced user of debugging tools, your suggestions
27332 for improvement of @value{GDBN} are welcome in any case.
27335 @node Bug Reporting
27336 @section How to Report Bugs
27337 @cindex bug reports
27338 @cindex @value{GDBN} bugs, reporting
27340 A number of companies and individuals offer support for @sc{gnu} products.
27341 If you obtained @value{GDBN} from a support organization, we recommend you
27342 contact that organization first.
27344 You can find contact information for many support companies and
27345 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
27347 @c should add a web page ref...
27350 @ifset BUGURL_DEFAULT
27351 In any event, we also recommend that you submit bug reports for
27352 @value{GDBN}. The preferred method is to submit them directly using
27353 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
27354 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
27357 @strong{Do not send bug reports to @samp{info-gdb}, or to
27358 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
27359 not want to receive bug reports. Those that do have arranged to receive
27362 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
27363 serves as a repeater. The mailing list and the newsgroup carry exactly
27364 the same messages. Often people think of posting bug reports to the
27365 newsgroup instead of mailing them. This appears to work, but it has one
27366 problem which can be crucial: a newsgroup posting often lacks a mail
27367 path back to the sender. Thus, if we need to ask for more information,
27368 we may be unable to reach you. For this reason, it is better to send
27369 bug reports to the mailing list.
27371 @ifclear BUGURL_DEFAULT
27372 In any event, we also recommend that you submit bug reports for
27373 @value{GDBN} to @value{BUGURL}.
27377 The fundamental principle of reporting bugs usefully is this:
27378 @strong{report all the facts}. If you are not sure whether to state a
27379 fact or leave it out, state it!
27381 Often people omit facts because they think they know what causes the
27382 problem and assume that some details do not matter. Thus, you might
27383 assume that the name of the variable you use in an example does not matter.
27384 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
27385 stray memory reference which happens to fetch from the location where that
27386 name is stored in memory; perhaps, if the name were different, the contents
27387 of that location would fool the debugger into doing the right thing despite
27388 the bug. Play it safe and give a specific, complete example. That is the
27389 easiest thing for you to do, and the most helpful.
27391 Keep in mind that the purpose of a bug report is to enable us to fix the
27392 bug. It may be that the bug has been reported previously, but neither
27393 you nor we can know that unless your bug report is complete and
27396 Sometimes people give a few sketchy facts and ask, ``Does this ring a
27397 bell?'' Those bug reports are useless, and we urge everyone to
27398 @emph{refuse to respond to them} except to chide the sender to report
27401 To enable us to fix the bug, you should include all these things:
27405 The version of @value{GDBN}. @value{GDBN} announces it if you start
27406 with no arguments; you can also print it at any time using @code{show
27409 Without this, we will not know whether there is any point in looking for
27410 the bug in the current version of @value{GDBN}.
27413 The type of machine you are using, and the operating system name and
27417 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
27418 ``@value{GCC}--2.8.1''.
27421 What compiler (and its version) was used to compile the program you are
27422 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
27423 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
27424 to get this information; for other compilers, see the documentation for
27428 The command arguments you gave the compiler to compile your example and
27429 observe the bug. For example, did you use @samp{-O}? To guarantee
27430 you will not omit something important, list them all. A copy of the
27431 Makefile (or the output from make) is sufficient.
27433 If we were to try to guess the arguments, we would probably guess wrong
27434 and then we might not encounter the bug.
27437 A complete input script, and all necessary source files, that will
27441 A description of what behavior you observe that you believe is
27442 incorrect. For example, ``It gets a fatal signal.''
27444 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
27445 will certainly notice it. But if the bug is incorrect output, we might
27446 not notice unless it is glaringly wrong. You might as well not give us
27447 a chance to make a mistake.
27449 Even if the problem you experience is a fatal signal, you should still
27450 say so explicitly. Suppose something strange is going on, such as, your
27451 copy of @value{GDBN} is out of synch, or you have encountered a bug in
27452 the C library on your system. (This has happened!) Your copy might
27453 crash and ours would not. If you told us to expect a crash, then when
27454 ours fails to crash, we would know that the bug was not happening for
27455 us. If you had not told us to expect a crash, then we would not be able
27456 to draw any conclusion from our observations.
27459 @cindex recording a session script
27460 To collect all this information, you can use a session recording program
27461 such as @command{script}, which is available on many Unix systems.
27462 Just run your @value{GDBN} session inside @command{script} and then
27463 include the @file{typescript} file with your bug report.
27465 Another way to record a @value{GDBN} session is to run @value{GDBN}
27466 inside Emacs and then save the entire buffer to a file.
27469 If you wish to suggest changes to the @value{GDBN} source, send us context
27470 diffs. If you even discuss something in the @value{GDBN} source, refer to
27471 it by context, not by line number.
27473 The line numbers in our development sources will not match those in your
27474 sources. Your line numbers would convey no useful information to us.
27478 Here are some things that are not necessary:
27482 A description of the envelope of the bug.
27484 Often people who encounter a bug spend a lot of time investigating
27485 which changes to the input file will make the bug go away and which
27486 changes will not affect it.
27488 This is often time consuming and not very useful, because the way we
27489 will find the bug is by running a single example under the debugger
27490 with breakpoints, not by pure deduction from a series of examples.
27491 We recommend that you save your time for something else.
27493 Of course, if you can find a simpler example to report @emph{instead}
27494 of the original one, that is a convenience for us. Errors in the
27495 output will be easier to spot, running under the debugger will take
27496 less time, and so on.
27498 However, simplification is not vital; if you do not want to do this,
27499 report the bug anyway and send us the entire test case you used.
27502 A patch for the bug.
27504 A patch for the bug does help us if it is a good one. But do not omit
27505 the necessary information, such as the test case, on the assumption that
27506 a patch is all we need. We might see problems with your patch and decide
27507 to fix the problem another way, or we might not understand it at all.
27509 Sometimes with a program as complicated as @value{GDBN} it is very hard to
27510 construct an example that will make the program follow a certain path
27511 through the code. If you do not send us the example, we will not be able
27512 to construct one, so we will not be able to verify that the bug is fixed.
27514 And if we cannot understand what bug you are trying to fix, or why your
27515 patch should be an improvement, we will not install it. A test case will
27516 help us to understand.
27519 A guess about what the bug is or what it depends on.
27521 Such guesses are usually wrong. Even we cannot guess right about such
27522 things without first using the debugger to find the facts.
27525 @c The readline documentation is distributed with the readline code
27526 @c and consists of the two following files:
27528 @c inc-hist.texinfo
27529 @c Use -I with makeinfo to point to the appropriate directory,
27530 @c environment var TEXINPUTS with TeX.
27531 @include rluser.texi
27532 @include inc-hist.texinfo
27535 @node Formatting Documentation
27536 @appendix Formatting Documentation
27538 @cindex @value{GDBN} reference card
27539 @cindex reference card
27540 The @value{GDBN} 4 release includes an already-formatted reference card, ready
27541 for printing with PostScript or Ghostscript, in the @file{gdb}
27542 subdirectory of the main source directory@footnote{In
27543 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
27544 release.}. If you can use PostScript or Ghostscript with your printer,
27545 you can print the reference card immediately with @file{refcard.ps}.
27547 The release also includes the source for the reference card. You
27548 can format it, using @TeX{}, by typing:
27554 The @value{GDBN} reference card is designed to print in @dfn{landscape}
27555 mode on US ``letter'' size paper;
27556 that is, on a sheet 11 inches wide by 8.5 inches
27557 high. You will need to specify this form of printing as an option to
27558 your @sc{dvi} output program.
27560 @cindex documentation
27562 All the documentation for @value{GDBN} comes as part of the machine-readable
27563 distribution. The documentation is written in Texinfo format, which is
27564 a documentation system that uses a single source file to produce both
27565 on-line information and a printed manual. You can use one of the Info
27566 formatting commands to create the on-line version of the documentation
27567 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
27569 @value{GDBN} includes an already formatted copy of the on-line Info
27570 version of this manual in the @file{gdb} subdirectory. The main Info
27571 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
27572 subordinate files matching @samp{gdb.info*} in the same directory. If
27573 necessary, you can print out these files, or read them with any editor;
27574 but they are easier to read using the @code{info} subsystem in @sc{gnu}
27575 Emacs or the standalone @code{info} program, available as part of the
27576 @sc{gnu} Texinfo distribution.
27578 If you want to format these Info files yourself, you need one of the
27579 Info formatting programs, such as @code{texinfo-format-buffer} or
27582 If you have @code{makeinfo} installed, and are in the top level
27583 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
27584 version @value{GDBVN}), you can make the Info file by typing:
27591 If you want to typeset and print copies of this manual, you need @TeX{},
27592 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
27593 Texinfo definitions file.
27595 @TeX{} is a typesetting program; it does not print files directly, but
27596 produces output files called @sc{dvi} files. To print a typeset
27597 document, you need a program to print @sc{dvi} files. If your system
27598 has @TeX{} installed, chances are it has such a program. The precise
27599 command to use depends on your system; @kbd{lpr -d} is common; another
27600 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
27601 require a file name without any extension or a @samp{.dvi} extension.
27603 @TeX{} also requires a macro definitions file called
27604 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
27605 written in Texinfo format. On its own, @TeX{} cannot either read or
27606 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
27607 and is located in the @file{gdb-@var{version-number}/texinfo}
27610 If you have @TeX{} and a @sc{dvi} printer program installed, you can
27611 typeset and print this manual. First switch to the @file{gdb}
27612 subdirectory of the main source directory (for example, to
27613 @file{gdb-@value{GDBVN}/gdb}) and type:
27619 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
27621 @node Installing GDB
27622 @appendix Installing @value{GDBN}
27623 @cindex installation
27626 * Requirements:: Requirements for building @value{GDBN}
27627 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
27628 * Separate Objdir:: Compiling @value{GDBN} in another directory
27629 * Config Names:: Specifying names for hosts and targets
27630 * Configure Options:: Summary of options for configure
27631 * System-wide configuration:: Having a system-wide init file
27635 @section Requirements for Building @value{GDBN}
27636 @cindex building @value{GDBN}, requirements for
27638 Building @value{GDBN} requires various tools and packages to be available.
27639 Other packages will be used only if they are found.
27641 @heading Tools/Packages Necessary for Building @value{GDBN}
27643 @item ISO C90 compiler
27644 @value{GDBN} is written in ISO C90. It should be buildable with any
27645 working C90 compiler, e.g.@: GCC.
27649 @heading Tools/Packages Optional for Building @value{GDBN}
27653 @value{GDBN} can use the Expat XML parsing library. This library may be
27654 included with your operating system distribution; if it is not, you
27655 can get the latest version from @url{http://expat.sourceforge.net}.
27656 The @file{configure} script will search for this library in several
27657 standard locations; if it is installed in an unusual path, you can
27658 use the @option{--with-libexpat-prefix} option to specify its location.
27664 Remote protocol memory maps (@pxref{Memory Map Format})
27666 Target descriptions (@pxref{Target Descriptions})
27668 Remote shared library lists (@pxref{Library List Format})
27670 MS-Windows shared libraries (@pxref{Shared Libraries})
27674 @cindex compressed debug sections
27675 @value{GDBN} will use the @samp{zlib} library, if available, to read
27676 compressed debug sections. Some linkers, such as GNU gold, are capable
27677 of producing binaries with compressed debug sections. If @value{GDBN}
27678 is compiled with @samp{zlib}, it will be able to read the debug
27679 information in such binaries.
27681 The @samp{zlib} library is likely included with your operating system
27682 distribution; if it is not, you can get the latest version from
27683 @url{http://zlib.net}.
27686 @value{GDBN}'s features related to character sets (@pxref{Character
27687 Sets}) require a functioning @code{iconv} implementation. If you are
27688 on a GNU system, then this is provided by the GNU C Library. Some
27689 other systems also provide a working @code{iconv}.
27691 On systems with @code{iconv}, you can install GNU Libiconv. If you
27692 have previously installed Libiconv, you can use the
27693 @option{--with-libiconv-prefix} option to configure.
27695 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
27696 arrange to build Libiconv if a directory named @file{libiconv} appears
27697 in the top-most source directory. If Libiconv is built this way, and
27698 if the operating system does not provide a suitable @code{iconv}
27699 implementation, then the just-built library will automatically be used
27700 by @value{GDBN}. One easy way to set this up is to download GNU
27701 Libiconv, unpack it, and then rename the directory holding the
27702 Libiconv source code to @samp{libiconv}.
27705 @node Running Configure
27706 @section Invoking the @value{GDBN} @file{configure} Script
27707 @cindex configuring @value{GDBN}
27708 @value{GDBN} comes with a @file{configure} script that automates the process
27709 of preparing @value{GDBN} for installation; you can then use @code{make} to
27710 build the @code{gdb} program.
27712 @c irrelevant in info file; it's as current as the code it lives with.
27713 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
27714 look at the @file{README} file in the sources; we may have improved the
27715 installation procedures since publishing this manual.}
27718 The @value{GDBN} distribution includes all the source code you need for
27719 @value{GDBN} in a single directory, whose name is usually composed by
27720 appending the version number to @samp{gdb}.
27722 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
27723 @file{gdb-@value{GDBVN}} directory. That directory contains:
27726 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
27727 script for configuring @value{GDBN} and all its supporting libraries
27729 @item gdb-@value{GDBVN}/gdb
27730 the source specific to @value{GDBN} itself
27732 @item gdb-@value{GDBVN}/bfd
27733 source for the Binary File Descriptor library
27735 @item gdb-@value{GDBVN}/include
27736 @sc{gnu} include files
27738 @item gdb-@value{GDBVN}/libiberty
27739 source for the @samp{-liberty} free software library
27741 @item gdb-@value{GDBVN}/opcodes
27742 source for the library of opcode tables and disassemblers
27744 @item gdb-@value{GDBVN}/readline
27745 source for the @sc{gnu} command-line interface
27747 @item gdb-@value{GDBVN}/glob
27748 source for the @sc{gnu} filename pattern-matching subroutine
27750 @item gdb-@value{GDBVN}/mmalloc
27751 source for the @sc{gnu} memory-mapped malloc package
27754 The simplest way to configure and build @value{GDBN} is to run @file{configure}
27755 from the @file{gdb-@var{version-number}} source directory, which in
27756 this example is the @file{gdb-@value{GDBVN}} directory.
27758 First switch to the @file{gdb-@var{version-number}} source directory
27759 if you are not already in it; then run @file{configure}. Pass the
27760 identifier for the platform on which @value{GDBN} will run as an
27766 cd gdb-@value{GDBVN}
27767 ./configure @var{host}
27772 where @var{host} is an identifier such as @samp{sun4} or
27773 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
27774 (You can often leave off @var{host}; @file{configure} tries to guess the
27775 correct value by examining your system.)
27777 Running @samp{configure @var{host}} and then running @code{make} builds the
27778 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
27779 libraries, then @code{gdb} itself. The configured source files, and the
27780 binaries, are left in the corresponding source directories.
27783 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
27784 system does not recognize this automatically when you run a different
27785 shell, you may need to run @code{sh} on it explicitly:
27788 sh configure @var{host}
27791 If you run @file{configure} from a directory that contains source
27792 directories for multiple libraries or programs, such as the
27793 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
27795 creates configuration files for every directory level underneath (unless
27796 you tell it not to, with the @samp{--norecursion} option).
27798 You should run the @file{configure} script from the top directory in the
27799 source tree, the @file{gdb-@var{version-number}} directory. If you run
27800 @file{configure} from one of the subdirectories, you will configure only
27801 that subdirectory. That is usually not what you want. In particular,
27802 if you run the first @file{configure} from the @file{gdb} subdirectory
27803 of the @file{gdb-@var{version-number}} directory, you will omit the
27804 configuration of @file{bfd}, @file{readline}, and other sibling
27805 directories of the @file{gdb} subdirectory. This leads to build errors
27806 about missing include files such as @file{bfd/bfd.h}.
27808 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
27809 However, you should make sure that the shell on your path (named by
27810 the @samp{SHELL} environment variable) is publicly readable. Remember
27811 that @value{GDBN} uses the shell to start your program---some systems refuse to
27812 let @value{GDBN} debug child processes whose programs are not readable.
27814 @node Separate Objdir
27815 @section Compiling @value{GDBN} in Another Directory
27817 If you want to run @value{GDBN} versions for several host or target machines,
27818 you need a different @code{gdb} compiled for each combination of
27819 host and target. @file{configure} is designed to make this easy by
27820 allowing you to generate each configuration in a separate subdirectory,
27821 rather than in the source directory. If your @code{make} program
27822 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
27823 @code{make} in each of these directories builds the @code{gdb}
27824 program specified there.
27826 To build @code{gdb} in a separate directory, run @file{configure}
27827 with the @samp{--srcdir} option to specify where to find the source.
27828 (You also need to specify a path to find @file{configure}
27829 itself from your working directory. If the path to @file{configure}
27830 would be the same as the argument to @samp{--srcdir}, you can leave out
27831 the @samp{--srcdir} option; it is assumed.)
27833 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
27834 separate directory for a Sun 4 like this:
27838 cd gdb-@value{GDBVN}
27841 ../gdb-@value{GDBVN}/configure sun4
27846 When @file{configure} builds a configuration using a remote source
27847 directory, it creates a tree for the binaries with the same structure
27848 (and using the same names) as the tree under the source directory. In
27849 the example, you'd find the Sun 4 library @file{libiberty.a} in the
27850 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
27851 @file{gdb-sun4/gdb}.
27853 Make sure that your path to the @file{configure} script has just one
27854 instance of @file{gdb} in it. If your path to @file{configure} looks
27855 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
27856 one subdirectory of @value{GDBN}, not the whole package. This leads to
27857 build errors about missing include files such as @file{bfd/bfd.h}.
27859 One popular reason to build several @value{GDBN} configurations in separate
27860 directories is to configure @value{GDBN} for cross-compiling (where
27861 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
27862 programs that run on another machine---the @dfn{target}).
27863 You specify a cross-debugging target by
27864 giving the @samp{--target=@var{target}} option to @file{configure}.
27866 When you run @code{make} to build a program or library, you must run
27867 it in a configured directory---whatever directory you were in when you
27868 called @file{configure} (or one of its subdirectories).
27870 The @code{Makefile} that @file{configure} generates in each source
27871 directory also runs recursively. If you type @code{make} in a source
27872 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
27873 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
27874 will build all the required libraries, and then build GDB.
27876 When you have multiple hosts or targets configured in separate
27877 directories, you can run @code{make} on them in parallel (for example,
27878 if they are NFS-mounted on each of the hosts); they will not interfere
27882 @section Specifying Names for Hosts and Targets
27884 The specifications used for hosts and targets in the @file{configure}
27885 script are based on a three-part naming scheme, but some short predefined
27886 aliases are also supported. The full naming scheme encodes three pieces
27887 of information in the following pattern:
27890 @var{architecture}-@var{vendor}-@var{os}
27893 For example, you can use the alias @code{sun4} as a @var{host} argument,
27894 or as the value for @var{target} in a @code{--target=@var{target}}
27895 option. The equivalent full name is @samp{sparc-sun-sunos4}.
27897 The @file{configure} script accompanying @value{GDBN} does not provide
27898 any query facility to list all supported host and target names or
27899 aliases. @file{configure} calls the Bourne shell script
27900 @code{config.sub} to map abbreviations to full names; you can read the
27901 script, if you wish, or you can use it to test your guesses on
27902 abbreviations---for example:
27905 % sh config.sub i386-linux
27907 % sh config.sub alpha-linux
27908 alpha-unknown-linux-gnu
27909 % sh config.sub hp9k700
27911 % sh config.sub sun4
27912 sparc-sun-sunos4.1.1
27913 % sh config.sub sun3
27914 m68k-sun-sunos4.1.1
27915 % sh config.sub i986v
27916 Invalid configuration `i986v': machine `i986v' not recognized
27920 @code{config.sub} is also distributed in the @value{GDBN} source
27921 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
27923 @node Configure Options
27924 @section @file{configure} Options
27926 Here is a summary of the @file{configure} options and arguments that
27927 are most often useful for building @value{GDBN}. @file{configure} also has
27928 several other options not listed here. @inforef{What Configure
27929 Does,,configure.info}, for a full explanation of @file{configure}.
27932 configure @r{[}--help@r{]}
27933 @r{[}--prefix=@var{dir}@r{]}
27934 @r{[}--exec-prefix=@var{dir}@r{]}
27935 @r{[}--srcdir=@var{dirname}@r{]}
27936 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
27937 @r{[}--target=@var{target}@r{]}
27942 You may introduce options with a single @samp{-} rather than
27943 @samp{--} if you prefer; but you may abbreviate option names if you use
27948 Display a quick summary of how to invoke @file{configure}.
27950 @item --prefix=@var{dir}
27951 Configure the source to install programs and files under directory
27954 @item --exec-prefix=@var{dir}
27955 Configure the source to install programs under directory
27958 @c avoid splitting the warning from the explanation:
27960 @item --srcdir=@var{dirname}
27961 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
27962 @code{make} that implements the @code{VPATH} feature.}@*
27963 Use this option to make configurations in directories separate from the
27964 @value{GDBN} source directories. Among other things, you can use this to
27965 build (or maintain) several configurations simultaneously, in separate
27966 directories. @file{configure} writes configuration-specific files in
27967 the current directory, but arranges for them to use the source in the
27968 directory @var{dirname}. @file{configure} creates directories under
27969 the working directory in parallel to the source directories below
27972 @item --norecursion
27973 Configure only the directory level where @file{configure} is executed; do not
27974 propagate configuration to subdirectories.
27976 @item --target=@var{target}
27977 Configure @value{GDBN} for cross-debugging programs running on the specified
27978 @var{target}. Without this option, @value{GDBN} is configured to debug
27979 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
27981 There is no convenient way to generate a list of all available targets.
27983 @item @var{host} @dots{}
27984 Configure @value{GDBN} to run on the specified @var{host}.
27986 There is no convenient way to generate a list of all available hosts.
27989 There are many other options available as well, but they are generally
27990 needed for special purposes only.
27992 @node System-wide configuration
27993 @section System-wide configuration and settings
27994 @cindex system-wide init file
27996 @value{GDBN} can be configured to have a system-wide init file;
27997 this file will be read and executed at startup (@pxref{Startup, , What
27998 @value{GDBN} does during startup}).
28000 Here is the corresponding configure option:
28003 @item --with-system-gdbinit=@var{file}
28004 Specify that the default location of the system-wide init file is
28008 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
28009 it may be subject to relocation. Two possible cases:
28013 If the default location of this init file contains @file{$prefix},
28014 it will be subject to relocation. Suppose that the configure options
28015 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
28016 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
28017 init file is looked for as @file{$install/etc/gdbinit} instead of
28018 @file{$prefix/etc/gdbinit}.
28021 By contrast, if the default location does not contain the prefix,
28022 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
28023 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
28024 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
28025 wherever @value{GDBN} is installed.
28028 @node Maintenance Commands
28029 @appendix Maintenance Commands
28030 @cindex maintenance commands
28031 @cindex internal commands
28033 In addition to commands intended for @value{GDBN} users, @value{GDBN}
28034 includes a number of commands intended for @value{GDBN} developers,
28035 that are not documented elsewhere in this manual. These commands are
28036 provided here for reference. (For commands that turn on debugging
28037 messages, see @ref{Debugging Output}.)
28040 @kindex maint agent
28041 @kindex maint agent-eval
28042 @item maint agent @var{expression}
28043 @itemx maint agent-eval @var{expression}
28044 Translate the given @var{expression} into remote agent bytecodes.
28045 This command is useful for debugging the Agent Expression mechanism
28046 (@pxref{Agent Expressions}). The @samp{agent} version produces an
28047 expression useful for data collection, such as by tracepoints, while
28048 @samp{maint agent-eval} produces an expression that evaluates directly
28049 to a result. For instance, a collection expression for @code{globa +
28050 globb} will include bytecodes to record four bytes of memory at each
28051 of the addresses of @code{globa} and @code{globb}, while discarding
28052 the result of the addition, while an evaluation expression will do the
28053 addition and return the sum.
28055 @kindex maint info breakpoints
28056 @item @anchor{maint info breakpoints}maint info breakpoints
28057 Using the same format as @samp{info breakpoints}, display both the
28058 breakpoints you've set explicitly, and those @value{GDBN} is using for
28059 internal purposes. Internal breakpoints are shown with negative
28060 breakpoint numbers. The type column identifies what kind of breakpoint
28065 Normal, explicitly set breakpoint.
28068 Normal, explicitly set watchpoint.
28071 Internal breakpoint, used to handle correctly stepping through
28072 @code{longjmp} calls.
28074 @item longjmp resume
28075 Internal breakpoint at the target of a @code{longjmp}.
28078 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
28081 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
28084 Shared library events.
28088 @kindex set displaced-stepping
28089 @kindex show displaced-stepping
28090 @cindex displaced stepping support
28091 @cindex out-of-line single-stepping
28092 @item set displaced-stepping
28093 @itemx show displaced-stepping
28094 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
28095 if the target supports it. Displaced stepping is a way to single-step
28096 over breakpoints without removing them from the inferior, by executing
28097 an out-of-line copy of the instruction that was originally at the
28098 breakpoint location. It is also known as out-of-line single-stepping.
28101 @item set displaced-stepping on
28102 If the target architecture supports it, @value{GDBN} will use
28103 displaced stepping to step over breakpoints.
28105 @item set displaced-stepping off
28106 @value{GDBN} will not use displaced stepping to step over breakpoints,
28107 even if such is supported by the target architecture.
28109 @cindex non-stop mode, and @samp{set displaced-stepping}
28110 @item set displaced-stepping auto
28111 This is the default mode. @value{GDBN} will use displaced stepping
28112 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
28113 architecture supports displaced stepping.
28116 @kindex maint check-symtabs
28117 @item maint check-symtabs
28118 Check the consistency of psymtabs and symtabs.
28120 @kindex maint cplus first_component
28121 @item maint cplus first_component @var{name}
28122 Print the first C@t{++} class/namespace component of @var{name}.
28124 @kindex maint cplus namespace
28125 @item maint cplus namespace
28126 Print the list of possible C@t{++} namespaces.
28128 @kindex maint demangle
28129 @item maint demangle @var{name}
28130 Demangle a C@t{++} or Objective-C mangled @var{name}.
28132 @kindex maint deprecate
28133 @kindex maint undeprecate
28134 @cindex deprecated commands
28135 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
28136 @itemx maint undeprecate @var{command}
28137 Deprecate or undeprecate the named @var{command}. Deprecated commands
28138 cause @value{GDBN} to issue a warning when you use them. The optional
28139 argument @var{replacement} says which newer command should be used in
28140 favor of the deprecated one; if it is given, @value{GDBN} will mention
28141 the replacement as part of the warning.
28143 @kindex maint dump-me
28144 @item maint dump-me
28145 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
28146 Cause a fatal signal in the debugger and force it to dump its core.
28147 This is supported only on systems which support aborting a program
28148 with the @code{SIGQUIT} signal.
28150 @kindex maint internal-error
28151 @kindex maint internal-warning
28152 @item maint internal-error @r{[}@var{message-text}@r{]}
28153 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
28154 Cause @value{GDBN} to call the internal function @code{internal_error}
28155 or @code{internal_warning} and hence behave as though an internal error
28156 or internal warning has been detected. In addition to reporting the
28157 internal problem, these functions give the user the opportunity to
28158 either quit @value{GDBN} or create a core file of the current
28159 @value{GDBN} session.
28161 These commands take an optional parameter @var{message-text} that is
28162 used as the text of the error or warning message.
28164 Here's an example of using @code{internal-error}:
28167 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
28168 @dots{}/maint.c:121: internal-error: testing, 1, 2
28169 A problem internal to GDB has been detected. Further
28170 debugging may prove unreliable.
28171 Quit this debugging session? (y or n) @kbd{n}
28172 Create a core file? (y or n) @kbd{n}
28176 @cindex @value{GDBN} internal error
28177 @cindex internal errors, control of @value{GDBN} behavior
28179 @kindex maint set internal-error
28180 @kindex maint show internal-error
28181 @kindex maint set internal-warning
28182 @kindex maint show internal-warning
28183 @item maint set internal-error @var{action} [ask|yes|no]
28184 @itemx maint show internal-error @var{action}
28185 @itemx maint set internal-warning @var{action} [ask|yes|no]
28186 @itemx maint show internal-warning @var{action}
28187 When @value{GDBN} reports an internal problem (error or warning) it
28188 gives the user the opportunity to both quit @value{GDBN} and create a
28189 core file of the current @value{GDBN} session. These commands let you
28190 override the default behaviour for each particular @var{action},
28191 described in the table below.
28195 You can specify that @value{GDBN} should always (yes) or never (no)
28196 quit. The default is to ask the user what to do.
28199 You can specify that @value{GDBN} should always (yes) or never (no)
28200 create a core file. The default is to ask the user what to do.
28203 @kindex maint packet
28204 @item maint packet @var{text}
28205 If @value{GDBN} is talking to an inferior via the serial protocol,
28206 then this command sends the string @var{text} to the inferior, and
28207 displays the response packet. @value{GDBN} supplies the initial
28208 @samp{$} character, the terminating @samp{#} character, and the
28211 @kindex maint print architecture
28212 @item maint print architecture @r{[}@var{file}@r{]}
28213 Print the entire architecture configuration. The optional argument
28214 @var{file} names the file where the output goes.
28216 @kindex maint print c-tdesc
28217 @item maint print c-tdesc
28218 Print the current target description (@pxref{Target Descriptions}) as
28219 a C source file. The created source file can be used in @value{GDBN}
28220 when an XML parser is not available to parse the description.
28222 @kindex maint print dummy-frames
28223 @item maint print dummy-frames
28224 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
28227 (@value{GDBP}) @kbd{b add}
28229 (@value{GDBP}) @kbd{print add(2,3)}
28230 Breakpoint 2, add (a=2, b=3) at @dots{}
28232 The program being debugged stopped while in a function called from GDB.
28234 (@value{GDBP}) @kbd{maint print dummy-frames}
28235 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
28236 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
28237 call_lo=0x01014000 call_hi=0x01014001
28241 Takes an optional file parameter.
28243 @kindex maint print registers
28244 @kindex maint print raw-registers
28245 @kindex maint print cooked-registers
28246 @kindex maint print register-groups
28247 @item maint print registers @r{[}@var{file}@r{]}
28248 @itemx maint print raw-registers @r{[}@var{file}@r{]}
28249 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
28250 @itemx maint print register-groups @r{[}@var{file}@r{]}
28251 Print @value{GDBN}'s internal register data structures.
28253 The command @code{maint print raw-registers} includes the contents of
28254 the raw register cache; the command @code{maint print cooked-registers}
28255 includes the (cooked) value of all registers; and the command
28256 @code{maint print register-groups} includes the groups that each
28257 register is a member of. @xref{Registers,, Registers, gdbint,
28258 @value{GDBN} Internals}.
28260 These commands take an optional parameter, a file name to which to
28261 write the information.
28263 @kindex maint print reggroups
28264 @item maint print reggroups @r{[}@var{file}@r{]}
28265 Print @value{GDBN}'s internal register group data structures. The
28266 optional argument @var{file} tells to what file to write the
28269 The register groups info looks like this:
28272 (@value{GDBP}) @kbd{maint print reggroups}
28285 This command forces @value{GDBN} to flush its internal register cache.
28287 @kindex maint print objfiles
28288 @cindex info for known object files
28289 @item maint print objfiles
28290 Print a dump of all known object files. For each object file, this
28291 command prints its name, address in memory, and all of its psymtabs
28294 @kindex maint print statistics
28295 @cindex bcache statistics
28296 @item maint print statistics
28297 This command prints, for each object file in the program, various data
28298 about that object file followed by the byte cache (@dfn{bcache})
28299 statistics for the object file. The objfile data includes the number
28300 of minimal, partial, full, and stabs symbols, the number of types
28301 defined by the objfile, the number of as yet unexpanded psym tables,
28302 the number of line tables and string tables, and the amount of memory
28303 used by the various tables. The bcache statistics include the counts,
28304 sizes, and counts of duplicates of all and unique objects, max,
28305 average, and median entry size, total memory used and its overhead and
28306 savings, and various measures of the hash table size and chain
28309 @kindex maint print target-stack
28310 @cindex target stack description
28311 @item maint print target-stack
28312 A @dfn{target} is an interface between the debugger and a particular
28313 kind of file or process. Targets can be stacked in @dfn{strata},
28314 so that more than one target can potentially respond to a request.
28315 In particular, memory accesses will walk down the stack of targets
28316 until they find a target that is interested in handling that particular
28319 This command prints a short description of each layer that was pushed on
28320 the @dfn{target stack}, starting from the top layer down to the bottom one.
28322 @kindex maint print type
28323 @cindex type chain of a data type
28324 @item maint print type @var{expr}
28325 Print the type chain for a type specified by @var{expr}. The argument
28326 can be either a type name or a symbol. If it is a symbol, the type of
28327 that symbol is described. The type chain produced by this command is
28328 a recursive definition of the data type as stored in @value{GDBN}'s
28329 data structures, including its flags and contained types.
28331 @kindex maint set dwarf2 max-cache-age
28332 @kindex maint show dwarf2 max-cache-age
28333 @item maint set dwarf2 max-cache-age
28334 @itemx maint show dwarf2 max-cache-age
28335 Control the DWARF 2 compilation unit cache.
28337 @cindex DWARF 2 compilation units cache
28338 In object files with inter-compilation-unit references, such as those
28339 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
28340 reader needs to frequently refer to previously read compilation units.
28341 This setting controls how long a compilation unit will remain in the
28342 cache if it is not referenced. A higher limit means that cached
28343 compilation units will be stored in memory longer, and more total
28344 memory will be used. Setting it to zero disables caching, which will
28345 slow down @value{GDBN} startup, but reduce memory consumption.
28347 @kindex maint set profile
28348 @kindex maint show profile
28349 @cindex profiling GDB
28350 @item maint set profile
28351 @itemx maint show profile
28352 Control profiling of @value{GDBN}.
28354 Profiling will be disabled until you use the @samp{maint set profile}
28355 command to enable it. When you enable profiling, the system will begin
28356 collecting timing and execution count data; when you disable profiling or
28357 exit @value{GDBN}, the results will be written to a log file. Remember that
28358 if you use profiling, @value{GDBN} will overwrite the profiling log file
28359 (often called @file{gmon.out}). If you have a record of important profiling
28360 data in a @file{gmon.out} file, be sure to move it to a safe location.
28362 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
28363 compiled with the @samp{-pg} compiler option.
28365 @kindex maint set show-debug-regs
28366 @kindex maint show show-debug-regs
28367 @cindex hardware debug registers
28368 @item maint set show-debug-regs
28369 @itemx maint show show-debug-regs
28370 Control whether to show variables that mirror the hardware debug
28371 registers. Use @code{ON} to enable, @code{OFF} to disable. If
28372 enabled, the debug registers values are shown when @value{GDBN} inserts or
28373 removes a hardware breakpoint or watchpoint, and when the inferior
28374 triggers a hardware-assisted breakpoint or watchpoint.
28376 @kindex maint space
28377 @cindex memory used by commands
28379 Control whether to display memory usage for each command. If set to a
28380 nonzero value, @value{GDBN} will display how much memory each command
28381 took, following the command's own output. This can also be requested
28382 by invoking @value{GDBN} with the @option{--statistics} command-line
28383 switch (@pxref{Mode Options}).
28386 @cindex time of command execution
28388 Control whether to display the execution time for each command. If
28389 set to a nonzero value, @value{GDBN} will display how much time it
28390 took to execute each command, following the command's own output.
28391 The time is not printed for the commands that run the target, since
28392 there's no mechanism currently to compute how much time was spend
28393 by @value{GDBN} and how much time was spend by the program been debugged.
28394 it's not possibly currently
28395 This can also be requested by invoking @value{GDBN} with the
28396 @option{--statistics} command-line switch (@pxref{Mode Options}).
28398 @kindex maint translate-address
28399 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
28400 Find the symbol stored at the location specified by the address
28401 @var{addr} and an optional section name @var{section}. If found,
28402 @value{GDBN} prints the name of the closest symbol and an offset from
28403 the symbol's location to the specified address. This is similar to
28404 the @code{info address} command (@pxref{Symbols}), except that this
28405 command also allows to find symbols in other sections.
28407 If section was not specified, the section in which the symbol was found
28408 is also printed. For dynamically linked executables, the name of
28409 executable or shared library containing the symbol is printed as well.
28413 The following command is useful for non-interactive invocations of
28414 @value{GDBN}, such as in the test suite.
28417 @item set watchdog @var{nsec}
28418 @kindex set watchdog
28419 @cindex watchdog timer
28420 @cindex timeout for commands
28421 Set the maximum number of seconds @value{GDBN} will wait for the
28422 target operation to finish. If this time expires, @value{GDBN}
28423 reports and error and the command is aborted.
28425 @item show watchdog
28426 Show the current setting of the target wait timeout.
28429 @node Remote Protocol
28430 @appendix @value{GDBN} Remote Serial Protocol
28435 * Stop Reply Packets::
28436 * General Query Packets::
28437 * Architecture-Specific Protocol Details::
28438 * Tracepoint Packets::
28439 * Host I/O Packets::
28441 * Notification Packets::
28442 * Remote Non-Stop::
28443 * Packet Acknowledgment::
28445 * File-I/O Remote Protocol Extension::
28446 * Library List Format::
28447 * Memory Map Format::
28448 * Thread List Format::
28454 There may be occasions when you need to know something about the
28455 protocol---for example, if there is only one serial port to your target
28456 machine, you might want your program to do something special if it
28457 recognizes a packet meant for @value{GDBN}.
28459 In the examples below, @samp{->} and @samp{<-} are used to indicate
28460 transmitted and received data, respectively.
28462 @cindex protocol, @value{GDBN} remote serial
28463 @cindex serial protocol, @value{GDBN} remote
28464 @cindex remote serial protocol
28465 All @value{GDBN} commands and responses (other than acknowledgments
28466 and notifications, see @ref{Notification Packets}) are sent as a
28467 @var{packet}. A @var{packet} is introduced with the character
28468 @samp{$}, the actual @var{packet-data}, and the terminating character
28469 @samp{#} followed by a two-digit @var{checksum}:
28472 @code{$}@var{packet-data}@code{#}@var{checksum}
28476 @cindex checksum, for @value{GDBN} remote
28478 The two-digit @var{checksum} is computed as the modulo 256 sum of all
28479 characters between the leading @samp{$} and the trailing @samp{#} (an
28480 eight bit unsigned checksum).
28482 Implementors should note that prior to @value{GDBN} 5.0 the protocol
28483 specification also included an optional two-digit @var{sequence-id}:
28486 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
28489 @cindex sequence-id, for @value{GDBN} remote
28491 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
28492 has never output @var{sequence-id}s. Stubs that handle packets added
28493 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
28495 When either the host or the target machine receives a packet, the first
28496 response expected is an acknowledgment: either @samp{+} (to indicate
28497 the package was received correctly) or @samp{-} (to request
28501 -> @code{$}@var{packet-data}@code{#}@var{checksum}
28506 The @samp{+}/@samp{-} acknowledgments can be disabled
28507 once a connection is established.
28508 @xref{Packet Acknowledgment}, for details.
28510 The host (@value{GDBN}) sends @var{command}s, and the target (the
28511 debugging stub incorporated in your program) sends a @var{response}. In
28512 the case of step and continue @var{command}s, the response is only sent
28513 when the operation has completed, and the target has again stopped all
28514 threads in all attached processes. This is the default all-stop mode
28515 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
28516 execution mode; see @ref{Remote Non-Stop}, for details.
28518 @var{packet-data} consists of a sequence of characters with the
28519 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
28522 @cindex remote protocol, field separator
28523 Fields within the packet should be separated using @samp{,} @samp{;} or
28524 @samp{:}. Except where otherwise noted all numbers are represented in
28525 @sc{hex} with leading zeros suppressed.
28527 Implementors should note that prior to @value{GDBN} 5.0, the character
28528 @samp{:} could not appear as the third character in a packet (as it
28529 would potentially conflict with the @var{sequence-id}).
28531 @cindex remote protocol, binary data
28532 @anchor{Binary Data}
28533 Binary data in most packets is encoded either as two hexadecimal
28534 digits per byte of binary data. This allowed the traditional remote
28535 protocol to work over connections which were only seven-bit clean.
28536 Some packets designed more recently assume an eight-bit clean
28537 connection, and use a more efficient encoding to send and receive
28540 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
28541 as an escape character. Any escaped byte is transmitted as the escape
28542 character followed by the original character XORed with @code{0x20}.
28543 For example, the byte @code{0x7d} would be transmitted as the two
28544 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
28545 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
28546 @samp{@}}) must always be escaped. Responses sent by the stub
28547 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
28548 is not interpreted as the start of a run-length encoded sequence
28551 Response @var{data} can be run-length encoded to save space.
28552 Run-length encoding replaces runs of identical characters with one
28553 instance of the repeated character, followed by a @samp{*} and a
28554 repeat count. The repeat count is itself sent encoded, to avoid
28555 binary characters in @var{data}: a value of @var{n} is sent as
28556 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
28557 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
28558 code 32) for a repeat count of 3. (This is because run-length
28559 encoding starts to win for counts 3 or more.) Thus, for example,
28560 @samp{0* } is a run-length encoding of ``0000'': the space character
28561 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
28564 The printable characters @samp{#} and @samp{$} or with a numeric value
28565 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
28566 seven repeats (@samp{$}) can be expanded using a repeat count of only
28567 five (@samp{"}). For example, @samp{00000000} can be encoded as
28570 The error response returned for some packets includes a two character
28571 error number. That number is not well defined.
28573 @cindex empty response, for unsupported packets
28574 For any @var{command} not supported by the stub, an empty response
28575 (@samp{$#00}) should be returned. That way it is possible to extend the
28576 protocol. A newer @value{GDBN} can tell if a packet is supported based
28579 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
28580 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
28586 The following table provides a complete list of all currently defined
28587 @var{command}s and their corresponding response @var{data}.
28588 @xref{File-I/O Remote Protocol Extension}, for details about the File
28589 I/O extension of the remote protocol.
28591 Each packet's description has a template showing the packet's overall
28592 syntax, followed by an explanation of the packet's meaning. We
28593 include spaces in some of the templates for clarity; these are not
28594 part of the packet's syntax. No @value{GDBN} packet uses spaces to
28595 separate its components. For example, a template like @samp{foo
28596 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
28597 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
28598 @var{baz}. @value{GDBN} does not transmit a space character between the
28599 @samp{foo} and the @var{bar}, or between the @var{bar} and the
28602 @cindex @var{thread-id}, in remote protocol
28603 @anchor{thread-id syntax}
28604 Several packets and replies include a @var{thread-id} field to identify
28605 a thread. Normally these are positive numbers with a target-specific
28606 interpretation, formatted as big-endian hex strings. A @var{thread-id}
28607 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
28610 In addition, the remote protocol supports a multiprocess feature in
28611 which the @var{thread-id} syntax is extended to optionally include both
28612 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
28613 The @var{pid} (process) and @var{tid} (thread) components each have the
28614 format described above: a positive number with target-specific
28615 interpretation formatted as a big-endian hex string, literal @samp{-1}
28616 to indicate all processes or threads (respectively), or @samp{0} to
28617 indicate an arbitrary process or thread. Specifying just a process, as
28618 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
28619 error to specify all processes but a specific thread, such as
28620 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
28621 for those packets and replies explicitly documented to include a process
28622 ID, rather than a @var{thread-id}.
28624 The multiprocess @var{thread-id} syntax extensions are only used if both
28625 @value{GDBN} and the stub report support for the @samp{multiprocess}
28626 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
28629 Note that all packet forms beginning with an upper- or lower-case
28630 letter, other than those described here, are reserved for future use.
28632 Here are the packet descriptions.
28637 @cindex @samp{!} packet
28638 @anchor{extended mode}
28639 Enable extended mode. In extended mode, the remote server is made
28640 persistent. The @samp{R} packet is used to restart the program being
28646 The remote target both supports and has enabled extended mode.
28650 @cindex @samp{?} packet
28651 Indicate the reason the target halted. The reply is the same as for
28652 step and continue. This packet has a special interpretation when the
28653 target is in non-stop mode; see @ref{Remote Non-Stop}.
28656 @xref{Stop Reply Packets}, for the reply specifications.
28658 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
28659 @cindex @samp{A} packet
28660 Initialized @code{argv[]} array passed into program. @var{arglen}
28661 specifies the number of bytes in the hex encoded byte stream
28662 @var{arg}. See @code{gdbserver} for more details.
28667 The arguments were set.
28673 @cindex @samp{b} packet
28674 (Don't use this packet; its behavior is not well-defined.)
28675 Change the serial line speed to @var{baud}.
28677 JTC: @emph{When does the transport layer state change? When it's
28678 received, or after the ACK is transmitted. In either case, there are
28679 problems if the command or the acknowledgment packet is dropped.}
28681 Stan: @emph{If people really wanted to add something like this, and get
28682 it working for the first time, they ought to modify ser-unix.c to send
28683 some kind of out-of-band message to a specially-setup stub and have the
28684 switch happen "in between" packets, so that from remote protocol's point
28685 of view, nothing actually happened.}
28687 @item B @var{addr},@var{mode}
28688 @cindex @samp{B} packet
28689 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
28690 breakpoint at @var{addr}.
28692 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
28693 (@pxref{insert breakpoint or watchpoint packet}).
28695 @cindex @samp{bc} packet
28698 Backward continue. Execute the target system in reverse. No parameter.
28699 @xref{Reverse Execution}, for more information.
28702 @xref{Stop Reply Packets}, for the reply specifications.
28704 @cindex @samp{bs} packet
28707 Backward single step. Execute one instruction in reverse. No parameter.
28708 @xref{Reverse Execution}, for more information.
28711 @xref{Stop Reply Packets}, for the reply specifications.
28713 @item c @r{[}@var{addr}@r{]}
28714 @cindex @samp{c} packet
28715 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
28716 resume at current address.
28719 @xref{Stop Reply Packets}, for the reply specifications.
28721 @item C @var{sig}@r{[};@var{addr}@r{]}
28722 @cindex @samp{C} packet
28723 Continue with signal @var{sig} (hex signal number). If
28724 @samp{;@var{addr}} is omitted, resume at same address.
28727 @xref{Stop Reply Packets}, for the reply specifications.
28730 @cindex @samp{d} packet
28733 Don't use this packet; instead, define a general set packet
28734 (@pxref{General Query Packets}).
28738 @cindex @samp{D} packet
28739 The first form of the packet is used to detach @value{GDBN} from the
28740 remote system. It is sent to the remote target
28741 before @value{GDBN} disconnects via the @code{detach} command.
28743 The second form, including a process ID, is used when multiprocess
28744 protocol extensions are enabled (@pxref{multiprocess extensions}), to
28745 detach only a specific process. The @var{pid} is specified as a
28746 big-endian hex string.
28756 @item F @var{RC},@var{EE},@var{CF};@var{XX}
28757 @cindex @samp{F} packet
28758 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
28759 This is part of the File-I/O protocol extension. @xref{File-I/O
28760 Remote Protocol Extension}, for the specification.
28763 @anchor{read registers packet}
28764 @cindex @samp{g} packet
28765 Read general registers.
28769 @item @var{XX@dots{}}
28770 Each byte of register data is described by two hex digits. The bytes
28771 with the register are transmitted in target byte order. The size of
28772 each register and their position within the @samp{g} packet are
28773 determined by the @value{GDBN} internal gdbarch functions
28774 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
28775 specification of several standard @samp{g} packets is specified below.
28780 @item G @var{XX@dots{}}
28781 @cindex @samp{G} packet
28782 Write general registers. @xref{read registers packet}, for a
28783 description of the @var{XX@dots{}} data.
28793 @item H @var{c} @var{thread-id}
28794 @cindex @samp{H} packet
28795 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
28796 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
28797 should be @samp{c} for step and continue operations, @samp{g} for other
28798 operations. The thread designator @var{thread-id} has the format and
28799 interpretation described in @ref{thread-id syntax}.
28810 @c 'H': How restrictive (or permissive) is the thread model. If a
28811 @c thread is selected and stopped, are other threads allowed
28812 @c to continue to execute? As I mentioned above, I think the
28813 @c semantics of each command when a thread is selected must be
28814 @c described. For example:
28816 @c 'g': If the stub supports threads and a specific thread is
28817 @c selected, returns the register block from that thread;
28818 @c otherwise returns current registers.
28820 @c 'G' If the stub supports threads and a specific thread is
28821 @c selected, sets the registers of the register block of
28822 @c that thread; otherwise sets current registers.
28824 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
28825 @anchor{cycle step packet}
28826 @cindex @samp{i} packet
28827 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
28828 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
28829 step starting at that address.
28832 @cindex @samp{I} packet
28833 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
28837 @cindex @samp{k} packet
28840 FIXME: @emph{There is no description of how to operate when a specific
28841 thread context has been selected (i.e.@: does 'k' kill only that
28844 @item m @var{addr},@var{length}
28845 @cindex @samp{m} packet
28846 Read @var{length} bytes of memory starting at address @var{addr}.
28847 Note that @var{addr} may not be aligned to any particular boundary.
28849 The stub need not use any particular size or alignment when gathering
28850 data from memory for the response; even if @var{addr} is word-aligned
28851 and @var{length} is a multiple of the word size, the stub is free to
28852 use byte accesses, or not. For this reason, this packet may not be
28853 suitable for accessing memory-mapped I/O devices.
28854 @cindex alignment of remote memory accesses
28855 @cindex size of remote memory accesses
28856 @cindex memory, alignment and size of remote accesses
28860 @item @var{XX@dots{}}
28861 Memory contents; each byte is transmitted as a two-digit hexadecimal
28862 number. The reply may contain fewer bytes than requested if the
28863 server was able to read only part of the region of memory.
28868 @item M @var{addr},@var{length}:@var{XX@dots{}}
28869 @cindex @samp{M} packet
28870 Write @var{length} bytes of memory starting at address @var{addr}.
28871 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
28872 hexadecimal number.
28879 for an error (this includes the case where only part of the data was
28884 @cindex @samp{p} packet
28885 Read the value of register @var{n}; @var{n} is in hex.
28886 @xref{read registers packet}, for a description of how the returned
28887 register value is encoded.
28891 @item @var{XX@dots{}}
28892 the register's value
28896 Indicating an unrecognized @var{query}.
28899 @item P @var{n@dots{}}=@var{r@dots{}}
28900 @anchor{write register packet}
28901 @cindex @samp{P} packet
28902 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
28903 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
28904 digits for each byte in the register (target byte order).
28914 @item q @var{name} @var{params}@dots{}
28915 @itemx Q @var{name} @var{params}@dots{}
28916 @cindex @samp{q} packet
28917 @cindex @samp{Q} packet
28918 General query (@samp{q}) and set (@samp{Q}). These packets are
28919 described fully in @ref{General Query Packets}.
28922 @cindex @samp{r} packet
28923 Reset the entire system.
28925 Don't use this packet; use the @samp{R} packet instead.
28928 @cindex @samp{R} packet
28929 Restart the program being debugged. @var{XX}, while needed, is ignored.
28930 This packet is only available in extended mode (@pxref{extended mode}).
28932 The @samp{R} packet has no reply.
28934 @item s @r{[}@var{addr}@r{]}
28935 @cindex @samp{s} packet
28936 Single step. @var{addr} is the address at which to resume. If
28937 @var{addr} is omitted, resume at same address.
28940 @xref{Stop Reply Packets}, for the reply specifications.
28942 @item S @var{sig}@r{[};@var{addr}@r{]}
28943 @anchor{step with signal packet}
28944 @cindex @samp{S} packet
28945 Step with signal. This is analogous to the @samp{C} packet, but
28946 requests a single-step, rather than a normal resumption of execution.
28949 @xref{Stop Reply Packets}, for the reply specifications.
28951 @item t @var{addr}:@var{PP},@var{MM}
28952 @cindex @samp{t} packet
28953 Search backwards starting at address @var{addr} for a match with pattern
28954 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
28955 @var{addr} must be at least 3 digits.
28957 @item T @var{thread-id}
28958 @cindex @samp{T} packet
28959 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
28964 thread is still alive
28970 Packets starting with @samp{v} are identified by a multi-letter name,
28971 up to the first @samp{;} or @samp{?} (or the end of the packet).
28973 @item vAttach;@var{pid}
28974 @cindex @samp{vAttach} packet
28975 Attach to a new process with the specified process ID @var{pid}.
28976 The process ID is a
28977 hexadecimal integer identifying the process. In all-stop mode, all
28978 threads in the attached process are stopped; in non-stop mode, it may be
28979 attached without being stopped if that is supported by the target.
28981 @c In non-stop mode, on a successful vAttach, the stub should set the
28982 @c current thread to a thread of the newly-attached process. After
28983 @c attaching, GDB queries for the attached process's thread ID with qC.
28984 @c Also note that, from a user perspective, whether or not the
28985 @c target is stopped on attach in non-stop mode depends on whether you
28986 @c use the foreground or background version of the attach command, not
28987 @c on what vAttach does; GDB does the right thing with respect to either
28988 @c stopping or restarting threads.
28990 This packet is only available in extended mode (@pxref{extended mode}).
28996 @item @r{Any stop packet}
28997 for success in all-stop mode (@pxref{Stop Reply Packets})
28999 for success in non-stop mode (@pxref{Remote Non-Stop})
29002 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
29003 @cindex @samp{vCont} packet
29004 Resume the inferior, specifying different actions for each thread.
29005 If an action is specified with no @var{thread-id}, then it is applied to any
29006 threads that don't have a specific action specified; if no default action is
29007 specified then other threads should remain stopped in all-stop mode and
29008 in their current state in non-stop mode.
29009 Specifying multiple
29010 default actions is an error; specifying no actions is also an error.
29011 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
29013 Currently supported actions are:
29019 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
29023 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
29028 The optional argument @var{addr} normally associated with the
29029 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
29030 not supported in @samp{vCont}.
29032 The @samp{t} action is only relevant in non-stop mode
29033 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
29034 A stop reply should be generated for any affected thread not already stopped.
29035 When a thread is stopped by means of a @samp{t} action,
29036 the corresponding stop reply should indicate that the thread has stopped with
29037 signal @samp{0}, regardless of whether the target uses some other signal
29038 as an implementation detail.
29041 @xref{Stop Reply Packets}, for the reply specifications.
29044 @cindex @samp{vCont?} packet
29045 Request a list of actions supported by the @samp{vCont} packet.
29049 @item vCont@r{[};@var{action}@dots{}@r{]}
29050 The @samp{vCont} packet is supported. Each @var{action} is a supported
29051 command in the @samp{vCont} packet.
29053 The @samp{vCont} packet is not supported.
29056 @item vFile:@var{operation}:@var{parameter}@dots{}
29057 @cindex @samp{vFile} packet
29058 Perform a file operation on the target system. For details,
29059 see @ref{Host I/O Packets}.
29061 @item vFlashErase:@var{addr},@var{length}
29062 @cindex @samp{vFlashErase} packet
29063 Direct the stub to erase @var{length} bytes of flash starting at
29064 @var{addr}. The region may enclose any number of flash blocks, but
29065 its start and end must fall on block boundaries, as indicated by the
29066 flash block size appearing in the memory map (@pxref{Memory Map
29067 Format}). @value{GDBN} groups flash memory programming operations
29068 together, and sends a @samp{vFlashDone} request after each group; the
29069 stub is allowed to delay erase operation until the @samp{vFlashDone}
29070 packet is received.
29072 The stub must support @samp{vCont} if it reports support for
29073 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
29074 this case @samp{vCont} actions can be specified to apply to all threads
29075 in a process by using the @samp{p@var{pid}.-1} form of the
29086 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
29087 @cindex @samp{vFlashWrite} packet
29088 Direct the stub to write data to flash address @var{addr}. The data
29089 is passed in binary form using the same encoding as for the @samp{X}
29090 packet (@pxref{Binary Data}). The memory ranges specified by
29091 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
29092 not overlap, and must appear in order of increasing addresses
29093 (although @samp{vFlashErase} packets for higher addresses may already
29094 have been received; the ordering is guaranteed only between
29095 @samp{vFlashWrite} packets). If a packet writes to an address that was
29096 neither erased by a preceding @samp{vFlashErase} packet nor by some other
29097 target-specific method, the results are unpredictable.
29105 for vFlashWrite addressing non-flash memory
29111 @cindex @samp{vFlashDone} packet
29112 Indicate to the stub that flash programming operation is finished.
29113 The stub is permitted to delay or batch the effects of a group of
29114 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
29115 @samp{vFlashDone} packet is received. The contents of the affected
29116 regions of flash memory are unpredictable until the @samp{vFlashDone}
29117 request is completed.
29119 @item vKill;@var{pid}
29120 @cindex @samp{vKill} packet
29121 Kill the process with the specified process ID. @var{pid} is a
29122 hexadecimal integer identifying the process. This packet is used in
29123 preference to @samp{k} when multiprocess protocol extensions are
29124 supported; see @ref{multiprocess extensions}.
29134 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
29135 @cindex @samp{vRun} packet
29136 Run the program @var{filename}, passing it each @var{argument} on its
29137 command line. The file and arguments are hex-encoded strings. If
29138 @var{filename} is an empty string, the stub may use a default program
29139 (e.g.@: the last program run). The program is created in the stopped
29142 @c FIXME: What about non-stop mode?
29144 This packet is only available in extended mode (@pxref{extended mode}).
29150 @item @r{Any stop packet}
29151 for success (@pxref{Stop Reply Packets})
29155 @anchor{vStopped packet}
29156 @cindex @samp{vStopped} packet
29158 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
29159 reply and prompt for the stub to report another one.
29163 @item @r{Any stop packet}
29164 if there is another unreported stop event (@pxref{Stop Reply Packets})
29166 if there are no unreported stop events
29169 @item X @var{addr},@var{length}:@var{XX@dots{}}
29171 @cindex @samp{X} packet
29172 Write data to memory, where the data is transmitted in binary.
29173 @var{addr} is address, @var{length} is number of bytes,
29174 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
29184 @item z @var{type},@var{addr},@var{kind}
29185 @itemx Z @var{type},@var{addr},@var{kind}
29186 @anchor{insert breakpoint or watchpoint packet}
29187 @cindex @samp{z} packet
29188 @cindex @samp{Z} packets
29189 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
29190 watchpoint starting at address @var{address} of kind @var{kind}.
29192 Each breakpoint and watchpoint packet @var{type} is documented
29195 @emph{Implementation notes: A remote target shall return an empty string
29196 for an unrecognized breakpoint or watchpoint packet @var{type}. A
29197 remote target shall support either both or neither of a given
29198 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
29199 avoid potential problems with duplicate packets, the operations should
29200 be implemented in an idempotent way.}
29202 @item z0,@var{addr},@var{kind}
29203 @itemx Z0,@var{addr},@var{kind}
29204 @cindex @samp{z0} packet
29205 @cindex @samp{Z0} packet
29206 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
29207 @var{addr} of type @var{kind}.
29209 A memory breakpoint is implemented by replacing the instruction at
29210 @var{addr} with a software breakpoint or trap instruction. The
29211 @var{kind} is target-specific and typically indicates the size of
29212 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
29213 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
29214 architectures have additional meanings for @var{kind};
29215 see @ref{Architecture-Specific Protocol Details}.
29217 @emph{Implementation note: It is possible for a target to copy or move
29218 code that contains memory breakpoints (e.g., when implementing
29219 overlays). The behavior of this packet, in the presence of such a
29220 target, is not defined.}
29232 @item z1,@var{addr},@var{kind}
29233 @itemx Z1,@var{addr},@var{kind}
29234 @cindex @samp{z1} packet
29235 @cindex @samp{Z1} packet
29236 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
29237 address @var{addr}.
29239 A hardware breakpoint is implemented using a mechanism that is not
29240 dependant on being able to modify the target's memory. @var{kind}
29241 has the same meaning as in @samp{Z0} packets.
29243 @emph{Implementation note: A hardware breakpoint is not affected by code
29256 @item z2,@var{addr},@var{kind}
29257 @itemx Z2,@var{addr},@var{kind}
29258 @cindex @samp{z2} packet
29259 @cindex @samp{Z2} packet
29260 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
29261 @var{kind} is interpreted as the number of bytes to watch.
29273 @item z3,@var{addr},@var{kind}
29274 @itemx Z3,@var{addr},@var{kind}
29275 @cindex @samp{z3} packet
29276 @cindex @samp{Z3} packet
29277 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
29278 @var{kind} is interpreted as the number of bytes to watch.
29290 @item z4,@var{addr},@var{kind}
29291 @itemx Z4,@var{addr},@var{kind}
29292 @cindex @samp{z4} packet
29293 @cindex @samp{Z4} packet
29294 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
29295 @var{kind} is interpreted as the number of bytes to watch.
29309 @node Stop Reply Packets
29310 @section Stop Reply Packets
29311 @cindex stop reply packets
29313 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
29314 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
29315 receive any of the below as a reply. Except for @samp{?}
29316 and @samp{vStopped}, that reply is only returned
29317 when the target halts. In the below the exact meaning of @dfn{signal
29318 number} is defined by the header @file{include/gdb/signals.h} in the
29319 @value{GDBN} source code.
29321 As in the description of request packets, we include spaces in the
29322 reply templates for clarity; these are not part of the reply packet's
29323 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
29329 The program received signal number @var{AA} (a two-digit hexadecimal
29330 number). This is equivalent to a @samp{T} response with no
29331 @var{n}:@var{r} pairs.
29333 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
29334 @cindex @samp{T} packet reply
29335 The program received signal number @var{AA} (a two-digit hexadecimal
29336 number). This is equivalent to an @samp{S} response, except that the
29337 @samp{@var{n}:@var{r}} pairs can carry values of important registers
29338 and other information directly in the stop reply packet, reducing
29339 round-trip latency. Single-step and breakpoint traps are reported
29340 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
29344 If @var{n} is a hexadecimal number, it is a register number, and the
29345 corresponding @var{r} gives that register's value. @var{r} is a
29346 series of bytes in target byte order, with each byte given by a
29347 two-digit hex number.
29350 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
29351 the stopped thread, as specified in @ref{thread-id syntax}.
29354 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
29355 the core on which the stop event was detected.
29358 If @var{n} is a recognized @dfn{stop reason}, it describes a more
29359 specific event that stopped the target. The currently defined stop
29360 reasons are listed below. @var{aa} should be @samp{05}, the trap
29361 signal. At most one stop reason should be present.
29364 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
29365 and go on to the next; this allows us to extend the protocol in the
29369 The currently defined stop reasons are:
29375 The packet indicates a watchpoint hit, and @var{r} is the data address, in
29378 @cindex shared library events, remote reply
29380 The packet indicates that the loaded libraries have changed.
29381 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
29382 list of loaded libraries. @var{r} is ignored.
29384 @cindex replay log events, remote reply
29386 The packet indicates that the target cannot continue replaying
29387 logged execution events, because it has reached the end (or the
29388 beginning when executing backward) of the log. The value of @var{r}
29389 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
29390 for more information.
29394 @itemx W @var{AA} ; process:@var{pid}
29395 The process exited, and @var{AA} is the exit status. This is only
29396 applicable to certain targets.
29398 The second form of the response, including the process ID of the exited
29399 process, can be used only when @value{GDBN} has reported support for
29400 multiprocess protocol extensions; see @ref{multiprocess extensions}.
29401 The @var{pid} is formatted as a big-endian hex string.
29404 @itemx X @var{AA} ; process:@var{pid}
29405 The process terminated with signal @var{AA}.
29407 The second form of the response, including the process ID of the
29408 terminated process, can be used only when @value{GDBN} has reported
29409 support for multiprocess protocol extensions; see @ref{multiprocess
29410 extensions}. The @var{pid} is formatted as a big-endian hex string.
29412 @item O @var{XX}@dots{}
29413 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
29414 written as the program's console output. This can happen at any time
29415 while the program is running and the debugger should continue to wait
29416 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
29418 @item F @var{call-id},@var{parameter}@dots{}
29419 @var{call-id} is the identifier which says which host system call should
29420 be called. This is just the name of the function. Translation into the
29421 correct system call is only applicable as it's defined in @value{GDBN}.
29422 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
29425 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
29426 this very system call.
29428 The target replies with this packet when it expects @value{GDBN} to
29429 call a host system call on behalf of the target. @value{GDBN} replies
29430 with an appropriate @samp{F} packet and keeps up waiting for the next
29431 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
29432 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
29433 Protocol Extension}, for more details.
29437 @node General Query Packets
29438 @section General Query Packets
29439 @cindex remote query requests
29441 Packets starting with @samp{q} are @dfn{general query packets};
29442 packets starting with @samp{Q} are @dfn{general set packets}. General
29443 query and set packets are a semi-unified form for retrieving and
29444 sending information to and from the stub.
29446 The initial letter of a query or set packet is followed by a name
29447 indicating what sort of thing the packet applies to. For example,
29448 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
29449 definitions with the stub. These packet names follow some
29454 The name must not contain commas, colons or semicolons.
29456 Most @value{GDBN} query and set packets have a leading upper case
29459 The names of custom vendor packets should use a company prefix, in
29460 lower case, followed by a period. For example, packets designed at
29461 the Acme Corporation might begin with @samp{qacme.foo} (for querying
29462 foos) or @samp{Qacme.bar} (for setting bars).
29465 The name of a query or set packet should be separated from any
29466 parameters by a @samp{:}; the parameters themselves should be
29467 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
29468 full packet name, and check for a separator or the end of the packet,
29469 in case two packet names share a common prefix. New packets should not begin
29470 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
29471 packets predate these conventions, and have arguments without any terminator
29472 for the packet name; we suspect they are in widespread use in places that
29473 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
29474 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
29477 Like the descriptions of the other packets, each description here
29478 has a template showing the packet's overall syntax, followed by an
29479 explanation of the packet's meaning. We include spaces in some of the
29480 templates for clarity; these are not part of the packet's syntax. No
29481 @value{GDBN} packet uses spaces to separate its components.
29483 Here are the currently defined query and set packets:
29488 @cindex current thread, remote request
29489 @cindex @samp{qC} packet
29490 Return the current thread ID.
29494 @item QC @var{thread-id}
29495 Where @var{thread-id} is a thread ID as documented in
29496 @ref{thread-id syntax}.
29497 @item @r{(anything else)}
29498 Any other reply implies the old thread ID.
29501 @item qCRC:@var{addr},@var{length}
29502 @cindex CRC of memory block, remote request
29503 @cindex @samp{qCRC} packet
29504 Compute the CRC checksum of a block of memory using CRC-32 defined in
29505 IEEE 802.3. The CRC is computed byte at a time, taking the most
29506 significant bit of each byte first. The initial pattern code
29507 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
29509 @emph{Note:} This is the same CRC used in validating separate debug
29510 files (@pxref{Separate Debug Files, , Debugging Information in Separate
29511 Files}). However the algorithm is slightly different. When validating
29512 separate debug files, the CRC is computed taking the @emph{least}
29513 significant bit of each byte first, and the final result is inverted to
29514 detect trailing zeros.
29519 An error (such as memory fault)
29520 @item C @var{crc32}
29521 The specified memory region's checksum is @var{crc32}.
29525 @itemx qsThreadInfo
29526 @cindex list active threads, remote request
29527 @cindex @samp{qfThreadInfo} packet
29528 @cindex @samp{qsThreadInfo} packet
29529 Obtain a list of all active thread IDs from the target (OS). Since there
29530 may be too many active threads to fit into one reply packet, this query
29531 works iteratively: it may require more than one query/reply sequence to
29532 obtain the entire list of threads. The first query of the sequence will
29533 be the @samp{qfThreadInfo} query; subsequent queries in the
29534 sequence will be the @samp{qsThreadInfo} query.
29536 NOTE: This packet replaces the @samp{qL} query (see below).
29540 @item m @var{thread-id}
29542 @item m @var{thread-id},@var{thread-id}@dots{}
29543 a comma-separated list of thread IDs
29545 (lower case letter @samp{L}) denotes end of list.
29548 In response to each query, the target will reply with a list of one or
29549 more thread IDs, separated by commas.
29550 @value{GDBN} will respond to each reply with a request for more thread
29551 ids (using the @samp{qs} form of the query), until the target responds
29552 with @samp{l} (lower-case el, for @dfn{last}).
29553 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
29556 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
29557 @cindex get thread-local storage address, remote request
29558 @cindex @samp{qGetTLSAddr} packet
29559 Fetch the address associated with thread local storage specified
29560 by @var{thread-id}, @var{offset}, and @var{lm}.
29562 @var{thread-id} is the thread ID associated with the
29563 thread for which to fetch the TLS address. @xref{thread-id syntax}.
29565 @var{offset} is the (big endian, hex encoded) offset associated with the
29566 thread local variable. (This offset is obtained from the debug
29567 information associated with the variable.)
29569 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
29570 the load module associated with the thread local storage. For example,
29571 a @sc{gnu}/Linux system will pass the link map address of the shared
29572 object associated with the thread local storage under consideration.
29573 Other operating environments may choose to represent the load module
29574 differently, so the precise meaning of this parameter will vary.
29578 @item @var{XX}@dots{}
29579 Hex encoded (big endian) bytes representing the address of the thread
29580 local storage requested.
29583 An error occurred. @var{nn} are hex digits.
29586 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
29589 @item qL @var{startflag} @var{threadcount} @var{nextthread}
29590 Obtain thread information from RTOS. Where: @var{startflag} (one hex
29591 digit) is one to indicate the first query and zero to indicate a
29592 subsequent query; @var{threadcount} (two hex digits) is the maximum
29593 number of threads the response packet can contain; and @var{nextthread}
29594 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
29595 returned in the response as @var{argthread}.
29597 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
29601 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
29602 Where: @var{count} (two hex digits) is the number of threads being
29603 returned; @var{done} (one hex digit) is zero to indicate more threads
29604 and one indicates no further threads; @var{argthreadid} (eight hex
29605 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
29606 is a sequence of thread IDs from the target. @var{threadid} (eight hex
29607 digits). See @code{remote.c:parse_threadlist_response()}.
29611 @cindex section offsets, remote request
29612 @cindex @samp{qOffsets} packet
29613 Get section offsets that the target used when relocating the downloaded
29618 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
29619 Relocate the @code{Text} section by @var{xxx} from its original address.
29620 Relocate the @code{Data} section by @var{yyy} from its original address.
29621 If the object file format provides segment information (e.g.@: @sc{elf}
29622 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
29623 segments by the supplied offsets.
29625 @emph{Note: while a @code{Bss} offset may be included in the response,
29626 @value{GDBN} ignores this and instead applies the @code{Data} offset
29627 to the @code{Bss} section.}
29629 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
29630 Relocate the first segment of the object file, which conventionally
29631 contains program code, to a starting address of @var{xxx}. If
29632 @samp{DataSeg} is specified, relocate the second segment, which
29633 conventionally contains modifiable data, to a starting address of
29634 @var{yyy}. @value{GDBN} will report an error if the object file
29635 does not contain segment information, or does not contain at least
29636 as many segments as mentioned in the reply. Extra segments are
29637 kept at fixed offsets relative to the last relocated segment.
29640 @item qP @var{mode} @var{thread-id}
29641 @cindex thread information, remote request
29642 @cindex @samp{qP} packet
29643 Returns information on @var{thread-id}. Where: @var{mode} is a hex
29644 encoded 32 bit mode; @var{thread-id} is a thread ID
29645 (@pxref{thread-id syntax}).
29647 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
29650 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
29654 @cindex non-stop mode, remote request
29655 @cindex @samp{QNonStop} packet
29657 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
29658 @xref{Remote Non-Stop}, for more information.
29663 The request succeeded.
29666 An error occurred. @var{nn} are hex digits.
29669 An empty reply indicates that @samp{QNonStop} is not supported by
29673 This packet is not probed by default; the remote stub must request it,
29674 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29675 Use of this packet is controlled by the @code{set non-stop} command;
29676 @pxref{Non-Stop Mode}.
29678 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
29679 @cindex pass signals to inferior, remote request
29680 @cindex @samp{QPassSignals} packet
29681 @anchor{QPassSignals}
29682 Each listed @var{signal} should be passed directly to the inferior process.
29683 Signals are numbered identically to continue packets and stop replies
29684 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
29685 strictly greater than the previous item. These signals do not need to stop
29686 the inferior, or be reported to @value{GDBN}. All other signals should be
29687 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
29688 combine; any earlier @samp{QPassSignals} list is completely replaced by the
29689 new list. This packet improves performance when using @samp{handle
29690 @var{signal} nostop noprint pass}.
29695 The request succeeded.
29698 An error occurred. @var{nn} are hex digits.
29701 An empty reply indicates that @samp{QPassSignals} is not supported by
29705 Use of this packet is controlled by the @code{set remote pass-signals}
29706 command (@pxref{Remote Configuration, set remote pass-signals}).
29707 This packet is not probed by default; the remote stub must request it,
29708 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29710 @item qRcmd,@var{command}
29711 @cindex execute remote command, remote request
29712 @cindex @samp{qRcmd} packet
29713 @var{command} (hex encoded) is passed to the local interpreter for
29714 execution. Invalid commands should be reported using the output
29715 string. Before the final result packet, the target may also respond
29716 with a number of intermediate @samp{O@var{output}} console output
29717 packets. @emph{Implementors should note that providing access to a
29718 stubs's interpreter may have security implications}.
29723 A command response with no output.
29725 A command response with the hex encoded output string @var{OUTPUT}.
29727 Indicate a badly formed request.
29729 An empty reply indicates that @samp{qRcmd} is not recognized.
29732 (Note that the @code{qRcmd} packet's name is separated from the
29733 command by a @samp{,}, not a @samp{:}, contrary to the naming
29734 conventions above. Please don't use this packet as a model for new
29737 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
29738 @cindex searching memory, in remote debugging
29739 @cindex @samp{qSearch:memory} packet
29740 @anchor{qSearch memory}
29741 Search @var{length} bytes at @var{address} for @var{search-pattern}.
29742 @var{address} and @var{length} are encoded in hex.
29743 @var{search-pattern} is a sequence of bytes, hex encoded.
29748 The pattern was not found.
29750 The pattern was found at @var{address}.
29752 A badly formed request or an error was encountered while searching memory.
29754 An empty reply indicates that @samp{qSearch:memory} is not recognized.
29757 @item QStartNoAckMode
29758 @cindex @samp{QStartNoAckMode} packet
29759 @anchor{QStartNoAckMode}
29760 Request that the remote stub disable the normal @samp{+}/@samp{-}
29761 protocol acknowledgments (@pxref{Packet Acknowledgment}).
29766 The stub has switched to no-acknowledgment mode.
29767 @value{GDBN} acknowledges this reponse,
29768 but neither the stub nor @value{GDBN} shall send or expect further
29769 @samp{+}/@samp{-} acknowledgments in the current connection.
29771 An empty reply indicates that the stub does not support no-acknowledgment mode.
29774 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
29775 @cindex supported packets, remote query
29776 @cindex features of the remote protocol
29777 @cindex @samp{qSupported} packet
29778 @anchor{qSupported}
29779 Tell the remote stub about features supported by @value{GDBN}, and
29780 query the stub for features it supports. This packet allows
29781 @value{GDBN} and the remote stub to take advantage of each others'
29782 features. @samp{qSupported} also consolidates multiple feature probes
29783 at startup, to improve @value{GDBN} performance---a single larger
29784 packet performs better than multiple smaller probe packets on
29785 high-latency links. Some features may enable behavior which must not
29786 be on by default, e.g.@: because it would confuse older clients or
29787 stubs. Other features may describe packets which could be
29788 automatically probed for, but are not. These features must be
29789 reported before @value{GDBN} will use them. This ``default
29790 unsupported'' behavior is not appropriate for all packets, but it
29791 helps to keep the initial connection time under control with new
29792 versions of @value{GDBN} which support increasing numbers of packets.
29796 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
29797 The stub supports or does not support each returned @var{stubfeature},
29798 depending on the form of each @var{stubfeature} (see below for the
29801 An empty reply indicates that @samp{qSupported} is not recognized,
29802 or that no features needed to be reported to @value{GDBN}.
29805 The allowed forms for each feature (either a @var{gdbfeature} in the
29806 @samp{qSupported} packet, or a @var{stubfeature} in the response)
29810 @item @var{name}=@var{value}
29811 The remote protocol feature @var{name} is supported, and associated
29812 with the specified @var{value}. The format of @var{value} depends
29813 on the feature, but it must not include a semicolon.
29815 The remote protocol feature @var{name} is supported, and does not
29816 need an associated value.
29818 The remote protocol feature @var{name} is not supported.
29820 The remote protocol feature @var{name} may be supported, and
29821 @value{GDBN} should auto-detect support in some other way when it is
29822 needed. This form will not be used for @var{gdbfeature} notifications,
29823 but may be used for @var{stubfeature} responses.
29826 Whenever the stub receives a @samp{qSupported} request, the
29827 supplied set of @value{GDBN} features should override any previous
29828 request. This allows @value{GDBN} to put the stub in a known
29829 state, even if the stub had previously been communicating with
29830 a different version of @value{GDBN}.
29832 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
29837 This feature indicates whether @value{GDBN} supports multiprocess
29838 extensions to the remote protocol. @value{GDBN} does not use such
29839 extensions unless the stub also reports that it supports them by
29840 including @samp{multiprocess+} in its @samp{qSupported} reply.
29841 @xref{multiprocess extensions}, for details.
29844 Stubs should ignore any unknown values for
29845 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
29846 packet supports receiving packets of unlimited length (earlier
29847 versions of @value{GDBN} may reject overly long responses). Additional values
29848 for @var{gdbfeature} may be defined in the future to let the stub take
29849 advantage of new features in @value{GDBN}, e.g.@: incompatible
29850 improvements in the remote protocol---the @samp{multiprocess} feature is
29851 an example of such a feature. The stub's reply should be independent
29852 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
29853 describes all the features it supports, and then the stub replies with
29854 all the features it supports.
29856 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
29857 responses, as long as each response uses one of the standard forms.
29859 Some features are flags. A stub which supports a flag feature
29860 should respond with a @samp{+} form response. Other features
29861 require values, and the stub should respond with an @samp{=}
29864 Each feature has a default value, which @value{GDBN} will use if
29865 @samp{qSupported} is not available or if the feature is not mentioned
29866 in the @samp{qSupported} response. The default values are fixed; a
29867 stub is free to omit any feature responses that match the defaults.
29869 Not all features can be probed, but for those which can, the probing
29870 mechanism is useful: in some cases, a stub's internal
29871 architecture may not allow the protocol layer to know some information
29872 about the underlying target in advance. This is especially common in
29873 stubs which may be configured for multiple targets.
29875 These are the currently defined stub features and their properties:
29877 @multitable @columnfractions 0.35 0.2 0.12 0.2
29878 @c NOTE: The first row should be @headitem, but we do not yet require
29879 @c a new enough version of Texinfo (4.7) to use @headitem.
29881 @tab Value Required
29885 @item @samp{PacketSize}
29890 @item @samp{qXfer:auxv:read}
29895 @item @samp{qXfer:features:read}
29900 @item @samp{qXfer:libraries:read}
29905 @item @samp{qXfer:memory-map:read}
29910 @item @samp{qXfer:spu:read}
29915 @item @samp{qXfer:spu:write}
29920 @item @samp{qXfer:siginfo:read}
29925 @item @samp{qXfer:siginfo:write}
29930 @item @samp{qXfer:threads:read}
29936 @item @samp{QNonStop}
29941 @item @samp{QPassSignals}
29946 @item @samp{QStartNoAckMode}
29951 @item @samp{multiprocess}
29956 @item @samp{ConditionalTracepoints}
29961 @item @samp{ReverseContinue}
29966 @item @samp{ReverseStep}
29973 These are the currently defined stub features, in more detail:
29976 @cindex packet size, remote protocol
29977 @item PacketSize=@var{bytes}
29978 The remote stub can accept packets up to at least @var{bytes} in
29979 length. @value{GDBN} will send packets up to this size for bulk
29980 transfers, and will never send larger packets. This is a limit on the
29981 data characters in the packet, including the frame and checksum.
29982 There is no trailing NUL byte in a remote protocol packet; if the stub
29983 stores packets in a NUL-terminated format, it should allow an extra
29984 byte in its buffer for the NUL. If this stub feature is not supported,
29985 @value{GDBN} guesses based on the size of the @samp{g} packet response.
29987 @item qXfer:auxv:read
29988 The remote stub understands the @samp{qXfer:auxv:read} packet
29989 (@pxref{qXfer auxiliary vector read}).
29991 @item qXfer:features:read
29992 The remote stub understands the @samp{qXfer:features:read} packet
29993 (@pxref{qXfer target description read}).
29995 @item qXfer:libraries:read
29996 The remote stub understands the @samp{qXfer:libraries:read} packet
29997 (@pxref{qXfer library list read}).
29999 @item qXfer:memory-map:read
30000 The remote stub understands the @samp{qXfer:memory-map:read} packet
30001 (@pxref{qXfer memory map read}).
30003 @item qXfer:spu:read
30004 The remote stub understands the @samp{qXfer:spu:read} packet
30005 (@pxref{qXfer spu read}).
30007 @item qXfer:spu:write
30008 The remote stub understands the @samp{qXfer:spu:write} packet
30009 (@pxref{qXfer spu write}).
30011 @item qXfer:siginfo:read
30012 The remote stub understands the @samp{qXfer:siginfo:read} packet
30013 (@pxref{qXfer siginfo read}).
30015 @item qXfer:siginfo:write
30016 The remote stub understands the @samp{qXfer:siginfo:write} packet
30017 (@pxref{qXfer siginfo write}).
30019 @item qXfer:threads:read
30020 The remote stub understands the @samp{qXfer:threads:read} packet
30021 (@pxref{qXfer threads read}).
30024 The remote stub understands the @samp{QNonStop} packet
30025 (@pxref{QNonStop}).
30028 The remote stub understands the @samp{QPassSignals} packet
30029 (@pxref{QPassSignals}).
30031 @item QStartNoAckMode
30032 The remote stub understands the @samp{QStartNoAckMode} packet and
30033 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
30036 @anchor{multiprocess extensions}
30037 @cindex multiprocess extensions, in remote protocol
30038 The remote stub understands the multiprocess extensions to the remote
30039 protocol syntax. The multiprocess extensions affect the syntax of
30040 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
30041 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
30042 replies. Note that reporting this feature indicates support for the
30043 syntactic extensions only, not that the stub necessarily supports
30044 debugging of more than one process at a time. The stub must not use
30045 multiprocess extensions in packet replies unless @value{GDBN} has also
30046 indicated it supports them in its @samp{qSupported} request.
30048 @item qXfer:osdata:read
30049 The remote stub understands the @samp{qXfer:osdata:read} packet
30050 ((@pxref{qXfer osdata read}).
30052 @item ConditionalTracepoints
30053 The remote stub accepts and implements conditional expressions defined
30054 for tracepoints (@pxref{Tracepoint Conditions}).
30056 @item ReverseContinue
30057 The remote stub accepts and implements the reverse continue packet
30061 The remote stub accepts and implements the reverse step packet
30067 @cindex symbol lookup, remote request
30068 @cindex @samp{qSymbol} packet
30069 Notify the target that @value{GDBN} is prepared to serve symbol lookup
30070 requests. Accept requests from the target for the values of symbols.
30075 The target does not need to look up any (more) symbols.
30076 @item qSymbol:@var{sym_name}
30077 The target requests the value of symbol @var{sym_name} (hex encoded).
30078 @value{GDBN} may provide the value by using the
30079 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
30083 @item qSymbol:@var{sym_value}:@var{sym_name}
30084 Set the value of @var{sym_name} to @var{sym_value}.
30086 @var{sym_name} (hex encoded) is the name of a symbol whose value the
30087 target has previously requested.
30089 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
30090 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
30096 The target does not need to look up any (more) symbols.
30097 @item qSymbol:@var{sym_name}
30098 The target requests the value of a new symbol @var{sym_name} (hex
30099 encoded). @value{GDBN} will continue to supply the values of symbols
30100 (if available), until the target ceases to request them.
30104 @item QTDisconnected
30110 @xref{Tracepoint Packets}.
30112 @item qThreadExtraInfo,@var{thread-id}
30113 @cindex thread attributes info, remote request
30114 @cindex @samp{qThreadExtraInfo} packet
30115 Obtain a printable string description of a thread's attributes from
30116 the target OS. @var{thread-id} is a thread ID;
30117 see @ref{thread-id syntax}. This
30118 string may contain anything that the target OS thinks is interesting
30119 for @value{GDBN} to tell the user about the thread. The string is
30120 displayed in @value{GDBN}'s @code{info threads} display. Some
30121 examples of possible thread extra info strings are @samp{Runnable}, or
30122 @samp{Blocked on Mutex}.
30126 @item @var{XX}@dots{}
30127 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
30128 comprising the printable string containing the extra information about
30129 the thread's attributes.
30132 (Note that the @code{qThreadExtraInfo} packet's name is separated from
30133 the command by a @samp{,}, not a @samp{:}, contrary to the naming
30134 conventions above. Please don't use this packet as a model for new
30146 @xref{Tracepoint Packets}.
30148 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
30149 @cindex read special object, remote request
30150 @cindex @samp{qXfer} packet
30151 @anchor{qXfer read}
30152 Read uninterpreted bytes from the target's special data area
30153 identified by the keyword @var{object}. Request @var{length} bytes
30154 starting at @var{offset} bytes into the data. The content and
30155 encoding of @var{annex} is specific to @var{object}; it can supply
30156 additional details about what data to access.
30158 Here are the specific requests of this form defined so far. All
30159 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
30160 formats, listed below.
30163 @item qXfer:auxv:read::@var{offset},@var{length}
30164 @anchor{qXfer auxiliary vector read}
30165 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
30166 auxiliary vector}. Note @var{annex} must be empty.
30168 This packet is not probed by default; the remote stub must request it,
30169 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
30171 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
30172 @anchor{qXfer target description read}
30173 Access the @dfn{target description}. @xref{Target Descriptions}. The
30174 annex specifies which XML document to access. The main description is
30175 always loaded from the @samp{target.xml} annex.
30177 This packet is not probed by default; the remote stub must request it,
30178 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
30180 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
30181 @anchor{qXfer library list read}
30182 Access the target's list of loaded libraries. @xref{Library List Format}.
30183 The annex part of the generic @samp{qXfer} packet must be empty
30184 (@pxref{qXfer read}).
30186 Targets which maintain a list of libraries in the program's memory do
30187 not need to implement this packet; it is designed for platforms where
30188 the operating system manages the list of loaded libraries.
30190 This packet is not probed by default; the remote stub must request it,
30191 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
30193 @item qXfer:memory-map:read::@var{offset},@var{length}
30194 @anchor{qXfer memory map read}
30195 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
30196 annex part of the generic @samp{qXfer} packet must be empty
30197 (@pxref{qXfer read}).
30199 This packet is not probed by default; the remote stub must request it,
30200 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
30202 @item qXfer:siginfo:read::@var{offset},@var{length}
30203 @anchor{qXfer siginfo read}
30204 Read contents of the extra signal information on the target
30205 system. The annex part of the generic @samp{qXfer} packet must be
30206 empty (@pxref{qXfer read}).
30208 This packet is not probed by default; the remote stub must request it,
30209 by supplying an appropriate @samp{qSupported} response
30210 (@pxref{qSupported}).
30212 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
30213 @anchor{qXfer spu read}
30214 Read contents of an @code{spufs} file on the target system. The
30215 annex specifies which file to read; it must be of the form
30216 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
30217 in the target process, and @var{name} identifes the @code{spufs} file
30218 in that context to be accessed.
30220 This packet is not probed by default; the remote stub must request it,
30221 by supplying an appropriate @samp{qSupported} response
30222 (@pxref{qSupported}).
30224 @item qXfer:threads:read::@var{offset},@var{length}
30225 @anchor{qXfer threads read}
30226 Access the list of threads on target. @xref{Thread List Format}. The
30227 annex part of the generic @samp{qXfer} packet must be empty
30228 (@pxref{qXfer read}).
30230 This packet is not probed by default; the remote stub must request it,
30231 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
30233 @item qXfer:osdata:read::@var{offset},@var{length}
30234 @anchor{qXfer osdata read}
30235 Access the target's @dfn{operating system information}.
30236 @xref{Operating System Information}.
30243 Data @var{data} (@pxref{Binary Data}) has been read from the
30244 target. There may be more data at a higher address (although
30245 it is permitted to return @samp{m} even for the last valid
30246 block of data, as long as at least one byte of data was read).
30247 @var{data} may have fewer bytes than the @var{length} in the
30251 Data @var{data} (@pxref{Binary Data}) has been read from the target.
30252 There is no more data to be read. @var{data} may have fewer bytes
30253 than the @var{length} in the request.
30256 The @var{offset} in the request is at the end of the data.
30257 There is no more data to be read.
30260 The request was malformed, or @var{annex} was invalid.
30263 The offset was invalid, or there was an error encountered reading the data.
30264 @var{nn} is a hex-encoded @code{errno} value.
30267 An empty reply indicates the @var{object} string was not recognized by
30268 the stub, or that the object does not support reading.
30271 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
30272 @cindex write data into object, remote request
30273 @anchor{qXfer write}
30274 Write uninterpreted bytes into the target's special data area
30275 identified by the keyword @var{object}, starting at @var{offset} bytes
30276 into the data. @var{data}@dots{} is the binary-encoded data
30277 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
30278 is specific to @var{object}; it can supply additional details about what data
30281 Here are the specific requests of this form defined so far. All
30282 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
30283 formats, listed below.
30286 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
30287 @anchor{qXfer siginfo write}
30288 Write @var{data} to the extra signal information on the target system.
30289 The annex part of the generic @samp{qXfer} packet must be
30290 empty (@pxref{qXfer write}).
30292 This packet is not probed by default; the remote stub must request it,
30293 by supplying an appropriate @samp{qSupported} response
30294 (@pxref{qSupported}).
30296 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
30297 @anchor{qXfer spu write}
30298 Write @var{data} to an @code{spufs} file on the target system. The
30299 annex specifies which file to write; it must be of the form
30300 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
30301 in the target process, and @var{name} identifes the @code{spufs} file
30302 in that context to be accessed.
30304 This packet is not probed by default; the remote stub must request it,
30305 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
30311 @var{nn} (hex encoded) is the number of bytes written.
30312 This may be fewer bytes than supplied in the request.
30315 The request was malformed, or @var{annex} was invalid.
30318 The offset was invalid, or there was an error encountered writing the data.
30319 @var{nn} is a hex-encoded @code{errno} value.
30322 An empty reply indicates the @var{object} string was not
30323 recognized by the stub, or that the object does not support writing.
30326 @item qXfer:@var{object}:@var{operation}:@dots{}
30327 Requests of this form may be added in the future. When a stub does
30328 not recognize the @var{object} keyword, or its support for
30329 @var{object} does not recognize the @var{operation} keyword, the stub
30330 must respond with an empty packet.
30332 @item qAttached:@var{pid}
30333 @cindex query attached, remote request
30334 @cindex @samp{qAttached} packet
30335 Return an indication of whether the remote server attached to an
30336 existing process or created a new process. When the multiprocess
30337 protocol extensions are supported (@pxref{multiprocess extensions}),
30338 @var{pid} is an integer in hexadecimal format identifying the target
30339 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
30340 the query packet will be simplified as @samp{qAttached}.
30342 This query is used, for example, to know whether the remote process
30343 should be detached or killed when a @value{GDBN} session is ended with
30344 the @code{quit} command.
30349 The remote server attached to an existing process.
30351 The remote server created a new process.
30353 A badly formed request or an error was encountered.
30358 @node Architecture-Specific Protocol Details
30359 @section Architecture-Specific Protocol Details
30361 This section describes how the remote protocol is applied to specific
30362 target architectures. Also see @ref{Standard Target Features}, for
30363 details of XML target descriptions for each architecture.
30367 @subsubsection Breakpoint Kinds
30369 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
30374 16-bit Thumb mode breakpoint.
30377 32-bit Thumb mode (Thumb-2) breakpoint.
30380 32-bit ARM mode breakpoint.
30386 @subsubsection Register Packet Format
30388 The following @code{g}/@code{G} packets have previously been defined.
30389 In the below, some thirty-two bit registers are transferred as
30390 sixty-four bits. Those registers should be zero/sign extended (which?)
30391 to fill the space allocated. Register bytes are transferred in target
30392 byte order. The two nibbles within a register byte are transferred
30393 most-significant - least-significant.
30399 All registers are transferred as thirty-two bit quantities in the order:
30400 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
30401 registers; fsr; fir; fp.
30405 All registers are transferred as sixty-four bit quantities (including
30406 thirty-two bit registers such as @code{sr}). The ordering is the same
30411 @node Tracepoint Packets
30412 @section Tracepoint Packets
30413 @cindex tracepoint packets
30414 @cindex packets, tracepoint
30416 Here we describe the packets @value{GDBN} uses to implement
30417 tracepoints (@pxref{Tracepoints}).
30421 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
30422 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
30423 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
30424 the tracepoint is disabled. @var{step} is the tracepoint's step
30425 count, and @var{pass} is its pass count. If an @samp{F} is present,
30426 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
30427 the number of bytes that the target should copy elsewhere to make room
30428 for the tracepoint. If an @samp{X} is present, it introduces a
30429 tracepoint condition, which consists of a hexadecimal length, followed
30430 by a comma and hex-encoded bytes, in a manner similar to action
30431 encodings as described below. If the trailing @samp{-} is present,
30432 further @samp{QTDP} packets will follow to specify this tracepoint's
30438 The packet was understood and carried out.
30440 The packet was not recognized.
30443 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
30444 Define actions to be taken when a tracepoint is hit. @var{n} and
30445 @var{addr} must be the same as in the initial @samp{QTDP} packet for
30446 this tracepoint. This packet may only be sent immediately after
30447 another @samp{QTDP} packet that ended with a @samp{-}. If the
30448 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
30449 specifying more actions for this tracepoint.
30451 In the series of action packets for a given tracepoint, at most one
30452 can have an @samp{S} before its first @var{action}. If such a packet
30453 is sent, it and the following packets define ``while-stepping''
30454 actions. Any prior packets define ordinary actions --- that is, those
30455 taken when the tracepoint is first hit. If no action packet has an
30456 @samp{S}, then all the packets in the series specify ordinary
30457 tracepoint actions.
30459 The @samp{@var{action}@dots{}} portion of the packet is a series of
30460 actions, concatenated without separators. Each action has one of the
30466 Collect the registers whose bits are set in @var{mask}. @var{mask} is
30467 a hexadecimal number whose @var{i}'th bit is set if register number
30468 @var{i} should be collected. (The least significant bit is numbered
30469 zero.) Note that @var{mask} may be any number of digits long; it may
30470 not fit in a 32-bit word.
30472 @item M @var{basereg},@var{offset},@var{len}
30473 Collect @var{len} bytes of memory starting at the address in register
30474 number @var{basereg}, plus @var{offset}. If @var{basereg} is
30475 @samp{-1}, then the range has a fixed address: @var{offset} is the
30476 address of the lowest byte to collect. The @var{basereg},
30477 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
30478 values (the @samp{-1} value for @var{basereg} is a special case).
30480 @item X @var{len},@var{expr}
30481 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
30482 it directs. @var{expr} is an agent expression, as described in
30483 @ref{Agent Expressions}. Each byte of the expression is encoded as a
30484 two-digit hex number in the packet; @var{len} is the number of bytes
30485 in the expression (and thus one-half the number of hex digits in the
30490 Any number of actions may be packed together in a single @samp{QTDP}
30491 packet, as long as the packet does not exceed the maximum packet
30492 length (400 bytes, for many stubs). There may be only one @samp{R}
30493 action per tracepoint, and it must precede any @samp{M} or @samp{X}
30494 actions. Any registers referred to by @samp{M} and @samp{X} actions
30495 must be collected by a preceding @samp{R} action. (The
30496 ``while-stepping'' actions are treated as if they were attached to a
30497 separate tracepoint, as far as these restrictions are concerned.)
30502 The packet was understood and carried out.
30504 The packet was not recognized.
30507 @item QTDV:@var{n}:@var{value}
30508 @cindex define trace state variable, remote request
30509 @cindex @samp{QTDV} packet
30510 Create a new trace state variable, number @var{n}, with an initial
30511 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
30512 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
30513 the option of not using this packet for initial values of zero; the
30514 target should simply create the trace state variables as they are
30515 mentioned in expressions.
30517 @item QTFrame:@var{n}
30518 Select the @var{n}'th tracepoint frame from the buffer, and use the
30519 register and memory contents recorded there to answer subsequent
30520 request packets from @value{GDBN}.
30522 A successful reply from the stub indicates that the stub has found the
30523 requested frame. The response is a series of parts, concatenated
30524 without separators, describing the frame we selected. Each part has
30525 one of the following forms:
30529 The selected frame is number @var{n} in the trace frame buffer;
30530 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
30531 was no frame matching the criteria in the request packet.
30534 The selected trace frame records a hit of tracepoint number @var{t};
30535 @var{t} is a hexadecimal number.
30539 @item QTFrame:pc:@var{addr}
30540 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
30541 currently selected frame whose PC is @var{addr};
30542 @var{addr} is a hexadecimal number.
30544 @item QTFrame:tdp:@var{t}
30545 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
30546 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
30547 is a hexadecimal number.
30549 @item QTFrame:range:@var{start}:@var{end}
30550 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
30551 currently selected frame whose PC is between @var{start} (inclusive)
30552 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
30555 @item QTFrame:outside:@var{start}:@var{end}
30556 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
30557 frame @emph{outside} the given range of addresses (exclusive).
30560 Begin the tracepoint experiment. Begin collecting data from tracepoint
30561 hits in the trace frame buffer.
30564 End the tracepoint experiment. Stop collecting trace frames.
30567 Clear the table of tracepoints, and empty the trace frame buffer.
30569 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
30570 Establish the given ranges of memory as ``transparent''. The stub
30571 will answer requests for these ranges from memory's current contents,
30572 if they were not collected as part of the tracepoint hit.
30574 @value{GDBN} uses this to mark read-only regions of memory, like those
30575 containing program code. Since these areas never change, they should
30576 still have the same contents they did when the tracepoint was hit, so
30577 there's no reason for the stub to refuse to provide their contents.
30579 @item QTDisconnected:@var{value}
30580 Set the choice to what to do with the tracing run when @value{GDBN}
30581 disconnects from the target. A @var{value} of 1 directs the target to
30582 continue the tracing run, while 0 tells the target to stop tracing if
30583 @value{GDBN} is no longer in the picture.
30586 Ask the stub if there is a trace experiment running right now.
30591 There is no trace experiment running.
30593 There is a trace experiment running.
30596 @item qTV:@var{var}
30597 @cindex trace state variable value, remote request
30598 @cindex @samp{qTV} packet
30599 Ask the stub for the value of the trace state variable number @var{var}.
30604 The value of the variable is @var{value}. This will be the current
30605 value of the variable if the user is examining a running target, or a
30606 saved value if the variable was collected in the trace frame that the
30607 user is looking at. Note that multiple requests may result in
30608 different reply values, such as when requesting values while the
30609 program is running.
30612 The value of the variable is unknown. This would occur, for example,
30613 if the user is examining a trace frame in which the requested variable
30619 These packets request data about tracepoints that are being used by
30620 the target. @value{GDBN} sends @code{qTfP} to get the first piece
30621 of data, and multiple @code{qTsP} to get additional pieces. Replies
30622 to these packets generally take the form of the @code{QTDP} packets
30623 that define tracepoints. (FIXME add detailed syntax)
30627 These packets request data about trace state variables that are on the
30628 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
30629 and multiple @code{qTsV} to get additional variables. Replies to
30630 these packets follow the syntax of the @code{QTDV} packets that define
30631 trace state variables.
30633 @item QTSave:@var{filename}
30634 This packet directs the target to save trace data to the file name
30635 @var{filename} in the target's filesystem. @var{filename} is encoded
30636 as a hex string; the interpretation of the file name (relative vs
30637 absolute, wild cards, etc) is up to the target.
30639 @item qTBuffer:@var{offset},@var{len}
30640 Return up to @var{len} bytes of the current contents of trace buffer,
30641 starting at @var{offset}. The trace buffer is treated as if it were
30642 a contiguous collection of traceframes, as per the trace file format.
30643 The reply consists as many hex-encoded bytes as the target can deliver
30644 in a packet; it is not an error to return fewer than were asked for.
30645 A reply consisting of just @code{l} indicates that no bytes are
30650 @node Host I/O Packets
30651 @section Host I/O Packets
30652 @cindex Host I/O, remote protocol
30653 @cindex file transfer, remote protocol
30655 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
30656 operations on the far side of a remote link. For example, Host I/O is
30657 used to upload and download files to a remote target with its own
30658 filesystem. Host I/O uses the same constant values and data structure
30659 layout as the target-initiated File-I/O protocol. However, the
30660 Host I/O packets are structured differently. The target-initiated
30661 protocol relies on target memory to store parameters and buffers.
30662 Host I/O requests are initiated by @value{GDBN}, and the
30663 target's memory is not involved. @xref{File-I/O Remote Protocol
30664 Extension}, for more details on the target-initiated protocol.
30666 The Host I/O request packets all encode a single operation along with
30667 its arguments. They have this format:
30671 @item vFile:@var{operation}: @var{parameter}@dots{}
30672 @var{operation} is the name of the particular request; the target
30673 should compare the entire packet name up to the second colon when checking
30674 for a supported operation. The format of @var{parameter} depends on
30675 the operation. Numbers are always passed in hexadecimal. Negative
30676 numbers have an explicit minus sign (i.e.@: two's complement is not
30677 used). Strings (e.g.@: filenames) are encoded as a series of
30678 hexadecimal bytes. The last argument to a system call may be a
30679 buffer of escaped binary data (@pxref{Binary Data}).
30683 The valid responses to Host I/O packets are:
30687 @item F @var{result} [, @var{errno}] [; @var{attachment}]
30688 @var{result} is the integer value returned by this operation, usually
30689 non-negative for success and -1 for errors. If an error has occured,
30690 @var{errno} will be included in the result. @var{errno} will have a
30691 value defined by the File-I/O protocol (@pxref{Errno Values}). For
30692 operations which return data, @var{attachment} supplies the data as a
30693 binary buffer. Binary buffers in response packets are escaped in the
30694 normal way (@pxref{Binary Data}). See the individual packet
30695 documentation for the interpretation of @var{result} and
30699 An empty response indicates that this operation is not recognized.
30703 These are the supported Host I/O operations:
30706 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
30707 Open a file at @var{pathname} and return a file descriptor for it, or
30708 return -1 if an error occurs. @var{pathname} is a string,
30709 @var{flags} is an integer indicating a mask of open flags
30710 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
30711 of mode bits to use if the file is created (@pxref{mode_t Values}).
30712 @xref{open}, for details of the open flags and mode values.
30714 @item vFile:close: @var{fd}
30715 Close the open file corresponding to @var{fd} and return 0, or
30716 -1 if an error occurs.
30718 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
30719 Read data from the open file corresponding to @var{fd}. Up to
30720 @var{count} bytes will be read from the file, starting at @var{offset}
30721 relative to the start of the file. The target may read fewer bytes;
30722 common reasons include packet size limits and an end-of-file
30723 condition. The number of bytes read is returned. Zero should only be
30724 returned for a successful read at the end of the file, or if
30725 @var{count} was zero.
30727 The data read should be returned as a binary attachment on success.
30728 If zero bytes were read, the response should include an empty binary
30729 attachment (i.e.@: a trailing semicolon). The return value is the
30730 number of target bytes read; the binary attachment may be longer if
30731 some characters were escaped.
30733 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
30734 Write @var{data} (a binary buffer) to the open file corresponding
30735 to @var{fd}. Start the write at @var{offset} from the start of the
30736 file. Unlike many @code{write} system calls, there is no
30737 separate @var{count} argument; the length of @var{data} in the
30738 packet is used. @samp{vFile:write} returns the number of bytes written,
30739 which may be shorter than the length of @var{data}, or -1 if an
30742 @item vFile:unlink: @var{pathname}
30743 Delete the file at @var{pathname} on the target. Return 0,
30744 or -1 if an error occurs. @var{pathname} is a string.
30749 @section Interrupts
30750 @cindex interrupts (remote protocol)
30752 When a program on the remote target is running, @value{GDBN} may
30753 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
30754 a @code{BREAK} followed by @code{g},
30755 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
30757 The precise meaning of @code{BREAK} is defined by the transport
30758 mechanism and may, in fact, be undefined. @value{GDBN} does not
30759 currently define a @code{BREAK} mechanism for any of the network
30760 interfaces except for TCP, in which case @value{GDBN} sends the
30761 @code{telnet} BREAK sequence.
30763 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
30764 transport mechanisms. It is represented by sending the single byte
30765 @code{0x03} without any of the usual packet overhead described in
30766 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
30767 transmitted as part of a packet, it is considered to be packet data
30768 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
30769 (@pxref{X packet}), used for binary downloads, may include an unescaped
30770 @code{0x03} as part of its packet.
30772 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
30773 When Linux kernel receives this sequence from serial port,
30774 it stops execution and connects to gdb.
30776 Stubs are not required to recognize these interrupt mechanisms and the
30777 precise meaning associated with receipt of the interrupt is
30778 implementation defined. If the target supports debugging of multiple
30779 threads and/or processes, it should attempt to interrupt all
30780 currently-executing threads and processes.
30781 If the stub is successful at interrupting the
30782 running program, it should send one of the stop
30783 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
30784 of successfully stopping the program in all-stop mode, and a stop reply
30785 for each stopped thread in non-stop mode.
30786 Interrupts received while the
30787 program is stopped are discarded.
30789 @node Notification Packets
30790 @section Notification Packets
30791 @cindex notification packets
30792 @cindex packets, notification
30794 The @value{GDBN} remote serial protocol includes @dfn{notifications},
30795 packets that require no acknowledgment. Both the GDB and the stub
30796 may send notifications (although the only notifications defined at
30797 present are sent by the stub). Notifications carry information
30798 without incurring the round-trip latency of an acknowledgment, and so
30799 are useful for low-impact communications where occasional packet loss
30802 A notification packet has the form @samp{% @var{data} #
30803 @var{checksum}}, where @var{data} is the content of the notification,
30804 and @var{checksum} is a checksum of @var{data}, computed and formatted
30805 as for ordinary @value{GDBN} packets. A notification's @var{data}
30806 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
30807 receiving a notification, the recipient sends no @samp{+} or @samp{-}
30808 to acknowledge the notification's receipt or to report its corruption.
30810 Every notification's @var{data} begins with a name, which contains no
30811 colon characters, followed by a colon character.
30813 Recipients should silently ignore corrupted notifications and
30814 notifications they do not understand. Recipients should restart
30815 timeout periods on receipt of a well-formed notification, whether or
30816 not they understand it.
30818 Senders should only send the notifications described here when this
30819 protocol description specifies that they are permitted. In the
30820 future, we may extend the protocol to permit existing notifications in
30821 new contexts; this rule helps older senders avoid confusing newer
30824 (Older versions of @value{GDBN} ignore bytes received until they see
30825 the @samp{$} byte that begins an ordinary packet, so new stubs may
30826 transmit notifications without fear of confusing older clients. There
30827 are no notifications defined for @value{GDBN} to send at the moment, but we
30828 assume that most older stubs would ignore them, as well.)
30830 The following notification packets from the stub to @value{GDBN} are
30834 @item Stop: @var{reply}
30835 Report an asynchronous stop event in non-stop mode.
30836 The @var{reply} has the form of a stop reply, as
30837 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
30838 for information on how these notifications are acknowledged by
30842 @node Remote Non-Stop
30843 @section Remote Protocol Support for Non-Stop Mode
30845 @value{GDBN}'s remote protocol supports non-stop debugging of
30846 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
30847 supports non-stop mode, it should report that to @value{GDBN} by including
30848 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
30850 @value{GDBN} typically sends a @samp{QNonStop} packet only when
30851 establishing a new connection with the stub. Entering non-stop mode
30852 does not alter the state of any currently-running threads, but targets
30853 must stop all threads in any already-attached processes when entering
30854 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
30855 probe the target state after a mode change.
30857 In non-stop mode, when an attached process encounters an event that
30858 would otherwise be reported with a stop reply, it uses the
30859 asynchronous notification mechanism (@pxref{Notification Packets}) to
30860 inform @value{GDBN}. In contrast to all-stop mode, where all threads
30861 in all processes are stopped when a stop reply is sent, in non-stop
30862 mode only the thread reporting the stop event is stopped. That is,
30863 when reporting a @samp{S} or @samp{T} response to indicate completion
30864 of a step operation, hitting a breakpoint, or a fault, only the
30865 affected thread is stopped; any other still-running threads continue
30866 to run. When reporting a @samp{W} or @samp{X} response, all running
30867 threads belonging to other attached processes continue to run.
30869 Only one stop reply notification at a time may be pending; if
30870 additional stop events occur before @value{GDBN} has acknowledged the
30871 previous notification, they must be queued by the stub for later
30872 synchronous transmission in response to @samp{vStopped} packets from
30873 @value{GDBN}. Because the notification mechanism is unreliable,
30874 the stub is permitted to resend a stop reply notification
30875 if it believes @value{GDBN} may not have received it. @value{GDBN}
30876 ignores additional stop reply notifications received before it has
30877 finished processing a previous notification and the stub has completed
30878 sending any queued stop events.
30880 Otherwise, @value{GDBN} must be prepared to receive a stop reply
30881 notification at any time. Specifically, they may appear when
30882 @value{GDBN} is not otherwise reading input from the stub, or when
30883 @value{GDBN} is expecting to read a normal synchronous response or a
30884 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
30885 Notification packets are distinct from any other communication from
30886 the stub so there is no ambiguity.
30888 After receiving a stop reply notification, @value{GDBN} shall
30889 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
30890 as a regular, synchronous request to the stub. Such acknowledgment
30891 is not required to happen immediately, as @value{GDBN} is permitted to
30892 send other, unrelated packets to the stub first, which the stub should
30895 Upon receiving a @samp{vStopped} packet, if the stub has other queued
30896 stop events to report to @value{GDBN}, it shall respond by sending a
30897 normal stop reply response. @value{GDBN} shall then send another
30898 @samp{vStopped} packet to solicit further responses; again, it is
30899 permitted to send other, unrelated packets as well which the stub
30900 should process normally.
30902 If the stub receives a @samp{vStopped} packet and there are no
30903 additional stop events to report, the stub shall return an @samp{OK}
30904 response. At this point, if further stop events occur, the stub shall
30905 send a new stop reply notification, @value{GDBN} shall accept the
30906 notification, and the process shall be repeated.
30908 In non-stop mode, the target shall respond to the @samp{?} packet as
30909 follows. First, any incomplete stop reply notification/@samp{vStopped}
30910 sequence in progress is abandoned. The target must begin a new
30911 sequence reporting stop events for all stopped threads, whether or not
30912 it has previously reported those events to @value{GDBN}. The first
30913 stop reply is sent as a synchronous reply to the @samp{?} packet, and
30914 subsequent stop replies are sent as responses to @samp{vStopped} packets
30915 using the mechanism described above. The target must not send
30916 asynchronous stop reply notifications until the sequence is complete.
30917 If all threads are running when the target receives the @samp{?} packet,
30918 or if the target is not attached to any process, it shall respond
30921 @node Packet Acknowledgment
30922 @section Packet Acknowledgment
30924 @cindex acknowledgment, for @value{GDBN} remote
30925 @cindex packet acknowledgment, for @value{GDBN} remote
30926 By default, when either the host or the target machine receives a packet,
30927 the first response expected is an acknowledgment: either @samp{+} (to indicate
30928 the package was received correctly) or @samp{-} (to request retransmission).
30929 This mechanism allows the @value{GDBN} remote protocol to operate over
30930 unreliable transport mechanisms, such as a serial line.
30932 In cases where the transport mechanism is itself reliable (such as a pipe or
30933 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
30934 It may be desirable to disable them in that case to reduce communication
30935 overhead, or for other reasons. This can be accomplished by means of the
30936 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
30938 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
30939 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
30940 and response format still includes the normal checksum, as described in
30941 @ref{Overview}, but the checksum may be ignored by the receiver.
30943 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
30944 no-acknowledgment mode, it should report that to @value{GDBN}
30945 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
30946 @pxref{qSupported}.
30947 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
30948 disabled via the @code{set remote noack-packet off} command
30949 (@pxref{Remote Configuration}),
30950 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
30951 Only then may the stub actually turn off packet acknowledgments.
30952 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
30953 response, which can be safely ignored by the stub.
30955 Note that @code{set remote noack-packet} command only affects negotiation
30956 between @value{GDBN} and the stub when subsequent connections are made;
30957 it does not affect the protocol acknowledgment state for any current
30959 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
30960 new connection is established,
30961 there is also no protocol request to re-enable the acknowledgments
30962 for the current connection, once disabled.
30967 Example sequence of a target being re-started. Notice how the restart
30968 does not get any direct output:
30973 @emph{target restarts}
30976 <- @code{T001:1234123412341234}
30980 Example sequence of a target being stepped by a single instruction:
30983 -> @code{G1445@dots{}}
30988 <- @code{T001:1234123412341234}
30992 <- @code{1455@dots{}}
30996 @node File-I/O Remote Protocol Extension
30997 @section File-I/O Remote Protocol Extension
30998 @cindex File-I/O remote protocol extension
31001 * File-I/O Overview::
31002 * Protocol Basics::
31003 * The F Request Packet::
31004 * The F Reply Packet::
31005 * The Ctrl-C Message::
31007 * List of Supported Calls::
31008 * Protocol-specific Representation of Datatypes::
31010 * File-I/O Examples::
31013 @node File-I/O Overview
31014 @subsection File-I/O Overview
31015 @cindex file-i/o overview
31017 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
31018 target to use the host's file system and console I/O to perform various
31019 system calls. System calls on the target system are translated into a
31020 remote protocol packet to the host system, which then performs the needed
31021 actions and returns a response packet to the target system.
31022 This simulates file system operations even on targets that lack file systems.
31024 The protocol is defined to be independent of both the host and target systems.
31025 It uses its own internal representation of datatypes and values. Both
31026 @value{GDBN} and the target's @value{GDBN} stub are responsible for
31027 translating the system-dependent value representations into the internal
31028 protocol representations when data is transmitted.
31030 The communication is synchronous. A system call is possible only when
31031 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
31032 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
31033 the target is stopped to allow deterministic access to the target's
31034 memory. Therefore File-I/O is not interruptible by target signals. On
31035 the other hand, it is possible to interrupt File-I/O by a user interrupt
31036 (@samp{Ctrl-C}) within @value{GDBN}.
31038 The target's request to perform a host system call does not finish
31039 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
31040 after finishing the system call, the target returns to continuing the
31041 previous activity (continue, step). No additional continue or step
31042 request from @value{GDBN} is required.
31045 (@value{GDBP}) continue
31046 <- target requests 'system call X'
31047 target is stopped, @value{GDBN} executes system call
31048 -> @value{GDBN} returns result
31049 ... target continues, @value{GDBN} returns to wait for the target
31050 <- target hits breakpoint and sends a Txx packet
31053 The protocol only supports I/O on the console and to regular files on
31054 the host file system. Character or block special devices, pipes,
31055 named pipes, sockets or any other communication method on the host
31056 system are not supported by this protocol.
31058 File I/O is not supported in non-stop mode.
31060 @node Protocol Basics
31061 @subsection Protocol Basics
31062 @cindex protocol basics, file-i/o
31064 The File-I/O protocol uses the @code{F} packet as the request as well
31065 as reply packet. Since a File-I/O system call can only occur when
31066 @value{GDBN} is waiting for a response from the continuing or stepping target,
31067 the File-I/O request is a reply that @value{GDBN} has to expect as a result
31068 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
31069 This @code{F} packet contains all information needed to allow @value{GDBN}
31070 to call the appropriate host system call:
31074 A unique identifier for the requested system call.
31077 All parameters to the system call. Pointers are given as addresses
31078 in the target memory address space. Pointers to strings are given as
31079 pointer/length pair. Numerical values are given as they are.
31080 Numerical control flags are given in a protocol-specific representation.
31084 At this point, @value{GDBN} has to perform the following actions.
31088 If the parameters include pointer values to data needed as input to a
31089 system call, @value{GDBN} requests this data from the target with a
31090 standard @code{m} packet request. This additional communication has to be
31091 expected by the target implementation and is handled as any other @code{m}
31095 @value{GDBN} translates all value from protocol representation to host
31096 representation as needed. Datatypes are coerced into the host types.
31099 @value{GDBN} calls the system call.
31102 It then coerces datatypes back to protocol representation.
31105 If the system call is expected to return data in buffer space specified
31106 by pointer parameters to the call, the data is transmitted to the
31107 target using a @code{M} or @code{X} packet. This packet has to be expected
31108 by the target implementation and is handled as any other @code{M} or @code{X}
31113 Eventually @value{GDBN} replies with another @code{F} packet which contains all
31114 necessary information for the target to continue. This at least contains
31121 @code{errno}, if has been changed by the system call.
31128 After having done the needed type and value coercion, the target continues
31129 the latest continue or step action.
31131 @node The F Request Packet
31132 @subsection The @code{F} Request Packet
31133 @cindex file-i/o request packet
31134 @cindex @code{F} request packet
31136 The @code{F} request packet has the following format:
31139 @item F@var{call-id},@var{parameter@dots{}}
31141 @var{call-id} is the identifier to indicate the host system call to be called.
31142 This is just the name of the function.
31144 @var{parameter@dots{}} are the parameters to the system call.
31145 Parameters are hexadecimal integer values, either the actual values in case
31146 of scalar datatypes, pointers to target buffer space in case of compound
31147 datatypes and unspecified memory areas, or pointer/length pairs in case
31148 of string parameters. These are appended to the @var{call-id} as a
31149 comma-delimited list. All values are transmitted in ASCII
31150 string representation, pointer/length pairs separated by a slash.
31156 @node The F Reply Packet
31157 @subsection The @code{F} Reply Packet
31158 @cindex file-i/o reply packet
31159 @cindex @code{F} reply packet
31161 The @code{F} reply packet has the following format:
31165 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
31167 @var{retcode} is the return code of the system call as hexadecimal value.
31169 @var{errno} is the @code{errno} set by the call, in protocol-specific
31171 This parameter can be omitted if the call was successful.
31173 @var{Ctrl-C flag} is only sent if the user requested a break. In this
31174 case, @var{errno} must be sent as well, even if the call was successful.
31175 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
31182 or, if the call was interrupted before the host call has been performed:
31189 assuming 4 is the protocol-specific representation of @code{EINTR}.
31194 @node The Ctrl-C Message
31195 @subsection The @samp{Ctrl-C} Message
31196 @cindex ctrl-c message, in file-i/o protocol
31198 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
31199 reply packet (@pxref{The F Reply Packet}),
31200 the target should behave as if it had
31201 gotten a break message. The meaning for the target is ``system call
31202 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
31203 (as with a break message) and return to @value{GDBN} with a @code{T02}
31206 It's important for the target to know in which
31207 state the system call was interrupted. There are two possible cases:
31211 The system call hasn't been performed on the host yet.
31214 The system call on the host has been finished.
31218 These two states can be distinguished by the target by the value of the
31219 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
31220 call hasn't been performed. This is equivalent to the @code{EINTR} handling
31221 on POSIX systems. In any other case, the target may presume that the
31222 system call has been finished --- successfully or not --- and should behave
31223 as if the break message arrived right after the system call.
31225 @value{GDBN} must behave reliably. If the system call has not been called
31226 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
31227 @code{errno} in the packet. If the system call on the host has been finished
31228 before the user requests a break, the full action must be finished by
31229 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
31230 The @code{F} packet may only be sent when either nothing has happened
31231 or the full action has been completed.
31234 @subsection Console I/O
31235 @cindex console i/o as part of file-i/o
31237 By default and if not explicitly closed by the target system, the file
31238 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
31239 on the @value{GDBN} console is handled as any other file output operation
31240 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
31241 by @value{GDBN} so that after the target read request from file descriptor
31242 0 all following typing is buffered until either one of the following
31247 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
31249 system call is treated as finished.
31252 The user presses @key{RET}. This is treated as end of input with a trailing
31256 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
31257 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
31261 If the user has typed more characters than fit in the buffer given to
31262 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
31263 either another @code{read(0, @dots{})} is requested by the target, or debugging
31264 is stopped at the user's request.
31267 @node List of Supported Calls
31268 @subsection List of Supported Calls
31269 @cindex list of supported file-i/o calls
31286 @unnumberedsubsubsec open
31287 @cindex open, file-i/o system call
31292 int open(const char *pathname, int flags);
31293 int open(const char *pathname, int flags, mode_t mode);
31297 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
31300 @var{flags} is the bitwise @code{OR} of the following values:
31304 If the file does not exist it will be created. The host
31305 rules apply as far as file ownership and time stamps
31309 When used with @code{O_CREAT}, if the file already exists it is
31310 an error and open() fails.
31313 If the file already exists and the open mode allows
31314 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
31315 truncated to zero length.
31318 The file is opened in append mode.
31321 The file is opened for reading only.
31324 The file is opened for writing only.
31327 The file is opened for reading and writing.
31331 Other bits are silently ignored.
31335 @var{mode} is the bitwise @code{OR} of the following values:
31339 User has read permission.
31342 User has write permission.
31345 Group has read permission.
31348 Group has write permission.
31351 Others have read permission.
31354 Others have write permission.
31358 Other bits are silently ignored.
31361 @item Return value:
31362 @code{open} returns the new file descriptor or -1 if an error
31369 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
31372 @var{pathname} refers to a directory.
31375 The requested access is not allowed.
31378 @var{pathname} was too long.
31381 A directory component in @var{pathname} does not exist.
31384 @var{pathname} refers to a device, pipe, named pipe or socket.
31387 @var{pathname} refers to a file on a read-only filesystem and
31388 write access was requested.
31391 @var{pathname} is an invalid pointer value.
31394 No space on device to create the file.
31397 The process already has the maximum number of files open.
31400 The limit on the total number of files open on the system
31404 The call was interrupted by the user.
31410 @unnumberedsubsubsec close
31411 @cindex close, file-i/o system call
31420 @samp{Fclose,@var{fd}}
31422 @item Return value:
31423 @code{close} returns zero on success, or -1 if an error occurred.
31429 @var{fd} isn't a valid open file descriptor.
31432 The call was interrupted by the user.
31438 @unnumberedsubsubsec read
31439 @cindex read, file-i/o system call
31444 int read(int fd, void *buf, unsigned int count);
31448 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
31450 @item Return value:
31451 On success, the number of bytes read is returned.
31452 Zero indicates end of file. If count is zero, read
31453 returns zero as well. On error, -1 is returned.
31459 @var{fd} is not a valid file descriptor or is not open for
31463 @var{bufptr} is an invalid pointer value.
31466 The call was interrupted by the user.
31472 @unnumberedsubsubsec write
31473 @cindex write, file-i/o system call
31478 int write(int fd, const void *buf, unsigned int count);
31482 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
31484 @item Return value:
31485 On success, the number of bytes written are returned.
31486 Zero indicates nothing was written. On error, -1
31493 @var{fd} is not a valid file descriptor or is not open for
31497 @var{bufptr} is an invalid pointer value.
31500 An attempt was made to write a file that exceeds the
31501 host-specific maximum file size allowed.
31504 No space on device to write the data.
31507 The call was interrupted by the user.
31513 @unnumberedsubsubsec lseek
31514 @cindex lseek, file-i/o system call
31519 long lseek (int fd, long offset, int flag);
31523 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
31525 @var{flag} is one of:
31529 The offset is set to @var{offset} bytes.
31532 The offset is set to its current location plus @var{offset}
31536 The offset is set to the size of the file plus @var{offset}
31540 @item Return value:
31541 On success, the resulting unsigned offset in bytes from
31542 the beginning of the file is returned. Otherwise, a
31543 value of -1 is returned.
31549 @var{fd} is not a valid open file descriptor.
31552 @var{fd} is associated with the @value{GDBN} console.
31555 @var{flag} is not a proper value.
31558 The call was interrupted by the user.
31564 @unnumberedsubsubsec rename
31565 @cindex rename, file-i/o system call
31570 int rename(const char *oldpath, const char *newpath);
31574 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
31576 @item Return value:
31577 On success, zero is returned. On error, -1 is returned.
31583 @var{newpath} is an existing directory, but @var{oldpath} is not a
31587 @var{newpath} is a non-empty directory.
31590 @var{oldpath} or @var{newpath} is a directory that is in use by some
31594 An attempt was made to make a directory a subdirectory
31598 A component used as a directory in @var{oldpath} or new
31599 path is not a directory. Or @var{oldpath} is a directory
31600 and @var{newpath} exists but is not a directory.
31603 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
31606 No access to the file or the path of the file.
31610 @var{oldpath} or @var{newpath} was too long.
31613 A directory component in @var{oldpath} or @var{newpath} does not exist.
31616 The file is on a read-only filesystem.
31619 The device containing the file has no room for the new
31623 The call was interrupted by the user.
31629 @unnumberedsubsubsec unlink
31630 @cindex unlink, file-i/o system call
31635 int unlink(const char *pathname);
31639 @samp{Funlink,@var{pathnameptr}/@var{len}}
31641 @item Return value:
31642 On success, zero is returned. On error, -1 is returned.
31648 No access to the file or the path of the file.
31651 The system does not allow unlinking of directories.
31654 The file @var{pathname} cannot be unlinked because it's
31655 being used by another process.
31658 @var{pathnameptr} is an invalid pointer value.
31661 @var{pathname} was too long.
31664 A directory component in @var{pathname} does not exist.
31667 A component of the path is not a directory.
31670 The file is on a read-only filesystem.
31673 The call was interrupted by the user.
31679 @unnumberedsubsubsec stat/fstat
31680 @cindex fstat, file-i/o system call
31681 @cindex stat, file-i/o system call
31686 int stat(const char *pathname, struct stat *buf);
31687 int fstat(int fd, struct stat *buf);
31691 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
31692 @samp{Ffstat,@var{fd},@var{bufptr}}
31694 @item Return value:
31695 On success, zero is returned. On error, -1 is returned.
31701 @var{fd} is not a valid open file.
31704 A directory component in @var{pathname} does not exist or the
31705 path is an empty string.
31708 A component of the path is not a directory.
31711 @var{pathnameptr} is an invalid pointer value.
31714 No access to the file or the path of the file.
31717 @var{pathname} was too long.
31720 The call was interrupted by the user.
31726 @unnumberedsubsubsec gettimeofday
31727 @cindex gettimeofday, file-i/o system call
31732 int gettimeofday(struct timeval *tv, void *tz);
31736 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
31738 @item Return value:
31739 On success, 0 is returned, -1 otherwise.
31745 @var{tz} is a non-NULL pointer.
31748 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
31754 @unnumberedsubsubsec isatty
31755 @cindex isatty, file-i/o system call
31760 int isatty(int fd);
31764 @samp{Fisatty,@var{fd}}
31766 @item Return value:
31767 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
31773 The call was interrupted by the user.
31778 Note that the @code{isatty} call is treated as a special case: it returns
31779 1 to the target if the file descriptor is attached
31780 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
31781 would require implementing @code{ioctl} and would be more complex than
31786 @unnumberedsubsubsec system
31787 @cindex system, file-i/o system call
31792 int system(const char *command);
31796 @samp{Fsystem,@var{commandptr}/@var{len}}
31798 @item Return value:
31799 If @var{len} is zero, the return value indicates whether a shell is
31800 available. A zero return value indicates a shell is not available.
31801 For non-zero @var{len}, the value returned is -1 on error and the
31802 return status of the command otherwise. Only the exit status of the
31803 command is returned, which is extracted from the host's @code{system}
31804 return value by calling @code{WEXITSTATUS(retval)}. In case
31805 @file{/bin/sh} could not be executed, 127 is returned.
31811 The call was interrupted by the user.
31816 @value{GDBN} takes over the full task of calling the necessary host calls
31817 to perform the @code{system} call. The return value of @code{system} on
31818 the host is simplified before it's returned
31819 to the target. Any termination signal information from the child process
31820 is discarded, and the return value consists
31821 entirely of the exit status of the called command.
31823 Due to security concerns, the @code{system} call is by default refused
31824 by @value{GDBN}. The user has to allow this call explicitly with the
31825 @code{set remote system-call-allowed 1} command.
31828 @item set remote system-call-allowed
31829 @kindex set remote system-call-allowed
31830 Control whether to allow the @code{system} calls in the File I/O
31831 protocol for the remote target. The default is zero (disabled).
31833 @item show remote system-call-allowed
31834 @kindex show remote system-call-allowed
31835 Show whether the @code{system} calls are allowed in the File I/O
31839 @node Protocol-specific Representation of Datatypes
31840 @subsection Protocol-specific Representation of Datatypes
31841 @cindex protocol-specific representation of datatypes, in file-i/o protocol
31844 * Integral Datatypes::
31846 * Memory Transfer::
31851 @node Integral Datatypes
31852 @unnumberedsubsubsec Integral Datatypes
31853 @cindex integral datatypes, in file-i/o protocol
31855 The integral datatypes used in the system calls are @code{int},
31856 @code{unsigned int}, @code{long}, @code{unsigned long},
31857 @code{mode_t}, and @code{time_t}.
31859 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
31860 implemented as 32 bit values in this protocol.
31862 @code{long} and @code{unsigned long} are implemented as 64 bit types.
31864 @xref{Limits}, for corresponding MIN and MAX values (similar to those
31865 in @file{limits.h}) to allow range checking on host and target.
31867 @code{time_t} datatypes are defined as seconds since the Epoch.
31869 All integral datatypes transferred as part of a memory read or write of a
31870 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
31873 @node Pointer Values
31874 @unnumberedsubsubsec Pointer Values
31875 @cindex pointer values, in file-i/o protocol
31877 Pointers to target data are transmitted as they are. An exception
31878 is made for pointers to buffers for which the length isn't
31879 transmitted as part of the function call, namely strings. Strings
31880 are transmitted as a pointer/length pair, both as hex values, e.g.@:
31887 which is a pointer to data of length 18 bytes at position 0x1aaf.
31888 The length is defined as the full string length in bytes, including
31889 the trailing null byte. For example, the string @code{"hello world"}
31890 at address 0x123456 is transmitted as
31896 @node Memory Transfer
31897 @unnumberedsubsubsec Memory Transfer
31898 @cindex memory transfer, in file-i/o protocol
31900 Structured data which is transferred using a memory read or write (for
31901 example, a @code{struct stat}) is expected to be in a protocol-specific format
31902 with all scalar multibyte datatypes being big endian. Translation to
31903 this representation needs to be done both by the target before the @code{F}
31904 packet is sent, and by @value{GDBN} before
31905 it transfers memory to the target. Transferred pointers to structured
31906 data should point to the already-coerced data at any time.
31910 @unnumberedsubsubsec struct stat
31911 @cindex struct stat, in file-i/o protocol
31913 The buffer of type @code{struct stat} used by the target and @value{GDBN}
31914 is defined as follows:
31918 unsigned int st_dev; /* device */
31919 unsigned int st_ino; /* inode */
31920 mode_t st_mode; /* protection */
31921 unsigned int st_nlink; /* number of hard links */
31922 unsigned int st_uid; /* user ID of owner */
31923 unsigned int st_gid; /* group ID of owner */
31924 unsigned int st_rdev; /* device type (if inode device) */
31925 unsigned long st_size; /* total size, in bytes */
31926 unsigned long st_blksize; /* blocksize for filesystem I/O */
31927 unsigned long st_blocks; /* number of blocks allocated */
31928 time_t st_atime; /* time of last access */
31929 time_t st_mtime; /* time of last modification */
31930 time_t st_ctime; /* time of last change */
31934 The integral datatypes conform to the definitions given in the
31935 appropriate section (see @ref{Integral Datatypes}, for details) so this
31936 structure is of size 64 bytes.
31938 The values of several fields have a restricted meaning and/or
31944 A value of 0 represents a file, 1 the console.
31947 No valid meaning for the target. Transmitted unchanged.
31950 Valid mode bits are described in @ref{Constants}. Any other
31951 bits have currently no meaning for the target.
31956 No valid meaning for the target. Transmitted unchanged.
31961 These values have a host and file system dependent
31962 accuracy. Especially on Windows hosts, the file system may not
31963 support exact timing values.
31966 The target gets a @code{struct stat} of the above representation and is
31967 responsible for coercing it to the target representation before
31970 Note that due to size differences between the host, target, and protocol
31971 representations of @code{struct stat} members, these members could eventually
31972 get truncated on the target.
31974 @node struct timeval
31975 @unnumberedsubsubsec struct timeval
31976 @cindex struct timeval, in file-i/o protocol
31978 The buffer of type @code{struct timeval} used by the File-I/O protocol
31979 is defined as follows:
31983 time_t tv_sec; /* second */
31984 long tv_usec; /* microsecond */
31988 The integral datatypes conform to the definitions given in the
31989 appropriate section (see @ref{Integral Datatypes}, for details) so this
31990 structure is of size 8 bytes.
31993 @subsection Constants
31994 @cindex constants, in file-i/o protocol
31996 The following values are used for the constants inside of the
31997 protocol. @value{GDBN} and target are responsible for translating these
31998 values before and after the call as needed.
32009 @unnumberedsubsubsec Open Flags
32010 @cindex open flags, in file-i/o protocol
32012 All values are given in hexadecimal representation.
32024 @node mode_t Values
32025 @unnumberedsubsubsec mode_t Values
32026 @cindex mode_t values, in file-i/o protocol
32028 All values are given in octal representation.
32045 @unnumberedsubsubsec Errno Values
32046 @cindex errno values, in file-i/o protocol
32048 All values are given in decimal representation.
32073 @code{EUNKNOWN} is used as a fallback error value if a host system returns
32074 any error value not in the list of supported error numbers.
32077 @unnumberedsubsubsec Lseek Flags
32078 @cindex lseek flags, in file-i/o protocol
32087 @unnumberedsubsubsec Limits
32088 @cindex limits, in file-i/o protocol
32090 All values are given in decimal representation.
32093 INT_MIN -2147483648
32095 UINT_MAX 4294967295
32096 LONG_MIN -9223372036854775808
32097 LONG_MAX 9223372036854775807
32098 ULONG_MAX 18446744073709551615
32101 @node File-I/O Examples
32102 @subsection File-I/O Examples
32103 @cindex file-i/o examples
32105 Example sequence of a write call, file descriptor 3, buffer is at target
32106 address 0x1234, 6 bytes should be written:
32109 <- @code{Fwrite,3,1234,6}
32110 @emph{request memory read from target}
32113 @emph{return "6 bytes written"}
32117 Example sequence of a read call, file descriptor 3, buffer is at target
32118 address 0x1234, 6 bytes should be read:
32121 <- @code{Fread,3,1234,6}
32122 @emph{request memory write to target}
32123 -> @code{X1234,6:XXXXXX}
32124 @emph{return "6 bytes read"}
32128 Example sequence of a read call, call fails on the host due to invalid
32129 file descriptor (@code{EBADF}):
32132 <- @code{Fread,3,1234,6}
32136 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
32140 <- @code{Fread,3,1234,6}
32145 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
32149 <- @code{Fread,3,1234,6}
32150 -> @code{X1234,6:XXXXXX}
32154 @node Library List Format
32155 @section Library List Format
32156 @cindex library list format, remote protocol
32158 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
32159 same process as your application to manage libraries. In this case,
32160 @value{GDBN} can use the loader's symbol table and normal memory
32161 operations to maintain a list of shared libraries. On other
32162 platforms, the operating system manages loaded libraries.
32163 @value{GDBN} can not retrieve the list of currently loaded libraries
32164 through memory operations, so it uses the @samp{qXfer:libraries:read}
32165 packet (@pxref{qXfer library list read}) instead. The remote stub
32166 queries the target's operating system and reports which libraries
32169 The @samp{qXfer:libraries:read} packet returns an XML document which
32170 lists loaded libraries and their offsets. Each library has an
32171 associated name and one or more segment or section base addresses,
32172 which report where the library was loaded in memory.
32174 For the common case of libraries that are fully linked binaries, the
32175 library should have a list of segments. If the target supports
32176 dynamic linking of a relocatable object file, its library XML element
32177 should instead include a list of allocated sections. The segment or
32178 section bases are start addresses, not relocation offsets; they do not
32179 depend on the library's link-time base addresses.
32181 @value{GDBN} must be linked with the Expat library to support XML
32182 library lists. @xref{Expat}.
32184 A simple memory map, with one loaded library relocated by a single
32185 offset, looks like this:
32189 <library name="/lib/libc.so.6">
32190 <segment address="0x10000000"/>
32195 Another simple memory map, with one loaded library with three
32196 allocated sections (.text, .data, .bss), looks like this:
32200 <library name="sharedlib.o">
32201 <section address="0x10000000"/>
32202 <section address="0x20000000"/>
32203 <section address="0x30000000"/>
32208 The format of a library list is described by this DTD:
32211 <!-- library-list: Root element with versioning -->
32212 <!ELEMENT library-list (library)*>
32213 <!ATTLIST library-list version CDATA #FIXED "1.0">
32214 <!ELEMENT library (segment*, section*)>
32215 <!ATTLIST library name CDATA #REQUIRED>
32216 <!ELEMENT segment EMPTY>
32217 <!ATTLIST segment address CDATA #REQUIRED>
32218 <!ELEMENT section EMPTY>
32219 <!ATTLIST section address CDATA #REQUIRED>
32222 In addition, segments and section descriptors cannot be mixed within a
32223 single library element, and you must supply at least one segment or
32224 section for each library.
32226 @node Memory Map Format
32227 @section Memory Map Format
32228 @cindex memory map format
32230 To be able to write into flash memory, @value{GDBN} needs to obtain a
32231 memory map from the target. This section describes the format of the
32234 The memory map is obtained using the @samp{qXfer:memory-map:read}
32235 (@pxref{qXfer memory map read}) packet and is an XML document that
32236 lists memory regions.
32238 @value{GDBN} must be linked with the Expat library to support XML
32239 memory maps. @xref{Expat}.
32241 The top-level structure of the document is shown below:
32244 <?xml version="1.0"?>
32245 <!DOCTYPE memory-map
32246 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
32247 "http://sourceware.org/gdb/gdb-memory-map.dtd">
32253 Each region can be either:
32258 A region of RAM starting at @var{addr} and extending for @var{length}
32262 <memory type="ram" start="@var{addr}" length="@var{length}"/>
32267 A region of read-only memory:
32270 <memory type="rom" start="@var{addr}" length="@var{length}"/>
32275 A region of flash memory, with erasure blocks @var{blocksize}
32279 <memory type="flash" start="@var{addr}" length="@var{length}">
32280 <property name="blocksize">@var{blocksize}</property>
32286 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
32287 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
32288 packets to write to addresses in such ranges.
32290 The formal DTD for memory map format is given below:
32293 <!-- ................................................... -->
32294 <!-- Memory Map XML DTD ................................ -->
32295 <!-- File: memory-map.dtd .............................. -->
32296 <!-- .................................... .............. -->
32297 <!-- memory-map.dtd -->
32298 <!-- memory-map: Root element with versioning -->
32299 <!ELEMENT memory-map (memory | property)>
32300 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
32301 <!ELEMENT memory (property)>
32302 <!-- memory: Specifies a memory region,
32303 and its type, or device. -->
32304 <!ATTLIST memory type CDATA #REQUIRED
32305 start CDATA #REQUIRED
32306 length CDATA #REQUIRED
32307 device CDATA #IMPLIED>
32308 <!-- property: Generic attribute tag -->
32309 <!ELEMENT property (#PCDATA | property)*>
32310 <!ATTLIST property name CDATA #REQUIRED>
32313 @node Thread List Format
32314 @section Thread List Format
32315 @cindex thread list format
32317 To efficiently update the list of threads and their attributes,
32318 @value{GDBN} issues the @samp{qXfer:threads:read} packet
32319 (@pxref{qXfer threads read}) and obtains the XML document with
32320 the following structure:
32323 <?xml version="1.0"?>
32325 <thread id="id" core="0">
32326 ... description ...
32331 Each @samp{thread} element must have the @samp{id} attribute that
32332 identifies the thread (@pxref{thread-id syntax}). The
32333 @samp{core} attribute, if present, specifies which processor core
32334 the thread was last executing on. The content of the of @samp{thread}
32335 element is interpreted as human-readable auxilliary information.
32337 @include agentexpr.texi
32339 @node Trace File Format
32340 @appendix Trace File Format
32341 @cindex trace file format
32343 The trace file comes in three parts: a header, a textual description
32344 section, and a trace frame section with binary data.
32346 The header has the form @code{\x7fTRACE0\n}. The first byte is
32347 @code{0x7f} so as to indicate that the file contains binary data,
32348 while the @code{0} is a version number that may have different values
32351 The description section consists of multiple lines of @sc{ascii} text
32352 separated by newline characters (@code{0xa}). The lines may include a
32353 variety of optional descriptive or context-setting information, such
32354 as tracepoint definitions or register set size. @value{GDBN} will
32355 ignore any line that it does not recognize. An empty line marks the end
32358 @c FIXME add some specific types of data
32360 The trace frame section consists of a number of consecutive frames.
32361 Each frame begins with a two-byte tracepoint number, followed by a
32362 four-byte size giving the amount of data in the frame. The data in
32363 the frame consists of a number of blocks, each introduced by a
32364 character indicating its type (at least register, memory, and trace
32365 state variable). The data in this section is raw binary, not a
32366 hexadecimal or other encoding; its endianness matches the target's
32369 @c FIXME bi-arch may require endianness/arch info in description section
32372 @item R @var{bytes}
32373 Register block. The number and ordering of bytes matches that of a
32374 @code{g} packet in the remote protocol. Note that these are the
32375 actual bytes, in target order and @value{GDBN} register order, not a
32376 hexadecimal encoding.
32378 @item M @var{address} @var{length} @var{bytes}...
32379 Memory block. This is a contiguous block of memory, at the 8-byte
32380 address @var{address}, with a 2-byte length @var{length}, followed by
32381 @var{length} bytes.
32383 @item V @var{number} @var{value}
32384 Trace state variable block. This records the 8-byte signed value
32385 @var{value} of trace state variable numbered @var{number}.
32389 Future enhancements of the trace file format may include additional types
32392 @node Target Descriptions
32393 @appendix Target Descriptions
32394 @cindex target descriptions
32396 @strong{Warning:} target descriptions are still under active development,
32397 and the contents and format may change between @value{GDBN} releases.
32398 The format is expected to stabilize in the future.
32400 One of the challenges of using @value{GDBN} to debug embedded systems
32401 is that there are so many minor variants of each processor
32402 architecture in use. It is common practice for vendors to start with
32403 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
32404 and then make changes to adapt it to a particular market niche. Some
32405 architectures have hundreds of variants, available from dozens of
32406 vendors. This leads to a number of problems:
32410 With so many different customized processors, it is difficult for
32411 the @value{GDBN} maintainers to keep up with the changes.
32413 Since individual variants may have short lifetimes or limited
32414 audiences, it may not be worthwhile to carry information about every
32415 variant in the @value{GDBN} source tree.
32417 When @value{GDBN} does support the architecture of the embedded system
32418 at hand, the task of finding the correct architecture name to give the
32419 @command{set architecture} command can be error-prone.
32422 To address these problems, the @value{GDBN} remote protocol allows a
32423 target system to not only identify itself to @value{GDBN}, but to
32424 actually describe its own features. This lets @value{GDBN} support
32425 processor variants it has never seen before --- to the extent that the
32426 descriptions are accurate, and that @value{GDBN} understands them.
32428 @value{GDBN} must be linked with the Expat library to support XML
32429 target descriptions. @xref{Expat}.
32432 * Retrieving Descriptions:: How descriptions are fetched from a target.
32433 * Target Description Format:: The contents of a target description.
32434 * Predefined Target Types:: Standard types available for target
32436 * Standard Target Features:: Features @value{GDBN} knows about.
32439 @node Retrieving Descriptions
32440 @section Retrieving Descriptions
32442 Target descriptions can be read from the target automatically, or
32443 specified by the user manually. The default behavior is to read the
32444 description from the target. @value{GDBN} retrieves it via the remote
32445 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
32446 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
32447 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
32448 XML document, of the form described in @ref{Target Description
32451 Alternatively, you can specify a file to read for the target description.
32452 If a file is set, the target will not be queried. The commands to
32453 specify a file are:
32456 @cindex set tdesc filename
32457 @item set tdesc filename @var{path}
32458 Read the target description from @var{path}.
32460 @cindex unset tdesc filename
32461 @item unset tdesc filename
32462 Do not read the XML target description from a file. @value{GDBN}
32463 will use the description supplied by the current target.
32465 @cindex show tdesc filename
32466 @item show tdesc filename
32467 Show the filename to read for a target description, if any.
32471 @node Target Description Format
32472 @section Target Description Format
32473 @cindex target descriptions, XML format
32475 A target description annex is an @uref{http://www.w3.org/XML/, XML}
32476 document which complies with the Document Type Definition provided in
32477 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
32478 means you can use generally available tools like @command{xmllint} to
32479 check that your feature descriptions are well-formed and valid.
32480 However, to help people unfamiliar with XML write descriptions for
32481 their targets, we also describe the grammar here.
32483 Target descriptions can identify the architecture of the remote target
32484 and (for some architectures) provide information about custom register
32485 sets. They can also identify the OS ABI of the remote target.
32486 @value{GDBN} can use this information to autoconfigure for your
32487 target, or to warn you if you connect to an unsupported target.
32489 Here is a simple target description:
32492 <target version="1.0">
32493 <architecture>i386:x86-64</architecture>
32498 This minimal description only says that the target uses
32499 the x86-64 architecture.
32501 A target description has the following overall form, with [ ] marking
32502 optional elements and @dots{} marking repeatable elements. The elements
32503 are explained further below.
32506 <?xml version="1.0"?>
32507 <!DOCTYPE target SYSTEM "gdb-target.dtd">
32508 <target version="1.0">
32509 @r{[}@var{architecture}@r{]}
32510 @r{[}@var{osabi}@r{]}
32511 @r{[}@var{compatible}@r{]}
32512 @r{[}@var{feature}@dots{}@r{]}
32517 The description is generally insensitive to whitespace and line
32518 breaks, under the usual common-sense rules. The XML version
32519 declaration and document type declaration can generally be omitted
32520 (@value{GDBN} does not require them), but specifying them may be
32521 useful for XML validation tools. The @samp{version} attribute for
32522 @samp{<target>} may also be omitted, but we recommend
32523 including it; if future versions of @value{GDBN} use an incompatible
32524 revision of @file{gdb-target.dtd}, they will detect and report
32525 the version mismatch.
32527 @subsection Inclusion
32528 @cindex target descriptions, inclusion
32531 @cindex <xi:include>
32534 It can sometimes be valuable to split a target description up into
32535 several different annexes, either for organizational purposes, or to
32536 share files between different possible target descriptions. You can
32537 divide a description into multiple files by replacing any element of
32538 the target description with an inclusion directive of the form:
32541 <xi:include href="@var{document}"/>
32545 When @value{GDBN} encounters an element of this form, it will retrieve
32546 the named XML @var{document}, and replace the inclusion directive with
32547 the contents of that document. If the current description was read
32548 using @samp{qXfer}, then so will be the included document;
32549 @var{document} will be interpreted as the name of an annex. If the
32550 current description was read from a file, @value{GDBN} will look for
32551 @var{document} as a file in the same directory where it found the
32552 original description.
32554 @subsection Architecture
32555 @cindex <architecture>
32557 An @samp{<architecture>} element has this form:
32560 <architecture>@var{arch}</architecture>
32563 @var{arch} is one of the architectures from the set accepted by
32564 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
32567 @cindex @code{<osabi>}
32569 This optional field was introduced in @value{GDBN} version 7.0.
32570 Previous versions of @value{GDBN} ignore it.
32572 An @samp{<osabi>} element has this form:
32575 <osabi>@var{abi-name}</osabi>
32578 @var{abi-name} is an OS ABI name from the same selection accepted by
32579 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
32581 @subsection Compatible Architecture
32582 @cindex @code{<compatible>}
32584 This optional field was introduced in @value{GDBN} version 7.0.
32585 Previous versions of @value{GDBN} ignore it.
32587 A @samp{<compatible>} element has this form:
32590 <compatible>@var{arch}</compatible>
32593 @var{arch} is one of the architectures from the set accepted by
32594 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
32596 A @samp{<compatible>} element is used to specify that the target
32597 is able to run binaries in some other than the main target architecture
32598 given by the @samp{<architecture>} element. For example, on the
32599 Cell Broadband Engine, the main architecture is @code{powerpc:common}
32600 or @code{powerpc:common64}, but the system is able to run binaries
32601 in the @code{spu} architecture as well. The way to describe this
32602 capability with @samp{<compatible>} is as follows:
32605 <architecture>powerpc:common</architecture>
32606 <compatible>spu</compatible>
32609 @subsection Features
32612 Each @samp{<feature>} describes some logical portion of the target
32613 system. Features are currently used to describe available CPU
32614 registers and the types of their contents. A @samp{<feature>} element
32618 <feature name="@var{name}">
32619 @r{[}@var{type}@dots{}@r{]}
32625 Each feature's name should be unique within the description. The name
32626 of a feature does not matter unless @value{GDBN} has some special
32627 knowledge of the contents of that feature; if it does, the feature
32628 should have its standard name. @xref{Standard Target Features}.
32632 Any register's value is a collection of bits which @value{GDBN} must
32633 interpret. The default interpretation is a two's complement integer,
32634 but other types can be requested by name in the register description.
32635 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
32636 Target Types}), and the description can define additional composite types.
32638 Each type element must have an @samp{id} attribute, which gives
32639 a unique (within the containing @samp{<feature>}) name to the type.
32640 Types must be defined before they are used.
32643 Some targets offer vector registers, which can be treated as arrays
32644 of scalar elements. These types are written as @samp{<vector>} elements,
32645 specifying the array element type, @var{type}, and the number of elements,
32649 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
32653 If a register's value is usefully viewed in multiple ways, define it
32654 with a union type containing the useful representations. The
32655 @samp{<union>} element contains one or more @samp{<field>} elements,
32656 each of which has a @var{name} and a @var{type}:
32659 <union id="@var{id}">
32660 <field name="@var{name}" type="@var{type}"/>
32665 @subsection Registers
32668 Each register is represented as an element with this form:
32671 <reg name="@var{name}"
32672 bitsize="@var{size}"
32673 @r{[}regnum="@var{num}"@r{]}
32674 @r{[}save-restore="@var{save-restore}"@r{]}
32675 @r{[}type="@var{type}"@r{]}
32676 @r{[}group="@var{group}"@r{]}/>
32680 The components are as follows:
32685 The register's name; it must be unique within the target description.
32688 The register's size, in bits.
32691 The register's number. If omitted, a register's number is one greater
32692 than that of the previous register (either in the current feature or in
32693 a preceeding feature); the first register in the target description
32694 defaults to zero. This register number is used to read or write
32695 the register; e.g.@: it is used in the remote @code{p} and @code{P}
32696 packets, and registers appear in the @code{g} and @code{G} packets
32697 in order of increasing register number.
32700 Whether the register should be preserved across inferior function
32701 calls; this must be either @code{yes} or @code{no}. The default is
32702 @code{yes}, which is appropriate for most registers except for
32703 some system control registers; this is not related to the target's
32707 The type of the register. @var{type} may be a predefined type, a type
32708 defined in the current feature, or one of the special types @code{int}
32709 and @code{float}. @code{int} is an integer type of the correct size
32710 for @var{bitsize}, and @code{float} is a floating point type (in the
32711 architecture's normal floating point format) of the correct size for
32712 @var{bitsize}. The default is @code{int}.
32715 The register group to which this register belongs. @var{group} must
32716 be either @code{general}, @code{float}, or @code{vector}. If no
32717 @var{group} is specified, @value{GDBN} will not display the register
32718 in @code{info registers}.
32722 @node Predefined Target Types
32723 @section Predefined Target Types
32724 @cindex target descriptions, predefined types
32726 Type definitions in the self-description can build up composite types
32727 from basic building blocks, but can not define fundamental types. Instead,
32728 standard identifiers are provided by @value{GDBN} for the fundamental
32729 types. The currently supported types are:
32738 Signed integer types holding the specified number of bits.
32745 Unsigned integer types holding the specified number of bits.
32749 Pointers to unspecified code and data. The program counter and
32750 any dedicated return address register may be marked as code
32751 pointers; printing a code pointer converts it into a symbolic
32752 address. The stack pointer and any dedicated address registers
32753 may be marked as data pointers.
32756 Single precision IEEE floating point.
32759 Double precision IEEE floating point.
32762 The 12-byte extended precision format used by ARM FPA registers.
32765 The 10-byte extended precision format used by x87 registers.
32768 32bit @sc{eflags} register used by x86.
32771 32bit @sc{mxcsr} register used by x86.
32775 @node Standard Target Features
32776 @section Standard Target Features
32777 @cindex target descriptions, standard features
32779 A target description must contain either no registers or all the
32780 target's registers. If the description contains no registers, then
32781 @value{GDBN} will assume a default register layout, selected based on
32782 the architecture. If the description contains any registers, the
32783 default layout will not be used; the standard registers must be
32784 described in the target description, in such a way that @value{GDBN}
32785 can recognize them.
32787 This is accomplished by giving specific names to feature elements
32788 which contain standard registers. @value{GDBN} will look for features
32789 with those names and verify that they contain the expected registers;
32790 if any known feature is missing required registers, or if any required
32791 feature is missing, @value{GDBN} will reject the target
32792 description. You can add additional registers to any of the
32793 standard features --- @value{GDBN} will display them just as if
32794 they were added to an unrecognized feature.
32796 This section lists the known features and their expected contents.
32797 Sample XML documents for these features are included in the
32798 @value{GDBN} source tree, in the directory @file{gdb/features}.
32800 Names recognized by @value{GDBN} should include the name of the
32801 company or organization which selected the name, and the overall
32802 architecture to which the feature applies; so e.g.@: the feature
32803 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
32805 The names of registers are not case sensitive for the purpose
32806 of recognizing standard features, but @value{GDBN} will only display
32807 registers using the capitalization used in the description.
32814 * PowerPC Features::
32819 @subsection ARM Features
32820 @cindex target descriptions, ARM features
32822 The @samp{org.gnu.gdb.arm.core} feature is required for ARM targets.
32823 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
32824 @samp{lr}, @samp{pc}, and @samp{cpsr}.
32826 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
32827 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
32829 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
32830 it should contain at least registers @samp{wR0} through @samp{wR15} and
32831 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
32832 @samp{wCSSF}, and @samp{wCASF} registers are optional.
32834 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
32835 should contain at least registers @samp{d0} through @samp{d15}. If
32836 they are present, @samp{d16} through @samp{d31} should also be included.
32837 @value{GDBN} will synthesize the single-precision registers from
32838 halves of the double-precision registers.
32840 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
32841 need to contain registers; it instructs @value{GDBN} to display the
32842 VFP double-precision registers as vectors and to synthesize the
32843 quad-precision registers from pairs of double-precision registers.
32844 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
32845 be present and include 32 double-precision registers.
32847 @node i386 Features
32848 @subsection i386 Features
32849 @cindex target descriptions, i386 features
32851 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
32852 targets. It should describe the following registers:
32856 @samp{eax} through @samp{edi} plus @samp{eip} for i386
32858 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
32860 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
32861 @samp{fs}, @samp{gs}
32863 @samp{st0} through @samp{st7}
32865 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
32866 @samp{foseg}, @samp{fooff} and @samp{fop}
32869 The register sets may be different, depending on the target.
32871 The @samp{org.gnu.gdb.i386.sse} feature is required. It should
32872 describe registers:
32876 @samp{xmm0} through @samp{xmm7} for i386
32878 @samp{xmm0} through @samp{xmm15} for amd64
32883 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
32884 describe a single register, @samp{orig_eax}.
32886 @node MIPS Features
32887 @subsection MIPS Features
32888 @cindex target descriptions, MIPS features
32890 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
32891 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
32892 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
32895 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
32896 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
32897 registers. They may be 32-bit or 64-bit depending on the target.
32899 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
32900 it may be optional in a future version of @value{GDBN}. It should
32901 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
32902 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
32904 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
32905 contain a single register, @samp{restart}, which is used by the
32906 Linux kernel to control restartable syscalls.
32908 @node M68K Features
32909 @subsection M68K Features
32910 @cindex target descriptions, M68K features
32913 @item @samp{org.gnu.gdb.m68k.core}
32914 @itemx @samp{org.gnu.gdb.coldfire.core}
32915 @itemx @samp{org.gnu.gdb.fido.core}
32916 One of those features must be always present.
32917 The feature that is present determines which flavor of m68k is
32918 used. The feature that is present should contain registers
32919 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
32920 @samp{sp}, @samp{ps} and @samp{pc}.
32922 @item @samp{org.gnu.gdb.coldfire.fp}
32923 This feature is optional. If present, it should contain registers
32924 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
32928 @node PowerPC Features
32929 @subsection PowerPC Features
32930 @cindex target descriptions, PowerPC features
32932 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
32933 targets. It should contain registers @samp{r0} through @samp{r31},
32934 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
32935 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
32937 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
32938 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
32940 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
32941 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
32944 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
32945 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
32946 will combine these registers with the floating point registers
32947 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
32948 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
32949 through @samp{vs63}, the set of vector registers for POWER7.
32951 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
32952 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
32953 @samp{spefscr}. SPE targets should provide 32-bit registers in
32954 @samp{org.gnu.gdb.power.core} and provide the upper halves in
32955 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
32956 these to present registers @samp{ev0} through @samp{ev31} to the
32959 @node Operating System Information
32960 @appendix Operating System Information
32961 @cindex operating system information
32967 Users of @value{GDBN} often wish to obtain information about the state of
32968 the operating system running on the target---for example the list of
32969 processes, or the list of open files. This section describes the
32970 mechanism that makes it possible. This mechanism is similar to the
32971 target features mechanism (@pxref{Target Descriptions}), but focuses
32972 on a different aspect of target.
32974 Operating system information is retrived from the target via the
32975 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
32976 read}). The object name in the request should be @samp{osdata}, and
32977 the @var{annex} identifies the data to be fetched.
32980 @appendixsection Process list
32981 @cindex operating system information, process list
32983 When requesting the process list, the @var{annex} field in the
32984 @samp{qXfer} request should be @samp{processes}. The returned data is
32985 an XML document. The formal syntax of this document is defined in
32986 @file{gdb/features/osdata.dtd}.
32988 An example document is:
32991 <?xml version="1.0"?>
32992 <!DOCTYPE target SYSTEM "osdata.dtd">
32993 <osdata type="processes">
32995 <column name="pid">1</column>
32996 <column name="user">root</column>
32997 <column name="command">/sbin/init</column>
32998 <column name="cores">1,2,3</column>
33003 Each item should include a column whose name is @samp{pid}. The value
33004 of that column should identify the process on the target. The
33005 @samp{user} and @samp{command} columns are optional, and will be
33006 displayed by @value{GDBN}. The @samp{cores} column, if present,
33007 should contain a comma-separated list of cores that this process
33008 is running on. Target may provide additional columns,
33009 which @value{GDBN} currently ignores.
33023 % I think something like @colophon should be in texinfo. In the
33025 \long\def\colophon{\hbox to0pt{}\vfill
33026 \centerline{The body of this manual is set in}
33027 \centerline{\fontname\tenrm,}
33028 \centerline{with headings in {\bf\fontname\tenbf}}
33029 \centerline{and examples in {\tt\fontname\tentt}.}
33030 \centerline{{\it\fontname\tenit\/},}
33031 \centerline{{\bf\fontname\tenbf}, and}
33032 \centerline{{\sl\fontname\tensl\/}}
33033 \centerline{are used for emphasis.}\vfill}
33035 % Blame: doc@cygnus.com, 1991.