1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-2013 Free Software Foundation, Inc.
5 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
6 @c of @set vars. However, you can override filename with makeinfo -o.
13 @settitle Debugging with @value{GDBN}
14 @setchapternewpage odd
23 @c To avoid file-name clashes between index.html and Index.html, when
24 @c the manual is produced on a Posix host and then moved to a
25 @c case-insensitive filesystem (e.g., MS-Windows), we separate the
26 @c indices into two: Concept Index and all the rest.
30 @c readline appendices use @vindex, @findex and @ftable,
31 @c annotate.texi and gdbmi use @findex.
34 @c !!set GDB manual's edition---not the same as GDB version!
35 @c This is updated by GNU Press.
38 @c !!set GDB edit command default editor
41 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
43 @c This is a dir.info fragment to support semi-automated addition of
44 @c manuals to an info tree.
45 @dircategory Software development
47 * Gdb: (gdb). The GNU debugger.
51 @c man begin COPYRIGHT
52 Copyright @copyright{} 1988-2013 Free Software Foundation, Inc.
54 Permission is granted to copy, distribute and/or modify this document
55 under the terms of the GNU Free Documentation License, Version 1.3 or
56 any later version published by the Free Software Foundation; with the
57 Invariant Sections being ``Free Software'' and ``Free Software Needs
58 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
59 and with the Back-Cover Texts as in (a) below.
61 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
62 this GNU Manual. Buying copies from GNU Press supports the FSF in
63 developing GNU and promoting software freedom.''
68 This file documents the @sc{gnu} debugger @value{GDBN}.
70 This is the @value{EDITION} Edition, of @cite{Debugging with
71 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
72 @ifset VERSION_PACKAGE
73 @value{VERSION_PACKAGE}
75 Version @value{GDBVN}.
81 @title Debugging with @value{GDBN}
82 @subtitle The @sc{gnu} Source-Level Debugger
84 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
85 @ifset VERSION_PACKAGE
87 @subtitle @value{VERSION_PACKAGE}
89 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
93 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
94 \hfill {\it Debugging with @value{GDBN}}\par
95 \hfill \TeX{}info \texinfoversion\par
99 @vskip 0pt plus 1filll
100 Published by the Free Software Foundation @*
101 51 Franklin Street, Fifth Floor,
102 Boston, MA 02110-1301, USA@*
103 ISBN 978-0-9831592-3-0 @*
110 @node Top, Summary, (dir), (dir)
112 @top Debugging with @value{GDBN}
114 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
116 This is the @value{EDITION} Edition, for @value{GDBN}
117 @ifset VERSION_PACKAGE
118 @value{VERSION_PACKAGE}
120 Version @value{GDBVN}.
122 Copyright (C) 1988-2013 Free Software Foundation, Inc.
124 This edition of the GDB manual is dedicated to the memory of Fred
125 Fish. Fred was a long-standing contributor to GDB and to Free
126 software in general. We will miss him.
129 * Summary:: Summary of @value{GDBN}
130 * Sample Session:: A sample @value{GDBN} session
132 * Invocation:: Getting in and out of @value{GDBN}
133 * Commands:: @value{GDBN} commands
134 * Running:: Running programs under @value{GDBN}
135 * Stopping:: Stopping and continuing
136 * Reverse Execution:: Running programs backward
137 * Process Record and Replay:: Recording inferior's execution and replaying it
138 * Stack:: Examining the stack
139 * Source:: Examining source files
140 * Data:: Examining data
141 * Optimized Code:: Debugging optimized code
142 * Macros:: Preprocessor Macros
143 * Tracepoints:: Debugging remote targets non-intrusively
144 * Overlays:: Debugging programs that use overlays
146 * Languages:: Using @value{GDBN} with different languages
148 * Symbols:: Examining the symbol table
149 * Altering:: Altering execution
150 * GDB Files:: @value{GDBN} files
151 * Targets:: Specifying a debugging target
152 * Remote Debugging:: Debugging remote programs
153 * Configurations:: Configuration-specific information
154 * Controlling GDB:: Controlling @value{GDBN}
155 * Extending GDB:: Extending @value{GDBN}
156 * Interpreters:: Command Interpreters
157 * TUI:: @value{GDBN} Text User Interface
158 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
159 * GDB/MI:: @value{GDBN}'s Machine Interface.
160 * Annotations:: @value{GDBN}'s annotation interface.
161 * JIT Interface:: Using the JIT debugging interface.
162 * In-Process Agent:: In-Process Agent
164 * GDB Bugs:: Reporting bugs in @value{GDBN}
166 @ifset SYSTEM_READLINE
167 * Command Line Editing: (rluserman). Command Line Editing
168 * Using History Interactively: (history). Using History Interactively
170 @ifclear SYSTEM_READLINE
171 * Command Line Editing:: Command Line Editing
172 * Using History Interactively:: Using History Interactively
174 * In Memoriam:: In Memoriam
175 * Formatting Documentation:: How to format and print @value{GDBN} documentation
176 * Installing GDB:: Installing GDB
177 * Maintenance Commands:: Maintenance Commands
178 * Remote Protocol:: GDB Remote Serial Protocol
179 * Agent Expressions:: The GDB Agent Expression Mechanism
180 * Target Descriptions:: How targets can describe themselves to
182 * Operating System Information:: Getting additional information from
184 * Trace File Format:: GDB trace file format
185 * Index Section Format:: .gdb_index section format
186 * Man Pages:: Manual pages
187 * Copying:: GNU General Public License says
188 how you can copy and share GDB
189 * GNU Free Documentation License:: The license for this documentation
190 * Concept Index:: Index of @value{GDBN} concepts
191 * Command and Variable Index:: Index of @value{GDBN} commands, variables,
192 functions, and Python data types
200 @unnumbered Summary of @value{GDBN}
202 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
203 going on ``inside'' another program while it executes---or what another
204 program was doing at the moment it crashed.
206 @value{GDBN} can do four main kinds of things (plus other things in support of
207 these) to help you catch bugs in the act:
211 Start your program, specifying anything that might affect its behavior.
214 Make your program stop on specified conditions.
217 Examine what has happened, when your program has stopped.
220 Change things in your program, so you can experiment with correcting the
221 effects of one bug and go on to learn about another.
224 You can use @value{GDBN} to debug programs written in C and C@t{++}.
225 For more information, see @ref{Supported Languages,,Supported Languages}.
226 For more information, see @ref{C,,C and C++}.
228 Support for D is partial. For information on D, see
232 Support for Modula-2 is partial. For information on Modula-2, see
233 @ref{Modula-2,,Modula-2}.
235 Support for OpenCL C is partial. For information on OpenCL C, see
236 @ref{OpenCL C,,OpenCL C}.
239 Debugging Pascal programs which use sets, subranges, file variables, or
240 nested functions does not currently work. @value{GDBN} does not support
241 entering expressions, printing values, or similar features using Pascal
245 @value{GDBN} can be used to debug programs written in Fortran, although
246 it may be necessary to refer to some variables with a trailing
249 @value{GDBN} can be used to debug programs written in Objective-C,
250 using either the Apple/NeXT or the GNU Objective-C runtime.
253 * Free Software:: Freely redistributable software
254 * Free Documentation:: Free Software Needs Free Documentation
255 * Contributors:: Contributors to GDB
259 @unnumberedsec Free Software
261 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
262 General Public License
263 (GPL). The GPL gives you the freedom to copy or adapt a licensed
264 program---but every person getting a copy also gets with it the
265 freedom to modify that copy (which means that they must get access to
266 the source code), and the freedom to distribute further copies.
267 Typical software companies use copyrights to limit your freedoms; the
268 Free Software Foundation uses the GPL to preserve these freedoms.
270 Fundamentally, the General Public License is a license which says that
271 you have these freedoms and that you cannot take these freedoms away
274 @node Free Documentation
275 @unnumberedsec Free Software Needs Free Documentation
277 The biggest deficiency in the free software community today is not in
278 the software---it is the lack of good free documentation that we can
279 include with the free software. Many of our most important
280 programs do not come with free reference manuals and free introductory
281 texts. Documentation is an essential part of any software package;
282 when an important free software package does not come with a free
283 manual and a free tutorial, that is a major gap. We have many such
286 Consider Perl, for instance. The tutorial manuals that people
287 normally use are non-free. How did this come about? Because the
288 authors of those manuals published them with restrictive terms---no
289 copying, no modification, source files not available---which exclude
290 them from the free software world.
292 That wasn't the first time this sort of thing happened, and it was far
293 from the last. Many times we have heard a GNU user eagerly describe a
294 manual that he is writing, his intended contribution to the community,
295 only to learn that he had ruined everything by signing a publication
296 contract to make it non-free.
298 Free documentation, like free software, is a matter of freedom, not
299 price. The problem with the non-free manual is not that publishers
300 charge a price for printed copies---that in itself is fine. (The Free
301 Software Foundation sells printed copies of manuals, too.) The
302 problem is the restrictions on the use of the manual. Free manuals
303 are available in source code form, and give you permission to copy and
304 modify. Non-free manuals do not allow this.
306 The criteria of freedom for a free manual are roughly the same as for
307 free software. Redistribution (including the normal kinds of
308 commercial redistribution) must be permitted, so that the manual can
309 accompany every copy of the program, both on-line and on paper.
311 Permission for modification of the technical content is crucial too.
312 When people modify the software, adding or changing features, if they
313 are conscientious they will change the manual too---so they can
314 provide accurate and clear documentation for the modified program. A
315 manual that leaves you no choice but to write a new manual to document
316 a changed version of the program is not really available to our
319 Some kinds of limits on the way modification is handled are
320 acceptable. For example, requirements to preserve the original
321 author's copyright notice, the distribution terms, or the list of
322 authors, are ok. It is also no problem to require modified versions
323 to include notice that they were modified. Even entire sections that
324 may not be deleted or changed are acceptable, as long as they deal
325 with nontechnical topics (like this one). These kinds of restrictions
326 are acceptable because they don't obstruct the community's normal use
329 However, it must be possible to modify all the @emph{technical}
330 content of the manual, and then distribute the result in all the usual
331 media, through all the usual channels. Otherwise, the restrictions
332 obstruct the use of the manual, it is not free, and we need another
333 manual to replace it.
335 Please spread the word about this issue. Our community continues to
336 lose manuals to proprietary publishing. If we spread the word that
337 free software needs free reference manuals and free tutorials, perhaps
338 the next person who wants to contribute by writing documentation will
339 realize, before it is too late, that only free manuals contribute to
340 the free software community.
342 If you are writing documentation, please insist on publishing it under
343 the GNU Free Documentation License or another free documentation
344 license. Remember that this decision requires your approval---you
345 don't have to let the publisher decide. Some commercial publishers
346 will use a free license if you insist, but they will not propose the
347 option; it is up to you to raise the issue and say firmly that this is
348 what you want. If the publisher you are dealing with refuses, please
349 try other publishers. If you're not sure whether a proposed license
350 is free, write to @email{licensing@@gnu.org}.
352 You can encourage commercial publishers to sell more free, copylefted
353 manuals and tutorials by buying them, and particularly by buying
354 copies from the publishers that paid for their writing or for major
355 improvements. Meanwhile, try to avoid buying non-free documentation
356 at all. Check the distribution terms of a manual before you buy it,
357 and insist that whoever seeks your business must respect your freedom.
358 Check the history of the book, and try to reward the publishers that
359 have paid or pay the authors to work on it.
361 The Free Software Foundation maintains a list of free documentation
362 published by other publishers, at
363 @url{http://www.fsf.org/doc/other-free-books.html}.
366 @unnumberedsec Contributors to @value{GDBN}
368 Richard Stallman was the original author of @value{GDBN}, and of many
369 other @sc{gnu} programs. Many others have contributed to its
370 development. This section attempts to credit major contributors. One
371 of the virtues of free software is that everyone is free to contribute
372 to it; with regret, we cannot actually acknowledge everyone here. The
373 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
374 blow-by-blow account.
376 Changes much prior to version 2.0 are lost in the mists of time.
379 @emph{Plea:} Additions to this section are particularly welcome. If you
380 or your friends (or enemies, to be evenhanded) have been unfairly
381 omitted from this list, we would like to add your names!
384 So that they may not regard their many labors as thankless, we
385 particularly thank those who shepherded @value{GDBN} through major
387 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
388 Jim Blandy (release 4.18);
389 Jason Molenda (release 4.17);
390 Stan Shebs (release 4.14);
391 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
392 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
393 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
394 Jim Kingdon (releases 3.5, 3.4, and 3.3);
395 and Randy Smith (releases 3.2, 3.1, and 3.0).
397 Richard Stallman, assisted at various times by Peter TerMaat, Chris
398 Hanson, and Richard Mlynarik, handled releases through 2.8.
400 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
401 in @value{GDBN}, with significant additional contributions from Per
402 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
403 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
404 much general update work leading to release 3.0).
406 @value{GDBN} uses the BFD subroutine library to examine multiple
407 object-file formats; BFD was a joint project of David V.
408 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
410 David Johnson wrote the original COFF support; Pace Willison did
411 the original support for encapsulated COFF.
413 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
415 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
416 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
418 Jean-Daniel Fekete contributed Sun 386i support.
419 Chris Hanson improved the HP9000 support.
420 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
421 David Johnson contributed Encore Umax support.
422 Jyrki Kuoppala contributed Altos 3068 support.
423 Jeff Law contributed HP PA and SOM support.
424 Keith Packard contributed NS32K support.
425 Doug Rabson contributed Acorn Risc Machine support.
426 Bob Rusk contributed Harris Nighthawk CX-UX support.
427 Chris Smith contributed Convex support (and Fortran debugging).
428 Jonathan Stone contributed Pyramid support.
429 Michael Tiemann contributed SPARC support.
430 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
431 Pace Willison contributed Intel 386 support.
432 Jay Vosburgh contributed Symmetry support.
433 Marko Mlinar contributed OpenRISC 1000 support.
435 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
437 Rich Schaefer and Peter Schauer helped with support of SunOS shared
440 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
441 about several machine instruction sets.
443 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
444 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
445 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
446 and RDI targets, respectively.
448 Brian Fox is the author of the readline libraries providing
449 command-line editing and command history.
451 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
452 Modula-2 support, and contributed the Languages chapter of this manual.
454 Fred Fish wrote most of the support for Unix System Vr4.
455 He also enhanced the command-completion support to cover C@t{++} overloaded
458 Hitachi America (now Renesas America), Ltd. sponsored the support for
459 H8/300, H8/500, and Super-H processors.
461 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
463 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
466 Toshiba sponsored the support for the TX39 Mips processor.
468 Matsushita sponsored the support for the MN10200 and MN10300 processors.
470 Fujitsu sponsored the support for SPARClite and FR30 processors.
472 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
475 Michael Snyder added support for tracepoints.
477 Stu Grossman wrote gdbserver.
479 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
480 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
482 The following people at the Hewlett-Packard Company contributed
483 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
484 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
485 compiler, and the Text User Interface (nee Terminal User Interface):
486 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
487 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
488 provided HP-specific information in this manual.
490 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
491 Robert Hoehne made significant contributions to the DJGPP port.
493 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
494 development since 1991. Cygnus engineers who have worked on @value{GDBN}
495 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
496 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
497 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
498 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
499 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
500 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
501 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
502 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
503 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
504 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
505 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
506 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
507 Zuhn have made contributions both large and small.
509 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
510 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
512 Jim Blandy added support for preprocessor macros, while working for Red
515 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
516 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
517 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
518 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
519 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
520 with the migration of old architectures to this new framework.
522 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
523 unwinder framework, this consisting of a fresh new design featuring
524 frame IDs, independent frame sniffers, and the sentinel frame. Mark
525 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
526 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
527 trad unwinders. The architecture-specific changes, each involving a
528 complete rewrite of the architecture's frame code, were carried out by
529 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
530 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
531 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
532 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
535 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
536 Tensilica, Inc.@: contributed support for Xtensa processors. Others
537 who have worked on the Xtensa port of @value{GDBN} in the past include
538 Steve Tjiang, John Newlin, and Scott Foehner.
540 Michael Eager and staff of Xilinx, Inc., contributed support for the
541 Xilinx MicroBlaze architecture.
544 @chapter A Sample @value{GDBN} Session
546 You can use this manual at your leisure to read all about @value{GDBN}.
547 However, a handful of commands are enough to get started using the
548 debugger. This chapter illustrates those commands.
551 In this sample session, we emphasize user input like this: @b{input},
552 to make it easier to pick out from the surrounding output.
555 @c FIXME: this example may not be appropriate for some configs, where
556 @c FIXME...primary interest is in remote use.
558 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
559 processor) exhibits the following bug: sometimes, when we change its
560 quote strings from the default, the commands used to capture one macro
561 definition within another stop working. In the following short @code{m4}
562 session, we define a macro @code{foo} which expands to @code{0000}; we
563 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
564 same thing. However, when we change the open quote string to
565 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
566 procedure fails to define a new synonym @code{baz}:
575 @b{define(bar,defn(`foo'))}
579 @b{changequote(<QUOTE>,<UNQUOTE>)}
581 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
584 m4: End of input: 0: fatal error: EOF in string
588 Let us use @value{GDBN} to try to see what is going on.
591 $ @b{@value{GDBP} m4}
592 @c FIXME: this falsifies the exact text played out, to permit smallbook
593 @c FIXME... format to come out better.
594 @value{GDBN} is free software and you are welcome to distribute copies
595 of it under certain conditions; type "show copying" to see
597 There is absolutely no warranty for @value{GDBN}; type "show warranty"
600 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
605 @value{GDBN} reads only enough symbol data to know where to find the
606 rest when needed; as a result, the first prompt comes up very quickly.
607 We now tell @value{GDBN} to use a narrower display width than usual, so
608 that examples fit in this manual.
611 (@value{GDBP}) @b{set width 70}
615 We need to see how the @code{m4} built-in @code{changequote} works.
616 Having looked at the source, we know the relevant subroutine is
617 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
618 @code{break} command.
621 (@value{GDBP}) @b{break m4_changequote}
622 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
626 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
627 control; as long as control does not reach the @code{m4_changequote}
628 subroutine, the program runs as usual:
631 (@value{GDBP}) @b{run}
632 Starting program: /work/Editorial/gdb/gnu/m4/m4
640 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
641 suspends execution of @code{m4}, displaying information about the
642 context where it stops.
645 @b{changequote(<QUOTE>,<UNQUOTE>)}
647 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
649 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
653 Now we use the command @code{n} (@code{next}) to advance execution to
654 the next line of the current function.
658 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
663 @code{set_quotes} looks like a promising subroutine. We can go into it
664 by using the command @code{s} (@code{step}) instead of @code{next}.
665 @code{step} goes to the next line to be executed in @emph{any}
666 subroutine, so it steps into @code{set_quotes}.
670 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
672 530 if (lquote != def_lquote)
676 The display that shows the subroutine where @code{m4} is now
677 suspended (and its arguments) is called a stack frame display. It
678 shows a summary of the stack. We can use the @code{backtrace}
679 command (which can also be spelled @code{bt}), to see where we are
680 in the stack as a whole: the @code{backtrace} command displays a
681 stack frame for each active subroutine.
684 (@value{GDBP}) @b{bt}
685 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
687 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
689 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
690 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
692 #4 0x79dc in expand_input () at macro.c:40
693 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
697 We step through a few more lines to see what happens. The first two
698 times, we can use @samp{s}; the next two times we use @code{n} to avoid
699 falling into the @code{xstrdup} subroutine.
703 0x3b5c 532 if (rquote != def_rquote)
705 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
706 def_lquote : xstrdup(lq);
708 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
711 538 len_lquote = strlen(rquote);
715 The last line displayed looks a little odd; we can examine the variables
716 @code{lquote} and @code{rquote} to see if they are in fact the new left
717 and right quotes we specified. We use the command @code{p}
718 (@code{print}) to see their values.
721 (@value{GDBP}) @b{p lquote}
722 $1 = 0x35d40 "<QUOTE>"
723 (@value{GDBP}) @b{p rquote}
724 $2 = 0x35d50 "<UNQUOTE>"
728 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
729 To look at some context, we can display ten lines of source
730 surrounding the current line with the @code{l} (@code{list}) command.
736 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
738 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
741 538 len_lquote = strlen(rquote);
742 539 len_rquote = strlen(lquote);
749 Let us step past the two lines that set @code{len_lquote} and
750 @code{len_rquote}, and then examine the values of those variables.
754 539 len_rquote = strlen(lquote);
757 (@value{GDBP}) @b{p len_lquote}
759 (@value{GDBP}) @b{p len_rquote}
764 That certainly looks wrong, assuming @code{len_lquote} and
765 @code{len_rquote} are meant to be the lengths of @code{lquote} and
766 @code{rquote} respectively. We can set them to better values using
767 the @code{p} command, since it can print the value of
768 any expression---and that expression can include subroutine calls and
772 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
774 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
779 Is that enough to fix the problem of using the new quotes with the
780 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
781 executing with the @code{c} (@code{continue}) command, and then try the
782 example that caused trouble initially:
788 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
795 Success! The new quotes now work just as well as the default ones. The
796 problem seems to have been just the two typos defining the wrong
797 lengths. We allow @code{m4} exit by giving it an EOF as input:
801 Program exited normally.
805 The message @samp{Program exited normally.} is from @value{GDBN}; it
806 indicates @code{m4} has finished executing. We can end our @value{GDBN}
807 session with the @value{GDBN} @code{quit} command.
810 (@value{GDBP}) @b{quit}
814 @chapter Getting In and Out of @value{GDBN}
816 This chapter discusses how to start @value{GDBN}, and how to get out of it.
820 type @samp{@value{GDBP}} to start @value{GDBN}.
822 type @kbd{quit} or @kbd{Ctrl-d} to exit.
826 * Invoking GDB:: How to start @value{GDBN}
827 * Quitting GDB:: How to quit @value{GDBN}
828 * Shell Commands:: How to use shell commands inside @value{GDBN}
829 * Logging Output:: How to log @value{GDBN}'s output to a file
833 @section Invoking @value{GDBN}
835 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
836 @value{GDBN} reads commands from the terminal until you tell it to exit.
838 You can also run @code{@value{GDBP}} with a variety of arguments and options,
839 to specify more of your debugging environment at the outset.
841 The command-line options described here are designed
842 to cover a variety of situations; in some environments, some of these
843 options may effectively be unavailable.
845 The most usual way to start @value{GDBN} is with one argument,
846 specifying an executable program:
849 @value{GDBP} @var{program}
853 You can also start with both an executable program and a core file
857 @value{GDBP} @var{program} @var{core}
860 You can, instead, specify a process ID as a second argument, if you want
861 to debug a running process:
864 @value{GDBP} @var{program} 1234
868 would attach @value{GDBN} to process @code{1234} (unless you also have a file
869 named @file{1234}; @value{GDBN} does check for a core file first).
871 Taking advantage of the second command-line argument requires a fairly
872 complete operating system; when you use @value{GDBN} as a remote
873 debugger attached to a bare board, there may not be any notion of
874 ``process'', and there is often no way to get a core dump. @value{GDBN}
875 will warn you if it is unable to attach or to read core dumps.
877 You can optionally have @code{@value{GDBP}} pass any arguments after the
878 executable file to the inferior using @code{--args}. This option stops
881 @value{GDBP} --args gcc -O2 -c foo.c
883 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
884 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
886 You can run @code{@value{GDBP}} without printing the front material, which describes
887 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
894 You can further control how @value{GDBN} starts up by using command-line
895 options. @value{GDBN} itself can remind you of the options available.
905 to display all available options and briefly describe their use
906 (@samp{@value{GDBP} -h} is a shorter equivalent).
908 All options and command line arguments you give are processed
909 in sequential order. The order makes a difference when the
910 @samp{-x} option is used.
914 * File Options:: Choosing files
915 * Mode Options:: Choosing modes
916 * Startup:: What @value{GDBN} does during startup
920 @subsection Choosing Files
922 When @value{GDBN} starts, it reads any arguments other than options as
923 specifying an executable file and core file (or process ID). This is
924 the same as if the arguments were specified by the @samp{-se} and
925 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
926 first argument that does not have an associated option flag as
927 equivalent to the @samp{-se} option followed by that argument; and the
928 second argument that does not have an associated option flag, if any, as
929 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
930 If the second argument begins with a decimal digit, @value{GDBN} will
931 first attempt to attach to it as a process, and if that fails, attempt
932 to open it as a corefile. If you have a corefile whose name begins with
933 a digit, you can prevent @value{GDBN} from treating it as a pid by
934 prefixing it with @file{./}, e.g.@: @file{./12345}.
936 If @value{GDBN} has not been configured to included core file support,
937 such as for most embedded targets, then it will complain about a second
938 argument and ignore it.
940 Many options have both long and short forms; both are shown in the
941 following list. @value{GDBN} also recognizes the long forms if you truncate
942 them, so long as enough of the option is present to be unambiguous.
943 (If you prefer, you can flag option arguments with @samp{--} rather
944 than @samp{-}, though we illustrate the more usual convention.)
946 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
947 @c way, both those who look for -foo and --foo in the index, will find
951 @item -symbols @var{file}
953 @cindex @code{--symbols}
955 Read symbol table from file @var{file}.
957 @item -exec @var{file}
959 @cindex @code{--exec}
961 Use file @var{file} as the executable file to execute when appropriate,
962 and for examining pure data in conjunction with a core dump.
966 Read symbol table from file @var{file} and use it as the executable
969 @item -core @var{file}
971 @cindex @code{--core}
973 Use file @var{file} as a core dump to examine.
975 @item -pid @var{number}
976 @itemx -p @var{number}
979 Connect to process ID @var{number}, as with the @code{attach} command.
981 @item -command @var{file}
983 @cindex @code{--command}
985 Execute commands from file @var{file}. The contents of this file is
986 evaluated exactly as the @code{source} command would.
987 @xref{Command Files,, Command files}.
989 @item -eval-command @var{command}
990 @itemx -ex @var{command}
991 @cindex @code{--eval-command}
993 Execute a single @value{GDBN} command.
995 This option may be used multiple times to call multiple commands. It may
996 also be interleaved with @samp{-command} as required.
999 @value{GDBP} -ex 'target sim' -ex 'load' \
1000 -x setbreakpoints -ex 'run' a.out
1003 @item -init-command @var{file}
1004 @itemx -ix @var{file}
1005 @cindex @code{--init-command}
1007 Execute commands from file @var{file} before loading the inferior (but
1008 after loading gdbinit files).
1011 @item -init-eval-command @var{command}
1012 @itemx -iex @var{command}
1013 @cindex @code{--init-eval-command}
1015 Execute a single @value{GDBN} command before loading the inferior (but
1016 after loading gdbinit files).
1019 @item -directory @var{directory}
1020 @itemx -d @var{directory}
1021 @cindex @code{--directory}
1023 Add @var{directory} to the path to search for source and script files.
1027 @cindex @code{--readnow}
1029 Read each symbol file's entire symbol table immediately, rather than
1030 the default, which is to read it incrementally as it is needed.
1031 This makes startup slower, but makes future operations faster.
1036 @subsection Choosing Modes
1038 You can run @value{GDBN} in various alternative modes---for example, in
1039 batch mode or quiet mode.
1047 Do not execute commands found in any initialization file.
1048 There are three init files, loaded in the following order:
1051 @item @file{system.gdbinit}
1052 This is the system-wide init file.
1053 Its location is specified with the @code{--with-system-gdbinit}
1054 configure option (@pxref{System-wide configuration}).
1055 It is loaded first when @value{GDBN} starts, before command line options
1056 have been processed.
1057 @item @file{~/.gdbinit}
1058 This is the init file in your home directory.
1059 It is loaded next, after @file{system.gdbinit}, and before
1060 command options have been processed.
1061 @item @file{./.gdbinit}
1062 This is the init file in the current directory.
1063 It is loaded last, after command line options other than @code{-x} and
1064 @code{-ex} have been processed. Command line options @code{-x} and
1065 @code{-ex} are processed last, after @file{./.gdbinit} has been loaded.
1068 For further documentation on startup processing, @xref{Startup}.
1069 For documentation on how to write command files,
1070 @xref{Command Files,,Command Files}.
1075 Do not execute commands found in @file{~/.gdbinit}, the init file
1076 in your home directory.
1082 @cindex @code{--quiet}
1083 @cindex @code{--silent}
1085 ``Quiet''. Do not print the introductory and copyright messages. These
1086 messages are also suppressed in batch mode.
1089 @cindex @code{--batch}
1090 Run in batch mode. Exit with status @code{0} after processing all the
1091 command files specified with @samp{-x} (and all commands from
1092 initialization files, if not inhibited with @samp{-n}). Exit with
1093 nonzero status if an error occurs in executing the @value{GDBN} commands
1094 in the command files. Batch mode also disables pagination, sets unlimited
1095 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1096 off} were in effect (@pxref{Messages/Warnings}).
1098 Batch mode may be useful for running @value{GDBN} as a filter, for
1099 example to download and run a program on another computer; in order to
1100 make this more useful, the message
1103 Program exited normally.
1107 (which is ordinarily issued whenever a program running under
1108 @value{GDBN} control terminates) is not issued when running in batch
1112 @cindex @code{--batch-silent}
1113 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1114 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1115 unaffected). This is much quieter than @samp{-silent} and would be useless
1116 for an interactive session.
1118 This is particularly useful when using targets that give @samp{Loading section}
1119 messages, for example.
1121 Note that targets that give their output via @value{GDBN}, as opposed to
1122 writing directly to @code{stdout}, will also be made silent.
1124 @item -return-child-result
1125 @cindex @code{--return-child-result}
1126 The return code from @value{GDBN} will be the return code from the child
1127 process (the process being debugged), with the following exceptions:
1131 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1132 internal error. In this case the exit code is the same as it would have been
1133 without @samp{-return-child-result}.
1135 The user quits with an explicit value. E.g., @samp{quit 1}.
1137 The child process never runs, or is not allowed to terminate, in which case
1138 the exit code will be -1.
1141 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1142 when @value{GDBN} is being used as a remote program loader or simulator
1147 @cindex @code{--nowindows}
1149 ``No windows''. If @value{GDBN} comes with a graphical user interface
1150 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1151 interface. If no GUI is available, this option has no effect.
1155 @cindex @code{--windows}
1157 If @value{GDBN} includes a GUI, then this option requires it to be
1160 @item -cd @var{directory}
1162 Run @value{GDBN} using @var{directory} as its working directory,
1163 instead of the current directory.
1165 @item -data-directory @var{directory}
1166 @cindex @code{--data-directory}
1167 Run @value{GDBN} using @var{directory} as its data directory.
1168 The data directory is where @value{GDBN} searches for its
1169 auxiliary files. @xref{Data Files}.
1173 @cindex @code{--fullname}
1175 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1176 subprocess. It tells @value{GDBN} to output the full file name and line
1177 number in a standard, recognizable fashion each time a stack frame is
1178 displayed (which includes each time your program stops). This
1179 recognizable format looks like two @samp{\032} characters, followed by
1180 the file name, line number and character position separated by colons,
1181 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1182 @samp{\032} characters as a signal to display the source code for the
1185 @item -annotate @var{level}
1186 @cindex @code{--annotate}
1187 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1188 effect is identical to using @samp{set annotate @var{level}}
1189 (@pxref{Annotations}). The annotation @var{level} controls how much
1190 information @value{GDBN} prints together with its prompt, values of
1191 expressions, source lines, and other types of output. Level 0 is the
1192 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1193 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1194 that control @value{GDBN}, and level 2 has been deprecated.
1196 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1200 @cindex @code{--args}
1201 Change interpretation of command line so that arguments following the
1202 executable file are passed as command line arguments to the inferior.
1203 This option stops option processing.
1205 @item -baud @var{bps}
1207 @cindex @code{--baud}
1209 Set the line speed (baud rate or bits per second) of any serial
1210 interface used by @value{GDBN} for remote debugging.
1212 @item -l @var{timeout}
1214 Set the timeout (in seconds) of any communication used by @value{GDBN}
1215 for remote debugging.
1217 @item -tty @var{device}
1218 @itemx -t @var{device}
1219 @cindex @code{--tty}
1221 Run using @var{device} for your program's standard input and output.
1222 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1224 @c resolve the situation of these eventually
1226 @cindex @code{--tui}
1227 Activate the @dfn{Text User Interface} when starting. The Text User
1228 Interface manages several text windows on the terminal, showing
1229 source, assembly, registers and @value{GDBN} command outputs
1230 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1231 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1232 Using @value{GDBN} under @sc{gnu} Emacs}).
1235 @c @cindex @code{--xdb}
1236 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1237 @c For information, see the file @file{xdb_trans.html}, which is usually
1238 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1241 @item -interpreter @var{interp}
1242 @cindex @code{--interpreter}
1243 Use the interpreter @var{interp} for interface with the controlling
1244 program or device. This option is meant to be set by programs which
1245 communicate with @value{GDBN} using it as a back end.
1246 @xref{Interpreters, , Command Interpreters}.
1248 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1249 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1250 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1251 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1252 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1253 @sc{gdb/mi} interfaces are no longer supported.
1256 @cindex @code{--write}
1257 Open the executable and core files for both reading and writing. This
1258 is equivalent to the @samp{set write on} command inside @value{GDBN}
1262 @cindex @code{--statistics}
1263 This option causes @value{GDBN} to print statistics about time and
1264 memory usage after it completes each command and returns to the prompt.
1267 @cindex @code{--version}
1268 This option causes @value{GDBN} to print its version number and
1269 no-warranty blurb, and exit.
1271 @item -configuration
1272 @cindex @code{--configuration}
1273 This option causes @value{GDBN} to print details about its build-time
1274 configuration parameters, and then exit. These details can be
1275 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1280 @subsection What @value{GDBN} Does During Startup
1281 @cindex @value{GDBN} startup
1283 Here's the description of what @value{GDBN} does during session startup:
1287 Sets up the command interpreter as specified by the command line
1288 (@pxref{Mode Options, interpreter}).
1292 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1293 used when building @value{GDBN}; @pxref{System-wide configuration,
1294 ,System-wide configuration and settings}) and executes all the commands in
1297 @anchor{Home Directory Init File}
1299 Reads the init file (if any) in your home directory@footnote{On
1300 DOS/Windows systems, the home directory is the one pointed to by the
1301 @code{HOME} environment variable.} and executes all the commands in
1304 @anchor{Option -init-eval-command}
1306 Executes commands and command files specified by the @samp{-iex} and
1307 @samp{-ix} options in their specified order. Usually you should use the
1308 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1309 settings before @value{GDBN} init files get executed and before inferior
1313 Processes command line options and operands.
1315 @anchor{Init File in the Current Directory during Startup}
1317 Reads and executes the commands from init file (if any) in the current
1318 working directory as long as @samp{set auto-load local-gdbinit} is set to
1319 @samp{on} (@pxref{Init File in the Current Directory}).
1320 This is only done if the current directory is
1321 different from your home directory. Thus, you can have more than one
1322 init file, one generic in your home directory, and another, specific
1323 to the program you are debugging, in the directory where you invoke
1327 If the command line specified a program to debug, or a process to
1328 attach to, or a core file, @value{GDBN} loads any auto-loaded
1329 scripts provided for the program or for its loaded shared libraries.
1330 @xref{Auto-loading}.
1332 If you wish to disable the auto-loading during startup,
1333 you must do something like the following:
1336 $ gdb -iex "set auto-load python-scripts off" myprogram
1339 Option @samp{-ex} does not work because the auto-loading is then turned
1343 Executes commands and command files specified by the @samp{-ex} and
1344 @samp{-x} options in their specified order. @xref{Command Files}, for
1345 more details about @value{GDBN} command files.
1348 Reads the command history recorded in the @dfn{history file}.
1349 @xref{Command History}, for more details about the command history and the
1350 files where @value{GDBN} records it.
1353 Init files use the same syntax as @dfn{command files} (@pxref{Command
1354 Files}) and are processed by @value{GDBN} in the same way. The init
1355 file in your home directory can set options (such as @samp{set
1356 complaints}) that affect subsequent processing of command line options
1357 and operands. Init files are not executed if you use the @samp{-nx}
1358 option (@pxref{Mode Options, ,Choosing Modes}).
1360 To display the list of init files loaded by gdb at startup, you
1361 can use @kbd{gdb --help}.
1363 @cindex init file name
1364 @cindex @file{.gdbinit}
1365 @cindex @file{gdb.ini}
1366 The @value{GDBN} init files are normally called @file{.gdbinit}.
1367 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1368 the limitations of file names imposed by DOS filesystems. The Windows
1369 port of @value{GDBN} uses the standard name, but if it finds a
1370 @file{gdb.ini} file in your home directory, it warns you about that
1371 and suggests to rename the file to the standard name.
1375 @section Quitting @value{GDBN}
1376 @cindex exiting @value{GDBN}
1377 @cindex leaving @value{GDBN}
1380 @kindex quit @r{[}@var{expression}@r{]}
1381 @kindex q @r{(@code{quit})}
1382 @item quit @r{[}@var{expression}@r{]}
1384 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1385 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1386 do not supply @var{expression}, @value{GDBN} will terminate normally;
1387 otherwise it will terminate using the result of @var{expression} as the
1392 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1393 terminates the action of any @value{GDBN} command that is in progress and
1394 returns to @value{GDBN} command level. It is safe to type the interrupt
1395 character at any time because @value{GDBN} does not allow it to take effect
1396 until a time when it is safe.
1398 If you have been using @value{GDBN} to control an attached process or
1399 device, you can release it with the @code{detach} command
1400 (@pxref{Attach, ,Debugging an Already-running Process}).
1402 @node Shell Commands
1403 @section Shell Commands
1405 If you need to execute occasional shell commands during your
1406 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1407 just use the @code{shell} command.
1412 @cindex shell escape
1413 @item shell @var{command-string}
1414 @itemx !@var{command-string}
1415 Invoke a standard shell to execute @var{command-string}.
1416 Note that no space is needed between @code{!} and @var{command-string}.
1417 If it exists, the environment variable @code{SHELL} determines which
1418 shell to run. Otherwise @value{GDBN} uses the default shell
1419 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1422 The utility @code{make} is often needed in development environments.
1423 You do not have to use the @code{shell} command for this purpose in
1428 @cindex calling make
1429 @item make @var{make-args}
1430 Execute the @code{make} program with the specified
1431 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1434 @node Logging Output
1435 @section Logging Output
1436 @cindex logging @value{GDBN} output
1437 @cindex save @value{GDBN} output to a file
1439 You may want to save the output of @value{GDBN} commands to a file.
1440 There are several commands to control @value{GDBN}'s logging.
1444 @item set logging on
1446 @item set logging off
1448 @cindex logging file name
1449 @item set logging file @var{file}
1450 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1451 @item set logging overwrite [on|off]
1452 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1453 you want @code{set logging on} to overwrite the logfile instead.
1454 @item set logging redirect [on|off]
1455 By default, @value{GDBN} output will go to both the terminal and the logfile.
1456 Set @code{redirect} if you want output to go only to the log file.
1457 @kindex show logging
1459 Show the current values of the logging settings.
1463 @chapter @value{GDBN} Commands
1465 You can abbreviate a @value{GDBN} command to the first few letters of the command
1466 name, if that abbreviation is unambiguous; and you can repeat certain
1467 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1468 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1469 show you the alternatives available, if there is more than one possibility).
1472 * Command Syntax:: How to give commands to @value{GDBN}
1473 * Completion:: Command completion
1474 * Help:: How to ask @value{GDBN} for help
1477 @node Command Syntax
1478 @section Command Syntax
1480 A @value{GDBN} command is a single line of input. There is no limit on
1481 how long it can be. It starts with a command name, which is followed by
1482 arguments whose meaning depends on the command name. For example, the
1483 command @code{step} accepts an argument which is the number of times to
1484 step, as in @samp{step 5}. You can also use the @code{step} command
1485 with no arguments. Some commands do not allow any arguments.
1487 @cindex abbreviation
1488 @value{GDBN} command names may always be truncated if that abbreviation is
1489 unambiguous. Other possible command abbreviations are listed in the
1490 documentation for individual commands. In some cases, even ambiguous
1491 abbreviations are allowed; for example, @code{s} is specially defined as
1492 equivalent to @code{step} even though there are other commands whose
1493 names start with @code{s}. You can test abbreviations by using them as
1494 arguments to the @code{help} command.
1496 @cindex repeating commands
1497 @kindex RET @r{(repeat last command)}
1498 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1499 repeat the previous command. Certain commands (for example, @code{run})
1500 will not repeat this way; these are commands whose unintentional
1501 repetition might cause trouble and which you are unlikely to want to
1502 repeat. User-defined commands can disable this feature; see
1503 @ref{Define, dont-repeat}.
1505 The @code{list} and @code{x} commands, when you repeat them with
1506 @key{RET}, construct new arguments rather than repeating
1507 exactly as typed. This permits easy scanning of source or memory.
1509 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1510 output, in a way similar to the common utility @code{more}
1511 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1512 @key{RET} too many in this situation, @value{GDBN} disables command
1513 repetition after any command that generates this sort of display.
1515 @kindex # @r{(a comment)}
1517 Any text from a @kbd{#} to the end of the line is a comment; it does
1518 nothing. This is useful mainly in command files (@pxref{Command
1519 Files,,Command Files}).
1521 @cindex repeating command sequences
1522 @kindex Ctrl-o @r{(operate-and-get-next)}
1523 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1524 commands. This command accepts the current line, like @key{RET}, and
1525 then fetches the next line relative to the current line from the history
1529 @section Command Completion
1532 @cindex word completion
1533 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1534 only one possibility; it can also show you what the valid possibilities
1535 are for the next word in a command, at any time. This works for @value{GDBN}
1536 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1538 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1539 of a word. If there is only one possibility, @value{GDBN} fills in the
1540 word, and waits for you to finish the command (or press @key{RET} to
1541 enter it). For example, if you type
1543 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1544 @c complete accuracy in these examples; space introduced for clarity.
1545 @c If texinfo enhancements make it unnecessary, it would be nice to
1546 @c replace " @key" by "@key" in the following...
1548 (@value{GDBP}) info bre @key{TAB}
1552 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1553 the only @code{info} subcommand beginning with @samp{bre}:
1556 (@value{GDBP}) info breakpoints
1560 You can either press @key{RET} at this point, to run the @code{info
1561 breakpoints} command, or backspace and enter something else, if
1562 @samp{breakpoints} does not look like the command you expected. (If you
1563 were sure you wanted @code{info breakpoints} in the first place, you
1564 might as well just type @key{RET} immediately after @samp{info bre},
1565 to exploit command abbreviations rather than command completion).
1567 If there is more than one possibility for the next word when you press
1568 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1569 characters and try again, or just press @key{TAB} a second time;
1570 @value{GDBN} displays all the possible completions for that word. For
1571 example, you might want to set a breakpoint on a subroutine whose name
1572 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1573 just sounds the bell. Typing @key{TAB} again displays all the
1574 function names in your program that begin with those characters, for
1578 (@value{GDBP}) b make_ @key{TAB}
1579 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1580 make_a_section_from_file make_environ
1581 make_abs_section make_function_type
1582 make_blockvector make_pointer_type
1583 make_cleanup make_reference_type
1584 make_command make_symbol_completion_list
1585 (@value{GDBP}) b make_
1589 After displaying the available possibilities, @value{GDBN} copies your
1590 partial input (@samp{b make_} in the example) so you can finish the
1593 If you just want to see the list of alternatives in the first place, you
1594 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1595 means @kbd{@key{META} ?}. You can type this either by holding down a
1596 key designated as the @key{META} shift on your keyboard (if there is
1597 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1599 @cindex quotes in commands
1600 @cindex completion of quoted strings
1601 Sometimes the string you need, while logically a ``word'', may contain
1602 parentheses or other characters that @value{GDBN} normally excludes from
1603 its notion of a word. To permit word completion to work in this
1604 situation, you may enclose words in @code{'} (single quote marks) in
1605 @value{GDBN} commands.
1607 The most likely situation where you might need this is in typing the
1608 name of a C@t{++} function. This is because C@t{++} allows function
1609 overloading (multiple definitions of the same function, distinguished
1610 by argument type). For example, when you want to set a breakpoint you
1611 may need to distinguish whether you mean the version of @code{name}
1612 that takes an @code{int} parameter, @code{name(int)}, or the version
1613 that takes a @code{float} parameter, @code{name(float)}. To use the
1614 word-completion facilities in this situation, type a single quote
1615 @code{'} at the beginning of the function name. This alerts
1616 @value{GDBN} that it may need to consider more information than usual
1617 when you press @key{TAB} or @kbd{M-?} to request word completion:
1620 (@value{GDBP}) b 'bubble( @kbd{M-?}
1621 bubble(double,double) bubble(int,int)
1622 (@value{GDBP}) b 'bubble(
1625 In some cases, @value{GDBN} can tell that completing a name requires using
1626 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1627 completing as much as it can) if you do not type the quote in the first
1631 (@value{GDBP}) b bub @key{TAB}
1632 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1633 (@value{GDBP}) b 'bubble(
1637 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1638 you have not yet started typing the argument list when you ask for
1639 completion on an overloaded symbol.
1641 For more information about overloaded functions, see @ref{C Plus Plus
1642 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1643 overload-resolution off} to disable overload resolution;
1644 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1646 @cindex completion of structure field names
1647 @cindex structure field name completion
1648 @cindex completion of union field names
1649 @cindex union field name completion
1650 When completing in an expression which looks up a field in a
1651 structure, @value{GDBN} also tries@footnote{The completer can be
1652 confused by certain kinds of invalid expressions. Also, it only
1653 examines the static type of the expression, not the dynamic type.} to
1654 limit completions to the field names available in the type of the
1658 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1659 magic to_fputs to_rewind
1660 to_data to_isatty to_write
1661 to_delete to_put to_write_async_safe
1666 This is because the @code{gdb_stdout} is a variable of the type
1667 @code{struct ui_file} that is defined in @value{GDBN} sources as
1674 ui_file_flush_ftype *to_flush;
1675 ui_file_write_ftype *to_write;
1676 ui_file_write_async_safe_ftype *to_write_async_safe;
1677 ui_file_fputs_ftype *to_fputs;
1678 ui_file_read_ftype *to_read;
1679 ui_file_delete_ftype *to_delete;
1680 ui_file_isatty_ftype *to_isatty;
1681 ui_file_rewind_ftype *to_rewind;
1682 ui_file_put_ftype *to_put;
1689 @section Getting Help
1690 @cindex online documentation
1693 You can always ask @value{GDBN} itself for information on its commands,
1694 using the command @code{help}.
1697 @kindex h @r{(@code{help})}
1700 You can use @code{help} (abbreviated @code{h}) with no arguments to
1701 display a short list of named classes of commands:
1705 List of classes of commands:
1707 aliases -- Aliases of other commands
1708 breakpoints -- Making program stop at certain points
1709 data -- Examining data
1710 files -- Specifying and examining files
1711 internals -- Maintenance commands
1712 obscure -- Obscure features
1713 running -- Running the program
1714 stack -- Examining the stack
1715 status -- Status inquiries
1716 support -- Support facilities
1717 tracepoints -- Tracing of program execution without
1718 stopping the program
1719 user-defined -- User-defined commands
1721 Type "help" followed by a class name for a list of
1722 commands in that class.
1723 Type "help" followed by command name for full
1725 Command name abbreviations are allowed if unambiguous.
1728 @c the above line break eliminates huge line overfull...
1730 @item help @var{class}
1731 Using one of the general help classes as an argument, you can get a
1732 list of the individual commands in that class. For example, here is the
1733 help display for the class @code{status}:
1736 (@value{GDBP}) help status
1741 @c Line break in "show" line falsifies real output, but needed
1742 @c to fit in smallbook page size.
1743 info -- Generic command for showing things
1744 about the program being debugged
1745 show -- Generic command for showing things
1748 Type "help" followed by command name for full
1750 Command name abbreviations are allowed if unambiguous.
1754 @item help @var{command}
1755 With a command name as @code{help} argument, @value{GDBN} displays a
1756 short paragraph on how to use that command.
1759 @item apropos @var{args}
1760 The @code{apropos} command searches through all of the @value{GDBN}
1761 commands, and their documentation, for the regular expression specified in
1762 @var{args}. It prints out all matches found. For example:
1773 alias -- Define a new command that is an alias of an existing command
1774 aliases -- Aliases of other commands
1775 d -- Delete some breakpoints or auto-display expressions
1776 del -- Delete some breakpoints or auto-display expressions
1777 delete -- Delete some breakpoints or auto-display expressions
1782 @item complete @var{args}
1783 The @code{complete @var{args}} command lists all the possible completions
1784 for the beginning of a command. Use @var{args} to specify the beginning of the
1785 command you want completed. For example:
1791 @noindent results in:
1802 @noindent This is intended for use by @sc{gnu} Emacs.
1805 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1806 and @code{show} to inquire about the state of your program, or the state
1807 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1808 manual introduces each of them in the appropriate context. The listings
1809 under @code{info} and under @code{show} in the Command, Variable, and
1810 Function Index point to all the sub-commands. @xref{Command and Variable
1816 @kindex i @r{(@code{info})}
1818 This command (abbreviated @code{i}) is for describing the state of your
1819 program. For example, you can show the arguments passed to a function
1820 with @code{info args}, list the registers currently in use with @code{info
1821 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1822 You can get a complete list of the @code{info} sub-commands with
1823 @w{@code{help info}}.
1827 You can assign the result of an expression to an environment variable with
1828 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1829 @code{set prompt $}.
1833 In contrast to @code{info}, @code{show} is for describing the state of
1834 @value{GDBN} itself.
1835 You can change most of the things you can @code{show}, by using the
1836 related command @code{set}; for example, you can control what number
1837 system is used for displays with @code{set radix}, or simply inquire
1838 which is currently in use with @code{show radix}.
1841 To display all the settable parameters and their current
1842 values, you can use @code{show} with no arguments; you may also use
1843 @code{info set}. Both commands produce the same display.
1844 @c FIXME: "info set" violates the rule that "info" is for state of
1845 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1846 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1850 Here are several miscellaneous @code{show} subcommands, all of which are
1851 exceptional in lacking corresponding @code{set} commands:
1854 @kindex show version
1855 @cindex @value{GDBN} version number
1857 Show what version of @value{GDBN} is running. You should include this
1858 information in @value{GDBN} bug-reports. If multiple versions of
1859 @value{GDBN} are in use at your site, you may need to determine which
1860 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1861 commands are introduced, and old ones may wither away. Also, many
1862 system vendors ship variant versions of @value{GDBN}, and there are
1863 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1864 The version number is the same as the one announced when you start
1867 @kindex show copying
1868 @kindex info copying
1869 @cindex display @value{GDBN} copyright
1872 Display information about permission for copying @value{GDBN}.
1874 @kindex show warranty
1875 @kindex info warranty
1877 @itemx info warranty
1878 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1879 if your version of @value{GDBN} comes with one.
1881 @kindex show configuration
1882 @item show configuration
1883 Display detailed information about the way @value{GDBN} was configured
1884 when it was built. This displays the optional arguments passed to the
1885 @file{configure} script and also configuration parameters detected
1886 automatically by @command{configure}. When reporting a @value{GDBN}
1887 bug (@pxref{GDB Bugs}), it is important to include this information in
1893 @chapter Running Programs Under @value{GDBN}
1895 When you run a program under @value{GDBN}, you must first generate
1896 debugging information when you compile it.
1898 You may start @value{GDBN} with its arguments, if any, in an environment
1899 of your choice. If you are doing native debugging, you may redirect
1900 your program's input and output, debug an already running process, or
1901 kill a child process.
1904 * Compilation:: Compiling for debugging
1905 * Starting:: Starting your program
1906 * Arguments:: Your program's arguments
1907 * Environment:: Your program's environment
1909 * Working Directory:: Your program's working directory
1910 * Input/Output:: Your program's input and output
1911 * Attach:: Debugging an already-running process
1912 * Kill Process:: Killing the child process
1914 * Inferiors and Programs:: Debugging multiple inferiors and programs
1915 * Threads:: Debugging programs with multiple threads
1916 * Forks:: Debugging forks
1917 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1921 @section Compiling for Debugging
1923 In order to debug a program effectively, you need to generate
1924 debugging information when you compile it. This debugging information
1925 is stored in the object file; it describes the data type of each
1926 variable or function and the correspondence between source line numbers
1927 and addresses in the executable code.
1929 To request debugging information, specify the @samp{-g} option when you run
1932 Programs that are to be shipped to your customers are compiled with
1933 optimizations, using the @samp{-O} compiler option. However, some
1934 compilers are unable to handle the @samp{-g} and @samp{-O} options
1935 together. Using those compilers, you cannot generate optimized
1936 executables containing debugging information.
1938 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1939 without @samp{-O}, making it possible to debug optimized code. We
1940 recommend that you @emph{always} use @samp{-g} whenever you compile a
1941 program. You may think your program is correct, but there is no sense
1942 in pushing your luck. For more information, see @ref{Optimized Code}.
1944 Older versions of the @sc{gnu} C compiler permitted a variant option
1945 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1946 format; if your @sc{gnu} C compiler has this option, do not use it.
1948 @value{GDBN} knows about preprocessor macros and can show you their
1949 expansion (@pxref{Macros}). Most compilers do not include information
1950 about preprocessor macros in the debugging information if you specify
1951 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
1952 the @sc{gnu} C compiler, provides macro information if you are using
1953 the DWARF debugging format, and specify the option @option{-g3}.
1955 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
1956 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}, for more
1957 information on @value{NGCC} options affecting debug information.
1959 You will have the best debugging experience if you use the latest
1960 version of the DWARF debugging format that your compiler supports.
1961 DWARF is currently the most expressive and best supported debugging
1962 format in @value{GDBN}.
1966 @section Starting your Program
1972 @kindex r @r{(@code{run})}
1975 Use the @code{run} command to start your program under @value{GDBN}.
1976 You must first specify the program name (except on VxWorks) with an
1977 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1978 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1979 (@pxref{Files, ,Commands to Specify Files}).
1983 If you are running your program in an execution environment that
1984 supports processes, @code{run} creates an inferior process and makes
1985 that process run your program. In some environments without processes,
1986 @code{run} jumps to the start of your program. Other targets,
1987 like @samp{remote}, are always running. If you get an error
1988 message like this one:
1991 The "remote" target does not support "run".
1992 Try "help target" or "continue".
1996 then use @code{continue} to run your program. You may need @code{load}
1997 first (@pxref{load}).
1999 The execution of a program is affected by certain information it
2000 receives from its superior. @value{GDBN} provides ways to specify this
2001 information, which you must do @emph{before} starting your program. (You
2002 can change it after starting your program, but such changes only affect
2003 your program the next time you start it.) This information may be
2004 divided into four categories:
2007 @item The @emph{arguments.}
2008 Specify the arguments to give your program as the arguments of the
2009 @code{run} command. If a shell is available on your target, the shell
2010 is used to pass the arguments, so that you may use normal conventions
2011 (such as wildcard expansion or variable substitution) in describing
2013 In Unix systems, you can control which shell is used with the
2014 @code{SHELL} environment variable. If you do not define @code{SHELL},
2015 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2016 use of any shell with the @code{set startup-with-shell} command (see
2019 @item The @emph{environment.}
2020 Your program normally inherits its environment from @value{GDBN}, but you can
2021 use the @value{GDBN} commands @code{set environment} and @code{unset
2022 environment} to change parts of the environment that affect
2023 your program. @xref{Environment, ,Your Program's Environment}.
2025 @item The @emph{working directory.}
2026 Your program inherits its working directory from @value{GDBN}. You can set
2027 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
2028 @xref{Working Directory, ,Your Program's Working Directory}.
2030 @item The @emph{standard input and output.}
2031 Your program normally uses the same device for standard input and
2032 standard output as @value{GDBN} is using. You can redirect input and output
2033 in the @code{run} command line, or you can use the @code{tty} command to
2034 set a different device for your program.
2035 @xref{Input/Output, ,Your Program's Input and Output}.
2038 @emph{Warning:} While input and output redirection work, you cannot use
2039 pipes to pass the output of the program you are debugging to another
2040 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2044 When you issue the @code{run} command, your program begins to execute
2045 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2046 of how to arrange for your program to stop. Once your program has
2047 stopped, you may call functions in your program, using the @code{print}
2048 or @code{call} commands. @xref{Data, ,Examining Data}.
2050 If the modification time of your symbol file has changed since the last
2051 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2052 table, and reads it again. When it does this, @value{GDBN} tries to retain
2053 your current breakpoints.
2058 @cindex run to main procedure
2059 The name of the main procedure can vary from language to language.
2060 With C or C@t{++}, the main procedure name is always @code{main}, but
2061 other languages such as Ada do not require a specific name for their
2062 main procedure. The debugger provides a convenient way to start the
2063 execution of the program and to stop at the beginning of the main
2064 procedure, depending on the language used.
2066 The @samp{start} command does the equivalent of setting a temporary
2067 breakpoint at the beginning of the main procedure and then invoking
2068 the @samp{run} command.
2070 @cindex elaboration phase
2071 Some programs contain an @dfn{elaboration} phase where some startup code is
2072 executed before the main procedure is called. This depends on the
2073 languages used to write your program. In C@t{++}, for instance,
2074 constructors for static and global objects are executed before
2075 @code{main} is called. It is therefore possible that the debugger stops
2076 before reaching the main procedure. However, the temporary breakpoint
2077 will remain to halt execution.
2079 Specify the arguments to give to your program as arguments to the
2080 @samp{start} command. These arguments will be given verbatim to the
2081 underlying @samp{run} command. Note that the same arguments will be
2082 reused if no argument is provided during subsequent calls to
2083 @samp{start} or @samp{run}.
2085 It is sometimes necessary to debug the program during elaboration. In
2086 these cases, using the @code{start} command would stop the execution of
2087 your program too late, as the program would have already completed the
2088 elaboration phase. Under these circumstances, insert breakpoints in your
2089 elaboration code before running your program.
2091 @kindex set exec-wrapper
2092 @item set exec-wrapper @var{wrapper}
2093 @itemx show exec-wrapper
2094 @itemx unset exec-wrapper
2095 When @samp{exec-wrapper} is set, the specified wrapper is used to
2096 launch programs for debugging. @value{GDBN} starts your program
2097 with a shell command of the form @kbd{exec @var{wrapper}
2098 @var{program}}. Quoting is added to @var{program} and its
2099 arguments, but not to @var{wrapper}, so you should add quotes if
2100 appropriate for your shell. The wrapper runs until it executes
2101 your program, and then @value{GDBN} takes control.
2103 You can use any program that eventually calls @code{execve} with
2104 its arguments as a wrapper. Several standard Unix utilities do
2105 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2106 with @code{exec "$@@"} will also work.
2108 For example, you can use @code{env} to pass an environment variable to
2109 the debugged program, without setting the variable in your shell's
2113 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2117 This command is available when debugging locally on most targets, excluding
2118 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2120 @kindex set startup-with-shell
2121 @item set startup-with-shell
2122 @itemx set startup-with-shell on
2123 @itemx set startup-with-shell off
2124 @itemx show set startup-with-shell
2125 On Unix systems, by default, if a shell is available on your target,
2126 @value{GDBN}) uses it to start your program. Arguments of the
2127 @code{run} command are passed to the shell, which does variable
2128 substitution, expands wildcard characters and performs redirection of
2129 I/O. In some circumstances, it may be useful to disable such use of a
2130 shell, for example, when debugging the shell itself or diagnosing
2131 startup failures such as:
2135 Starting program: ./a.out
2136 During startup program terminated with signal SIGSEGV, Segmentation fault.
2140 which indicates the shell or the wrapper specified with
2141 @samp{exec-wrapper} crashed, not your program. Most often, this is
2142 caused by something odd in your shell's non-interactive mode
2143 initialization file---such as @file{.cshrc} for C-shell,
2144 $@file{.zshenv} for the Z shell, or the file specified in the
2145 @samp{BASH_ENV} environment variable for BASH.
2147 @kindex set disable-randomization
2148 @item set disable-randomization
2149 @itemx set disable-randomization on
2150 This option (enabled by default in @value{GDBN}) will turn off the native
2151 randomization of the virtual address space of the started program. This option
2152 is useful for multiple debugging sessions to make the execution better
2153 reproducible and memory addresses reusable across debugging sessions.
2155 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2156 On @sc{gnu}/Linux you can get the same behavior using
2159 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2162 @item set disable-randomization off
2163 Leave the behavior of the started executable unchanged. Some bugs rear their
2164 ugly heads only when the program is loaded at certain addresses. If your bug
2165 disappears when you run the program under @value{GDBN}, that might be because
2166 @value{GDBN} by default disables the address randomization on platforms, such
2167 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2168 disable-randomization off} to try to reproduce such elusive bugs.
2170 On targets where it is available, virtual address space randomization
2171 protects the programs against certain kinds of security attacks. In these
2172 cases the attacker needs to know the exact location of a concrete executable
2173 code. Randomizing its location makes it impossible to inject jumps misusing
2174 a code at its expected addresses.
2176 Prelinking shared libraries provides a startup performance advantage but it
2177 makes addresses in these libraries predictable for privileged processes by
2178 having just unprivileged access at the target system. Reading the shared
2179 library binary gives enough information for assembling the malicious code
2180 misusing it. Still even a prelinked shared library can get loaded at a new
2181 random address just requiring the regular relocation process during the
2182 startup. Shared libraries not already prelinked are always loaded at
2183 a randomly chosen address.
2185 Position independent executables (PIE) contain position independent code
2186 similar to the shared libraries and therefore such executables get loaded at
2187 a randomly chosen address upon startup. PIE executables always load even
2188 already prelinked shared libraries at a random address. You can build such
2189 executable using @command{gcc -fPIE -pie}.
2191 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2192 (as long as the randomization is enabled).
2194 @item show disable-randomization
2195 Show the current setting of the explicit disable of the native randomization of
2196 the virtual address space of the started program.
2201 @section Your Program's Arguments
2203 @cindex arguments (to your program)
2204 The arguments to your program can be specified by the arguments of the
2206 They are passed to a shell, which expands wildcard characters and
2207 performs redirection of I/O, and thence to your program. Your
2208 @code{SHELL} environment variable (if it exists) specifies what shell
2209 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2210 the default shell (@file{/bin/sh} on Unix).
2212 On non-Unix systems, the program is usually invoked directly by
2213 @value{GDBN}, which emulates I/O redirection via the appropriate system
2214 calls, and the wildcard characters are expanded by the startup code of
2215 the program, not by the shell.
2217 @code{run} with no arguments uses the same arguments used by the previous
2218 @code{run}, or those set by the @code{set args} command.
2223 Specify the arguments to be used the next time your program is run. If
2224 @code{set args} has no arguments, @code{run} executes your program
2225 with no arguments. Once you have run your program with arguments,
2226 using @code{set args} before the next @code{run} is the only way to run
2227 it again without arguments.
2231 Show the arguments to give your program when it is started.
2235 @section Your Program's Environment
2237 @cindex environment (of your program)
2238 The @dfn{environment} consists of a set of environment variables and
2239 their values. Environment variables conventionally record such things as
2240 your user name, your home directory, your terminal type, and your search
2241 path for programs to run. Usually you set up environment variables with
2242 the shell and they are inherited by all the other programs you run. When
2243 debugging, it can be useful to try running your program with a modified
2244 environment without having to start @value{GDBN} over again.
2248 @item path @var{directory}
2249 Add @var{directory} to the front of the @code{PATH} environment variable
2250 (the search path for executables) that will be passed to your program.
2251 The value of @code{PATH} used by @value{GDBN} does not change.
2252 You may specify several directory names, separated by whitespace or by a
2253 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2254 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2255 is moved to the front, so it is searched sooner.
2257 You can use the string @samp{$cwd} to refer to whatever is the current
2258 working directory at the time @value{GDBN} searches the path. If you
2259 use @samp{.} instead, it refers to the directory where you executed the
2260 @code{path} command. @value{GDBN} replaces @samp{.} in the
2261 @var{directory} argument (with the current path) before adding
2262 @var{directory} to the search path.
2263 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2264 @c document that, since repeating it would be a no-op.
2268 Display the list of search paths for executables (the @code{PATH}
2269 environment variable).
2271 @kindex show environment
2272 @item show environment @r{[}@var{varname}@r{]}
2273 Print the value of environment variable @var{varname} to be given to
2274 your program when it starts. If you do not supply @var{varname},
2275 print the names and values of all environment variables to be given to
2276 your program. You can abbreviate @code{environment} as @code{env}.
2278 @kindex set environment
2279 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2280 Set environment variable @var{varname} to @var{value}. The value
2281 changes for your program only, not for @value{GDBN} itself. @var{value} may
2282 be any string; the values of environment variables are just strings, and
2283 any interpretation is supplied by your program itself. The @var{value}
2284 parameter is optional; if it is eliminated, the variable is set to a
2286 @c "any string" here does not include leading, trailing
2287 @c blanks. Gnu asks: does anyone care?
2289 For example, this command:
2296 tells the debugged program, when subsequently run, that its user is named
2297 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2298 are not actually required.)
2300 @kindex unset environment
2301 @item unset environment @var{varname}
2302 Remove variable @var{varname} from the environment to be passed to your
2303 program. This is different from @samp{set env @var{varname} =};
2304 @code{unset environment} removes the variable from the environment,
2305 rather than assigning it an empty value.
2308 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2309 the shell indicated by your @code{SHELL} environment variable if it
2310 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2311 names a shell that runs an initialization file when started
2312 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2313 for the Z shell, or the file specified in the @samp{BASH_ENV}
2314 environment variable for BASH---any variables you set in that file
2315 affect your program. You may wish to move setting of environment
2316 variables to files that are only run when you sign on, such as
2317 @file{.login} or @file{.profile}.
2319 @node Working Directory
2320 @section Your Program's Working Directory
2322 @cindex working directory (of your program)
2323 Each time you start your program with @code{run}, it inherits its
2324 working directory from the current working directory of @value{GDBN}.
2325 The @value{GDBN} working directory is initially whatever it inherited
2326 from its parent process (typically the shell), but you can specify a new
2327 working directory in @value{GDBN} with the @code{cd} command.
2329 The @value{GDBN} working directory also serves as a default for the commands
2330 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2335 @cindex change working directory
2336 @item cd @r{[}@var{directory}@r{]}
2337 Set the @value{GDBN} working directory to @var{directory}. If not
2338 given, @var{directory} uses @file{'~'}.
2342 Print the @value{GDBN} working directory.
2345 It is generally impossible to find the current working directory of
2346 the process being debugged (since a program can change its directory
2347 during its run). If you work on a system where @value{GDBN} is
2348 configured with the @file{/proc} support, you can use the @code{info
2349 proc} command (@pxref{SVR4 Process Information}) to find out the
2350 current working directory of the debuggee.
2353 @section Your Program's Input and Output
2358 By default, the program you run under @value{GDBN} does input and output to
2359 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2360 to its own terminal modes to interact with you, but it records the terminal
2361 modes your program was using and switches back to them when you continue
2362 running your program.
2365 @kindex info terminal
2367 Displays information recorded by @value{GDBN} about the terminal modes your
2371 You can redirect your program's input and/or output using shell
2372 redirection with the @code{run} command. For example,
2379 starts your program, diverting its output to the file @file{outfile}.
2382 @cindex controlling terminal
2383 Another way to specify where your program should do input and output is
2384 with the @code{tty} command. This command accepts a file name as
2385 argument, and causes this file to be the default for future @code{run}
2386 commands. It also resets the controlling terminal for the child
2387 process, for future @code{run} commands. For example,
2394 directs that processes started with subsequent @code{run} commands
2395 default to do input and output on the terminal @file{/dev/ttyb} and have
2396 that as their controlling terminal.
2398 An explicit redirection in @code{run} overrides the @code{tty} command's
2399 effect on the input/output device, but not its effect on the controlling
2402 When you use the @code{tty} command or redirect input in the @code{run}
2403 command, only the input @emph{for your program} is affected. The input
2404 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2405 for @code{set inferior-tty}.
2407 @cindex inferior tty
2408 @cindex set inferior controlling terminal
2409 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2410 display the name of the terminal that will be used for future runs of your
2414 @item set inferior-tty /dev/ttyb
2415 @kindex set inferior-tty
2416 Set the tty for the program being debugged to /dev/ttyb.
2418 @item show inferior-tty
2419 @kindex show inferior-tty
2420 Show the current tty for the program being debugged.
2424 @section Debugging an Already-running Process
2429 @item attach @var{process-id}
2430 This command attaches to a running process---one that was started
2431 outside @value{GDBN}. (@code{info files} shows your active
2432 targets.) The command takes as argument a process ID. The usual way to
2433 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2434 or with the @samp{jobs -l} shell command.
2436 @code{attach} does not repeat if you press @key{RET} a second time after
2437 executing the command.
2440 To use @code{attach}, your program must be running in an environment
2441 which supports processes; for example, @code{attach} does not work for
2442 programs on bare-board targets that lack an operating system. You must
2443 also have permission to send the process a signal.
2445 When you use @code{attach}, the debugger finds the program running in
2446 the process first by looking in the current working directory, then (if
2447 the program is not found) by using the source file search path
2448 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2449 the @code{file} command to load the program. @xref{Files, ,Commands to
2452 The first thing @value{GDBN} does after arranging to debug the specified
2453 process is to stop it. You can examine and modify an attached process
2454 with all the @value{GDBN} commands that are ordinarily available when
2455 you start processes with @code{run}. You can insert breakpoints; you
2456 can step and continue; you can modify storage. If you would rather the
2457 process continue running, you may use the @code{continue} command after
2458 attaching @value{GDBN} to the process.
2463 When you have finished debugging the attached process, you can use the
2464 @code{detach} command to release it from @value{GDBN} control. Detaching
2465 the process continues its execution. After the @code{detach} command,
2466 that process and @value{GDBN} become completely independent once more, and you
2467 are ready to @code{attach} another process or start one with @code{run}.
2468 @code{detach} does not repeat if you press @key{RET} again after
2469 executing the command.
2472 If you exit @value{GDBN} while you have an attached process, you detach
2473 that process. If you use the @code{run} command, you kill that process.
2474 By default, @value{GDBN} asks for confirmation if you try to do either of these
2475 things; you can control whether or not you need to confirm by using the
2476 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2480 @section Killing the Child Process
2485 Kill the child process in which your program is running under @value{GDBN}.
2488 This command is useful if you wish to debug a core dump instead of a
2489 running process. @value{GDBN} ignores any core dump file while your program
2492 On some operating systems, a program cannot be executed outside @value{GDBN}
2493 while you have breakpoints set on it inside @value{GDBN}. You can use the
2494 @code{kill} command in this situation to permit running your program
2495 outside the debugger.
2497 The @code{kill} command is also useful if you wish to recompile and
2498 relink your program, since on many systems it is impossible to modify an
2499 executable file while it is running in a process. In this case, when you
2500 next type @code{run}, @value{GDBN} notices that the file has changed, and
2501 reads the symbol table again (while trying to preserve your current
2502 breakpoint settings).
2504 @node Inferiors and Programs
2505 @section Debugging Multiple Inferiors and Programs
2507 @value{GDBN} lets you run and debug multiple programs in a single
2508 session. In addition, @value{GDBN} on some systems may let you run
2509 several programs simultaneously (otherwise you have to exit from one
2510 before starting another). In the most general case, you can have
2511 multiple threads of execution in each of multiple processes, launched
2512 from multiple executables.
2515 @value{GDBN} represents the state of each program execution with an
2516 object called an @dfn{inferior}. An inferior typically corresponds to
2517 a process, but is more general and applies also to targets that do not
2518 have processes. Inferiors may be created before a process runs, and
2519 may be retained after a process exits. Inferiors have unique
2520 identifiers that are different from process ids. Usually each
2521 inferior will also have its own distinct address space, although some
2522 embedded targets may have several inferiors running in different parts
2523 of a single address space. Each inferior may in turn have multiple
2524 threads running in it.
2526 To find out what inferiors exist at any moment, use @w{@code{info
2530 @kindex info inferiors
2531 @item info inferiors
2532 Print a list of all inferiors currently being managed by @value{GDBN}.
2534 @value{GDBN} displays for each inferior (in this order):
2538 the inferior number assigned by @value{GDBN}
2541 the target system's inferior identifier
2544 the name of the executable the inferior is running.
2549 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2550 indicates the current inferior.
2554 @c end table here to get a little more width for example
2557 (@value{GDBP}) info inferiors
2558 Num Description Executable
2559 2 process 2307 hello
2560 * 1 process 3401 goodbye
2563 To switch focus between inferiors, use the @code{inferior} command:
2566 @kindex inferior @var{infno}
2567 @item inferior @var{infno}
2568 Make inferior number @var{infno} the current inferior. The argument
2569 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2570 in the first field of the @samp{info inferiors} display.
2574 You can get multiple executables into a debugging session via the
2575 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2576 systems @value{GDBN} can add inferiors to the debug session
2577 automatically by following calls to @code{fork} and @code{exec}. To
2578 remove inferiors from the debugging session use the
2579 @w{@code{remove-inferiors}} command.
2582 @kindex add-inferior
2583 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2584 Adds @var{n} inferiors to be run using @var{executable} as the
2585 executable. @var{n} defaults to 1. If no executable is specified,
2586 the inferiors begins empty, with no program. You can still assign or
2587 change the program assigned to the inferior at any time by using the
2588 @code{file} command with the executable name as its argument.
2590 @kindex clone-inferior
2591 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2592 Adds @var{n} inferiors ready to execute the same program as inferior
2593 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2594 number of the current inferior. This is a convenient command when you
2595 want to run another instance of the inferior you are debugging.
2598 (@value{GDBP}) info inferiors
2599 Num Description Executable
2600 * 1 process 29964 helloworld
2601 (@value{GDBP}) clone-inferior
2604 (@value{GDBP}) info inferiors
2605 Num Description Executable
2607 * 1 process 29964 helloworld
2610 You can now simply switch focus to inferior 2 and run it.
2612 @kindex remove-inferiors
2613 @item remove-inferiors @var{infno}@dots{}
2614 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2615 possible to remove an inferior that is running with this command. For
2616 those, use the @code{kill} or @code{detach} command first.
2620 To quit debugging one of the running inferiors that is not the current
2621 inferior, you can either detach from it by using the @w{@code{detach
2622 inferior}} command (allowing it to run independently), or kill it
2623 using the @w{@code{kill inferiors}} command:
2626 @kindex detach inferiors @var{infno}@dots{}
2627 @item detach inferior @var{infno}@dots{}
2628 Detach from the inferior or inferiors identified by @value{GDBN}
2629 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2630 still stays on the list of inferiors shown by @code{info inferiors},
2631 but its Description will show @samp{<null>}.
2633 @kindex kill inferiors @var{infno}@dots{}
2634 @item kill inferiors @var{infno}@dots{}
2635 Kill the inferior or inferiors identified by @value{GDBN} inferior
2636 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2637 stays on the list of inferiors shown by @code{info inferiors}, but its
2638 Description will show @samp{<null>}.
2641 After the successful completion of a command such as @code{detach},
2642 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2643 a normal process exit, the inferior is still valid and listed with
2644 @code{info inferiors}, ready to be restarted.
2647 To be notified when inferiors are started or exit under @value{GDBN}'s
2648 control use @w{@code{set print inferior-events}}:
2651 @kindex set print inferior-events
2652 @cindex print messages on inferior start and exit
2653 @item set print inferior-events
2654 @itemx set print inferior-events on
2655 @itemx set print inferior-events off
2656 The @code{set print inferior-events} command allows you to enable or
2657 disable printing of messages when @value{GDBN} notices that new
2658 inferiors have started or that inferiors have exited or have been
2659 detached. By default, these messages will not be printed.
2661 @kindex show print inferior-events
2662 @item show print inferior-events
2663 Show whether messages will be printed when @value{GDBN} detects that
2664 inferiors have started, exited or have been detached.
2667 Many commands will work the same with multiple programs as with a
2668 single program: e.g., @code{print myglobal} will simply display the
2669 value of @code{myglobal} in the current inferior.
2672 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2673 get more info about the relationship of inferiors, programs, address
2674 spaces in a debug session. You can do that with the @w{@code{maint
2675 info program-spaces}} command.
2678 @kindex maint info program-spaces
2679 @item maint info program-spaces
2680 Print a list of all program spaces currently being managed by
2683 @value{GDBN} displays for each program space (in this order):
2687 the program space number assigned by @value{GDBN}
2690 the name of the executable loaded into the program space, with e.g.,
2691 the @code{file} command.
2696 An asterisk @samp{*} preceding the @value{GDBN} program space number
2697 indicates the current program space.
2699 In addition, below each program space line, @value{GDBN} prints extra
2700 information that isn't suitable to display in tabular form. For
2701 example, the list of inferiors bound to the program space.
2704 (@value{GDBP}) maint info program-spaces
2707 Bound inferiors: ID 1 (process 21561)
2711 Here we can see that no inferior is running the program @code{hello},
2712 while @code{process 21561} is running the program @code{goodbye}. On
2713 some targets, it is possible that multiple inferiors are bound to the
2714 same program space. The most common example is that of debugging both
2715 the parent and child processes of a @code{vfork} call. For example,
2718 (@value{GDBP}) maint info program-spaces
2721 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2724 Here, both inferior 2 and inferior 1 are running in the same program
2725 space as a result of inferior 1 having executed a @code{vfork} call.
2729 @section Debugging Programs with Multiple Threads
2731 @cindex threads of execution
2732 @cindex multiple threads
2733 @cindex switching threads
2734 In some operating systems, such as HP-UX and Solaris, a single program
2735 may have more than one @dfn{thread} of execution. The precise semantics
2736 of threads differ from one operating system to another, but in general
2737 the threads of a single program are akin to multiple processes---except
2738 that they share one address space (that is, they can all examine and
2739 modify the same variables). On the other hand, each thread has its own
2740 registers and execution stack, and perhaps private memory.
2742 @value{GDBN} provides these facilities for debugging multi-thread
2746 @item automatic notification of new threads
2747 @item @samp{thread @var{threadno}}, a command to switch among threads
2748 @item @samp{info threads}, a command to inquire about existing threads
2749 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2750 a command to apply a command to a list of threads
2751 @item thread-specific breakpoints
2752 @item @samp{set print thread-events}, which controls printing of
2753 messages on thread start and exit.
2754 @item @samp{set libthread-db-search-path @var{path}}, which lets
2755 the user specify which @code{libthread_db} to use if the default choice
2756 isn't compatible with the program.
2760 @emph{Warning:} These facilities are not yet available on every
2761 @value{GDBN} configuration where the operating system supports threads.
2762 If your @value{GDBN} does not support threads, these commands have no
2763 effect. For example, a system without thread support shows no output
2764 from @samp{info threads}, and always rejects the @code{thread} command,
2768 (@value{GDBP}) info threads
2769 (@value{GDBP}) thread 1
2770 Thread ID 1 not known. Use the "info threads" command to
2771 see the IDs of currently known threads.
2773 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2774 @c doesn't support threads"?
2777 @cindex focus of debugging
2778 @cindex current thread
2779 The @value{GDBN} thread debugging facility allows you to observe all
2780 threads while your program runs---but whenever @value{GDBN} takes
2781 control, one thread in particular is always the focus of debugging.
2782 This thread is called the @dfn{current thread}. Debugging commands show
2783 program information from the perspective of the current thread.
2785 @cindex @code{New} @var{systag} message
2786 @cindex thread identifier (system)
2787 @c FIXME-implementors!! It would be more helpful if the [New...] message
2788 @c included GDB's numeric thread handle, so you could just go to that
2789 @c thread without first checking `info threads'.
2790 Whenever @value{GDBN} detects a new thread in your program, it displays
2791 the target system's identification for the thread with a message in the
2792 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2793 whose form varies depending on the particular system. For example, on
2794 @sc{gnu}/Linux, you might see
2797 [New Thread 0x41e02940 (LWP 25582)]
2801 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2802 the @var{systag} is simply something like @samp{process 368}, with no
2805 @c FIXME!! (1) Does the [New...] message appear even for the very first
2806 @c thread of a program, or does it only appear for the
2807 @c second---i.e.@: when it becomes obvious we have a multithread
2809 @c (2) *Is* there necessarily a first thread always? Or do some
2810 @c multithread systems permit starting a program with multiple
2811 @c threads ab initio?
2813 @cindex thread number
2814 @cindex thread identifier (GDB)
2815 For debugging purposes, @value{GDBN} associates its own thread
2816 number---always a single integer---with each thread in your program.
2819 @kindex info threads
2820 @item info threads @r{[}@var{id}@dots{}@r{]}
2821 Display a summary of all threads currently in your program. Optional
2822 argument @var{id}@dots{} is one or more thread ids separated by spaces, and
2823 means to print information only about the specified thread or threads.
2824 @value{GDBN} displays for each thread (in this order):
2828 the thread number assigned by @value{GDBN}
2831 the target system's thread identifier (@var{systag})
2834 the thread's name, if one is known. A thread can either be named by
2835 the user (see @code{thread name}, below), or, in some cases, by the
2839 the current stack frame summary for that thread
2843 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2844 indicates the current thread.
2848 @c end table here to get a little more width for example
2851 (@value{GDBP}) info threads
2853 3 process 35 thread 27 0x34e5 in sigpause ()
2854 2 process 35 thread 23 0x34e5 in sigpause ()
2855 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2859 On Solaris, you can display more information about user threads with a
2860 Solaris-specific command:
2863 @item maint info sol-threads
2864 @kindex maint info sol-threads
2865 @cindex thread info (Solaris)
2866 Display info on Solaris user threads.
2870 @kindex thread @var{threadno}
2871 @item thread @var{threadno}
2872 Make thread number @var{threadno} the current thread. The command
2873 argument @var{threadno} is the internal @value{GDBN} thread number, as
2874 shown in the first field of the @samp{info threads} display.
2875 @value{GDBN} responds by displaying the system identifier of the thread
2876 you selected, and its current stack frame summary:
2879 (@value{GDBP}) thread 2
2880 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
2881 #0 some_function (ignore=0x0) at example.c:8
2882 8 printf ("hello\n");
2886 As with the @samp{[New @dots{}]} message, the form of the text after
2887 @samp{Switching to} depends on your system's conventions for identifying
2890 @vindex $_thread@r{, convenience variable}
2891 The debugger convenience variable @samp{$_thread} contains the number
2892 of the current thread. You may find this useful in writing breakpoint
2893 conditional expressions, command scripts, and so forth. See
2894 @xref{Convenience Vars,, Convenience Variables}, for general
2895 information on convenience variables.
2897 @kindex thread apply
2898 @cindex apply command to several threads
2899 @item thread apply [@var{threadno} | all] @var{command}
2900 The @code{thread apply} command allows you to apply the named
2901 @var{command} to one or more threads. Specify the numbers of the
2902 threads that you want affected with the command argument
2903 @var{threadno}. It can be a single thread number, one of the numbers
2904 shown in the first field of the @samp{info threads} display; or it
2905 could be a range of thread numbers, as in @code{2-4}. To apply a
2906 command to all threads, type @kbd{thread apply all @var{command}}.
2909 @cindex name a thread
2910 @item thread name [@var{name}]
2911 This command assigns a name to the current thread. If no argument is
2912 given, any existing user-specified name is removed. The thread name
2913 appears in the @samp{info threads} display.
2915 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
2916 determine the name of the thread as given by the OS. On these
2917 systems, a name specified with @samp{thread name} will override the
2918 system-give name, and removing the user-specified name will cause
2919 @value{GDBN} to once again display the system-specified name.
2922 @cindex search for a thread
2923 @item thread find [@var{regexp}]
2924 Search for and display thread ids whose name or @var{systag}
2925 matches the supplied regular expression.
2927 As well as being the complement to the @samp{thread name} command,
2928 this command also allows you to identify a thread by its target
2929 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
2933 (@value{GDBN}) thread find 26688
2934 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
2935 (@value{GDBN}) info thread 4
2937 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
2940 @kindex set print thread-events
2941 @cindex print messages on thread start and exit
2942 @item set print thread-events
2943 @itemx set print thread-events on
2944 @itemx set print thread-events off
2945 The @code{set print thread-events} command allows you to enable or
2946 disable printing of messages when @value{GDBN} notices that new threads have
2947 started or that threads have exited. By default, these messages will
2948 be printed if detection of these events is supported by the target.
2949 Note that these messages cannot be disabled on all targets.
2951 @kindex show print thread-events
2952 @item show print thread-events
2953 Show whether messages will be printed when @value{GDBN} detects that threads
2954 have started and exited.
2957 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2958 more information about how @value{GDBN} behaves when you stop and start
2959 programs with multiple threads.
2961 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2962 watchpoints in programs with multiple threads.
2964 @anchor{set libthread-db-search-path}
2966 @kindex set libthread-db-search-path
2967 @cindex search path for @code{libthread_db}
2968 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2969 If this variable is set, @var{path} is a colon-separated list of
2970 directories @value{GDBN} will use to search for @code{libthread_db}.
2971 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2972 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
2973 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
2976 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2977 @code{libthread_db} library to obtain information about threads in the
2978 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2979 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
2980 specific thread debugging library loading is enabled
2981 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
2983 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
2984 refers to the default system directories that are
2985 normally searched for loading shared libraries. The @samp{$sdir} entry
2986 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
2987 (@pxref{libthread_db.so.1 file}).
2989 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
2990 refers to the directory from which @code{libpthread}
2991 was loaded in the inferior process.
2993 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2994 @value{GDBN} attempts to initialize it with the current inferior process.
2995 If this initialization fails (which could happen because of a version
2996 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2997 will unload @code{libthread_db}, and continue with the next directory.
2998 If none of @code{libthread_db} libraries initialize successfully,
2999 @value{GDBN} will issue a warning and thread debugging will be disabled.
3001 Setting @code{libthread-db-search-path} is currently implemented
3002 only on some platforms.
3004 @kindex show libthread-db-search-path
3005 @item show libthread-db-search-path
3006 Display current libthread_db search path.
3008 @kindex set debug libthread-db
3009 @kindex show debug libthread-db
3010 @cindex debugging @code{libthread_db}
3011 @item set debug libthread-db
3012 @itemx show debug libthread-db
3013 Turns on or off display of @code{libthread_db}-related events.
3014 Use @code{1} to enable, @code{0} to disable.
3018 @section Debugging Forks
3020 @cindex fork, debugging programs which call
3021 @cindex multiple processes
3022 @cindex processes, multiple
3023 On most systems, @value{GDBN} has no special support for debugging
3024 programs which create additional processes using the @code{fork}
3025 function. When a program forks, @value{GDBN} will continue to debug the
3026 parent process and the child process will run unimpeded. If you have
3027 set a breakpoint in any code which the child then executes, the child
3028 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3029 will cause it to terminate.
3031 However, if you want to debug the child process there is a workaround
3032 which isn't too painful. Put a call to @code{sleep} in the code which
3033 the child process executes after the fork. It may be useful to sleep
3034 only if a certain environment variable is set, or a certain file exists,
3035 so that the delay need not occur when you don't want to run @value{GDBN}
3036 on the child. While the child is sleeping, use the @code{ps} program to
3037 get its process ID. Then tell @value{GDBN} (a new invocation of
3038 @value{GDBN} if you are also debugging the parent process) to attach to
3039 the child process (@pxref{Attach}). From that point on you can debug
3040 the child process just like any other process which you attached to.
3042 On some systems, @value{GDBN} provides support for debugging programs that
3043 create additional processes using the @code{fork} or @code{vfork} functions.
3044 Currently, the only platforms with this feature are HP-UX (11.x and later
3045 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
3047 By default, when a program forks, @value{GDBN} will continue to debug
3048 the parent process and the child process will run unimpeded.
3050 If you want to follow the child process instead of the parent process,
3051 use the command @w{@code{set follow-fork-mode}}.
3054 @kindex set follow-fork-mode
3055 @item set follow-fork-mode @var{mode}
3056 Set the debugger response to a program call of @code{fork} or
3057 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3058 process. The @var{mode} argument can be:
3062 The original process is debugged after a fork. The child process runs
3063 unimpeded. This is the default.
3066 The new process is debugged after a fork. The parent process runs
3071 @kindex show follow-fork-mode
3072 @item show follow-fork-mode
3073 Display the current debugger response to a @code{fork} or @code{vfork} call.
3076 @cindex debugging multiple processes
3077 On Linux, if you want to debug both the parent and child processes, use the
3078 command @w{@code{set detach-on-fork}}.
3081 @kindex set detach-on-fork
3082 @item set detach-on-fork @var{mode}
3083 Tells gdb whether to detach one of the processes after a fork, or
3084 retain debugger control over them both.
3088 The child process (or parent process, depending on the value of
3089 @code{follow-fork-mode}) will be detached and allowed to run
3090 independently. This is the default.
3093 Both processes will be held under the control of @value{GDBN}.
3094 One process (child or parent, depending on the value of
3095 @code{follow-fork-mode}) is debugged as usual, while the other
3100 @kindex show detach-on-fork
3101 @item show detach-on-fork
3102 Show whether detach-on-fork mode is on/off.
3105 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3106 will retain control of all forked processes (including nested forks).
3107 You can list the forked processes under the control of @value{GDBN} by
3108 using the @w{@code{info inferiors}} command, and switch from one fork
3109 to another by using the @code{inferior} command (@pxref{Inferiors and
3110 Programs, ,Debugging Multiple Inferiors and Programs}).
3112 To quit debugging one of the forked processes, you can either detach
3113 from it by using the @w{@code{detach inferiors}} command (allowing it
3114 to run independently), or kill it using the @w{@code{kill inferiors}}
3115 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3118 If you ask to debug a child process and a @code{vfork} is followed by an
3119 @code{exec}, @value{GDBN} executes the new target up to the first
3120 breakpoint in the new target. If you have a breakpoint set on
3121 @code{main} in your original program, the breakpoint will also be set on
3122 the child process's @code{main}.
3124 On some systems, when a child process is spawned by @code{vfork}, you
3125 cannot debug the child or parent until an @code{exec} call completes.
3127 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3128 call executes, the new target restarts. To restart the parent
3129 process, use the @code{file} command with the parent executable name
3130 as its argument. By default, after an @code{exec} call executes,
3131 @value{GDBN} discards the symbols of the previous executable image.
3132 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3136 @kindex set follow-exec-mode
3137 @item set follow-exec-mode @var{mode}
3139 Set debugger response to a program call of @code{exec}. An
3140 @code{exec} call replaces the program image of a process.
3142 @code{follow-exec-mode} can be:
3146 @value{GDBN} creates a new inferior and rebinds the process to this
3147 new inferior. The program the process was running before the
3148 @code{exec} call can be restarted afterwards by restarting the
3154 (@value{GDBP}) info inferiors
3156 Id Description Executable
3159 process 12020 is executing new program: prog2
3160 Program exited normally.
3161 (@value{GDBP}) info inferiors
3162 Id Description Executable
3168 @value{GDBN} keeps the process bound to the same inferior. The new
3169 executable image replaces the previous executable loaded in the
3170 inferior. Restarting the inferior after the @code{exec} call, with
3171 e.g., the @code{run} command, restarts the executable the process was
3172 running after the @code{exec} call. This is the default mode.
3177 (@value{GDBP}) info inferiors
3178 Id Description Executable
3181 process 12020 is executing new program: prog2
3182 Program exited normally.
3183 (@value{GDBP}) info inferiors
3184 Id Description Executable
3191 You can use the @code{catch} command to make @value{GDBN} stop whenever
3192 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3193 Catchpoints, ,Setting Catchpoints}.
3195 @node Checkpoint/Restart
3196 @section Setting a @emph{Bookmark} to Return to Later
3201 @cindex snapshot of a process
3202 @cindex rewind program state
3204 On certain operating systems@footnote{Currently, only
3205 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3206 program's state, called a @dfn{checkpoint}, and come back to it
3209 Returning to a checkpoint effectively undoes everything that has
3210 happened in the program since the @code{checkpoint} was saved. This
3211 includes changes in memory, registers, and even (within some limits)
3212 system state. Effectively, it is like going back in time to the
3213 moment when the checkpoint was saved.
3215 Thus, if you're stepping thru a program and you think you're
3216 getting close to the point where things go wrong, you can save
3217 a checkpoint. Then, if you accidentally go too far and miss
3218 the critical statement, instead of having to restart your program
3219 from the beginning, you can just go back to the checkpoint and
3220 start again from there.
3222 This can be especially useful if it takes a lot of time or
3223 steps to reach the point where you think the bug occurs.
3225 To use the @code{checkpoint}/@code{restart} method of debugging:
3230 Save a snapshot of the debugged program's current execution state.
3231 The @code{checkpoint} command takes no arguments, but each checkpoint
3232 is assigned a small integer id, similar to a breakpoint id.
3234 @kindex info checkpoints
3235 @item info checkpoints
3236 List the checkpoints that have been saved in the current debugging
3237 session. For each checkpoint, the following information will be
3244 @item Source line, or label
3247 @kindex restart @var{checkpoint-id}
3248 @item restart @var{checkpoint-id}
3249 Restore the program state that was saved as checkpoint number
3250 @var{checkpoint-id}. All program variables, registers, stack frames
3251 etc.@: will be returned to the values that they had when the checkpoint
3252 was saved. In essence, gdb will ``wind back the clock'' to the point
3253 in time when the checkpoint was saved.
3255 Note that breakpoints, @value{GDBN} variables, command history etc.
3256 are not affected by restoring a checkpoint. In general, a checkpoint
3257 only restores things that reside in the program being debugged, not in
3260 @kindex delete checkpoint @var{checkpoint-id}
3261 @item delete checkpoint @var{checkpoint-id}
3262 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3266 Returning to a previously saved checkpoint will restore the user state
3267 of the program being debugged, plus a significant subset of the system
3268 (OS) state, including file pointers. It won't ``un-write'' data from
3269 a file, but it will rewind the file pointer to the previous location,
3270 so that the previously written data can be overwritten. For files
3271 opened in read mode, the pointer will also be restored so that the
3272 previously read data can be read again.
3274 Of course, characters that have been sent to a printer (or other
3275 external device) cannot be ``snatched back'', and characters received
3276 from eg.@: a serial device can be removed from internal program buffers,
3277 but they cannot be ``pushed back'' into the serial pipeline, ready to
3278 be received again. Similarly, the actual contents of files that have
3279 been changed cannot be restored (at this time).
3281 However, within those constraints, you actually can ``rewind'' your
3282 program to a previously saved point in time, and begin debugging it
3283 again --- and you can change the course of events so as to debug a
3284 different execution path this time.
3286 @cindex checkpoints and process id
3287 Finally, there is one bit of internal program state that will be
3288 different when you return to a checkpoint --- the program's process
3289 id. Each checkpoint will have a unique process id (or @var{pid}),
3290 and each will be different from the program's original @var{pid}.
3291 If your program has saved a local copy of its process id, this could
3292 potentially pose a problem.
3294 @subsection A Non-obvious Benefit of Using Checkpoints
3296 On some systems such as @sc{gnu}/Linux, address space randomization
3297 is performed on new processes for security reasons. This makes it
3298 difficult or impossible to set a breakpoint, or watchpoint, on an
3299 absolute address if you have to restart the program, since the
3300 absolute location of a symbol will change from one execution to the
3303 A checkpoint, however, is an @emph{identical} copy of a process.
3304 Therefore if you create a checkpoint at (eg.@:) the start of main,
3305 and simply return to that checkpoint instead of restarting the
3306 process, you can avoid the effects of address randomization and
3307 your symbols will all stay in the same place.
3310 @chapter Stopping and Continuing
3312 The principal purposes of using a debugger are so that you can stop your
3313 program before it terminates; or so that, if your program runs into
3314 trouble, you can investigate and find out why.
3316 Inside @value{GDBN}, your program may stop for any of several reasons,
3317 such as a signal, a breakpoint, or reaching a new line after a
3318 @value{GDBN} command such as @code{step}. You may then examine and
3319 change variables, set new breakpoints or remove old ones, and then
3320 continue execution. Usually, the messages shown by @value{GDBN} provide
3321 ample explanation of the status of your program---but you can also
3322 explicitly request this information at any time.
3325 @kindex info program
3327 Display information about the status of your program: whether it is
3328 running or not, what process it is, and why it stopped.
3332 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3333 * Continuing and Stepping:: Resuming execution
3334 * Skipping Over Functions and Files::
3335 Skipping over functions and files
3337 * Thread Stops:: Stopping and starting multi-thread programs
3341 @section Breakpoints, Watchpoints, and Catchpoints
3344 A @dfn{breakpoint} makes your program stop whenever a certain point in
3345 the program is reached. For each breakpoint, you can add conditions to
3346 control in finer detail whether your program stops. You can set
3347 breakpoints with the @code{break} command and its variants (@pxref{Set
3348 Breaks, ,Setting Breakpoints}), to specify the place where your program
3349 should stop by line number, function name or exact address in the
3352 On some systems, you can set breakpoints in shared libraries before
3353 the executable is run. There is a minor limitation on HP-UX systems:
3354 you must wait until the executable is run in order to set breakpoints
3355 in shared library routines that are not called directly by the program
3356 (for example, routines that are arguments in a @code{pthread_create}
3360 @cindex data breakpoints
3361 @cindex memory tracing
3362 @cindex breakpoint on memory address
3363 @cindex breakpoint on variable modification
3364 A @dfn{watchpoint} is a special breakpoint that stops your program
3365 when the value of an expression changes. The expression may be a value
3366 of a variable, or it could involve values of one or more variables
3367 combined by operators, such as @samp{a + b}. This is sometimes called
3368 @dfn{data breakpoints}. You must use a different command to set
3369 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3370 from that, you can manage a watchpoint like any other breakpoint: you
3371 enable, disable, and delete both breakpoints and watchpoints using the
3374 You can arrange to have values from your program displayed automatically
3375 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3379 @cindex breakpoint on events
3380 A @dfn{catchpoint} is another special breakpoint that stops your program
3381 when a certain kind of event occurs, such as the throwing of a C@t{++}
3382 exception or the loading of a library. As with watchpoints, you use a
3383 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3384 Catchpoints}), but aside from that, you can manage a catchpoint like any
3385 other breakpoint. (To stop when your program receives a signal, use the
3386 @code{handle} command; see @ref{Signals, ,Signals}.)
3388 @cindex breakpoint numbers
3389 @cindex numbers for breakpoints
3390 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3391 catchpoint when you create it; these numbers are successive integers
3392 starting with one. In many of the commands for controlling various
3393 features of breakpoints you use the breakpoint number to say which
3394 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3395 @dfn{disabled}; if disabled, it has no effect on your program until you
3398 @cindex breakpoint ranges
3399 @cindex ranges of breakpoints
3400 Some @value{GDBN} commands accept a range of breakpoints on which to
3401 operate. A breakpoint range is either a single breakpoint number, like
3402 @samp{5}, or two such numbers, in increasing order, separated by a
3403 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3404 all breakpoints in that range are operated on.
3407 * Set Breaks:: Setting breakpoints
3408 * Set Watchpoints:: Setting watchpoints
3409 * Set Catchpoints:: Setting catchpoints
3410 * Delete Breaks:: Deleting breakpoints
3411 * Disabling:: Disabling breakpoints
3412 * Conditions:: Break conditions
3413 * Break Commands:: Breakpoint command lists
3414 * Dynamic Printf:: Dynamic printf
3415 * Save Breakpoints:: How to save breakpoints in a file
3416 * Static Probe Points:: Listing static probe points
3417 * Error in Breakpoints:: ``Cannot insert breakpoints''
3418 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3422 @subsection Setting Breakpoints
3424 @c FIXME LMB what does GDB do if no code on line of breakpt?
3425 @c consider in particular declaration with/without initialization.
3427 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3430 @kindex b @r{(@code{break})}
3431 @vindex $bpnum@r{, convenience variable}
3432 @cindex latest breakpoint
3433 Breakpoints are set with the @code{break} command (abbreviated
3434 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3435 number of the breakpoint you've set most recently; see @ref{Convenience
3436 Vars,, Convenience Variables}, for a discussion of what you can do with
3437 convenience variables.
3440 @item break @var{location}
3441 Set a breakpoint at the given @var{location}, which can specify a
3442 function name, a line number, or an address of an instruction.
3443 (@xref{Specify Location}, for a list of all the possible ways to
3444 specify a @var{location}.) The breakpoint will stop your program just
3445 before it executes any of the code in the specified @var{location}.
3447 When using source languages that permit overloading of symbols, such as
3448 C@t{++}, a function name may refer to more than one possible place to break.
3449 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3452 It is also possible to insert a breakpoint that will stop the program
3453 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3454 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3457 When called without any arguments, @code{break} sets a breakpoint at
3458 the next instruction to be executed in the selected stack frame
3459 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3460 innermost, this makes your program stop as soon as control
3461 returns to that frame. This is similar to the effect of a
3462 @code{finish} command in the frame inside the selected frame---except
3463 that @code{finish} does not leave an active breakpoint. If you use
3464 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3465 the next time it reaches the current location; this may be useful
3468 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3469 least one instruction has been executed. If it did not do this, you
3470 would be unable to proceed past a breakpoint without first disabling the
3471 breakpoint. This rule applies whether or not the breakpoint already
3472 existed when your program stopped.
3474 @item break @dots{} if @var{cond}
3475 Set a breakpoint with condition @var{cond}; evaluate the expression
3476 @var{cond} each time the breakpoint is reached, and stop only if the
3477 value is nonzero---that is, if @var{cond} evaluates as true.
3478 @samp{@dots{}} stands for one of the possible arguments described
3479 above (or no argument) specifying where to break. @xref{Conditions,
3480 ,Break Conditions}, for more information on breakpoint conditions.
3483 @item tbreak @var{args}
3484 Set a breakpoint enabled only for one stop. @var{args} are the
3485 same as for the @code{break} command, and the breakpoint is set in the same
3486 way, but the breakpoint is automatically deleted after the first time your
3487 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3490 @cindex hardware breakpoints
3491 @item hbreak @var{args}
3492 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3493 @code{break} command and the breakpoint is set in the same way, but the
3494 breakpoint requires hardware support and some target hardware may not
3495 have this support. The main purpose of this is EPROM/ROM code
3496 debugging, so you can set a breakpoint at an instruction without
3497 changing the instruction. This can be used with the new trap-generation
3498 provided by SPARClite DSU and most x86-based targets. These targets
3499 will generate traps when a program accesses some data or instruction
3500 address that is assigned to the debug registers. However the hardware
3501 breakpoint registers can take a limited number of breakpoints. For
3502 example, on the DSU, only two data breakpoints can be set at a time, and
3503 @value{GDBN} will reject this command if more than two are used. Delete
3504 or disable unused hardware breakpoints before setting new ones
3505 (@pxref{Disabling, ,Disabling Breakpoints}).
3506 @xref{Conditions, ,Break Conditions}.
3507 For remote targets, you can restrict the number of hardware
3508 breakpoints @value{GDBN} will use, see @ref{set remote
3509 hardware-breakpoint-limit}.
3512 @item thbreak @var{args}
3513 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3514 are the same as for the @code{hbreak} command and the breakpoint is set in
3515 the same way. However, like the @code{tbreak} command,
3516 the breakpoint is automatically deleted after the
3517 first time your program stops there. Also, like the @code{hbreak}
3518 command, the breakpoint requires hardware support and some target hardware
3519 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3520 See also @ref{Conditions, ,Break Conditions}.
3523 @cindex regular expression
3524 @cindex breakpoints at functions matching a regexp
3525 @cindex set breakpoints in many functions
3526 @item rbreak @var{regex}
3527 Set breakpoints on all functions matching the regular expression
3528 @var{regex}. This command sets an unconditional breakpoint on all
3529 matches, printing a list of all breakpoints it set. Once these
3530 breakpoints are set, they are treated just like the breakpoints set with
3531 the @code{break} command. You can delete them, disable them, or make
3532 them conditional the same way as any other breakpoint.
3534 The syntax of the regular expression is the standard one used with tools
3535 like @file{grep}. Note that this is different from the syntax used by
3536 shells, so for instance @code{foo*} matches all functions that include
3537 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3538 @code{.*} leading and trailing the regular expression you supply, so to
3539 match only functions that begin with @code{foo}, use @code{^foo}.
3541 @cindex non-member C@t{++} functions, set breakpoint in
3542 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3543 breakpoints on overloaded functions that are not members of any special
3546 @cindex set breakpoints on all functions
3547 The @code{rbreak} command can be used to set breakpoints in
3548 @strong{all} the functions in a program, like this:
3551 (@value{GDBP}) rbreak .
3554 @item rbreak @var{file}:@var{regex}
3555 If @code{rbreak} is called with a filename qualification, it limits
3556 the search for functions matching the given regular expression to the
3557 specified @var{file}. This can be used, for example, to set breakpoints on
3558 every function in a given file:
3561 (@value{GDBP}) rbreak file.c:.
3564 The colon separating the filename qualifier from the regex may
3565 optionally be surrounded by spaces.
3567 @kindex info breakpoints
3568 @cindex @code{$_} and @code{info breakpoints}
3569 @item info breakpoints @r{[}@var{n}@dots{}@r{]}
3570 @itemx info break @r{[}@var{n}@dots{}@r{]}
3571 Print a table of all breakpoints, watchpoints, and catchpoints set and
3572 not deleted. Optional argument @var{n} means print information only
3573 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3574 For each breakpoint, following columns are printed:
3577 @item Breakpoint Numbers
3579 Breakpoint, watchpoint, or catchpoint.
3581 Whether the breakpoint is marked to be disabled or deleted when hit.
3582 @item Enabled or Disabled
3583 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3584 that are not enabled.
3586 Where the breakpoint is in your program, as a memory address. For a
3587 pending breakpoint whose address is not yet known, this field will
3588 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3589 library that has the symbol or line referred by breakpoint is loaded.
3590 See below for details. A breakpoint with several locations will
3591 have @samp{<MULTIPLE>} in this field---see below for details.
3593 Where the breakpoint is in the source for your program, as a file and
3594 line number. For a pending breakpoint, the original string passed to
3595 the breakpoint command will be listed as it cannot be resolved until
3596 the appropriate shared library is loaded in the future.
3600 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
3601 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
3602 @value{GDBN} on the host's side. If it is ``target'', then the condition
3603 is evaluated by the target. The @code{info break} command shows
3604 the condition on the line following the affected breakpoint, together with
3605 its condition evaluation mode in between parentheses.
3607 Breakpoint commands, if any, are listed after that. A pending breakpoint is
3608 allowed to have a condition specified for it. The condition is not parsed for
3609 validity until a shared library is loaded that allows the pending
3610 breakpoint to resolve to a valid location.
3613 @code{info break} with a breakpoint
3614 number @var{n} as argument lists only that breakpoint. The
3615 convenience variable @code{$_} and the default examining-address for
3616 the @code{x} command are set to the address of the last breakpoint
3617 listed (@pxref{Memory, ,Examining Memory}).
3620 @code{info break} displays a count of the number of times the breakpoint
3621 has been hit. This is especially useful in conjunction with the
3622 @code{ignore} command. You can ignore a large number of breakpoint
3623 hits, look at the breakpoint info to see how many times the breakpoint
3624 was hit, and then run again, ignoring one less than that number. This
3625 will get you quickly to the last hit of that breakpoint.
3628 For a breakpoints with an enable count (xref) greater than 1,
3629 @code{info break} also displays that count.
3633 @value{GDBN} allows you to set any number of breakpoints at the same place in
3634 your program. There is nothing silly or meaningless about this. When
3635 the breakpoints are conditional, this is even useful
3636 (@pxref{Conditions, ,Break Conditions}).
3638 @cindex multiple locations, breakpoints
3639 @cindex breakpoints, multiple locations
3640 It is possible that a breakpoint corresponds to several locations
3641 in your program. Examples of this situation are:
3645 Multiple functions in the program may have the same name.
3648 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3649 instances of the function body, used in different cases.
3652 For a C@t{++} template function, a given line in the function can
3653 correspond to any number of instantiations.
3656 For an inlined function, a given source line can correspond to
3657 several places where that function is inlined.
3660 In all those cases, @value{GDBN} will insert a breakpoint at all
3661 the relevant locations.
3663 A breakpoint with multiple locations is displayed in the breakpoint
3664 table using several rows---one header row, followed by one row for
3665 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3666 address column. The rows for individual locations contain the actual
3667 addresses for locations, and show the functions to which those
3668 locations belong. The number column for a location is of the form
3669 @var{breakpoint-number}.@var{location-number}.
3674 Num Type Disp Enb Address What
3675 1 breakpoint keep y <MULTIPLE>
3677 breakpoint already hit 1 time
3678 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3679 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3682 Each location can be individually enabled or disabled by passing
3683 @var{breakpoint-number}.@var{location-number} as argument to the
3684 @code{enable} and @code{disable} commands. Note that you cannot
3685 delete the individual locations from the list, you can only delete the
3686 entire list of locations that belong to their parent breakpoint (with
3687 the @kbd{delete @var{num}} command, where @var{num} is the number of
3688 the parent breakpoint, 1 in the above example). Disabling or enabling
3689 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3690 that belong to that breakpoint.
3692 @cindex pending breakpoints
3693 It's quite common to have a breakpoint inside a shared library.
3694 Shared libraries can be loaded and unloaded explicitly,
3695 and possibly repeatedly, as the program is executed. To support
3696 this use case, @value{GDBN} updates breakpoint locations whenever
3697 any shared library is loaded or unloaded. Typically, you would
3698 set a breakpoint in a shared library at the beginning of your
3699 debugging session, when the library is not loaded, and when the
3700 symbols from the library are not available. When you try to set
3701 breakpoint, @value{GDBN} will ask you if you want to set
3702 a so called @dfn{pending breakpoint}---breakpoint whose address
3703 is not yet resolved.
3705 After the program is run, whenever a new shared library is loaded,
3706 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3707 shared library contains the symbol or line referred to by some
3708 pending breakpoint, that breakpoint is resolved and becomes an
3709 ordinary breakpoint. When a library is unloaded, all breakpoints
3710 that refer to its symbols or source lines become pending again.
3712 This logic works for breakpoints with multiple locations, too. For
3713 example, if you have a breakpoint in a C@t{++} template function, and
3714 a newly loaded shared library has an instantiation of that template,
3715 a new location is added to the list of locations for the breakpoint.
3717 Except for having unresolved address, pending breakpoints do not
3718 differ from regular breakpoints. You can set conditions or commands,
3719 enable and disable them and perform other breakpoint operations.
3721 @value{GDBN} provides some additional commands for controlling what
3722 happens when the @samp{break} command cannot resolve breakpoint
3723 address specification to an address:
3725 @kindex set breakpoint pending
3726 @kindex show breakpoint pending
3728 @item set breakpoint pending auto
3729 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3730 location, it queries you whether a pending breakpoint should be created.
3732 @item set breakpoint pending on
3733 This indicates that an unrecognized breakpoint location should automatically
3734 result in a pending breakpoint being created.
3736 @item set breakpoint pending off
3737 This indicates that pending breakpoints are not to be created. Any
3738 unrecognized breakpoint location results in an error. This setting does
3739 not affect any pending breakpoints previously created.
3741 @item show breakpoint pending
3742 Show the current behavior setting for creating pending breakpoints.
3745 The settings above only affect the @code{break} command and its
3746 variants. Once breakpoint is set, it will be automatically updated
3747 as shared libraries are loaded and unloaded.
3749 @cindex automatic hardware breakpoints
3750 For some targets, @value{GDBN} can automatically decide if hardware or
3751 software breakpoints should be used, depending on whether the
3752 breakpoint address is read-only or read-write. This applies to
3753 breakpoints set with the @code{break} command as well as to internal
3754 breakpoints set by commands like @code{next} and @code{finish}. For
3755 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3758 You can control this automatic behaviour with the following commands::
3760 @kindex set breakpoint auto-hw
3761 @kindex show breakpoint auto-hw
3763 @item set breakpoint auto-hw on
3764 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3765 will try to use the target memory map to decide if software or hardware
3766 breakpoint must be used.
3768 @item set breakpoint auto-hw off
3769 This indicates @value{GDBN} should not automatically select breakpoint
3770 type. If the target provides a memory map, @value{GDBN} will warn when
3771 trying to set software breakpoint at a read-only address.
3774 @value{GDBN} normally implements breakpoints by replacing the program code
3775 at the breakpoint address with a special instruction, which, when
3776 executed, given control to the debugger. By default, the program
3777 code is so modified only when the program is resumed. As soon as
3778 the program stops, @value{GDBN} restores the original instructions. This
3779 behaviour guards against leaving breakpoints inserted in the
3780 target should gdb abrubptly disconnect. However, with slow remote
3781 targets, inserting and removing breakpoint can reduce the performance.
3782 This behavior can be controlled with the following commands::
3784 @kindex set breakpoint always-inserted
3785 @kindex show breakpoint always-inserted
3787 @item set breakpoint always-inserted off
3788 All breakpoints, including newly added by the user, are inserted in
3789 the target only when the target is resumed. All breakpoints are
3790 removed from the target when it stops.
3792 @item set breakpoint always-inserted on
3793 Causes all breakpoints to be inserted in the target at all times. If
3794 the user adds a new breakpoint, or changes an existing breakpoint, the
3795 breakpoints in the target are updated immediately. A breakpoint is
3796 removed from the target only when breakpoint itself is removed.
3798 @cindex non-stop mode, and @code{breakpoint always-inserted}
3799 @item set breakpoint always-inserted auto
3800 This is the default mode. If @value{GDBN} is controlling the inferior
3801 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3802 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3803 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3804 @code{breakpoint always-inserted} mode is off.
3807 @value{GDBN} handles conditional breakpoints by evaluating these conditions
3808 when a breakpoint breaks. If the condition is true, then the process being
3809 debugged stops, otherwise the process is resumed.
3811 If the target supports evaluating conditions on its end, @value{GDBN} may
3812 download the breakpoint, together with its conditions, to it.
3814 This feature can be controlled via the following commands:
3816 @kindex set breakpoint condition-evaluation
3817 @kindex show breakpoint condition-evaluation
3819 @item set breakpoint condition-evaluation host
3820 This option commands @value{GDBN} to evaluate the breakpoint
3821 conditions on the host's side. Unconditional breakpoints are sent to
3822 the target which in turn receives the triggers and reports them back to GDB
3823 for condition evaluation. This is the standard evaluation mode.
3825 @item set breakpoint condition-evaluation target
3826 This option commands @value{GDBN} to download breakpoint conditions
3827 to the target at the moment of their insertion. The target
3828 is responsible for evaluating the conditional expression and reporting
3829 breakpoint stop events back to @value{GDBN} whenever the condition
3830 is true. Due to limitations of target-side evaluation, some conditions
3831 cannot be evaluated there, e.g., conditions that depend on local data
3832 that is only known to the host. Examples include
3833 conditional expressions involving convenience variables, complex types
3834 that cannot be handled by the agent expression parser and expressions
3835 that are too long to be sent over to the target, specially when the
3836 target is a remote system. In these cases, the conditions will be
3837 evaluated by @value{GDBN}.
3839 @item set breakpoint condition-evaluation auto
3840 This is the default mode. If the target supports evaluating breakpoint
3841 conditions on its end, @value{GDBN} will download breakpoint conditions to
3842 the target (limitations mentioned previously apply). If the target does
3843 not support breakpoint condition evaluation, then @value{GDBN} will fallback
3844 to evaluating all these conditions on the host's side.
3848 @cindex negative breakpoint numbers
3849 @cindex internal @value{GDBN} breakpoints
3850 @value{GDBN} itself sometimes sets breakpoints in your program for
3851 special purposes, such as proper handling of @code{longjmp} (in C
3852 programs). These internal breakpoints are assigned negative numbers,
3853 starting with @code{-1}; @samp{info breakpoints} does not display them.
3854 You can see these breakpoints with the @value{GDBN} maintenance command
3855 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3858 @node Set Watchpoints
3859 @subsection Setting Watchpoints
3861 @cindex setting watchpoints
3862 You can use a watchpoint to stop execution whenever the value of an
3863 expression changes, without having to predict a particular place where
3864 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3865 The expression may be as simple as the value of a single variable, or
3866 as complex as many variables combined by operators. Examples include:
3870 A reference to the value of a single variable.
3873 An address cast to an appropriate data type. For example,
3874 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3875 address (assuming an @code{int} occupies 4 bytes).
3878 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3879 expression can use any operators valid in the program's native
3880 language (@pxref{Languages}).
3883 You can set a watchpoint on an expression even if the expression can
3884 not be evaluated yet. For instance, you can set a watchpoint on
3885 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3886 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3887 the expression produces a valid value. If the expression becomes
3888 valid in some other way than changing a variable (e.g.@: if the memory
3889 pointed to by @samp{*global_ptr} becomes readable as the result of a
3890 @code{malloc} call), @value{GDBN} may not stop until the next time
3891 the expression changes.
3893 @cindex software watchpoints
3894 @cindex hardware watchpoints
3895 Depending on your system, watchpoints may be implemented in software or
3896 hardware. @value{GDBN} does software watchpointing by single-stepping your
3897 program and testing the variable's value each time, which is hundreds of
3898 times slower than normal execution. (But this may still be worth it, to
3899 catch errors where you have no clue what part of your program is the
3902 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3903 x86-based targets, @value{GDBN} includes support for hardware
3904 watchpoints, which do not slow down the running of your program.
3908 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3909 Set a watchpoint for an expression. @value{GDBN} will break when the
3910 expression @var{expr} is written into by the program and its value
3911 changes. The simplest (and the most popular) use of this command is
3912 to watch the value of a single variable:
3915 (@value{GDBP}) watch foo
3918 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3919 argument, @value{GDBN} breaks only when the thread identified by
3920 @var{threadnum} changes the value of @var{expr}. If any other threads
3921 change the value of @var{expr}, @value{GDBN} will not break. Note
3922 that watchpoints restricted to a single thread in this way only work
3923 with Hardware Watchpoints.
3925 Ordinarily a watchpoint respects the scope of variables in @var{expr}
3926 (see below). The @code{-location} argument tells @value{GDBN} to
3927 instead watch the memory referred to by @var{expr}. In this case,
3928 @value{GDBN} will evaluate @var{expr}, take the address of the result,
3929 and watch the memory at that address. The type of the result is used
3930 to determine the size of the watched memory. If the expression's
3931 result does not have an address, then @value{GDBN} will print an
3934 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
3935 of masked watchpoints, if the current architecture supports this
3936 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
3937 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
3938 to an address to watch. The mask specifies that some bits of an address
3939 (the bits which are reset in the mask) should be ignored when matching
3940 the address accessed by the inferior against the watchpoint address.
3941 Thus, a masked watchpoint watches many addresses simultaneously---those
3942 addresses whose unmasked bits are identical to the unmasked bits in the
3943 watchpoint address. The @code{mask} argument implies @code{-location}.
3947 (@value{GDBP}) watch foo mask 0xffff00ff
3948 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
3952 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3953 Set a watchpoint that will break when the value of @var{expr} is read
3957 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3958 Set a watchpoint that will break when @var{expr} is either read from
3959 or written into by the program.
3961 @kindex info watchpoints @r{[}@var{n}@dots{}@r{]}
3962 @item info watchpoints @r{[}@var{n}@dots{}@r{]}
3963 This command prints a list of watchpoints, using the same format as
3964 @code{info break} (@pxref{Set Breaks}).
3967 If you watch for a change in a numerically entered address you need to
3968 dereference it, as the address itself is just a constant number which will
3969 never change. @value{GDBN} refuses to create a watchpoint that watches
3970 a never-changing value:
3973 (@value{GDBP}) watch 0x600850
3974 Cannot watch constant value 0x600850.
3975 (@value{GDBP}) watch *(int *) 0x600850
3976 Watchpoint 1: *(int *) 6293584
3979 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3980 watchpoints execute very quickly, and the debugger reports a change in
3981 value at the exact instruction where the change occurs. If @value{GDBN}
3982 cannot set a hardware watchpoint, it sets a software watchpoint, which
3983 executes more slowly and reports the change in value at the next
3984 @emph{statement}, not the instruction, after the change occurs.
3986 @cindex use only software watchpoints
3987 You can force @value{GDBN} to use only software watchpoints with the
3988 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3989 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3990 the underlying system supports them. (Note that hardware-assisted
3991 watchpoints that were set @emph{before} setting
3992 @code{can-use-hw-watchpoints} to zero will still use the hardware
3993 mechanism of watching expression values.)
3996 @item set can-use-hw-watchpoints
3997 @kindex set can-use-hw-watchpoints
3998 Set whether or not to use hardware watchpoints.
4000 @item show can-use-hw-watchpoints
4001 @kindex show can-use-hw-watchpoints
4002 Show the current mode of using hardware watchpoints.
4005 For remote targets, you can restrict the number of hardware
4006 watchpoints @value{GDBN} will use, see @ref{set remote
4007 hardware-breakpoint-limit}.
4009 When you issue the @code{watch} command, @value{GDBN} reports
4012 Hardware watchpoint @var{num}: @var{expr}
4016 if it was able to set a hardware watchpoint.
4018 Currently, the @code{awatch} and @code{rwatch} commands can only set
4019 hardware watchpoints, because accesses to data that don't change the
4020 value of the watched expression cannot be detected without examining
4021 every instruction as it is being executed, and @value{GDBN} does not do
4022 that currently. If @value{GDBN} finds that it is unable to set a
4023 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4024 will print a message like this:
4027 Expression cannot be implemented with read/access watchpoint.
4030 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4031 data type of the watched expression is wider than what a hardware
4032 watchpoint on the target machine can handle. For example, some systems
4033 can only watch regions that are up to 4 bytes wide; on such systems you
4034 cannot set hardware watchpoints for an expression that yields a
4035 double-precision floating-point number (which is typically 8 bytes
4036 wide). As a work-around, it might be possible to break the large region
4037 into a series of smaller ones and watch them with separate watchpoints.
4039 If you set too many hardware watchpoints, @value{GDBN} might be unable
4040 to insert all of them when you resume the execution of your program.
4041 Since the precise number of active watchpoints is unknown until such
4042 time as the program is about to be resumed, @value{GDBN} might not be
4043 able to warn you about this when you set the watchpoints, and the
4044 warning will be printed only when the program is resumed:
4047 Hardware watchpoint @var{num}: Could not insert watchpoint
4051 If this happens, delete or disable some of the watchpoints.
4053 Watching complex expressions that reference many variables can also
4054 exhaust the resources available for hardware-assisted watchpoints.
4055 That's because @value{GDBN} needs to watch every variable in the
4056 expression with separately allocated resources.
4058 If you call a function interactively using @code{print} or @code{call},
4059 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4060 kind of breakpoint or the call completes.
4062 @value{GDBN} automatically deletes watchpoints that watch local
4063 (automatic) variables, or expressions that involve such variables, when
4064 they go out of scope, that is, when the execution leaves the block in
4065 which these variables were defined. In particular, when the program
4066 being debugged terminates, @emph{all} local variables go out of scope,
4067 and so only watchpoints that watch global variables remain set. If you
4068 rerun the program, you will need to set all such watchpoints again. One
4069 way of doing that would be to set a code breakpoint at the entry to the
4070 @code{main} function and when it breaks, set all the watchpoints.
4072 @cindex watchpoints and threads
4073 @cindex threads and watchpoints
4074 In multi-threaded programs, watchpoints will detect changes to the
4075 watched expression from every thread.
4078 @emph{Warning:} In multi-threaded programs, software watchpoints
4079 have only limited usefulness. If @value{GDBN} creates a software
4080 watchpoint, it can only watch the value of an expression @emph{in a
4081 single thread}. If you are confident that the expression can only
4082 change due to the current thread's activity (and if you are also
4083 confident that no other thread can become current), then you can use
4084 software watchpoints as usual. However, @value{GDBN} may not notice
4085 when a non-current thread's activity changes the expression. (Hardware
4086 watchpoints, in contrast, watch an expression in all threads.)
4089 @xref{set remote hardware-watchpoint-limit}.
4091 @node Set Catchpoints
4092 @subsection Setting Catchpoints
4093 @cindex catchpoints, setting
4094 @cindex exception handlers
4095 @cindex event handling
4097 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4098 kinds of program events, such as C@t{++} exceptions or the loading of a
4099 shared library. Use the @code{catch} command to set a catchpoint.
4103 @item catch @var{event}
4104 Stop when @var{event} occurs. @var{event} can be any of the following:
4107 @item throw @r{[}@var{regexp}@r{]}
4108 @itemx rethrow @r{[}@var{regexp}@r{]}
4109 @itemx catch @r{[}@var{regexp}@r{]}
4110 @cindex stop on C@t{++} exceptions
4111 The throwing, re-throwing, or catching of a C@t{++} exception.
4113 If @var{regexp} is given, then only exceptions whose type matches the
4114 regular expression will be caught.
4116 @vindex $_exception@r{, convenience variable}
4117 The convenience variable @code{$_exception} is available at an
4118 exception-related catchpoint, on some systems. This holds the
4119 exception being thrown.
4121 There are currently some limitations to C@t{++} exception handling in
4126 The support for these commands is system-dependent. Currently, only
4127 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4131 The regular expression feature and the @code{$_exception} convenience
4132 variable rely on the presence of some SDT probes in @code{libstdc++}.
4133 If these probes are not present, then these features cannot be used.
4134 These probes were first available in the GCC 4.8 release, but whether
4135 or not they are available in your GCC also depends on how it was
4139 The @code{$_exception} convenience variable is only valid at the
4140 instruction at which an exception-related catchpoint is set.
4143 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4144 location in the system library which implements runtime exception
4145 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4146 (@pxref{Selection}) to get to your code.
4149 If you call a function interactively, @value{GDBN} normally returns
4150 control to you when the function has finished executing. If the call
4151 raises an exception, however, the call may bypass the mechanism that
4152 returns control to you and cause your program either to abort or to
4153 simply continue running until it hits a breakpoint, catches a signal
4154 that @value{GDBN} is listening for, or exits. This is the case even if
4155 you set a catchpoint for the exception; catchpoints on exceptions are
4156 disabled within interactive calls. @xref{Calling}, for information on
4157 controlling this with @code{set unwind-on-terminating-exception}.
4160 You cannot raise an exception interactively.
4163 You cannot install an exception handler interactively.
4167 @cindex Ada exception catching
4168 @cindex catch Ada exceptions
4169 An Ada exception being raised. If an exception name is specified
4170 at the end of the command (eg @code{catch exception Program_Error}),
4171 the debugger will stop only when this specific exception is raised.
4172 Otherwise, the debugger stops execution when any Ada exception is raised.
4174 When inserting an exception catchpoint on a user-defined exception whose
4175 name is identical to one of the exceptions defined by the language, the
4176 fully qualified name must be used as the exception name. Otherwise,
4177 @value{GDBN} will assume that it should stop on the pre-defined exception
4178 rather than the user-defined one. For instance, assuming an exception
4179 called @code{Constraint_Error} is defined in package @code{Pck}, then
4180 the command to use to catch such exceptions is @kbd{catch exception
4181 Pck.Constraint_Error}.
4183 @item exception unhandled
4184 An exception that was raised but is not handled by the program.
4187 A failed Ada assertion.
4190 @cindex break on fork/exec
4191 A call to @code{exec}. This is currently only available for HP-UX
4195 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
4196 @cindex break on a system call.
4197 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4198 syscall is a mechanism for application programs to request a service
4199 from the operating system (OS) or one of the OS system services.
4200 @value{GDBN} can catch some or all of the syscalls issued by the
4201 debuggee, and show the related information for each syscall. If no
4202 argument is specified, calls to and returns from all system calls
4205 @var{name} can be any system call name that is valid for the
4206 underlying OS. Just what syscalls are valid depends on the OS. On
4207 GNU and Unix systems, you can find the full list of valid syscall
4208 names on @file{/usr/include/asm/unistd.h}.
4210 @c For MS-Windows, the syscall names and the corresponding numbers
4211 @c can be found, e.g., on this URL:
4212 @c http://www.metasploit.com/users/opcode/syscalls.html
4213 @c but we don't support Windows syscalls yet.
4215 Normally, @value{GDBN} knows in advance which syscalls are valid for
4216 each OS, so you can use the @value{GDBN} command-line completion
4217 facilities (@pxref{Completion,, command completion}) to list the
4220 You may also specify the system call numerically. A syscall's
4221 number is the value passed to the OS's syscall dispatcher to
4222 identify the requested service. When you specify the syscall by its
4223 name, @value{GDBN} uses its database of syscalls to convert the name
4224 into the corresponding numeric code, but using the number directly
4225 may be useful if @value{GDBN}'s database does not have the complete
4226 list of syscalls on your system (e.g., because @value{GDBN} lags
4227 behind the OS upgrades).
4229 The example below illustrates how this command works if you don't provide
4233 (@value{GDBP}) catch syscall
4234 Catchpoint 1 (syscall)
4236 Starting program: /tmp/catch-syscall
4238 Catchpoint 1 (call to syscall 'close'), \
4239 0xffffe424 in __kernel_vsyscall ()
4243 Catchpoint 1 (returned from syscall 'close'), \
4244 0xffffe424 in __kernel_vsyscall ()
4248 Here is an example of catching a system call by name:
4251 (@value{GDBP}) catch syscall chroot
4252 Catchpoint 1 (syscall 'chroot' [61])
4254 Starting program: /tmp/catch-syscall
4256 Catchpoint 1 (call to syscall 'chroot'), \
4257 0xffffe424 in __kernel_vsyscall ()
4261 Catchpoint 1 (returned from syscall 'chroot'), \
4262 0xffffe424 in __kernel_vsyscall ()
4266 An example of specifying a system call numerically. In the case
4267 below, the syscall number has a corresponding entry in the XML
4268 file, so @value{GDBN} finds its name and prints it:
4271 (@value{GDBP}) catch syscall 252
4272 Catchpoint 1 (syscall(s) 'exit_group')
4274 Starting program: /tmp/catch-syscall
4276 Catchpoint 1 (call to syscall 'exit_group'), \
4277 0xffffe424 in __kernel_vsyscall ()
4281 Program exited normally.
4285 However, there can be situations when there is no corresponding name
4286 in XML file for that syscall number. In this case, @value{GDBN} prints
4287 a warning message saying that it was not able to find the syscall name,
4288 but the catchpoint will be set anyway. See the example below:
4291 (@value{GDBP}) catch syscall 764
4292 warning: The number '764' does not represent a known syscall.
4293 Catchpoint 2 (syscall 764)
4297 If you configure @value{GDBN} using the @samp{--without-expat} option,
4298 it will not be able to display syscall names. Also, if your
4299 architecture does not have an XML file describing its system calls,
4300 you will not be able to see the syscall names. It is important to
4301 notice that these two features are used for accessing the syscall
4302 name database. In either case, you will see a warning like this:
4305 (@value{GDBP}) catch syscall
4306 warning: Could not open "syscalls/i386-linux.xml"
4307 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4308 GDB will not be able to display syscall names.
4309 Catchpoint 1 (syscall)
4313 Of course, the file name will change depending on your architecture and system.
4315 Still using the example above, you can also try to catch a syscall by its
4316 number. In this case, you would see something like:
4319 (@value{GDBP}) catch syscall 252
4320 Catchpoint 1 (syscall(s) 252)
4323 Again, in this case @value{GDBN} would not be able to display syscall's names.
4326 A call to @code{fork}. This is currently only available for HP-UX
4330 A call to @code{vfork}. This is currently only available for HP-UX
4333 @item load @r{[}regexp@r{]}
4334 @itemx unload @r{[}regexp@r{]}
4335 The loading or unloading of a shared library. If @var{regexp} is
4336 given, then the catchpoint will stop only if the regular expression
4337 matches one of the affected libraries.
4339 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4340 The delivery of a signal.
4342 With no arguments, this catchpoint will catch any signal that is not
4343 used internally by @value{GDBN}, specifically, all signals except
4344 @samp{SIGTRAP} and @samp{SIGINT}.
4346 With the argument @samp{all}, all signals, including those used by
4347 @value{GDBN}, will be caught. This argument cannot be used with other
4350 Otherwise, the arguments are a list of signal names as given to
4351 @code{handle} (@pxref{Signals}). Only signals specified in this list
4354 One reason that @code{catch signal} can be more useful than
4355 @code{handle} is that you can attach commands and conditions to the
4358 When a signal is caught by a catchpoint, the signal's @code{stop} and
4359 @code{print} settings, as specified by @code{handle}, are ignored.
4360 However, whether the signal is still delivered to the inferior depends
4361 on the @code{pass} setting; this can be changed in the catchpoint's
4366 @item tcatch @var{event}
4367 Set a catchpoint that is enabled only for one stop. The catchpoint is
4368 automatically deleted after the first time the event is caught.
4372 Use the @code{info break} command to list the current catchpoints.
4376 @subsection Deleting Breakpoints
4378 @cindex clearing breakpoints, watchpoints, catchpoints
4379 @cindex deleting breakpoints, watchpoints, catchpoints
4380 It is often necessary to eliminate a breakpoint, watchpoint, or
4381 catchpoint once it has done its job and you no longer want your program
4382 to stop there. This is called @dfn{deleting} the breakpoint. A
4383 breakpoint that has been deleted no longer exists; it is forgotten.
4385 With the @code{clear} command you can delete breakpoints according to
4386 where they are in your program. With the @code{delete} command you can
4387 delete individual breakpoints, watchpoints, or catchpoints by specifying
4388 their breakpoint numbers.
4390 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4391 automatically ignores breakpoints on the first instruction to be executed
4392 when you continue execution without changing the execution address.
4397 Delete any breakpoints at the next instruction to be executed in the
4398 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4399 the innermost frame is selected, this is a good way to delete a
4400 breakpoint where your program just stopped.
4402 @item clear @var{location}
4403 Delete any breakpoints set at the specified @var{location}.
4404 @xref{Specify Location}, for the various forms of @var{location}; the
4405 most useful ones are listed below:
4408 @item clear @var{function}
4409 @itemx clear @var{filename}:@var{function}
4410 Delete any breakpoints set at entry to the named @var{function}.
4412 @item clear @var{linenum}
4413 @itemx clear @var{filename}:@var{linenum}
4414 Delete any breakpoints set at or within the code of the specified
4415 @var{linenum} of the specified @var{filename}.
4418 @cindex delete breakpoints
4420 @kindex d @r{(@code{delete})}
4421 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4422 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4423 ranges specified as arguments. If no argument is specified, delete all
4424 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4425 confirm off}). You can abbreviate this command as @code{d}.
4429 @subsection Disabling Breakpoints
4431 @cindex enable/disable a breakpoint
4432 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4433 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4434 it had been deleted, but remembers the information on the breakpoint so
4435 that you can @dfn{enable} it again later.
4437 You disable and enable breakpoints, watchpoints, and catchpoints with
4438 the @code{enable} and @code{disable} commands, optionally specifying
4439 one or more breakpoint numbers as arguments. Use @code{info break} to
4440 print a list of all breakpoints, watchpoints, and catchpoints if you
4441 do not know which numbers to use.
4443 Disabling and enabling a breakpoint that has multiple locations
4444 affects all of its locations.
4446 A breakpoint, watchpoint, or catchpoint can have any of several
4447 different states of enablement:
4451 Enabled. The breakpoint stops your program. A breakpoint set
4452 with the @code{break} command starts out in this state.
4454 Disabled. The breakpoint has no effect on your program.
4456 Enabled once. The breakpoint stops your program, but then becomes
4459 Enabled for a count. The breakpoint stops your program for the next
4460 N times, then becomes disabled.
4462 Enabled for deletion. The breakpoint stops your program, but
4463 immediately after it does so it is deleted permanently. A breakpoint
4464 set with the @code{tbreak} command starts out in this state.
4467 You can use the following commands to enable or disable breakpoints,
4468 watchpoints, and catchpoints:
4472 @kindex dis @r{(@code{disable})}
4473 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4474 Disable the specified breakpoints---or all breakpoints, if none are
4475 listed. A disabled breakpoint has no effect but is not forgotten. All
4476 options such as ignore-counts, conditions and commands are remembered in
4477 case the breakpoint is enabled again later. You may abbreviate
4478 @code{disable} as @code{dis}.
4481 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4482 Enable the specified breakpoints (or all defined breakpoints). They
4483 become effective once again in stopping your program.
4485 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4486 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4487 of these breakpoints immediately after stopping your program.
4489 @item enable @r{[}breakpoints@r{]} count @var{count} @var{range}@dots{}
4490 Enable the specified breakpoints temporarily. @value{GDBN} records
4491 @var{count} with each of the specified breakpoints, and decrements a
4492 breakpoint's count when it is hit. When any count reaches 0,
4493 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4494 count (@pxref{Conditions, ,Break Conditions}), that will be
4495 decremented to 0 before @var{count} is affected.
4497 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4498 Enable the specified breakpoints to work once, then die. @value{GDBN}
4499 deletes any of these breakpoints as soon as your program stops there.
4500 Breakpoints set by the @code{tbreak} command start out in this state.
4503 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4504 @c confusing: tbreak is also initially enabled.
4505 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4506 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4507 subsequently, they become disabled or enabled only when you use one of
4508 the commands above. (The command @code{until} can set and delete a
4509 breakpoint of its own, but it does not change the state of your other
4510 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4514 @subsection Break Conditions
4515 @cindex conditional breakpoints
4516 @cindex breakpoint conditions
4518 @c FIXME what is scope of break condition expr? Context where wanted?
4519 @c in particular for a watchpoint?
4520 The simplest sort of breakpoint breaks every time your program reaches a
4521 specified place. You can also specify a @dfn{condition} for a
4522 breakpoint. A condition is just a Boolean expression in your
4523 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4524 a condition evaluates the expression each time your program reaches it,
4525 and your program stops only if the condition is @emph{true}.
4527 This is the converse of using assertions for program validation; in that
4528 situation, you want to stop when the assertion is violated---that is,
4529 when the condition is false. In C, if you want to test an assertion expressed
4530 by the condition @var{assert}, you should set the condition
4531 @samp{! @var{assert}} on the appropriate breakpoint.
4533 Conditions are also accepted for watchpoints; you may not need them,
4534 since a watchpoint is inspecting the value of an expression anyhow---but
4535 it might be simpler, say, to just set a watchpoint on a variable name,
4536 and specify a condition that tests whether the new value is an interesting
4539 Break conditions can have side effects, and may even call functions in
4540 your program. This can be useful, for example, to activate functions
4541 that log program progress, or to use your own print functions to
4542 format special data structures. The effects are completely predictable
4543 unless there is another enabled breakpoint at the same address. (In
4544 that case, @value{GDBN} might see the other breakpoint first and stop your
4545 program without checking the condition of this one.) Note that
4546 breakpoint commands are usually more convenient and flexible than break
4548 purpose of performing side effects when a breakpoint is reached
4549 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4551 Breakpoint conditions can also be evaluated on the target's side if
4552 the target supports it. Instead of evaluating the conditions locally,
4553 @value{GDBN} encodes the expression into an agent expression
4554 (@pxref{Agent Expressions}) suitable for execution on the target,
4555 independently of @value{GDBN}. Global variables become raw memory
4556 locations, locals become stack accesses, and so forth.
4558 In this case, @value{GDBN} will only be notified of a breakpoint trigger
4559 when its condition evaluates to true. This mechanism may provide faster
4560 response times depending on the performance characteristics of the target
4561 since it does not need to keep @value{GDBN} informed about
4562 every breakpoint trigger, even those with false conditions.
4564 Break conditions can be specified when a breakpoint is set, by using
4565 @samp{if} in the arguments to the @code{break} command. @xref{Set
4566 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4567 with the @code{condition} command.
4569 You can also use the @code{if} keyword with the @code{watch} command.
4570 The @code{catch} command does not recognize the @code{if} keyword;
4571 @code{condition} is the only way to impose a further condition on a
4576 @item condition @var{bnum} @var{expression}
4577 Specify @var{expression} as the break condition for breakpoint,
4578 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4579 breakpoint @var{bnum} stops your program only if the value of
4580 @var{expression} is true (nonzero, in C). When you use
4581 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4582 syntactic correctness, and to determine whether symbols in it have
4583 referents in the context of your breakpoint. If @var{expression} uses
4584 symbols not referenced in the context of the breakpoint, @value{GDBN}
4585 prints an error message:
4588 No symbol "foo" in current context.
4593 not actually evaluate @var{expression} at the time the @code{condition}
4594 command (or a command that sets a breakpoint with a condition, like
4595 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4597 @item condition @var{bnum}
4598 Remove the condition from breakpoint number @var{bnum}. It becomes
4599 an ordinary unconditional breakpoint.
4602 @cindex ignore count (of breakpoint)
4603 A special case of a breakpoint condition is to stop only when the
4604 breakpoint has been reached a certain number of times. This is so
4605 useful that there is a special way to do it, using the @dfn{ignore
4606 count} of the breakpoint. Every breakpoint has an ignore count, which
4607 is an integer. Most of the time, the ignore count is zero, and
4608 therefore has no effect. But if your program reaches a breakpoint whose
4609 ignore count is positive, then instead of stopping, it just decrements
4610 the ignore count by one and continues. As a result, if the ignore count
4611 value is @var{n}, the breakpoint does not stop the next @var{n} times
4612 your program reaches it.
4616 @item ignore @var{bnum} @var{count}
4617 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4618 The next @var{count} times the breakpoint is reached, your program's
4619 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4622 To make the breakpoint stop the next time it is reached, specify
4625 When you use @code{continue} to resume execution of your program from a
4626 breakpoint, you can specify an ignore count directly as an argument to
4627 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4628 Stepping,,Continuing and Stepping}.
4630 If a breakpoint has a positive ignore count and a condition, the
4631 condition is not checked. Once the ignore count reaches zero,
4632 @value{GDBN} resumes checking the condition.
4634 You could achieve the effect of the ignore count with a condition such
4635 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4636 is decremented each time. @xref{Convenience Vars, ,Convenience
4640 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4643 @node Break Commands
4644 @subsection Breakpoint Command Lists
4646 @cindex breakpoint commands
4647 You can give any breakpoint (or watchpoint or catchpoint) a series of
4648 commands to execute when your program stops due to that breakpoint. For
4649 example, you might want to print the values of certain expressions, or
4650 enable other breakpoints.
4654 @kindex end@r{ (breakpoint commands)}
4655 @item commands @r{[}@var{range}@dots{}@r{]}
4656 @itemx @dots{} @var{command-list} @dots{}
4658 Specify a list of commands for the given breakpoints. The commands
4659 themselves appear on the following lines. Type a line containing just
4660 @code{end} to terminate the commands.
4662 To remove all commands from a breakpoint, type @code{commands} and
4663 follow it immediately with @code{end}; that is, give no commands.
4665 With no argument, @code{commands} refers to the last breakpoint,
4666 watchpoint, or catchpoint set (not to the breakpoint most recently
4667 encountered). If the most recent breakpoints were set with a single
4668 command, then the @code{commands} will apply to all the breakpoints
4669 set by that command. This applies to breakpoints set by
4670 @code{rbreak}, and also applies when a single @code{break} command
4671 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4675 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4676 disabled within a @var{command-list}.
4678 You can use breakpoint commands to start your program up again. Simply
4679 use the @code{continue} command, or @code{step}, or any other command
4680 that resumes execution.
4682 Any other commands in the command list, after a command that resumes
4683 execution, are ignored. This is because any time you resume execution
4684 (even with a simple @code{next} or @code{step}), you may encounter
4685 another breakpoint---which could have its own command list, leading to
4686 ambiguities about which list to execute.
4689 If the first command you specify in a command list is @code{silent}, the
4690 usual message about stopping at a breakpoint is not printed. This may
4691 be desirable for breakpoints that are to print a specific message and
4692 then continue. If none of the remaining commands print anything, you
4693 see no sign that the breakpoint was reached. @code{silent} is
4694 meaningful only at the beginning of a breakpoint command list.
4696 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4697 print precisely controlled output, and are often useful in silent
4698 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4700 For example, here is how you could use breakpoint commands to print the
4701 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4707 printf "x is %d\n",x
4712 One application for breakpoint commands is to compensate for one bug so
4713 you can test for another. Put a breakpoint just after the erroneous line
4714 of code, give it a condition to detect the case in which something
4715 erroneous has been done, and give it commands to assign correct values
4716 to any variables that need them. End with the @code{continue} command
4717 so that your program does not stop, and start with the @code{silent}
4718 command so that no output is produced. Here is an example:
4729 @node Dynamic Printf
4730 @subsection Dynamic Printf
4732 @cindex dynamic printf
4734 The dynamic printf command @code{dprintf} combines a breakpoint with
4735 formatted printing of your program's data to give you the effect of
4736 inserting @code{printf} calls into your program on-the-fly, without
4737 having to recompile it.
4739 In its most basic form, the output goes to the GDB console. However,
4740 you can set the variable @code{dprintf-style} for alternate handling.
4741 For instance, you can ask to format the output by calling your
4742 program's @code{printf} function. This has the advantage that the
4743 characters go to the program's output device, so they can recorded in
4744 redirects to files and so forth.
4746 If you are doing remote debugging with a stub or agent, you can also
4747 ask to have the printf handled by the remote agent. In addition to
4748 ensuring that the output goes to the remote program's device along
4749 with any other output the program might produce, you can also ask that
4750 the dprintf remain active even after disconnecting from the remote
4751 target. Using the stub/agent is also more efficient, as it can do
4752 everything without needing to communicate with @value{GDBN}.
4756 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
4757 Whenever execution reaches @var{location}, print the values of one or
4758 more @var{expressions} under the control of the string @var{template}.
4759 To print several values, separate them with commas.
4761 @item set dprintf-style @var{style}
4762 Set the dprintf output to be handled in one of several different
4763 styles enumerated below. A change of style affects all existing
4764 dynamic printfs immediately. (If you need individual control over the
4765 print commands, simply define normal breakpoints with
4766 explicitly-supplied command lists.)
4769 @kindex dprintf-style gdb
4770 Handle the output using the @value{GDBN} @code{printf} command.
4773 @kindex dprintf-style call
4774 Handle the output by calling a function in your program (normally
4778 @kindex dprintf-style agent
4779 Have the remote debugging agent (such as @code{gdbserver}) handle
4780 the output itself. This style is only available for agents that
4781 support running commands on the target.
4783 @item set dprintf-function @var{function}
4784 Set the function to call if the dprintf style is @code{call}. By
4785 default its value is @code{printf}. You may set it to any expression.
4786 that @value{GDBN} can evaluate to a function, as per the @code{call}
4789 @item set dprintf-channel @var{channel}
4790 Set a ``channel'' for dprintf. If set to a non-empty value,
4791 @value{GDBN} will evaluate it as an expression and pass the result as
4792 a first argument to the @code{dprintf-function}, in the manner of
4793 @code{fprintf} and similar functions. Otherwise, the dprintf format
4794 string will be the first argument, in the manner of @code{printf}.
4796 As an example, if you wanted @code{dprintf} output to go to a logfile
4797 that is a standard I/O stream assigned to the variable @code{mylog},
4798 you could do the following:
4801 (gdb) set dprintf-style call
4802 (gdb) set dprintf-function fprintf
4803 (gdb) set dprintf-channel mylog
4804 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
4805 Dprintf 1 at 0x123456: file main.c, line 25.
4807 1 dprintf keep y 0x00123456 in main at main.c:25
4808 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
4813 Note that the @code{info break} displays the dynamic printf commands
4814 as normal breakpoint commands; you can thus easily see the effect of
4815 the variable settings.
4817 @item set disconnected-dprintf on
4818 @itemx set disconnected-dprintf off
4819 @kindex set disconnected-dprintf
4820 Choose whether @code{dprintf} commands should continue to run if
4821 @value{GDBN} has disconnected from the target. This only applies
4822 if the @code{dprintf-style} is @code{agent}.
4824 @item show disconnected-dprintf off
4825 @kindex show disconnected-dprintf
4826 Show the current choice for disconnected @code{dprintf}.
4830 @value{GDBN} does not check the validity of function and channel,
4831 relying on you to supply values that are meaningful for the contexts
4832 in which they are being used. For instance, the function and channel
4833 may be the values of local variables, but if that is the case, then
4834 all enabled dynamic prints must be at locations within the scope of
4835 those locals. If evaluation fails, @value{GDBN} will report an error.
4837 @node Save Breakpoints
4838 @subsection How to save breakpoints to a file
4840 To save breakpoint definitions to a file use the @w{@code{save
4841 breakpoints}} command.
4844 @kindex save breakpoints
4845 @cindex save breakpoints to a file for future sessions
4846 @item save breakpoints [@var{filename}]
4847 This command saves all current breakpoint definitions together with
4848 their commands and ignore counts, into a file @file{@var{filename}}
4849 suitable for use in a later debugging session. This includes all
4850 types of breakpoints (breakpoints, watchpoints, catchpoints,
4851 tracepoints). To read the saved breakpoint definitions, use the
4852 @code{source} command (@pxref{Command Files}). Note that watchpoints
4853 with expressions involving local variables may fail to be recreated
4854 because it may not be possible to access the context where the
4855 watchpoint is valid anymore. Because the saved breakpoint definitions
4856 are simply a sequence of @value{GDBN} commands that recreate the
4857 breakpoints, you can edit the file in your favorite editing program,
4858 and remove the breakpoint definitions you're not interested in, or
4859 that can no longer be recreated.
4862 @node Static Probe Points
4863 @subsection Static Probe Points
4865 @cindex static probe point, SystemTap
4866 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
4867 for Statically Defined Tracing, and the probes are designed to have a tiny
4868 runtime code and data footprint, and no dynamic relocations. They are
4869 usable from assembly, C and C@t{++} languages. See
4870 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
4871 for a good reference on how the @acronym{SDT} probes are implemented.
4873 Currently, @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
4874 @acronym{SDT} probes are supported on ELF-compatible systems. See
4875 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
4876 for more information on how to add @code{SystemTap} @acronym{SDT} probes
4877 in your applications.
4879 @cindex semaphores on static probe points
4880 Some probes have an associated semaphore variable; for instance, this
4881 happens automatically if you defined your probe using a DTrace-style
4882 @file{.d} file. If your probe has a semaphore, @value{GDBN} will
4883 automatically enable it when you specify a breakpoint using the
4884 @samp{-probe-stap} notation. But, if you put a breakpoint at a probe's
4885 location by some other method (e.g., @code{break file:line}), then
4886 @value{GDBN} will not automatically set the semaphore.
4888 You can examine the available static static probes using @code{info
4889 probes}, with optional arguments:
4893 @item info probes stap @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
4894 If given, @var{provider} is a regular expression used to match against provider
4895 names when selecting which probes to list. If omitted, probes by all
4896 probes from all providers are listed.
4898 If given, @var{name} is a regular expression to match against probe names
4899 when selecting which probes to list. If omitted, probe names are not
4900 considered when deciding whether to display them.
4902 If given, @var{objfile} is a regular expression used to select which
4903 object files (executable or shared libraries) to examine. If not
4904 given, all object files are considered.
4906 @item info probes all
4907 List the available static probes, from all types.
4910 @vindex $_probe_arg@r{, convenience variable}
4911 A probe may specify up to twelve arguments. These are available at the
4912 point at which the probe is defined---that is, when the current PC is
4913 at the probe's location. The arguments are available using the
4914 convenience variables (@pxref{Convenience Vars})
4915 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. Each probe argument is
4916 an integer of the appropriate size; types are not preserved. The
4917 convenience variable @code{$_probe_argc} holds the number of arguments
4918 at the current probe point.
4920 These variables are always available, but attempts to access them at
4921 any location other than a probe point will cause @value{GDBN} to give
4925 @c @ifclear BARETARGET
4926 @node Error in Breakpoints
4927 @subsection ``Cannot insert breakpoints''
4929 If you request too many active hardware-assisted breakpoints and
4930 watchpoints, you will see this error message:
4932 @c FIXME: the precise wording of this message may change; the relevant
4933 @c source change is not committed yet (Sep 3, 1999).
4935 Stopped; cannot insert breakpoints.
4936 You may have requested too many hardware breakpoints and watchpoints.
4940 This message is printed when you attempt to resume the program, since
4941 only then @value{GDBN} knows exactly how many hardware breakpoints and
4942 watchpoints it needs to insert.
4944 When this message is printed, you need to disable or remove some of the
4945 hardware-assisted breakpoints and watchpoints, and then continue.
4947 @node Breakpoint-related Warnings
4948 @subsection ``Breakpoint address adjusted...''
4949 @cindex breakpoint address adjusted
4951 Some processor architectures place constraints on the addresses at
4952 which breakpoints may be placed. For architectures thus constrained,
4953 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4954 with the constraints dictated by the architecture.
4956 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4957 a VLIW architecture in which a number of RISC-like instructions may be
4958 bundled together for parallel execution. The FR-V architecture
4959 constrains the location of a breakpoint instruction within such a
4960 bundle to the instruction with the lowest address. @value{GDBN}
4961 honors this constraint by adjusting a breakpoint's address to the
4962 first in the bundle.
4964 It is not uncommon for optimized code to have bundles which contain
4965 instructions from different source statements, thus it may happen that
4966 a breakpoint's address will be adjusted from one source statement to
4967 another. Since this adjustment may significantly alter @value{GDBN}'s
4968 breakpoint related behavior from what the user expects, a warning is
4969 printed when the breakpoint is first set and also when the breakpoint
4972 A warning like the one below is printed when setting a breakpoint
4973 that's been subject to address adjustment:
4976 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4979 Such warnings are printed both for user settable and @value{GDBN}'s
4980 internal breakpoints. If you see one of these warnings, you should
4981 verify that a breakpoint set at the adjusted address will have the
4982 desired affect. If not, the breakpoint in question may be removed and
4983 other breakpoints may be set which will have the desired behavior.
4984 E.g., it may be sufficient to place the breakpoint at a later
4985 instruction. A conditional breakpoint may also be useful in some
4986 cases to prevent the breakpoint from triggering too often.
4988 @value{GDBN} will also issue a warning when stopping at one of these
4989 adjusted breakpoints:
4992 warning: Breakpoint 1 address previously adjusted from 0x00010414
4996 When this warning is encountered, it may be too late to take remedial
4997 action except in cases where the breakpoint is hit earlier or more
4998 frequently than expected.
5000 @node Continuing and Stepping
5001 @section Continuing and Stepping
5005 @cindex resuming execution
5006 @dfn{Continuing} means resuming program execution until your program
5007 completes normally. In contrast, @dfn{stepping} means executing just
5008 one more ``step'' of your program, where ``step'' may mean either one
5009 line of source code, or one machine instruction (depending on what
5010 particular command you use). Either when continuing or when stepping,
5011 your program may stop even sooner, due to a breakpoint or a signal. (If
5012 it stops due to a signal, you may want to use @code{handle}, or use
5013 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
5017 @kindex c @r{(@code{continue})}
5018 @kindex fg @r{(resume foreground execution)}
5019 @item continue @r{[}@var{ignore-count}@r{]}
5020 @itemx c @r{[}@var{ignore-count}@r{]}
5021 @itemx fg @r{[}@var{ignore-count}@r{]}
5022 Resume program execution, at the address where your program last stopped;
5023 any breakpoints set at that address are bypassed. The optional argument
5024 @var{ignore-count} allows you to specify a further number of times to
5025 ignore a breakpoint at this location; its effect is like that of
5026 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5028 The argument @var{ignore-count} is meaningful only when your program
5029 stopped due to a breakpoint. At other times, the argument to
5030 @code{continue} is ignored.
5032 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5033 debugged program is deemed to be the foreground program) are provided
5034 purely for convenience, and have exactly the same behavior as
5038 To resume execution at a different place, you can use @code{return}
5039 (@pxref{Returning, ,Returning from a Function}) to go back to the
5040 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5041 Different Address}) to go to an arbitrary location in your program.
5043 A typical technique for using stepping is to set a breakpoint
5044 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5045 beginning of the function or the section of your program where a problem
5046 is believed to lie, run your program until it stops at that breakpoint,
5047 and then step through the suspect area, examining the variables that are
5048 interesting, until you see the problem happen.
5052 @kindex s @r{(@code{step})}
5054 Continue running your program until control reaches a different source
5055 line, then stop it and return control to @value{GDBN}. This command is
5056 abbreviated @code{s}.
5059 @c "without debugging information" is imprecise; actually "without line
5060 @c numbers in the debugging information". (gcc -g1 has debugging info but
5061 @c not line numbers). But it seems complex to try to make that
5062 @c distinction here.
5063 @emph{Warning:} If you use the @code{step} command while control is
5064 within a function that was compiled without debugging information,
5065 execution proceeds until control reaches a function that does have
5066 debugging information. Likewise, it will not step into a function which
5067 is compiled without debugging information. To step through functions
5068 without debugging information, use the @code{stepi} command, described
5072 The @code{step} command only stops at the first instruction of a source
5073 line. This prevents the multiple stops that could otherwise occur in
5074 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5075 to stop if a function that has debugging information is called within
5076 the line. In other words, @code{step} @emph{steps inside} any functions
5077 called within the line.
5079 Also, the @code{step} command only enters a function if there is line
5080 number information for the function. Otherwise it acts like the
5081 @code{next} command. This avoids problems when using @code{cc -gl}
5082 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5083 was any debugging information about the routine.
5085 @item step @var{count}
5086 Continue running as in @code{step}, but do so @var{count} times. If a
5087 breakpoint is reached, or a signal not related to stepping occurs before
5088 @var{count} steps, stepping stops right away.
5091 @kindex n @r{(@code{next})}
5092 @item next @r{[}@var{count}@r{]}
5093 Continue to the next source line in the current (innermost) stack frame.
5094 This is similar to @code{step}, but function calls that appear within
5095 the line of code are executed without stopping. Execution stops when
5096 control reaches a different line of code at the original stack level
5097 that was executing when you gave the @code{next} command. This command
5098 is abbreviated @code{n}.
5100 An argument @var{count} is a repeat count, as for @code{step}.
5103 @c FIX ME!! Do we delete this, or is there a way it fits in with
5104 @c the following paragraph? --- Vctoria
5106 @c @code{next} within a function that lacks debugging information acts like
5107 @c @code{step}, but any function calls appearing within the code of the
5108 @c function are executed without stopping.
5110 The @code{next} command only stops at the first instruction of a
5111 source line. This prevents multiple stops that could otherwise occur in
5112 @code{switch} statements, @code{for} loops, etc.
5114 @kindex set step-mode
5116 @cindex functions without line info, and stepping
5117 @cindex stepping into functions with no line info
5118 @itemx set step-mode on
5119 The @code{set step-mode on} command causes the @code{step} command to
5120 stop at the first instruction of a function which contains no debug line
5121 information rather than stepping over it.
5123 This is useful in cases where you may be interested in inspecting the
5124 machine instructions of a function which has no symbolic info and do not
5125 want @value{GDBN} to automatically skip over this function.
5127 @item set step-mode off
5128 Causes the @code{step} command to step over any functions which contains no
5129 debug information. This is the default.
5131 @item show step-mode
5132 Show whether @value{GDBN} will stop in or step over functions without
5133 source line debug information.
5136 @kindex fin @r{(@code{finish})}
5138 Continue running until just after function in the selected stack frame
5139 returns. Print the returned value (if any). This command can be
5140 abbreviated as @code{fin}.
5142 Contrast this with the @code{return} command (@pxref{Returning,
5143 ,Returning from a Function}).
5146 @kindex u @r{(@code{until})}
5147 @cindex run until specified location
5150 Continue running until a source line past the current line, in the
5151 current stack frame, is reached. This command is used to avoid single
5152 stepping through a loop more than once. It is like the @code{next}
5153 command, except that when @code{until} encounters a jump, it
5154 automatically continues execution until the program counter is greater
5155 than the address of the jump.
5157 This means that when you reach the end of a loop after single stepping
5158 though it, @code{until} makes your program continue execution until it
5159 exits the loop. In contrast, a @code{next} command at the end of a loop
5160 simply steps back to the beginning of the loop, which forces you to step
5161 through the next iteration.
5163 @code{until} always stops your program if it attempts to exit the current
5166 @code{until} may produce somewhat counterintuitive results if the order
5167 of machine code does not match the order of the source lines. For
5168 example, in the following excerpt from a debugging session, the @code{f}
5169 (@code{frame}) command shows that execution is stopped at line
5170 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5174 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5176 (@value{GDBP}) until
5177 195 for ( ; argc > 0; NEXTARG) @{
5180 This happened because, for execution efficiency, the compiler had
5181 generated code for the loop closure test at the end, rather than the
5182 start, of the loop---even though the test in a C @code{for}-loop is
5183 written before the body of the loop. The @code{until} command appeared
5184 to step back to the beginning of the loop when it advanced to this
5185 expression; however, it has not really gone to an earlier
5186 statement---not in terms of the actual machine code.
5188 @code{until} with no argument works by means of single
5189 instruction stepping, and hence is slower than @code{until} with an
5192 @item until @var{location}
5193 @itemx u @var{location}
5194 Continue running your program until either the specified location is
5195 reached, or the current stack frame returns. @var{location} is any of
5196 the forms described in @ref{Specify Location}.
5197 This form of the command uses temporary breakpoints, and
5198 hence is quicker than @code{until} without an argument. The specified
5199 location is actually reached only if it is in the current frame. This
5200 implies that @code{until} can be used to skip over recursive function
5201 invocations. For instance in the code below, if the current location is
5202 line @code{96}, issuing @code{until 99} will execute the program up to
5203 line @code{99} in the same invocation of factorial, i.e., after the inner
5204 invocations have returned.
5207 94 int factorial (int value)
5209 96 if (value > 1) @{
5210 97 value *= factorial (value - 1);
5217 @kindex advance @var{location}
5218 @item advance @var{location}
5219 Continue running the program up to the given @var{location}. An argument is
5220 required, which should be of one of the forms described in
5221 @ref{Specify Location}.
5222 Execution will also stop upon exit from the current stack
5223 frame. This command is similar to @code{until}, but @code{advance} will
5224 not skip over recursive function calls, and the target location doesn't
5225 have to be in the same frame as the current one.
5229 @kindex si @r{(@code{stepi})}
5231 @itemx stepi @var{arg}
5233 Execute one machine instruction, then stop and return to the debugger.
5235 It is often useful to do @samp{display/i $pc} when stepping by machine
5236 instructions. This makes @value{GDBN} automatically display the next
5237 instruction to be executed, each time your program stops. @xref{Auto
5238 Display,, Automatic Display}.
5240 An argument is a repeat count, as in @code{step}.
5244 @kindex ni @r{(@code{nexti})}
5246 @itemx nexti @var{arg}
5248 Execute one machine instruction, but if it is a function call,
5249 proceed until the function returns.
5251 An argument is a repeat count, as in @code{next}.
5255 @anchor{range stepping}
5256 @cindex range stepping
5257 @cindex target-assisted range stepping
5258 By default, and if available, @value{GDBN} makes use of
5259 target-assisted @dfn{range stepping}. In other words, whenever you
5260 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5261 tells the target to step the corresponding range of instruction
5262 addresses instead of issuing multiple single-steps. This speeds up
5263 line stepping, particularly for remote targets. Ideally, there should
5264 be no reason you would want to turn range stepping off. However, it's
5265 possible that a bug in the debug info, a bug in the remote stub (for
5266 remote targets), or even a bug in @value{GDBN} could make line
5267 stepping behave incorrectly when target-assisted range stepping is
5268 enabled. You can use the following command to turn off range stepping
5272 @kindex set range-stepping
5273 @kindex show range-stepping
5274 @item set range-stepping
5275 @itemx show range-stepping
5276 Control whether range stepping is enabled.
5278 If @code{on}, and the target supports it, @value{GDBN} tells the
5279 target to step a range of addresses itself, instead of issuing
5280 multiple single-steps. If @code{off}, @value{GDBN} always issues
5281 single-steps, even if range stepping is supported by the target. The
5282 default is @code{on}.
5286 @node Skipping Over Functions and Files
5287 @section Skipping Over Functions and Files
5288 @cindex skipping over functions and files
5290 The program you are debugging may contain some functions which are
5291 uninteresting to debug. The @code{skip} comand lets you tell @value{GDBN} to
5292 skip a function or all functions in a file when stepping.
5294 For example, consider the following C function:
5305 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5306 are not interested in stepping through @code{boring}. If you run @code{step}
5307 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5308 step over both @code{foo} and @code{boring}!
5310 One solution is to @code{step} into @code{boring} and use the @code{finish}
5311 command to immediately exit it. But this can become tedious if @code{boring}
5312 is called from many places.
5314 A more flexible solution is to execute @kbd{skip boring}. This instructs
5315 @value{GDBN} never to step into @code{boring}. Now when you execute
5316 @code{step} at line 103, you'll step over @code{boring} and directly into
5319 You can also instruct @value{GDBN} to skip all functions in a file, with, for
5320 example, @code{skip file boring.c}.
5323 @kindex skip function
5324 @item skip @r{[}@var{linespec}@r{]}
5325 @itemx skip function @r{[}@var{linespec}@r{]}
5326 After running this command, the function named by @var{linespec} or the
5327 function containing the line named by @var{linespec} will be skipped over when
5328 stepping. @xref{Specify Location}.
5330 If you do not specify @var{linespec}, the function you're currently debugging
5333 (If you have a function called @code{file} that you want to skip, use
5334 @kbd{skip function file}.)
5337 @item skip file @r{[}@var{filename}@r{]}
5338 After running this command, any function whose source lives in @var{filename}
5339 will be skipped over when stepping.
5341 If you do not specify @var{filename}, functions whose source lives in the file
5342 you're currently debugging will be skipped.
5345 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5346 These are the commands for managing your list of skips:
5350 @item info skip @r{[}@var{range}@r{]}
5351 Print details about the specified skip(s). If @var{range} is not specified,
5352 print a table with details about all functions and files marked for skipping.
5353 @code{info skip} prints the following information about each skip:
5357 A number identifying this skip.
5359 The type of this skip, either @samp{function} or @samp{file}.
5360 @item Enabled or Disabled
5361 Enabled skips are marked with @samp{y}. Disabled skips are marked with @samp{n}.
5363 For function skips, this column indicates the address in memory of the function
5364 being skipped. If you've set a function skip on a function which has not yet
5365 been loaded, this field will contain @samp{<PENDING>}. Once a shared library
5366 which has the function is loaded, @code{info skip} will show the function's
5369 For file skips, this field contains the filename being skipped. For functions
5370 skips, this field contains the function name and its line number in the file
5371 where it is defined.
5375 @item skip delete @r{[}@var{range}@r{]}
5376 Delete the specified skip(s). If @var{range} is not specified, delete all
5380 @item skip enable @r{[}@var{range}@r{]}
5381 Enable the specified skip(s). If @var{range} is not specified, enable all
5384 @kindex skip disable
5385 @item skip disable @r{[}@var{range}@r{]}
5386 Disable the specified skip(s). If @var{range} is not specified, disable all
5395 A signal is an asynchronous event that can happen in a program. The
5396 operating system defines the possible kinds of signals, and gives each
5397 kind a name and a number. For example, in Unix @code{SIGINT} is the
5398 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
5399 @code{SIGSEGV} is the signal a program gets from referencing a place in
5400 memory far away from all the areas in use; @code{SIGALRM} occurs when
5401 the alarm clock timer goes off (which happens only if your program has
5402 requested an alarm).
5404 @cindex fatal signals
5405 Some signals, including @code{SIGALRM}, are a normal part of the
5406 functioning of your program. Others, such as @code{SIGSEGV}, indicate
5407 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
5408 program has not specified in advance some other way to handle the signal.
5409 @code{SIGINT} does not indicate an error in your program, but it is normally
5410 fatal so it can carry out the purpose of the interrupt: to kill the program.
5412 @value{GDBN} has the ability to detect any occurrence of a signal in your
5413 program. You can tell @value{GDBN} in advance what to do for each kind of
5416 @cindex handling signals
5417 Normally, @value{GDBN} is set up to let the non-erroneous signals like
5418 @code{SIGALRM} be silently passed to your program
5419 (so as not to interfere with their role in the program's functioning)
5420 but to stop your program immediately whenever an error signal happens.
5421 You can change these settings with the @code{handle} command.
5424 @kindex info signals
5428 Print a table of all the kinds of signals and how @value{GDBN} has been told to
5429 handle each one. You can use this to see the signal numbers of all
5430 the defined types of signals.
5432 @item info signals @var{sig}
5433 Similar, but print information only about the specified signal number.
5435 @code{info handle} is an alias for @code{info signals}.
5437 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5438 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
5439 for details about this command.
5442 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
5443 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
5444 can be the number of a signal or its name (with or without the
5445 @samp{SIG} at the beginning); a list of signal numbers of the form
5446 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
5447 known signals. Optional arguments @var{keywords}, described below,
5448 say what change to make.
5452 The keywords allowed by the @code{handle} command can be abbreviated.
5453 Their full names are:
5457 @value{GDBN} should not stop your program when this signal happens. It may
5458 still print a message telling you that the signal has come in.
5461 @value{GDBN} should stop your program when this signal happens. This implies
5462 the @code{print} keyword as well.
5465 @value{GDBN} should print a message when this signal happens.
5468 @value{GDBN} should not mention the occurrence of the signal at all. This
5469 implies the @code{nostop} keyword as well.
5473 @value{GDBN} should allow your program to see this signal; your program
5474 can handle the signal, or else it may terminate if the signal is fatal
5475 and not handled. @code{pass} and @code{noignore} are synonyms.
5479 @value{GDBN} should not allow your program to see this signal.
5480 @code{nopass} and @code{ignore} are synonyms.
5484 When a signal stops your program, the signal is not visible to the
5486 continue. Your program sees the signal then, if @code{pass} is in
5487 effect for the signal in question @emph{at that time}. In other words,
5488 after @value{GDBN} reports a signal, you can use the @code{handle}
5489 command with @code{pass} or @code{nopass} to control whether your
5490 program sees that signal when you continue.
5492 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
5493 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
5494 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
5497 You can also use the @code{signal} command to prevent your program from
5498 seeing a signal, or cause it to see a signal it normally would not see,
5499 or to give it any signal at any time. For example, if your program stopped
5500 due to some sort of memory reference error, you might store correct
5501 values into the erroneous variables and continue, hoping to see more
5502 execution; but your program would probably terminate immediately as
5503 a result of the fatal signal once it saw the signal. To prevent this,
5504 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
5507 @cindex extra signal information
5508 @anchor{extra signal information}
5510 On some targets, @value{GDBN} can inspect extra signal information
5511 associated with the intercepted signal, before it is actually
5512 delivered to the program being debugged. This information is exported
5513 by the convenience variable @code{$_siginfo}, and consists of data
5514 that is passed by the kernel to the signal handler at the time of the
5515 receipt of a signal. The data type of the information itself is
5516 target dependent. You can see the data type using the @code{ptype
5517 $_siginfo} command. On Unix systems, it typically corresponds to the
5518 standard @code{siginfo_t} type, as defined in the @file{signal.h}
5521 Here's an example, on a @sc{gnu}/Linux system, printing the stray
5522 referenced address that raised a segmentation fault.
5526 (@value{GDBP}) continue
5527 Program received signal SIGSEGV, Segmentation fault.
5528 0x0000000000400766 in main ()
5530 (@value{GDBP}) ptype $_siginfo
5537 struct @{...@} _kill;
5538 struct @{...@} _timer;
5540 struct @{...@} _sigchld;
5541 struct @{...@} _sigfault;
5542 struct @{...@} _sigpoll;
5545 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5549 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5550 $1 = (void *) 0x7ffff7ff7000
5554 Depending on target support, @code{$_siginfo} may also be writable.
5557 @section Stopping and Starting Multi-thread Programs
5559 @cindex stopped threads
5560 @cindex threads, stopped
5562 @cindex continuing threads
5563 @cindex threads, continuing
5565 @value{GDBN} supports debugging programs with multiple threads
5566 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
5567 are two modes of controlling execution of your program within the
5568 debugger. In the default mode, referred to as @dfn{all-stop mode},
5569 when any thread in your program stops (for example, at a breakpoint
5570 or while being stepped), all other threads in the program are also stopped by
5571 @value{GDBN}. On some targets, @value{GDBN} also supports
5572 @dfn{non-stop mode}, in which other threads can continue to run freely while
5573 you examine the stopped thread in the debugger.
5576 * All-Stop Mode:: All threads stop when GDB takes control
5577 * Non-Stop Mode:: Other threads continue to execute
5578 * Background Execution:: Running your program asynchronously
5579 * Thread-Specific Breakpoints:: Controlling breakpoints
5580 * Interrupted System Calls:: GDB may interfere with system calls
5581 * Observer Mode:: GDB does not alter program behavior
5585 @subsection All-Stop Mode
5587 @cindex all-stop mode
5589 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
5590 @emph{all} threads of execution stop, not just the current thread. This
5591 allows you to examine the overall state of the program, including
5592 switching between threads, without worrying that things may change
5595 Conversely, whenever you restart the program, @emph{all} threads start
5596 executing. @emph{This is true even when single-stepping} with commands
5597 like @code{step} or @code{next}.
5599 In particular, @value{GDBN} cannot single-step all threads in lockstep.
5600 Since thread scheduling is up to your debugging target's operating
5601 system (not controlled by @value{GDBN}), other threads may
5602 execute more than one statement while the current thread completes a
5603 single step. Moreover, in general other threads stop in the middle of a
5604 statement, rather than at a clean statement boundary, when the program
5607 You might even find your program stopped in another thread after
5608 continuing or even single-stepping. This happens whenever some other
5609 thread runs into a breakpoint, a signal, or an exception before the
5610 first thread completes whatever you requested.
5612 @cindex automatic thread selection
5613 @cindex switching threads automatically
5614 @cindex threads, automatic switching
5615 Whenever @value{GDBN} stops your program, due to a breakpoint or a
5616 signal, it automatically selects the thread where that breakpoint or
5617 signal happened. @value{GDBN} alerts you to the context switch with a
5618 message such as @samp{[Switching to Thread @var{n}]} to identify the
5621 On some OSes, you can modify @value{GDBN}'s default behavior by
5622 locking the OS scheduler to allow only a single thread to run.
5625 @item set scheduler-locking @var{mode}
5626 @cindex scheduler locking mode
5627 @cindex lock scheduler
5628 Set the scheduler locking mode. If it is @code{off}, then there is no
5629 locking and any thread may run at any time. If @code{on}, then only the
5630 current thread may run when the inferior is resumed. The @code{step}
5631 mode optimizes for single-stepping; it prevents other threads
5632 from preempting the current thread while you are stepping, so that
5633 the focus of debugging does not change unexpectedly.
5634 Other threads only rarely (or never) get a chance to run
5635 when you step. They are more likely to run when you @samp{next} over a
5636 function call, and they are completely free to run when you use commands
5637 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
5638 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
5639 the current thread away from the thread that you are debugging.
5641 @item show scheduler-locking
5642 Display the current scheduler locking mode.
5645 @cindex resume threads of multiple processes simultaneously
5646 By default, when you issue one of the execution commands such as
5647 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5648 threads of the current inferior to run. For example, if @value{GDBN}
5649 is attached to two inferiors, each with two threads, the
5650 @code{continue} command resumes only the two threads of the current
5651 inferior. This is useful, for example, when you debug a program that
5652 forks and you want to hold the parent stopped (so that, for instance,
5653 it doesn't run to exit), while you debug the child. In other
5654 situations, you may not be interested in inspecting the current state
5655 of any of the processes @value{GDBN} is attached to, and you may want
5656 to resume them all until some breakpoint is hit. In the latter case,
5657 you can instruct @value{GDBN} to allow all threads of all the
5658 inferiors to run with the @w{@code{set schedule-multiple}} command.
5661 @kindex set schedule-multiple
5662 @item set schedule-multiple
5663 Set the mode for allowing threads of multiple processes to be resumed
5664 when an execution command is issued. When @code{on}, all threads of
5665 all processes are allowed to run. When @code{off}, only the threads
5666 of the current process are resumed. The default is @code{off}. The
5667 @code{scheduler-locking} mode takes precedence when set to @code{on},
5668 or while you are stepping and set to @code{step}.
5670 @item show schedule-multiple
5671 Display the current mode for resuming the execution of threads of
5676 @subsection Non-Stop Mode
5678 @cindex non-stop mode
5680 @c This section is really only a place-holder, and needs to be expanded
5681 @c with more details.
5683 For some multi-threaded targets, @value{GDBN} supports an optional
5684 mode of operation in which you can examine stopped program threads in
5685 the debugger while other threads continue to execute freely. This
5686 minimizes intrusion when debugging live systems, such as programs
5687 where some threads have real-time constraints or must continue to
5688 respond to external events. This is referred to as @dfn{non-stop} mode.
5690 In non-stop mode, when a thread stops to report a debugging event,
5691 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5692 threads as well, in contrast to the all-stop mode behavior. Additionally,
5693 execution commands such as @code{continue} and @code{step} apply by default
5694 only to the current thread in non-stop mode, rather than all threads as
5695 in all-stop mode. This allows you to control threads explicitly in
5696 ways that are not possible in all-stop mode --- for example, stepping
5697 one thread while allowing others to run freely, stepping
5698 one thread while holding all others stopped, or stepping several threads
5699 independently and simultaneously.
5701 To enter non-stop mode, use this sequence of commands before you run
5702 or attach to your program:
5705 # Enable the async interface.
5708 # If using the CLI, pagination breaks non-stop.
5711 # Finally, turn it on!
5715 You can use these commands to manipulate the non-stop mode setting:
5718 @kindex set non-stop
5719 @item set non-stop on
5720 Enable selection of non-stop mode.
5721 @item set non-stop off
5722 Disable selection of non-stop mode.
5723 @kindex show non-stop
5725 Show the current non-stop enablement setting.
5728 Note these commands only reflect whether non-stop mode is enabled,
5729 not whether the currently-executing program is being run in non-stop mode.
5730 In particular, the @code{set non-stop} preference is only consulted when
5731 @value{GDBN} starts or connects to the target program, and it is generally
5732 not possible to switch modes once debugging has started. Furthermore,
5733 since not all targets support non-stop mode, even when you have enabled
5734 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5737 In non-stop mode, all execution commands apply only to the current thread
5738 by default. That is, @code{continue} only continues one thread.
5739 To continue all threads, issue @code{continue -a} or @code{c -a}.
5741 You can use @value{GDBN}'s background execution commands
5742 (@pxref{Background Execution}) to run some threads in the background
5743 while you continue to examine or step others from @value{GDBN}.
5744 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5745 always executed asynchronously in non-stop mode.
5747 Suspending execution is done with the @code{interrupt} command when
5748 running in the background, or @kbd{Ctrl-c} during foreground execution.
5749 In all-stop mode, this stops the whole process;
5750 but in non-stop mode the interrupt applies only to the current thread.
5751 To stop the whole program, use @code{interrupt -a}.
5753 Other execution commands do not currently support the @code{-a} option.
5755 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5756 that thread current, as it does in all-stop mode. This is because the
5757 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5758 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5759 changed to a different thread just as you entered a command to operate on the
5760 previously current thread.
5762 @node Background Execution
5763 @subsection Background Execution
5765 @cindex foreground execution
5766 @cindex background execution
5767 @cindex asynchronous execution
5768 @cindex execution, foreground, background and asynchronous
5770 @value{GDBN}'s execution commands have two variants: the normal
5771 foreground (synchronous) behavior, and a background
5772 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5773 the program to report that some thread has stopped before prompting for
5774 another command. In background execution, @value{GDBN} immediately gives
5775 a command prompt so that you can issue other commands while your program runs.
5777 You need to explicitly enable asynchronous mode before you can use
5778 background execution commands. You can use these commands to
5779 manipulate the asynchronous mode setting:
5782 @kindex set target-async
5783 @item set target-async on
5784 Enable asynchronous mode.
5785 @item set target-async off
5786 Disable asynchronous mode.
5787 @kindex show target-async
5788 @item show target-async
5789 Show the current target-async setting.
5792 If the target doesn't support async mode, @value{GDBN} issues an error
5793 message if you attempt to use the background execution commands.
5795 To specify background execution, add a @code{&} to the command. For example,
5796 the background form of the @code{continue} command is @code{continue&}, or
5797 just @code{c&}. The execution commands that accept background execution
5803 @xref{Starting, , Starting your Program}.
5807 @xref{Attach, , Debugging an Already-running Process}.
5811 @xref{Continuing and Stepping, step}.
5815 @xref{Continuing and Stepping, stepi}.
5819 @xref{Continuing and Stepping, next}.
5823 @xref{Continuing and Stepping, nexti}.
5827 @xref{Continuing and Stepping, continue}.
5831 @xref{Continuing and Stepping, finish}.
5835 @xref{Continuing and Stepping, until}.
5839 Background execution is especially useful in conjunction with non-stop
5840 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5841 However, you can also use these commands in the normal all-stop mode with
5842 the restriction that you cannot issue another execution command until the
5843 previous one finishes. Examples of commands that are valid in all-stop
5844 mode while the program is running include @code{help} and @code{info break}.
5846 You can interrupt your program while it is running in the background by
5847 using the @code{interrupt} command.
5854 Suspend execution of the running program. In all-stop mode,
5855 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5856 only the current thread. To stop the whole program in non-stop mode,
5857 use @code{interrupt -a}.
5860 @node Thread-Specific Breakpoints
5861 @subsection Thread-Specific Breakpoints
5863 When your program has multiple threads (@pxref{Threads,, Debugging
5864 Programs with Multiple Threads}), you can choose whether to set
5865 breakpoints on all threads, or on a particular thread.
5868 @cindex breakpoints and threads
5869 @cindex thread breakpoints
5870 @kindex break @dots{} thread @var{threadno}
5871 @item break @var{linespec} thread @var{threadno}
5872 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5873 @var{linespec} specifies source lines; there are several ways of
5874 writing them (@pxref{Specify Location}), but the effect is always to
5875 specify some source line.
5877 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5878 to specify that you only want @value{GDBN} to stop the program when a
5879 particular thread reaches this breakpoint. @var{threadno} is one of the
5880 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5881 column of the @samp{info threads} display.
5883 If you do not specify @samp{thread @var{threadno}} when you set a
5884 breakpoint, the breakpoint applies to @emph{all} threads of your
5887 You can use the @code{thread} qualifier on conditional breakpoints as
5888 well; in this case, place @samp{thread @var{threadno}} before or
5889 after the breakpoint condition, like this:
5892 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5897 Thread-specific breakpoints are automatically deleted when
5898 @value{GDBN} detects the corresponding thread is no longer in the
5899 thread list. For example:
5903 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
5906 There are several ways for a thread to disappear, such as a regular
5907 thread exit, but also when you detach from the process with the
5908 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
5909 Process}), or if @value{GDBN} loses the remote connection
5910 (@pxref{Remote Debugging}), etc. Note that with some targets,
5911 @value{GDBN} is only able to detect a thread has exited when the user
5912 explictly asks for the thread list with the @code{info threads}
5915 @node Interrupted System Calls
5916 @subsection Interrupted System Calls
5918 @cindex thread breakpoints and system calls
5919 @cindex system calls and thread breakpoints
5920 @cindex premature return from system calls
5921 There is an unfortunate side effect when using @value{GDBN} to debug
5922 multi-threaded programs. If one thread stops for a
5923 breakpoint, or for some other reason, and another thread is blocked in a
5924 system call, then the system call may return prematurely. This is a
5925 consequence of the interaction between multiple threads and the signals
5926 that @value{GDBN} uses to implement breakpoints and other events that
5929 To handle this problem, your program should check the return value of
5930 each system call and react appropriately. This is good programming
5933 For example, do not write code like this:
5939 The call to @code{sleep} will return early if a different thread stops
5940 at a breakpoint or for some other reason.
5942 Instead, write this:
5947 unslept = sleep (unslept);
5950 A system call is allowed to return early, so the system is still
5951 conforming to its specification. But @value{GDBN} does cause your
5952 multi-threaded program to behave differently than it would without
5955 Also, @value{GDBN} uses internal breakpoints in the thread library to
5956 monitor certain events such as thread creation and thread destruction.
5957 When such an event happens, a system call in another thread may return
5958 prematurely, even though your program does not appear to stop.
5961 @subsection Observer Mode
5963 If you want to build on non-stop mode and observe program behavior
5964 without any chance of disruption by @value{GDBN}, you can set
5965 variables to disable all of the debugger's attempts to modify state,
5966 whether by writing memory, inserting breakpoints, etc. These operate
5967 at a low level, intercepting operations from all commands.
5969 When all of these are set to @code{off}, then @value{GDBN} is said to
5970 be @dfn{observer mode}. As a convenience, the variable
5971 @code{observer} can be set to disable these, plus enable non-stop
5974 Note that @value{GDBN} will not prevent you from making nonsensical
5975 combinations of these settings. For instance, if you have enabled
5976 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
5977 then breakpoints that work by writing trap instructions into the code
5978 stream will still not be able to be placed.
5983 @item set observer on
5984 @itemx set observer off
5985 When set to @code{on}, this disables all the permission variables
5986 below (except for @code{insert-fast-tracepoints}), plus enables
5987 non-stop debugging. Setting this to @code{off} switches back to
5988 normal debugging, though remaining in non-stop mode.
5991 Show whether observer mode is on or off.
5993 @kindex may-write-registers
5994 @item set may-write-registers on
5995 @itemx set may-write-registers off
5996 This controls whether @value{GDBN} will attempt to alter the values of
5997 registers, such as with assignment expressions in @code{print}, or the
5998 @code{jump} command. It defaults to @code{on}.
6000 @item show may-write-registers
6001 Show the current permission to write registers.
6003 @kindex may-write-memory
6004 @item set may-write-memory on
6005 @itemx set may-write-memory off
6006 This controls whether @value{GDBN} will attempt to alter the contents
6007 of memory, such as with assignment expressions in @code{print}. It
6008 defaults to @code{on}.
6010 @item show may-write-memory
6011 Show the current permission to write memory.
6013 @kindex may-insert-breakpoints
6014 @item set may-insert-breakpoints on
6015 @itemx set may-insert-breakpoints off
6016 This controls whether @value{GDBN} will attempt to insert breakpoints.
6017 This affects all breakpoints, including internal breakpoints defined
6018 by @value{GDBN}. It defaults to @code{on}.
6020 @item show may-insert-breakpoints
6021 Show the current permission to insert breakpoints.
6023 @kindex may-insert-tracepoints
6024 @item set may-insert-tracepoints on
6025 @itemx set may-insert-tracepoints off
6026 This controls whether @value{GDBN} will attempt to insert (regular)
6027 tracepoints at the beginning of a tracing experiment. It affects only
6028 non-fast tracepoints, fast tracepoints being under the control of
6029 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6031 @item show may-insert-tracepoints
6032 Show the current permission to insert tracepoints.
6034 @kindex may-insert-fast-tracepoints
6035 @item set may-insert-fast-tracepoints on
6036 @itemx set may-insert-fast-tracepoints off
6037 This controls whether @value{GDBN} will attempt to insert fast
6038 tracepoints at the beginning of a tracing experiment. It affects only
6039 fast tracepoints, regular (non-fast) tracepoints being under the
6040 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6042 @item show may-insert-fast-tracepoints
6043 Show the current permission to insert fast tracepoints.
6045 @kindex may-interrupt
6046 @item set may-interrupt on
6047 @itemx set may-interrupt off
6048 This controls whether @value{GDBN} will attempt to interrupt or stop
6049 program execution. When this variable is @code{off}, the
6050 @code{interrupt} command will have no effect, nor will
6051 @kbd{Ctrl-c}. It defaults to @code{on}.
6053 @item show may-interrupt
6054 Show the current permission to interrupt or stop the program.
6058 @node Reverse Execution
6059 @chapter Running programs backward
6060 @cindex reverse execution
6061 @cindex running programs backward
6063 When you are debugging a program, it is not unusual to realize that
6064 you have gone too far, and some event of interest has already happened.
6065 If the target environment supports it, @value{GDBN} can allow you to
6066 ``rewind'' the program by running it backward.
6068 A target environment that supports reverse execution should be able
6069 to ``undo'' the changes in machine state that have taken place as the
6070 program was executing normally. Variables, registers etc.@: should
6071 revert to their previous values. Obviously this requires a great
6072 deal of sophistication on the part of the target environment; not
6073 all target environments can support reverse execution.
6075 When a program is executed in reverse, the instructions that
6076 have most recently been executed are ``un-executed'', in reverse
6077 order. The program counter runs backward, following the previous
6078 thread of execution in reverse. As each instruction is ``un-executed'',
6079 the values of memory and/or registers that were changed by that
6080 instruction are reverted to their previous states. After executing
6081 a piece of source code in reverse, all side effects of that code
6082 should be ``undone'', and all variables should be returned to their
6083 prior values@footnote{
6084 Note that some side effects are easier to undo than others. For instance,
6085 memory and registers are relatively easy, but device I/O is hard. Some
6086 targets may be able undo things like device I/O, and some may not.
6088 The contract between @value{GDBN} and the reverse executing target
6089 requires only that the target do something reasonable when
6090 @value{GDBN} tells it to execute backwards, and then report the
6091 results back to @value{GDBN}. Whatever the target reports back to
6092 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6093 assumes that the memory and registers that the target reports are in a
6094 consistant state, but @value{GDBN} accepts whatever it is given.
6097 If you are debugging in a target environment that supports
6098 reverse execution, @value{GDBN} provides the following commands.
6101 @kindex reverse-continue
6102 @kindex rc @r{(@code{reverse-continue})}
6103 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6104 @itemx rc @r{[}@var{ignore-count}@r{]}
6105 Beginning at the point where your program last stopped, start executing
6106 in reverse. Reverse execution will stop for breakpoints and synchronous
6107 exceptions (signals), just like normal execution. Behavior of
6108 asynchronous signals depends on the target environment.
6110 @kindex reverse-step
6111 @kindex rs @r{(@code{step})}
6112 @item reverse-step @r{[}@var{count}@r{]}
6113 Run the program backward until control reaches the start of a
6114 different source line; then stop it, and return control to @value{GDBN}.
6116 Like the @code{step} command, @code{reverse-step} will only stop
6117 at the beginning of a source line. It ``un-executes'' the previously
6118 executed source line. If the previous source line included calls to
6119 debuggable functions, @code{reverse-step} will step (backward) into
6120 the called function, stopping at the beginning of the @emph{last}
6121 statement in the called function (typically a return statement).
6123 Also, as with the @code{step} command, if non-debuggable functions are
6124 called, @code{reverse-step} will run thru them backward without stopping.
6126 @kindex reverse-stepi
6127 @kindex rsi @r{(@code{reverse-stepi})}
6128 @item reverse-stepi @r{[}@var{count}@r{]}
6129 Reverse-execute one machine instruction. Note that the instruction
6130 to be reverse-executed is @emph{not} the one pointed to by the program
6131 counter, but the instruction executed prior to that one. For instance,
6132 if the last instruction was a jump, @code{reverse-stepi} will take you
6133 back from the destination of the jump to the jump instruction itself.
6135 @kindex reverse-next
6136 @kindex rn @r{(@code{reverse-next})}
6137 @item reverse-next @r{[}@var{count}@r{]}
6138 Run backward to the beginning of the previous line executed in
6139 the current (innermost) stack frame. If the line contains function
6140 calls, they will be ``un-executed'' without stopping. Starting from
6141 the first line of a function, @code{reverse-next} will take you back
6142 to the caller of that function, @emph{before} the function was called,
6143 just as the normal @code{next} command would take you from the last
6144 line of a function back to its return to its caller
6145 @footnote{Unless the code is too heavily optimized.}.
6147 @kindex reverse-nexti
6148 @kindex rni @r{(@code{reverse-nexti})}
6149 @item reverse-nexti @r{[}@var{count}@r{]}
6150 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6151 in reverse, except that called functions are ``un-executed'' atomically.
6152 That is, if the previously executed instruction was a return from
6153 another function, @code{reverse-nexti} will continue to execute
6154 in reverse until the call to that function (from the current stack
6157 @kindex reverse-finish
6158 @item reverse-finish
6159 Just as the @code{finish} command takes you to the point where the
6160 current function returns, @code{reverse-finish} takes you to the point
6161 where it was called. Instead of ending up at the end of the current
6162 function invocation, you end up at the beginning.
6164 @kindex set exec-direction
6165 @item set exec-direction
6166 Set the direction of target execution.
6167 @item set exec-direction reverse
6168 @cindex execute forward or backward in time
6169 @value{GDBN} will perform all execution commands in reverse, until the
6170 exec-direction mode is changed to ``forward''. Affected commands include
6171 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6172 command cannot be used in reverse mode.
6173 @item set exec-direction forward
6174 @value{GDBN} will perform all execution commands in the normal fashion.
6175 This is the default.
6179 @node Process Record and Replay
6180 @chapter Recording Inferior's Execution and Replaying It
6181 @cindex process record and replay
6182 @cindex recording inferior's execution and replaying it
6184 On some platforms, @value{GDBN} provides a special @dfn{process record
6185 and replay} target that can record a log of the process execution, and
6186 replay it later with both forward and reverse execution commands.
6189 When this target is in use, if the execution log includes the record
6190 for the next instruction, @value{GDBN} will debug in @dfn{replay
6191 mode}. In the replay mode, the inferior does not really execute code
6192 instructions. Instead, all the events that normally happen during
6193 code execution are taken from the execution log. While code is not
6194 really executed in replay mode, the values of registers (including the
6195 program counter register) and the memory of the inferior are still
6196 changed as they normally would. Their contents are taken from the
6200 If the record for the next instruction is not in the execution log,
6201 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6202 inferior executes normally, and @value{GDBN} records the execution log
6205 The process record and replay target supports reverse execution
6206 (@pxref{Reverse Execution}), even if the platform on which the
6207 inferior runs does not. However, the reverse execution is limited in
6208 this case by the range of the instructions recorded in the execution
6209 log. In other words, reverse execution on platforms that don't
6210 support it directly can only be done in the replay mode.
6212 When debugging in the reverse direction, @value{GDBN} will work in
6213 replay mode as long as the execution log includes the record for the
6214 previous instruction; otherwise, it will work in record mode, if the
6215 platform supports reverse execution, or stop if not.
6217 For architecture environments that support process record and replay,
6218 @value{GDBN} provides the following commands:
6221 @kindex target record
6222 @kindex target record-full
6223 @kindex target record-btrace
6226 @kindex record btrace
6230 @item record @var{method}
6231 This command starts the process record and replay target. The
6232 recording method can be specified as parameter. Without a parameter
6233 the command uses the @code{full} recording method. The following
6234 recording methods are available:
6238 Full record/replay recording using @value{GDBN}'s software record and
6239 replay implementation. This method allows replaying and reverse
6243 Hardware-supported instruction recording. This method does not allow
6244 replaying and reverse execution.
6246 This recording method may not be available on all processors.
6249 The process record and replay target can only debug a process that is
6250 already running. Therefore, you need first to start the process with
6251 the @kbd{run} or @kbd{start} commands, and then start the recording
6252 with the @kbd{record @var{method}} command.
6254 Both @code{record @var{method}} and @code{rec @var{method}} are
6255 aliases of @code{target record-@var{method}}.
6257 @cindex displaced stepping, and process record and replay
6258 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6259 will be automatically disabled when process record and replay target
6260 is started. That's because the process record and replay target
6261 doesn't support displaced stepping.
6263 @cindex non-stop mode, and process record and replay
6264 @cindex asynchronous execution, and process record and replay
6265 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6266 the asynchronous execution mode (@pxref{Background Execution}), not
6267 all recording methods are available. The @code{full} recording method
6268 does not support these two modes.
6273 Stop the process record and replay target. When process record and
6274 replay target stops, the entire execution log will be deleted and the
6275 inferior will either be terminated, or will remain in its final state.
6277 When you stop the process record and replay target in record mode (at
6278 the end of the execution log), the inferior will be stopped at the
6279 next instruction that would have been recorded. In other words, if
6280 you record for a while and then stop recording, the inferior process
6281 will be left in the same state as if the recording never happened.
6283 On the other hand, if the process record and replay target is stopped
6284 while in replay mode (that is, not at the end of the execution log,
6285 but at some earlier point), the inferior process will become ``live''
6286 at that earlier state, and it will then be possible to continue the
6287 usual ``live'' debugging of the process from that state.
6289 When the inferior process exits, or @value{GDBN} detaches from it,
6290 process record and replay target will automatically stop itself.
6294 Go to a specific location in the execution log. There are several
6295 ways to specify the location to go to:
6298 @item record goto begin
6299 @itemx record goto start
6300 Go to the beginning of the execution log.
6302 @item record goto end
6303 Go to the end of the execution log.
6305 @item record goto @var{n}
6306 Go to instruction number @var{n} in the execution log.
6310 @item record save @var{filename}
6311 Save the execution log to a file @file{@var{filename}}.
6312 Default filename is @file{gdb_record.@var{process_id}}, where
6313 @var{process_id} is the process ID of the inferior.
6315 This command may not be available for all recording methods.
6317 @kindex record restore
6318 @item record restore @var{filename}
6319 Restore the execution log from a file @file{@var{filename}}.
6320 File must have been created with @code{record save}.
6322 @kindex set record full
6323 @item set record full insn-number-max @var{limit}
6324 @itemx set record full insn-number-max unlimited
6325 Set the limit of instructions to be recorded for the @code{full}
6326 recording method. Default value is 200000.
6328 If @var{limit} is a positive number, then @value{GDBN} will start
6329 deleting instructions from the log once the number of the record
6330 instructions becomes greater than @var{limit}. For every new recorded
6331 instruction, @value{GDBN} will delete the earliest recorded
6332 instruction to keep the number of recorded instructions at the limit.
6333 (Since deleting recorded instructions loses information, @value{GDBN}
6334 lets you control what happens when the limit is reached, by means of
6335 the @code{stop-at-limit} option, described below.)
6337 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
6338 delete recorded instructions from the execution log. The number of
6339 recorded instructions is limited only by the available memory.
6341 @kindex show record full
6342 @item show record full insn-number-max
6343 Show the limit of instructions to be recorded with the @code{full}
6346 @item set record full stop-at-limit
6347 Control the behavior of the @code{full} recording method when the
6348 number of recorded instructions reaches the limit. If ON (the
6349 default), @value{GDBN} will stop when the limit is reached for the
6350 first time and ask you whether you want to stop the inferior or
6351 continue running it and recording the execution log. If you decide
6352 to continue recording, each new recorded instruction will cause the
6353 oldest one to be deleted.
6355 If this option is OFF, @value{GDBN} will automatically delete the
6356 oldest record to make room for each new one, without asking.
6358 @item show record full stop-at-limit
6359 Show the current setting of @code{stop-at-limit}.
6361 @item set record full memory-query
6362 Control the behavior when @value{GDBN} is unable to record memory
6363 changes caused by an instruction for the @code{full} recording method.
6364 If ON, @value{GDBN} will query whether to stop the inferior in that
6367 If this option is OFF (the default), @value{GDBN} will automatically
6368 ignore the effect of such instructions on memory. Later, when
6369 @value{GDBN} replays this execution log, it will mark the log of this
6370 instruction as not accessible, and it will not affect the replay
6373 @item show record full memory-query
6374 Show the current setting of @code{memory-query}.
6378 Show various statistics about the recording depending on the recording
6383 For the @code{full} recording method, it shows the state of process
6384 record and its in-memory execution log buffer, including:
6388 Whether in record mode or replay mode.
6390 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
6392 Highest recorded instruction number.
6394 Current instruction about to be replayed (if in replay mode).
6396 Number of instructions contained in the execution log.
6398 Maximum number of instructions that may be contained in the execution log.
6402 For the @code{btrace} recording method, it shows the number of
6403 instructions that have been recorded and the number of blocks of
6404 sequential control-flow that is formed by the recorded instructions.
6407 @kindex record delete
6410 When record target runs in replay mode (``in the past''), delete the
6411 subsequent execution log and begin to record a new execution log starting
6412 from the current address. This means you will abandon the previously
6413 recorded ``future'' and begin recording a new ``future''.
6415 @kindex record instruction-history
6416 @kindex rec instruction-history
6417 @item record instruction-history
6418 Disassembles instructions from the recorded execution log. By
6419 default, ten instructions are disassembled. This can be changed using
6420 the @code{set record instruction-history-size} command. Instructions
6421 are printed in execution order. There are several ways to specify
6422 what part of the execution log to disassemble:
6425 @item record instruction-history @var{insn}
6426 Disassembles ten instructions starting from instruction number
6429 @item record instruction-history @var{insn}, +/-@var{n}
6430 Disassembles @var{n} instructions around instruction number
6431 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
6432 @var{n} instructions after instruction number @var{insn}. If
6433 @var{n} is preceded with @code{-}, disassembles @var{n}
6434 instructions before instruction number @var{insn}.
6436 @item record instruction-history
6437 Disassembles ten more instructions after the last disassembly.
6439 @item record instruction-history -
6440 Disassembles ten more instructions before the last disassembly.
6442 @item record instruction-history @var{begin} @var{end}
6443 Disassembles instructions beginning with instruction number
6444 @var{begin} until instruction number @var{end}. The instruction
6445 number @var{end} is not included.
6448 This command may not be available for all recording methods.
6451 @item set record instruction-history-size @var{size}
6452 @itemx set record instruction-history-size unlimited
6453 Define how many instructions to disassemble in the @code{record
6454 instruction-history} command. The default value is 10.
6455 A @var{size} of @code{unlimited} means unlimited instructions.
6458 @item show record instruction-history-size
6459 Show how many instructions to disassemble in the @code{record
6460 instruction-history} command.
6462 @kindex record function-call-history
6463 @kindex rec function-call-history
6464 @item record function-call-history
6465 Prints the execution history at function granularity. It prints one
6466 line for each sequence of instructions that belong to the same
6467 function giving the name of that function, the source lines
6468 for this instruction sequence (if the @code{/l} modifier is
6469 specified), and the instructions numbers that form the sequence (if
6470 the @code{/i} modifier is specified).
6473 (@value{GDBP}) @b{list 1, 10}
6484 (@value{GDBP}) @b{record function-call-history /l}
6490 By default, ten lines are printed. This can be changed using the
6491 @code{set record function-call-history-size} command. Functions are
6492 printed in execution order. There are several ways to specify what
6496 @item record function-call-history @var{func}
6497 Prints ten functions starting from function number @var{func}.
6499 @item record function-call-history @var{func}, +/-@var{n}
6500 Prints @var{n} functions around function number @var{func}. If
6501 @var{n} is preceded with @code{+}, prints @var{n} functions after
6502 function number @var{func}. If @var{n} is preceded with @code{-},
6503 prints @var{n} functions before function number @var{func}.
6505 @item record function-call-history
6506 Prints ten more functions after the last ten-line print.
6508 @item record function-call-history -
6509 Prints ten more functions before the last ten-line print.
6511 @item record function-call-history @var{begin} @var{end}
6512 Prints functions beginning with function number @var{begin} until
6513 function number @var{end}. The function number @var{end} is not
6517 This command may not be available for all recording methods.
6519 @item set record function-call-history-size @var{size}
6520 @itemx set record function-call-history-size unlimited
6521 Define how many lines to print in the
6522 @code{record function-call-history} command. The default value is 10.
6523 A size of @code{unlimited} means unlimited lines.
6525 @item show record function-call-history-size
6526 Show how many lines to print in the
6527 @code{record function-call-history} command.
6532 @chapter Examining the Stack
6534 When your program has stopped, the first thing you need to know is where it
6535 stopped and how it got there.
6538 Each time your program performs a function call, information about the call
6540 That information includes the location of the call in your program,
6541 the arguments of the call,
6542 and the local variables of the function being called.
6543 The information is saved in a block of data called a @dfn{stack frame}.
6544 The stack frames are allocated in a region of memory called the @dfn{call
6547 When your program stops, the @value{GDBN} commands for examining the
6548 stack allow you to see all of this information.
6550 @cindex selected frame
6551 One of the stack frames is @dfn{selected} by @value{GDBN} and many
6552 @value{GDBN} commands refer implicitly to the selected frame. In
6553 particular, whenever you ask @value{GDBN} for the value of a variable in
6554 your program, the value is found in the selected frame. There are
6555 special @value{GDBN} commands to select whichever frame you are
6556 interested in. @xref{Selection, ,Selecting a Frame}.
6558 When your program stops, @value{GDBN} automatically selects the
6559 currently executing frame and describes it briefly, similar to the
6560 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
6563 * Frames:: Stack frames
6564 * Backtrace:: Backtraces
6565 * Frame Filter Management:: Managing frame filters
6566 * Selection:: Selecting a frame
6567 * Frame Info:: Information on a frame
6572 @section Stack Frames
6574 @cindex frame, definition
6576 The call stack is divided up into contiguous pieces called @dfn{stack
6577 frames}, or @dfn{frames} for short; each frame is the data associated
6578 with one call to one function. The frame contains the arguments given
6579 to the function, the function's local variables, and the address at
6580 which the function is executing.
6582 @cindex initial frame
6583 @cindex outermost frame
6584 @cindex innermost frame
6585 When your program is started, the stack has only one frame, that of the
6586 function @code{main}. This is called the @dfn{initial} frame or the
6587 @dfn{outermost} frame. Each time a function is called, a new frame is
6588 made. Each time a function returns, the frame for that function invocation
6589 is eliminated. If a function is recursive, there can be many frames for
6590 the same function. The frame for the function in which execution is
6591 actually occurring is called the @dfn{innermost} frame. This is the most
6592 recently created of all the stack frames that still exist.
6594 @cindex frame pointer
6595 Inside your program, stack frames are identified by their addresses. A
6596 stack frame consists of many bytes, each of which has its own address; each
6597 kind of computer has a convention for choosing one byte whose
6598 address serves as the address of the frame. Usually this address is kept
6599 in a register called the @dfn{frame pointer register}
6600 (@pxref{Registers, $fp}) while execution is going on in that frame.
6602 @cindex frame number
6603 @value{GDBN} assigns numbers to all existing stack frames, starting with
6604 zero for the innermost frame, one for the frame that called it,
6605 and so on upward. These numbers do not really exist in your program;
6606 they are assigned by @value{GDBN} to give you a way of designating stack
6607 frames in @value{GDBN} commands.
6609 @c The -fomit-frame-pointer below perennially causes hbox overflow
6610 @c underflow problems.
6611 @cindex frameless execution
6612 Some compilers provide a way to compile functions so that they operate
6613 without stack frames. (For example, the @value{NGCC} option
6615 @samp{-fomit-frame-pointer}
6617 generates functions without a frame.)
6618 This is occasionally done with heavily used library functions to save
6619 the frame setup time. @value{GDBN} has limited facilities for dealing
6620 with these function invocations. If the innermost function invocation
6621 has no stack frame, @value{GDBN} nevertheless regards it as though
6622 it had a separate frame, which is numbered zero as usual, allowing
6623 correct tracing of the function call chain. However, @value{GDBN} has
6624 no provision for frameless functions elsewhere in the stack.
6627 @kindex frame@r{, command}
6628 @cindex current stack frame
6629 @item frame @var{args}
6630 The @code{frame} command allows you to move from one stack frame to another,
6631 and to print the stack frame you select. @var{args} may be either the
6632 address of the frame or the stack frame number. Without an argument,
6633 @code{frame} prints the current stack frame.
6635 @kindex select-frame
6636 @cindex selecting frame silently
6638 The @code{select-frame} command allows you to move from one stack frame
6639 to another without printing the frame. This is the silent version of
6647 @cindex call stack traces
6648 A backtrace is a summary of how your program got where it is. It shows one
6649 line per frame, for many frames, starting with the currently executing
6650 frame (frame zero), followed by its caller (frame one), and on up the
6653 @anchor{backtrace-command}
6656 @kindex bt @r{(@code{backtrace})}
6659 Print a backtrace of the entire stack: one line per frame for all
6660 frames in the stack.
6662 You can stop the backtrace at any time by typing the system interrupt
6663 character, normally @kbd{Ctrl-c}.
6665 @item backtrace @var{n}
6667 Similar, but print only the innermost @var{n} frames.
6669 @item backtrace -@var{n}
6671 Similar, but print only the outermost @var{n} frames.
6673 @item backtrace full
6675 @itemx bt full @var{n}
6676 @itemx bt full -@var{n}
6677 Print the values of the local variables also. @var{n} specifies the
6678 number of frames to print, as described above.
6680 @item backtrace no-filters
6681 @itemx bt no-filters
6682 @itemx bt no-filters @var{n}
6683 @itemx bt no-filters -@var{n}
6684 @itemx bt no-filters full
6685 @itemx bt no-filters full @var{n}
6686 @itemx bt no-filters full -@var{n}
6687 Do not run Python frame filters on this backtrace. @xref{Frame
6688 Filter API}, for more information. Additionally use @ref{disable
6689 frame-filter all} to turn off all frame filters. This is only
6690 relevant when @value{GDBN} has been configured with @code{Python}
6696 The names @code{where} and @code{info stack} (abbreviated @code{info s})
6697 are additional aliases for @code{backtrace}.
6699 @cindex multiple threads, backtrace
6700 In a multi-threaded program, @value{GDBN} by default shows the
6701 backtrace only for the current thread. To display the backtrace for
6702 several or all of the threads, use the command @code{thread apply}
6703 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
6704 apply all backtrace}, @value{GDBN} will display the backtrace for all
6705 the threads; this is handy when you debug a core dump of a
6706 multi-threaded program.
6708 Each line in the backtrace shows the frame number and the function name.
6709 The program counter value is also shown---unless you use @code{set
6710 print address off}. The backtrace also shows the source file name and
6711 line number, as well as the arguments to the function. The program
6712 counter value is omitted if it is at the beginning of the code for that
6715 Here is an example of a backtrace. It was made with the command
6716 @samp{bt 3}, so it shows the innermost three frames.
6720 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6722 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
6723 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
6725 (More stack frames follow...)
6730 The display for frame zero does not begin with a program counter
6731 value, indicating that your program has stopped at the beginning of the
6732 code for line @code{993} of @code{builtin.c}.
6735 The value of parameter @code{data} in frame 1 has been replaced by
6736 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
6737 only if it is a scalar (integer, pointer, enumeration, etc). See command
6738 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
6739 on how to configure the way function parameter values are printed.
6741 @cindex optimized out, in backtrace
6742 @cindex function call arguments, optimized out
6743 If your program was compiled with optimizations, some compilers will
6744 optimize away arguments passed to functions if those arguments are
6745 never used after the call. Such optimizations generate code that
6746 passes arguments through registers, but doesn't store those arguments
6747 in the stack frame. @value{GDBN} has no way of displaying such
6748 arguments in stack frames other than the innermost one. Here's what
6749 such a backtrace might look like:
6753 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6755 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
6756 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
6758 (More stack frames follow...)
6763 The values of arguments that were not saved in their stack frames are
6764 shown as @samp{<optimized out>}.
6766 If you need to display the values of such optimized-out arguments,
6767 either deduce that from other variables whose values depend on the one
6768 you are interested in, or recompile without optimizations.
6770 @cindex backtrace beyond @code{main} function
6771 @cindex program entry point
6772 @cindex startup code, and backtrace
6773 Most programs have a standard user entry point---a place where system
6774 libraries and startup code transition into user code. For C this is
6775 @code{main}@footnote{
6776 Note that embedded programs (the so-called ``free-standing''
6777 environment) are not required to have a @code{main} function as the
6778 entry point. They could even have multiple entry points.}.
6779 When @value{GDBN} finds the entry function in a backtrace
6780 it will terminate the backtrace, to avoid tracing into highly
6781 system-specific (and generally uninteresting) code.
6783 If you need to examine the startup code, or limit the number of levels
6784 in a backtrace, you can change this behavior:
6787 @item set backtrace past-main
6788 @itemx set backtrace past-main on
6789 @kindex set backtrace
6790 Backtraces will continue past the user entry point.
6792 @item set backtrace past-main off
6793 Backtraces will stop when they encounter the user entry point. This is the
6796 @item show backtrace past-main
6797 @kindex show backtrace
6798 Display the current user entry point backtrace policy.
6800 @item set backtrace past-entry
6801 @itemx set backtrace past-entry on
6802 Backtraces will continue past the internal entry point of an application.
6803 This entry point is encoded by the linker when the application is built,
6804 and is likely before the user entry point @code{main} (or equivalent) is called.
6806 @item set backtrace past-entry off
6807 Backtraces will stop when they encounter the internal entry point of an
6808 application. This is the default.
6810 @item show backtrace past-entry
6811 Display the current internal entry point backtrace policy.
6813 @item set backtrace limit @var{n}
6814 @itemx set backtrace limit 0
6815 @itemx set backtrace limit unlimited
6816 @cindex backtrace limit
6817 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
6818 or zero means unlimited levels.
6820 @item show backtrace limit
6821 Display the current limit on backtrace levels.
6824 You can control how file names are displayed.
6827 @item set filename-display
6828 @itemx set filename-display relative
6829 @cindex filename-display
6830 Display file names relative to the compilation directory. This is the default.
6832 @item set filename-display basename
6833 Display only basename of a filename.
6835 @item set filename-display absolute
6836 Display an absolute filename.
6838 @item show filename-display
6839 Show the current way to display filenames.
6842 @node Frame Filter Management
6843 @section Management of Frame Filters.
6844 @cindex managing frame filters
6846 Frame filters are Python based utilities to manage and decorate the
6847 output of frames. @xref{Frame Filter API}, for further information.
6849 Managing frame filters is performed by several commands available
6850 within @value{GDBN}, detailed here.
6853 @kindex info frame-filter
6854 @item info frame-filter
6855 Print a list of installed frame filters from all dictionaries, showing
6856 their name, priority and enabled status.
6858 @kindex disable frame-filter
6859 @anchor{disable frame-filter all}
6860 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
6861 Disable a frame filter in the dictionary matching
6862 @var{filter-dictionary}, or @code{all}, and @var{filter-name}.
6863 @var{filter-dictionary} may be @code{all}, @code{global},
6864 @code{progspace} or the name of the object file where the frame filter
6865 dictionary resides. When @code{all} is specified, all frame filters
6866 across all dictionaries are disabled. @var{filter-name} is the name
6867 of the frame filter and is used when @code{all} is not the option for
6868 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
6869 may be enabled again later.
6871 @kindex enable frame-filter
6872 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
6873 Enable a frame filter in the dictionary matching
6874 @var{filter-dictionary}, or @code{all}, and @var{filter-name}.
6875 @var{filter-dictionary} may be @code{all}, @code{global},
6876 @code{progspace} or the name of the object file where the frame filter
6877 dictionary resides. When @code{all} is specified, all frame filters across
6878 all dictionaries are enabled. @var{filter-name} is the name of the frame
6879 filter and is used when @code{all} is not the option for
6880 @var{filter-dictionary}.
6885 (gdb) info frame-filter
6887 global frame-filters:
6888 Priority Enabled Name
6889 1000 No PrimaryFunctionFilter
6892 progspace /build/test frame-filters:
6893 Priority Enabled Name
6894 100 Yes ProgspaceFilter
6896 objfile /build/test frame-filters:
6897 Priority Enabled Name
6898 999 Yes BuildProgra Filter
6900 (gdb) disable frame-filter /build/test BuildProgramFilter
6901 (gdb) info frame-filter
6903 global frame-filters:
6904 Priority Enabled Name
6905 1000 No PrimaryFunctionFilter
6908 progspace /build/test frame-filters:
6909 Priority Enabled Name
6910 100 Yes ProgspaceFilter
6912 objfile /build/test frame-filters:
6913 Priority Enabled Name
6914 999 No BuildProgramFilter
6916 (gdb) enable frame-filter global PrimaryFunctionFilter
6917 (gdb) info frame-filter
6919 global frame-filters:
6920 Priority Enabled Name
6921 1000 Yes PrimaryFunctionFilter
6924 progspace /build/test frame-filters:
6925 Priority Enabled Name
6926 100 Yes ProgspaceFilter
6928 objfile /build/test frame-filters:
6929 Priority Enabled Name
6930 999 No BuildProgramFilter
6933 @kindex set frame-filter priority
6934 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
6935 Set the @var{priority} of a frame filter in the dictionary matching
6936 @var{filter-dictionary}, and the frame filter name matching
6937 @var{filter-name}. @var{filter-dictionary} may be @code{global},
6938 @code{progspace} or the name of the object file where the frame filter
6939 dictionary resides. @var{priority} is an integer.
6941 @kindex show frame-filter priority
6942 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
6943 Show the @var{priority} of a frame filter in the dictionary matching
6944 @var{filter-dictionary}, and the frame filter name matching
6945 @var{filter-name}. @var{filter-dictionary} may be @code{global},
6946 @code{progspace} or the name of the object file where the frame filter
6952 (gdb) info frame-filter
6954 global frame-filters:
6955 Priority Enabled Name
6956 1000 Yes PrimaryFunctionFilter
6959 progspace /build/test frame-filters:
6960 Priority Enabled Name
6961 100 Yes ProgspaceFilter
6963 objfile /build/test frame-filters:
6964 Priority Enabled Name
6965 999 No BuildProgramFilter
6967 (gdb) set frame-filter priority global Reverse 50
6968 (gdb) info frame-filter
6970 global frame-filters:
6971 Priority Enabled Name
6972 1000 Yes PrimaryFunctionFilter
6975 progspace /build/test frame-filters:
6976 Priority Enabled Name
6977 100 Yes ProgspaceFilter
6979 objfile /build/test frame-filters:
6980 Priority Enabled Name
6981 999 No BuildProgramFilter
6986 @section Selecting a Frame
6988 Most commands for examining the stack and other data in your program work on
6989 whichever stack frame is selected at the moment. Here are the commands for
6990 selecting a stack frame; all of them finish by printing a brief description
6991 of the stack frame just selected.
6994 @kindex frame@r{, selecting}
6995 @kindex f @r{(@code{frame})}
6998 Select frame number @var{n}. Recall that frame zero is the innermost
6999 (currently executing) frame, frame one is the frame that called the
7000 innermost one, and so on. The highest-numbered frame is the one for
7003 @item frame @var{addr}
7005 Select the frame at address @var{addr}. This is useful mainly if the
7006 chaining of stack frames has been damaged by a bug, making it
7007 impossible for @value{GDBN} to assign numbers properly to all frames. In
7008 addition, this can be useful when your program has multiple stacks and
7009 switches between them.
7011 On the SPARC architecture, @code{frame} needs two addresses to
7012 select an arbitrary frame: a frame pointer and a stack pointer.
7014 On the @acronym{MIPS} and Alpha architecture, it needs two addresses: a stack
7015 pointer and a program counter.
7017 On the 29k architecture, it needs three addresses: a register stack
7018 pointer, a program counter, and a memory stack pointer.
7022 Move @var{n} frames up the stack. For positive numbers @var{n}, this
7023 advances toward the outermost frame, to higher frame numbers, to frames
7024 that have existed longer. @var{n} defaults to one.
7027 @kindex do @r{(@code{down})}
7029 Move @var{n} frames down the stack. For positive numbers @var{n}, this
7030 advances toward the innermost frame, to lower frame numbers, to frames
7031 that were created more recently. @var{n} defaults to one. You may
7032 abbreviate @code{down} as @code{do}.
7035 All of these commands end by printing two lines of output describing the
7036 frame. The first line shows the frame number, the function name, the
7037 arguments, and the source file and line number of execution in that
7038 frame. The second line shows the text of that source line.
7046 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
7048 10 read_input_file (argv[i]);
7052 After such a printout, the @code{list} command with no arguments
7053 prints ten lines centered on the point of execution in the frame.
7054 You can also edit the program at the point of execution with your favorite
7055 editing program by typing @code{edit}.
7056 @xref{List, ,Printing Source Lines},
7060 @kindex down-silently
7062 @item up-silently @var{n}
7063 @itemx down-silently @var{n}
7064 These two commands are variants of @code{up} and @code{down},
7065 respectively; they differ in that they do their work silently, without
7066 causing display of the new frame. They are intended primarily for use
7067 in @value{GDBN} command scripts, where the output might be unnecessary and
7072 @section Information About a Frame
7074 There are several other commands to print information about the selected
7080 When used without any argument, this command does not change which
7081 frame is selected, but prints a brief description of the currently
7082 selected stack frame. It can be abbreviated @code{f}. With an
7083 argument, this command is used to select a stack frame.
7084 @xref{Selection, ,Selecting a Frame}.
7087 @kindex info f @r{(@code{info frame})}
7090 This command prints a verbose description of the selected stack frame,
7095 the address of the frame
7097 the address of the next frame down (called by this frame)
7099 the address of the next frame up (caller of this frame)
7101 the language in which the source code corresponding to this frame is written
7103 the address of the frame's arguments
7105 the address of the frame's local variables
7107 the program counter saved in it (the address of execution in the caller frame)
7109 which registers were saved in the frame
7112 @noindent The verbose description is useful when
7113 something has gone wrong that has made the stack format fail to fit
7114 the usual conventions.
7116 @item info frame @var{addr}
7117 @itemx info f @var{addr}
7118 Print a verbose description of the frame at address @var{addr}, without
7119 selecting that frame. The selected frame remains unchanged by this
7120 command. This requires the same kind of address (more than one for some
7121 architectures) that you specify in the @code{frame} command.
7122 @xref{Selection, ,Selecting a Frame}.
7126 Print the arguments of the selected frame, each on a separate line.
7130 Print the local variables of the selected frame, each on a separate
7131 line. These are all variables (declared either static or automatic)
7132 accessible at the point of execution of the selected frame.
7138 @chapter Examining Source Files
7140 @value{GDBN} can print parts of your program's source, since the debugging
7141 information recorded in the program tells @value{GDBN} what source files were
7142 used to build it. When your program stops, @value{GDBN} spontaneously prints
7143 the line where it stopped. Likewise, when you select a stack frame
7144 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
7145 execution in that frame has stopped. You can print other portions of
7146 source files by explicit command.
7148 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
7149 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
7150 @value{GDBN} under @sc{gnu} Emacs}.
7153 * List:: Printing source lines
7154 * Specify Location:: How to specify code locations
7155 * Edit:: Editing source files
7156 * Search:: Searching source files
7157 * Source Path:: Specifying source directories
7158 * Machine Code:: Source and machine code
7162 @section Printing Source Lines
7165 @kindex l @r{(@code{list})}
7166 To print lines from a source file, use the @code{list} command
7167 (abbreviated @code{l}). By default, ten lines are printed.
7168 There are several ways to specify what part of the file you want to
7169 print; see @ref{Specify Location}, for the full list.
7171 Here are the forms of the @code{list} command most commonly used:
7174 @item list @var{linenum}
7175 Print lines centered around line number @var{linenum} in the
7176 current source file.
7178 @item list @var{function}
7179 Print lines centered around the beginning of function
7183 Print more lines. If the last lines printed were printed with a
7184 @code{list} command, this prints lines following the last lines
7185 printed; however, if the last line printed was a solitary line printed
7186 as part of displaying a stack frame (@pxref{Stack, ,Examining the
7187 Stack}), this prints lines centered around that line.
7190 Print lines just before the lines last printed.
7193 @cindex @code{list}, how many lines to display
7194 By default, @value{GDBN} prints ten source lines with any of these forms of
7195 the @code{list} command. You can change this using @code{set listsize}:
7198 @kindex set listsize
7199 @item set listsize @var{count}
7200 @itemx set listsize unlimited
7201 Make the @code{list} command display @var{count} source lines (unless
7202 the @code{list} argument explicitly specifies some other number).
7203 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
7205 @kindex show listsize
7207 Display the number of lines that @code{list} prints.
7210 Repeating a @code{list} command with @key{RET} discards the argument,
7211 so it is equivalent to typing just @code{list}. This is more useful
7212 than listing the same lines again. An exception is made for an
7213 argument of @samp{-}; that argument is preserved in repetition so that
7214 each repetition moves up in the source file.
7216 In general, the @code{list} command expects you to supply zero, one or two
7217 @dfn{linespecs}. Linespecs specify source lines; there are several ways
7218 of writing them (@pxref{Specify Location}), but the effect is always
7219 to specify some source line.
7221 Here is a complete description of the possible arguments for @code{list}:
7224 @item list @var{linespec}
7225 Print lines centered around the line specified by @var{linespec}.
7227 @item list @var{first},@var{last}
7228 Print lines from @var{first} to @var{last}. Both arguments are
7229 linespecs. When a @code{list} command has two linespecs, and the
7230 source file of the second linespec is omitted, this refers to
7231 the same source file as the first linespec.
7233 @item list ,@var{last}
7234 Print lines ending with @var{last}.
7236 @item list @var{first},
7237 Print lines starting with @var{first}.
7240 Print lines just after the lines last printed.
7243 Print lines just before the lines last printed.
7246 As described in the preceding table.
7249 @node Specify Location
7250 @section Specifying a Location
7251 @cindex specifying location
7254 Several @value{GDBN} commands accept arguments that specify a location
7255 of your program's code. Since @value{GDBN} is a source-level
7256 debugger, a location usually specifies some line in the source code;
7257 for that reason, locations are also known as @dfn{linespecs}.
7259 Here are all the different ways of specifying a code location that
7260 @value{GDBN} understands:
7264 Specifies the line number @var{linenum} of the current source file.
7267 @itemx +@var{offset}
7268 Specifies the line @var{offset} lines before or after the @dfn{current
7269 line}. For the @code{list} command, the current line is the last one
7270 printed; for the breakpoint commands, this is the line at which
7271 execution stopped in the currently selected @dfn{stack frame}
7272 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
7273 used as the second of the two linespecs in a @code{list} command,
7274 this specifies the line @var{offset} lines up or down from the first
7277 @item @var{filename}:@var{linenum}
7278 Specifies the line @var{linenum} in the source file @var{filename}.
7279 If @var{filename} is a relative file name, then it will match any
7280 source file name with the same trailing components. For example, if
7281 @var{filename} is @samp{gcc/expr.c}, then it will match source file
7282 name of @file{/build/trunk/gcc/expr.c}, but not
7283 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
7285 @item @var{function}
7286 Specifies the line that begins the body of the function @var{function}.
7287 For example, in C, this is the line with the open brace.
7289 @item @var{function}:@var{label}
7290 Specifies the line where @var{label} appears in @var{function}.
7292 @item @var{filename}:@var{function}
7293 Specifies the line that begins the body of the function @var{function}
7294 in the file @var{filename}. You only need the file name with a
7295 function name to avoid ambiguity when there are identically named
7296 functions in different source files.
7299 Specifies the line at which the label named @var{label} appears.
7300 @value{GDBN} searches for the label in the function corresponding to
7301 the currently selected stack frame. If there is no current selected
7302 stack frame (for instance, if the inferior is not running), then
7303 @value{GDBN} will not search for a label.
7305 @item *@var{address}
7306 Specifies the program address @var{address}. For line-oriented
7307 commands, such as @code{list} and @code{edit}, this specifies a source
7308 line that contains @var{address}. For @code{break} and other
7309 breakpoint oriented commands, this can be used to set breakpoints in
7310 parts of your program which do not have debugging information or
7313 Here @var{address} may be any expression valid in the current working
7314 language (@pxref{Languages, working language}) that specifies a code
7315 address. In addition, as a convenience, @value{GDBN} extends the
7316 semantics of expressions used in locations to cover the situations
7317 that frequently happen during debugging. Here are the various forms
7321 @item @var{expression}
7322 Any expression valid in the current working language.
7324 @item @var{funcaddr}
7325 An address of a function or procedure derived from its name. In C,
7326 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
7327 simply the function's name @var{function} (and actually a special case
7328 of a valid expression). In Pascal and Modula-2, this is
7329 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
7330 (although the Pascal form also works).
7332 This form specifies the address of the function's first instruction,
7333 before the stack frame and arguments have been set up.
7335 @item '@var{filename}'::@var{funcaddr}
7336 Like @var{funcaddr} above, but also specifies the name of the source
7337 file explicitly. This is useful if the name of the function does not
7338 specify the function unambiguously, e.g., if there are several
7339 functions with identical names in different source files.
7342 @cindex breakpoint at static probe point
7343 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
7344 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
7345 applications to embed static probes. @xref{Static Probe Points}, for more
7346 information on finding and using static probes. This form of linespec
7347 specifies the location of such a static probe.
7349 If @var{objfile} is given, only probes coming from that shared library
7350 or executable matching @var{objfile} as a regular expression are considered.
7351 If @var{provider} is given, then only probes from that provider are considered.
7352 If several probes match the spec, @value{GDBN} will insert a breakpoint at
7353 each one of those probes.
7359 @section Editing Source Files
7360 @cindex editing source files
7363 @kindex e @r{(@code{edit})}
7364 To edit the lines in a source file, use the @code{edit} command.
7365 The editing program of your choice
7366 is invoked with the current line set to
7367 the active line in the program.
7368 Alternatively, there are several ways to specify what part of the file you
7369 want to print if you want to see other parts of the program:
7372 @item edit @var{location}
7373 Edit the source file specified by @code{location}. Editing starts at
7374 that @var{location}, e.g., at the specified source line of the
7375 specified file. @xref{Specify Location}, for all the possible forms
7376 of the @var{location} argument; here are the forms of the @code{edit}
7377 command most commonly used:
7380 @item edit @var{number}
7381 Edit the current source file with @var{number} as the active line number.
7383 @item edit @var{function}
7384 Edit the file containing @var{function} at the beginning of its definition.
7389 @subsection Choosing your Editor
7390 You can customize @value{GDBN} to use any editor you want
7392 The only restriction is that your editor (say @code{ex}), recognizes the
7393 following command-line syntax:
7395 ex +@var{number} file
7397 The optional numeric value +@var{number} specifies the number of the line in
7398 the file where to start editing.}.
7399 By default, it is @file{@value{EDITOR}}, but you can change this
7400 by setting the environment variable @code{EDITOR} before using
7401 @value{GDBN}. For example, to configure @value{GDBN} to use the
7402 @code{vi} editor, you could use these commands with the @code{sh} shell:
7408 or in the @code{csh} shell,
7410 setenv EDITOR /usr/bin/vi
7415 @section Searching Source Files
7416 @cindex searching source files
7418 There are two commands for searching through the current source file for a
7423 @kindex forward-search
7424 @kindex fo @r{(@code{forward-search})}
7425 @item forward-search @var{regexp}
7426 @itemx search @var{regexp}
7427 The command @samp{forward-search @var{regexp}} checks each line,
7428 starting with the one following the last line listed, for a match for
7429 @var{regexp}. It lists the line that is found. You can use the
7430 synonym @samp{search @var{regexp}} or abbreviate the command name as
7433 @kindex reverse-search
7434 @item reverse-search @var{regexp}
7435 The command @samp{reverse-search @var{regexp}} checks each line, starting
7436 with the one before the last line listed and going backward, for a match
7437 for @var{regexp}. It lists the line that is found. You can abbreviate
7438 this command as @code{rev}.
7442 @section Specifying Source Directories
7445 @cindex directories for source files
7446 Executable programs sometimes do not record the directories of the source
7447 files from which they were compiled, just the names. Even when they do,
7448 the directories could be moved between the compilation and your debugging
7449 session. @value{GDBN} has a list of directories to search for source files;
7450 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
7451 it tries all the directories in the list, in the order they are present
7452 in the list, until it finds a file with the desired name.
7454 For example, suppose an executable references the file
7455 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
7456 @file{/mnt/cross}. The file is first looked up literally; if this
7457 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
7458 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
7459 message is printed. @value{GDBN} does not look up the parts of the
7460 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
7461 Likewise, the subdirectories of the source path are not searched: if
7462 the source path is @file{/mnt/cross}, and the binary refers to
7463 @file{foo.c}, @value{GDBN} would not find it under
7464 @file{/mnt/cross/usr/src/foo-1.0/lib}.
7466 Plain file names, relative file names with leading directories, file
7467 names containing dots, etc.@: are all treated as described above; for
7468 instance, if the source path is @file{/mnt/cross}, and the source file
7469 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
7470 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
7471 that---@file{/mnt/cross/foo.c}.
7473 Note that the executable search path is @emph{not} used to locate the
7476 Whenever you reset or rearrange the source path, @value{GDBN} clears out
7477 any information it has cached about where source files are found and where
7478 each line is in the file.
7482 When you start @value{GDBN}, its source path includes only @samp{cdir}
7483 and @samp{cwd}, in that order.
7484 To add other directories, use the @code{directory} command.
7486 The search path is used to find both program source files and @value{GDBN}
7487 script files (read using the @samp{-command} option and @samp{source} command).
7489 In addition to the source path, @value{GDBN} provides a set of commands
7490 that manage a list of source path substitution rules. A @dfn{substitution
7491 rule} specifies how to rewrite source directories stored in the program's
7492 debug information in case the sources were moved to a different
7493 directory between compilation and debugging. A rule is made of
7494 two strings, the first specifying what needs to be rewritten in
7495 the path, and the second specifying how it should be rewritten.
7496 In @ref{set substitute-path}, we name these two parts @var{from} and
7497 @var{to} respectively. @value{GDBN} does a simple string replacement
7498 of @var{from} with @var{to} at the start of the directory part of the
7499 source file name, and uses that result instead of the original file
7500 name to look up the sources.
7502 Using the previous example, suppose the @file{foo-1.0} tree has been
7503 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
7504 @value{GDBN} to replace @file{/usr/src} in all source path names with
7505 @file{/mnt/cross}. The first lookup will then be
7506 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
7507 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
7508 substitution rule, use the @code{set substitute-path} command
7509 (@pxref{set substitute-path}).
7511 To avoid unexpected substitution results, a rule is applied only if the
7512 @var{from} part of the directory name ends at a directory separator.
7513 For instance, a rule substituting @file{/usr/source} into
7514 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
7515 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
7516 is applied only at the beginning of the directory name, this rule will
7517 not be applied to @file{/root/usr/source/baz.c} either.
7519 In many cases, you can achieve the same result using the @code{directory}
7520 command. However, @code{set substitute-path} can be more efficient in
7521 the case where the sources are organized in a complex tree with multiple
7522 subdirectories. With the @code{directory} command, you need to add each
7523 subdirectory of your project. If you moved the entire tree while
7524 preserving its internal organization, then @code{set substitute-path}
7525 allows you to direct the debugger to all the sources with one single
7528 @code{set substitute-path} is also more than just a shortcut command.
7529 The source path is only used if the file at the original location no
7530 longer exists. On the other hand, @code{set substitute-path} modifies
7531 the debugger behavior to look at the rewritten location instead. So, if
7532 for any reason a source file that is not relevant to your executable is
7533 located at the original location, a substitution rule is the only
7534 method available to point @value{GDBN} at the new location.
7536 @cindex @samp{--with-relocated-sources}
7537 @cindex default source path substitution
7538 You can configure a default source path substitution rule by
7539 configuring @value{GDBN} with the
7540 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
7541 should be the name of a directory under @value{GDBN}'s configured
7542 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
7543 directory names in debug information under @var{dir} will be adjusted
7544 automatically if the installed @value{GDBN} is moved to a new
7545 location. This is useful if @value{GDBN}, libraries or executables
7546 with debug information and corresponding source code are being moved
7550 @item directory @var{dirname} @dots{}
7551 @item dir @var{dirname} @dots{}
7552 Add directory @var{dirname} to the front of the source path. Several
7553 directory names may be given to this command, separated by @samp{:}
7554 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
7555 part of absolute file names) or
7556 whitespace. You may specify a directory that is already in the source
7557 path; this moves it forward, so @value{GDBN} searches it sooner.
7561 @vindex $cdir@r{, convenience variable}
7562 @vindex $cwd@r{, convenience variable}
7563 @cindex compilation directory
7564 @cindex current directory
7565 @cindex working directory
7566 @cindex directory, current
7567 @cindex directory, compilation
7568 You can use the string @samp{$cdir} to refer to the compilation
7569 directory (if one is recorded), and @samp{$cwd} to refer to the current
7570 working directory. @samp{$cwd} is not the same as @samp{.}---the former
7571 tracks the current working directory as it changes during your @value{GDBN}
7572 session, while the latter is immediately expanded to the current
7573 directory at the time you add an entry to the source path.
7576 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
7578 @c RET-repeat for @code{directory} is explicitly disabled, but since
7579 @c repeating it would be a no-op we do not say that. (thanks to RMS)
7581 @item set directories @var{path-list}
7582 @kindex set directories
7583 Set the source path to @var{path-list}.
7584 @samp{$cdir:$cwd} are added if missing.
7586 @item show directories
7587 @kindex show directories
7588 Print the source path: show which directories it contains.
7590 @anchor{set substitute-path}
7591 @item set substitute-path @var{from} @var{to}
7592 @kindex set substitute-path
7593 Define a source path substitution rule, and add it at the end of the
7594 current list of existing substitution rules. If a rule with the same
7595 @var{from} was already defined, then the old rule is also deleted.
7597 For example, if the file @file{/foo/bar/baz.c} was moved to
7598 @file{/mnt/cross/baz.c}, then the command
7601 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
7605 will tell @value{GDBN} to replace @samp{/usr/src} with
7606 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
7607 @file{baz.c} even though it was moved.
7609 In the case when more than one substitution rule have been defined,
7610 the rules are evaluated one by one in the order where they have been
7611 defined. The first one matching, if any, is selected to perform
7614 For instance, if we had entered the following commands:
7617 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
7618 (@value{GDBP}) set substitute-path /usr/src /mnt/src
7622 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
7623 @file{/mnt/include/defs.h} by using the first rule. However, it would
7624 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
7625 @file{/mnt/src/lib/foo.c}.
7628 @item unset substitute-path [path]
7629 @kindex unset substitute-path
7630 If a path is specified, search the current list of substitution rules
7631 for a rule that would rewrite that path. Delete that rule if found.
7632 A warning is emitted by the debugger if no rule could be found.
7634 If no path is specified, then all substitution rules are deleted.
7636 @item show substitute-path [path]
7637 @kindex show substitute-path
7638 If a path is specified, then print the source path substitution rule
7639 which would rewrite that path, if any.
7641 If no path is specified, then print all existing source path substitution
7646 If your source path is cluttered with directories that are no longer of
7647 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
7648 versions of source. You can correct the situation as follows:
7652 Use @code{directory} with no argument to reset the source path to its default value.
7655 Use @code{directory} with suitable arguments to reinstall the
7656 directories you want in the source path. You can add all the
7657 directories in one command.
7661 @section Source and Machine Code
7662 @cindex source line and its code address
7664 You can use the command @code{info line} to map source lines to program
7665 addresses (and vice versa), and the command @code{disassemble} to display
7666 a range of addresses as machine instructions. You can use the command
7667 @code{set disassemble-next-line} to set whether to disassemble next
7668 source line when execution stops. When run under @sc{gnu} Emacs
7669 mode, the @code{info line} command causes the arrow to point to the
7670 line specified. Also, @code{info line} prints addresses in symbolic form as
7675 @item info line @var{linespec}
7676 Print the starting and ending addresses of the compiled code for
7677 source line @var{linespec}. You can specify source lines in any of
7678 the ways documented in @ref{Specify Location}.
7681 For example, we can use @code{info line} to discover the location of
7682 the object code for the first line of function
7683 @code{m4_changequote}:
7685 @c FIXME: I think this example should also show the addresses in
7686 @c symbolic form, as they usually would be displayed.
7688 (@value{GDBP}) info line m4_changequote
7689 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
7693 @cindex code address and its source line
7694 We can also inquire (using @code{*@var{addr}} as the form for
7695 @var{linespec}) what source line covers a particular address:
7697 (@value{GDBP}) info line *0x63ff
7698 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
7701 @cindex @code{$_} and @code{info line}
7702 @cindex @code{x} command, default address
7703 @kindex x@r{(examine), and} info line
7704 After @code{info line}, the default address for the @code{x} command
7705 is changed to the starting address of the line, so that @samp{x/i} is
7706 sufficient to begin examining the machine code (@pxref{Memory,
7707 ,Examining Memory}). Also, this address is saved as the value of the
7708 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
7713 @cindex assembly instructions
7714 @cindex instructions, assembly
7715 @cindex machine instructions
7716 @cindex listing machine instructions
7718 @itemx disassemble /m
7719 @itemx disassemble /r
7720 This specialized command dumps a range of memory as machine
7721 instructions. It can also print mixed source+disassembly by specifying
7722 the @code{/m} modifier and print the raw instructions in hex as well as
7723 in symbolic form by specifying the @code{/r}.
7724 The default memory range is the function surrounding the
7725 program counter of the selected frame. A single argument to this
7726 command is a program counter value; @value{GDBN} dumps the function
7727 surrounding this value. When two arguments are given, they should
7728 be separated by a comma, possibly surrounded by whitespace. The
7729 arguments specify a range of addresses to dump, in one of two forms:
7732 @item @var{start},@var{end}
7733 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
7734 @item @var{start},+@var{length}
7735 the addresses from @var{start} (inclusive) to
7736 @code{@var{start}+@var{length}} (exclusive).
7740 When 2 arguments are specified, the name of the function is also
7741 printed (since there could be several functions in the given range).
7743 The argument(s) can be any expression yielding a numeric value, such as
7744 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
7746 If the range of memory being disassembled contains current program counter,
7747 the instruction at that location is shown with a @code{=>} marker.
7750 The following example shows the disassembly of a range of addresses of
7751 HP PA-RISC 2.0 code:
7754 (@value{GDBP}) disas 0x32c4, 0x32e4
7755 Dump of assembler code from 0x32c4 to 0x32e4:
7756 0x32c4 <main+204>: addil 0,dp
7757 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
7758 0x32cc <main+212>: ldil 0x3000,r31
7759 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
7760 0x32d4 <main+220>: ldo 0(r31),rp
7761 0x32d8 <main+224>: addil -0x800,dp
7762 0x32dc <main+228>: ldo 0x588(r1),r26
7763 0x32e0 <main+232>: ldil 0x3000,r31
7764 End of assembler dump.
7767 Here is an example showing mixed source+assembly for Intel x86, when the
7768 program is stopped just after function prologue:
7771 (@value{GDBP}) disas /m main
7772 Dump of assembler code for function main:
7774 0x08048330 <+0>: push %ebp
7775 0x08048331 <+1>: mov %esp,%ebp
7776 0x08048333 <+3>: sub $0x8,%esp
7777 0x08048336 <+6>: and $0xfffffff0,%esp
7778 0x08048339 <+9>: sub $0x10,%esp
7780 6 printf ("Hello.\n");
7781 => 0x0804833c <+12>: movl $0x8048440,(%esp)
7782 0x08048343 <+19>: call 0x8048284 <puts@@plt>
7786 0x08048348 <+24>: mov $0x0,%eax
7787 0x0804834d <+29>: leave
7788 0x0804834e <+30>: ret
7790 End of assembler dump.
7793 Here is another example showing raw instructions in hex for AMD x86-64,
7796 (gdb) disas /r 0x400281,+10
7797 Dump of assembler code from 0x400281 to 0x40028b:
7798 0x0000000000400281: 38 36 cmp %dh,(%rsi)
7799 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
7800 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
7801 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
7802 End of assembler dump.
7805 Addresses cannot be specified as a linespec (@pxref{Specify Location}).
7806 So, for example, if you want to disassemble function @code{bar}
7807 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
7808 and not @samp{disassemble foo.c:bar}.
7810 Some architectures have more than one commonly-used set of instruction
7811 mnemonics or other syntax.
7813 For programs that were dynamically linked and use shared libraries,
7814 instructions that call functions or branch to locations in the shared
7815 libraries might show a seemingly bogus location---it's actually a
7816 location of the relocation table. On some architectures, @value{GDBN}
7817 might be able to resolve these to actual function names.
7820 @kindex set disassembly-flavor
7821 @cindex Intel disassembly flavor
7822 @cindex AT&T disassembly flavor
7823 @item set disassembly-flavor @var{instruction-set}
7824 Select the instruction set to use when disassembling the
7825 program via the @code{disassemble} or @code{x/i} commands.
7827 Currently this command is only defined for the Intel x86 family. You
7828 can set @var{instruction-set} to either @code{intel} or @code{att}.
7829 The default is @code{att}, the AT&T flavor used by default by Unix
7830 assemblers for x86-based targets.
7832 @kindex show disassembly-flavor
7833 @item show disassembly-flavor
7834 Show the current setting of the disassembly flavor.
7838 @kindex set disassemble-next-line
7839 @kindex show disassemble-next-line
7840 @item set disassemble-next-line
7841 @itemx show disassemble-next-line
7842 Control whether or not @value{GDBN} will disassemble the next source
7843 line or instruction when execution stops. If ON, @value{GDBN} will
7844 display disassembly of the next source line when execution of the
7845 program being debugged stops. This is @emph{in addition} to
7846 displaying the source line itself, which @value{GDBN} always does if
7847 possible. If the next source line cannot be displayed for some reason
7848 (e.g., if @value{GDBN} cannot find the source file, or there's no line
7849 info in the debug info), @value{GDBN} will display disassembly of the
7850 next @emph{instruction} instead of showing the next source line. If
7851 AUTO, @value{GDBN} will display disassembly of next instruction only
7852 if the source line cannot be displayed. This setting causes
7853 @value{GDBN} to display some feedback when you step through a function
7854 with no line info or whose source file is unavailable. The default is
7855 OFF, which means never display the disassembly of the next line or
7861 @chapter Examining Data
7863 @cindex printing data
7864 @cindex examining data
7867 The usual way to examine data in your program is with the @code{print}
7868 command (abbreviated @code{p}), or its synonym @code{inspect}. It
7869 evaluates and prints the value of an expression of the language your
7870 program is written in (@pxref{Languages, ,Using @value{GDBN} with
7871 Different Languages}). It may also print the expression using a
7872 Python-based pretty-printer (@pxref{Pretty Printing}).
7875 @item print @var{expr}
7876 @itemx print /@var{f} @var{expr}
7877 @var{expr} is an expression (in the source language). By default the
7878 value of @var{expr} is printed in a format appropriate to its data type;
7879 you can choose a different format by specifying @samp{/@var{f}}, where
7880 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
7884 @itemx print /@var{f}
7885 @cindex reprint the last value
7886 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
7887 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
7888 conveniently inspect the same value in an alternative format.
7891 A more low-level way of examining data is with the @code{x} command.
7892 It examines data in memory at a specified address and prints it in a
7893 specified format. @xref{Memory, ,Examining Memory}.
7895 If you are interested in information about types, or about how the
7896 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
7897 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
7900 @cindex exploring hierarchical data structures
7902 Another way of examining values of expressions and type information is
7903 through the Python extension command @code{explore} (available only if
7904 the @value{GDBN} build is configured with @code{--with-python}). It
7905 offers an interactive way to start at the highest level (or, the most
7906 abstract level) of the data type of an expression (or, the data type
7907 itself) and explore all the way down to leaf scalar values/fields
7908 embedded in the higher level data types.
7911 @item explore @var{arg}
7912 @var{arg} is either an expression (in the source language), or a type
7913 visible in the current context of the program being debugged.
7916 The working of the @code{explore} command can be illustrated with an
7917 example. If a data type @code{struct ComplexStruct} is defined in your
7927 struct ComplexStruct
7929 struct SimpleStruct *ss_p;
7935 followed by variable declarations as
7938 struct SimpleStruct ss = @{ 10, 1.11 @};
7939 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
7943 then, the value of the variable @code{cs} can be explored using the
7944 @code{explore} command as follows.
7948 The value of `cs' is a struct/class of type `struct ComplexStruct' with
7949 the following fields:
7951 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
7952 arr = <Enter 1 to explore this field of type `int [10]'>
7954 Enter the field number of choice:
7958 Since the fields of @code{cs} are not scalar values, you are being
7959 prompted to chose the field you want to explore. Let's say you choose
7960 the field @code{ss_p} by entering @code{0}. Then, since this field is a
7961 pointer, you will be asked if it is pointing to a single value. From
7962 the declaration of @code{cs} above, it is indeed pointing to a single
7963 value, hence you enter @code{y}. If you enter @code{n}, then you will
7964 be asked if it were pointing to an array of values, in which case this
7965 field will be explored as if it were an array.
7968 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
7969 Continue exploring it as a pointer to a single value [y/n]: y
7970 The value of `*(cs.ss_p)' is a struct/class of type `struct
7971 SimpleStruct' with the following fields:
7973 i = 10 .. (Value of type `int')
7974 d = 1.1100000000000001 .. (Value of type `double')
7976 Press enter to return to parent value:
7980 If the field @code{arr} of @code{cs} was chosen for exploration by
7981 entering @code{1} earlier, then since it is as array, you will be
7982 prompted to enter the index of the element in the array that you want
7986 `cs.arr' is an array of `int'.
7987 Enter the index of the element you want to explore in `cs.arr': 5
7989 `(cs.arr)[5]' is a scalar value of type `int'.
7993 Press enter to return to parent value:
7996 In general, at any stage of exploration, you can go deeper towards the
7997 leaf values by responding to the prompts appropriately, or hit the
7998 return key to return to the enclosing data structure (the @i{higher}
7999 level data structure).
8001 Similar to exploring values, you can use the @code{explore} command to
8002 explore types. Instead of specifying a value (which is typically a
8003 variable name or an expression valid in the current context of the
8004 program being debugged), you specify a type name. If you consider the
8005 same example as above, your can explore the type
8006 @code{struct ComplexStruct} by passing the argument
8007 @code{struct ComplexStruct} to the @code{explore} command.
8010 (gdb) explore struct ComplexStruct
8014 By responding to the prompts appropriately in the subsequent interactive
8015 session, you can explore the type @code{struct ComplexStruct} in a
8016 manner similar to how the value @code{cs} was explored in the above
8019 The @code{explore} command also has two sub-commands,
8020 @code{explore value} and @code{explore type}. The former sub-command is
8021 a way to explicitly specify that value exploration of the argument is
8022 being invoked, while the latter is a way to explicitly specify that type
8023 exploration of the argument is being invoked.
8026 @item explore value @var{expr}
8027 @cindex explore value
8028 This sub-command of @code{explore} explores the value of the
8029 expression @var{expr} (if @var{expr} is an expression valid in the
8030 current context of the program being debugged). The behavior of this
8031 command is identical to that of the behavior of the @code{explore}
8032 command being passed the argument @var{expr}.
8034 @item explore type @var{arg}
8035 @cindex explore type
8036 This sub-command of @code{explore} explores the type of @var{arg} (if
8037 @var{arg} is a type visible in the current context of program being
8038 debugged), or the type of the value/expression @var{arg} (if @var{arg}
8039 is an expression valid in the current context of the program being
8040 debugged). If @var{arg} is a type, then the behavior of this command is
8041 identical to that of the @code{explore} command being passed the
8042 argument @var{arg}. If @var{arg} is an expression, then the behavior of
8043 this command will be identical to that of the @code{explore} command
8044 being passed the type of @var{arg} as the argument.
8048 * Expressions:: Expressions
8049 * Ambiguous Expressions:: Ambiguous Expressions
8050 * Variables:: Program variables
8051 * Arrays:: Artificial arrays
8052 * Output Formats:: Output formats
8053 * Memory:: Examining memory
8054 * Auto Display:: Automatic display
8055 * Print Settings:: Print settings
8056 * Pretty Printing:: Python pretty printing
8057 * Value History:: Value history
8058 * Convenience Vars:: Convenience variables
8059 * Convenience Funs:: Convenience functions
8060 * Registers:: Registers
8061 * Floating Point Hardware:: Floating point hardware
8062 * Vector Unit:: Vector Unit
8063 * OS Information:: Auxiliary data provided by operating system
8064 * Memory Region Attributes:: Memory region attributes
8065 * Dump/Restore Files:: Copy between memory and a file
8066 * Core File Generation:: Cause a program dump its core
8067 * Character Sets:: Debugging programs that use a different
8068 character set than GDB does
8069 * Caching Remote Data:: Data caching for remote targets
8070 * Searching Memory:: Searching memory for a sequence of bytes
8074 @section Expressions
8077 @code{print} and many other @value{GDBN} commands accept an expression and
8078 compute its value. Any kind of constant, variable or operator defined
8079 by the programming language you are using is valid in an expression in
8080 @value{GDBN}. This includes conditional expressions, function calls,
8081 casts, and string constants. It also includes preprocessor macros, if
8082 you compiled your program to include this information; see
8085 @cindex arrays in expressions
8086 @value{GDBN} supports array constants in expressions input by
8087 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
8088 you can use the command @code{print @{1, 2, 3@}} to create an array
8089 of three integers. If you pass an array to a function or assign it
8090 to a program variable, @value{GDBN} copies the array to memory that
8091 is @code{malloc}ed in the target program.
8093 Because C is so widespread, most of the expressions shown in examples in
8094 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
8095 Languages}, for information on how to use expressions in other
8098 In this section, we discuss operators that you can use in @value{GDBN}
8099 expressions regardless of your programming language.
8101 @cindex casts, in expressions
8102 Casts are supported in all languages, not just in C, because it is so
8103 useful to cast a number into a pointer in order to examine a structure
8104 at that address in memory.
8105 @c FIXME: casts supported---Mod2 true?
8107 @value{GDBN} supports these operators, in addition to those common
8108 to programming languages:
8112 @samp{@@} is a binary operator for treating parts of memory as arrays.
8113 @xref{Arrays, ,Artificial Arrays}, for more information.
8116 @samp{::} allows you to specify a variable in terms of the file or
8117 function where it is defined. @xref{Variables, ,Program Variables}.
8119 @cindex @{@var{type}@}
8120 @cindex type casting memory
8121 @cindex memory, viewing as typed object
8122 @cindex casts, to view memory
8123 @item @{@var{type}@} @var{addr}
8124 Refers to an object of type @var{type} stored at address @var{addr} in
8125 memory. @var{addr} may be any expression whose value is an integer or
8126 pointer (but parentheses are required around binary operators, just as in
8127 a cast). This construct is allowed regardless of what kind of data is
8128 normally supposed to reside at @var{addr}.
8131 @node Ambiguous Expressions
8132 @section Ambiguous Expressions
8133 @cindex ambiguous expressions
8135 Expressions can sometimes contain some ambiguous elements. For instance,
8136 some programming languages (notably Ada, C@t{++} and Objective-C) permit
8137 a single function name to be defined several times, for application in
8138 different contexts. This is called @dfn{overloading}. Another example
8139 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
8140 templates and is typically instantiated several times, resulting in
8141 the same function name being defined in different contexts.
8143 In some cases and depending on the language, it is possible to adjust
8144 the expression to remove the ambiguity. For instance in C@t{++}, you
8145 can specify the signature of the function you want to break on, as in
8146 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
8147 qualified name of your function often makes the expression unambiguous
8150 When an ambiguity that needs to be resolved is detected, the debugger
8151 has the capability to display a menu of numbered choices for each
8152 possibility, and then waits for the selection with the prompt @samp{>}.
8153 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
8154 aborts the current command. If the command in which the expression was
8155 used allows more than one choice to be selected, the next option in the
8156 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
8159 For example, the following session excerpt shows an attempt to set a
8160 breakpoint at the overloaded symbol @code{String::after}.
8161 We choose three particular definitions of that function name:
8163 @c FIXME! This is likely to change to show arg type lists, at least
8166 (@value{GDBP}) b String::after
8169 [2] file:String.cc; line number:867
8170 [3] file:String.cc; line number:860
8171 [4] file:String.cc; line number:875
8172 [5] file:String.cc; line number:853
8173 [6] file:String.cc; line number:846
8174 [7] file:String.cc; line number:735
8176 Breakpoint 1 at 0xb26c: file String.cc, line 867.
8177 Breakpoint 2 at 0xb344: file String.cc, line 875.
8178 Breakpoint 3 at 0xafcc: file String.cc, line 846.
8179 Multiple breakpoints were set.
8180 Use the "delete" command to delete unwanted
8187 @kindex set multiple-symbols
8188 @item set multiple-symbols @var{mode}
8189 @cindex multiple-symbols menu
8191 This option allows you to adjust the debugger behavior when an expression
8194 By default, @var{mode} is set to @code{all}. If the command with which
8195 the expression is used allows more than one choice, then @value{GDBN}
8196 automatically selects all possible choices. For instance, inserting
8197 a breakpoint on a function using an ambiguous name results in a breakpoint
8198 inserted on each possible match. However, if a unique choice must be made,
8199 then @value{GDBN} uses the menu to help you disambiguate the expression.
8200 For instance, printing the address of an overloaded function will result
8201 in the use of the menu.
8203 When @var{mode} is set to @code{ask}, the debugger always uses the menu
8204 when an ambiguity is detected.
8206 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
8207 an error due to the ambiguity and the command is aborted.
8209 @kindex show multiple-symbols
8210 @item show multiple-symbols
8211 Show the current value of the @code{multiple-symbols} setting.
8215 @section Program Variables
8217 The most common kind of expression to use is the name of a variable
8220 Variables in expressions are understood in the selected stack frame
8221 (@pxref{Selection, ,Selecting a Frame}); they must be either:
8225 global (or file-static)
8232 visible according to the scope rules of the
8233 programming language from the point of execution in that frame
8236 @noindent This means that in the function
8251 you can examine and use the variable @code{a} whenever your program is
8252 executing within the function @code{foo}, but you can only use or
8253 examine the variable @code{b} while your program is executing inside
8254 the block where @code{b} is declared.
8256 @cindex variable name conflict
8257 There is an exception: you can refer to a variable or function whose
8258 scope is a single source file even if the current execution point is not
8259 in this file. But it is possible to have more than one such variable or
8260 function with the same name (in different source files). If that
8261 happens, referring to that name has unpredictable effects. If you wish,
8262 you can specify a static variable in a particular function or file by
8263 using the colon-colon (@code{::}) notation:
8265 @cindex colon-colon, context for variables/functions
8267 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
8268 @cindex @code{::}, context for variables/functions
8271 @var{file}::@var{variable}
8272 @var{function}::@var{variable}
8276 Here @var{file} or @var{function} is the name of the context for the
8277 static @var{variable}. In the case of file names, you can use quotes to
8278 make sure @value{GDBN} parses the file name as a single word---for example,
8279 to print a global value of @code{x} defined in @file{f2.c}:
8282 (@value{GDBP}) p 'f2.c'::x
8285 The @code{::} notation is normally used for referring to
8286 static variables, since you typically disambiguate uses of local variables
8287 in functions by selecting the appropriate frame and using the
8288 simple name of the variable. However, you may also use this notation
8289 to refer to local variables in frames enclosing the selected frame:
8298 process (a); /* Stop here */
8309 For example, if there is a breakpoint at the commented line,
8310 here is what you might see
8311 when the program stops after executing the call @code{bar(0)}:
8316 (@value{GDBP}) p bar::a
8319 #2 0x080483d0 in foo (a=5) at foobar.c:12
8322 (@value{GDBP}) p bar::a
8326 @cindex C@t{++} scope resolution
8327 These uses of @samp{::} are very rarely in conflict with the very similar
8328 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
8329 scope resolution operator in @value{GDBN} expressions.
8330 @c FIXME: Um, so what happens in one of those rare cases where it's in
8333 @cindex wrong values
8334 @cindex variable values, wrong
8335 @cindex function entry/exit, wrong values of variables
8336 @cindex optimized code, wrong values of variables
8338 @emph{Warning:} Occasionally, a local variable may appear to have the
8339 wrong value at certain points in a function---just after entry to a new
8340 scope, and just before exit.
8342 You may see this problem when you are stepping by machine instructions.
8343 This is because, on most machines, it takes more than one instruction to
8344 set up a stack frame (including local variable definitions); if you are
8345 stepping by machine instructions, variables may appear to have the wrong
8346 values until the stack frame is completely built. On exit, it usually
8347 also takes more than one machine instruction to destroy a stack frame;
8348 after you begin stepping through that group of instructions, local
8349 variable definitions may be gone.
8351 This may also happen when the compiler does significant optimizations.
8352 To be sure of always seeing accurate values, turn off all optimization
8355 @cindex ``No symbol "foo" in current context''
8356 Another possible effect of compiler optimizations is to optimize
8357 unused variables out of existence, or assign variables to registers (as
8358 opposed to memory addresses). Depending on the support for such cases
8359 offered by the debug info format used by the compiler, @value{GDBN}
8360 might not be able to display values for such local variables. If that
8361 happens, @value{GDBN} will print a message like this:
8364 No symbol "foo" in current context.
8367 To solve such problems, either recompile without optimizations, or use a
8368 different debug info format, if the compiler supports several such
8369 formats. @xref{Compilation}, for more information on choosing compiler
8370 options. @xref{C, ,C and C@t{++}}, for more information about debug
8371 info formats that are best suited to C@t{++} programs.
8373 If you ask to print an object whose contents are unknown to
8374 @value{GDBN}, e.g., because its data type is not completely specified
8375 by the debug information, @value{GDBN} will say @samp{<incomplete
8376 type>}. @xref{Symbols, incomplete type}, for more about this.
8378 If you append @kbd{@@entry} string to a function parameter name you get its
8379 value at the time the function got called. If the value is not available an
8380 error message is printed. Entry values are available only with some compilers.
8381 Entry values are normally also printed at the function parameter list according
8382 to @ref{set print entry-values}.
8385 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
8391 (gdb) print i@@entry
8395 Strings are identified as arrays of @code{char} values without specified
8396 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
8397 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
8398 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
8399 defines literal string type @code{"char"} as @code{char} without a sign.
8404 signed char var1[] = "A";
8407 You get during debugging
8412 $2 = @{65 'A', 0 '\0'@}
8416 @section Artificial Arrays
8418 @cindex artificial array
8420 @kindex @@@r{, referencing memory as an array}
8421 It is often useful to print out several successive objects of the
8422 same type in memory; a section of an array, or an array of
8423 dynamically determined size for which only a pointer exists in the
8426 You can do this by referring to a contiguous span of memory as an
8427 @dfn{artificial array}, using the binary operator @samp{@@}. The left
8428 operand of @samp{@@} should be the first element of the desired array
8429 and be an individual object. The right operand should be the desired length
8430 of the array. The result is an array value whose elements are all of
8431 the type of the left argument. The first element is actually the left
8432 argument; the second element comes from bytes of memory immediately
8433 following those that hold the first element, and so on. Here is an
8434 example. If a program says
8437 int *array = (int *) malloc (len * sizeof (int));
8441 you can print the contents of @code{array} with
8447 The left operand of @samp{@@} must reside in memory. Array values made
8448 with @samp{@@} in this way behave just like other arrays in terms of
8449 subscripting, and are coerced to pointers when used in expressions.
8450 Artificial arrays most often appear in expressions via the value history
8451 (@pxref{Value History, ,Value History}), after printing one out.
8453 Another way to create an artificial array is to use a cast.
8454 This re-interprets a value as if it were an array.
8455 The value need not be in memory:
8457 (@value{GDBP}) p/x (short[2])0x12345678
8458 $1 = @{0x1234, 0x5678@}
8461 As a convenience, if you leave the array length out (as in
8462 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
8463 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
8465 (@value{GDBP}) p/x (short[])0x12345678
8466 $2 = @{0x1234, 0x5678@}
8469 Sometimes the artificial array mechanism is not quite enough; in
8470 moderately complex data structures, the elements of interest may not
8471 actually be adjacent---for example, if you are interested in the values
8472 of pointers in an array. One useful work-around in this situation is
8473 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
8474 Variables}) as a counter in an expression that prints the first
8475 interesting value, and then repeat that expression via @key{RET}. For
8476 instance, suppose you have an array @code{dtab} of pointers to
8477 structures, and you are interested in the values of a field @code{fv}
8478 in each structure. Here is an example of what you might type:
8488 @node Output Formats
8489 @section Output Formats
8491 @cindex formatted output
8492 @cindex output formats
8493 By default, @value{GDBN} prints a value according to its data type. Sometimes
8494 this is not what you want. For example, you might want to print a number
8495 in hex, or a pointer in decimal. Or you might want to view data in memory
8496 at a certain address as a character string or as an instruction. To do
8497 these things, specify an @dfn{output format} when you print a value.
8499 The simplest use of output formats is to say how to print a value
8500 already computed. This is done by starting the arguments of the
8501 @code{print} command with a slash and a format letter. The format
8502 letters supported are:
8506 Regard the bits of the value as an integer, and print the integer in
8510 Print as integer in signed decimal.
8513 Print as integer in unsigned decimal.
8516 Print as integer in octal.
8519 Print as integer in binary. The letter @samp{t} stands for ``two''.
8520 @footnote{@samp{b} cannot be used because these format letters are also
8521 used with the @code{x} command, where @samp{b} stands for ``byte'';
8522 see @ref{Memory,,Examining Memory}.}
8525 @cindex unknown address, locating
8526 @cindex locate address
8527 Print as an address, both absolute in hexadecimal and as an offset from
8528 the nearest preceding symbol. You can use this format used to discover
8529 where (in what function) an unknown address is located:
8532 (@value{GDBP}) p/a 0x54320
8533 $3 = 0x54320 <_initialize_vx+396>
8537 The command @code{info symbol 0x54320} yields similar results.
8538 @xref{Symbols, info symbol}.
8541 Regard as an integer and print it as a character constant. This
8542 prints both the numerical value and its character representation. The
8543 character representation is replaced with the octal escape @samp{\nnn}
8544 for characters outside the 7-bit @sc{ascii} range.
8546 Without this format, @value{GDBN} displays @code{char},
8547 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
8548 constants. Single-byte members of vectors are displayed as integer
8552 Regard the bits of the value as a floating point number and print
8553 using typical floating point syntax.
8556 @cindex printing strings
8557 @cindex printing byte arrays
8558 Regard as a string, if possible. With this format, pointers to single-byte
8559 data are displayed as null-terminated strings and arrays of single-byte data
8560 are displayed as fixed-length strings. Other values are displayed in their
8563 Without this format, @value{GDBN} displays pointers to and arrays of
8564 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
8565 strings. Single-byte members of a vector are displayed as an integer
8569 Like @samp{x} formatting, the value is treated as an integer and
8570 printed as hexadecimal, but leading zeros are printed to pad the value
8571 to the size of the integer type.
8574 @cindex raw printing
8575 Print using the @samp{raw} formatting. By default, @value{GDBN} will
8576 use a Python-based pretty-printer, if one is available (@pxref{Pretty
8577 Printing}). This typically results in a higher-level display of the
8578 value's contents. The @samp{r} format bypasses any Python
8579 pretty-printer which might exist.
8582 For example, to print the program counter in hex (@pxref{Registers}), type
8589 Note that no space is required before the slash; this is because command
8590 names in @value{GDBN} cannot contain a slash.
8592 To reprint the last value in the value history with a different format,
8593 you can use the @code{print} command with just a format and no
8594 expression. For example, @samp{p/x} reprints the last value in hex.
8597 @section Examining Memory
8599 You can use the command @code{x} (for ``examine'') to examine memory in
8600 any of several formats, independently of your program's data types.
8602 @cindex examining memory
8604 @kindex x @r{(examine memory)}
8605 @item x/@var{nfu} @var{addr}
8608 Use the @code{x} command to examine memory.
8611 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
8612 much memory to display and how to format it; @var{addr} is an
8613 expression giving the address where you want to start displaying memory.
8614 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
8615 Several commands set convenient defaults for @var{addr}.
8618 @item @var{n}, the repeat count
8619 The repeat count is a decimal integer; the default is 1. It specifies
8620 how much memory (counting by units @var{u}) to display.
8621 @c This really is **decimal**; unaffected by 'set radix' as of GDB
8624 @item @var{f}, the display format
8625 The display format is one of the formats used by @code{print}
8626 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
8627 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
8628 The default is @samp{x} (hexadecimal) initially. The default changes
8629 each time you use either @code{x} or @code{print}.
8631 @item @var{u}, the unit size
8632 The unit size is any of
8638 Halfwords (two bytes).
8640 Words (four bytes). This is the initial default.
8642 Giant words (eight bytes).
8645 Each time you specify a unit size with @code{x}, that size becomes the
8646 default unit the next time you use @code{x}. For the @samp{i} format,
8647 the unit size is ignored and is normally not written. For the @samp{s} format,
8648 the unit size defaults to @samp{b}, unless it is explicitly given.
8649 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
8650 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
8651 Note that the results depend on the programming language of the
8652 current compilation unit. If the language is C, the @samp{s}
8653 modifier will use the UTF-16 encoding while @samp{w} will use
8654 UTF-32. The encoding is set by the programming language and cannot
8657 @item @var{addr}, starting display address
8658 @var{addr} is the address where you want @value{GDBN} to begin displaying
8659 memory. The expression need not have a pointer value (though it may);
8660 it is always interpreted as an integer address of a byte of memory.
8661 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
8662 @var{addr} is usually just after the last address examined---but several
8663 other commands also set the default address: @code{info breakpoints} (to
8664 the address of the last breakpoint listed), @code{info line} (to the
8665 starting address of a line), and @code{print} (if you use it to display
8666 a value from memory).
8669 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
8670 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
8671 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
8672 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
8673 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
8675 Since the letters indicating unit sizes are all distinct from the
8676 letters specifying output formats, you do not have to remember whether
8677 unit size or format comes first; either order works. The output
8678 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
8679 (However, the count @var{n} must come first; @samp{wx4} does not work.)
8681 Even though the unit size @var{u} is ignored for the formats @samp{s}
8682 and @samp{i}, you might still want to use a count @var{n}; for example,
8683 @samp{3i} specifies that you want to see three machine instructions,
8684 including any operands. For convenience, especially when used with
8685 the @code{display} command, the @samp{i} format also prints branch delay
8686 slot instructions, if any, beyond the count specified, which immediately
8687 follow the last instruction that is within the count. The command
8688 @code{disassemble} gives an alternative way of inspecting machine
8689 instructions; see @ref{Machine Code,,Source and Machine Code}.
8691 All the defaults for the arguments to @code{x} are designed to make it
8692 easy to continue scanning memory with minimal specifications each time
8693 you use @code{x}. For example, after you have inspected three machine
8694 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
8695 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
8696 the repeat count @var{n} is used again; the other arguments default as
8697 for successive uses of @code{x}.
8699 When examining machine instructions, the instruction at current program
8700 counter is shown with a @code{=>} marker. For example:
8703 (@value{GDBP}) x/5i $pc-6
8704 0x804837f <main+11>: mov %esp,%ebp
8705 0x8048381 <main+13>: push %ecx
8706 0x8048382 <main+14>: sub $0x4,%esp
8707 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
8708 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
8711 @cindex @code{$_}, @code{$__}, and value history
8712 The addresses and contents printed by the @code{x} command are not saved
8713 in the value history because there is often too much of them and they
8714 would get in the way. Instead, @value{GDBN} makes these values available for
8715 subsequent use in expressions as values of the convenience variables
8716 @code{$_} and @code{$__}. After an @code{x} command, the last address
8717 examined is available for use in expressions in the convenience variable
8718 @code{$_}. The contents of that address, as examined, are available in
8719 the convenience variable @code{$__}.
8721 If the @code{x} command has a repeat count, the address and contents saved
8722 are from the last memory unit printed; this is not the same as the last
8723 address printed if several units were printed on the last line of output.
8725 @cindex remote memory comparison
8726 @cindex verify remote memory image
8727 When you are debugging a program running on a remote target machine
8728 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
8729 remote machine's memory against the executable file you downloaded to
8730 the target. The @code{compare-sections} command is provided for such
8734 @kindex compare-sections
8735 @item compare-sections @r{[}@var{section-name}@r{]}
8736 Compare the data of a loadable section @var{section-name} in the
8737 executable file of the program being debugged with the same section in
8738 the remote machine's memory, and report any mismatches. With no
8739 arguments, compares all loadable sections. This command's
8740 availability depends on the target's support for the @code{"qCRC"}
8745 @section Automatic Display
8746 @cindex automatic display
8747 @cindex display of expressions
8749 If you find that you want to print the value of an expression frequently
8750 (to see how it changes), you might want to add it to the @dfn{automatic
8751 display list} so that @value{GDBN} prints its value each time your program stops.
8752 Each expression added to the list is given a number to identify it;
8753 to remove an expression from the list, you specify that number.
8754 The automatic display looks like this:
8758 3: bar[5] = (struct hack *) 0x3804
8762 This display shows item numbers, expressions and their current values. As with
8763 displays you request manually using @code{x} or @code{print}, you can
8764 specify the output format you prefer; in fact, @code{display} decides
8765 whether to use @code{print} or @code{x} depending your format
8766 specification---it uses @code{x} if you specify either the @samp{i}
8767 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
8771 @item display @var{expr}
8772 Add the expression @var{expr} to the list of expressions to display
8773 each time your program stops. @xref{Expressions, ,Expressions}.
8775 @code{display} does not repeat if you press @key{RET} again after using it.
8777 @item display/@var{fmt} @var{expr}
8778 For @var{fmt} specifying only a display format and not a size or
8779 count, add the expression @var{expr} to the auto-display list but
8780 arrange to display it each time in the specified format @var{fmt}.
8781 @xref{Output Formats,,Output Formats}.
8783 @item display/@var{fmt} @var{addr}
8784 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
8785 number of units, add the expression @var{addr} as a memory address to
8786 be examined each time your program stops. Examining means in effect
8787 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
8790 For example, @samp{display/i $pc} can be helpful, to see the machine
8791 instruction about to be executed each time execution stops (@samp{$pc}
8792 is a common name for the program counter; @pxref{Registers, ,Registers}).
8795 @kindex delete display
8797 @item undisplay @var{dnums}@dots{}
8798 @itemx delete display @var{dnums}@dots{}
8799 Remove items from the list of expressions to display. Specify the
8800 numbers of the displays that you want affected with the command
8801 argument @var{dnums}. It can be a single display number, one of the
8802 numbers shown in the first field of the @samp{info display} display;
8803 or it could be a range of display numbers, as in @code{2-4}.
8805 @code{undisplay} does not repeat if you press @key{RET} after using it.
8806 (Otherwise you would just get the error @samp{No display number @dots{}}.)
8808 @kindex disable display
8809 @item disable display @var{dnums}@dots{}
8810 Disable the display of item numbers @var{dnums}. A disabled display
8811 item is not printed automatically, but is not forgotten. It may be
8812 enabled again later. Specify the numbers of the displays that you
8813 want affected with the command argument @var{dnums}. It can be a
8814 single display number, one of the numbers shown in the first field of
8815 the @samp{info display} display; or it could be a range of display
8816 numbers, as in @code{2-4}.
8818 @kindex enable display
8819 @item enable display @var{dnums}@dots{}
8820 Enable display of item numbers @var{dnums}. It becomes effective once
8821 again in auto display of its expression, until you specify otherwise.
8822 Specify the numbers of the displays that you want affected with the
8823 command argument @var{dnums}. It can be a single display number, one
8824 of the numbers shown in the first field of the @samp{info display}
8825 display; or it could be a range of display numbers, as in @code{2-4}.
8828 Display the current values of the expressions on the list, just as is
8829 done when your program stops.
8831 @kindex info display
8833 Print the list of expressions previously set up to display
8834 automatically, each one with its item number, but without showing the
8835 values. This includes disabled expressions, which are marked as such.
8836 It also includes expressions which would not be displayed right now
8837 because they refer to automatic variables not currently available.
8840 @cindex display disabled out of scope
8841 If a display expression refers to local variables, then it does not make
8842 sense outside the lexical context for which it was set up. Such an
8843 expression is disabled when execution enters a context where one of its
8844 variables is not defined. For example, if you give the command
8845 @code{display last_char} while inside a function with an argument
8846 @code{last_char}, @value{GDBN} displays this argument while your program
8847 continues to stop inside that function. When it stops elsewhere---where
8848 there is no variable @code{last_char}---the display is disabled
8849 automatically. The next time your program stops where @code{last_char}
8850 is meaningful, you can enable the display expression once again.
8852 @node Print Settings
8853 @section Print Settings
8855 @cindex format options
8856 @cindex print settings
8857 @value{GDBN} provides the following ways to control how arrays, structures,
8858 and symbols are printed.
8861 These settings are useful for debugging programs in any language:
8865 @item set print address
8866 @itemx set print address on
8867 @cindex print/don't print memory addresses
8868 @value{GDBN} prints memory addresses showing the location of stack
8869 traces, structure values, pointer values, breakpoints, and so forth,
8870 even when it also displays the contents of those addresses. The default
8871 is @code{on}. For example, this is what a stack frame display looks like with
8872 @code{set print address on}:
8877 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
8879 530 if (lquote != def_lquote)
8883 @item set print address off
8884 Do not print addresses when displaying their contents. For example,
8885 this is the same stack frame displayed with @code{set print address off}:
8889 (@value{GDBP}) set print addr off
8891 #0 set_quotes (lq="<<", rq=">>") at input.c:530
8892 530 if (lquote != def_lquote)
8896 You can use @samp{set print address off} to eliminate all machine
8897 dependent displays from the @value{GDBN} interface. For example, with
8898 @code{print address off}, you should get the same text for backtraces on
8899 all machines---whether or not they involve pointer arguments.
8902 @item show print address
8903 Show whether or not addresses are to be printed.
8906 When @value{GDBN} prints a symbolic address, it normally prints the
8907 closest earlier symbol plus an offset. If that symbol does not uniquely
8908 identify the address (for example, it is a name whose scope is a single
8909 source file), you may need to clarify. One way to do this is with
8910 @code{info line}, for example @samp{info line *0x4537}. Alternately,
8911 you can set @value{GDBN} to print the source file and line number when
8912 it prints a symbolic address:
8915 @item set print symbol-filename on
8916 @cindex source file and line of a symbol
8917 @cindex symbol, source file and line
8918 Tell @value{GDBN} to print the source file name and line number of a
8919 symbol in the symbolic form of an address.
8921 @item set print symbol-filename off
8922 Do not print source file name and line number of a symbol. This is the
8925 @item show print symbol-filename
8926 Show whether or not @value{GDBN} will print the source file name and
8927 line number of a symbol in the symbolic form of an address.
8930 Another situation where it is helpful to show symbol filenames and line
8931 numbers is when disassembling code; @value{GDBN} shows you the line
8932 number and source file that corresponds to each instruction.
8934 Also, you may wish to see the symbolic form only if the address being
8935 printed is reasonably close to the closest earlier symbol:
8938 @item set print max-symbolic-offset @var{max-offset}
8939 @itemx set print max-symbolic-offset unlimited
8940 @cindex maximum value for offset of closest symbol
8941 Tell @value{GDBN} to only display the symbolic form of an address if the
8942 offset between the closest earlier symbol and the address is less than
8943 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
8944 to always print the symbolic form of an address if any symbol precedes
8945 it. Zero is equivalent to @code{unlimited}.
8947 @item show print max-symbolic-offset
8948 Ask how large the maximum offset is that @value{GDBN} prints in a
8952 @cindex wild pointer, interpreting
8953 @cindex pointer, finding referent
8954 If you have a pointer and you are not sure where it points, try
8955 @samp{set print symbol-filename on}. Then you can determine the name
8956 and source file location of the variable where it points, using
8957 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
8958 For example, here @value{GDBN} shows that a variable @code{ptt} points
8959 at another variable @code{t}, defined in @file{hi2.c}:
8962 (@value{GDBP}) set print symbol-filename on
8963 (@value{GDBP}) p/a ptt
8964 $4 = 0xe008 <t in hi2.c>
8968 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
8969 does not show the symbol name and filename of the referent, even with
8970 the appropriate @code{set print} options turned on.
8973 You can also enable @samp{/a}-like formatting all the time using
8974 @samp{set print symbol on}:
8977 @item set print symbol on
8978 Tell @value{GDBN} to print the symbol corresponding to an address, if
8981 @item set print symbol off
8982 Tell @value{GDBN} not to print the symbol corresponding to an
8983 address. In this mode, @value{GDBN} will still print the symbol
8984 corresponding to pointers to functions. This is the default.
8986 @item show print symbol
8987 Show whether @value{GDBN} will display the symbol corresponding to an
8991 Other settings control how different kinds of objects are printed:
8994 @item set print array
8995 @itemx set print array on
8996 @cindex pretty print arrays
8997 Pretty print arrays. This format is more convenient to read,
8998 but uses more space. The default is off.
9000 @item set print array off
9001 Return to compressed format for arrays.
9003 @item show print array
9004 Show whether compressed or pretty format is selected for displaying
9007 @cindex print array indexes
9008 @item set print array-indexes
9009 @itemx set print array-indexes on
9010 Print the index of each element when displaying arrays. May be more
9011 convenient to locate a given element in the array or quickly find the
9012 index of a given element in that printed array. The default is off.
9014 @item set print array-indexes off
9015 Stop printing element indexes when displaying arrays.
9017 @item show print array-indexes
9018 Show whether the index of each element is printed when displaying
9021 @item set print elements @var{number-of-elements}
9022 @itemx set print elements unlimited
9023 @cindex number of array elements to print
9024 @cindex limit on number of printed array elements
9025 Set a limit on how many elements of an array @value{GDBN} will print.
9026 If @value{GDBN} is printing a large array, it stops printing after it has
9027 printed the number of elements set by the @code{set print elements} command.
9028 This limit also applies to the display of strings.
9029 When @value{GDBN} starts, this limit is set to 200.
9030 Setting @var{number-of-elements} to @code{unlimited} or zero means
9031 that the number of elements to print is unlimited.
9033 @item show print elements
9034 Display the number of elements of a large array that @value{GDBN} will print.
9035 If the number is 0, then the printing is unlimited.
9037 @item set print frame-arguments @var{value}
9038 @kindex set print frame-arguments
9039 @cindex printing frame argument values
9040 @cindex print all frame argument values
9041 @cindex print frame argument values for scalars only
9042 @cindex do not print frame argument values
9043 This command allows to control how the values of arguments are printed
9044 when the debugger prints a frame (@pxref{Frames}). The possible
9049 The values of all arguments are printed.
9052 Print the value of an argument only if it is a scalar. The value of more
9053 complex arguments such as arrays, structures, unions, etc, is replaced
9054 by @code{@dots{}}. This is the default. Here is an example where
9055 only scalar arguments are shown:
9058 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
9063 None of the argument values are printed. Instead, the value of each argument
9064 is replaced by @code{@dots{}}. In this case, the example above now becomes:
9067 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
9072 By default, only scalar arguments are printed. This command can be used
9073 to configure the debugger to print the value of all arguments, regardless
9074 of their type. However, it is often advantageous to not print the value
9075 of more complex parameters. For instance, it reduces the amount of
9076 information printed in each frame, making the backtrace more readable.
9077 Also, it improves performance when displaying Ada frames, because
9078 the computation of large arguments can sometimes be CPU-intensive,
9079 especially in large applications. Setting @code{print frame-arguments}
9080 to @code{scalars} (the default) or @code{none} avoids this computation,
9081 thus speeding up the display of each Ada frame.
9083 @item show print frame-arguments
9084 Show how the value of arguments should be displayed when printing a frame.
9086 @item set print raw frame-arguments on
9087 Print frame arguments in raw, non pretty-printed, form.
9089 @item set print raw frame-arguments off
9090 Print frame arguments in pretty-printed form, if there is a pretty-printer
9091 for the value (@pxref{Pretty Printing}),
9092 otherwise print the value in raw form.
9093 This is the default.
9095 @item show print raw frame-arguments
9096 Show whether to print frame arguments in raw form.
9098 @anchor{set print entry-values}
9099 @item set print entry-values @var{value}
9100 @kindex set print entry-values
9101 Set printing of frame argument values at function entry. In some cases
9102 @value{GDBN} can determine the value of function argument which was passed by
9103 the function caller, even if the value was modified inside the called function
9104 and therefore is different. With optimized code, the current value could be
9105 unavailable, but the entry value may still be known.
9107 The default value is @code{default} (see below for its description). Older
9108 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
9109 this feature will behave in the @code{default} setting the same way as with the
9112 This functionality is currently supported only by DWARF 2 debugging format and
9113 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
9114 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
9117 The @var{value} parameter can be one of the following:
9121 Print only actual parameter values, never print values from function entry
9125 #0 different (val=6)
9126 #0 lost (val=<optimized out>)
9128 #0 invalid (val=<optimized out>)
9132 Print only parameter values from function entry point. The actual parameter
9133 values are never printed.
9135 #0 equal (val@@entry=5)
9136 #0 different (val@@entry=5)
9137 #0 lost (val@@entry=5)
9138 #0 born (val@@entry=<optimized out>)
9139 #0 invalid (val@@entry=<optimized out>)
9143 Print only parameter values from function entry point. If value from function
9144 entry point is not known while the actual value is known, print the actual
9145 value for such parameter.
9147 #0 equal (val@@entry=5)
9148 #0 different (val@@entry=5)
9149 #0 lost (val@@entry=5)
9151 #0 invalid (val@@entry=<optimized out>)
9155 Print actual parameter values. If actual parameter value is not known while
9156 value from function entry point is known, print the entry point value for such
9160 #0 different (val=6)
9161 #0 lost (val@@entry=5)
9163 #0 invalid (val=<optimized out>)
9167 Always print both the actual parameter value and its value from function entry
9168 point, even if values of one or both are not available due to compiler
9171 #0 equal (val=5, val@@entry=5)
9172 #0 different (val=6, val@@entry=5)
9173 #0 lost (val=<optimized out>, val@@entry=5)
9174 #0 born (val=10, val@@entry=<optimized out>)
9175 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
9179 Print the actual parameter value if it is known and also its value from
9180 function entry point if it is known. If neither is known, print for the actual
9181 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
9182 values are known and identical, print the shortened
9183 @code{param=param@@entry=VALUE} notation.
9185 #0 equal (val=val@@entry=5)
9186 #0 different (val=6, val@@entry=5)
9187 #0 lost (val@@entry=5)
9189 #0 invalid (val=<optimized out>)
9193 Always print the actual parameter value. Print also its value from function
9194 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
9195 if both values are known and identical, print the shortened
9196 @code{param=param@@entry=VALUE} notation.
9198 #0 equal (val=val@@entry=5)
9199 #0 different (val=6, val@@entry=5)
9200 #0 lost (val=<optimized out>, val@@entry=5)
9202 #0 invalid (val=<optimized out>)
9206 For analysis messages on possible failures of frame argument values at function
9207 entry resolution see @ref{set debug entry-values}.
9209 @item show print entry-values
9210 Show the method being used for printing of frame argument values at function
9213 @item set print repeats @var{number-of-repeats}
9214 @itemx set print repeats unlimited
9215 @cindex repeated array elements
9216 Set the threshold for suppressing display of repeated array
9217 elements. When the number of consecutive identical elements of an
9218 array exceeds the threshold, @value{GDBN} prints the string
9219 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
9220 identical repetitions, instead of displaying the identical elements
9221 themselves. Setting the threshold to @code{unlimited} or zero will
9222 cause all elements to be individually printed. The default threshold
9225 @item show print repeats
9226 Display the current threshold for printing repeated identical
9229 @item set print null-stop
9230 @cindex @sc{null} elements in arrays
9231 Cause @value{GDBN} to stop printing the characters of an array when the first
9232 @sc{null} is encountered. This is useful when large arrays actually
9233 contain only short strings.
9236 @item show print null-stop
9237 Show whether @value{GDBN} stops printing an array on the first
9238 @sc{null} character.
9240 @item set print pretty on
9241 @cindex print structures in indented form
9242 @cindex indentation in structure display
9243 Cause @value{GDBN} to print structures in an indented format with one member
9244 per line, like this:
9259 @item set print pretty off
9260 Cause @value{GDBN} to print structures in a compact format, like this:
9264 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
9265 meat = 0x54 "Pork"@}
9270 This is the default format.
9272 @item show print pretty
9273 Show which format @value{GDBN} is using to print structures.
9275 @item set print sevenbit-strings on
9276 @cindex eight-bit characters in strings
9277 @cindex octal escapes in strings
9278 Print using only seven-bit characters; if this option is set,
9279 @value{GDBN} displays any eight-bit characters (in strings or
9280 character values) using the notation @code{\}@var{nnn}. This setting is
9281 best if you are working in English (@sc{ascii}) and you use the
9282 high-order bit of characters as a marker or ``meta'' bit.
9284 @item set print sevenbit-strings off
9285 Print full eight-bit characters. This allows the use of more
9286 international character sets, and is the default.
9288 @item show print sevenbit-strings
9289 Show whether or not @value{GDBN} is printing only seven-bit characters.
9291 @item set print union on
9292 @cindex unions in structures, printing
9293 Tell @value{GDBN} to print unions which are contained in structures
9294 and other unions. This is the default setting.
9296 @item set print union off
9297 Tell @value{GDBN} not to print unions which are contained in
9298 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
9301 @item show print union
9302 Ask @value{GDBN} whether or not it will print unions which are contained in
9303 structures and other unions.
9305 For example, given the declarations
9308 typedef enum @{Tree, Bug@} Species;
9309 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
9310 typedef enum @{Caterpillar, Cocoon, Butterfly@}
9321 struct thing foo = @{Tree, @{Acorn@}@};
9325 with @code{set print union on} in effect @samp{p foo} would print
9328 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
9332 and with @code{set print union off} in effect it would print
9335 $1 = @{it = Tree, form = @{...@}@}
9339 @code{set print union} affects programs written in C-like languages
9345 These settings are of interest when debugging C@t{++} programs:
9348 @cindex demangling C@t{++} names
9349 @item set print demangle
9350 @itemx set print demangle on
9351 Print C@t{++} names in their source form rather than in the encoded
9352 (``mangled'') form passed to the assembler and linker for type-safe
9353 linkage. The default is on.
9355 @item show print demangle
9356 Show whether C@t{++} names are printed in mangled or demangled form.
9358 @item set print asm-demangle
9359 @itemx set print asm-demangle on
9360 Print C@t{++} names in their source form rather than their mangled form, even
9361 in assembler code printouts such as instruction disassemblies.
9364 @item show print asm-demangle
9365 Show whether C@t{++} names in assembly listings are printed in mangled
9368 @cindex C@t{++} symbol decoding style
9369 @cindex symbol decoding style, C@t{++}
9370 @kindex set demangle-style
9371 @item set demangle-style @var{style}
9372 Choose among several encoding schemes used by different compilers to
9373 represent C@t{++} names. The choices for @var{style} are currently:
9377 Allow @value{GDBN} to choose a decoding style by inspecting your program.
9378 This is the default.
9381 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
9384 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
9387 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
9390 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
9391 @strong{Warning:} this setting alone is not sufficient to allow
9392 debugging @code{cfront}-generated executables. @value{GDBN} would
9393 require further enhancement to permit that.
9396 If you omit @var{style}, you will see a list of possible formats.
9398 @item show demangle-style
9399 Display the encoding style currently in use for decoding C@t{++} symbols.
9401 @item set print object
9402 @itemx set print object on
9403 @cindex derived type of an object, printing
9404 @cindex display derived types
9405 When displaying a pointer to an object, identify the @emph{actual}
9406 (derived) type of the object rather than the @emph{declared} type, using
9407 the virtual function table. Note that the virtual function table is
9408 required---this feature can only work for objects that have run-time
9409 type identification; a single virtual method in the object's declared
9410 type is sufficient. Note that this setting is also taken into account when
9411 working with variable objects via MI (@pxref{GDB/MI}).
9413 @item set print object off
9414 Display only the declared type of objects, without reference to the
9415 virtual function table. This is the default setting.
9417 @item show print object
9418 Show whether actual, or declared, object types are displayed.
9420 @item set print static-members
9421 @itemx set print static-members on
9422 @cindex static members of C@t{++} objects
9423 Print static members when displaying a C@t{++} object. The default is on.
9425 @item set print static-members off
9426 Do not print static members when displaying a C@t{++} object.
9428 @item show print static-members
9429 Show whether C@t{++} static members are printed or not.
9431 @item set print pascal_static-members
9432 @itemx set print pascal_static-members on
9433 @cindex static members of Pascal objects
9434 @cindex Pascal objects, static members display
9435 Print static members when displaying a Pascal object. The default is on.
9437 @item set print pascal_static-members off
9438 Do not print static members when displaying a Pascal object.
9440 @item show print pascal_static-members
9441 Show whether Pascal static members are printed or not.
9443 @c These don't work with HP ANSI C++ yet.
9444 @item set print vtbl
9445 @itemx set print vtbl on
9446 @cindex pretty print C@t{++} virtual function tables
9447 @cindex virtual functions (C@t{++}) display
9448 @cindex VTBL display
9449 Pretty print C@t{++} virtual function tables. The default is off.
9450 (The @code{vtbl} commands do not work on programs compiled with the HP
9451 ANSI C@t{++} compiler (@code{aCC}).)
9453 @item set print vtbl off
9454 Do not pretty print C@t{++} virtual function tables.
9456 @item show print vtbl
9457 Show whether C@t{++} virtual function tables are pretty printed, or not.
9460 @node Pretty Printing
9461 @section Pretty Printing
9463 @value{GDBN} provides a mechanism to allow pretty-printing of values using
9464 Python code. It greatly simplifies the display of complex objects. This
9465 mechanism works for both MI and the CLI.
9468 * Pretty-Printer Introduction:: Introduction to pretty-printers
9469 * Pretty-Printer Example:: An example pretty-printer
9470 * Pretty-Printer Commands:: Pretty-printer commands
9473 @node Pretty-Printer Introduction
9474 @subsection Pretty-Printer Introduction
9476 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
9477 registered for the value. If there is then @value{GDBN} invokes the
9478 pretty-printer to print the value. Otherwise the value is printed normally.
9480 Pretty-printers are normally named. This makes them easy to manage.
9481 The @samp{info pretty-printer} command will list all the installed
9482 pretty-printers with their names.
9483 If a pretty-printer can handle multiple data types, then its
9484 @dfn{subprinters} are the printers for the individual data types.
9485 Each such subprinter has its own name.
9486 The format of the name is @var{printer-name};@var{subprinter-name}.
9488 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
9489 Typically they are automatically loaded and registered when the corresponding
9490 debug information is loaded, thus making them available without having to
9491 do anything special.
9493 There are three places where a pretty-printer can be registered.
9497 Pretty-printers registered globally are available when debugging
9501 Pretty-printers registered with a program space are available only
9502 when debugging that program.
9503 @xref{Progspaces In Python}, for more details on program spaces in Python.
9506 Pretty-printers registered with an objfile are loaded and unloaded
9507 with the corresponding objfile (e.g., shared library).
9508 @xref{Objfiles In Python}, for more details on objfiles in Python.
9511 @xref{Selecting Pretty-Printers}, for further information on how
9512 pretty-printers are selected,
9514 @xref{Writing a Pretty-Printer}, for implementing pretty printers
9517 @node Pretty-Printer Example
9518 @subsection Pretty-Printer Example
9520 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
9523 (@value{GDBP}) print s
9525 static npos = 4294967295,
9527 <std::allocator<char>> = @{
9528 <__gnu_cxx::new_allocator<char>> = @{
9529 <No data fields>@}, <No data fields>
9531 members of std::basic_string<char, std::char_traits<char>,
9532 std::allocator<char> >::_Alloc_hider:
9533 _M_p = 0x804a014 "abcd"
9538 With a pretty-printer for @code{std::string} only the contents are printed:
9541 (@value{GDBP}) print s
9545 @node Pretty-Printer Commands
9546 @subsection Pretty-Printer Commands
9547 @cindex pretty-printer commands
9550 @kindex info pretty-printer
9551 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9552 Print the list of installed pretty-printers.
9553 This includes disabled pretty-printers, which are marked as such.
9555 @var{object-regexp} is a regular expression matching the objects
9556 whose pretty-printers to list.
9557 Objects can be @code{global}, the program space's file
9558 (@pxref{Progspaces In Python}),
9559 and the object files within that program space (@pxref{Objfiles In Python}).
9560 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
9561 looks up a printer from these three objects.
9563 @var{name-regexp} is a regular expression matching the name of the printers
9566 @kindex disable pretty-printer
9567 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9568 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9569 A disabled pretty-printer is not forgotten, it may be enabled again later.
9571 @kindex enable pretty-printer
9572 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9573 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9578 Suppose we have three pretty-printers installed: one from library1.so
9579 named @code{foo} that prints objects of type @code{foo}, and
9580 another from library2.so named @code{bar} that prints two types of objects,
9581 @code{bar1} and @code{bar2}.
9584 (gdb) info pretty-printer
9591 (gdb) info pretty-printer library2
9596 (gdb) disable pretty-printer library1
9598 2 of 3 printers enabled
9599 (gdb) info pretty-printer
9606 (gdb) disable pretty-printer library2 bar:bar1
9608 1 of 3 printers enabled
9609 (gdb) info pretty-printer library2
9616 (gdb) disable pretty-printer library2 bar
9618 0 of 3 printers enabled
9619 (gdb) info pretty-printer library2
9628 Note that for @code{bar} the entire printer can be disabled,
9629 as can each individual subprinter.
9632 @section Value History
9634 @cindex value history
9635 @cindex history of values printed by @value{GDBN}
9636 Values printed by the @code{print} command are saved in the @value{GDBN}
9637 @dfn{value history}. This allows you to refer to them in other expressions.
9638 Values are kept until the symbol table is re-read or discarded
9639 (for example with the @code{file} or @code{symbol-file} commands).
9640 When the symbol table changes, the value history is discarded,
9641 since the values may contain pointers back to the types defined in the
9646 @cindex history number
9647 The values printed are given @dfn{history numbers} by which you can
9648 refer to them. These are successive integers starting with one.
9649 @code{print} shows you the history number assigned to a value by
9650 printing @samp{$@var{num} = } before the value; here @var{num} is the
9653 To refer to any previous value, use @samp{$} followed by the value's
9654 history number. The way @code{print} labels its output is designed to
9655 remind you of this. Just @code{$} refers to the most recent value in
9656 the history, and @code{$$} refers to the value before that.
9657 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
9658 is the value just prior to @code{$$}, @code{$$1} is equivalent to
9659 @code{$$}, and @code{$$0} is equivalent to @code{$}.
9661 For example, suppose you have just printed a pointer to a structure and
9662 want to see the contents of the structure. It suffices to type
9668 If you have a chain of structures where the component @code{next} points
9669 to the next one, you can print the contents of the next one with this:
9676 You can print successive links in the chain by repeating this
9677 command---which you can do by just typing @key{RET}.
9679 Note that the history records values, not expressions. If the value of
9680 @code{x} is 4 and you type these commands:
9688 then the value recorded in the value history by the @code{print} command
9689 remains 4 even though the value of @code{x} has changed.
9694 Print the last ten values in the value history, with their item numbers.
9695 This is like @samp{p@ $$9} repeated ten times, except that @code{show
9696 values} does not change the history.
9698 @item show values @var{n}
9699 Print ten history values centered on history item number @var{n}.
9702 Print ten history values just after the values last printed. If no more
9703 values are available, @code{show values +} produces no display.
9706 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
9707 same effect as @samp{show values +}.
9709 @node Convenience Vars
9710 @section Convenience Variables
9712 @cindex convenience variables
9713 @cindex user-defined variables
9714 @value{GDBN} provides @dfn{convenience variables} that you can use within
9715 @value{GDBN} to hold on to a value and refer to it later. These variables
9716 exist entirely within @value{GDBN}; they are not part of your program, and
9717 setting a convenience variable has no direct effect on further execution
9718 of your program. That is why you can use them freely.
9720 Convenience variables are prefixed with @samp{$}. Any name preceded by
9721 @samp{$} can be used for a convenience variable, unless it is one of
9722 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
9723 (Value history references, in contrast, are @emph{numbers} preceded
9724 by @samp{$}. @xref{Value History, ,Value History}.)
9726 You can save a value in a convenience variable with an assignment
9727 expression, just as you would set a variable in your program.
9731 set $foo = *object_ptr
9735 would save in @code{$foo} the value contained in the object pointed to by
9738 Using a convenience variable for the first time creates it, but its
9739 value is @code{void} until you assign a new value. You can alter the
9740 value with another assignment at any time.
9742 Convenience variables have no fixed types. You can assign a convenience
9743 variable any type of value, including structures and arrays, even if
9744 that variable already has a value of a different type. The convenience
9745 variable, when used as an expression, has the type of its current value.
9748 @kindex show convenience
9749 @cindex show all user variables and functions
9750 @item show convenience
9751 Print a list of convenience variables used so far, and their values,
9752 as well as a list of the convenience functions.
9753 Abbreviated @code{show conv}.
9755 @kindex init-if-undefined
9756 @cindex convenience variables, initializing
9757 @item init-if-undefined $@var{variable} = @var{expression}
9758 Set a convenience variable if it has not already been set. This is useful
9759 for user-defined commands that keep some state. It is similar, in concept,
9760 to using local static variables with initializers in C (except that
9761 convenience variables are global). It can also be used to allow users to
9762 override default values used in a command script.
9764 If the variable is already defined then the expression is not evaluated so
9765 any side-effects do not occur.
9768 One of the ways to use a convenience variable is as a counter to be
9769 incremented or a pointer to be advanced. For example, to print
9770 a field from successive elements of an array of structures:
9774 print bar[$i++]->contents
9778 Repeat that command by typing @key{RET}.
9780 Some convenience variables are created automatically by @value{GDBN} and given
9781 values likely to be useful.
9784 @vindex $_@r{, convenience variable}
9786 The variable @code{$_} is automatically set by the @code{x} command to
9787 the last address examined (@pxref{Memory, ,Examining Memory}). Other
9788 commands which provide a default address for @code{x} to examine also
9789 set @code{$_} to that address; these commands include @code{info line}
9790 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
9791 except when set by the @code{x} command, in which case it is a pointer
9792 to the type of @code{$__}.
9794 @vindex $__@r{, convenience variable}
9796 The variable @code{$__} is automatically set by the @code{x} command
9797 to the value found in the last address examined. Its type is chosen
9798 to match the format in which the data was printed.
9801 @vindex $_exitcode@r{, convenience variable}
9802 When the program being debugged terminates normally, @value{GDBN}
9803 automatically sets this variable to the exit code of the program, and
9804 resets @code{$_exitsignal} to @code{void}.
9807 @vindex $_exitsignal@r{, convenience variable}
9808 When the program being debugged dies due to an uncaught signal,
9809 @value{GDBN} automatically sets this variable to that signal's number,
9810 and resets @code{$_exitcode} to @code{void}.
9812 To distinguish between whether the program being debugged has exited
9813 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
9814 @code{$_exitsignal} is not @code{void}), the convenience function
9815 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
9816 Functions}). For example, considering the following source code:
9822 main (int argc, char *argv[])
9829 A valid way of telling whether the program being debugged has exited
9830 or signalled would be:
9833 (@value{GDBP}) define has_exited_or_signalled
9834 Type commands for definition of ``has_exited_or_signalled''.
9835 End with a line saying just ``end''.
9836 >if $_isvoid ($_exitsignal)
9837 >echo The program has exited\n
9839 >echo The program has signalled\n
9845 Program terminated with signal SIGALRM, Alarm clock.
9846 The program no longer exists.
9847 (@value{GDBP}) has_exited_or_signalled
9848 The program has signalled
9851 As can be seen, @value{GDBN} correctly informs that the program being
9852 debugged has signalled, since it calls @code{raise} and raises a
9853 @code{SIGALRM} signal. If the program being debugged had not called
9854 @code{raise}, then @value{GDBN} would report a normal exit:
9857 (@value{GDBP}) has_exited_or_signalled
9858 The program has exited
9862 The variable @code{$_exception} is set to the exception object being
9863 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
9866 @itemx $_probe_arg0@dots{}$_probe_arg11
9867 Arguments to a static probe. @xref{Static Probe Points}.
9870 @vindex $_sdata@r{, inspect, convenience variable}
9871 The variable @code{$_sdata} contains extra collected static tracepoint
9872 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
9873 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
9874 if extra static tracepoint data has not been collected.
9877 @vindex $_siginfo@r{, convenience variable}
9878 The variable @code{$_siginfo} contains extra signal information
9879 (@pxref{extra signal information}). Note that @code{$_siginfo}
9880 could be empty, if the application has not yet received any signals.
9881 For example, it will be empty before you execute the @code{run} command.
9884 @vindex $_tlb@r{, convenience variable}
9885 The variable @code{$_tlb} is automatically set when debugging
9886 applications running on MS-Windows in native mode or connected to
9887 gdbserver that supports the @code{qGetTIBAddr} request.
9888 @xref{General Query Packets}.
9889 This variable contains the address of the thread information block.
9893 On HP-UX systems, if you refer to a function or variable name that
9894 begins with a dollar sign, @value{GDBN} searches for a user or system
9895 name first, before it searches for a convenience variable.
9897 @node Convenience Funs
9898 @section Convenience Functions
9900 @cindex convenience functions
9901 @value{GDBN} also supplies some @dfn{convenience functions}. These
9902 have a syntax similar to convenience variables. A convenience
9903 function can be used in an expression just like an ordinary function;
9904 however, a convenience function is implemented internally to
9907 These functions do not require @value{GDBN} to be configured with
9908 @code{Python} support, which means that they are always available.
9912 @item $_isvoid (@var{expr})
9913 @findex $_isvoid@r{, convenience function}
9914 Return one if the expression @var{expr} is @code{void}. Otherwise it
9917 A @code{void} expression is an expression where the type of the result
9918 is @code{void}. For example, you can examine a convenience variable
9919 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
9923 (@value{GDBP}) print $_exitcode
9925 (@value{GDBP}) print $_isvoid ($_exitcode)
9928 Starting program: ./a.out
9929 [Inferior 1 (process 29572) exited normally]
9930 (@value{GDBP}) print $_exitcode
9932 (@value{GDBP}) print $_isvoid ($_exitcode)
9936 In the example above, we used @code{$_isvoid} to check whether
9937 @code{$_exitcode} is @code{void} before and after the execution of the
9938 program being debugged. Before the execution there is no exit code to
9939 be examined, therefore @code{$_exitcode} is @code{void}. After the
9940 execution the program being debugged returned zero, therefore
9941 @code{$_exitcode} is zero, which means that it is not @code{void}
9944 The @code{void} expression can also be a call of a function from the
9945 program being debugged. For example, given the following function:
9954 The result of calling it inside @value{GDBN} is @code{void}:
9957 (@value{GDBP}) print foo ()
9959 (@value{GDBP}) print $_isvoid (foo ())
9961 (@value{GDBP}) set $v = foo ()
9962 (@value{GDBP}) print $v
9964 (@value{GDBP}) print $_isvoid ($v)
9970 These functions require @value{GDBN} to be configured with
9971 @code{Python} support.
9975 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
9976 @findex $_memeq@r{, convenience function}
9977 Returns one if the @var{length} bytes at the addresses given by
9978 @var{buf1} and @var{buf2} are equal.
9979 Otherwise it returns zero.
9981 @item $_regex(@var{str}, @var{regex})
9982 @findex $_regex@r{, convenience function}
9983 Returns one if the string @var{str} matches the regular expression
9984 @var{regex}. Otherwise it returns zero.
9985 The syntax of the regular expression is that specified by @code{Python}'s
9986 regular expression support.
9988 @item $_streq(@var{str1}, @var{str2})
9989 @findex $_streq@r{, convenience function}
9990 Returns one if the strings @var{str1} and @var{str2} are equal.
9991 Otherwise it returns zero.
9993 @item $_strlen(@var{str})
9994 @findex $_strlen@r{, convenience function}
9995 Returns the length of string @var{str}.
9999 @value{GDBN} provides the ability to list and get help on
10000 convenience functions.
10003 @item help function
10004 @kindex help function
10005 @cindex show all convenience functions
10006 Print a list of all convenience functions.
10013 You can refer to machine register contents, in expressions, as variables
10014 with names starting with @samp{$}. The names of registers are different
10015 for each machine; use @code{info registers} to see the names used on
10019 @kindex info registers
10020 @item info registers
10021 Print the names and values of all registers except floating-point
10022 and vector registers (in the selected stack frame).
10024 @kindex info all-registers
10025 @cindex floating point registers
10026 @item info all-registers
10027 Print the names and values of all registers, including floating-point
10028 and vector registers (in the selected stack frame).
10030 @item info registers @var{regname} @dots{}
10031 Print the @dfn{relativized} value of each specified register @var{regname}.
10032 As discussed in detail below, register values are normally relative to
10033 the selected stack frame. @var{regname} may be any register name valid on
10034 the machine you are using, with or without the initial @samp{$}.
10037 @cindex stack pointer register
10038 @cindex program counter register
10039 @cindex process status register
10040 @cindex frame pointer register
10041 @cindex standard registers
10042 @value{GDBN} has four ``standard'' register names that are available (in
10043 expressions) on most machines---whenever they do not conflict with an
10044 architecture's canonical mnemonics for registers. The register names
10045 @code{$pc} and @code{$sp} are used for the program counter register and
10046 the stack pointer. @code{$fp} is used for a register that contains a
10047 pointer to the current stack frame, and @code{$ps} is used for a
10048 register that contains the processor status. For example,
10049 you could print the program counter in hex with
10056 or print the instruction to be executed next with
10063 or add four to the stack pointer@footnote{This is a way of removing
10064 one word from the stack, on machines where stacks grow downward in
10065 memory (most machines, nowadays). This assumes that the innermost
10066 stack frame is selected; setting @code{$sp} is not allowed when other
10067 stack frames are selected. To pop entire frames off the stack,
10068 regardless of machine architecture, use @code{return};
10069 see @ref{Returning, ,Returning from a Function}.} with
10075 Whenever possible, these four standard register names are available on
10076 your machine even though the machine has different canonical mnemonics,
10077 so long as there is no conflict. The @code{info registers} command
10078 shows the canonical names. For example, on the SPARC, @code{info
10079 registers} displays the processor status register as @code{$psr} but you
10080 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
10081 is an alias for the @sc{eflags} register.
10083 @value{GDBN} always considers the contents of an ordinary register as an
10084 integer when the register is examined in this way. Some machines have
10085 special registers which can hold nothing but floating point; these
10086 registers are considered to have floating point values. There is no way
10087 to refer to the contents of an ordinary register as floating point value
10088 (although you can @emph{print} it as a floating point value with
10089 @samp{print/f $@var{regname}}).
10091 Some registers have distinct ``raw'' and ``virtual'' data formats. This
10092 means that the data format in which the register contents are saved by
10093 the operating system is not the same one that your program normally
10094 sees. For example, the registers of the 68881 floating point
10095 coprocessor are always saved in ``extended'' (raw) format, but all C
10096 programs expect to work with ``double'' (virtual) format. In such
10097 cases, @value{GDBN} normally works with the virtual format only (the format
10098 that makes sense for your program), but the @code{info registers} command
10099 prints the data in both formats.
10101 @cindex SSE registers (x86)
10102 @cindex MMX registers (x86)
10103 Some machines have special registers whose contents can be interpreted
10104 in several different ways. For example, modern x86-based machines
10105 have SSE and MMX registers that can hold several values packed
10106 together in several different formats. @value{GDBN} refers to such
10107 registers in @code{struct} notation:
10110 (@value{GDBP}) print $xmm1
10112 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
10113 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
10114 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
10115 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
10116 v4_int32 = @{0, 20657912, 11, 13@},
10117 v2_int64 = @{88725056443645952, 55834574859@},
10118 uint128 = 0x0000000d0000000b013b36f800000000
10123 To set values of such registers, you need to tell @value{GDBN} which
10124 view of the register you wish to change, as if you were assigning
10125 value to a @code{struct} member:
10128 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
10131 Normally, register values are relative to the selected stack frame
10132 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
10133 value that the register would contain if all stack frames farther in
10134 were exited and their saved registers restored. In order to see the
10135 true contents of hardware registers, you must select the innermost
10136 frame (with @samp{frame 0}).
10138 @cindex caller-saved registers
10139 @cindex call-clobbered registers
10140 @cindex volatile registers
10141 @cindex <not saved> values
10142 Usually ABIs reserve some registers as not needed to be saved by the
10143 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
10144 registers). It may therefore not be possible for @value{GDBN} to know
10145 the value a register had before the call (in other words, in the outer
10146 frame), if the register value has since been changed by the callee.
10147 @value{GDBN} tries to deduce where the inner frame saved
10148 (``callee-saved'') registers, from the debug info, unwind info, or the
10149 machine code generated by your compiler. If some register is not
10150 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
10151 its own knowledge of the ABI, or because the debug/unwind info
10152 explicitly says the register's value is undefined), @value{GDBN}
10153 displays @w{@samp{<not saved>}} as the register's value. With targets
10154 that @value{GDBN} has no knowledge of the register saving convention,
10155 if a register was not saved by the callee, then its value and location
10156 in the outer frame are assumed to be the same of the inner frame.
10157 This is usually harmless, because if the register is call-clobbered,
10158 the caller either does not care what is in the register after the
10159 call, or has code to restore the value that it does care about. Note,
10160 however, that if you change such a register in the outer frame, you
10161 may also be affecting the inner frame. Also, the more ``outer'' the
10162 frame is you're looking at, the more likely a call-clobbered
10163 register's value is to be wrong, in the sense that it doesn't actually
10164 represent the value the register had just before the call.
10166 @node Floating Point Hardware
10167 @section Floating Point Hardware
10168 @cindex floating point
10170 Depending on the configuration, @value{GDBN} may be able to give
10171 you more information about the status of the floating point hardware.
10176 Display hardware-dependent information about the floating
10177 point unit. The exact contents and layout vary depending on the
10178 floating point chip. Currently, @samp{info float} is supported on
10179 the ARM and x86 machines.
10183 @section Vector Unit
10184 @cindex vector unit
10186 Depending on the configuration, @value{GDBN} may be able to give you
10187 more information about the status of the vector unit.
10190 @kindex info vector
10192 Display information about the vector unit. The exact contents and
10193 layout vary depending on the hardware.
10196 @node OS Information
10197 @section Operating System Auxiliary Information
10198 @cindex OS information
10200 @value{GDBN} provides interfaces to useful OS facilities that can help
10201 you debug your program.
10203 @cindex auxiliary vector
10204 @cindex vector, auxiliary
10205 Some operating systems supply an @dfn{auxiliary vector} to programs at
10206 startup. This is akin to the arguments and environment that you
10207 specify for a program, but contains a system-dependent variety of
10208 binary values that tell system libraries important details about the
10209 hardware, operating system, and process. Each value's purpose is
10210 identified by an integer tag; the meanings are well-known but system-specific.
10211 Depending on the configuration and operating system facilities,
10212 @value{GDBN} may be able to show you this information. For remote
10213 targets, this functionality may further depend on the remote stub's
10214 support of the @samp{qXfer:auxv:read} packet, see
10215 @ref{qXfer auxiliary vector read}.
10220 Display the auxiliary vector of the inferior, which can be either a
10221 live process or a core dump file. @value{GDBN} prints each tag value
10222 numerically, and also shows names and text descriptions for recognized
10223 tags. Some values in the vector are numbers, some bit masks, and some
10224 pointers to strings or other data. @value{GDBN} displays each value in the
10225 most appropriate form for a recognized tag, and in hexadecimal for
10226 an unrecognized tag.
10229 On some targets, @value{GDBN} can access operating system-specific
10230 information and show it to you. The types of information available
10231 will differ depending on the type of operating system running on the
10232 target. The mechanism used to fetch the data is described in
10233 @ref{Operating System Information}. For remote targets, this
10234 functionality depends on the remote stub's support of the
10235 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
10239 @item info os @var{infotype}
10241 Display OS information of the requested type.
10243 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
10245 @anchor{linux info os infotypes}
10247 @kindex info os processes
10249 Display the list of processes on the target. For each process,
10250 @value{GDBN} prints the process identifier, the name of the user, the
10251 command corresponding to the process, and the list of processor cores
10252 that the process is currently running on. (To understand what these
10253 properties mean, for this and the following info types, please consult
10254 the general @sc{gnu}/Linux documentation.)
10256 @kindex info os procgroups
10258 Display the list of process groups on the target. For each process,
10259 @value{GDBN} prints the identifier of the process group that it belongs
10260 to, the command corresponding to the process group leader, the process
10261 identifier, and the command line of the process. The list is sorted
10262 first by the process group identifier, then by the process identifier,
10263 so that processes belonging to the same process group are grouped together
10264 and the process group leader is listed first.
10266 @kindex info os threads
10268 Display the list of threads running on the target. For each thread,
10269 @value{GDBN} prints the identifier of the process that the thread
10270 belongs to, the command of the process, the thread identifier, and the
10271 processor core that it is currently running on. The main thread of a
10272 process is not listed.
10274 @kindex info os files
10276 Display the list of open file descriptors on the target. For each
10277 file descriptor, @value{GDBN} prints the identifier of the process
10278 owning the descriptor, the command of the owning process, the value
10279 of the descriptor, and the target of the descriptor.
10281 @kindex info os sockets
10283 Display the list of Internet-domain sockets on the target. For each
10284 socket, @value{GDBN} prints the address and port of the local and
10285 remote endpoints, the current state of the connection, the creator of
10286 the socket, the IP address family of the socket, and the type of the
10289 @kindex info os shm
10291 Display the list of all System V shared-memory regions on the target.
10292 For each shared-memory region, @value{GDBN} prints the region key,
10293 the shared-memory identifier, the access permissions, the size of the
10294 region, the process that created the region, the process that last
10295 attached to or detached from the region, the current number of live
10296 attaches to the region, and the times at which the region was last
10297 attached to, detach from, and changed.
10299 @kindex info os semaphores
10301 Display the list of all System V semaphore sets on the target. For each
10302 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
10303 set identifier, the access permissions, the number of semaphores in the
10304 set, the user and group of the owner and creator of the semaphore set,
10305 and the times at which the semaphore set was operated upon and changed.
10307 @kindex info os msg
10309 Display the list of all System V message queues on the target. For each
10310 message queue, @value{GDBN} prints the message queue key, the message
10311 queue identifier, the access permissions, the current number of bytes
10312 on the queue, the current number of messages on the queue, the processes
10313 that last sent and received a message on the queue, the user and group
10314 of the owner and creator of the message queue, the times at which a
10315 message was last sent and received on the queue, and the time at which
10316 the message queue was last changed.
10318 @kindex info os modules
10320 Display the list of all loaded kernel modules on the target. For each
10321 module, @value{GDBN} prints the module name, the size of the module in
10322 bytes, the number of times the module is used, the dependencies of the
10323 module, the status of the module, and the address of the loaded module
10328 If @var{infotype} is omitted, then list the possible values for
10329 @var{infotype} and the kind of OS information available for each
10330 @var{infotype}. If the target does not return a list of possible
10331 types, this command will report an error.
10334 @node Memory Region Attributes
10335 @section Memory Region Attributes
10336 @cindex memory region attributes
10338 @dfn{Memory region attributes} allow you to describe special handling
10339 required by regions of your target's memory. @value{GDBN} uses
10340 attributes to determine whether to allow certain types of memory
10341 accesses; whether to use specific width accesses; and whether to cache
10342 target memory. By default the description of memory regions is
10343 fetched from the target (if the current target supports this), but the
10344 user can override the fetched regions.
10346 Defined memory regions can be individually enabled and disabled. When a
10347 memory region is disabled, @value{GDBN} uses the default attributes when
10348 accessing memory in that region. Similarly, if no memory regions have
10349 been defined, @value{GDBN} uses the default attributes when accessing
10352 When a memory region is defined, it is given a number to identify it;
10353 to enable, disable, or remove a memory region, you specify that number.
10357 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
10358 Define a memory region bounded by @var{lower} and @var{upper} with
10359 attributes @var{attributes}@dots{}, and add it to the list of regions
10360 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
10361 case: it is treated as the target's maximum memory address.
10362 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
10365 Discard any user changes to the memory regions and use target-supplied
10366 regions, if available, or no regions if the target does not support.
10369 @item delete mem @var{nums}@dots{}
10370 Remove memory regions @var{nums}@dots{} from the list of regions
10371 monitored by @value{GDBN}.
10373 @kindex disable mem
10374 @item disable mem @var{nums}@dots{}
10375 Disable monitoring of memory regions @var{nums}@dots{}.
10376 A disabled memory region is not forgotten.
10377 It may be enabled again later.
10380 @item enable mem @var{nums}@dots{}
10381 Enable monitoring of memory regions @var{nums}@dots{}.
10385 Print a table of all defined memory regions, with the following columns
10389 @item Memory Region Number
10390 @item Enabled or Disabled.
10391 Enabled memory regions are marked with @samp{y}.
10392 Disabled memory regions are marked with @samp{n}.
10395 The address defining the inclusive lower bound of the memory region.
10398 The address defining the exclusive upper bound of the memory region.
10401 The list of attributes set for this memory region.
10406 @subsection Attributes
10408 @subsubsection Memory Access Mode
10409 The access mode attributes set whether @value{GDBN} may make read or
10410 write accesses to a memory region.
10412 While these attributes prevent @value{GDBN} from performing invalid
10413 memory accesses, they do nothing to prevent the target system, I/O DMA,
10414 etc.@: from accessing memory.
10418 Memory is read only.
10420 Memory is write only.
10422 Memory is read/write. This is the default.
10425 @subsubsection Memory Access Size
10426 The access size attribute tells @value{GDBN} to use specific sized
10427 accesses in the memory region. Often memory mapped device registers
10428 require specific sized accesses. If no access size attribute is
10429 specified, @value{GDBN} may use accesses of any size.
10433 Use 8 bit memory accesses.
10435 Use 16 bit memory accesses.
10437 Use 32 bit memory accesses.
10439 Use 64 bit memory accesses.
10442 @c @subsubsection Hardware/Software Breakpoints
10443 @c The hardware/software breakpoint attributes set whether @value{GDBN}
10444 @c will use hardware or software breakpoints for the internal breakpoints
10445 @c used by the step, next, finish, until, etc. commands.
10449 @c Always use hardware breakpoints
10450 @c @item swbreak (default)
10453 @subsubsection Data Cache
10454 The data cache attributes set whether @value{GDBN} will cache target
10455 memory. While this generally improves performance by reducing debug
10456 protocol overhead, it can lead to incorrect results because @value{GDBN}
10457 does not know about volatile variables or memory mapped device
10462 Enable @value{GDBN} to cache target memory.
10464 Disable @value{GDBN} from caching target memory. This is the default.
10467 @subsection Memory Access Checking
10468 @value{GDBN} can be instructed to refuse accesses to memory that is
10469 not explicitly described. This can be useful if accessing such
10470 regions has undesired effects for a specific target, or to provide
10471 better error checking. The following commands control this behaviour.
10474 @kindex set mem inaccessible-by-default
10475 @item set mem inaccessible-by-default [on|off]
10476 If @code{on} is specified, make @value{GDBN} treat memory not
10477 explicitly described by the memory ranges as non-existent and refuse accesses
10478 to such memory. The checks are only performed if there's at least one
10479 memory range defined. If @code{off} is specified, make @value{GDBN}
10480 treat the memory not explicitly described by the memory ranges as RAM.
10481 The default value is @code{on}.
10482 @kindex show mem inaccessible-by-default
10483 @item show mem inaccessible-by-default
10484 Show the current handling of accesses to unknown memory.
10488 @c @subsubsection Memory Write Verification
10489 @c The memory write verification attributes set whether @value{GDBN}
10490 @c will re-reads data after each write to verify the write was successful.
10494 @c @item noverify (default)
10497 @node Dump/Restore Files
10498 @section Copy Between Memory and a File
10499 @cindex dump/restore files
10500 @cindex append data to a file
10501 @cindex dump data to a file
10502 @cindex restore data from a file
10504 You can use the commands @code{dump}, @code{append}, and
10505 @code{restore} to copy data between target memory and a file. The
10506 @code{dump} and @code{append} commands write data to a file, and the
10507 @code{restore} command reads data from a file back into the inferior's
10508 memory. Files may be in binary, Motorola S-record, Intel hex, or
10509 Tektronix Hex format; however, @value{GDBN} can only append to binary
10515 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
10516 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
10517 Dump the contents of memory from @var{start_addr} to @var{end_addr},
10518 or the value of @var{expr}, to @var{filename} in the given format.
10520 The @var{format} parameter may be any one of:
10527 Motorola S-record format.
10529 Tektronix Hex format.
10532 @value{GDBN} uses the same definitions of these formats as the
10533 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
10534 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
10538 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
10539 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
10540 Append the contents of memory from @var{start_addr} to @var{end_addr},
10541 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
10542 (@value{GDBN} can only append data to files in raw binary form.)
10545 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
10546 Restore the contents of file @var{filename} into memory. The
10547 @code{restore} command can automatically recognize any known @sc{bfd}
10548 file format, except for raw binary. To restore a raw binary file you
10549 must specify the optional keyword @code{binary} after the filename.
10551 If @var{bias} is non-zero, its value will be added to the addresses
10552 contained in the file. Binary files always start at address zero, so
10553 they will be restored at address @var{bias}. Other bfd files have
10554 a built-in location; they will be restored at offset @var{bias}
10555 from that location.
10557 If @var{start} and/or @var{end} are non-zero, then only data between
10558 file offset @var{start} and file offset @var{end} will be restored.
10559 These offsets are relative to the addresses in the file, before
10560 the @var{bias} argument is applied.
10564 @node Core File Generation
10565 @section How to Produce a Core File from Your Program
10566 @cindex dump core from inferior
10568 A @dfn{core file} or @dfn{core dump} is a file that records the memory
10569 image of a running process and its process status (register values
10570 etc.). Its primary use is post-mortem debugging of a program that
10571 crashed while it ran outside a debugger. A program that crashes
10572 automatically produces a core file, unless this feature is disabled by
10573 the user. @xref{Files}, for information on invoking @value{GDBN} in
10574 the post-mortem debugging mode.
10576 Occasionally, you may wish to produce a core file of the program you
10577 are debugging in order to preserve a snapshot of its state.
10578 @value{GDBN} has a special command for that.
10582 @kindex generate-core-file
10583 @item generate-core-file [@var{file}]
10584 @itemx gcore [@var{file}]
10585 Produce a core dump of the inferior process. The optional argument
10586 @var{file} specifies the file name where to put the core dump. If not
10587 specified, the file name defaults to @file{core.@var{pid}}, where
10588 @var{pid} is the inferior process ID.
10590 Note that this command is implemented only for some systems (as of
10591 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
10594 @node Character Sets
10595 @section Character Sets
10596 @cindex character sets
10598 @cindex translating between character sets
10599 @cindex host character set
10600 @cindex target character set
10602 If the program you are debugging uses a different character set to
10603 represent characters and strings than the one @value{GDBN} uses itself,
10604 @value{GDBN} can automatically translate between the character sets for
10605 you. The character set @value{GDBN} uses we call the @dfn{host
10606 character set}; the one the inferior program uses we call the
10607 @dfn{target character set}.
10609 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
10610 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
10611 remote protocol (@pxref{Remote Debugging}) to debug a program
10612 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
10613 then the host character set is Latin-1, and the target character set is
10614 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
10615 target-charset EBCDIC-US}, then @value{GDBN} translates between
10616 @sc{ebcdic} and Latin 1 as you print character or string values, or use
10617 character and string literals in expressions.
10619 @value{GDBN} has no way to automatically recognize which character set
10620 the inferior program uses; you must tell it, using the @code{set
10621 target-charset} command, described below.
10623 Here are the commands for controlling @value{GDBN}'s character set
10627 @item set target-charset @var{charset}
10628 @kindex set target-charset
10629 Set the current target character set to @var{charset}. To display the
10630 list of supported target character sets, type
10631 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
10633 @item set host-charset @var{charset}
10634 @kindex set host-charset
10635 Set the current host character set to @var{charset}.
10637 By default, @value{GDBN} uses a host character set appropriate to the
10638 system it is running on; you can override that default using the
10639 @code{set host-charset} command. On some systems, @value{GDBN} cannot
10640 automatically determine the appropriate host character set. In this
10641 case, @value{GDBN} uses @samp{UTF-8}.
10643 @value{GDBN} can only use certain character sets as its host character
10644 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
10645 @value{GDBN} will list the host character sets it supports.
10647 @item set charset @var{charset}
10648 @kindex set charset
10649 Set the current host and target character sets to @var{charset}. As
10650 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
10651 @value{GDBN} will list the names of the character sets that can be used
10652 for both host and target.
10655 @kindex show charset
10656 Show the names of the current host and target character sets.
10658 @item show host-charset
10659 @kindex show host-charset
10660 Show the name of the current host character set.
10662 @item show target-charset
10663 @kindex show target-charset
10664 Show the name of the current target character set.
10666 @item set target-wide-charset @var{charset}
10667 @kindex set target-wide-charset
10668 Set the current target's wide character set to @var{charset}. This is
10669 the character set used by the target's @code{wchar_t} type. To
10670 display the list of supported wide character sets, type
10671 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
10673 @item show target-wide-charset
10674 @kindex show target-wide-charset
10675 Show the name of the current target's wide character set.
10678 Here is an example of @value{GDBN}'s character set support in action.
10679 Assume that the following source code has been placed in the file
10680 @file{charset-test.c}:
10686 = @{72, 101, 108, 108, 111, 44, 32, 119,
10687 111, 114, 108, 100, 33, 10, 0@};
10688 char ibm1047_hello[]
10689 = @{200, 133, 147, 147, 150, 107, 64, 166,
10690 150, 153, 147, 132, 90, 37, 0@};
10694 printf ("Hello, world!\n");
10698 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
10699 containing the string @samp{Hello, world!} followed by a newline,
10700 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
10702 We compile the program, and invoke the debugger on it:
10705 $ gcc -g charset-test.c -o charset-test
10706 $ gdb -nw charset-test
10707 GNU gdb 2001-12-19-cvs
10708 Copyright 2001 Free Software Foundation, Inc.
10713 We can use the @code{show charset} command to see what character sets
10714 @value{GDBN} is currently using to interpret and display characters and
10718 (@value{GDBP}) show charset
10719 The current host and target character set is `ISO-8859-1'.
10723 For the sake of printing this manual, let's use @sc{ascii} as our
10724 initial character set:
10726 (@value{GDBP}) set charset ASCII
10727 (@value{GDBP}) show charset
10728 The current host and target character set is `ASCII'.
10732 Let's assume that @sc{ascii} is indeed the correct character set for our
10733 host system --- in other words, let's assume that if @value{GDBN} prints
10734 characters using the @sc{ascii} character set, our terminal will display
10735 them properly. Since our current target character set is also
10736 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
10739 (@value{GDBP}) print ascii_hello
10740 $1 = 0x401698 "Hello, world!\n"
10741 (@value{GDBP}) print ascii_hello[0]
10746 @value{GDBN} uses the target character set for character and string
10747 literals you use in expressions:
10750 (@value{GDBP}) print '+'
10755 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
10758 @value{GDBN} relies on the user to tell it which character set the
10759 target program uses. If we print @code{ibm1047_hello} while our target
10760 character set is still @sc{ascii}, we get jibberish:
10763 (@value{GDBP}) print ibm1047_hello
10764 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
10765 (@value{GDBP}) print ibm1047_hello[0]
10770 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
10771 @value{GDBN} tells us the character sets it supports:
10774 (@value{GDBP}) set target-charset
10775 ASCII EBCDIC-US IBM1047 ISO-8859-1
10776 (@value{GDBP}) set target-charset
10779 We can select @sc{ibm1047} as our target character set, and examine the
10780 program's strings again. Now the @sc{ascii} string is wrong, but
10781 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
10782 target character set, @sc{ibm1047}, to the host character set,
10783 @sc{ascii}, and they display correctly:
10786 (@value{GDBP}) set target-charset IBM1047
10787 (@value{GDBP}) show charset
10788 The current host character set is `ASCII'.
10789 The current target character set is `IBM1047'.
10790 (@value{GDBP}) print ascii_hello
10791 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
10792 (@value{GDBP}) print ascii_hello[0]
10794 (@value{GDBP}) print ibm1047_hello
10795 $8 = 0x4016a8 "Hello, world!\n"
10796 (@value{GDBP}) print ibm1047_hello[0]
10801 As above, @value{GDBN} uses the target character set for character and
10802 string literals you use in expressions:
10805 (@value{GDBP}) print '+'
10810 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
10813 @node Caching Remote Data
10814 @section Caching Data of Remote Targets
10815 @cindex caching data of remote targets
10817 @value{GDBN} caches data exchanged between the debugger and a
10818 remote target (@pxref{Remote Debugging}). Such caching generally improves
10819 performance, because it reduces the overhead of the remote protocol by
10820 bundling memory reads and writes into large chunks. Unfortunately, simply
10821 caching everything would lead to incorrect results, since @value{GDBN}
10822 does not necessarily know anything about volatile values, memory-mapped I/O
10823 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
10824 memory can be changed @emph{while} a gdb command is executing.
10825 Therefore, by default, @value{GDBN} only caches data
10826 known to be on the stack@footnote{In non-stop mode, it is moderately
10827 rare for a running thread to modify the stack of a stopped thread
10828 in a way that would interfere with a backtrace, and caching of
10829 stack reads provides a significant speed up of remote backtraces.}.
10830 Other regions of memory can be explicitly marked as
10831 cacheable; see @pxref{Memory Region Attributes}.
10834 @kindex set remotecache
10835 @item set remotecache on
10836 @itemx set remotecache off
10837 This option no longer does anything; it exists for compatibility
10840 @kindex show remotecache
10841 @item show remotecache
10842 Show the current state of the obsolete remotecache flag.
10844 @kindex set stack-cache
10845 @item set stack-cache on
10846 @itemx set stack-cache off
10847 Enable or disable caching of stack accesses. When @code{ON}, use
10848 caching. By default, this option is @code{ON}.
10850 @kindex show stack-cache
10851 @item show stack-cache
10852 Show the current state of data caching for memory accesses.
10854 @kindex info dcache
10855 @item info dcache @r{[}line@r{]}
10856 Print the information about the data cache performance. The
10857 information displayed includes the dcache width and depth, and for
10858 each cache line, its number, address, and how many times it was
10859 referenced. This command is useful for debugging the data cache
10862 If a line number is specified, the contents of that line will be
10865 @item set dcache size @var{size}
10866 @cindex dcache size
10867 @kindex set dcache size
10868 Set maximum number of entries in dcache (dcache depth above).
10870 @item set dcache line-size @var{line-size}
10871 @cindex dcache line-size
10872 @kindex set dcache line-size
10873 Set number of bytes each dcache entry caches (dcache width above).
10874 Must be a power of 2.
10876 @item show dcache size
10877 @kindex show dcache size
10878 Show maximum number of dcache entries. See also @ref{Caching Remote Data, info dcache}.
10880 @item show dcache line-size
10881 @kindex show dcache line-size
10882 Show default size of dcache lines. See also @ref{Caching Remote Data, info dcache}.
10886 @node Searching Memory
10887 @section Search Memory
10888 @cindex searching memory
10890 Memory can be searched for a particular sequence of bytes with the
10891 @code{find} command.
10895 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
10896 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
10897 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
10898 etc. The search begins at address @var{start_addr} and continues for either
10899 @var{len} bytes or through to @var{end_addr} inclusive.
10902 @var{s} and @var{n} are optional parameters.
10903 They may be specified in either order, apart or together.
10906 @item @var{s}, search query size
10907 The size of each search query value.
10913 halfwords (two bytes)
10917 giant words (eight bytes)
10920 All values are interpreted in the current language.
10921 This means, for example, that if the current source language is C/C@t{++}
10922 then searching for the string ``hello'' includes the trailing '\0'.
10924 If the value size is not specified, it is taken from the
10925 value's type in the current language.
10926 This is useful when one wants to specify the search
10927 pattern as a mixture of types.
10928 Note that this means, for example, that in the case of C-like languages
10929 a search for an untyped 0x42 will search for @samp{(int) 0x42}
10930 which is typically four bytes.
10932 @item @var{n}, maximum number of finds
10933 The maximum number of matches to print. The default is to print all finds.
10936 You can use strings as search values. Quote them with double-quotes
10938 The string value is copied into the search pattern byte by byte,
10939 regardless of the endianness of the target and the size specification.
10941 The address of each match found is printed as well as a count of the
10942 number of matches found.
10944 The address of the last value found is stored in convenience variable
10946 A count of the number of matches is stored in @samp{$numfound}.
10948 For example, if stopped at the @code{printf} in this function:
10954 static char hello[] = "hello-hello";
10955 static struct @{ char c; short s; int i; @}
10956 __attribute__ ((packed)) mixed
10957 = @{ 'c', 0x1234, 0x87654321 @};
10958 printf ("%s\n", hello);
10963 you get during debugging:
10966 (gdb) find &hello[0], +sizeof(hello), "hello"
10967 0x804956d <hello.1620+6>
10969 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
10970 0x8049567 <hello.1620>
10971 0x804956d <hello.1620+6>
10973 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
10974 0x8049567 <hello.1620>
10976 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
10977 0x8049560 <mixed.1625>
10979 (gdb) print $numfound
10982 $2 = (void *) 0x8049560
10985 @node Optimized Code
10986 @chapter Debugging Optimized Code
10987 @cindex optimized code, debugging
10988 @cindex debugging optimized code
10990 Almost all compilers support optimization. With optimization
10991 disabled, the compiler generates assembly code that corresponds
10992 directly to your source code, in a simplistic way. As the compiler
10993 applies more powerful optimizations, the generated assembly code
10994 diverges from your original source code. With help from debugging
10995 information generated by the compiler, @value{GDBN} can map from
10996 the running program back to constructs from your original source.
10998 @value{GDBN} is more accurate with optimization disabled. If you
10999 can recompile without optimization, it is easier to follow the
11000 progress of your program during debugging. But, there are many cases
11001 where you may need to debug an optimized version.
11003 When you debug a program compiled with @samp{-g -O}, remember that the
11004 optimizer has rearranged your code; the debugger shows you what is
11005 really there. Do not be too surprised when the execution path does not
11006 exactly match your source file! An extreme example: if you define a
11007 variable, but never use it, @value{GDBN} never sees that
11008 variable---because the compiler optimizes it out of existence.
11010 Some things do not work as well with @samp{-g -O} as with just
11011 @samp{-g}, particularly on machines with instruction scheduling. If in
11012 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
11013 please report it to us as a bug (including a test case!).
11014 @xref{Variables}, for more information about debugging optimized code.
11017 * Inline Functions:: How @value{GDBN} presents inlining
11018 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
11021 @node Inline Functions
11022 @section Inline Functions
11023 @cindex inline functions, debugging
11025 @dfn{Inlining} is an optimization that inserts a copy of the function
11026 body directly at each call site, instead of jumping to a shared
11027 routine. @value{GDBN} displays inlined functions just like
11028 non-inlined functions. They appear in backtraces. You can view their
11029 arguments and local variables, step into them with @code{step}, skip
11030 them with @code{next}, and escape from them with @code{finish}.
11031 You can check whether a function was inlined by using the
11032 @code{info frame} command.
11034 For @value{GDBN} to support inlined functions, the compiler must
11035 record information about inlining in the debug information ---
11036 @value{NGCC} using the @sc{dwarf 2} format does this, and several
11037 other compilers do also. @value{GDBN} only supports inlined functions
11038 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
11039 do not emit two required attributes (@samp{DW_AT_call_file} and
11040 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
11041 function calls with earlier versions of @value{NGCC}. It instead
11042 displays the arguments and local variables of inlined functions as
11043 local variables in the caller.
11045 The body of an inlined function is directly included at its call site;
11046 unlike a non-inlined function, there are no instructions devoted to
11047 the call. @value{GDBN} still pretends that the call site and the
11048 start of the inlined function are different instructions. Stepping to
11049 the call site shows the call site, and then stepping again shows
11050 the first line of the inlined function, even though no additional
11051 instructions are executed.
11053 This makes source-level debugging much clearer; you can see both the
11054 context of the call and then the effect of the call. Only stepping by
11055 a single instruction using @code{stepi} or @code{nexti} does not do
11056 this; single instruction steps always show the inlined body.
11058 There are some ways that @value{GDBN} does not pretend that inlined
11059 function calls are the same as normal calls:
11063 Setting breakpoints at the call site of an inlined function may not
11064 work, because the call site does not contain any code. @value{GDBN}
11065 may incorrectly move the breakpoint to the next line of the enclosing
11066 function, after the call. This limitation will be removed in a future
11067 version of @value{GDBN}; until then, set a breakpoint on an earlier line
11068 or inside the inlined function instead.
11071 @value{GDBN} cannot locate the return value of inlined calls after
11072 using the @code{finish} command. This is a limitation of compiler-generated
11073 debugging information; after @code{finish}, you can step to the next line
11074 and print a variable where your program stored the return value.
11078 @node Tail Call Frames
11079 @section Tail Call Frames
11080 @cindex tail call frames, debugging
11082 Function @code{B} can call function @code{C} in its very last statement. In
11083 unoptimized compilation the call of @code{C} is immediately followed by return
11084 instruction at the end of @code{B} code. Optimizing compiler may replace the
11085 call and return in function @code{B} into one jump to function @code{C}
11086 instead. Such use of a jump instruction is called @dfn{tail call}.
11088 During execution of function @code{C}, there will be no indication in the
11089 function call stack frames that it was tail-called from @code{B}. If function
11090 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
11091 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
11092 some cases @value{GDBN} can determine that @code{C} was tail-called from
11093 @code{B}, and it will then create fictitious call frame for that, with the
11094 return address set up as if @code{B} called @code{C} normally.
11096 This functionality is currently supported only by DWARF 2 debugging format and
11097 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
11098 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
11101 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
11102 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
11106 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
11108 Stack level 1, frame at 0x7fffffffda30:
11109 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
11110 tail call frame, caller of frame at 0x7fffffffda30
11111 source language c++.
11112 Arglist at unknown address.
11113 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
11116 The detection of all the possible code path executions can find them ambiguous.
11117 There is no execution history stored (possible @ref{Reverse Execution} is never
11118 used for this purpose) and the last known caller could have reached the known
11119 callee by multiple different jump sequences. In such case @value{GDBN} still
11120 tries to show at least all the unambiguous top tail callers and all the
11121 unambiguous bottom tail calees, if any.
11124 @anchor{set debug entry-values}
11125 @item set debug entry-values
11126 @kindex set debug entry-values
11127 When set to on, enables printing of analysis messages for both frame argument
11128 values at function entry and tail calls. It will show all the possible valid
11129 tail calls code paths it has considered. It will also print the intersection
11130 of them with the final unambiguous (possibly partial or even empty) code path
11133 @item show debug entry-values
11134 @kindex show debug entry-values
11135 Show the current state of analysis messages printing for both frame argument
11136 values at function entry and tail calls.
11139 The analysis messages for tail calls can for example show why the virtual tail
11140 call frame for function @code{c} has not been recognized (due to the indirect
11141 reference by variable @code{x}):
11144 static void __attribute__((noinline, noclone)) c (void);
11145 void (*x) (void) = c;
11146 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
11147 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
11148 int main (void) @{ x (); return 0; @}
11150 Breakpoint 1, DW_OP_GNU_entry_value resolving cannot find
11151 DW_TAG_GNU_call_site 0x40039a in main
11153 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
11156 #1 0x000000000040039a in main () at t.c:5
11159 Another possibility is an ambiguous virtual tail call frames resolution:
11163 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
11164 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
11165 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
11166 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
11167 static void __attribute__((noinline, noclone)) b (void)
11168 @{ if (i) c (); else e (); @}
11169 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
11170 int main (void) @{ a (); return 0; @}
11172 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
11173 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
11174 tailcall: reduced: 0x4004d2(a) |
11177 #1 0x00000000004004d2 in a () at t.c:8
11178 #2 0x0000000000400395 in main () at t.c:9
11181 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
11182 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
11184 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
11185 @ifset HAVE_MAKEINFO_CLICK
11186 @set ARROW @click{}
11187 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
11188 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
11190 @ifclear HAVE_MAKEINFO_CLICK
11192 @set CALLSEQ1B @value{CALLSEQ1A}
11193 @set CALLSEQ2B @value{CALLSEQ2A}
11196 Frames #0 and #2 are real, #1 is a virtual tail call frame.
11197 The code can have possible execution paths @value{CALLSEQ1B} or
11198 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
11200 @code{initial:} state shows some random possible calling sequence @value{GDBN}
11201 has found. It then finds another possible calling sequcen - that one is
11202 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
11203 printed as the @code{reduced:} calling sequence. That one could have many
11204 futher @code{compare:} and @code{reduced:} statements as long as there remain
11205 any non-ambiguous sequence entries.
11207 For the frame of function @code{b} in both cases there are different possible
11208 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
11209 also ambigous. The only non-ambiguous frame is the one for function @code{a},
11210 therefore this one is displayed to the user while the ambiguous frames are
11213 There can be also reasons why printing of frame argument values at function
11218 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
11219 static void __attribute__((noinline, noclone)) a (int i);
11220 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
11221 static void __attribute__((noinline, noclone)) a (int i)
11222 @{ if (i) b (i - 1); else c (0); @}
11223 int main (void) @{ a (5); return 0; @}
11226 #0 c (i=i@@entry=0) at t.c:2
11227 #1 0x0000000000400428 in a (DW_OP_GNU_entry_value resolving has found
11228 function "a" at 0x400420 can call itself via tail calls
11229 i=<optimized out>) at t.c:6
11230 #2 0x000000000040036e in main () at t.c:7
11233 @value{GDBN} cannot find out from the inferior state if and how many times did
11234 function @code{a} call itself (via function @code{b}) as these calls would be
11235 tail calls. Such tail calls would modify thue @code{i} variable, therefore
11236 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
11237 prints @code{<optimized out>} instead.
11240 @chapter C Preprocessor Macros
11242 Some languages, such as C and C@t{++}, provide a way to define and invoke
11243 ``preprocessor macros'' which expand into strings of tokens.
11244 @value{GDBN} can evaluate expressions containing macro invocations, show
11245 the result of macro expansion, and show a macro's definition, including
11246 where it was defined.
11248 You may need to compile your program specially to provide @value{GDBN}
11249 with information about preprocessor macros. Most compilers do not
11250 include macros in their debugging information, even when you compile
11251 with the @option{-g} flag. @xref{Compilation}.
11253 A program may define a macro at one point, remove that definition later,
11254 and then provide a different definition after that. Thus, at different
11255 points in the program, a macro may have different definitions, or have
11256 no definition at all. If there is a current stack frame, @value{GDBN}
11257 uses the macros in scope at that frame's source code line. Otherwise,
11258 @value{GDBN} uses the macros in scope at the current listing location;
11261 Whenever @value{GDBN} evaluates an expression, it always expands any
11262 macro invocations present in the expression. @value{GDBN} also provides
11263 the following commands for working with macros explicitly.
11267 @kindex macro expand
11268 @cindex macro expansion, showing the results of preprocessor
11269 @cindex preprocessor macro expansion, showing the results of
11270 @cindex expanding preprocessor macros
11271 @item macro expand @var{expression}
11272 @itemx macro exp @var{expression}
11273 Show the results of expanding all preprocessor macro invocations in
11274 @var{expression}. Since @value{GDBN} simply expands macros, but does
11275 not parse the result, @var{expression} need not be a valid expression;
11276 it can be any string of tokens.
11279 @item macro expand-once @var{expression}
11280 @itemx macro exp1 @var{expression}
11281 @cindex expand macro once
11282 @i{(This command is not yet implemented.)} Show the results of
11283 expanding those preprocessor macro invocations that appear explicitly in
11284 @var{expression}. Macro invocations appearing in that expansion are
11285 left unchanged. This command allows you to see the effect of a
11286 particular macro more clearly, without being confused by further
11287 expansions. Since @value{GDBN} simply expands macros, but does not
11288 parse the result, @var{expression} need not be a valid expression; it
11289 can be any string of tokens.
11292 @cindex macro definition, showing
11293 @cindex definition of a macro, showing
11294 @cindex macros, from debug info
11295 @item info macro [-a|-all] [--] @var{macro}
11296 Show the current definition or all definitions of the named @var{macro},
11297 and describe the source location or compiler command-line where that
11298 definition was established. The optional double dash is to signify the end of
11299 argument processing and the beginning of @var{macro} for non C-like macros where
11300 the macro may begin with a hyphen.
11302 @kindex info macros
11303 @item info macros @var{linespec}
11304 Show all macro definitions that are in effect at the location specified
11305 by @var{linespec}, and describe the source location or compiler
11306 command-line where those definitions were established.
11308 @kindex macro define
11309 @cindex user-defined macros
11310 @cindex defining macros interactively
11311 @cindex macros, user-defined
11312 @item macro define @var{macro} @var{replacement-list}
11313 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
11314 Introduce a definition for a preprocessor macro named @var{macro},
11315 invocations of which are replaced by the tokens given in
11316 @var{replacement-list}. The first form of this command defines an
11317 ``object-like'' macro, which takes no arguments; the second form
11318 defines a ``function-like'' macro, which takes the arguments given in
11321 A definition introduced by this command is in scope in every
11322 expression evaluated in @value{GDBN}, until it is removed with the
11323 @code{macro undef} command, described below. The definition overrides
11324 all definitions for @var{macro} present in the program being debugged,
11325 as well as any previous user-supplied definition.
11327 @kindex macro undef
11328 @item macro undef @var{macro}
11329 Remove any user-supplied definition for the macro named @var{macro}.
11330 This command only affects definitions provided with the @code{macro
11331 define} command, described above; it cannot remove definitions present
11332 in the program being debugged.
11336 List all the macros defined using the @code{macro define} command.
11339 @cindex macros, example of debugging with
11340 Here is a transcript showing the above commands in action. First, we
11341 show our source files:
11346 #include "sample.h"
11349 #define ADD(x) (M + x)
11354 printf ("Hello, world!\n");
11356 printf ("We're so creative.\n");
11358 printf ("Goodbye, world!\n");
11365 Now, we compile the program using the @sc{gnu} C compiler,
11366 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
11367 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
11368 and @option{-gdwarf-4}; we recommend always choosing the most recent
11369 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
11370 includes information about preprocessor macros in the debugging
11374 $ gcc -gdwarf-2 -g3 sample.c -o sample
11378 Now, we start @value{GDBN} on our sample program:
11382 GNU gdb 2002-05-06-cvs
11383 Copyright 2002 Free Software Foundation, Inc.
11384 GDB is free software, @dots{}
11388 We can expand macros and examine their definitions, even when the
11389 program is not running. @value{GDBN} uses the current listing position
11390 to decide which macro definitions are in scope:
11393 (@value{GDBP}) list main
11396 5 #define ADD(x) (M + x)
11401 10 printf ("Hello, world!\n");
11403 12 printf ("We're so creative.\n");
11404 (@value{GDBP}) info macro ADD
11405 Defined at /home/jimb/gdb/macros/play/sample.c:5
11406 #define ADD(x) (M + x)
11407 (@value{GDBP}) info macro Q
11408 Defined at /home/jimb/gdb/macros/play/sample.h:1
11409 included at /home/jimb/gdb/macros/play/sample.c:2
11411 (@value{GDBP}) macro expand ADD(1)
11412 expands to: (42 + 1)
11413 (@value{GDBP}) macro expand-once ADD(1)
11414 expands to: once (M + 1)
11418 In the example above, note that @code{macro expand-once} expands only
11419 the macro invocation explicit in the original text --- the invocation of
11420 @code{ADD} --- but does not expand the invocation of the macro @code{M},
11421 which was introduced by @code{ADD}.
11423 Once the program is running, @value{GDBN} uses the macro definitions in
11424 force at the source line of the current stack frame:
11427 (@value{GDBP}) break main
11428 Breakpoint 1 at 0x8048370: file sample.c, line 10.
11430 Starting program: /home/jimb/gdb/macros/play/sample
11432 Breakpoint 1, main () at sample.c:10
11433 10 printf ("Hello, world!\n");
11437 At line 10, the definition of the macro @code{N} at line 9 is in force:
11440 (@value{GDBP}) info macro N
11441 Defined at /home/jimb/gdb/macros/play/sample.c:9
11443 (@value{GDBP}) macro expand N Q M
11444 expands to: 28 < 42
11445 (@value{GDBP}) print N Q M
11450 As we step over directives that remove @code{N}'s definition, and then
11451 give it a new definition, @value{GDBN} finds the definition (or lack
11452 thereof) in force at each point:
11455 (@value{GDBP}) next
11457 12 printf ("We're so creative.\n");
11458 (@value{GDBP}) info macro N
11459 The symbol `N' has no definition as a C/C++ preprocessor macro
11460 at /home/jimb/gdb/macros/play/sample.c:12
11461 (@value{GDBP}) next
11463 14 printf ("Goodbye, world!\n");
11464 (@value{GDBP}) info macro N
11465 Defined at /home/jimb/gdb/macros/play/sample.c:13
11467 (@value{GDBP}) macro expand N Q M
11468 expands to: 1729 < 42
11469 (@value{GDBP}) print N Q M
11474 In addition to source files, macros can be defined on the compilation command
11475 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
11476 such a way, @value{GDBN} displays the location of their definition as line zero
11477 of the source file submitted to the compiler.
11480 (@value{GDBP}) info macro __STDC__
11481 Defined at /home/jimb/gdb/macros/play/sample.c:0
11488 @chapter Tracepoints
11489 @c This chapter is based on the documentation written by Michael
11490 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
11492 @cindex tracepoints
11493 In some applications, it is not feasible for the debugger to interrupt
11494 the program's execution long enough for the developer to learn
11495 anything helpful about its behavior. If the program's correctness
11496 depends on its real-time behavior, delays introduced by a debugger
11497 might cause the program to change its behavior drastically, or perhaps
11498 fail, even when the code itself is correct. It is useful to be able
11499 to observe the program's behavior without interrupting it.
11501 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
11502 specify locations in the program, called @dfn{tracepoints}, and
11503 arbitrary expressions to evaluate when those tracepoints are reached.
11504 Later, using the @code{tfind} command, you can examine the values
11505 those expressions had when the program hit the tracepoints. The
11506 expressions may also denote objects in memory---structures or arrays,
11507 for example---whose values @value{GDBN} should record; while visiting
11508 a particular tracepoint, you may inspect those objects as if they were
11509 in memory at that moment. However, because @value{GDBN} records these
11510 values without interacting with you, it can do so quickly and
11511 unobtrusively, hopefully not disturbing the program's behavior.
11513 The tracepoint facility is currently available only for remote
11514 targets. @xref{Targets}. In addition, your remote target must know
11515 how to collect trace data. This functionality is implemented in the
11516 remote stub; however, none of the stubs distributed with @value{GDBN}
11517 support tracepoints as of this writing. The format of the remote
11518 packets used to implement tracepoints are described in @ref{Tracepoint
11521 It is also possible to get trace data from a file, in a manner reminiscent
11522 of corefiles; you specify the filename, and use @code{tfind} to search
11523 through the file. @xref{Trace Files}, for more details.
11525 This chapter describes the tracepoint commands and features.
11528 * Set Tracepoints::
11529 * Analyze Collected Data::
11530 * Tracepoint Variables::
11534 @node Set Tracepoints
11535 @section Commands to Set Tracepoints
11537 Before running such a @dfn{trace experiment}, an arbitrary number of
11538 tracepoints can be set. A tracepoint is actually a special type of
11539 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
11540 standard breakpoint commands. For instance, as with breakpoints,
11541 tracepoint numbers are successive integers starting from one, and many
11542 of the commands associated with tracepoints take the tracepoint number
11543 as their argument, to identify which tracepoint to work on.
11545 For each tracepoint, you can specify, in advance, some arbitrary set
11546 of data that you want the target to collect in the trace buffer when
11547 it hits that tracepoint. The collected data can include registers,
11548 local variables, or global data. Later, you can use @value{GDBN}
11549 commands to examine the values these data had at the time the
11550 tracepoint was hit.
11552 Tracepoints do not support every breakpoint feature. Ignore counts on
11553 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
11554 commands when they are hit. Tracepoints may not be thread-specific
11557 @cindex fast tracepoints
11558 Some targets may support @dfn{fast tracepoints}, which are inserted in
11559 a different way (such as with a jump instead of a trap), that is
11560 faster but possibly restricted in where they may be installed.
11562 @cindex static tracepoints
11563 @cindex markers, static tracepoints
11564 @cindex probing markers, static tracepoints
11565 Regular and fast tracepoints are dynamic tracing facilities, meaning
11566 that they can be used to insert tracepoints at (almost) any location
11567 in the target. Some targets may also support controlling @dfn{static
11568 tracepoints} from @value{GDBN}. With static tracing, a set of
11569 instrumentation points, also known as @dfn{markers}, are embedded in
11570 the target program, and can be activated or deactivated by name or
11571 address. These are usually placed at locations which facilitate
11572 investigating what the target is actually doing. @value{GDBN}'s
11573 support for static tracing includes being able to list instrumentation
11574 points, and attach them with @value{GDBN} defined high level
11575 tracepoints that expose the whole range of convenience of
11576 @value{GDBN}'s tracepoints support. Namely, support for collecting
11577 registers values and values of global or local (to the instrumentation
11578 point) variables; tracepoint conditions and trace state variables.
11579 The act of installing a @value{GDBN} static tracepoint on an
11580 instrumentation point, or marker, is referred to as @dfn{probing} a
11581 static tracepoint marker.
11583 @code{gdbserver} supports tracepoints on some target systems.
11584 @xref{Server,,Tracepoints support in @code{gdbserver}}.
11586 This section describes commands to set tracepoints and associated
11587 conditions and actions.
11590 * Create and Delete Tracepoints::
11591 * Enable and Disable Tracepoints::
11592 * Tracepoint Passcounts::
11593 * Tracepoint Conditions::
11594 * Trace State Variables::
11595 * Tracepoint Actions::
11596 * Listing Tracepoints::
11597 * Listing Static Tracepoint Markers::
11598 * Starting and Stopping Trace Experiments::
11599 * Tracepoint Restrictions::
11602 @node Create and Delete Tracepoints
11603 @subsection Create and Delete Tracepoints
11606 @cindex set tracepoint
11608 @item trace @var{location}
11609 The @code{trace} command is very similar to the @code{break} command.
11610 Its argument @var{location} can be a source line, a function name, or
11611 an address in the target program. @xref{Specify Location}. The
11612 @code{trace} command defines a tracepoint, which is a point in the
11613 target program where the debugger will briefly stop, collect some
11614 data, and then allow the program to continue. Setting a tracepoint or
11615 changing its actions takes effect immediately if the remote stub
11616 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
11618 If remote stub doesn't support the @samp{InstallInTrace} feature, all
11619 these changes don't take effect until the next @code{tstart}
11620 command, and once a trace experiment is running, further changes will
11621 not have any effect until the next trace experiment starts. In addition,
11622 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
11623 address is not yet resolved. (This is similar to pending breakpoints.)
11624 Pending tracepoints are not downloaded to the target and not installed
11625 until they are resolved. The resolution of pending tracepoints requires
11626 @value{GDBN} support---when debugging with the remote target, and
11627 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
11628 tracing}), pending tracepoints can not be resolved (and downloaded to
11629 the remote stub) while @value{GDBN} is disconnected.
11631 Here are some examples of using the @code{trace} command:
11634 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
11636 (@value{GDBP}) @b{trace +2} // 2 lines forward
11638 (@value{GDBP}) @b{trace my_function} // first source line of function
11640 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
11642 (@value{GDBP}) @b{trace *0x2117c4} // an address
11646 You can abbreviate @code{trace} as @code{tr}.
11648 @item trace @var{location} if @var{cond}
11649 Set a tracepoint with condition @var{cond}; evaluate the expression
11650 @var{cond} each time the tracepoint is reached, and collect data only
11651 if the value is nonzero---that is, if @var{cond} evaluates as true.
11652 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
11653 information on tracepoint conditions.
11655 @item ftrace @var{location} [ if @var{cond} ]
11656 @cindex set fast tracepoint
11657 @cindex fast tracepoints, setting
11659 The @code{ftrace} command sets a fast tracepoint. For targets that
11660 support them, fast tracepoints will use a more efficient but possibly
11661 less general technique to trigger data collection, such as a jump
11662 instruction instead of a trap, or some sort of hardware support. It
11663 may not be possible to create a fast tracepoint at the desired
11664 location, in which case the command will exit with an explanatory
11667 @value{GDBN} handles arguments to @code{ftrace} exactly as for
11670 On 32-bit x86-architecture systems, fast tracepoints normally need to
11671 be placed at an instruction that is 5 bytes or longer, but can be
11672 placed at 4-byte instructions if the low 64K of memory of the target
11673 program is available to install trampolines. Some Unix-type systems,
11674 such as @sc{gnu}/Linux, exclude low addresses from the program's
11675 address space; but for instance with the Linux kernel it is possible
11676 to let @value{GDBN} use this area by doing a @command{sysctl} command
11677 to set the @code{mmap_min_addr} kernel parameter, as in
11680 sudo sysctl -w vm.mmap_min_addr=32768
11684 which sets the low address to 32K, which leaves plenty of room for
11685 trampolines. The minimum address should be set to a page boundary.
11687 @item strace @var{location} [ if @var{cond} ]
11688 @cindex set static tracepoint
11689 @cindex static tracepoints, setting
11690 @cindex probe static tracepoint marker
11692 The @code{strace} command sets a static tracepoint. For targets that
11693 support it, setting a static tracepoint probes a static
11694 instrumentation point, or marker, found at @var{location}. It may not
11695 be possible to set a static tracepoint at the desired location, in
11696 which case the command will exit with an explanatory message.
11698 @value{GDBN} handles arguments to @code{strace} exactly as for
11699 @code{trace}, with the addition that the user can also specify
11700 @code{-m @var{marker}} as @var{location}. This probes the marker
11701 identified by the @var{marker} string identifier. This identifier
11702 depends on the static tracepoint backend library your program is
11703 using. You can find all the marker identifiers in the @samp{ID} field
11704 of the @code{info static-tracepoint-markers} command output.
11705 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
11706 Markers}. For example, in the following small program using the UST
11712 trace_mark(ust, bar33, "str %s", "FOOBAZ");
11717 the marker id is composed of joining the first two arguments to the
11718 @code{trace_mark} call with a slash, which translates to:
11721 (@value{GDBP}) info static-tracepoint-markers
11722 Cnt Enb ID Address What
11723 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
11729 so you may probe the marker above with:
11732 (@value{GDBP}) strace -m ust/bar33
11735 Static tracepoints accept an extra collect action --- @code{collect
11736 $_sdata}. This collects arbitrary user data passed in the probe point
11737 call to the tracing library. In the UST example above, you'll see
11738 that the third argument to @code{trace_mark} is a printf-like format
11739 string. The user data is then the result of running that formating
11740 string against the following arguments. Note that @code{info
11741 static-tracepoint-markers} command output lists that format string in
11742 the @samp{Data:} field.
11744 You can inspect this data when analyzing the trace buffer, by printing
11745 the $_sdata variable like any other variable available to
11746 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
11749 @cindex last tracepoint number
11750 @cindex recent tracepoint number
11751 @cindex tracepoint number
11752 The convenience variable @code{$tpnum} records the tracepoint number
11753 of the most recently set tracepoint.
11755 @kindex delete tracepoint
11756 @cindex tracepoint deletion
11757 @item delete tracepoint @r{[}@var{num}@r{]}
11758 Permanently delete one or more tracepoints. With no argument, the
11759 default is to delete all tracepoints. Note that the regular
11760 @code{delete} command can remove tracepoints also.
11765 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
11767 (@value{GDBP}) @b{delete trace} // remove all tracepoints
11771 You can abbreviate this command as @code{del tr}.
11774 @node Enable and Disable Tracepoints
11775 @subsection Enable and Disable Tracepoints
11777 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
11780 @kindex disable tracepoint
11781 @item disable tracepoint @r{[}@var{num}@r{]}
11782 Disable tracepoint @var{num}, or all tracepoints if no argument
11783 @var{num} is given. A disabled tracepoint will have no effect during
11784 a trace experiment, but it is not forgotten. You can re-enable
11785 a disabled tracepoint using the @code{enable tracepoint} command.
11786 If the command is issued during a trace experiment and the debug target
11787 has support for disabling tracepoints during a trace experiment, then the
11788 change will be effective immediately. Otherwise, it will be applied to the
11789 next trace experiment.
11791 @kindex enable tracepoint
11792 @item enable tracepoint @r{[}@var{num}@r{]}
11793 Enable tracepoint @var{num}, or all tracepoints. If this command is
11794 issued during a trace experiment and the debug target supports enabling
11795 tracepoints during a trace experiment, then the enabled tracepoints will
11796 become effective immediately. Otherwise, they will become effective the
11797 next time a trace experiment is run.
11800 @node Tracepoint Passcounts
11801 @subsection Tracepoint Passcounts
11805 @cindex tracepoint pass count
11806 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
11807 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
11808 automatically stop a trace experiment. If a tracepoint's passcount is
11809 @var{n}, then the trace experiment will be automatically stopped on
11810 the @var{n}'th time that tracepoint is hit. If the tracepoint number
11811 @var{num} is not specified, the @code{passcount} command sets the
11812 passcount of the most recently defined tracepoint. If no passcount is
11813 given, the trace experiment will run until stopped explicitly by the
11819 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
11820 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
11822 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
11823 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
11824 (@value{GDBP}) @b{trace foo}
11825 (@value{GDBP}) @b{pass 3}
11826 (@value{GDBP}) @b{trace bar}
11827 (@value{GDBP}) @b{pass 2}
11828 (@value{GDBP}) @b{trace baz}
11829 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
11830 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
11831 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
11832 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
11836 @node Tracepoint Conditions
11837 @subsection Tracepoint Conditions
11838 @cindex conditional tracepoints
11839 @cindex tracepoint conditions
11841 The simplest sort of tracepoint collects data every time your program
11842 reaches a specified place. You can also specify a @dfn{condition} for
11843 a tracepoint. A condition is just a Boolean expression in your
11844 programming language (@pxref{Expressions, ,Expressions}). A
11845 tracepoint with a condition evaluates the expression each time your
11846 program reaches it, and data collection happens only if the condition
11849 Tracepoint conditions can be specified when a tracepoint is set, by
11850 using @samp{if} in the arguments to the @code{trace} command.
11851 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
11852 also be set or changed at any time with the @code{condition} command,
11853 just as with breakpoints.
11855 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
11856 the conditional expression itself. Instead, @value{GDBN} encodes the
11857 expression into an agent expression (@pxref{Agent Expressions})
11858 suitable for execution on the target, independently of @value{GDBN}.
11859 Global variables become raw memory locations, locals become stack
11860 accesses, and so forth.
11862 For instance, suppose you have a function that is usually called
11863 frequently, but should not be called after an error has occurred. You
11864 could use the following tracepoint command to collect data about calls
11865 of that function that happen while the error code is propagating
11866 through the program; an unconditional tracepoint could end up
11867 collecting thousands of useless trace frames that you would have to
11871 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
11874 @node Trace State Variables
11875 @subsection Trace State Variables
11876 @cindex trace state variables
11878 A @dfn{trace state variable} is a special type of variable that is
11879 created and managed by target-side code. The syntax is the same as
11880 that for GDB's convenience variables (a string prefixed with ``$''),
11881 but they are stored on the target. They must be created explicitly,
11882 using a @code{tvariable} command. They are always 64-bit signed
11885 Trace state variables are remembered by @value{GDBN}, and downloaded
11886 to the target along with tracepoint information when the trace
11887 experiment starts. There are no intrinsic limits on the number of
11888 trace state variables, beyond memory limitations of the target.
11890 @cindex convenience variables, and trace state variables
11891 Although trace state variables are managed by the target, you can use
11892 them in print commands and expressions as if they were convenience
11893 variables; @value{GDBN} will get the current value from the target
11894 while the trace experiment is running. Trace state variables share
11895 the same namespace as other ``$'' variables, which means that you
11896 cannot have trace state variables with names like @code{$23} or
11897 @code{$pc}, nor can you have a trace state variable and a convenience
11898 variable with the same name.
11902 @item tvariable $@var{name} [ = @var{expression} ]
11904 The @code{tvariable} command creates a new trace state variable named
11905 @code{$@var{name}}, and optionally gives it an initial value of
11906 @var{expression}. @var{expression} is evaluated when this command is
11907 entered; the result will be converted to an integer if possible,
11908 otherwise @value{GDBN} will report an error. A subsequent
11909 @code{tvariable} command specifying the same name does not create a
11910 variable, but instead assigns the supplied initial value to the
11911 existing variable of that name, overwriting any previous initial
11912 value. The default initial value is 0.
11914 @item info tvariables
11915 @kindex info tvariables
11916 List all the trace state variables along with their initial values.
11917 Their current values may also be displayed, if the trace experiment is
11920 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
11921 @kindex delete tvariable
11922 Delete the given trace state variables, or all of them if no arguments
11927 @node Tracepoint Actions
11928 @subsection Tracepoint Action Lists
11932 @cindex tracepoint actions
11933 @item actions @r{[}@var{num}@r{]}
11934 This command will prompt for a list of actions to be taken when the
11935 tracepoint is hit. If the tracepoint number @var{num} is not
11936 specified, this command sets the actions for the one that was most
11937 recently defined (so that you can define a tracepoint and then say
11938 @code{actions} without bothering about its number). You specify the
11939 actions themselves on the following lines, one action at a time, and
11940 terminate the actions list with a line containing just @code{end}. So
11941 far, the only defined actions are @code{collect}, @code{teval}, and
11942 @code{while-stepping}.
11944 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
11945 Commands, ,Breakpoint Command Lists}), except that only the defined
11946 actions are allowed; any other @value{GDBN} command is rejected.
11948 @cindex remove actions from a tracepoint
11949 To remove all actions from a tracepoint, type @samp{actions @var{num}}
11950 and follow it immediately with @samp{end}.
11953 (@value{GDBP}) @b{collect @var{data}} // collect some data
11955 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
11957 (@value{GDBP}) @b{end} // signals the end of actions.
11960 In the following example, the action list begins with @code{collect}
11961 commands indicating the things to be collected when the tracepoint is
11962 hit. Then, in order to single-step and collect additional data
11963 following the tracepoint, a @code{while-stepping} command is used,
11964 followed by the list of things to be collected after each step in a
11965 sequence of single steps. The @code{while-stepping} command is
11966 terminated by its own separate @code{end} command. Lastly, the action
11967 list is terminated by an @code{end} command.
11970 (@value{GDBP}) @b{trace foo}
11971 (@value{GDBP}) @b{actions}
11972 Enter actions for tracepoint 1, one per line:
11975 > while-stepping 12
11976 > collect $pc, arr[i]
11981 @kindex collect @r{(tracepoints)}
11982 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
11983 Collect values of the given expressions when the tracepoint is hit.
11984 This command accepts a comma-separated list of any valid expressions.
11985 In addition to global, static, or local variables, the following
11986 special arguments are supported:
11990 Collect all registers.
11993 Collect all function arguments.
11996 Collect all local variables.
11999 Collect the return address. This is helpful if you want to see more
12003 Collects the number of arguments from the static probe at which the
12004 tracepoint is located.
12005 @xref{Static Probe Points}.
12007 @item $_probe_arg@var{n}
12008 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
12009 from the static probe at which the tracepoint is located.
12010 @xref{Static Probe Points}.
12013 @vindex $_sdata@r{, collect}
12014 Collect static tracepoint marker specific data. Only available for
12015 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
12016 Lists}. On the UST static tracepoints library backend, an
12017 instrumentation point resembles a @code{printf} function call. The
12018 tracing library is able to collect user specified data formatted to a
12019 character string using the format provided by the programmer that
12020 instrumented the program. Other backends have similar mechanisms.
12021 Here's an example of a UST marker call:
12024 const char master_name[] = "$your_name";
12025 trace_mark(channel1, marker1, "hello %s", master_name)
12028 In this case, collecting @code{$_sdata} collects the string
12029 @samp{hello $yourname}. When analyzing the trace buffer, you can
12030 inspect @samp{$_sdata} like any other variable available to
12034 You can give several consecutive @code{collect} commands, each one
12035 with a single argument, or one @code{collect} command with several
12036 arguments separated by commas; the effect is the same.
12038 The optional @var{mods} changes the usual handling of the arguments.
12039 @code{s} requests that pointers to chars be handled as strings, in
12040 particular collecting the contents of the memory being pointed at, up
12041 to the first zero. The upper bound is by default the value of the
12042 @code{print elements} variable; if @code{s} is followed by a decimal
12043 number, that is the upper bound instead. So for instance
12044 @samp{collect/s25 mystr} collects as many as 25 characters at
12047 The command @code{info scope} (@pxref{Symbols, info scope}) is
12048 particularly useful for figuring out what data to collect.
12050 @kindex teval @r{(tracepoints)}
12051 @item teval @var{expr1}, @var{expr2}, @dots{}
12052 Evaluate the given expressions when the tracepoint is hit. This
12053 command accepts a comma-separated list of expressions. The results
12054 are discarded, so this is mainly useful for assigning values to trace
12055 state variables (@pxref{Trace State Variables}) without adding those
12056 values to the trace buffer, as would be the case if the @code{collect}
12059 @kindex while-stepping @r{(tracepoints)}
12060 @item while-stepping @var{n}
12061 Perform @var{n} single-step instruction traces after the tracepoint,
12062 collecting new data after each step. The @code{while-stepping}
12063 command is followed by the list of what to collect while stepping
12064 (followed by its own @code{end} command):
12067 > while-stepping 12
12068 > collect $regs, myglobal
12074 Note that @code{$pc} is not automatically collected by
12075 @code{while-stepping}; you need to explicitly collect that register if
12076 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
12079 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
12080 @kindex set default-collect
12081 @cindex default collection action
12082 This variable is a list of expressions to collect at each tracepoint
12083 hit. It is effectively an additional @code{collect} action prepended
12084 to every tracepoint action list. The expressions are parsed
12085 individually for each tracepoint, so for instance a variable named
12086 @code{xyz} may be interpreted as a global for one tracepoint, and a
12087 local for another, as appropriate to the tracepoint's location.
12089 @item show default-collect
12090 @kindex show default-collect
12091 Show the list of expressions that are collected by default at each
12096 @node Listing Tracepoints
12097 @subsection Listing Tracepoints
12100 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
12101 @kindex info tp @r{[}@var{n}@dots{}@r{]}
12102 @cindex information about tracepoints
12103 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
12104 Display information about the tracepoint @var{num}. If you don't
12105 specify a tracepoint number, displays information about all the
12106 tracepoints defined so far. The format is similar to that used for
12107 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
12108 command, simply restricting itself to tracepoints.
12110 A tracepoint's listing may include additional information specific to
12115 its passcount as given by the @code{passcount @var{n}} command
12118 the state about installed on target of each location
12122 (@value{GDBP}) @b{info trace}
12123 Num Type Disp Enb Address What
12124 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
12126 collect globfoo, $regs
12131 2 tracepoint keep y <MULTIPLE>
12133 2.1 y 0x0804859c in func4 at change-loc.h:35
12134 installed on target
12135 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
12136 installed on target
12137 2.3 y <PENDING> set_tracepoint
12138 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
12139 not installed on target
12144 This command can be abbreviated @code{info tp}.
12147 @node Listing Static Tracepoint Markers
12148 @subsection Listing Static Tracepoint Markers
12151 @kindex info static-tracepoint-markers
12152 @cindex information about static tracepoint markers
12153 @item info static-tracepoint-markers
12154 Display information about all static tracepoint markers defined in the
12157 For each marker, the following columns are printed:
12161 An incrementing counter, output to help readability. This is not a
12164 The marker ID, as reported by the target.
12165 @item Enabled or Disabled
12166 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
12167 that are not enabled.
12169 Where the marker is in your program, as a memory address.
12171 Where the marker is in the source for your program, as a file and line
12172 number. If the debug information included in the program does not
12173 allow @value{GDBN} to locate the source of the marker, this column
12174 will be left blank.
12178 In addition, the following information may be printed for each marker:
12182 User data passed to the tracing library by the marker call. In the
12183 UST backend, this is the format string passed as argument to the
12185 @item Static tracepoints probing the marker
12186 The list of static tracepoints attached to the marker.
12190 (@value{GDBP}) info static-tracepoint-markers
12191 Cnt ID Enb Address What
12192 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
12193 Data: number1 %d number2 %d
12194 Probed by static tracepoints: #2
12195 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
12201 @node Starting and Stopping Trace Experiments
12202 @subsection Starting and Stopping Trace Experiments
12205 @kindex tstart [ @var{notes} ]
12206 @cindex start a new trace experiment
12207 @cindex collected data discarded
12209 This command starts the trace experiment, and begins collecting data.
12210 It has the side effect of discarding all the data collected in the
12211 trace buffer during the previous trace experiment. If any arguments
12212 are supplied, they are taken as a note and stored with the trace
12213 experiment's state. The notes may be arbitrary text, and are
12214 especially useful with disconnected tracing in a multi-user context;
12215 the notes can explain what the trace is doing, supply user contact
12216 information, and so forth.
12218 @kindex tstop [ @var{notes} ]
12219 @cindex stop a running trace experiment
12221 This command stops the trace experiment. If any arguments are
12222 supplied, they are recorded with the experiment as a note. This is
12223 useful if you are stopping a trace started by someone else, for
12224 instance if the trace is interfering with the system's behavior and
12225 needs to be stopped quickly.
12227 @strong{Note}: a trace experiment and data collection may stop
12228 automatically if any tracepoint's passcount is reached
12229 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
12232 @cindex status of trace data collection
12233 @cindex trace experiment, status of
12235 This command displays the status of the current trace data
12239 Here is an example of the commands we described so far:
12242 (@value{GDBP}) @b{trace gdb_c_test}
12243 (@value{GDBP}) @b{actions}
12244 Enter actions for tracepoint #1, one per line.
12245 > collect $regs,$locals,$args
12246 > while-stepping 11
12250 (@value{GDBP}) @b{tstart}
12251 [time passes @dots{}]
12252 (@value{GDBP}) @b{tstop}
12255 @anchor{disconnected tracing}
12256 @cindex disconnected tracing
12257 You can choose to continue running the trace experiment even if
12258 @value{GDBN} disconnects from the target, voluntarily or
12259 involuntarily. For commands such as @code{detach}, the debugger will
12260 ask what you want to do with the trace. But for unexpected
12261 terminations (@value{GDBN} crash, network outage), it would be
12262 unfortunate to lose hard-won trace data, so the variable
12263 @code{disconnected-tracing} lets you decide whether the trace should
12264 continue running without @value{GDBN}.
12267 @item set disconnected-tracing on
12268 @itemx set disconnected-tracing off
12269 @kindex set disconnected-tracing
12270 Choose whether a tracing run should continue to run if @value{GDBN}
12271 has disconnected from the target. Note that @code{detach} or
12272 @code{quit} will ask you directly what to do about a running trace no
12273 matter what this variable's setting, so the variable is mainly useful
12274 for handling unexpected situations, such as loss of the network.
12276 @item show disconnected-tracing
12277 @kindex show disconnected-tracing
12278 Show the current choice for disconnected tracing.
12282 When you reconnect to the target, the trace experiment may or may not
12283 still be running; it might have filled the trace buffer in the
12284 meantime, or stopped for one of the other reasons. If it is running,
12285 it will continue after reconnection.
12287 Upon reconnection, the target will upload information about the
12288 tracepoints in effect. @value{GDBN} will then compare that
12289 information to the set of tracepoints currently defined, and attempt
12290 to match them up, allowing for the possibility that the numbers may
12291 have changed due to creation and deletion in the meantime. If one of
12292 the target's tracepoints does not match any in @value{GDBN}, the
12293 debugger will create a new tracepoint, so that you have a number with
12294 which to specify that tracepoint. This matching-up process is
12295 necessarily heuristic, and it may result in useless tracepoints being
12296 created; you may simply delete them if they are of no use.
12298 @cindex circular trace buffer
12299 If your target agent supports a @dfn{circular trace buffer}, then you
12300 can run a trace experiment indefinitely without filling the trace
12301 buffer; when space runs out, the agent deletes already-collected trace
12302 frames, oldest first, until there is enough room to continue
12303 collecting. This is especially useful if your tracepoints are being
12304 hit too often, and your trace gets terminated prematurely because the
12305 buffer is full. To ask for a circular trace buffer, simply set
12306 @samp{circular-trace-buffer} to on. You can set this at any time,
12307 including during tracing; if the agent can do it, it will change
12308 buffer handling on the fly, otherwise it will not take effect until
12312 @item set circular-trace-buffer on
12313 @itemx set circular-trace-buffer off
12314 @kindex set circular-trace-buffer
12315 Choose whether a tracing run should use a linear or circular buffer
12316 for trace data. A linear buffer will not lose any trace data, but may
12317 fill up prematurely, while a circular buffer will discard old trace
12318 data, but it will have always room for the latest tracepoint hits.
12320 @item show circular-trace-buffer
12321 @kindex show circular-trace-buffer
12322 Show the current choice for the trace buffer. Note that this may not
12323 match the agent's current buffer handling, nor is it guaranteed to
12324 match the setting that might have been in effect during a past run,
12325 for instance if you are looking at frames from a trace file.
12330 @item set trace-buffer-size @var{n}
12331 @itemx set trace-buffer-size unlimited
12332 @kindex set trace-buffer-size
12333 Request that the target use a trace buffer of @var{n} bytes. Not all
12334 targets will honor the request; they may have a compiled-in size for
12335 the trace buffer, or some other limitation. Set to a value of
12336 @code{unlimited} or @code{-1} to let the target use whatever size it
12337 likes. This is also the default.
12339 @item show trace-buffer-size
12340 @kindex show trace-buffer-size
12341 Show the current requested size for the trace buffer. Note that this
12342 will only match the actual size if the target supports size-setting,
12343 and was able to handle the requested size. For instance, if the
12344 target can only change buffer size between runs, this variable will
12345 not reflect the change until the next run starts. Use @code{tstatus}
12346 to get a report of the actual buffer size.
12350 @item set trace-user @var{text}
12351 @kindex set trace-user
12353 @item show trace-user
12354 @kindex show trace-user
12356 @item set trace-notes @var{text}
12357 @kindex set trace-notes
12358 Set the trace run's notes.
12360 @item show trace-notes
12361 @kindex show trace-notes
12362 Show the trace run's notes.
12364 @item set trace-stop-notes @var{text}
12365 @kindex set trace-stop-notes
12366 Set the trace run's stop notes. The handling of the note is as for
12367 @code{tstop} arguments; the set command is convenient way to fix a
12368 stop note that is mistaken or incomplete.
12370 @item show trace-stop-notes
12371 @kindex show trace-stop-notes
12372 Show the trace run's stop notes.
12376 @node Tracepoint Restrictions
12377 @subsection Tracepoint Restrictions
12379 @cindex tracepoint restrictions
12380 There are a number of restrictions on the use of tracepoints. As
12381 described above, tracepoint data gathering occurs on the target
12382 without interaction from @value{GDBN}. Thus the full capabilities of
12383 the debugger are not available during data gathering, and then at data
12384 examination time, you will be limited by only having what was
12385 collected. The following items describe some common problems, but it
12386 is not exhaustive, and you may run into additional difficulties not
12392 Tracepoint expressions are intended to gather objects (lvalues). Thus
12393 the full flexibility of GDB's expression evaluator is not available.
12394 You cannot call functions, cast objects to aggregate types, access
12395 convenience variables or modify values (except by assignment to trace
12396 state variables). Some language features may implicitly call
12397 functions (for instance Objective-C fields with accessors), and therefore
12398 cannot be collected either.
12401 Collection of local variables, either individually or in bulk with
12402 @code{$locals} or @code{$args}, during @code{while-stepping} may
12403 behave erratically. The stepping action may enter a new scope (for
12404 instance by stepping into a function), or the location of the variable
12405 may change (for instance it is loaded into a register). The
12406 tracepoint data recorded uses the location information for the
12407 variables that is correct for the tracepoint location. When the
12408 tracepoint is created, it is not possible, in general, to determine
12409 where the steps of a @code{while-stepping} sequence will advance the
12410 program---particularly if a conditional branch is stepped.
12413 Collection of an incompletely-initialized or partially-destroyed object
12414 may result in something that @value{GDBN} cannot display, or displays
12415 in a misleading way.
12418 When @value{GDBN} displays a pointer to character it automatically
12419 dereferences the pointer to also display characters of the string
12420 being pointed to. However, collecting the pointer during tracing does
12421 not automatically collect the string. You need to explicitly
12422 dereference the pointer and provide size information if you want to
12423 collect not only the pointer, but the memory pointed to. For example,
12424 @code{*ptr@@50} can be used to collect the 50 element array pointed to
12428 It is not possible to collect a complete stack backtrace at a
12429 tracepoint. Instead, you may collect the registers and a few hundred
12430 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
12431 (adjust to use the name of the actual stack pointer register on your
12432 target architecture, and the amount of stack you wish to capture).
12433 Then the @code{backtrace} command will show a partial backtrace when
12434 using a trace frame. The number of stack frames that can be examined
12435 depends on the sizes of the frames in the collected stack. Note that
12436 if you ask for a block so large that it goes past the bottom of the
12437 stack, the target agent may report an error trying to read from an
12441 If you do not collect registers at a tracepoint, @value{GDBN} can
12442 infer that the value of @code{$pc} must be the same as the address of
12443 the tracepoint and use that when you are looking at a trace frame
12444 for that tracepoint. However, this cannot work if the tracepoint has
12445 multiple locations (for instance if it was set in a function that was
12446 inlined), or if it has a @code{while-stepping} loop. In those cases
12447 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
12452 @node Analyze Collected Data
12453 @section Using the Collected Data
12455 After the tracepoint experiment ends, you use @value{GDBN} commands
12456 for examining the trace data. The basic idea is that each tracepoint
12457 collects a trace @dfn{snapshot} every time it is hit and another
12458 snapshot every time it single-steps. All these snapshots are
12459 consecutively numbered from zero and go into a buffer, and you can
12460 examine them later. The way you examine them is to @dfn{focus} on a
12461 specific trace snapshot. When the remote stub is focused on a trace
12462 snapshot, it will respond to all @value{GDBN} requests for memory and
12463 registers by reading from the buffer which belongs to that snapshot,
12464 rather than from @emph{real} memory or registers of the program being
12465 debugged. This means that @strong{all} @value{GDBN} commands
12466 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
12467 behave as if we were currently debugging the program state as it was
12468 when the tracepoint occurred. Any requests for data that are not in
12469 the buffer will fail.
12472 * tfind:: How to select a trace snapshot
12473 * tdump:: How to display all data for a snapshot
12474 * save tracepoints:: How to save tracepoints for a future run
12478 @subsection @code{tfind @var{n}}
12481 @cindex select trace snapshot
12482 @cindex find trace snapshot
12483 The basic command for selecting a trace snapshot from the buffer is
12484 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
12485 counting from zero. If no argument @var{n} is given, the next
12486 snapshot is selected.
12488 Here are the various forms of using the @code{tfind} command.
12492 Find the first snapshot in the buffer. This is a synonym for
12493 @code{tfind 0} (since 0 is the number of the first snapshot).
12496 Stop debugging trace snapshots, resume @emph{live} debugging.
12499 Same as @samp{tfind none}.
12502 No argument means find the next trace snapshot.
12505 Find the previous trace snapshot before the current one. This permits
12506 retracing earlier steps.
12508 @item tfind tracepoint @var{num}
12509 Find the next snapshot associated with tracepoint @var{num}. Search
12510 proceeds forward from the last examined trace snapshot. If no
12511 argument @var{num} is given, it means find the next snapshot collected
12512 for the same tracepoint as the current snapshot.
12514 @item tfind pc @var{addr}
12515 Find the next snapshot associated with the value @var{addr} of the
12516 program counter. Search proceeds forward from the last examined trace
12517 snapshot. If no argument @var{addr} is given, it means find the next
12518 snapshot with the same value of PC as the current snapshot.
12520 @item tfind outside @var{addr1}, @var{addr2}
12521 Find the next snapshot whose PC is outside the given range of
12522 addresses (exclusive).
12524 @item tfind range @var{addr1}, @var{addr2}
12525 Find the next snapshot whose PC is between @var{addr1} and
12526 @var{addr2} (inclusive).
12528 @item tfind line @r{[}@var{file}:@r{]}@var{n}
12529 Find the next snapshot associated with the source line @var{n}. If
12530 the optional argument @var{file} is given, refer to line @var{n} in
12531 that source file. Search proceeds forward from the last examined
12532 trace snapshot. If no argument @var{n} is given, it means find the
12533 next line other than the one currently being examined; thus saying
12534 @code{tfind line} repeatedly can appear to have the same effect as
12535 stepping from line to line in a @emph{live} debugging session.
12538 The default arguments for the @code{tfind} commands are specifically
12539 designed to make it easy to scan through the trace buffer. For
12540 instance, @code{tfind} with no argument selects the next trace
12541 snapshot, and @code{tfind -} with no argument selects the previous
12542 trace snapshot. So, by giving one @code{tfind} command, and then
12543 simply hitting @key{RET} repeatedly you can examine all the trace
12544 snapshots in order. Or, by saying @code{tfind -} and then hitting
12545 @key{RET} repeatedly you can examine the snapshots in reverse order.
12546 The @code{tfind line} command with no argument selects the snapshot
12547 for the next source line executed. The @code{tfind pc} command with
12548 no argument selects the next snapshot with the same program counter
12549 (PC) as the current frame. The @code{tfind tracepoint} command with
12550 no argument selects the next trace snapshot collected by the same
12551 tracepoint as the current one.
12553 In addition to letting you scan through the trace buffer manually,
12554 these commands make it easy to construct @value{GDBN} scripts that
12555 scan through the trace buffer and print out whatever collected data
12556 you are interested in. Thus, if we want to examine the PC, FP, and SP
12557 registers from each trace frame in the buffer, we can say this:
12560 (@value{GDBP}) @b{tfind start}
12561 (@value{GDBP}) @b{while ($trace_frame != -1)}
12562 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
12563 $trace_frame, $pc, $sp, $fp
12567 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
12568 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
12569 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
12570 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
12571 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
12572 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
12573 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
12574 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
12575 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
12576 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
12577 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
12580 Or, if we want to examine the variable @code{X} at each source line in
12584 (@value{GDBP}) @b{tfind start}
12585 (@value{GDBP}) @b{while ($trace_frame != -1)}
12586 > printf "Frame %d, X == %d\n", $trace_frame, X
12596 @subsection @code{tdump}
12598 @cindex dump all data collected at tracepoint
12599 @cindex tracepoint data, display
12601 This command takes no arguments. It prints all the data collected at
12602 the current trace snapshot.
12605 (@value{GDBP}) @b{trace 444}
12606 (@value{GDBP}) @b{actions}
12607 Enter actions for tracepoint #2, one per line:
12608 > collect $regs, $locals, $args, gdb_long_test
12611 (@value{GDBP}) @b{tstart}
12613 (@value{GDBP}) @b{tfind line 444}
12614 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
12616 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
12618 (@value{GDBP}) @b{tdump}
12619 Data collected at tracepoint 2, trace frame 1:
12620 d0 0xc4aa0085 -995491707
12624 d4 0x71aea3d 119204413
12627 d7 0x380035 3670069
12628 a0 0x19e24a 1696330
12629 a1 0x3000668 50333288
12631 a3 0x322000 3284992
12632 a4 0x3000698 50333336
12633 a5 0x1ad3cc 1758156
12634 fp 0x30bf3c 0x30bf3c
12635 sp 0x30bf34 0x30bf34
12637 pc 0x20b2c8 0x20b2c8
12641 p = 0x20e5b4 "gdb-test"
12648 gdb_long_test = 17 '\021'
12653 @code{tdump} works by scanning the tracepoint's current collection
12654 actions and printing the value of each expression listed. So
12655 @code{tdump} can fail, if after a run, you change the tracepoint's
12656 actions to mention variables that were not collected during the run.
12658 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
12659 uses the collected value of @code{$pc} to distinguish between trace
12660 frames that were collected at the tracepoint hit, and frames that were
12661 collected while stepping. This allows it to correctly choose whether
12662 to display the basic list of collections, or the collections from the
12663 body of the while-stepping loop. However, if @code{$pc} was not collected,
12664 then @code{tdump} will always attempt to dump using the basic collection
12665 list, and may fail if a while-stepping frame does not include all the
12666 same data that is collected at the tracepoint hit.
12667 @c This is getting pretty arcane, example would be good.
12669 @node save tracepoints
12670 @subsection @code{save tracepoints @var{filename}}
12671 @kindex save tracepoints
12672 @kindex save-tracepoints
12673 @cindex save tracepoints for future sessions
12675 This command saves all current tracepoint definitions together with
12676 their actions and passcounts, into a file @file{@var{filename}}
12677 suitable for use in a later debugging session. To read the saved
12678 tracepoint definitions, use the @code{source} command (@pxref{Command
12679 Files}). The @w{@code{save-tracepoints}} command is a deprecated
12680 alias for @w{@code{save tracepoints}}
12682 @node Tracepoint Variables
12683 @section Convenience Variables for Tracepoints
12684 @cindex tracepoint variables
12685 @cindex convenience variables for tracepoints
12688 @vindex $trace_frame
12689 @item (int) $trace_frame
12690 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
12691 snapshot is selected.
12693 @vindex $tracepoint
12694 @item (int) $tracepoint
12695 The tracepoint for the current trace snapshot.
12697 @vindex $trace_line
12698 @item (int) $trace_line
12699 The line number for the current trace snapshot.
12701 @vindex $trace_file
12702 @item (char []) $trace_file
12703 The source file for the current trace snapshot.
12705 @vindex $trace_func
12706 @item (char []) $trace_func
12707 The name of the function containing @code{$tracepoint}.
12710 Note: @code{$trace_file} is not suitable for use in @code{printf},
12711 use @code{output} instead.
12713 Here's a simple example of using these convenience variables for
12714 stepping through all the trace snapshots and printing some of their
12715 data. Note that these are not the same as trace state variables,
12716 which are managed by the target.
12719 (@value{GDBP}) @b{tfind start}
12721 (@value{GDBP}) @b{while $trace_frame != -1}
12722 > output $trace_file
12723 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
12729 @section Using Trace Files
12730 @cindex trace files
12732 In some situations, the target running a trace experiment may no
12733 longer be available; perhaps it crashed, or the hardware was needed
12734 for a different activity. To handle these cases, you can arrange to
12735 dump the trace data into a file, and later use that file as a source
12736 of trace data, via the @code{target tfile} command.
12741 @item tsave [ -r ] @var{filename}
12742 @itemx tsave [-ctf] @var{dirname}
12743 Save the trace data to @var{filename}. By default, this command
12744 assumes that @var{filename} refers to the host filesystem, so if
12745 necessary @value{GDBN} will copy raw trace data up from the target and
12746 then save it. If the target supports it, you can also supply the
12747 optional argument @code{-r} (``remote'') to direct the target to save
12748 the data directly into @var{filename} in its own filesystem, which may be
12749 more efficient if the trace buffer is very large. (Note, however, that
12750 @code{target tfile} can only read from files accessible to the host.)
12751 By default, this command will save trace frame in tfile format.
12752 You can supply the optional argument @code{-ctf} to save date in CTF
12753 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
12754 that can be shared by multiple debugging and tracing tools. Please go to
12755 @indicateurl{http://www.efficios.com/ctf} to get more information.
12757 @kindex target tfile
12761 @item target tfile @var{filename}
12762 @itemx target ctf @var{dirname}
12763 Use the file named @var{filename} or directory named @var{dirname} as
12764 a source of trace data. Commands that examine data work as they do with
12765 a live target, but it is not possible to run any new trace experiments.
12766 @code{tstatus} will report the state of the trace run at the moment
12767 the data was saved, as well as the current trace frame you are examining.
12768 @var{filename} or @var{dirname} must be on a filesystem accessible to
12772 (@value{GDBP}) target ctf ctf.ctf
12773 (@value{GDBP}) tfind
12774 Found trace frame 0, tracepoint 2
12775 39 ++a; /* set tracepoint 1 here */
12776 (@value{GDBP}) tdump
12777 Data collected at tracepoint 2, trace frame 0:
12781 c = @{"123", "456", "789", "123", "456", "789"@}
12782 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
12790 @chapter Debugging Programs That Use Overlays
12793 If your program is too large to fit completely in your target system's
12794 memory, you can sometimes use @dfn{overlays} to work around this
12795 problem. @value{GDBN} provides some support for debugging programs that
12799 * How Overlays Work:: A general explanation of overlays.
12800 * Overlay Commands:: Managing overlays in @value{GDBN}.
12801 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
12802 mapped by asking the inferior.
12803 * Overlay Sample Program:: A sample program using overlays.
12806 @node How Overlays Work
12807 @section How Overlays Work
12808 @cindex mapped overlays
12809 @cindex unmapped overlays
12810 @cindex load address, overlay's
12811 @cindex mapped address
12812 @cindex overlay area
12814 Suppose you have a computer whose instruction address space is only 64
12815 kilobytes long, but which has much more memory which can be accessed by
12816 other means: special instructions, segment registers, or memory
12817 management hardware, for example. Suppose further that you want to
12818 adapt a program which is larger than 64 kilobytes to run on this system.
12820 One solution is to identify modules of your program which are relatively
12821 independent, and need not call each other directly; call these modules
12822 @dfn{overlays}. Separate the overlays from the main program, and place
12823 their machine code in the larger memory. Place your main program in
12824 instruction memory, but leave at least enough space there to hold the
12825 largest overlay as well.
12827 Now, to call a function located in an overlay, you must first copy that
12828 overlay's machine code from the large memory into the space set aside
12829 for it in the instruction memory, and then jump to its entry point
12832 @c NB: In the below the mapped area's size is greater or equal to the
12833 @c size of all overlays. This is intentional to remind the developer
12834 @c that overlays don't necessarily need to be the same size.
12838 Data Instruction Larger
12839 Address Space Address Space Address Space
12840 +-----------+ +-----------+ +-----------+
12842 +-----------+ +-----------+ +-----------+<-- overlay 1
12843 | program | | main | .----| overlay 1 | load address
12844 | variables | | program | | +-----------+
12845 | and heap | | | | | |
12846 +-----------+ | | | +-----------+<-- overlay 2
12847 | | +-----------+ | | | load address
12848 +-----------+ | | | .-| overlay 2 |
12850 mapped --->+-----------+ | | +-----------+
12851 address | | | | | |
12852 | overlay | <-' | | |
12853 | area | <---' +-----------+<-- overlay 3
12854 | | <---. | | load address
12855 +-----------+ `--| overlay 3 |
12862 @anchor{A code overlay}A code overlay
12866 The diagram (@pxref{A code overlay}) shows a system with separate data
12867 and instruction address spaces. To map an overlay, the program copies
12868 its code from the larger address space to the instruction address space.
12869 Since the overlays shown here all use the same mapped address, only one
12870 may be mapped at a time. For a system with a single address space for
12871 data and instructions, the diagram would be similar, except that the
12872 program variables and heap would share an address space with the main
12873 program and the overlay area.
12875 An overlay loaded into instruction memory and ready for use is called a
12876 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
12877 instruction memory. An overlay not present (or only partially present)
12878 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
12879 is its address in the larger memory. The mapped address is also called
12880 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
12881 called the @dfn{load memory address}, or @dfn{LMA}.
12883 Unfortunately, overlays are not a completely transparent way to adapt a
12884 program to limited instruction memory. They introduce a new set of
12885 global constraints you must keep in mind as you design your program:
12890 Before calling or returning to a function in an overlay, your program
12891 must make sure that overlay is actually mapped. Otherwise, the call or
12892 return will transfer control to the right address, but in the wrong
12893 overlay, and your program will probably crash.
12896 If the process of mapping an overlay is expensive on your system, you
12897 will need to choose your overlays carefully to minimize their effect on
12898 your program's performance.
12901 The executable file you load onto your system must contain each
12902 overlay's instructions, appearing at the overlay's load address, not its
12903 mapped address. However, each overlay's instructions must be relocated
12904 and its symbols defined as if the overlay were at its mapped address.
12905 You can use GNU linker scripts to specify different load and relocation
12906 addresses for pieces of your program; see @ref{Overlay Description,,,
12907 ld.info, Using ld: the GNU linker}.
12910 The procedure for loading executable files onto your system must be able
12911 to load their contents into the larger address space as well as the
12912 instruction and data spaces.
12916 The overlay system described above is rather simple, and could be
12917 improved in many ways:
12922 If your system has suitable bank switch registers or memory management
12923 hardware, you could use those facilities to make an overlay's load area
12924 contents simply appear at their mapped address in instruction space.
12925 This would probably be faster than copying the overlay to its mapped
12926 area in the usual way.
12929 If your overlays are small enough, you could set aside more than one
12930 overlay area, and have more than one overlay mapped at a time.
12933 You can use overlays to manage data, as well as instructions. In
12934 general, data overlays are even less transparent to your design than
12935 code overlays: whereas code overlays only require care when you call or
12936 return to functions, data overlays require care every time you access
12937 the data. Also, if you change the contents of a data overlay, you
12938 must copy its contents back out to its load address before you can copy a
12939 different data overlay into the same mapped area.
12944 @node Overlay Commands
12945 @section Overlay Commands
12947 To use @value{GDBN}'s overlay support, each overlay in your program must
12948 correspond to a separate section of the executable file. The section's
12949 virtual memory address and load memory address must be the overlay's
12950 mapped and load addresses. Identifying overlays with sections allows
12951 @value{GDBN} to determine the appropriate address of a function or
12952 variable, depending on whether the overlay is mapped or not.
12954 @value{GDBN}'s overlay commands all start with the word @code{overlay};
12955 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
12960 Disable @value{GDBN}'s overlay support. When overlay support is
12961 disabled, @value{GDBN} assumes that all functions and variables are
12962 always present at their mapped addresses. By default, @value{GDBN}'s
12963 overlay support is disabled.
12965 @item overlay manual
12966 @cindex manual overlay debugging
12967 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
12968 relies on you to tell it which overlays are mapped, and which are not,
12969 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
12970 commands described below.
12972 @item overlay map-overlay @var{overlay}
12973 @itemx overlay map @var{overlay}
12974 @cindex map an overlay
12975 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
12976 be the name of the object file section containing the overlay. When an
12977 overlay is mapped, @value{GDBN} assumes it can find the overlay's
12978 functions and variables at their mapped addresses. @value{GDBN} assumes
12979 that any other overlays whose mapped ranges overlap that of
12980 @var{overlay} are now unmapped.
12982 @item overlay unmap-overlay @var{overlay}
12983 @itemx overlay unmap @var{overlay}
12984 @cindex unmap an overlay
12985 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
12986 must be the name of the object file section containing the overlay.
12987 When an overlay is unmapped, @value{GDBN} assumes it can find the
12988 overlay's functions and variables at their load addresses.
12991 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
12992 consults a data structure the overlay manager maintains in the inferior
12993 to see which overlays are mapped. For details, see @ref{Automatic
12994 Overlay Debugging}.
12996 @item overlay load-target
12997 @itemx overlay load
12998 @cindex reloading the overlay table
12999 Re-read the overlay table from the inferior. Normally, @value{GDBN}
13000 re-reads the table @value{GDBN} automatically each time the inferior
13001 stops, so this command should only be necessary if you have changed the
13002 overlay mapping yourself using @value{GDBN}. This command is only
13003 useful when using automatic overlay debugging.
13005 @item overlay list-overlays
13006 @itemx overlay list
13007 @cindex listing mapped overlays
13008 Display a list of the overlays currently mapped, along with their mapped
13009 addresses, load addresses, and sizes.
13013 Normally, when @value{GDBN} prints a code address, it includes the name
13014 of the function the address falls in:
13017 (@value{GDBP}) print main
13018 $3 = @{int ()@} 0x11a0 <main>
13021 When overlay debugging is enabled, @value{GDBN} recognizes code in
13022 unmapped overlays, and prints the names of unmapped functions with
13023 asterisks around them. For example, if @code{foo} is a function in an
13024 unmapped overlay, @value{GDBN} prints it this way:
13027 (@value{GDBP}) overlay list
13028 No sections are mapped.
13029 (@value{GDBP}) print foo
13030 $5 = @{int (int)@} 0x100000 <*foo*>
13033 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
13037 (@value{GDBP}) overlay list
13038 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
13039 mapped at 0x1016 - 0x104a
13040 (@value{GDBP}) print foo
13041 $6 = @{int (int)@} 0x1016 <foo>
13044 When overlay debugging is enabled, @value{GDBN} can find the correct
13045 address for functions and variables in an overlay, whether or not the
13046 overlay is mapped. This allows most @value{GDBN} commands, like
13047 @code{break} and @code{disassemble}, to work normally, even on unmapped
13048 code. However, @value{GDBN}'s breakpoint support has some limitations:
13052 @cindex breakpoints in overlays
13053 @cindex overlays, setting breakpoints in
13054 You can set breakpoints in functions in unmapped overlays, as long as
13055 @value{GDBN} can write to the overlay at its load address.
13057 @value{GDBN} can not set hardware or simulator-based breakpoints in
13058 unmapped overlays. However, if you set a breakpoint at the end of your
13059 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
13060 you are using manual overlay management), @value{GDBN} will re-set its
13061 breakpoints properly.
13065 @node Automatic Overlay Debugging
13066 @section Automatic Overlay Debugging
13067 @cindex automatic overlay debugging
13069 @value{GDBN} can automatically track which overlays are mapped and which
13070 are not, given some simple co-operation from the overlay manager in the
13071 inferior. If you enable automatic overlay debugging with the
13072 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
13073 looks in the inferior's memory for certain variables describing the
13074 current state of the overlays.
13076 Here are the variables your overlay manager must define to support
13077 @value{GDBN}'s automatic overlay debugging:
13081 @item @code{_ovly_table}:
13082 This variable must be an array of the following structures:
13087 /* The overlay's mapped address. */
13090 /* The size of the overlay, in bytes. */
13091 unsigned long size;
13093 /* The overlay's load address. */
13096 /* Non-zero if the overlay is currently mapped;
13098 unsigned long mapped;
13102 @item @code{_novlys}:
13103 This variable must be a four-byte signed integer, holding the total
13104 number of elements in @code{_ovly_table}.
13108 To decide whether a particular overlay is mapped or not, @value{GDBN}
13109 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
13110 @code{lma} members equal the VMA and LMA of the overlay's section in the
13111 executable file. When @value{GDBN} finds a matching entry, it consults
13112 the entry's @code{mapped} member to determine whether the overlay is
13115 In addition, your overlay manager may define a function called
13116 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
13117 will silently set a breakpoint there. If the overlay manager then
13118 calls this function whenever it has changed the overlay table, this
13119 will enable @value{GDBN} to accurately keep track of which overlays
13120 are in program memory, and update any breakpoints that may be set
13121 in overlays. This will allow breakpoints to work even if the
13122 overlays are kept in ROM or other non-writable memory while they
13123 are not being executed.
13125 @node Overlay Sample Program
13126 @section Overlay Sample Program
13127 @cindex overlay example program
13129 When linking a program which uses overlays, you must place the overlays
13130 at their load addresses, while relocating them to run at their mapped
13131 addresses. To do this, you must write a linker script (@pxref{Overlay
13132 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
13133 since linker scripts are specific to a particular host system, target
13134 architecture, and target memory layout, this manual cannot provide
13135 portable sample code demonstrating @value{GDBN}'s overlay support.
13137 However, the @value{GDBN} source distribution does contain an overlaid
13138 program, with linker scripts for a few systems, as part of its test
13139 suite. The program consists of the following files from
13140 @file{gdb/testsuite/gdb.base}:
13144 The main program file.
13146 A simple overlay manager, used by @file{overlays.c}.
13151 Overlay modules, loaded and used by @file{overlays.c}.
13154 Linker scripts for linking the test program on the @code{d10v-elf}
13155 and @code{m32r-elf} targets.
13158 You can build the test program using the @code{d10v-elf} GCC
13159 cross-compiler like this:
13162 $ d10v-elf-gcc -g -c overlays.c
13163 $ d10v-elf-gcc -g -c ovlymgr.c
13164 $ d10v-elf-gcc -g -c foo.c
13165 $ d10v-elf-gcc -g -c bar.c
13166 $ d10v-elf-gcc -g -c baz.c
13167 $ d10v-elf-gcc -g -c grbx.c
13168 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
13169 baz.o grbx.o -Wl,-Td10v.ld -o overlays
13172 The build process is identical for any other architecture, except that
13173 you must substitute the appropriate compiler and linker script for the
13174 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
13178 @chapter Using @value{GDBN} with Different Languages
13181 Although programming languages generally have common aspects, they are
13182 rarely expressed in the same manner. For instance, in ANSI C,
13183 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
13184 Modula-2, it is accomplished by @code{p^}. Values can also be
13185 represented (and displayed) differently. Hex numbers in C appear as
13186 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
13188 @cindex working language
13189 Language-specific information is built into @value{GDBN} for some languages,
13190 allowing you to express operations like the above in your program's
13191 native language, and allowing @value{GDBN} to output values in a manner
13192 consistent with the syntax of your program's native language. The
13193 language you use to build expressions is called the @dfn{working
13197 * Setting:: Switching between source languages
13198 * Show:: Displaying the language
13199 * Checks:: Type and range checks
13200 * Supported Languages:: Supported languages
13201 * Unsupported Languages:: Unsupported languages
13205 @section Switching Between Source Languages
13207 There are two ways to control the working language---either have @value{GDBN}
13208 set it automatically, or select it manually yourself. You can use the
13209 @code{set language} command for either purpose. On startup, @value{GDBN}
13210 defaults to setting the language automatically. The working language is
13211 used to determine how expressions you type are interpreted, how values
13214 In addition to the working language, every source file that
13215 @value{GDBN} knows about has its own working language. For some object
13216 file formats, the compiler might indicate which language a particular
13217 source file is in. However, most of the time @value{GDBN} infers the
13218 language from the name of the file. The language of a source file
13219 controls whether C@t{++} names are demangled---this way @code{backtrace} can
13220 show each frame appropriately for its own language. There is no way to
13221 set the language of a source file from within @value{GDBN}, but you can
13222 set the language associated with a filename extension. @xref{Show, ,
13223 Displaying the Language}.
13225 This is most commonly a problem when you use a program, such
13226 as @code{cfront} or @code{f2c}, that generates C but is written in
13227 another language. In that case, make the
13228 program use @code{#line} directives in its C output; that way
13229 @value{GDBN} will know the correct language of the source code of the original
13230 program, and will display that source code, not the generated C code.
13233 * Filenames:: Filename extensions and languages.
13234 * Manually:: Setting the working language manually
13235 * Automatically:: Having @value{GDBN} infer the source language
13239 @subsection List of Filename Extensions and Languages
13241 If a source file name ends in one of the following extensions, then
13242 @value{GDBN} infers that its language is the one indicated.
13260 C@t{++} source file
13266 Objective-C source file
13270 Fortran source file
13273 Modula-2 source file
13277 Assembler source file. This actually behaves almost like C, but
13278 @value{GDBN} does not skip over function prologues when stepping.
13281 In addition, you may set the language associated with a filename
13282 extension. @xref{Show, , Displaying the Language}.
13285 @subsection Setting the Working Language
13287 If you allow @value{GDBN} to set the language automatically,
13288 expressions are interpreted the same way in your debugging session and
13291 @kindex set language
13292 If you wish, you may set the language manually. To do this, issue the
13293 command @samp{set language @var{lang}}, where @var{lang} is the name of
13294 a language, such as
13295 @code{c} or @code{modula-2}.
13296 For a list of the supported languages, type @samp{set language}.
13298 Setting the language manually prevents @value{GDBN} from updating the working
13299 language automatically. This can lead to confusion if you try
13300 to debug a program when the working language is not the same as the
13301 source language, when an expression is acceptable to both
13302 languages---but means different things. For instance, if the current
13303 source file were written in C, and @value{GDBN} was parsing Modula-2, a
13311 might not have the effect you intended. In C, this means to add
13312 @code{b} and @code{c} and place the result in @code{a}. The result
13313 printed would be the value of @code{a}. In Modula-2, this means to compare
13314 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
13316 @node Automatically
13317 @subsection Having @value{GDBN} Infer the Source Language
13319 To have @value{GDBN} set the working language automatically, use
13320 @samp{set language local} or @samp{set language auto}. @value{GDBN}
13321 then infers the working language. That is, when your program stops in a
13322 frame (usually by encountering a breakpoint), @value{GDBN} sets the
13323 working language to the language recorded for the function in that
13324 frame. If the language for a frame is unknown (that is, if the function
13325 or block corresponding to the frame was defined in a source file that
13326 does not have a recognized extension), the current working language is
13327 not changed, and @value{GDBN} issues a warning.
13329 This may not seem necessary for most programs, which are written
13330 entirely in one source language. However, program modules and libraries
13331 written in one source language can be used by a main program written in
13332 a different source language. Using @samp{set language auto} in this
13333 case frees you from having to set the working language manually.
13336 @section Displaying the Language
13338 The following commands help you find out which language is the
13339 working language, and also what language source files were written in.
13342 @item show language
13343 @kindex show language
13344 Display the current working language. This is the
13345 language you can use with commands such as @code{print} to
13346 build and compute expressions that may involve variables in your program.
13349 @kindex info frame@r{, show the source language}
13350 Display the source language for this frame. This language becomes the
13351 working language if you use an identifier from this frame.
13352 @xref{Frame Info, ,Information about a Frame}, to identify the other
13353 information listed here.
13356 @kindex info source@r{, show the source language}
13357 Display the source language of this source file.
13358 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
13359 information listed here.
13362 In unusual circumstances, you may have source files with extensions
13363 not in the standard list. You can then set the extension associated
13364 with a language explicitly:
13367 @item set extension-language @var{ext} @var{language}
13368 @kindex set extension-language
13369 Tell @value{GDBN} that source files with extension @var{ext} are to be
13370 assumed as written in the source language @var{language}.
13372 @item info extensions
13373 @kindex info extensions
13374 List all the filename extensions and the associated languages.
13378 @section Type and Range Checking
13380 Some languages are designed to guard you against making seemingly common
13381 errors through a series of compile- and run-time checks. These include
13382 checking the type of arguments to functions and operators and making
13383 sure mathematical overflows are caught at run time. Checks such as
13384 these help to ensure a program's correctness once it has been compiled
13385 by eliminating type mismatches and providing active checks for range
13386 errors when your program is running.
13388 By default @value{GDBN} checks for these errors according to the
13389 rules of the current source language. Although @value{GDBN} does not check
13390 the statements in your program, it can check expressions entered directly
13391 into @value{GDBN} for evaluation via the @code{print} command, for example.
13394 * Type Checking:: An overview of type checking
13395 * Range Checking:: An overview of range checking
13398 @cindex type checking
13399 @cindex checks, type
13400 @node Type Checking
13401 @subsection An Overview of Type Checking
13403 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
13404 arguments to operators and functions have to be of the correct type,
13405 otherwise an error occurs. These checks prevent type mismatch
13406 errors from ever causing any run-time problems. For example,
13409 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
13411 (@value{GDBP}) print obj.my_method (0)
13414 (@value{GDBP}) print obj.my_method (0x1234)
13415 Cannot resolve method klass::my_method to any overloaded instance
13418 The second example fails because in C@t{++} the integer constant
13419 @samp{0x1234} is not type-compatible with the pointer parameter type.
13421 For the expressions you use in @value{GDBN} commands, you can tell
13422 @value{GDBN} to not enforce strict type checking or
13423 to treat any mismatches as errors and abandon the expression;
13424 When type checking is disabled, @value{GDBN} successfully evaluates
13425 expressions like the second example above.
13427 Even if type checking is off, there may be other reasons
13428 related to type that prevent @value{GDBN} from evaluating an expression.
13429 For instance, @value{GDBN} does not know how to add an @code{int} and
13430 a @code{struct foo}. These particular type errors have nothing to do
13431 with the language in use and usually arise from expressions which make
13432 little sense to evaluate anyway.
13434 @value{GDBN} provides some additional commands for controlling type checking:
13436 @kindex set check type
13437 @kindex show check type
13439 @item set check type on
13440 @itemx set check type off
13441 Set strict type checking on or off. If any type mismatches occur in
13442 evaluating an expression while type checking is on, @value{GDBN} prints a
13443 message and aborts evaluation of the expression.
13445 @item show check type
13446 Show the current setting of type checking and whether @value{GDBN}
13447 is enforcing strict type checking rules.
13450 @cindex range checking
13451 @cindex checks, range
13452 @node Range Checking
13453 @subsection An Overview of Range Checking
13455 In some languages (such as Modula-2), it is an error to exceed the
13456 bounds of a type; this is enforced with run-time checks. Such range
13457 checking is meant to ensure program correctness by making sure
13458 computations do not overflow, or indices on an array element access do
13459 not exceed the bounds of the array.
13461 For expressions you use in @value{GDBN} commands, you can tell
13462 @value{GDBN} to treat range errors in one of three ways: ignore them,
13463 always treat them as errors and abandon the expression, or issue
13464 warnings but evaluate the expression anyway.
13466 A range error can result from numerical overflow, from exceeding an
13467 array index bound, or when you type a constant that is not a member
13468 of any type. Some languages, however, do not treat overflows as an
13469 error. In many implementations of C, mathematical overflow causes the
13470 result to ``wrap around'' to lower values---for example, if @var{m} is
13471 the largest integer value, and @var{s} is the smallest, then
13474 @var{m} + 1 @result{} @var{s}
13477 This, too, is specific to individual languages, and in some cases
13478 specific to individual compilers or machines. @xref{Supported Languages, ,
13479 Supported Languages}, for further details on specific languages.
13481 @value{GDBN} provides some additional commands for controlling the range checker:
13483 @kindex set check range
13484 @kindex show check range
13486 @item set check range auto
13487 Set range checking on or off based on the current working language.
13488 @xref{Supported Languages, ,Supported Languages}, for the default settings for
13491 @item set check range on
13492 @itemx set check range off
13493 Set range checking on or off, overriding the default setting for the
13494 current working language. A warning is issued if the setting does not
13495 match the language default. If a range error occurs and range checking is on,
13496 then a message is printed and evaluation of the expression is aborted.
13498 @item set check range warn
13499 Output messages when the @value{GDBN} range checker detects a range error,
13500 but attempt to evaluate the expression anyway. Evaluating the
13501 expression may still be impossible for other reasons, such as accessing
13502 memory that the process does not own (a typical example from many Unix
13506 Show the current setting of the range checker, and whether or not it is
13507 being set automatically by @value{GDBN}.
13510 @node Supported Languages
13511 @section Supported Languages
13513 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran, Java,
13514 OpenCL C, Pascal, assembly, Modula-2, and Ada.
13515 @c This is false ...
13516 Some @value{GDBN} features may be used in expressions regardless of the
13517 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
13518 and the @samp{@{type@}addr} construct (@pxref{Expressions,
13519 ,Expressions}) can be used with the constructs of any supported
13522 The following sections detail to what degree each source language is
13523 supported by @value{GDBN}. These sections are not meant to be language
13524 tutorials or references, but serve only as a reference guide to what the
13525 @value{GDBN} expression parser accepts, and what input and output
13526 formats should look like for different languages. There are many good
13527 books written on each of these languages; please look to these for a
13528 language reference or tutorial.
13531 * C:: C and C@t{++}
13534 * Objective-C:: Objective-C
13535 * OpenCL C:: OpenCL C
13536 * Fortran:: Fortran
13538 * Modula-2:: Modula-2
13543 @subsection C and C@t{++}
13545 @cindex C and C@t{++}
13546 @cindex expressions in C or C@t{++}
13548 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
13549 to both languages. Whenever this is the case, we discuss those languages
13553 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
13554 @cindex @sc{gnu} C@t{++}
13555 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
13556 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
13557 effectively, you must compile your C@t{++} programs with a supported
13558 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
13559 compiler (@code{aCC}).
13562 * C Operators:: C and C@t{++} operators
13563 * C Constants:: C and C@t{++} constants
13564 * C Plus Plus Expressions:: C@t{++} expressions
13565 * C Defaults:: Default settings for C and C@t{++}
13566 * C Checks:: C and C@t{++} type and range checks
13567 * Debugging C:: @value{GDBN} and C
13568 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
13569 * Decimal Floating Point:: Numbers in Decimal Floating Point format
13573 @subsubsection C and C@t{++} Operators
13575 @cindex C and C@t{++} operators
13577 Operators must be defined on values of specific types. For instance,
13578 @code{+} is defined on numbers, but not on structures. Operators are
13579 often defined on groups of types.
13581 For the purposes of C and C@t{++}, the following definitions hold:
13586 @emph{Integral types} include @code{int} with any of its storage-class
13587 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
13590 @emph{Floating-point types} include @code{float}, @code{double}, and
13591 @code{long double} (if supported by the target platform).
13594 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
13597 @emph{Scalar types} include all of the above.
13602 The following operators are supported. They are listed here
13603 in order of increasing precedence:
13607 The comma or sequencing operator. Expressions in a comma-separated list
13608 are evaluated from left to right, with the result of the entire
13609 expression being the last expression evaluated.
13612 Assignment. The value of an assignment expression is the value
13613 assigned. Defined on scalar types.
13616 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
13617 and translated to @w{@code{@var{a} = @var{a op b}}}.
13618 @w{@code{@var{op}=}} and @code{=} have the same precedence.
13619 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
13620 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
13623 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
13624 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
13628 Logical @sc{or}. Defined on integral types.
13631 Logical @sc{and}. Defined on integral types.
13634 Bitwise @sc{or}. Defined on integral types.
13637 Bitwise exclusive-@sc{or}. Defined on integral types.
13640 Bitwise @sc{and}. Defined on integral types.
13643 Equality and inequality. Defined on scalar types. The value of these
13644 expressions is 0 for false and non-zero for true.
13646 @item <@r{, }>@r{, }<=@r{, }>=
13647 Less than, greater than, less than or equal, greater than or equal.
13648 Defined on scalar types. The value of these expressions is 0 for false
13649 and non-zero for true.
13652 left shift, and right shift. Defined on integral types.
13655 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
13658 Addition and subtraction. Defined on integral types, floating-point types and
13661 @item *@r{, }/@r{, }%
13662 Multiplication, division, and modulus. Multiplication and division are
13663 defined on integral and floating-point types. Modulus is defined on
13667 Increment and decrement. When appearing before a variable, the
13668 operation is performed before the variable is used in an expression;
13669 when appearing after it, the variable's value is used before the
13670 operation takes place.
13673 Pointer dereferencing. Defined on pointer types. Same precedence as
13677 Address operator. Defined on variables. Same precedence as @code{++}.
13679 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
13680 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
13681 to examine the address
13682 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
13686 Negative. Defined on integral and floating-point types. Same
13687 precedence as @code{++}.
13690 Logical negation. Defined on integral types. Same precedence as
13694 Bitwise complement operator. Defined on integral types. Same precedence as
13699 Structure member, and pointer-to-structure member. For convenience,
13700 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
13701 pointer based on the stored type information.
13702 Defined on @code{struct} and @code{union} data.
13705 Dereferences of pointers to members.
13708 Array indexing. @code{@var{a}[@var{i}]} is defined as
13709 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
13712 Function parameter list. Same precedence as @code{->}.
13715 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
13716 and @code{class} types.
13719 Doubled colons also represent the @value{GDBN} scope operator
13720 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
13724 If an operator is redefined in the user code, @value{GDBN} usually
13725 attempts to invoke the redefined version instead of using the operator's
13726 predefined meaning.
13729 @subsubsection C and C@t{++} Constants
13731 @cindex C and C@t{++} constants
13733 @value{GDBN} allows you to express the constants of C and C@t{++} in the
13738 Integer constants are a sequence of digits. Octal constants are
13739 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
13740 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
13741 @samp{l}, specifying that the constant should be treated as a
13745 Floating point constants are a sequence of digits, followed by a decimal
13746 point, followed by a sequence of digits, and optionally followed by an
13747 exponent. An exponent is of the form:
13748 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
13749 sequence of digits. The @samp{+} is optional for positive exponents.
13750 A floating-point constant may also end with a letter @samp{f} or
13751 @samp{F}, specifying that the constant should be treated as being of
13752 the @code{float} (as opposed to the default @code{double}) type; or with
13753 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
13757 Enumerated constants consist of enumerated identifiers, or their
13758 integral equivalents.
13761 Character constants are a single character surrounded by single quotes
13762 (@code{'}), or a number---the ordinal value of the corresponding character
13763 (usually its @sc{ascii} value). Within quotes, the single character may
13764 be represented by a letter or by @dfn{escape sequences}, which are of
13765 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
13766 of the character's ordinal value; or of the form @samp{\@var{x}}, where
13767 @samp{@var{x}} is a predefined special character---for example,
13768 @samp{\n} for newline.
13770 Wide character constants can be written by prefixing a character
13771 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
13772 form of @samp{x}. The target wide character set is used when
13773 computing the value of this constant (@pxref{Character Sets}).
13776 String constants are a sequence of character constants surrounded by
13777 double quotes (@code{"}). Any valid character constant (as described
13778 above) may appear. Double quotes within the string must be preceded by
13779 a backslash, so for instance @samp{"a\"b'c"} is a string of five
13782 Wide string constants can be written by prefixing a string constant
13783 with @samp{L}, as in C. The target wide character set is used when
13784 computing the value of this constant (@pxref{Character Sets}).
13787 Pointer constants are an integral value. You can also write pointers
13788 to constants using the C operator @samp{&}.
13791 Array constants are comma-separated lists surrounded by braces @samp{@{}
13792 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
13793 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
13794 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
13797 @node C Plus Plus Expressions
13798 @subsubsection C@t{++} Expressions
13800 @cindex expressions in C@t{++}
13801 @value{GDBN} expression handling can interpret most C@t{++} expressions.
13803 @cindex debugging C@t{++} programs
13804 @cindex C@t{++} compilers
13805 @cindex debug formats and C@t{++}
13806 @cindex @value{NGCC} and C@t{++}
13808 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
13809 the proper compiler and the proper debug format. Currently,
13810 @value{GDBN} works best when debugging C@t{++} code that is compiled
13811 with the most recent version of @value{NGCC} possible. The DWARF
13812 debugging format is preferred; @value{NGCC} defaults to this on most
13813 popular platforms. Other compilers and/or debug formats are likely to
13814 work badly or not at all when using @value{GDBN} to debug C@t{++}
13815 code. @xref{Compilation}.
13820 @cindex member functions
13822 Member function calls are allowed; you can use expressions like
13825 count = aml->GetOriginal(x, y)
13828 @vindex this@r{, inside C@t{++} member functions}
13829 @cindex namespace in C@t{++}
13831 While a member function is active (in the selected stack frame), your
13832 expressions have the same namespace available as the member function;
13833 that is, @value{GDBN} allows implicit references to the class instance
13834 pointer @code{this} following the same rules as C@t{++}. @code{using}
13835 declarations in the current scope are also respected by @value{GDBN}.
13837 @cindex call overloaded functions
13838 @cindex overloaded functions, calling
13839 @cindex type conversions in C@t{++}
13841 You can call overloaded functions; @value{GDBN} resolves the function
13842 call to the right definition, with some restrictions. @value{GDBN} does not
13843 perform overload resolution involving user-defined type conversions,
13844 calls to constructors, or instantiations of templates that do not exist
13845 in the program. It also cannot handle ellipsis argument lists or
13848 It does perform integral conversions and promotions, floating-point
13849 promotions, arithmetic conversions, pointer conversions, conversions of
13850 class objects to base classes, and standard conversions such as those of
13851 functions or arrays to pointers; it requires an exact match on the
13852 number of function arguments.
13854 Overload resolution is always performed, unless you have specified
13855 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
13856 ,@value{GDBN} Features for C@t{++}}.
13858 You must specify @code{set overload-resolution off} in order to use an
13859 explicit function signature to call an overloaded function, as in
13861 p 'foo(char,int)'('x', 13)
13864 The @value{GDBN} command-completion facility can simplify this;
13865 see @ref{Completion, ,Command Completion}.
13867 @cindex reference declarations
13869 @value{GDBN} understands variables declared as C@t{++} references; you can use
13870 them in expressions just as you do in C@t{++} source---they are automatically
13873 In the parameter list shown when @value{GDBN} displays a frame, the values of
13874 reference variables are not displayed (unlike other variables); this
13875 avoids clutter, since references are often used for large structures.
13876 The @emph{address} of a reference variable is always shown, unless
13877 you have specified @samp{set print address off}.
13880 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
13881 expressions can use it just as expressions in your program do. Since
13882 one scope may be defined in another, you can use @code{::} repeatedly if
13883 necessary, for example in an expression like
13884 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
13885 resolving name scope by reference to source files, in both C and C@t{++}
13886 debugging (@pxref{Variables, ,Program Variables}).
13889 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
13894 @subsubsection C and C@t{++} Defaults
13896 @cindex C and C@t{++} defaults
13898 If you allow @value{GDBN} to set range checking automatically, it
13899 defaults to @code{off} whenever the working language changes to
13900 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
13901 selects the working language.
13903 If you allow @value{GDBN} to set the language automatically, it
13904 recognizes source files whose names end with @file{.c}, @file{.C}, or
13905 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
13906 these files, it sets the working language to C or C@t{++}.
13907 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
13908 for further details.
13911 @subsubsection C and C@t{++} Type and Range Checks
13913 @cindex C and C@t{++} checks
13915 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
13916 checking is used. However, if you turn type checking off, @value{GDBN}
13917 will allow certain non-standard conversions, such as promoting integer
13918 constants to pointers.
13920 Range checking, if turned on, is done on mathematical operations. Array
13921 indices are not checked, since they are often used to index a pointer
13922 that is not itself an array.
13925 @subsubsection @value{GDBN} and C
13927 The @code{set print union} and @code{show print union} commands apply to
13928 the @code{union} type. When set to @samp{on}, any @code{union} that is
13929 inside a @code{struct} or @code{class} is also printed. Otherwise, it
13930 appears as @samp{@{...@}}.
13932 The @code{@@} operator aids in the debugging of dynamic arrays, formed
13933 with pointers and a memory allocation function. @xref{Expressions,
13936 @node Debugging C Plus Plus
13937 @subsubsection @value{GDBN} Features for C@t{++}
13939 @cindex commands for C@t{++}
13941 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
13942 designed specifically for use with C@t{++}. Here is a summary:
13945 @cindex break in overloaded functions
13946 @item @r{breakpoint menus}
13947 When you want a breakpoint in a function whose name is overloaded,
13948 @value{GDBN} has the capability to display a menu of possible breakpoint
13949 locations to help you specify which function definition you want.
13950 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
13952 @cindex overloading in C@t{++}
13953 @item rbreak @var{regex}
13954 Setting breakpoints using regular expressions is helpful for setting
13955 breakpoints on overloaded functions that are not members of any special
13957 @xref{Set Breaks, ,Setting Breakpoints}.
13959 @cindex C@t{++} exception handling
13961 @itemx catch rethrow
13963 Debug C@t{++} exception handling using these commands. @xref{Set
13964 Catchpoints, , Setting Catchpoints}.
13966 @cindex inheritance
13967 @item ptype @var{typename}
13968 Print inheritance relationships as well as other information for type
13970 @xref{Symbols, ,Examining the Symbol Table}.
13972 @item info vtbl @var{expression}.
13973 The @code{info vtbl} command can be used to display the virtual
13974 method tables of the object computed by @var{expression}. This shows
13975 one entry per virtual table; there may be multiple virtual tables when
13976 multiple inheritance is in use.
13978 @cindex C@t{++} symbol display
13979 @item set print demangle
13980 @itemx show print demangle
13981 @itemx set print asm-demangle
13982 @itemx show print asm-demangle
13983 Control whether C@t{++} symbols display in their source form, both when
13984 displaying code as C@t{++} source and when displaying disassemblies.
13985 @xref{Print Settings, ,Print Settings}.
13987 @item set print object
13988 @itemx show print object
13989 Choose whether to print derived (actual) or declared types of objects.
13990 @xref{Print Settings, ,Print Settings}.
13992 @item set print vtbl
13993 @itemx show print vtbl
13994 Control the format for printing virtual function tables.
13995 @xref{Print Settings, ,Print Settings}.
13996 (The @code{vtbl} commands do not work on programs compiled with the HP
13997 ANSI C@t{++} compiler (@code{aCC}).)
13999 @kindex set overload-resolution
14000 @cindex overloaded functions, overload resolution
14001 @item set overload-resolution on
14002 Enable overload resolution for C@t{++} expression evaluation. The default
14003 is on. For overloaded functions, @value{GDBN} evaluates the arguments
14004 and searches for a function whose signature matches the argument types,
14005 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
14006 Expressions, ,C@t{++} Expressions}, for details).
14007 If it cannot find a match, it emits a message.
14009 @item set overload-resolution off
14010 Disable overload resolution for C@t{++} expression evaluation. For
14011 overloaded functions that are not class member functions, @value{GDBN}
14012 chooses the first function of the specified name that it finds in the
14013 symbol table, whether or not its arguments are of the correct type. For
14014 overloaded functions that are class member functions, @value{GDBN}
14015 searches for a function whose signature @emph{exactly} matches the
14018 @kindex show overload-resolution
14019 @item show overload-resolution
14020 Show the current setting of overload resolution.
14022 @item @r{Overloaded symbol names}
14023 You can specify a particular definition of an overloaded symbol, using
14024 the same notation that is used to declare such symbols in C@t{++}: type
14025 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
14026 also use the @value{GDBN} command-line word completion facilities to list the
14027 available choices, or to finish the type list for you.
14028 @xref{Completion,, Command Completion}, for details on how to do this.
14031 @node Decimal Floating Point
14032 @subsubsection Decimal Floating Point format
14033 @cindex decimal floating point format
14035 @value{GDBN} can examine, set and perform computations with numbers in
14036 decimal floating point format, which in the C language correspond to the
14037 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
14038 specified by the extension to support decimal floating-point arithmetic.
14040 There are two encodings in use, depending on the architecture: BID (Binary
14041 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
14042 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
14045 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
14046 to manipulate decimal floating point numbers, it is not possible to convert
14047 (using a cast, for example) integers wider than 32-bit to decimal float.
14049 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
14050 point computations, error checking in decimal float operations ignores
14051 underflow, overflow and divide by zero exceptions.
14053 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
14054 to inspect @code{_Decimal128} values stored in floating point registers.
14055 See @ref{PowerPC,,PowerPC} for more details.
14061 @value{GDBN} can be used to debug programs written in D and compiled with
14062 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
14063 specific feature --- dynamic arrays.
14068 @cindex Go (programming language)
14069 @value{GDBN} can be used to debug programs written in Go and compiled with
14070 @file{gccgo} or @file{6g} compilers.
14072 Here is a summary of the Go-specific features and restrictions:
14075 @cindex current Go package
14076 @item The current Go package
14077 The name of the current package does not need to be specified when
14078 specifying global variables and functions.
14080 For example, given the program:
14084 var myglob = "Shall we?"
14090 When stopped inside @code{main} either of these work:
14094 (gdb) p main.myglob
14097 @cindex builtin Go types
14098 @item Builtin Go types
14099 The @code{string} type is recognized by @value{GDBN} and is printed
14102 @cindex builtin Go functions
14103 @item Builtin Go functions
14104 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
14105 function and handles it internally.
14107 @cindex restrictions on Go expressions
14108 @item Restrictions on Go expressions
14109 All Go operators are supported except @code{&^}.
14110 The Go @code{_} ``blank identifier'' is not supported.
14111 Automatic dereferencing of pointers is not supported.
14115 @subsection Objective-C
14117 @cindex Objective-C
14118 This section provides information about some commands and command
14119 options that are useful for debugging Objective-C code. See also
14120 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
14121 few more commands specific to Objective-C support.
14124 * Method Names in Commands::
14125 * The Print Command with Objective-C::
14128 @node Method Names in Commands
14129 @subsubsection Method Names in Commands
14131 The following commands have been extended to accept Objective-C method
14132 names as line specifications:
14134 @kindex clear@r{, and Objective-C}
14135 @kindex break@r{, and Objective-C}
14136 @kindex info line@r{, and Objective-C}
14137 @kindex jump@r{, and Objective-C}
14138 @kindex list@r{, and Objective-C}
14142 @item @code{info line}
14147 A fully qualified Objective-C method name is specified as
14150 -[@var{Class} @var{methodName}]
14153 where the minus sign is used to indicate an instance method and a
14154 plus sign (not shown) is used to indicate a class method. The class
14155 name @var{Class} and method name @var{methodName} are enclosed in
14156 brackets, similar to the way messages are specified in Objective-C
14157 source code. For example, to set a breakpoint at the @code{create}
14158 instance method of class @code{Fruit} in the program currently being
14162 break -[Fruit create]
14165 To list ten program lines around the @code{initialize} class method,
14169 list +[NSText initialize]
14172 In the current version of @value{GDBN}, the plus or minus sign is
14173 required. In future versions of @value{GDBN}, the plus or minus
14174 sign will be optional, but you can use it to narrow the search. It
14175 is also possible to specify just a method name:
14181 You must specify the complete method name, including any colons. If
14182 your program's source files contain more than one @code{create} method,
14183 you'll be presented with a numbered list of classes that implement that
14184 method. Indicate your choice by number, or type @samp{0} to exit if
14187 As another example, to clear a breakpoint established at the
14188 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
14191 clear -[NSWindow makeKeyAndOrderFront:]
14194 @node The Print Command with Objective-C
14195 @subsubsection The Print Command With Objective-C
14196 @cindex Objective-C, print objects
14197 @kindex print-object
14198 @kindex po @r{(@code{print-object})}
14200 The print command has also been extended to accept methods. For example:
14203 print -[@var{object} hash]
14206 @cindex print an Objective-C object description
14207 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
14209 will tell @value{GDBN} to send the @code{hash} message to @var{object}
14210 and print the result. Also, an additional command has been added,
14211 @code{print-object} or @code{po} for short, which is meant to print
14212 the description of an object. However, this command may only work
14213 with certain Objective-C libraries that have a particular hook
14214 function, @code{_NSPrintForDebugger}, defined.
14217 @subsection OpenCL C
14220 This section provides information about @value{GDBN}s OpenCL C support.
14223 * OpenCL C Datatypes::
14224 * OpenCL C Expressions::
14225 * OpenCL C Operators::
14228 @node OpenCL C Datatypes
14229 @subsubsection OpenCL C Datatypes
14231 @cindex OpenCL C Datatypes
14232 @value{GDBN} supports the builtin scalar and vector datatypes specified
14233 by OpenCL 1.1. In addition the half- and double-precision floating point
14234 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
14235 extensions are also known to @value{GDBN}.
14237 @node OpenCL C Expressions
14238 @subsubsection OpenCL C Expressions
14240 @cindex OpenCL C Expressions
14241 @value{GDBN} supports accesses to vector components including the access as
14242 lvalue where possible. Since OpenCL C is based on C99 most C expressions
14243 supported by @value{GDBN} can be used as well.
14245 @node OpenCL C Operators
14246 @subsubsection OpenCL C Operators
14248 @cindex OpenCL C Operators
14249 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
14253 @subsection Fortran
14254 @cindex Fortran-specific support in @value{GDBN}
14256 @value{GDBN} can be used to debug programs written in Fortran, but it
14257 currently supports only the features of Fortran 77 language.
14259 @cindex trailing underscore, in Fortran symbols
14260 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
14261 among them) append an underscore to the names of variables and
14262 functions. When you debug programs compiled by those compilers, you
14263 will need to refer to variables and functions with a trailing
14267 * Fortran Operators:: Fortran operators and expressions
14268 * Fortran Defaults:: Default settings for Fortran
14269 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
14272 @node Fortran Operators
14273 @subsubsection Fortran Operators and Expressions
14275 @cindex Fortran operators and expressions
14277 Operators must be defined on values of specific types. For instance,
14278 @code{+} is defined on numbers, but not on characters or other non-
14279 arithmetic types. Operators are often defined on groups of types.
14283 The exponentiation operator. It raises the first operand to the power
14287 The range operator. Normally used in the form of array(low:high) to
14288 represent a section of array.
14291 The access component operator. Normally used to access elements in derived
14292 types. Also suitable for unions. As unions aren't part of regular Fortran,
14293 this can only happen when accessing a register that uses a gdbarch-defined
14297 @node Fortran Defaults
14298 @subsubsection Fortran Defaults
14300 @cindex Fortran Defaults
14302 Fortran symbols are usually case-insensitive, so @value{GDBN} by
14303 default uses case-insensitive matches for Fortran symbols. You can
14304 change that with the @samp{set case-insensitive} command, see
14305 @ref{Symbols}, for the details.
14307 @node Special Fortran Commands
14308 @subsubsection Special Fortran Commands
14310 @cindex Special Fortran commands
14312 @value{GDBN} has some commands to support Fortran-specific features,
14313 such as displaying common blocks.
14316 @cindex @code{COMMON} blocks, Fortran
14317 @kindex info common
14318 @item info common @r{[}@var{common-name}@r{]}
14319 This command prints the values contained in the Fortran @code{COMMON}
14320 block whose name is @var{common-name}. With no argument, the names of
14321 all @code{COMMON} blocks visible at the current program location are
14328 @cindex Pascal support in @value{GDBN}, limitations
14329 Debugging Pascal programs which use sets, subranges, file variables, or
14330 nested functions does not currently work. @value{GDBN} does not support
14331 entering expressions, printing values, or similar features using Pascal
14334 The Pascal-specific command @code{set print pascal_static-members}
14335 controls whether static members of Pascal objects are displayed.
14336 @xref{Print Settings, pascal_static-members}.
14339 @subsection Modula-2
14341 @cindex Modula-2, @value{GDBN} support
14343 The extensions made to @value{GDBN} to support Modula-2 only support
14344 output from the @sc{gnu} Modula-2 compiler (which is currently being
14345 developed). Other Modula-2 compilers are not currently supported, and
14346 attempting to debug executables produced by them is most likely
14347 to give an error as @value{GDBN} reads in the executable's symbol
14350 @cindex expressions in Modula-2
14352 * M2 Operators:: Built-in operators
14353 * Built-In Func/Proc:: Built-in functions and procedures
14354 * M2 Constants:: Modula-2 constants
14355 * M2 Types:: Modula-2 types
14356 * M2 Defaults:: Default settings for Modula-2
14357 * Deviations:: Deviations from standard Modula-2
14358 * M2 Checks:: Modula-2 type and range checks
14359 * M2 Scope:: The scope operators @code{::} and @code{.}
14360 * GDB/M2:: @value{GDBN} and Modula-2
14364 @subsubsection Operators
14365 @cindex Modula-2 operators
14367 Operators must be defined on values of specific types. For instance,
14368 @code{+} is defined on numbers, but not on structures. Operators are
14369 often defined on groups of types. For the purposes of Modula-2, the
14370 following definitions hold:
14375 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
14379 @emph{Character types} consist of @code{CHAR} and its subranges.
14382 @emph{Floating-point types} consist of @code{REAL}.
14385 @emph{Pointer types} consist of anything declared as @code{POINTER TO
14389 @emph{Scalar types} consist of all of the above.
14392 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
14395 @emph{Boolean types} consist of @code{BOOLEAN}.
14399 The following operators are supported, and appear in order of
14400 increasing precedence:
14404 Function argument or array index separator.
14407 Assignment. The value of @var{var} @code{:=} @var{value} is
14411 Less than, greater than on integral, floating-point, or enumerated
14415 Less than or equal to, greater than or equal to
14416 on integral, floating-point and enumerated types, or set inclusion on
14417 set types. Same precedence as @code{<}.
14419 @item =@r{, }<>@r{, }#
14420 Equality and two ways of expressing inequality, valid on scalar types.
14421 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
14422 available for inequality, since @code{#} conflicts with the script
14426 Set membership. Defined on set types and the types of their members.
14427 Same precedence as @code{<}.
14430 Boolean disjunction. Defined on boolean types.
14433 Boolean conjunction. Defined on boolean types.
14436 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
14439 Addition and subtraction on integral and floating-point types, or union
14440 and difference on set types.
14443 Multiplication on integral and floating-point types, or set intersection
14447 Division on floating-point types, or symmetric set difference on set
14448 types. Same precedence as @code{*}.
14451 Integer division and remainder. Defined on integral types. Same
14452 precedence as @code{*}.
14455 Negative. Defined on @code{INTEGER} and @code{REAL} data.
14458 Pointer dereferencing. Defined on pointer types.
14461 Boolean negation. Defined on boolean types. Same precedence as
14465 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
14466 precedence as @code{^}.
14469 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
14472 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
14476 @value{GDBN} and Modula-2 scope operators.
14480 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
14481 treats the use of the operator @code{IN}, or the use of operators
14482 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
14483 @code{<=}, and @code{>=} on sets as an error.
14487 @node Built-In Func/Proc
14488 @subsubsection Built-in Functions and Procedures
14489 @cindex Modula-2 built-ins
14491 Modula-2 also makes available several built-in procedures and functions.
14492 In describing these, the following metavariables are used:
14497 represents an @code{ARRAY} variable.
14500 represents a @code{CHAR} constant or variable.
14503 represents a variable or constant of integral type.
14506 represents an identifier that belongs to a set. Generally used in the
14507 same function with the metavariable @var{s}. The type of @var{s} should
14508 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
14511 represents a variable or constant of integral or floating-point type.
14514 represents a variable or constant of floating-point type.
14520 represents a variable.
14523 represents a variable or constant of one of many types. See the
14524 explanation of the function for details.
14527 All Modula-2 built-in procedures also return a result, described below.
14531 Returns the absolute value of @var{n}.
14534 If @var{c} is a lower case letter, it returns its upper case
14535 equivalent, otherwise it returns its argument.
14538 Returns the character whose ordinal value is @var{i}.
14541 Decrements the value in the variable @var{v} by one. Returns the new value.
14543 @item DEC(@var{v},@var{i})
14544 Decrements the value in the variable @var{v} by @var{i}. Returns the
14547 @item EXCL(@var{m},@var{s})
14548 Removes the element @var{m} from the set @var{s}. Returns the new
14551 @item FLOAT(@var{i})
14552 Returns the floating point equivalent of the integer @var{i}.
14554 @item HIGH(@var{a})
14555 Returns the index of the last member of @var{a}.
14558 Increments the value in the variable @var{v} by one. Returns the new value.
14560 @item INC(@var{v},@var{i})
14561 Increments the value in the variable @var{v} by @var{i}. Returns the
14564 @item INCL(@var{m},@var{s})
14565 Adds the element @var{m} to the set @var{s} if it is not already
14566 there. Returns the new set.
14569 Returns the maximum value of the type @var{t}.
14572 Returns the minimum value of the type @var{t}.
14575 Returns boolean TRUE if @var{i} is an odd number.
14578 Returns the ordinal value of its argument. For example, the ordinal
14579 value of a character is its @sc{ascii} value (on machines supporting the
14580 @sc{ascii} character set). @var{x} must be of an ordered type, which include
14581 integral, character and enumerated types.
14583 @item SIZE(@var{x})
14584 Returns the size of its argument. @var{x} can be a variable or a type.
14586 @item TRUNC(@var{r})
14587 Returns the integral part of @var{r}.
14589 @item TSIZE(@var{x})
14590 Returns the size of its argument. @var{x} can be a variable or a type.
14592 @item VAL(@var{t},@var{i})
14593 Returns the member of the type @var{t} whose ordinal value is @var{i}.
14597 @emph{Warning:} Sets and their operations are not yet supported, so
14598 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
14602 @cindex Modula-2 constants
14604 @subsubsection Constants
14606 @value{GDBN} allows you to express the constants of Modula-2 in the following
14612 Integer constants are simply a sequence of digits. When used in an
14613 expression, a constant is interpreted to be type-compatible with the
14614 rest of the expression. Hexadecimal integers are specified by a
14615 trailing @samp{H}, and octal integers by a trailing @samp{B}.
14618 Floating point constants appear as a sequence of digits, followed by a
14619 decimal point and another sequence of digits. An optional exponent can
14620 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
14621 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
14622 digits of the floating point constant must be valid decimal (base 10)
14626 Character constants consist of a single character enclosed by a pair of
14627 like quotes, either single (@code{'}) or double (@code{"}). They may
14628 also be expressed by their ordinal value (their @sc{ascii} value, usually)
14629 followed by a @samp{C}.
14632 String constants consist of a sequence of characters enclosed by a
14633 pair of like quotes, either single (@code{'}) or double (@code{"}).
14634 Escape sequences in the style of C are also allowed. @xref{C
14635 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
14639 Enumerated constants consist of an enumerated identifier.
14642 Boolean constants consist of the identifiers @code{TRUE} and
14646 Pointer constants consist of integral values only.
14649 Set constants are not yet supported.
14653 @subsubsection Modula-2 Types
14654 @cindex Modula-2 types
14656 Currently @value{GDBN} can print the following data types in Modula-2
14657 syntax: array types, record types, set types, pointer types, procedure
14658 types, enumerated types, subrange types and base types. You can also
14659 print the contents of variables declared using these type.
14660 This section gives a number of simple source code examples together with
14661 sample @value{GDBN} sessions.
14663 The first example contains the following section of code:
14672 and you can request @value{GDBN} to interrogate the type and value of
14673 @code{r} and @code{s}.
14676 (@value{GDBP}) print s
14678 (@value{GDBP}) ptype s
14680 (@value{GDBP}) print r
14682 (@value{GDBP}) ptype r
14687 Likewise if your source code declares @code{s} as:
14691 s: SET ['A'..'Z'] ;
14695 then you may query the type of @code{s} by:
14698 (@value{GDBP}) ptype s
14699 type = SET ['A'..'Z']
14703 Note that at present you cannot interactively manipulate set
14704 expressions using the debugger.
14706 The following example shows how you might declare an array in Modula-2
14707 and how you can interact with @value{GDBN} to print its type and contents:
14711 s: ARRAY [-10..10] OF CHAR ;
14715 (@value{GDBP}) ptype s
14716 ARRAY [-10..10] OF CHAR
14719 Note that the array handling is not yet complete and although the type
14720 is printed correctly, expression handling still assumes that all
14721 arrays have a lower bound of zero and not @code{-10} as in the example
14724 Here are some more type related Modula-2 examples:
14728 colour = (blue, red, yellow, green) ;
14729 t = [blue..yellow] ;
14737 The @value{GDBN} interaction shows how you can query the data type
14738 and value of a variable.
14741 (@value{GDBP}) print s
14743 (@value{GDBP}) ptype t
14744 type = [blue..yellow]
14748 In this example a Modula-2 array is declared and its contents
14749 displayed. Observe that the contents are written in the same way as
14750 their @code{C} counterparts.
14754 s: ARRAY [1..5] OF CARDINAL ;
14760 (@value{GDBP}) print s
14761 $1 = @{1, 0, 0, 0, 0@}
14762 (@value{GDBP}) ptype s
14763 type = ARRAY [1..5] OF CARDINAL
14766 The Modula-2 language interface to @value{GDBN} also understands
14767 pointer types as shown in this example:
14771 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
14778 and you can request that @value{GDBN} describes the type of @code{s}.
14781 (@value{GDBP}) ptype s
14782 type = POINTER TO ARRAY [1..5] OF CARDINAL
14785 @value{GDBN} handles compound types as we can see in this example.
14786 Here we combine array types, record types, pointer types and subrange
14797 myarray = ARRAY myrange OF CARDINAL ;
14798 myrange = [-2..2] ;
14800 s: POINTER TO ARRAY myrange OF foo ;
14804 and you can ask @value{GDBN} to describe the type of @code{s} as shown
14808 (@value{GDBP}) ptype s
14809 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
14812 f3 : ARRAY [-2..2] OF CARDINAL;
14817 @subsubsection Modula-2 Defaults
14818 @cindex Modula-2 defaults
14820 If type and range checking are set automatically by @value{GDBN}, they
14821 both default to @code{on} whenever the working language changes to
14822 Modula-2. This happens regardless of whether you or @value{GDBN}
14823 selected the working language.
14825 If you allow @value{GDBN} to set the language automatically, then entering
14826 code compiled from a file whose name ends with @file{.mod} sets the
14827 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
14828 Infer the Source Language}, for further details.
14831 @subsubsection Deviations from Standard Modula-2
14832 @cindex Modula-2, deviations from
14834 A few changes have been made to make Modula-2 programs easier to debug.
14835 This is done primarily via loosening its type strictness:
14839 Unlike in standard Modula-2, pointer constants can be formed by
14840 integers. This allows you to modify pointer variables during
14841 debugging. (In standard Modula-2, the actual address contained in a
14842 pointer variable is hidden from you; it can only be modified
14843 through direct assignment to another pointer variable or expression that
14844 returned a pointer.)
14847 C escape sequences can be used in strings and characters to represent
14848 non-printable characters. @value{GDBN} prints out strings with these
14849 escape sequences embedded. Single non-printable characters are
14850 printed using the @samp{CHR(@var{nnn})} format.
14853 The assignment operator (@code{:=}) returns the value of its right-hand
14857 All built-in procedures both modify @emph{and} return their argument.
14861 @subsubsection Modula-2 Type and Range Checks
14862 @cindex Modula-2 checks
14865 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
14868 @c FIXME remove warning when type/range checks added
14870 @value{GDBN} considers two Modula-2 variables type equivalent if:
14874 They are of types that have been declared equivalent via a @code{TYPE
14875 @var{t1} = @var{t2}} statement
14878 They have been declared on the same line. (Note: This is true of the
14879 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
14882 As long as type checking is enabled, any attempt to combine variables
14883 whose types are not equivalent is an error.
14885 Range checking is done on all mathematical operations, assignment, array
14886 index bounds, and all built-in functions and procedures.
14889 @subsubsection The Scope Operators @code{::} and @code{.}
14891 @cindex @code{.}, Modula-2 scope operator
14892 @cindex colon, doubled as scope operator
14894 @vindex colon-colon@r{, in Modula-2}
14895 @c Info cannot handle :: but TeX can.
14898 @vindex ::@r{, in Modula-2}
14901 There are a few subtle differences between the Modula-2 scope operator
14902 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
14907 @var{module} . @var{id}
14908 @var{scope} :: @var{id}
14912 where @var{scope} is the name of a module or a procedure,
14913 @var{module} the name of a module, and @var{id} is any declared
14914 identifier within your program, except another module.
14916 Using the @code{::} operator makes @value{GDBN} search the scope
14917 specified by @var{scope} for the identifier @var{id}. If it is not
14918 found in the specified scope, then @value{GDBN} searches all scopes
14919 enclosing the one specified by @var{scope}.
14921 Using the @code{.} operator makes @value{GDBN} search the current scope for
14922 the identifier specified by @var{id} that was imported from the
14923 definition module specified by @var{module}. With this operator, it is
14924 an error if the identifier @var{id} was not imported from definition
14925 module @var{module}, or if @var{id} is not an identifier in
14929 @subsubsection @value{GDBN} and Modula-2
14931 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
14932 Five subcommands of @code{set print} and @code{show print} apply
14933 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
14934 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
14935 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
14936 analogue in Modula-2.
14938 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
14939 with any language, is not useful with Modula-2. Its
14940 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
14941 created in Modula-2 as they can in C or C@t{++}. However, because an
14942 address can be specified by an integral constant, the construct
14943 @samp{@{@var{type}@}@var{adrexp}} is still useful.
14945 @cindex @code{#} in Modula-2
14946 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
14947 interpreted as the beginning of a comment. Use @code{<>} instead.
14953 The extensions made to @value{GDBN} for Ada only support
14954 output from the @sc{gnu} Ada (GNAT) compiler.
14955 Other Ada compilers are not currently supported, and
14956 attempting to debug executables produced by them is most likely
14960 @cindex expressions in Ada
14962 * Ada Mode Intro:: General remarks on the Ada syntax
14963 and semantics supported by Ada mode
14965 * Omissions from Ada:: Restrictions on the Ada expression syntax.
14966 * Additions to Ada:: Extensions of the Ada expression syntax.
14967 * Stopping Before Main Program:: Debugging the program during elaboration.
14968 * Ada Tasks:: Listing and setting breakpoints in tasks.
14969 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
14970 * Ravenscar Profile:: Tasking Support when using the Ravenscar
14972 * Ada Glitches:: Known peculiarities of Ada mode.
14975 @node Ada Mode Intro
14976 @subsubsection Introduction
14977 @cindex Ada mode, general
14979 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
14980 syntax, with some extensions.
14981 The philosophy behind the design of this subset is
14985 That @value{GDBN} should provide basic literals and access to operations for
14986 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
14987 leaving more sophisticated computations to subprograms written into the
14988 program (which therefore may be called from @value{GDBN}).
14991 That type safety and strict adherence to Ada language restrictions
14992 are not particularly important to the @value{GDBN} user.
14995 That brevity is important to the @value{GDBN} user.
14998 Thus, for brevity, the debugger acts as if all names declared in
14999 user-written packages are directly visible, even if they are not visible
15000 according to Ada rules, thus making it unnecessary to fully qualify most
15001 names with their packages, regardless of context. Where this causes
15002 ambiguity, @value{GDBN} asks the user's intent.
15004 The debugger will start in Ada mode if it detects an Ada main program.
15005 As for other languages, it will enter Ada mode when stopped in a program that
15006 was translated from an Ada source file.
15008 While in Ada mode, you may use `@t{--}' for comments. This is useful
15009 mostly for documenting command files. The standard @value{GDBN} comment
15010 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
15011 middle (to allow based literals).
15013 The debugger supports limited overloading. Given a subprogram call in which
15014 the function symbol has multiple definitions, it will use the number of
15015 actual parameters and some information about their types to attempt to narrow
15016 the set of definitions. It also makes very limited use of context, preferring
15017 procedures to functions in the context of the @code{call} command, and
15018 functions to procedures elsewhere.
15020 @node Omissions from Ada
15021 @subsubsection Omissions from Ada
15022 @cindex Ada, omissions from
15024 Here are the notable omissions from the subset:
15028 Only a subset of the attributes are supported:
15032 @t{'First}, @t{'Last}, and @t{'Length}
15033 on array objects (not on types and subtypes).
15036 @t{'Min} and @t{'Max}.
15039 @t{'Pos} and @t{'Val}.
15045 @t{'Range} on array objects (not subtypes), but only as the right
15046 operand of the membership (@code{in}) operator.
15049 @t{'Access}, @t{'Unchecked_Access}, and
15050 @t{'Unrestricted_Access} (a GNAT extension).
15058 @code{Characters.Latin_1} are not available and
15059 concatenation is not implemented. Thus, escape characters in strings are
15060 not currently available.
15063 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
15064 equality of representations. They will generally work correctly
15065 for strings and arrays whose elements have integer or enumeration types.
15066 They may not work correctly for arrays whose element
15067 types have user-defined equality, for arrays of real values
15068 (in particular, IEEE-conformant floating point, because of negative
15069 zeroes and NaNs), and for arrays whose elements contain unused bits with
15070 indeterminate values.
15073 The other component-by-component array operations (@code{and}, @code{or},
15074 @code{xor}, @code{not}, and relational tests other than equality)
15075 are not implemented.
15078 @cindex array aggregates (Ada)
15079 @cindex record aggregates (Ada)
15080 @cindex aggregates (Ada)
15081 There is limited support for array and record aggregates. They are
15082 permitted only on the right sides of assignments, as in these examples:
15085 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
15086 (@value{GDBP}) set An_Array := (1, others => 0)
15087 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
15088 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
15089 (@value{GDBP}) set A_Record := (1, "Peter", True);
15090 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
15094 discriminant's value by assigning an aggregate has an
15095 undefined effect if that discriminant is used within the record.
15096 However, you can first modify discriminants by directly assigning to
15097 them (which normally would not be allowed in Ada), and then performing an
15098 aggregate assignment. For example, given a variable @code{A_Rec}
15099 declared to have a type such as:
15102 type Rec (Len : Small_Integer := 0) is record
15104 Vals : IntArray (1 .. Len);
15108 you can assign a value with a different size of @code{Vals} with two
15112 (@value{GDBP}) set A_Rec.Len := 4
15113 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
15116 As this example also illustrates, @value{GDBN} is very loose about the usual
15117 rules concerning aggregates. You may leave out some of the
15118 components of an array or record aggregate (such as the @code{Len}
15119 component in the assignment to @code{A_Rec} above); they will retain their
15120 original values upon assignment. You may freely use dynamic values as
15121 indices in component associations. You may even use overlapping or
15122 redundant component associations, although which component values are
15123 assigned in such cases is not defined.
15126 Calls to dispatching subprograms are not implemented.
15129 The overloading algorithm is much more limited (i.e., less selective)
15130 than that of real Ada. It makes only limited use of the context in
15131 which a subexpression appears to resolve its meaning, and it is much
15132 looser in its rules for allowing type matches. As a result, some
15133 function calls will be ambiguous, and the user will be asked to choose
15134 the proper resolution.
15137 The @code{new} operator is not implemented.
15140 Entry calls are not implemented.
15143 Aside from printing, arithmetic operations on the native VAX floating-point
15144 formats are not supported.
15147 It is not possible to slice a packed array.
15150 The names @code{True} and @code{False}, when not part of a qualified name,
15151 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
15153 Should your program
15154 redefine these names in a package or procedure (at best a dubious practice),
15155 you will have to use fully qualified names to access their new definitions.
15158 @node Additions to Ada
15159 @subsubsection Additions to Ada
15160 @cindex Ada, deviations from
15162 As it does for other languages, @value{GDBN} makes certain generic
15163 extensions to Ada (@pxref{Expressions}):
15167 If the expression @var{E} is a variable residing in memory (typically
15168 a local variable or array element) and @var{N} is a positive integer,
15169 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
15170 @var{N}-1 adjacent variables following it in memory as an array. In
15171 Ada, this operator is generally not necessary, since its prime use is
15172 in displaying parts of an array, and slicing will usually do this in
15173 Ada. However, there are occasional uses when debugging programs in
15174 which certain debugging information has been optimized away.
15177 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
15178 appears in function or file @var{B}.'' When @var{B} is a file name,
15179 you must typically surround it in single quotes.
15182 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
15183 @var{type} that appears at address @var{addr}.''
15186 A name starting with @samp{$} is a convenience variable
15187 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
15190 In addition, @value{GDBN} provides a few other shortcuts and outright
15191 additions specific to Ada:
15195 The assignment statement is allowed as an expression, returning
15196 its right-hand operand as its value. Thus, you may enter
15199 (@value{GDBP}) set x := y + 3
15200 (@value{GDBP}) print A(tmp := y + 1)
15204 The semicolon is allowed as an ``operator,'' returning as its value
15205 the value of its right-hand operand.
15206 This allows, for example,
15207 complex conditional breaks:
15210 (@value{GDBP}) break f
15211 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
15215 Rather than use catenation and symbolic character names to introduce special
15216 characters into strings, one may instead use a special bracket notation,
15217 which is also used to print strings. A sequence of characters of the form
15218 @samp{["@var{XX}"]} within a string or character literal denotes the
15219 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
15220 sequence of characters @samp{["""]} also denotes a single quotation mark
15221 in strings. For example,
15223 "One line.["0a"]Next line.["0a"]"
15226 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
15230 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
15231 @t{'Max} is optional (and is ignored in any case). For example, it is valid
15235 (@value{GDBP}) print 'max(x, y)
15239 When printing arrays, @value{GDBN} uses positional notation when the
15240 array has a lower bound of 1, and uses a modified named notation otherwise.
15241 For example, a one-dimensional array of three integers with a lower bound
15242 of 3 might print as
15249 That is, in contrast to valid Ada, only the first component has a @code{=>}
15253 You may abbreviate attributes in expressions with any unique,
15254 multi-character subsequence of
15255 their names (an exact match gets preference).
15256 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
15257 in place of @t{a'length}.
15260 @cindex quoting Ada internal identifiers
15261 Since Ada is case-insensitive, the debugger normally maps identifiers you type
15262 to lower case. The GNAT compiler uses upper-case characters for
15263 some of its internal identifiers, which are normally of no interest to users.
15264 For the rare occasions when you actually have to look at them,
15265 enclose them in angle brackets to avoid the lower-case mapping.
15268 (@value{GDBP}) print <JMPBUF_SAVE>[0]
15272 Printing an object of class-wide type or dereferencing an
15273 access-to-class-wide value will display all the components of the object's
15274 specific type (as indicated by its run-time tag). Likewise, component
15275 selection on such a value will operate on the specific type of the
15280 @node Stopping Before Main Program
15281 @subsubsection Stopping at the Very Beginning
15283 @cindex breakpointing Ada elaboration code
15284 It is sometimes necessary to debug the program during elaboration, and
15285 before reaching the main procedure.
15286 As defined in the Ada Reference
15287 Manual, the elaboration code is invoked from a procedure called
15288 @code{adainit}. To run your program up to the beginning of
15289 elaboration, simply use the following two commands:
15290 @code{tbreak adainit} and @code{run}.
15293 @subsubsection Extensions for Ada Tasks
15294 @cindex Ada, tasking
15296 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
15297 @value{GDBN} provides the following task-related commands:
15302 This command shows a list of current Ada tasks, as in the following example:
15309 (@value{GDBP}) info tasks
15310 ID TID P-ID Pri State Name
15311 1 8088000 0 15 Child Activation Wait main_task
15312 2 80a4000 1 15 Accept Statement b
15313 3 809a800 1 15 Child Activation Wait a
15314 * 4 80ae800 3 15 Runnable c
15319 In this listing, the asterisk before the last task indicates it to be the
15320 task currently being inspected.
15324 Represents @value{GDBN}'s internal task number.
15330 The parent's task ID (@value{GDBN}'s internal task number).
15333 The base priority of the task.
15336 Current state of the task.
15340 The task has been created but has not been activated. It cannot be
15344 The task is not blocked for any reason known to Ada. (It may be waiting
15345 for a mutex, though.) It is conceptually "executing" in normal mode.
15348 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
15349 that were waiting on terminate alternatives have been awakened and have
15350 terminated themselves.
15352 @item Child Activation Wait
15353 The task is waiting for created tasks to complete activation.
15355 @item Accept Statement
15356 The task is waiting on an accept or selective wait statement.
15358 @item Waiting on entry call
15359 The task is waiting on an entry call.
15361 @item Async Select Wait
15362 The task is waiting to start the abortable part of an asynchronous
15366 The task is waiting on a select statement with only a delay
15369 @item Child Termination Wait
15370 The task is sleeping having completed a master within itself, and is
15371 waiting for the tasks dependent on that master to become terminated or
15372 waiting on a terminate Phase.
15374 @item Wait Child in Term Alt
15375 The task is sleeping waiting for tasks on terminate alternatives to
15376 finish terminating.
15378 @item Accepting RV with @var{taskno}
15379 The task is accepting a rendez-vous with the task @var{taskno}.
15383 Name of the task in the program.
15387 @kindex info task @var{taskno}
15388 @item info task @var{taskno}
15389 This command shows detailled informations on the specified task, as in
15390 the following example:
15395 (@value{GDBP}) info tasks
15396 ID TID P-ID Pri State Name
15397 1 8077880 0 15 Child Activation Wait main_task
15398 * 2 807c468 1 15 Runnable task_1
15399 (@value{GDBP}) info task 2
15400 Ada Task: 0x807c468
15403 Parent: 1 (main_task)
15409 @kindex task@r{ (Ada)}
15410 @cindex current Ada task ID
15411 This command prints the ID of the current task.
15417 (@value{GDBP}) info tasks
15418 ID TID P-ID Pri State Name
15419 1 8077870 0 15 Child Activation Wait main_task
15420 * 2 807c458 1 15 Runnable t
15421 (@value{GDBP}) task
15422 [Current task is 2]
15425 @item task @var{taskno}
15426 @cindex Ada task switching
15427 This command is like the @code{thread @var{threadno}}
15428 command (@pxref{Threads}). It switches the context of debugging
15429 from the current task to the given task.
15435 (@value{GDBP}) info tasks
15436 ID TID P-ID Pri State Name
15437 1 8077870 0 15 Child Activation Wait main_task
15438 * 2 807c458 1 15 Runnable t
15439 (@value{GDBP}) task 1
15440 [Switching to task 1]
15441 #0 0x8067726 in pthread_cond_wait ()
15443 #0 0x8067726 in pthread_cond_wait ()
15444 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
15445 #2 0x805cb63 in system.task_primitives.operations.sleep ()
15446 #3 0x806153e in system.tasking.stages.activate_tasks ()
15447 #4 0x804aacc in un () at un.adb:5
15450 @item break @var{linespec} task @var{taskno}
15451 @itemx break @var{linespec} task @var{taskno} if @dots{}
15452 @cindex breakpoints and tasks, in Ada
15453 @cindex task breakpoints, in Ada
15454 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
15455 These commands are like the @code{break @dots{} thread @dots{}}
15456 command (@pxref{Thread Stops}).
15457 @var{linespec} specifies source lines, as described
15458 in @ref{Specify Location}.
15460 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
15461 to specify that you only want @value{GDBN} to stop the program when a
15462 particular Ada task reaches this breakpoint. @var{taskno} is one of the
15463 numeric task identifiers assigned by @value{GDBN}, shown in the first
15464 column of the @samp{info tasks} display.
15466 If you do not specify @samp{task @var{taskno}} when you set a
15467 breakpoint, the breakpoint applies to @emph{all} tasks of your
15470 You can use the @code{task} qualifier on conditional breakpoints as
15471 well; in this case, place @samp{task @var{taskno}} before the
15472 breakpoint condition (before the @code{if}).
15480 (@value{GDBP}) info tasks
15481 ID TID P-ID Pri State Name
15482 1 140022020 0 15 Child Activation Wait main_task
15483 2 140045060 1 15 Accept/Select Wait t2
15484 3 140044840 1 15 Runnable t1
15485 * 4 140056040 1 15 Runnable t3
15486 (@value{GDBP}) b 15 task 2
15487 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
15488 (@value{GDBP}) cont
15493 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
15495 (@value{GDBP}) info tasks
15496 ID TID P-ID Pri State Name
15497 1 140022020 0 15 Child Activation Wait main_task
15498 * 2 140045060 1 15 Runnable t2
15499 3 140044840 1 15 Runnable t1
15500 4 140056040 1 15 Delay Sleep t3
15504 @node Ada Tasks and Core Files
15505 @subsubsection Tasking Support when Debugging Core Files
15506 @cindex Ada tasking and core file debugging
15508 When inspecting a core file, as opposed to debugging a live program,
15509 tasking support may be limited or even unavailable, depending on
15510 the platform being used.
15511 For instance, on x86-linux, the list of tasks is available, but task
15512 switching is not supported. On Tru64, however, task switching will work
15515 On certain platforms, including Tru64, the debugger needs to perform some
15516 memory writes in order to provide Ada tasking support. When inspecting
15517 a core file, this means that the core file must be opened with read-write
15518 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
15519 Under these circumstances, you should make a backup copy of the core
15520 file before inspecting it with @value{GDBN}.
15522 @node Ravenscar Profile
15523 @subsubsection Tasking Support when using the Ravenscar Profile
15524 @cindex Ravenscar Profile
15526 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
15527 specifically designed for systems with safety-critical real-time
15531 @kindex set ravenscar task-switching on
15532 @cindex task switching with program using Ravenscar Profile
15533 @item set ravenscar task-switching on
15534 Allows task switching when debugging a program that uses the Ravenscar
15535 Profile. This is the default.
15537 @kindex set ravenscar task-switching off
15538 @item set ravenscar task-switching off
15539 Turn off task switching when debugging a program that uses the Ravenscar
15540 Profile. This is mostly intended to disable the code that adds support
15541 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
15542 the Ravenscar runtime is preventing @value{GDBN} from working properly.
15543 To be effective, this command should be run before the program is started.
15545 @kindex show ravenscar task-switching
15546 @item show ravenscar task-switching
15547 Show whether it is possible to switch from task to task in a program
15548 using the Ravenscar Profile.
15553 @subsubsection Known Peculiarities of Ada Mode
15554 @cindex Ada, problems
15556 Besides the omissions listed previously (@pxref{Omissions from Ada}),
15557 we know of several problems with and limitations of Ada mode in
15559 some of which will be fixed with planned future releases of the debugger
15560 and the GNU Ada compiler.
15564 Static constants that the compiler chooses not to materialize as objects in
15565 storage are invisible to the debugger.
15568 Named parameter associations in function argument lists are ignored (the
15569 argument lists are treated as positional).
15572 Many useful library packages are currently invisible to the debugger.
15575 Fixed-point arithmetic, conversions, input, and output is carried out using
15576 floating-point arithmetic, and may give results that only approximate those on
15580 The GNAT compiler never generates the prefix @code{Standard} for any of
15581 the standard symbols defined by the Ada language. @value{GDBN} knows about
15582 this: it will strip the prefix from names when you use it, and will never
15583 look for a name you have so qualified among local symbols, nor match against
15584 symbols in other packages or subprograms. If you have
15585 defined entities anywhere in your program other than parameters and
15586 local variables whose simple names match names in @code{Standard},
15587 GNAT's lack of qualification here can cause confusion. When this happens,
15588 you can usually resolve the confusion
15589 by qualifying the problematic names with package
15590 @code{Standard} explicitly.
15593 Older versions of the compiler sometimes generate erroneous debugging
15594 information, resulting in the debugger incorrectly printing the value
15595 of affected entities. In some cases, the debugger is able to work
15596 around an issue automatically. In other cases, the debugger is able
15597 to work around the issue, but the work-around has to be specifically
15600 @kindex set ada trust-PAD-over-XVS
15601 @kindex show ada trust-PAD-over-XVS
15604 @item set ada trust-PAD-over-XVS on
15605 Configure GDB to strictly follow the GNAT encoding when computing the
15606 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
15607 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
15608 a complete description of the encoding used by the GNAT compiler).
15609 This is the default.
15611 @item set ada trust-PAD-over-XVS off
15612 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
15613 sometimes prints the wrong value for certain entities, changing @code{ada
15614 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
15615 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
15616 @code{off}, but this incurs a slight performance penalty, so it is
15617 recommended to leave this setting to @code{on} unless necessary.
15621 @node Unsupported Languages
15622 @section Unsupported Languages
15624 @cindex unsupported languages
15625 @cindex minimal language
15626 In addition to the other fully-supported programming languages,
15627 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
15628 It does not represent a real programming language, but provides a set
15629 of capabilities close to what the C or assembly languages provide.
15630 This should allow most simple operations to be performed while debugging
15631 an application that uses a language currently not supported by @value{GDBN}.
15633 If the language is set to @code{auto}, @value{GDBN} will automatically
15634 select this language if the current frame corresponds to an unsupported
15638 @chapter Examining the Symbol Table
15640 The commands described in this chapter allow you to inquire about the
15641 symbols (names of variables, functions and types) defined in your
15642 program. This information is inherent in the text of your program and
15643 does not change as your program executes. @value{GDBN} finds it in your
15644 program's symbol table, in the file indicated when you started @value{GDBN}
15645 (@pxref{File Options, ,Choosing Files}), or by one of the
15646 file-management commands (@pxref{Files, ,Commands to Specify Files}).
15648 @cindex symbol names
15649 @cindex names of symbols
15650 @cindex quoting names
15651 Occasionally, you may need to refer to symbols that contain unusual
15652 characters, which @value{GDBN} ordinarily treats as word delimiters. The
15653 most frequent case is in referring to static variables in other
15654 source files (@pxref{Variables,,Program Variables}). File names
15655 are recorded in object files as debugging symbols, but @value{GDBN} would
15656 ordinarily parse a typical file name, like @file{foo.c}, as the three words
15657 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
15658 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
15665 looks up the value of @code{x} in the scope of the file @file{foo.c}.
15668 @cindex case-insensitive symbol names
15669 @cindex case sensitivity in symbol names
15670 @kindex set case-sensitive
15671 @item set case-sensitive on
15672 @itemx set case-sensitive off
15673 @itemx set case-sensitive auto
15674 Normally, when @value{GDBN} looks up symbols, it matches their names
15675 with case sensitivity determined by the current source language.
15676 Occasionally, you may wish to control that. The command @code{set
15677 case-sensitive} lets you do that by specifying @code{on} for
15678 case-sensitive matches or @code{off} for case-insensitive ones. If
15679 you specify @code{auto}, case sensitivity is reset to the default
15680 suitable for the source language. The default is case-sensitive
15681 matches for all languages except for Fortran, for which the default is
15682 case-insensitive matches.
15684 @kindex show case-sensitive
15685 @item show case-sensitive
15686 This command shows the current setting of case sensitivity for symbols
15689 @kindex set print type methods
15690 @item set print type methods
15691 @itemx set print type methods on
15692 @itemx set print type methods off
15693 Normally, when @value{GDBN} prints a class, it displays any methods
15694 declared in that class. You can control this behavior either by
15695 passing the appropriate flag to @code{ptype}, or using @command{set
15696 print type methods}. Specifying @code{on} will cause @value{GDBN} to
15697 display the methods; this is the default. Specifying @code{off} will
15698 cause @value{GDBN} to omit the methods.
15700 @kindex show print type methods
15701 @item show print type methods
15702 This command shows the current setting of method display when printing
15705 @kindex set print type typedefs
15706 @item set print type typedefs
15707 @itemx set print type typedefs on
15708 @itemx set print type typedefs off
15710 Normally, when @value{GDBN} prints a class, it displays any typedefs
15711 defined in that class. You can control this behavior either by
15712 passing the appropriate flag to @code{ptype}, or using @command{set
15713 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
15714 display the typedef definitions; this is the default. Specifying
15715 @code{off} will cause @value{GDBN} to omit the typedef definitions.
15716 Note that this controls whether the typedef definition itself is
15717 printed, not whether typedef names are substituted when printing other
15720 @kindex show print type typedefs
15721 @item show print type typedefs
15722 This command shows the current setting of typedef display when
15725 @kindex info address
15726 @cindex address of a symbol
15727 @item info address @var{symbol}
15728 Describe where the data for @var{symbol} is stored. For a register
15729 variable, this says which register it is kept in. For a non-register
15730 local variable, this prints the stack-frame offset at which the variable
15733 Note the contrast with @samp{print &@var{symbol}}, which does not work
15734 at all for a register variable, and for a stack local variable prints
15735 the exact address of the current instantiation of the variable.
15737 @kindex info symbol
15738 @cindex symbol from address
15739 @cindex closest symbol and offset for an address
15740 @item info symbol @var{addr}
15741 Print the name of a symbol which is stored at the address @var{addr}.
15742 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
15743 nearest symbol and an offset from it:
15746 (@value{GDBP}) info symbol 0x54320
15747 _initialize_vx + 396 in section .text
15751 This is the opposite of the @code{info address} command. You can use
15752 it to find out the name of a variable or a function given its address.
15754 For dynamically linked executables, the name of executable or shared
15755 library containing the symbol is also printed:
15758 (@value{GDBP}) info symbol 0x400225
15759 _start + 5 in section .text of /tmp/a.out
15760 (@value{GDBP}) info symbol 0x2aaaac2811cf
15761 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
15765 @item whatis[/@var{flags}] [@var{arg}]
15766 Print the data type of @var{arg}, which can be either an expression
15767 or a name of a data type. With no argument, print the data type of
15768 @code{$}, the last value in the value history.
15770 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
15771 is not actually evaluated, and any side-effecting operations (such as
15772 assignments or function calls) inside it do not take place.
15774 If @var{arg} is a variable or an expression, @code{whatis} prints its
15775 literal type as it is used in the source code. If the type was
15776 defined using a @code{typedef}, @code{whatis} will @emph{not} print
15777 the data type underlying the @code{typedef}. If the type of the
15778 variable or the expression is a compound data type, such as
15779 @code{struct} or @code{class}, @code{whatis} never prints their
15780 fields or methods. It just prints the @code{struct}/@code{class}
15781 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
15782 such a compound data type, use @code{ptype}.
15784 If @var{arg} is a type name that was defined using @code{typedef},
15785 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
15786 Unrolling means that @code{whatis} will show the underlying type used
15787 in the @code{typedef} declaration of @var{arg}. However, if that
15788 underlying type is also a @code{typedef}, @code{whatis} will not
15791 For C code, the type names may also have the form @samp{class
15792 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
15793 @var{union-tag}} or @samp{enum @var{enum-tag}}.
15795 @var{flags} can be used to modify how the type is displayed.
15796 Available flags are:
15800 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
15801 parameters and typedefs defined in a class when printing the class'
15802 members. The @code{/r} flag disables this.
15805 Do not print methods defined in the class.
15808 Print methods defined in the class. This is the default, but the flag
15809 exists in case you change the default with @command{set print type methods}.
15812 Do not print typedefs defined in the class. Note that this controls
15813 whether the typedef definition itself is printed, not whether typedef
15814 names are substituted when printing other types.
15817 Print typedefs defined in the class. This is the default, but the flag
15818 exists in case you change the default with @command{set print type typedefs}.
15822 @item ptype[/@var{flags}] [@var{arg}]
15823 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
15824 detailed description of the type, instead of just the name of the type.
15825 @xref{Expressions, ,Expressions}.
15827 Contrary to @code{whatis}, @code{ptype} always unrolls any
15828 @code{typedef}s in its argument declaration, whether the argument is
15829 a variable, expression, or a data type. This means that @code{ptype}
15830 of a variable or an expression will not print literally its type as
15831 present in the source code---use @code{whatis} for that. @code{typedef}s at
15832 the pointer or reference targets are also unrolled. Only @code{typedef}s of
15833 fields, methods and inner @code{class typedef}s of @code{struct}s,
15834 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
15836 For example, for this variable declaration:
15839 typedef double real_t;
15840 struct complex @{ real_t real; double imag; @};
15841 typedef struct complex complex_t;
15843 real_t *real_pointer_var;
15847 the two commands give this output:
15851 (@value{GDBP}) whatis var
15853 (@value{GDBP}) ptype var
15854 type = struct complex @{
15858 (@value{GDBP}) whatis complex_t
15859 type = struct complex
15860 (@value{GDBP}) whatis struct complex
15861 type = struct complex
15862 (@value{GDBP}) ptype struct complex
15863 type = struct complex @{
15867 (@value{GDBP}) whatis real_pointer_var
15869 (@value{GDBP}) ptype real_pointer_var
15875 As with @code{whatis}, using @code{ptype} without an argument refers to
15876 the type of @code{$}, the last value in the value history.
15878 @cindex incomplete type
15879 Sometimes, programs use opaque data types or incomplete specifications
15880 of complex data structure. If the debug information included in the
15881 program does not allow @value{GDBN} to display a full declaration of
15882 the data type, it will say @samp{<incomplete type>}. For example,
15883 given these declarations:
15887 struct foo *fooptr;
15891 but no definition for @code{struct foo} itself, @value{GDBN} will say:
15894 (@value{GDBP}) ptype foo
15895 $1 = <incomplete type>
15899 ``Incomplete type'' is C terminology for data types that are not
15900 completely specified.
15903 @item info types @var{regexp}
15905 Print a brief description of all types whose names match the regular
15906 expression @var{regexp} (or all types in your program, if you supply
15907 no argument). Each complete typename is matched as though it were a
15908 complete line; thus, @samp{i type value} gives information on all
15909 types in your program whose names include the string @code{value}, but
15910 @samp{i type ^value$} gives information only on types whose complete
15911 name is @code{value}.
15913 This command differs from @code{ptype} in two ways: first, like
15914 @code{whatis}, it does not print a detailed description; second, it
15915 lists all source files where a type is defined.
15917 @kindex info type-printers
15918 @item info type-printers
15919 Versions of @value{GDBN} that ship with Python scripting enabled may
15920 have ``type printers'' available. When using @command{ptype} or
15921 @command{whatis}, these printers are consulted when the name of a type
15922 is needed. @xref{Type Printing API}, for more information on writing
15925 @code{info type-printers} displays all the available type printers.
15927 @kindex enable type-printer
15928 @kindex disable type-printer
15929 @item enable type-printer @var{name}@dots{}
15930 @item disable type-printer @var{name}@dots{}
15931 These commands can be used to enable or disable type printers.
15934 @cindex local variables
15935 @item info scope @var{location}
15936 List all the variables local to a particular scope. This command
15937 accepts a @var{location} argument---a function name, a source line, or
15938 an address preceded by a @samp{*}, and prints all the variables local
15939 to the scope defined by that location. (@xref{Specify Location}, for
15940 details about supported forms of @var{location}.) For example:
15943 (@value{GDBP}) @b{info scope command_line_handler}
15944 Scope for command_line_handler:
15945 Symbol rl is an argument at stack/frame offset 8, length 4.
15946 Symbol linebuffer is in static storage at address 0x150a18, length 4.
15947 Symbol linelength is in static storage at address 0x150a1c, length 4.
15948 Symbol p is a local variable in register $esi, length 4.
15949 Symbol p1 is a local variable in register $ebx, length 4.
15950 Symbol nline is a local variable in register $edx, length 4.
15951 Symbol repeat is a local variable at frame offset -8, length 4.
15955 This command is especially useful for determining what data to collect
15956 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
15959 @kindex info source
15961 Show information about the current source file---that is, the source file for
15962 the function containing the current point of execution:
15965 the name of the source file, and the directory containing it,
15967 the directory it was compiled in,
15969 its length, in lines,
15971 which programming language it is written in,
15973 whether the executable includes debugging information for that file, and
15974 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
15976 whether the debugging information includes information about
15977 preprocessor macros.
15981 @kindex info sources
15983 Print the names of all source files in your program for which there is
15984 debugging information, organized into two lists: files whose symbols
15985 have already been read, and files whose symbols will be read when needed.
15987 @kindex info functions
15988 @item info functions
15989 Print the names and data types of all defined functions.
15991 @item info functions @var{regexp}
15992 Print the names and data types of all defined functions
15993 whose names contain a match for regular expression @var{regexp}.
15994 Thus, @samp{info fun step} finds all functions whose names
15995 include @code{step}; @samp{info fun ^step} finds those whose names
15996 start with @code{step}. If a function name contains characters
15997 that conflict with the regular expression language (e.g.@:
15998 @samp{operator*()}), they may be quoted with a backslash.
16000 @kindex info variables
16001 @item info variables
16002 Print the names and data types of all variables that are defined
16003 outside of functions (i.e.@: excluding local variables).
16005 @item info variables @var{regexp}
16006 Print the names and data types of all variables (except for local
16007 variables) whose names contain a match for regular expression
16010 @kindex info classes
16011 @cindex Objective-C, classes and selectors
16013 @itemx info classes @var{regexp}
16014 Display all Objective-C classes in your program, or
16015 (with the @var{regexp} argument) all those matching a particular regular
16018 @kindex info selectors
16019 @item info selectors
16020 @itemx info selectors @var{regexp}
16021 Display all Objective-C selectors in your program, or
16022 (with the @var{regexp} argument) all those matching a particular regular
16026 This was never implemented.
16027 @kindex info methods
16029 @itemx info methods @var{regexp}
16030 The @code{info methods} command permits the user to examine all defined
16031 methods within C@t{++} program, or (with the @var{regexp} argument) a
16032 specific set of methods found in the various C@t{++} classes. Many
16033 C@t{++} classes provide a large number of methods. Thus, the output
16034 from the @code{ptype} command can be overwhelming and hard to use. The
16035 @code{info-methods} command filters the methods, printing only those
16036 which match the regular-expression @var{regexp}.
16039 @cindex opaque data types
16040 @kindex set opaque-type-resolution
16041 @item set opaque-type-resolution on
16042 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
16043 declared as a pointer to a @code{struct}, @code{class}, or
16044 @code{union}---for example, @code{struct MyType *}---that is used in one
16045 source file although the full declaration of @code{struct MyType} is in
16046 another source file. The default is on.
16048 A change in the setting of this subcommand will not take effect until
16049 the next time symbols for a file are loaded.
16051 @item set opaque-type-resolution off
16052 Tell @value{GDBN} not to resolve opaque types. In this case, the type
16053 is printed as follows:
16055 @{<no data fields>@}
16058 @kindex show opaque-type-resolution
16059 @item show opaque-type-resolution
16060 Show whether opaque types are resolved or not.
16062 @kindex maint print symbols
16063 @cindex symbol dump
16064 @kindex maint print psymbols
16065 @cindex partial symbol dump
16066 @kindex maint print msymbols
16067 @cindex minimal symbol dump
16068 @item maint print symbols @var{filename}
16069 @itemx maint print psymbols @var{filename}
16070 @itemx maint print msymbols @var{filename}
16071 Write a dump of debugging symbol data into the file @var{filename}.
16072 These commands are used to debug the @value{GDBN} symbol-reading code. Only
16073 symbols with debugging data are included. If you use @samp{maint print
16074 symbols}, @value{GDBN} includes all the symbols for which it has already
16075 collected full details: that is, @var{filename} reflects symbols for
16076 only those files whose symbols @value{GDBN} has read. You can use the
16077 command @code{info sources} to find out which files these are. If you
16078 use @samp{maint print psymbols} instead, the dump shows information about
16079 symbols that @value{GDBN} only knows partially---that is, symbols defined in
16080 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
16081 @samp{maint print msymbols} dumps just the minimal symbol information
16082 required for each object file from which @value{GDBN} has read some symbols.
16083 @xref{Files, ,Commands to Specify Files}, for a discussion of how
16084 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
16086 @kindex maint info symtabs
16087 @kindex maint info psymtabs
16088 @cindex listing @value{GDBN}'s internal symbol tables
16089 @cindex symbol tables, listing @value{GDBN}'s internal
16090 @cindex full symbol tables, listing @value{GDBN}'s internal
16091 @cindex partial symbol tables, listing @value{GDBN}'s internal
16092 @item maint info symtabs @r{[} @var{regexp} @r{]}
16093 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
16095 List the @code{struct symtab} or @code{struct partial_symtab}
16096 structures whose names match @var{regexp}. If @var{regexp} is not
16097 given, list them all. The output includes expressions which you can
16098 copy into a @value{GDBN} debugging this one to examine a particular
16099 structure in more detail. For example:
16102 (@value{GDBP}) maint info psymtabs dwarf2read
16103 @{ objfile /home/gnu/build/gdb/gdb
16104 ((struct objfile *) 0x82e69d0)
16105 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
16106 ((struct partial_symtab *) 0x8474b10)
16109 text addresses 0x814d3c8 -- 0x8158074
16110 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
16111 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
16112 dependencies (none)
16115 (@value{GDBP}) maint info symtabs
16119 We see that there is one partial symbol table whose filename contains
16120 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
16121 and we see that @value{GDBN} has not read in any symtabs yet at all.
16122 If we set a breakpoint on a function, that will cause @value{GDBN} to
16123 read the symtab for the compilation unit containing that function:
16126 (@value{GDBP}) break dwarf2_psymtab_to_symtab
16127 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
16129 (@value{GDBP}) maint info symtabs
16130 @{ objfile /home/gnu/build/gdb/gdb
16131 ((struct objfile *) 0x82e69d0)
16132 @{ symtab /home/gnu/src/gdb/dwarf2read.c
16133 ((struct symtab *) 0x86c1f38)
16136 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
16137 linetable ((struct linetable *) 0x8370fa0)
16138 debugformat DWARF 2
16147 @chapter Altering Execution
16149 Once you think you have found an error in your program, you might want to
16150 find out for certain whether correcting the apparent error would lead to
16151 correct results in the rest of the run. You can find the answer by
16152 experiment, using the @value{GDBN} features for altering execution of the
16155 For example, you can store new values into variables or memory
16156 locations, give your program a signal, restart it at a different
16157 address, or even return prematurely from a function.
16160 * Assignment:: Assignment to variables
16161 * Jumping:: Continuing at a different address
16162 * Signaling:: Giving your program a signal
16163 * Returning:: Returning from a function
16164 * Calling:: Calling your program's functions
16165 * Patching:: Patching your program
16169 @section Assignment to Variables
16172 @cindex setting variables
16173 To alter the value of a variable, evaluate an assignment expression.
16174 @xref{Expressions, ,Expressions}. For example,
16181 stores the value 4 into the variable @code{x}, and then prints the
16182 value of the assignment expression (which is 4).
16183 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
16184 information on operators in supported languages.
16186 @kindex set variable
16187 @cindex variables, setting
16188 If you are not interested in seeing the value of the assignment, use the
16189 @code{set} command instead of the @code{print} command. @code{set} is
16190 really the same as @code{print} except that the expression's value is
16191 not printed and is not put in the value history (@pxref{Value History,
16192 ,Value History}). The expression is evaluated only for its effects.
16194 If the beginning of the argument string of the @code{set} command
16195 appears identical to a @code{set} subcommand, use the @code{set
16196 variable} command instead of just @code{set}. This command is identical
16197 to @code{set} except for its lack of subcommands. For example, if your
16198 program has a variable @code{width}, you get an error if you try to set
16199 a new value with just @samp{set width=13}, because @value{GDBN} has the
16200 command @code{set width}:
16203 (@value{GDBP}) whatis width
16205 (@value{GDBP}) p width
16207 (@value{GDBP}) set width=47
16208 Invalid syntax in expression.
16212 The invalid expression, of course, is @samp{=47}. In
16213 order to actually set the program's variable @code{width}, use
16216 (@value{GDBP}) set var width=47
16219 Because the @code{set} command has many subcommands that can conflict
16220 with the names of program variables, it is a good idea to use the
16221 @code{set variable} command instead of just @code{set}. For example, if
16222 your program has a variable @code{g}, you run into problems if you try
16223 to set a new value with just @samp{set g=4}, because @value{GDBN} has
16224 the command @code{set gnutarget}, abbreviated @code{set g}:
16228 (@value{GDBP}) whatis g
16232 (@value{GDBP}) set g=4
16236 The program being debugged has been started already.
16237 Start it from the beginning? (y or n) y
16238 Starting program: /home/smith/cc_progs/a.out
16239 "/home/smith/cc_progs/a.out": can't open to read symbols:
16240 Invalid bfd target.
16241 (@value{GDBP}) show g
16242 The current BFD target is "=4".
16247 The program variable @code{g} did not change, and you silently set the
16248 @code{gnutarget} to an invalid value. In order to set the variable
16252 (@value{GDBP}) set var g=4
16255 @value{GDBN} allows more implicit conversions in assignments than C; you can
16256 freely store an integer value into a pointer variable or vice versa,
16257 and you can convert any structure to any other structure that is the
16258 same length or shorter.
16259 @comment FIXME: how do structs align/pad in these conversions?
16260 @comment /doc@cygnus.com 18dec1990
16262 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
16263 construct to generate a value of specified type at a specified address
16264 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
16265 to memory location @code{0x83040} as an integer (which implies a certain size
16266 and representation in memory), and
16269 set @{int@}0x83040 = 4
16273 stores the value 4 into that memory location.
16276 @section Continuing at a Different Address
16278 Ordinarily, when you continue your program, you do so at the place where
16279 it stopped, with the @code{continue} command. You can instead continue at
16280 an address of your own choosing, with the following commands:
16284 @kindex j @r{(@code{jump})}
16285 @item jump @var{linespec}
16286 @itemx j @var{linespec}
16287 @itemx jump @var{location}
16288 @itemx j @var{location}
16289 Resume execution at line @var{linespec} or at address given by
16290 @var{location}. Execution stops again immediately if there is a
16291 breakpoint there. @xref{Specify Location}, for a description of the
16292 different forms of @var{linespec} and @var{location}. It is common
16293 practice to use the @code{tbreak} command in conjunction with
16294 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
16296 The @code{jump} command does not change the current stack frame, or
16297 the stack pointer, or the contents of any memory location or any
16298 register other than the program counter. If line @var{linespec} is in
16299 a different function from the one currently executing, the results may
16300 be bizarre if the two functions expect different patterns of arguments or
16301 of local variables. For this reason, the @code{jump} command requests
16302 confirmation if the specified line is not in the function currently
16303 executing. However, even bizarre results are predictable if you are
16304 well acquainted with the machine-language code of your program.
16307 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
16308 On many systems, you can get much the same effect as the @code{jump}
16309 command by storing a new value into the register @code{$pc}. The
16310 difference is that this does not start your program running; it only
16311 changes the address of where it @emph{will} run when you continue. For
16319 makes the next @code{continue} command or stepping command execute at
16320 address @code{0x485}, rather than at the address where your program stopped.
16321 @xref{Continuing and Stepping, ,Continuing and Stepping}.
16323 The most common occasion to use the @code{jump} command is to back
16324 up---perhaps with more breakpoints set---over a portion of a program
16325 that has already executed, in order to examine its execution in more
16330 @section Giving your Program a Signal
16331 @cindex deliver a signal to a program
16335 @item signal @var{signal}
16336 Resume execution where your program stopped, but immediately give it the
16337 signal @var{signal}. @var{signal} can be the name or the number of a
16338 signal. For example, on many systems @code{signal 2} and @code{signal
16339 SIGINT} are both ways of sending an interrupt signal.
16341 Alternatively, if @var{signal} is zero, continue execution without
16342 giving a signal. This is useful when your program stopped on account of
16343 a signal and would ordinarily see the signal when resumed with the
16344 @code{continue} command; @samp{signal 0} causes it to resume without a
16347 @code{signal} does not repeat when you press @key{RET} a second time
16348 after executing the command.
16352 Invoking the @code{signal} command is not the same as invoking the
16353 @code{kill} utility from the shell. Sending a signal with @code{kill}
16354 causes @value{GDBN} to decide what to do with the signal depending on
16355 the signal handling tables (@pxref{Signals}). The @code{signal} command
16356 passes the signal directly to your program.
16360 @section Returning from a Function
16363 @cindex returning from a function
16366 @itemx return @var{expression}
16367 You can cancel execution of a function call with the @code{return}
16368 command. If you give an
16369 @var{expression} argument, its value is used as the function's return
16373 When you use @code{return}, @value{GDBN} discards the selected stack frame
16374 (and all frames within it). You can think of this as making the
16375 discarded frame return prematurely. If you wish to specify a value to
16376 be returned, give that value as the argument to @code{return}.
16378 This pops the selected stack frame (@pxref{Selection, ,Selecting a
16379 Frame}), and any other frames inside of it, leaving its caller as the
16380 innermost remaining frame. That frame becomes selected. The
16381 specified value is stored in the registers used for returning values
16384 The @code{return} command does not resume execution; it leaves the
16385 program stopped in the state that would exist if the function had just
16386 returned. In contrast, the @code{finish} command (@pxref{Continuing
16387 and Stepping, ,Continuing and Stepping}) resumes execution until the
16388 selected stack frame returns naturally.
16390 @value{GDBN} needs to know how the @var{expression} argument should be set for
16391 the inferior. The concrete registers assignment depends on the OS ABI and the
16392 type being returned by the selected stack frame. For example it is common for
16393 OS ABI to return floating point values in FPU registers while integer values in
16394 CPU registers. Still some ABIs return even floating point values in CPU
16395 registers. Larger integer widths (such as @code{long long int}) also have
16396 specific placement rules. @value{GDBN} already knows the OS ABI from its
16397 current target so it needs to find out also the type being returned to make the
16398 assignment into the right register(s).
16400 Normally, the selected stack frame has debug info. @value{GDBN} will always
16401 use the debug info instead of the implicit type of @var{expression} when the
16402 debug info is available. For example, if you type @kbd{return -1}, and the
16403 function in the current stack frame is declared to return a @code{long long
16404 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
16405 into a @code{long long int}:
16408 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
16410 (@value{GDBP}) return -1
16411 Make func return now? (y or n) y
16412 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
16413 43 printf ("result=%lld\n", func ());
16417 However, if the selected stack frame does not have a debug info, e.g., if the
16418 function was compiled without debug info, @value{GDBN} has to find out the type
16419 to return from user. Specifying a different type by mistake may set the value
16420 in different inferior registers than the caller code expects. For example,
16421 typing @kbd{return -1} with its implicit type @code{int} would set only a part
16422 of a @code{long long int} result for a debug info less function (on 32-bit
16423 architectures). Therefore the user is required to specify the return type by
16424 an appropriate cast explicitly:
16427 Breakpoint 2, 0x0040050b in func ()
16428 (@value{GDBP}) return -1
16429 Return value type not available for selected stack frame.
16430 Please use an explicit cast of the value to return.
16431 (@value{GDBP}) return (long long int) -1
16432 Make selected stack frame return now? (y or n) y
16433 #0 0x00400526 in main ()
16438 @section Calling Program Functions
16441 @cindex calling functions
16442 @cindex inferior functions, calling
16443 @item print @var{expr}
16444 Evaluate the expression @var{expr} and display the resulting value.
16445 @var{expr} may include calls to functions in the program being
16449 @item call @var{expr}
16450 Evaluate the expression @var{expr} without displaying @code{void}
16453 You can use this variant of the @code{print} command if you want to
16454 execute a function from your program that does not return anything
16455 (a.k.a.@: @dfn{a void function}), but without cluttering the output
16456 with @code{void} returned values that @value{GDBN} will otherwise
16457 print. If the result is not void, it is printed and saved in the
16461 It is possible for the function you call via the @code{print} or
16462 @code{call} command to generate a signal (e.g., if there's a bug in
16463 the function, or if you passed it incorrect arguments). What happens
16464 in that case is controlled by the @code{set unwindonsignal} command.
16466 Similarly, with a C@t{++} program it is possible for the function you
16467 call via the @code{print} or @code{call} command to generate an
16468 exception that is not handled due to the constraints of the dummy
16469 frame. In this case, any exception that is raised in the frame, but has
16470 an out-of-frame exception handler will not be found. GDB builds a
16471 dummy-frame for the inferior function call, and the unwinder cannot
16472 seek for exception handlers outside of this dummy-frame. What happens
16473 in that case is controlled by the
16474 @code{set unwind-on-terminating-exception} command.
16477 @item set unwindonsignal
16478 @kindex set unwindonsignal
16479 @cindex unwind stack in called functions
16480 @cindex call dummy stack unwinding
16481 Set unwinding of the stack if a signal is received while in a function
16482 that @value{GDBN} called in the program being debugged. If set to on,
16483 @value{GDBN} unwinds the stack it created for the call and restores
16484 the context to what it was before the call. If set to off (the
16485 default), @value{GDBN} stops in the frame where the signal was
16488 @item show unwindonsignal
16489 @kindex show unwindonsignal
16490 Show the current setting of stack unwinding in the functions called by
16493 @item set unwind-on-terminating-exception
16494 @kindex set unwind-on-terminating-exception
16495 @cindex unwind stack in called functions with unhandled exceptions
16496 @cindex call dummy stack unwinding on unhandled exception.
16497 Set unwinding of the stack if a C@t{++} exception is raised, but left
16498 unhandled while in a function that @value{GDBN} called in the program being
16499 debugged. If set to on (the default), @value{GDBN} unwinds the stack
16500 it created for the call and restores the context to what it was before
16501 the call. If set to off, @value{GDBN} the exception is delivered to
16502 the default C@t{++} exception handler and the inferior terminated.
16504 @item show unwind-on-terminating-exception
16505 @kindex show unwind-on-terminating-exception
16506 Show the current setting of stack unwinding in the functions called by
16511 @cindex weak alias functions
16512 Sometimes, a function you wish to call is actually a @dfn{weak alias}
16513 for another function. In such case, @value{GDBN} might not pick up
16514 the type information, including the types of the function arguments,
16515 which causes @value{GDBN} to call the inferior function incorrectly.
16516 As a result, the called function will function erroneously and may
16517 even crash. A solution to that is to use the name of the aliased
16521 @section Patching Programs
16523 @cindex patching binaries
16524 @cindex writing into executables
16525 @cindex writing into corefiles
16527 By default, @value{GDBN} opens the file containing your program's
16528 executable code (or the corefile) read-only. This prevents accidental
16529 alterations to machine code; but it also prevents you from intentionally
16530 patching your program's binary.
16532 If you'd like to be able to patch the binary, you can specify that
16533 explicitly with the @code{set write} command. For example, you might
16534 want to turn on internal debugging flags, or even to make emergency
16540 @itemx set write off
16541 If you specify @samp{set write on}, @value{GDBN} opens executable and
16542 core files for both reading and writing; if you specify @kbd{set write
16543 off} (the default), @value{GDBN} opens them read-only.
16545 If you have already loaded a file, you must load it again (using the
16546 @code{exec-file} or @code{core-file} command) after changing @code{set
16547 write}, for your new setting to take effect.
16551 Display whether executable files and core files are opened for writing
16552 as well as reading.
16556 @chapter @value{GDBN} Files
16558 @value{GDBN} needs to know the file name of the program to be debugged,
16559 both in order to read its symbol table and in order to start your
16560 program. To debug a core dump of a previous run, you must also tell
16561 @value{GDBN} the name of the core dump file.
16564 * Files:: Commands to specify files
16565 * Separate Debug Files:: Debugging information in separate files
16566 * MiniDebugInfo:: Debugging information in a special section
16567 * Index Files:: Index files speed up GDB
16568 * Symbol Errors:: Errors reading symbol files
16569 * Data Files:: GDB data files
16573 @section Commands to Specify Files
16575 @cindex symbol table
16576 @cindex core dump file
16578 You may want to specify executable and core dump file names. The usual
16579 way to do this is at start-up time, using the arguments to
16580 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
16581 Out of @value{GDBN}}).
16583 Occasionally it is necessary to change to a different file during a
16584 @value{GDBN} session. Or you may run @value{GDBN} and forget to
16585 specify a file you want to use. Or you are debugging a remote target
16586 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
16587 Program}). In these situations the @value{GDBN} commands to specify
16588 new files are useful.
16591 @cindex executable file
16593 @item file @var{filename}
16594 Use @var{filename} as the program to be debugged. It is read for its
16595 symbols and for the contents of pure memory. It is also the program
16596 executed when you use the @code{run} command. If you do not specify a
16597 directory and the file is not found in the @value{GDBN} working directory,
16598 @value{GDBN} uses the environment variable @code{PATH} as a list of
16599 directories to search, just as the shell does when looking for a program
16600 to run. You can change the value of this variable, for both @value{GDBN}
16601 and your program, using the @code{path} command.
16603 @cindex unlinked object files
16604 @cindex patching object files
16605 You can load unlinked object @file{.o} files into @value{GDBN} using
16606 the @code{file} command. You will not be able to ``run'' an object
16607 file, but you can disassemble functions and inspect variables. Also,
16608 if the underlying BFD functionality supports it, you could use
16609 @kbd{gdb -write} to patch object files using this technique. Note
16610 that @value{GDBN} can neither interpret nor modify relocations in this
16611 case, so branches and some initialized variables will appear to go to
16612 the wrong place. But this feature is still handy from time to time.
16615 @code{file} with no argument makes @value{GDBN} discard any information it
16616 has on both executable file and the symbol table.
16619 @item exec-file @r{[} @var{filename} @r{]}
16620 Specify that the program to be run (but not the symbol table) is found
16621 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
16622 if necessary to locate your program. Omitting @var{filename} means to
16623 discard information on the executable file.
16625 @kindex symbol-file
16626 @item symbol-file @r{[} @var{filename} @r{]}
16627 Read symbol table information from file @var{filename}. @code{PATH} is
16628 searched when necessary. Use the @code{file} command to get both symbol
16629 table and program to run from the same file.
16631 @code{symbol-file} with no argument clears out @value{GDBN} information on your
16632 program's symbol table.
16634 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
16635 some breakpoints and auto-display expressions. This is because they may
16636 contain pointers to the internal data recording symbols and data types,
16637 which are part of the old symbol table data being discarded inside
16640 @code{symbol-file} does not repeat if you press @key{RET} again after
16643 When @value{GDBN} is configured for a particular environment, it
16644 understands debugging information in whatever format is the standard
16645 generated for that environment; you may use either a @sc{gnu} compiler, or
16646 other compilers that adhere to the local conventions.
16647 Best results are usually obtained from @sc{gnu} compilers; for example,
16648 using @code{@value{NGCC}} you can generate debugging information for
16651 For most kinds of object files, with the exception of old SVR3 systems
16652 using COFF, the @code{symbol-file} command does not normally read the
16653 symbol table in full right away. Instead, it scans the symbol table
16654 quickly to find which source files and which symbols are present. The
16655 details are read later, one source file at a time, as they are needed.
16657 The purpose of this two-stage reading strategy is to make @value{GDBN}
16658 start up faster. For the most part, it is invisible except for
16659 occasional pauses while the symbol table details for a particular source
16660 file are being read. (The @code{set verbose} command can turn these
16661 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
16662 Warnings and Messages}.)
16664 We have not implemented the two-stage strategy for COFF yet. When the
16665 symbol table is stored in COFF format, @code{symbol-file} reads the
16666 symbol table data in full right away. Note that ``stabs-in-COFF''
16667 still does the two-stage strategy, since the debug info is actually
16671 @cindex reading symbols immediately
16672 @cindex symbols, reading immediately
16673 @item symbol-file @r{[} -readnow @r{]} @var{filename}
16674 @itemx file @r{[} -readnow @r{]} @var{filename}
16675 You can override the @value{GDBN} two-stage strategy for reading symbol
16676 tables by using the @samp{-readnow} option with any of the commands that
16677 load symbol table information, if you want to be sure @value{GDBN} has the
16678 entire symbol table available.
16680 @c FIXME: for now no mention of directories, since this seems to be in
16681 @c flux. 13mar1992 status is that in theory GDB would look either in
16682 @c current dir or in same dir as myprog; but issues like competing
16683 @c GDB's, or clutter in system dirs, mean that in practice right now
16684 @c only current dir is used. FFish says maybe a special GDB hierarchy
16685 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
16689 @item core-file @r{[}@var{filename}@r{]}
16691 Specify the whereabouts of a core dump file to be used as the ``contents
16692 of memory''. Traditionally, core files contain only some parts of the
16693 address space of the process that generated them; @value{GDBN} can access the
16694 executable file itself for other parts.
16696 @code{core-file} with no argument specifies that no core file is
16699 Note that the core file is ignored when your program is actually running
16700 under @value{GDBN}. So, if you have been running your program and you
16701 wish to debug a core file instead, you must kill the subprocess in which
16702 the program is running. To do this, use the @code{kill} command
16703 (@pxref{Kill Process, ,Killing the Child Process}).
16705 @kindex add-symbol-file
16706 @cindex dynamic linking
16707 @item add-symbol-file @var{filename} @var{address}
16708 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
16709 @itemx add-symbol-file @var{filename} @var{address} -s @var{section} @var{address} @dots{}
16710 The @code{add-symbol-file} command reads additional symbol table
16711 information from the file @var{filename}. You would use this command
16712 when @var{filename} has been dynamically loaded (by some other means)
16713 into the program that is running. @var{address} should be the memory
16714 address at which the file has been loaded; @value{GDBN} cannot figure
16715 this out for itself. You can additionally specify an arbitrary number
16716 of @samp{-s @var{section} @var{address}} pairs, to give an explicit
16717 section name and base address for that section. You can specify any
16718 @var{address} as an expression.
16720 The symbol table of the file @var{filename} is added to the symbol table
16721 originally read with the @code{symbol-file} command. You can use the
16722 @code{add-symbol-file} command any number of times; the new symbol data
16723 thus read is kept in addition to the old.
16725 Changes can be reverted using the command @code{remove-symbol-file}.
16727 @cindex relocatable object files, reading symbols from
16728 @cindex object files, relocatable, reading symbols from
16729 @cindex reading symbols from relocatable object files
16730 @cindex symbols, reading from relocatable object files
16731 @cindex @file{.o} files, reading symbols from
16732 Although @var{filename} is typically a shared library file, an
16733 executable file, or some other object file which has been fully
16734 relocated for loading into a process, you can also load symbolic
16735 information from relocatable @file{.o} files, as long as:
16739 the file's symbolic information refers only to linker symbols defined in
16740 that file, not to symbols defined by other object files,
16742 every section the file's symbolic information refers to has actually
16743 been loaded into the inferior, as it appears in the file, and
16745 you can determine the address at which every section was loaded, and
16746 provide these to the @code{add-symbol-file} command.
16750 Some embedded operating systems, like Sun Chorus and VxWorks, can load
16751 relocatable files into an already running program; such systems
16752 typically make the requirements above easy to meet. However, it's
16753 important to recognize that many native systems use complex link
16754 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
16755 assembly, for example) that make the requirements difficult to meet. In
16756 general, one cannot assume that using @code{add-symbol-file} to read a
16757 relocatable object file's symbolic information will have the same effect
16758 as linking the relocatable object file into the program in the normal
16761 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
16763 @kindex remove-symbol-file
16764 @item remove-symbol-file @var{filename}
16765 @item remove-symbol-file -a @var{address}
16766 Remove a symbol file added via the @code{add-symbol-file} command. The
16767 file to remove can be identified by its @var{filename} or by an @var{address}
16768 that lies within the boundaries of this symbol file in memory. Example:
16771 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
16772 add symbol table from file "/home/user/gdb/mylib.so" at
16773 .text_addr = 0x7ffff7ff9480
16775 Reading symbols from /home/user/gdb/mylib.so...done.
16776 (gdb) remove-symbol-file -a 0x7ffff7ff9480
16777 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
16782 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
16784 @kindex add-symbol-file-from-memory
16785 @cindex @code{syscall DSO}
16786 @cindex load symbols from memory
16787 @item add-symbol-file-from-memory @var{address}
16788 Load symbols from the given @var{address} in a dynamically loaded
16789 object file whose image is mapped directly into the inferior's memory.
16790 For example, the Linux kernel maps a @code{syscall DSO} into each
16791 process's address space; this DSO provides kernel-specific code for
16792 some system calls. The argument can be any expression whose
16793 evaluation yields the address of the file's shared object file header.
16794 For this command to work, you must have used @code{symbol-file} or
16795 @code{exec-file} commands in advance.
16797 @kindex add-shared-symbol-files
16799 @item add-shared-symbol-files @var{library-file}
16800 @itemx assf @var{library-file}
16801 The @code{add-shared-symbol-files} command can currently be used only
16802 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
16803 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
16804 @value{GDBN} automatically looks for shared libraries, however if
16805 @value{GDBN} does not find yours, you can invoke
16806 @code{add-shared-symbol-files}. It takes one argument: the shared
16807 library's file name. @code{assf} is a shorthand alias for
16808 @code{add-shared-symbol-files}.
16811 @item section @var{section} @var{addr}
16812 The @code{section} command changes the base address of the named
16813 @var{section} of the exec file to @var{addr}. This can be used if the
16814 exec file does not contain section addresses, (such as in the
16815 @code{a.out} format), or when the addresses specified in the file
16816 itself are wrong. Each section must be changed separately. The
16817 @code{info files} command, described below, lists all the sections and
16821 @kindex info target
16824 @code{info files} and @code{info target} are synonymous; both print the
16825 current target (@pxref{Targets, ,Specifying a Debugging Target}),
16826 including the names of the executable and core dump files currently in
16827 use by @value{GDBN}, and the files from which symbols were loaded. The
16828 command @code{help target} lists all possible targets rather than
16831 @kindex maint info sections
16832 @item maint info sections
16833 Another command that can give you extra information about program sections
16834 is @code{maint info sections}. In addition to the section information
16835 displayed by @code{info files}, this command displays the flags and file
16836 offset of each section in the executable and core dump files. In addition,
16837 @code{maint info sections} provides the following command options (which
16838 may be arbitrarily combined):
16842 Display sections for all loaded object files, including shared libraries.
16843 @item @var{sections}
16844 Display info only for named @var{sections}.
16845 @item @var{section-flags}
16846 Display info only for sections for which @var{section-flags} are true.
16847 The section flags that @value{GDBN} currently knows about are:
16850 Section will have space allocated in the process when loaded.
16851 Set for all sections except those containing debug information.
16853 Section will be loaded from the file into the child process memory.
16854 Set for pre-initialized code and data, clear for @code{.bss} sections.
16856 Section needs to be relocated before loading.
16858 Section cannot be modified by the child process.
16860 Section contains executable code only.
16862 Section contains data only (no executable code).
16864 Section will reside in ROM.
16866 Section contains data for constructor/destructor lists.
16868 Section is not empty.
16870 An instruction to the linker to not output the section.
16871 @item COFF_SHARED_LIBRARY
16872 A notification to the linker that the section contains
16873 COFF shared library information.
16875 Section contains common symbols.
16878 @kindex set trust-readonly-sections
16879 @cindex read-only sections
16880 @item set trust-readonly-sections on
16881 Tell @value{GDBN} that readonly sections in your object file
16882 really are read-only (i.e.@: that their contents will not change).
16883 In that case, @value{GDBN} can fetch values from these sections
16884 out of the object file, rather than from the target program.
16885 For some targets (notably embedded ones), this can be a significant
16886 enhancement to debugging performance.
16888 The default is off.
16890 @item set trust-readonly-sections off
16891 Tell @value{GDBN} not to trust readonly sections. This means that
16892 the contents of the section might change while the program is running,
16893 and must therefore be fetched from the target when needed.
16895 @item show trust-readonly-sections
16896 Show the current setting of trusting readonly sections.
16899 All file-specifying commands allow both absolute and relative file names
16900 as arguments. @value{GDBN} always converts the file name to an absolute file
16901 name and remembers it that way.
16903 @cindex shared libraries
16904 @anchor{Shared Libraries}
16905 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
16906 and IBM RS/6000 AIX shared libraries.
16908 On MS-Windows @value{GDBN} must be linked with the Expat library to support
16909 shared libraries. @xref{Expat}.
16911 @value{GDBN} automatically loads symbol definitions from shared libraries
16912 when you use the @code{run} command, or when you examine a core file.
16913 (Before you issue the @code{run} command, @value{GDBN} does not understand
16914 references to a function in a shared library, however---unless you are
16915 debugging a core file).
16917 On HP-UX, if the program loads a library explicitly, @value{GDBN}
16918 automatically loads the symbols at the time of the @code{shl_load} call.
16920 @c FIXME: some @value{GDBN} release may permit some refs to undef
16921 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
16922 @c FIXME...lib; check this from time to time when updating manual
16924 There are times, however, when you may wish to not automatically load
16925 symbol definitions from shared libraries, such as when they are
16926 particularly large or there are many of them.
16928 To control the automatic loading of shared library symbols, use the
16932 @kindex set auto-solib-add
16933 @item set auto-solib-add @var{mode}
16934 If @var{mode} is @code{on}, symbols from all shared object libraries
16935 will be loaded automatically when the inferior begins execution, you
16936 attach to an independently started inferior, or when the dynamic linker
16937 informs @value{GDBN} that a new library has been loaded. If @var{mode}
16938 is @code{off}, symbols must be loaded manually, using the
16939 @code{sharedlibrary} command. The default value is @code{on}.
16941 @cindex memory used for symbol tables
16942 If your program uses lots of shared libraries with debug info that
16943 takes large amounts of memory, you can decrease the @value{GDBN}
16944 memory footprint by preventing it from automatically loading the
16945 symbols from shared libraries. To that end, type @kbd{set
16946 auto-solib-add off} before running the inferior, then load each
16947 library whose debug symbols you do need with @kbd{sharedlibrary
16948 @var{regexp}}, where @var{regexp} is a regular expression that matches
16949 the libraries whose symbols you want to be loaded.
16951 @kindex show auto-solib-add
16952 @item show auto-solib-add
16953 Display the current autoloading mode.
16956 @cindex load shared library
16957 To explicitly load shared library symbols, use the @code{sharedlibrary}
16961 @kindex info sharedlibrary
16963 @item info share @var{regex}
16964 @itemx info sharedlibrary @var{regex}
16965 Print the names of the shared libraries which are currently loaded
16966 that match @var{regex}. If @var{regex} is omitted then print
16967 all shared libraries that are loaded.
16969 @kindex sharedlibrary
16971 @item sharedlibrary @var{regex}
16972 @itemx share @var{regex}
16973 Load shared object library symbols for files matching a
16974 Unix regular expression.
16975 As with files loaded automatically, it only loads shared libraries
16976 required by your program for a core file or after typing @code{run}. If
16977 @var{regex} is omitted all shared libraries required by your program are
16980 @item nosharedlibrary
16981 @kindex nosharedlibrary
16982 @cindex unload symbols from shared libraries
16983 Unload all shared object library symbols. This discards all symbols
16984 that have been loaded from all shared libraries. Symbols from shared
16985 libraries that were loaded by explicit user requests are not
16989 Sometimes you may wish that @value{GDBN} stops and gives you control
16990 when any of shared library events happen. The best way to do this is
16991 to use @code{catch load} and @code{catch unload} (@pxref{Set
16994 @value{GDBN} also supports the the @code{set stop-on-solib-events}
16995 command for this. This command exists for historical reasons. It is
16996 less useful than setting a catchpoint, because it does not allow for
16997 conditions or commands as a catchpoint does.
17000 @item set stop-on-solib-events
17001 @kindex set stop-on-solib-events
17002 This command controls whether @value{GDBN} should give you control
17003 when the dynamic linker notifies it about some shared library event.
17004 The most common event of interest is loading or unloading of a new
17007 @item show stop-on-solib-events
17008 @kindex show stop-on-solib-events
17009 Show whether @value{GDBN} stops and gives you control when shared
17010 library events happen.
17013 Shared libraries are also supported in many cross or remote debugging
17014 configurations. @value{GDBN} needs to have access to the target's libraries;
17015 this can be accomplished either by providing copies of the libraries
17016 on the host system, or by asking @value{GDBN} to automatically retrieve the
17017 libraries from the target. If copies of the target libraries are
17018 provided, they need to be the same as the target libraries, although the
17019 copies on the target can be stripped as long as the copies on the host are
17022 @cindex where to look for shared libraries
17023 For remote debugging, you need to tell @value{GDBN} where the target
17024 libraries are, so that it can load the correct copies---otherwise, it
17025 may try to load the host's libraries. @value{GDBN} has two variables
17026 to specify the search directories for target libraries.
17029 @cindex prefix for shared library file names
17030 @cindex system root, alternate
17031 @kindex set solib-absolute-prefix
17032 @kindex set sysroot
17033 @item set sysroot @var{path}
17034 Use @var{path} as the system root for the program being debugged. Any
17035 absolute shared library paths will be prefixed with @var{path}; many
17036 runtime loaders store the absolute paths to the shared library in the
17037 target program's memory. If you use @code{set sysroot} to find shared
17038 libraries, they need to be laid out in the same way that they are on
17039 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
17042 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
17043 retrieve the target libraries from the remote system. This is only
17044 supported when using a remote target that supports the @code{remote get}
17045 command (@pxref{File Transfer,,Sending files to a remote system}).
17046 The part of @var{path} following the initial @file{remote:}
17047 (if present) is used as system root prefix on the remote file system.
17048 @footnote{If you want to specify a local system root using a directory
17049 that happens to be named @file{remote:}, you need to use some equivalent
17050 variant of the name like @file{./remote:}.}
17052 For targets with an MS-DOS based filesystem, such as MS-Windows and
17053 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
17054 absolute file name with @var{path}. But first, on Unix hosts,
17055 @value{GDBN} converts all backslash directory separators into forward
17056 slashes, because the backslash is not a directory separator on Unix:
17059 c:\foo\bar.dll @result{} c:/foo/bar.dll
17062 Then, @value{GDBN} attempts prefixing the target file name with
17063 @var{path}, and looks for the resulting file name in the host file
17067 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
17070 If that does not find the shared library, @value{GDBN} tries removing
17071 the @samp{:} character from the drive spec, both for convenience, and,
17072 for the case of the host file system not supporting file names with
17076 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
17079 This makes it possible to have a system root that mirrors a target
17080 with more than one drive. E.g., you may want to setup your local
17081 copies of the target system shared libraries like so (note @samp{c} vs
17085 @file{/path/to/sysroot/c/sys/bin/foo.dll}
17086 @file{/path/to/sysroot/c/sys/bin/bar.dll}
17087 @file{/path/to/sysroot/z/sys/bin/bar.dll}
17091 and point the system root at @file{/path/to/sysroot}, so that
17092 @value{GDBN} can find the correct copies of both
17093 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
17095 If that still does not find the shared library, @value{GDBN} tries
17096 removing the whole drive spec from the target file name:
17099 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
17102 This last lookup makes it possible to not care about the drive name,
17103 if you don't want or need to.
17105 The @code{set solib-absolute-prefix} command is an alias for @code{set
17108 @cindex default system root
17109 @cindex @samp{--with-sysroot}
17110 You can set the default system root by using the configure-time
17111 @samp{--with-sysroot} option. If the system root is inside
17112 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
17113 @samp{--exec-prefix}), then the default system root will be updated
17114 automatically if the installed @value{GDBN} is moved to a new
17117 @kindex show sysroot
17119 Display the current shared library prefix.
17121 @kindex set solib-search-path
17122 @item set solib-search-path @var{path}
17123 If this variable is set, @var{path} is a colon-separated list of
17124 directories to search for shared libraries. @samp{solib-search-path}
17125 is used after @samp{sysroot} fails to locate the library, or if the
17126 path to the library is relative instead of absolute. If you want to
17127 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
17128 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
17129 finding your host's libraries. @samp{sysroot} is preferred; setting
17130 it to a nonexistent directory may interfere with automatic loading
17131 of shared library symbols.
17133 @kindex show solib-search-path
17134 @item show solib-search-path
17135 Display the current shared library search path.
17137 @cindex DOS file-name semantics of file names.
17138 @kindex set target-file-system-kind (unix|dos-based|auto)
17139 @kindex show target-file-system-kind
17140 @item set target-file-system-kind @var{kind}
17141 Set assumed file system kind for target reported file names.
17143 Shared library file names as reported by the target system may not
17144 make sense as is on the system @value{GDBN} is running on. For
17145 example, when remote debugging a target that has MS-DOS based file
17146 system semantics, from a Unix host, the target may be reporting to
17147 @value{GDBN} a list of loaded shared libraries with file names such as
17148 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
17149 drive letters, so the @samp{c:\} prefix is not normally understood as
17150 indicating an absolute file name, and neither is the backslash
17151 normally considered a directory separator character. In that case,
17152 the native file system would interpret this whole absolute file name
17153 as a relative file name with no directory components. This would make
17154 it impossible to point @value{GDBN} at a copy of the remote target's
17155 shared libraries on the host using @code{set sysroot}, and impractical
17156 with @code{set solib-search-path}. Setting
17157 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
17158 to interpret such file names similarly to how the target would, and to
17159 map them to file names valid on @value{GDBN}'s native file system
17160 semantics. The value of @var{kind} can be @code{"auto"}, in addition
17161 to one of the supported file system kinds. In that case, @value{GDBN}
17162 tries to determine the appropriate file system variant based on the
17163 current target's operating system (@pxref{ABI, ,Configuring the
17164 Current ABI}). The supported file system settings are:
17168 Instruct @value{GDBN} to assume the target file system is of Unix
17169 kind. Only file names starting the forward slash (@samp{/}) character
17170 are considered absolute, and the directory separator character is also
17174 Instruct @value{GDBN} to assume the target file system is DOS based.
17175 File names starting with either a forward slash, or a drive letter
17176 followed by a colon (e.g., @samp{c:}), are considered absolute, and
17177 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
17178 considered directory separators.
17181 Instruct @value{GDBN} to use the file system kind associated with the
17182 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
17183 This is the default.
17187 @cindex file name canonicalization
17188 @cindex base name differences
17189 When processing file names provided by the user, @value{GDBN}
17190 frequently needs to compare them to the file names recorded in the
17191 program's debug info. Normally, @value{GDBN} compares just the
17192 @dfn{base names} of the files as strings, which is reasonably fast
17193 even for very large programs. (The base name of a file is the last
17194 portion of its name, after stripping all the leading directories.)
17195 This shortcut in comparison is based upon the assumption that files
17196 cannot have more than one base name. This is usually true, but
17197 references to files that use symlinks or similar filesystem
17198 facilities violate that assumption. If your program records files
17199 using such facilities, or if you provide file names to @value{GDBN}
17200 using symlinks etc., you can set @code{basenames-may-differ} to
17201 @code{true} to instruct @value{GDBN} to completely canonicalize each
17202 pair of file names it needs to compare. This will make file-name
17203 comparisons accurate, but at a price of a significant slowdown.
17206 @item set basenames-may-differ
17207 @kindex set basenames-may-differ
17208 Set whether a source file may have multiple base names.
17210 @item show basenames-may-differ
17211 @kindex show basenames-may-differ
17212 Show whether a source file may have multiple base names.
17215 @node Separate Debug Files
17216 @section Debugging Information in Separate Files
17217 @cindex separate debugging information files
17218 @cindex debugging information in separate files
17219 @cindex @file{.debug} subdirectories
17220 @cindex debugging information directory, global
17221 @cindex global debugging information directories
17222 @cindex build ID, and separate debugging files
17223 @cindex @file{.build-id} directory
17225 @value{GDBN} allows you to put a program's debugging information in a
17226 file separate from the executable itself, in a way that allows
17227 @value{GDBN} to find and load the debugging information automatically.
17228 Since debugging information can be very large---sometimes larger
17229 than the executable code itself---some systems distribute debugging
17230 information for their executables in separate files, which users can
17231 install only when they need to debug a problem.
17233 @value{GDBN} supports two ways of specifying the separate debug info
17238 The executable contains a @dfn{debug link} that specifies the name of
17239 the separate debug info file. The separate debug file's name is
17240 usually @file{@var{executable}.debug}, where @var{executable} is the
17241 name of the corresponding executable file without leading directories
17242 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
17243 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
17244 checksum for the debug file, which @value{GDBN} uses to validate that
17245 the executable and the debug file came from the same build.
17248 The executable contains a @dfn{build ID}, a unique bit string that is
17249 also present in the corresponding debug info file. (This is supported
17250 only on some operating systems, notably those which use the ELF format
17251 for binary files and the @sc{gnu} Binutils.) For more details about
17252 this feature, see the description of the @option{--build-id}
17253 command-line option in @ref{Options, , Command Line Options, ld.info,
17254 The GNU Linker}. The debug info file's name is not specified
17255 explicitly by the build ID, but can be computed from the build ID, see
17259 Depending on the way the debug info file is specified, @value{GDBN}
17260 uses two different methods of looking for the debug file:
17264 For the ``debug link'' method, @value{GDBN} looks up the named file in
17265 the directory of the executable file, then in a subdirectory of that
17266 directory named @file{.debug}, and finally under each one of the global debug
17267 directories, in a subdirectory whose name is identical to the leading
17268 directories of the executable's absolute file name.
17271 For the ``build ID'' method, @value{GDBN} looks in the
17272 @file{.build-id} subdirectory of each one of the global debug directories for
17273 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
17274 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
17275 are the rest of the bit string. (Real build ID strings are 32 or more
17276 hex characters, not 10.)
17279 So, for example, suppose you ask @value{GDBN} to debug
17280 @file{/usr/bin/ls}, which has a debug link that specifies the
17281 file @file{ls.debug}, and a build ID whose value in hex is
17282 @code{abcdef1234}. If the list of the global debug directories includes
17283 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
17284 debug information files, in the indicated order:
17288 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
17290 @file{/usr/bin/ls.debug}
17292 @file{/usr/bin/.debug/ls.debug}
17294 @file{/usr/lib/debug/usr/bin/ls.debug}.
17297 @anchor{debug-file-directory}
17298 Global debugging info directories default to what is set by @value{GDBN}
17299 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
17300 you can also set the global debugging info directories, and view the list
17301 @value{GDBN} is currently using.
17305 @kindex set debug-file-directory
17306 @item set debug-file-directory @var{directories}
17307 Set the directories which @value{GDBN} searches for separate debugging
17308 information files to @var{directory}. Multiple path components can be set
17309 concatenating them by a path separator.
17311 @kindex show debug-file-directory
17312 @item show debug-file-directory
17313 Show the directories @value{GDBN} searches for separate debugging
17318 @cindex @code{.gnu_debuglink} sections
17319 @cindex debug link sections
17320 A debug link is a special section of the executable file named
17321 @code{.gnu_debuglink}. The section must contain:
17325 A filename, with any leading directory components removed, followed by
17328 zero to three bytes of padding, as needed to reach the next four-byte
17329 boundary within the section, and
17331 a four-byte CRC checksum, stored in the same endianness used for the
17332 executable file itself. The checksum is computed on the debugging
17333 information file's full contents by the function given below, passing
17334 zero as the @var{crc} argument.
17337 Any executable file format can carry a debug link, as long as it can
17338 contain a section named @code{.gnu_debuglink} with the contents
17341 @cindex @code{.note.gnu.build-id} sections
17342 @cindex build ID sections
17343 The build ID is a special section in the executable file (and in other
17344 ELF binary files that @value{GDBN} may consider). This section is
17345 often named @code{.note.gnu.build-id}, but that name is not mandatory.
17346 It contains unique identification for the built files---the ID remains
17347 the same across multiple builds of the same build tree. The default
17348 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
17349 content for the build ID string. The same section with an identical
17350 value is present in the original built binary with symbols, in its
17351 stripped variant, and in the separate debugging information file.
17353 The debugging information file itself should be an ordinary
17354 executable, containing a full set of linker symbols, sections, and
17355 debugging information. The sections of the debugging information file
17356 should have the same names, addresses, and sizes as the original file,
17357 but they need not contain any data---much like a @code{.bss} section
17358 in an ordinary executable.
17360 The @sc{gnu} binary utilities (Binutils) package includes the
17361 @samp{objcopy} utility that can produce
17362 the separated executable / debugging information file pairs using the
17363 following commands:
17366 @kbd{objcopy --only-keep-debug foo foo.debug}
17371 These commands remove the debugging
17372 information from the executable file @file{foo} and place it in the file
17373 @file{foo.debug}. You can use the first, second or both methods to link the
17378 The debug link method needs the following additional command to also leave
17379 behind a debug link in @file{foo}:
17382 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
17385 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
17386 a version of the @code{strip} command such that the command @kbd{strip foo -f
17387 foo.debug} has the same functionality as the two @code{objcopy} commands and
17388 the @code{ln -s} command above, together.
17391 Build ID gets embedded into the main executable using @code{ld --build-id} or
17392 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
17393 compatibility fixes for debug files separation are present in @sc{gnu} binary
17394 utilities (Binutils) package since version 2.18.
17399 @cindex CRC algorithm definition
17400 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
17401 IEEE 802.3 using the polynomial:
17403 @c TexInfo requires naked braces for multi-digit exponents for Tex
17404 @c output, but this causes HTML output to barf. HTML has to be set using
17405 @c raw commands. So we end up having to specify this equation in 2
17410 <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>
17411 + <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
17417 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
17418 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
17422 The function is computed byte at a time, taking the least
17423 significant bit of each byte first. The initial pattern
17424 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
17425 the final result is inverted to ensure trailing zeros also affect the
17428 @emph{Note:} This is the same CRC polynomial as used in handling the
17429 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
17430 , @value{GDBN} Remote Serial Protocol}). However in the
17431 case of the Remote Serial Protocol, the CRC is computed @emph{most}
17432 significant bit first, and the result is not inverted, so trailing
17433 zeros have no effect on the CRC value.
17435 To complete the description, we show below the code of the function
17436 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
17437 initially supplied @code{crc} argument means that an initial call to
17438 this function passing in zero will start computing the CRC using
17441 @kindex gnu_debuglink_crc32
17444 gnu_debuglink_crc32 (unsigned long crc,
17445 unsigned char *buf, size_t len)
17447 static const unsigned long crc32_table[256] =
17449 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
17450 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
17451 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
17452 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
17453 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
17454 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
17455 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
17456 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
17457 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
17458 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
17459 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
17460 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
17461 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
17462 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
17463 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
17464 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
17465 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
17466 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
17467 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
17468 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
17469 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
17470 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
17471 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
17472 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
17473 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
17474 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
17475 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
17476 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
17477 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
17478 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
17479 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
17480 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
17481 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
17482 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
17483 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
17484 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
17485 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
17486 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
17487 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
17488 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
17489 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
17490 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
17491 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
17492 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
17493 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
17494 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
17495 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
17496 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
17497 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
17498 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
17499 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
17502 unsigned char *end;
17504 crc = ~crc & 0xffffffff;
17505 for (end = buf + len; buf < end; ++buf)
17506 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
17507 return ~crc & 0xffffffff;
17512 This computation does not apply to the ``build ID'' method.
17514 @node MiniDebugInfo
17515 @section Debugging information in a special section
17516 @cindex separate debug sections
17517 @cindex @samp{.gnu_debugdata} section
17519 Some systems ship pre-built executables and libraries that have a
17520 special @samp{.gnu_debugdata} section. This feature is called
17521 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
17522 is used to supply extra symbols for backtraces.
17524 The intent of this section is to provide extra minimal debugging
17525 information for use in simple backtraces. It is not intended to be a
17526 replacement for full separate debugging information (@pxref{Separate
17527 Debug Files}). The example below shows the intended use; however,
17528 @value{GDBN} does not currently put restrictions on what sort of
17529 debugging information might be included in the section.
17531 @value{GDBN} has support for this extension. If the section exists,
17532 then it is used provided that no other source of debugging information
17533 can be found, and that @value{GDBN} was configured with LZMA support.
17535 This section can be easily created using @command{objcopy} and other
17536 standard utilities:
17539 # Extract the dynamic symbols from the main binary, there is no need
17540 # to also have these in the normal symbol table.
17541 nm -D @var{binary} --format=posix --defined-only \
17542 | awk '@{ print $1 @}' | sort > dynsyms
17544 # Extract all the text (i.e. function) symbols from the debuginfo.
17545 # (Note that we actually also accept "D" symbols, for the benefit
17546 # of platforms like PowerPC64 that use function descriptors.)
17547 nm @var{binary} --format=posix --defined-only \
17548 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
17551 # Keep all the function symbols not already in the dynamic symbol
17553 comm -13 dynsyms funcsyms > keep_symbols
17555 # Separate full debug info into debug binary.
17556 objcopy --only-keep-debug @var{binary} debug
17558 # Copy the full debuginfo, keeping only a minimal set of symbols and
17559 # removing some unnecessary sections.
17560 objcopy -S --remove-section .gdb_index --remove-section .comment \
17561 --keep-symbols=keep_symbols debug mini_debuginfo
17563 # Drop the full debug info from the original binary.
17564 strip --strip-all -R .comment @var{binary}
17566 # Inject the compressed data into the .gnu_debugdata section of the
17569 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
17573 @section Index Files Speed Up @value{GDBN}
17574 @cindex index files
17575 @cindex @samp{.gdb_index} section
17577 When @value{GDBN} finds a symbol file, it scans the symbols in the
17578 file in order to construct an internal symbol table. This lets most
17579 @value{GDBN} operations work quickly---at the cost of a delay early
17580 on. For large programs, this delay can be quite lengthy, so
17581 @value{GDBN} provides a way to build an index, which speeds up
17584 The index is stored as a section in the symbol file. @value{GDBN} can
17585 write the index to a file, then you can put it into the symbol file
17586 using @command{objcopy}.
17588 To create an index file, use the @code{save gdb-index} command:
17591 @item save gdb-index @var{directory}
17592 @kindex save gdb-index
17593 Create an index file for each symbol file currently known by
17594 @value{GDBN}. Each file is named after its corresponding symbol file,
17595 with @samp{.gdb-index} appended, and is written into the given
17599 Once you have created an index file you can merge it into your symbol
17600 file, here named @file{symfile}, using @command{objcopy}:
17603 $ objcopy --add-section .gdb_index=symfile.gdb-index \
17604 --set-section-flags .gdb_index=readonly symfile symfile
17607 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
17608 sections that have been deprecated. Usually they are deprecated because
17609 they are missing a new feature or have performance issues.
17610 To tell @value{GDBN} to use a deprecated index section anyway
17611 specify @code{set use-deprecated-index-sections on}.
17612 The default is @code{off}.
17613 This can speed up startup, but may result in some functionality being lost.
17614 @xref{Index Section Format}.
17616 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
17617 must be done before gdb reads the file. The following will not work:
17620 $ gdb -ex "set use-deprecated-index-sections on" <program>
17623 Instead you must do, for example,
17626 $ gdb -iex "set use-deprecated-index-sections on" <program>
17629 There are currently some limitation on indices. They only work when
17630 for DWARF debugging information, not stabs. And, they do not
17631 currently work for programs using Ada.
17633 @node Symbol Errors
17634 @section Errors Reading Symbol Files
17636 While reading a symbol file, @value{GDBN} occasionally encounters problems,
17637 such as symbol types it does not recognize, or known bugs in compiler
17638 output. By default, @value{GDBN} does not notify you of such problems, since
17639 they are relatively common and primarily of interest to people
17640 debugging compilers. If you are interested in seeing information
17641 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
17642 only one message about each such type of problem, no matter how many
17643 times the problem occurs; or you can ask @value{GDBN} to print more messages,
17644 to see how many times the problems occur, with the @code{set
17645 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
17648 The messages currently printed, and their meanings, include:
17651 @item inner block not inside outer block in @var{symbol}
17653 The symbol information shows where symbol scopes begin and end
17654 (such as at the start of a function or a block of statements). This
17655 error indicates that an inner scope block is not fully contained
17656 in its outer scope blocks.
17658 @value{GDBN} circumvents the problem by treating the inner block as if it had
17659 the same scope as the outer block. In the error message, @var{symbol}
17660 may be shown as ``@code{(don't know)}'' if the outer block is not a
17663 @item block at @var{address} out of order
17665 The symbol information for symbol scope blocks should occur in
17666 order of increasing addresses. This error indicates that it does not
17669 @value{GDBN} does not circumvent this problem, and has trouble
17670 locating symbols in the source file whose symbols it is reading. (You
17671 can often determine what source file is affected by specifying
17672 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
17675 @item bad block start address patched
17677 The symbol information for a symbol scope block has a start address
17678 smaller than the address of the preceding source line. This is known
17679 to occur in the SunOS 4.1.1 (and earlier) C compiler.
17681 @value{GDBN} circumvents the problem by treating the symbol scope block as
17682 starting on the previous source line.
17684 @item bad string table offset in symbol @var{n}
17687 Symbol number @var{n} contains a pointer into the string table which is
17688 larger than the size of the string table.
17690 @value{GDBN} circumvents the problem by considering the symbol to have the
17691 name @code{foo}, which may cause other problems if many symbols end up
17694 @item unknown symbol type @code{0x@var{nn}}
17696 The symbol information contains new data types that @value{GDBN} does
17697 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
17698 uncomprehended information, in hexadecimal.
17700 @value{GDBN} circumvents the error by ignoring this symbol information.
17701 This usually allows you to debug your program, though certain symbols
17702 are not accessible. If you encounter such a problem and feel like
17703 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
17704 on @code{complain}, then go up to the function @code{read_dbx_symtab}
17705 and examine @code{*bufp} to see the symbol.
17707 @item stub type has NULL name
17709 @value{GDBN} could not find the full definition for a struct or class.
17711 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
17712 The symbol information for a C@t{++} member function is missing some
17713 information that recent versions of the compiler should have output for
17716 @item info mismatch between compiler and debugger
17718 @value{GDBN} could not parse a type specification output by the compiler.
17723 @section GDB Data Files
17725 @cindex prefix for data files
17726 @value{GDBN} will sometimes read an auxiliary data file. These files
17727 are kept in a directory known as the @dfn{data directory}.
17729 You can set the data directory's name, and view the name @value{GDBN}
17730 is currently using.
17733 @kindex set data-directory
17734 @item set data-directory @var{directory}
17735 Set the directory which @value{GDBN} searches for auxiliary data files
17736 to @var{directory}.
17738 @kindex show data-directory
17739 @item show data-directory
17740 Show the directory @value{GDBN} searches for auxiliary data files.
17743 @cindex default data directory
17744 @cindex @samp{--with-gdb-datadir}
17745 You can set the default data directory by using the configure-time
17746 @samp{--with-gdb-datadir} option. If the data directory is inside
17747 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
17748 @samp{--exec-prefix}), then the default data directory will be updated
17749 automatically if the installed @value{GDBN} is moved to a new
17752 The data directory may also be specified with the
17753 @code{--data-directory} command line option.
17754 @xref{Mode Options}.
17757 @chapter Specifying a Debugging Target
17759 @cindex debugging target
17760 A @dfn{target} is the execution environment occupied by your program.
17762 Often, @value{GDBN} runs in the same host environment as your program;
17763 in that case, the debugging target is specified as a side effect when
17764 you use the @code{file} or @code{core} commands. When you need more
17765 flexibility---for example, running @value{GDBN} on a physically separate
17766 host, or controlling a standalone system over a serial port or a
17767 realtime system over a TCP/IP connection---you can use the @code{target}
17768 command to specify one of the target types configured for @value{GDBN}
17769 (@pxref{Target Commands, ,Commands for Managing Targets}).
17771 @cindex target architecture
17772 It is possible to build @value{GDBN} for several different @dfn{target
17773 architectures}. When @value{GDBN} is built like that, you can choose
17774 one of the available architectures with the @kbd{set architecture}
17778 @kindex set architecture
17779 @kindex show architecture
17780 @item set architecture @var{arch}
17781 This command sets the current target architecture to @var{arch}. The
17782 value of @var{arch} can be @code{"auto"}, in addition to one of the
17783 supported architectures.
17785 @item show architecture
17786 Show the current target architecture.
17788 @item set processor
17790 @kindex set processor
17791 @kindex show processor
17792 These are alias commands for, respectively, @code{set architecture}
17793 and @code{show architecture}.
17797 * Active Targets:: Active targets
17798 * Target Commands:: Commands for managing targets
17799 * Byte Order:: Choosing target byte order
17802 @node Active Targets
17803 @section Active Targets
17805 @cindex stacking targets
17806 @cindex active targets
17807 @cindex multiple targets
17809 There are multiple classes of targets such as: processes, executable files or
17810 recording sessions. Core files belong to the process class, making core file
17811 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
17812 on multiple active targets, one in each class. This allows you to (for
17813 example) start a process and inspect its activity, while still having access to
17814 the executable file after the process finishes. Or if you start process
17815 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
17816 presented a virtual layer of the recording target, while the process target
17817 remains stopped at the chronologically last point of the process execution.
17819 Use the @code{core-file} and @code{exec-file} commands to select a new core
17820 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
17821 specify as a target a process that is already running, use the @code{attach}
17822 command (@pxref{Attach, ,Debugging an Already-running Process}).
17824 @node Target Commands
17825 @section Commands for Managing Targets
17828 @item target @var{type} @var{parameters}
17829 Connects the @value{GDBN} host environment to a target machine or
17830 process. A target is typically a protocol for talking to debugging
17831 facilities. You use the argument @var{type} to specify the type or
17832 protocol of the target machine.
17834 Further @var{parameters} are interpreted by the target protocol, but
17835 typically include things like device names or host names to connect
17836 with, process numbers, and baud rates.
17838 The @code{target} command does not repeat if you press @key{RET} again
17839 after executing the command.
17841 @kindex help target
17843 Displays the names of all targets available. To display targets
17844 currently selected, use either @code{info target} or @code{info files}
17845 (@pxref{Files, ,Commands to Specify Files}).
17847 @item help target @var{name}
17848 Describe a particular target, including any parameters necessary to
17851 @kindex set gnutarget
17852 @item set gnutarget @var{args}
17853 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
17854 knows whether it is reading an @dfn{executable},
17855 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
17856 with the @code{set gnutarget} command. Unlike most @code{target} commands,
17857 with @code{gnutarget} the @code{target} refers to a program, not a machine.
17860 @emph{Warning:} To specify a file format with @code{set gnutarget},
17861 you must know the actual BFD name.
17865 @xref{Files, , Commands to Specify Files}.
17867 @kindex show gnutarget
17868 @item show gnutarget
17869 Use the @code{show gnutarget} command to display what file format
17870 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
17871 @value{GDBN} will determine the file format for each file automatically,
17872 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
17875 @cindex common targets
17876 Here are some common targets (available, or not, depending on the GDB
17881 @item target exec @var{program}
17882 @cindex executable file target
17883 An executable file. @samp{target exec @var{program}} is the same as
17884 @samp{exec-file @var{program}}.
17886 @item target core @var{filename}
17887 @cindex core dump file target
17888 A core dump file. @samp{target core @var{filename}} is the same as
17889 @samp{core-file @var{filename}}.
17891 @item target remote @var{medium}
17892 @cindex remote target
17893 A remote system connected to @value{GDBN} via a serial line or network
17894 connection. This command tells @value{GDBN} to use its own remote
17895 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
17897 For example, if you have a board connected to @file{/dev/ttya} on the
17898 machine running @value{GDBN}, you could say:
17901 target remote /dev/ttya
17904 @code{target remote} supports the @code{load} command. This is only
17905 useful if you have some other way of getting the stub to the target
17906 system, and you can put it somewhere in memory where it won't get
17907 clobbered by the download.
17909 @item target sim @r{[}@var{simargs}@r{]} @dots{}
17910 @cindex built-in simulator target
17911 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
17919 works; however, you cannot assume that a specific memory map, device
17920 drivers, or even basic I/O is available, although some simulators do
17921 provide these. For info about any processor-specific simulator details,
17922 see the appropriate section in @ref{Embedded Processors, ,Embedded
17927 Different targets are available on different configurations of @value{GDBN};
17928 your configuration may have more or fewer targets.
17930 Many remote targets require you to download the executable's code once
17931 you've successfully established a connection. You may wish to control
17932 various aspects of this process.
17937 @kindex set hash@r{, for remote monitors}
17938 @cindex hash mark while downloading
17939 This command controls whether a hash mark @samp{#} is displayed while
17940 downloading a file to the remote monitor. If on, a hash mark is
17941 displayed after each S-record is successfully downloaded to the
17945 @kindex show hash@r{, for remote monitors}
17946 Show the current status of displaying the hash mark.
17948 @item set debug monitor
17949 @kindex set debug monitor
17950 @cindex display remote monitor communications
17951 Enable or disable display of communications messages between
17952 @value{GDBN} and the remote monitor.
17954 @item show debug monitor
17955 @kindex show debug monitor
17956 Show the current status of displaying communications between
17957 @value{GDBN} and the remote monitor.
17962 @kindex load @var{filename}
17963 @item load @var{filename}
17965 Depending on what remote debugging facilities are configured into
17966 @value{GDBN}, the @code{load} command may be available. Where it exists, it
17967 is meant to make @var{filename} (an executable) available for debugging
17968 on the remote system---by downloading, or dynamic linking, for example.
17969 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
17970 the @code{add-symbol-file} command.
17972 If your @value{GDBN} does not have a @code{load} command, attempting to
17973 execute it gets the error message ``@code{You can't do that when your
17974 target is @dots{}}''
17976 The file is loaded at whatever address is specified in the executable.
17977 For some object file formats, you can specify the load address when you
17978 link the program; for other formats, like a.out, the object file format
17979 specifies a fixed address.
17980 @c FIXME! This would be a good place for an xref to the GNU linker doc.
17982 Depending on the remote side capabilities, @value{GDBN} may be able to
17983 load programs into flash memory.
17985 @code{load} does not repeat if you press @key{RET} again after using it.
17989 @section Choosing Target Byte Order
17991 @cindex choosing target byte order
17992 @cindex target byte order
17994 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
17995 offer the ability to run either big-endian or little-endian byte
17996 orders. Usually the executable or symbol will include a bit to
17997 designate the endian-ness, and you will not need to worry about
17998 which to use. However, you may still find it useful to adjust
17999 @value{GDBN}'s idea of processor endian-ness manually.
18003 @item set endian big
18004 Instruct @value{GDBN} to assume the target is big-endian.
18006 @item set endian little
18007 Instruct @value{GDBN} to assume the target is little-endian.
18009 @item set endian auto
18010 Instruct @value{GDBN} to use the byte order associated with the
18014 Display @value{GDBN}'s current idea of the target byte order.
18018 Note that these commands merely adjust interpretation of symbolic
18019 data on the host, and that they have absolutely no effect on the
18023 @node Remote Debugging
18024 @chapter Debugging Remote Programs
18025 @cindex remote debugging
18027 If you are trying to debug a program running on a machine that cannot run
18028 @value{GDBN} in the usual way, it is often useful to use remote debugging.
18029 For example, you might use remote debugging on an operating system kernel,
18030 or on a small system which does not have a general purpose operating system
18031 powerful enough to run a full-featured debugger.
18033 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
18034 to make this work with particular debugging targets. In addition,
18035 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
18036 but not specific to any particular target system) which you can use if you
18037 write the remote stubs---the code that runs on the remote system to
18038 communicate with @value{GDBN}.
18040 Other remote targets may be available in your
18041 configuration of @value{GDBN}; use @code{help target} to list them.
18044 * Connecting:: Connecting to a remote target
18045 * File Transfer:: Sending files to a remote system
18046 * Server:: Using the gdbserver program
18047 * Remote Configuration:: Remote configuration
18048 * Remote Stub:: Implementing a remote stub
18052 @section Connecting to a Remote Target
18054 On the @value{GDBN} host machine, you will need an unstripped copy of
18055 your program, since @value{GDBN} needs symbol and debugging information.
18056 Start up @value{GDBN} as usual, using the name of the local copy of your
18057 program as the first argument.
18059 @cindex @code{target remote}
18060 @value{GDBN} can communicate with the target over a serial line, or
18061 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
18062 each case, @value{GDBN} uses the same protocol for debugging your
18063 program; only the medium carrying the debugging packets varies. The
18064 @code{target remote} command establishes a connection to the target.
18065 Its arguments indicate which medium to use:
18069 @item target remote @var{serial-device}
18070 @cindex serial line, @code{target remote}
18071 Use @var{serial-device} to communicate with the target. For example,
18072 to use a serial line connected to the device named @file{/dev/ttyb}:
18075 target remote /dev/ttyb
18078 If you're using a serial line, you may want to give @value{GDBN} the
18079 @samp{--baud} option, or use the @code{set serial baud} command
18080 (@pxref{Remote Configuration, set serial baud}) before the
18081 @code{target} command.
18083 @item target remote @code{@var{host}:@var{port}}
18084 @itemx target remote @code{tcp:@var{host}:@var{port}}
18085 @cindex @acronym{TCP} port, @code{target remote}
18086 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
18087 The @var{host} may be either a host name or a numeric @acronym{IP}
18088 address; @var{port} must be a decimal number. The @var{host} could be
18089 the target machine itself, if it is directly connected to the net, or
18090 it might be a terminal server which in turn has a serial line to the
18093 For example, to connect to port 2828 on a terminal server named
18097 target remote manyfarms:2828
18100 If your remote target is actually running on the same machine as your
18101 debugger session (e.g.@: a simulator for your target running on the
18102 same host), you can omit the hostname. For example, to connect to
18103 port 1234 on your local machine:
18106 target remote :1234
18110 Note that the colon is still required here.
18112 @item target remote @code{udp:@var{host}:@var{port}}
18113 @cindex @acronym{UDP} port, @code{target remote}
18114 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
18115 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
18118 target remote udp:manyfarms:2828
18121 When using a @acronym{UDP} connection for remote debugging, you should
18122 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
18123 can silently drop packets on busy or unreliable networks, which will
18124 cause havoc with your debugging session.
18126 @item target remote | @var{command}
18127 @cindex pipe, @code{target remote} to
18128 Run @var{command} in the background and communicate with it using a
18129 pipe. The @var{command} is a shell command, to be parsed and expanded
18130 by the system's command shell, @code{/bin/sh}; it should expect remote
18131 protocol packets on its standard input, and send replies on its
18132 standard output. You could use this to run a stand-alone simulator
18133 that speaks the remote debugging protocol, to make net connections
18134 using programs like @code{ssh}, or for other similar tricks.
18136 If @var{command} closes its standard output (perhaps by exiting),
18137 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
18138 program has already exited, this will have no effect.)
18142 Once the connection has been established, you can use all the usual
18143 commands to examine and change data. The remote program is already
18144 running; you can use @kbd{step} and @kbd{continue}, and you do not
18145 need to use @kbd{run}.
18147 @cindex interrupting remote programs
18148 @cindex remote programs, interrupting
18149 Whenever @value{GDBN} is waiting for the remote program, if you type the
18150 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
18151 program. This may or may not succeed, depending in part on the hardware
18152 and the serial drivers the remote system uses. If you type the
18153 interrupt character once again, @value{GDBN} displays this prompt:
18156 Interrupted while waiting for the program.
18157 Give up (and stop debugging it)? (y or n)
18160 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
18161 (If you decide you want to try again later, you can use @samp{target
18162 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
18163 goes back to waiting.
18166 @kindex detach (remote)
18168 When you have finished debugging the remote program, you can use the
18169 @code{detach} command to release it from @value{GDBN} control.
18170 Detaching from the target normally resumes its execution, but the results
18171 will depend on your particular remote stub. After the @code{detach}
18172 command, @value{GDBN} is free to connect to another target.
18176 The @code{disconnect} command behaves like @code{detach}, except that
18177 the target is generally not resumed. It will wait for @value{GDBN}
18178 (this instance or another one) to connect and continue debugging. After
18179 the @code{disconnect} command, @value{GDBN} is again free to connect to
18182 @cindex send command to remote monitor
18183 @cindex extend @value{GDBN} for remote targets
18184 @cindex add new commands for external monitor
18186 @item monitor @var{cmd}
18187 This command allows you to send arbitrary commands directly to the
18188 remote monitor. Since @value{GDBN} doesn't care about the commands it
18189 sends like this, this command is the way to extend @value{GDBN}---you
18190 can add new commands that only the external monitor will understand
18194 @node File Transfer
18195 @section Sending files to a remote system
18196 @cindex remote target, file transfer
18197 @cindex file transfer
18198 @cindex sending files to remote systems
18200 Some remote targets offer the ability to transfer files over the same
18201 connection used to communicate with @value{GDBN}. This is convenient
18202 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
18203 running @code{gdbserver} over a network interface. For other targets,
18204 e.g.@: embedded devices with only a single serial port, this may be
18205 the only way to upload or download files.
18207 Not all remote targets support these commands.
18211 @item remote put @var{hostfile} @var{targetfile}
18212 Copy file @var{hostfile} from the host system (the machine running
18213 @value{GDBN}) to @var{targetfile} on the target system.
18216 @item remote get @var{targetfile} @var{hostfile}
18217 Copy file @var{targetfile} from the target system to @var{hostfile}
18218 on the host system.
18220 @kindex remote delete
18221 @item remote delete @var{targetfile}
18222 Delete @var{targetfile} from the target system.
18227 @section Using the @code{gdbserver} Program
18230 @cindex remote connection without stubs
18231 @code{gdbserver} is a control program for Unix-like systems, which
18232 allows you to connect your program with a remote @value{GDBN} via
18233 @code{target remote}---but without linking in the usual debugging stub.
18235 @code{gdbserver} is not a complete replacement for the debugging stubs,
18236 because it requires essentially the same operating-system facilities
18237 that @value{GDBN} itself does. In fact, a system that can run
18238 @code{gdbserver} to connect to a remote @value{GDBN} could also run
18239 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
18240 because it is a much smaller program than @value{GDBN} itself. It is
18241 also easier to port than all of @value{GDBN}, so you may be able to get
18242 started more quickly on a new system by using @code{gdbserver}.
18243 Finally, if you develop code for real-time systems, you may find that
18244 the tradeoffs involved in real-time operation make it more convenient to
18245 do as much development work as possible on another system, for example
18246 by cross-compiling. You can use @code{gdbserver} to make a similar
18247 choice for debugging.
18249 @value{GDBN} and @code{gdbserver} communicate via either a serial line
18250 or a TCP connection, using the standard @value{GDBN} remote serial
18254 @emph{Warning:} @code{gdbserver} does not have any built-in security.
18255 Do not run @code{gdbserver} connected to any public network; a
18256 @value{GDBN} connection to @code{gdbserver} provides access to the
18257 target system with the same privileges as the user running
18261 @subsection Running @code{gdbserver}
18262 @cindex arguments, to @code{gdbserver}
18263 @cindex @code{gdbserver}, command-line arguments
18265 Run @code{gdbserver} on the target system. You need a copy of the
18266 program you want to debug, including any libraries it requires.
18267 @code{gdbserver} does not need your program's symbol table, so you can
18268 strip the program if necessary to save space. @value{GDBN} on the host
18269 system does all the symbol handling.
18271 To use the server, you must tell it how to communicate with @value{GDBN};
18272 the name of your program; and the arguments for your program. The usual
18276 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
18279 @var{comm} is either a device name (to use a serial line), or a TCP
18280 hostname and portnumber, or @code{-} or @code{stdio} to use
18281 stdin/stdout of @code{gdbserver}.
18282 For example, to debug Emacs with the argument
18283 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
18287 target> gdbserver /dev/com1 emacs foo.txt
18290 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
18293 To use a TCP connection instead of a serial line:
18296 target> gdbserver host:2345 emacs foo.txt
18299 The only difference from the previous example is the first argument,
18300 specifying that you are communicating with the host @value{GDBN} via
18301 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
18302 expect a TCP connection from machine @samp{host} to local TCP port 2345.
18303 (Currently, the @samp{host} part is ignored.) You can choose any number
18304 you want for the port number as long as it does not conflict with any
18305 TCP ports already in use on the target system (for example, @code{23} is
18306 reserved for @code{telnet}).@footnote{If you choose a port number that
18307 conflicts with another service, @code{gdbserver} prints an error message
18308 and exits.} You must use the same port number with the host @value{GDBN}
18309 @code{target remote} command.
18311 The @code{stdio} connection is useful when starting @code{gdbserver}
18315 (gdb) target remote | ssh -T hostname gdbserver - hello
18318 The @samp{-T} option to ssh is provided because we don't need a remote pty,
18319 and we don't want escape-character handling. Ssh does this by default when
18320 a command is provided, the flag is provided to make it explicit.
18321 You could elide it if you want to.
18323 Programs started with stdio-connected gdbserver have @file{/dev/null} for
18324 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
18325 display through a pipe connected to gdbserver.
18326 Both @code{stdout} and @code{stderr} use the same pipe.
18328 @subsubsection Attaching to a Running Program
18329 @cindex attach to a program, @code{gdbserver}
18330 @cindex @option{--attach}, @code{gdbserver} option
18332 On some targets, @code{gdbserver} can also attach to running programs.
18333 This is accomplished via the @code{--attach} argument. The syntax is:
18336 target> gdbserver --attach @var{comm} @var{pid}
18339 @var{pid} is the process ID of a currently running process. It isn't necessary
18340 to point @code{gdbserver} at a binary for the running process.
18343 You can debug processes by name instead of process ID if your target has the
18344 @code{pidof} utility:
18347 target> gdbserver --attach @var{comm} `pidof @var{program}`
18350 In case more than one copy of @var{program} is running, or @var{program}
18351 has multiple threads, most versions of @code{pidof} support the
18352 @code{-s} option to only return the first process ID.
18354 @subsubsection Multi-Process Mode for @code{gdbserver}
18355 @cindex @code{gdbserver}, multiple processes
18356 @cindex multiple processes with @code{gdbserver}
18358 When you connect to @code{gdbserver} using @code{target remote},
18359 @code{gdbserver} debugs the specified program only once. When the
18360 program exits, or you detach from it, @value{GDBN} closes the connection
18361 and @code{gdbserver} exits.
18363 If you connect using @kbd{target extended-remote}, @code{gdbserver}
18364 enters multi-process mode. When the debugged program exits, or you
18365 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
18366 though no program is running. The @code{run} and @code{attach}
18367 commands instruct @code{gdbserver} to run or attach to a new program.
18368 The @code{run} command uses @code{set remote exec-file} (@pxref{set
18369 remote exec-file}) to select the program to run. Command line
18370 arguments are supported, except for wildcard expansion and I/O
18371 redirection (@pxref{Arguments}).
18373 @cindex @option{--multi}, @code{gdbserver} option
18374 To start @code{gdbserver} without supplying an initial command to run
18375 or process ID to attach, use the @option{--multi} command line option.
18376 Then you can connect using @kbd{target extended-remote} and start
18377 the program you want to debug.
18379 In multi-process mode @code{gdbserver} does not automatically exit unless you
18380 use the option @option{--once}. You can terminate it by using
18381 @code{monitor exit} (@pxref{Monitor Commands for gdbserver}). Note that the
18382 conditions under which @code{gdbserver} terminates depend on how @value{GDBN}
18383 connects to it (@kbd{target remote} or @kbd{target extended-remote}). The
18384 @option{--multi} option to @code{gdbserver} has no influence on that.
18386 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
18388 This section applies only when @code{gdbserver} is run to listen on a TCP port.
18390 @code{gdbserver} normally terminates after all of its debugged processes have
18391 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
18392 extended-remote}, @code{gdbserver} stays running even with no processes left.
18393 @value{GDBN} normally terminates the spawned debugged process on its exit,
18394 which normally also terminates @code{gdbserver} in the @kbd{target remote}
18395 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
18396 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
18397 stays running even in the @kbd{target remote} mode.
18399 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
18400 Such reconnecting is useful for features like @ref{disconnected tracing}. For
18401 completeness, at most one @value{GDBN} can be connected at a time.
18403 @cindex @option{--once}, @code{gdbserver} option
18404 By default, @code{gdbserver} keeps the listening TCP port open, so that
18405 subsequent connections are possible. However, if you start @code{gdbserver}
18406 with the @option{--once} option, it will stop listening for any further
18407 connection attempts after connecting to the first @value{GDBN} session. This
18408 means no further connections to @code{gdbserver} will be possible after the
18409 first one. It also means @code{gdbserver} will terminate after the first
18410 connection with remote @value{GDBN} has closed, even for unexpectedly closed
18411 connections and even in the @kbd{target extended-remote} mode. The
18412 @option{--once} option allows reusing the same port number for connecting to
18413 multiple instances of @code{gdbserver} running on the same host, since each
18414 instance closes its port after the first connection.
18416 @subsubsection Other Command-Line Arguments for @code{gdbserver}
18418 @cindex @option{--debug}, @code{gdbserver} option
18419 The @option{--debug} option tells @code{gdbserver} to display extra
18420 status information about the debugging process.
18421 @cindex @option{--remote-debug}, @code{gdbserver} option
18422 The @option{--remote-debug} option tells @code{gdbserver} to display
18423 remote protocol debug output. These options are intended for
18424 @code{gdbserver} development and for bug reports to the developers.
18426 @cindex @option{--wrapper}, @code{gdbserver} option
18427 The @option{--wrapper} option specifies a wrapper to launch programs
18428 for debugging. The option should be followed by the name of the
18429 wrapper, then any command-line arguments to pass to the wrapper, then
18430 @kbd{--} indicating the end of the wrapper arguments.
18432 @code{gdbserver} runs the specified wrapper program with a combined
18433 command line including the wrapper arguments, then the name of the
18434 program to debug, then any arguments to the program. The wrapper
18435 runs until it executes your program, and then @value{GDBN} gains control.
18437 You can use any program that eventually calls @code{execve} with
18438 its arguments as a wrapper. Several standard Unix utilities do
18439 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
18440 with @code{exec "$@@"} will also work.
18442 For example, you can use @code{env} to pass an environment variable to
18443 the debugged program, without setting the variable in @code{gdbserver}'s
18447 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
18450 @subsection Connecting to @code{gdbserver}
18452 Run @value{GDBN} on the host system.
18454 First make sure you have the necessary symbol files. Load symbols for
18455 your application using the @code{file} command before you connect. Use
18456 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
18457 was compiled with the correct sysroot using @code{--with-sysroot}).
18459 The symbol file and target libraries must exactly match the executable
18460 and libraries on the target, with one exception: the files on the host
18461 system should not be stripped, even if the files on the target system
18462 are. Mismatched or missing files will lead to confusing results
18463 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
18464 files may also prevent @code{gdbserver} from debugging multi-threaded
18467 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
18468 For TCP connections, you must start up @code{gdbserver} prior to using
18469 the @code{target remote} command. Otherwise you may get an error whose
18470 text depends on the host system, but which usually looks something like
18471 @samp{Connection refused}. Don't use the @code{load}
18472 command in @value{GDBN} when using @code{gdbserver}, since the program is
18473 already on the target.
18475 @subsection Monitor Commands for @code{gdbserver}
18476 @cindex monitor commands, for @code{gdbserver}
18477 @anchor{Monitor Commands for gdbserver}
18479 During a @value{GDBN} session using @code{gdbserver}, you can use the
18480 @code{monitor} command to send special requests to @code{gdbserver}.
18481 Here are the available commands.
18485 List the available monitor commands.
18487 @item monitor set debug 0
18488 @itemx monitor set debug 1
18489 Disable or enable general debugging messages.
18491 @item monitor set remote-debug 0
18492 @itemx monitor set remote-debug 1
18493 Disable or enable specific debugging messages associated with the remote
18494 protocol (@pxref{Remote Protocol}).
18496 @item monitor set libthread-db-search-path [PATH]
18497 @cindex gdbserver, search path for @code{libthread_db}
18498 When this command is issued, @var{path} is a colon-separated list of
18499 directories to search for @code{libthread_db} (@pxref{Threads,,set
18500 libthread-db-search-path}). If you omit @var{path},
18501 @samp{libthread-db-search-path} will be reset to its default value.
18503 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
18504 not supported in @code{gdbserver}.
18507 Tell gdbserver to exit immediately. This command should be followed by
18508 @code{disconnect} to close the debugging session. @code{gdbserver} will
18509 detach from any attached processes and kill any processes it created.
18510 Use @code{monitor exit} to terminate @code{gdbserver} at the end
18511 of a multi-process mode debug session.
18515 @subsection Tracepoints support in @code{gdbserver}
18516 @cindex tracepoints support in @code{gdbserver}
18518 On some targets, @code{gdbserver} supports tracepoints, fast
18519 tracepoints and static tracepoints.
18521 For fast or static tracepoints to work, a special library called the
18522 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
18523 This library is built and distributed as an integral part of
18524 @code{gdbserver}. In addition, support for static tracepoints
18525 requires building the in-process agent library with static tracepoints
18526 support. At present, the UST (LTTng Userspace Tracer,
18527 @url{http://lttng.org/ust}) tracing engine is supported. This support
18528 is automatically available if UST development headers are found in the
18529 standard include path when @code{gdbserver} is built, or if
18530 @code{gdbserver} was explicitly configured using @option{--with-ust}
18531 to point at such headers. You can explicitly disable the support
18532 using @option{--with-ust=no}.
18534 There are several ways to load the in-process agent in your program:
18537 @item Specifying it as dependency at link time
18539 You can link your program dynamically with the in-process agent
18540 library. On most systems, this is accomplished by adding
18541 @code{-linproctrace} to the link command.
18543 @item Using the system's preloading mechanisms
18545 You can force loading the in-process agent at startup time by using
18546 your system's support for preloading shared libraries. Many Unixes
18547 support the concept of preloading user defined libraries. In most
18548 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
18549 in the environment. See also the description of @code{gdbserver}'s
18550 @option{--wrapper} command line option.
18552 @item Using @value{GDBN} to force loading the agent at run time
18554 On some systems, you can force the inferior to load a shared library,
18555 by calling a dynamic loader function in the inferior that takes care
18556 of dynamically looking up and loading a shared library. On most Unix
18557 systems, the function is @code{dlopen}. You'll use the @code{call}
18558 command for that. For example:
18561 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
18564 Note that on most Unix systems, for the @code{dlopen} function to be
18565 available, the program needs to be linked with @code{-ldl}.
18568 On systems that have a userspace dynamic loader, like most Unix
18569 systems, when you connect to @code{gdbserver} using @code{target
18570 remote}, you'll find that the program is stopped at the dynamic
18571 loader's entry point, and no shared library has been loaded in the
18572 program's address space yet, including the in-process agent. In that
18573 case, before being able to use any of the fast or static tracepoints
18574 features, you need to let the loader run and load the shared
18575 libraries. The simplest way to do that is to run the program to the
18576 main procedure. E.g., if debugging a C or C@t{++} program, start
18577 @code{gdbserver} like so:
18580 $ gdbserver :9999 myprogram
18583 Start GDB and connect to @code{gdbserver} like so, and run to main:
18587 (@value{GDBP}) target remote myhost:9999
18588 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
18589 (@value{GDBP}) b main
18590 (@value{GDBP}) continue
18593 The in-process tracing agent library should now be loaded into the
18594 process; you can confirm it with the @code{info sharedlibrary}
18595 command, which will list @file{libinproctrace.so} as loaded in the
18596 process. You are now ready to install fast tracepoints, list static
18597 tracepoint markers, probe static tracepoints markers, and start
18600 @node Remote Configuration
18601 @section Remote Configuration
18604 @kindex show remote
18605 This section documents the configuration options available when
18606 debugging remote programs. For the options related to the File I/O
18607 extensions of the remote protocol, see @ref{system,
18608 system-call-allowed}.
18611 @item set remoteaddresssize @var{bits}
18612 @cindex address size for remote targets
18613 @cindex bits in remote address
18614 Set the maximum size of address in a memory packet to the specified
18615 number of bits. @value{GDBN} will mask off the address bits above
18616 that number, when it passes addresses to the remote target. The
18617 default value is the number of bits in the target's address.
18619 @item show remoteaddresssize
18620 Show the current value of remote address size in bits.
18622 @item set serial baud @var{n}
18623 @cindex baud rate for remote targets
18624 Set the baud rate for the remote serial I/O to @var{n} baud. The
18625 value is used to set the speed of the serial port used for debugging
18628 @item show serial baud
18629 Show the current speed of the remote connection.
18631 @item set remotebreak
18632 @cindex interrupt remote programs
18633 @cindex BREAK signal instead of Ctrl-C
18634 @anchor{set remotebreak}
18635 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
18636 when you type @kbd{Ctrl-c} to interrupt the program running
18637 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
18638 character instead. The default is off, since most remote systems
18639 expect to see @samp{Ctrl-C} as the interrupt signal.
18641 @item show remotebreak
18642 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
18643 interrupt the remote program.
18645 @item set remoteflow on
18646 @itemx set remoteflow off
18647 @kindex set remoteflow
18648 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
18649 on the serial port used to communicate to the remote target.
18651 @item show remoteflow
18652 @kindex show remoteflow
18653 Show the current setting of hardware flow control.
18655 @item set remotelogbase @var{base}
18656 Set the base (a.k.a.@: radix) of logging serial protocol
18657 communications to @var{base}. Supported values of @var{base} are:
18658 @code{ascii}, @code{octal}, and @code{hex}. The default is
18661 @item show remotelogbase
18662 Show the current setting of the radix for logging remote serial
18665 @item set remotelogfile @var{file}
18666 @cindex record serial communications on file
18667 Record remote serial communications on the named @var{file}. The
18668 default is not to record at all.
18670 @item show remotelogfile.
18671 Show the current setting of the file name on which to record the
18672 serial communications.
18674 @item set remotetimeout @var{num}
18675 @cindex timeout for serial communications
18676 @cindex remote timeout
18677 Set the timeout limit to wait for the remote target to respond to
18678 @var{num} seconds. The default is 2 seconds.
18680 @item show remotetimeout
18681 Show the current number of seconds to wait for the remote target
18684 @cindex limit hardware breakpoints and watchpoints
18685 @cindex remote target, limit break- and watchpoints
18686 @anchor{set remote hardware-watchpoint-limit}
18687 @anchor{set remote hardware-breakpoint-limit}
18688 @item set remote hardware-watchpoint-limit @var{limit}
18689 @itemx set remote hardware-breakpoint-limit @var{limit}
18690 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
18691 watchpoints. A limit of -1, the default, is treated as unlimited.
18693 @cindex limit hardware watchpoints length
18694 @cindex remote target, limit watchpoints length
18695 @anchor{set remote hardware-watchpoint-length-limit}
18696 @item set remote hardware-watchpoint-length-limit @var{limit}
18697 Restrict @value{GDBN} to using @var{limit} bytes for the maximum length of
18698 a remote hardware watchpoint. A limit of -1, the default, is treated
18701 @item show remote hardware-watchpoint-length-limit
18702 Show the current limit (in bytes) of the maximum length of
18703 a remote hardware watchpoint.
18705 @item set remote exec-file @var{filename}
18706 @itemx show remote exec-file
18707 @anchor{set remote exec-file}
18708 @cindex executable file, for remote target
18709 Select the file used for @code{run} with @code{target
18710 extended-remote}. This should be set to a filename valid on the
18711 target system. If it is not set, the target will use a default
18712 filename (e.g.@: the last program run).
18714 @item set remote interrupt-sequence
18715 @cindex interrupt remote programs
18716 @cindex select Ctrl-C, BREAK or BREAK-g
18717 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
18718 @samp{BREAK-g} as the
18719 sequence to the remote target in order to interrupt the execution.
18720 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
18721 is high level of serial line for some certain time.
18722 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
18723 It is @code{BREAK} signal followed by character @code{g}.
18725 @item show interrupt-sequence
18726 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
18727 is sent by @value{GDBN} to interrupt the remote program.
18728 @code{BREAK-g} is BREAK signal followed by @code{g} and
18729 also known as Magic SysRq g.
18731 @item set remote interrupt-on-connect
18732 @cindex send interrupt-sequence on start
18733 Specify whether interrupt-sequence is sent to remote target when
18734 @value{GDBN} connects to it. This is mostly needed when you debug
18735 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
18736 which is known as Magic SysRq g in order to connect @value{GDBN}.
18738 @item show interrupt-on-connect
18739 Show whether interrupt-sequence is sent
18740 to remote target when @value{GDBN} connects to it.
18744 @item set tcp auto-retry on
18745 @cindex auto-retry, for remote TCP target
18746 Enable auto-retry for remote TCP connections. This is useful if the remote
18747 debugging agent is launched in parallel with @value{GDBN}; there is a race
18748 condition because the agent may not become ready to accept the connection
18749 before @value{GDBN} attempts to connect. When auto-retry is
18750 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
18751 to establish the connection using the timeout specified by
18752 @code{set tcp connect-timeout}.
18754 @item set tcp auto-retry off
18755 Do not auto-retry failed TCP connections.
18757 @item show tcp auto-retry
18758 Show the current auto-retry setting.
18760 @item set tcp connect-timeout @var{seconds}
18761 @itemx set tcp connect-timeout unlimited
18762 @cindex connection timeout, for remote TCP target
18763 @cindex timeout, for remote target connection
18764 Set the timeout for establishing a TCP connection to the remote target to
18765 @var{seconds}. The timeout affects both polling to retry failed connections
18766 (enabled by @code{set tcp auto-retry on}) and waiting for connections
18767 that are merely slow to complete, and represents an approximate cumulative
18768 value. If @var{seconds} is @code{unlimited}, there is no timeout and
18769 @value{GDBN} will keep attempting to establish a connection forever,
18770 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
18772 @item show tcp connect-timeout
18773 Show the current connection timeout setting.
18776 @cindex remote packets, enabling and disabling
18777 The @value{GDBN} remote protocol autodetects the packets supported by
18778 your debugging stub. If you need to override the autodetection, you
18779 can use these commands to enable or disable individual packets. Each
18780 packet can be set to @samp{on} (the remote target supports this
18781 packet), @samp{off} (the remote target does not support this packet),
18782 or @samp{auto} (detect remote target support for this packet). They
18783 all default to @samp{auto}. For more information about each packet,
18784 see @ref{Remote Protocol}.
18786 During normal use, you should not have to use any of these commands.
18787 If you do, that may be a bug in your remote debugging stub, or a bug
18788 in @value{GDBN}. You may want to report the problem to the
18789 @value{GDBN} developers.
18791 For each packet @var{name}, the command to enable or disable the
18792 packet is @code{set remote @var{name}-packet}. The available settings
18795 @multitable @columnfractions 0.28 0.32 0.25
18798 @tab Related Features
18800 @item @code{fetch-register}
18802 @tab @code{info registers}
18804 @item @code{set-register}
18808 @item @code{binary-download}
18810 @tab @code{load}, @code{set}
18812 @item @code{read-aux-vector}
18813 @tab @code{qXfer:auxv:read}
18814 @tab @code{info auxv}
18816 @item @code{symbol-lookup}
18817 @tab @code{qSymbol}
18818 @tab Detecting multiple threads
18820 @item @code{attach}
18821 @tab @code{vAttach}
18824 @item @code{verbose-resume}
18826 @tab Stepping or resuming multiple threads
18832 @item @code{software-breakpoint}
18836 @item @code{hardware-breakpoint}
18840 @item @code{write-watchpoint}
18844 @item @code{read-watchpoint}
18848 @item @code{access-watchpoint}
18852 @item @code{target-features}
18853 @tab @code{qXfer:features:read}
18854 @tab @code{set architecture}
18856 @item @code{library-info}
18857 @tab @code{qXfer:libraries:read}
18858 @tab @code{info sharedlibrary}
18860 @item @code{memory-map}
18861 @tab @code{qXfer:memory-map:read}
18862 @tab @code{info mem}
18864 @item @code{read-sdata-object}
18865 @tab @code{qXfer:sdata:read}
18866 @tab @code{print $_sdata}
18868 @item @code{read-spu-object}
18869 @tab @code{qXfer:spu:read}
18870 @tab @code{info spu}
18872 @item @code{write-spu-object}
18873 @tab @code{qXfer:spu:write}
18874 @tab @code{info spu}
18876 @item @code{read-siginfo-object}
18877 @tab @code{qXfer:siginfo:read}
18878 @tab @code{print $_siginfo}
18880 @item @code{write-siginfo-object}
18881 @tab @code{qXfer:siginfo:write}
18882 @tab @code{set $_siginfo}
18884 @item @code{threads}
18885 @tab @code{qXfer:threads:read}
18886 @tab @code{info threads}
18888 @item @code{get-thread-local-@*storage-address}
18889 @tab @code{qGetTLSAddr}
18890 @tab Displaying @code{__thread} variables
18892 @item @code{get-thread-information-block-address}
18893 @tab @code{qGetTIBAddr}
18894 @tab Display MS-Windows Thread Information Block.
18896 @item @code{search-memory}
18897 @tab @code{qSearch:memory}
18900 @item @code{supported-packets}
18901 @tab @code{qSupported}
18902 @tab Remote communications parameters
18904 @item @code{pass-signals}
18905 @tab @code{QPassSignals}
18906 @tab @code{handle @var{signal}}
18908 @item @code{program-signals}
18909 @tab @code{QProgramSignals}
18910 @tab @code{handle @var{signal}}
18912 @item @code{hostio-close-packet}
18913 @tab @code{vFile:close}
18914 @tab @code{remote get}, @code{remote put}
18916 @item @code{hostio-open-packet}
18917 @tab @code{vFile:open}
18918 @tab @code{remote get}, @code{remote put}
18920 @item @code{hostio-pread-packet}
18921 @tab @code{vFile:pread}
18922 @tab @code{remote get}, @code{remote put}
18924 @item @code{hostio-pwrite-packet}
18925 @tab @code{vFile:pwrite}
18926 @tab @code{remote get}, @code{remote put}
18928 @item @code{hostio-unlink-packet}
18929 @tab @code{vFile:unlink}
18930 @tab @code{remote delete}
18932 @item @code{hostio-readlink-packet}
18933 @tab @code{vFile:readlink}
18936 @item @code{noack-packet}
18937 @tab @code{QStartNoAckMode}
18938 @tab Packet acknowledgment
18940 @item @code{osdata}
18941 @tab @code{qXfer:osdata:read}
18942 @tab @code{info os}
18944 @item @code{query-attached}
18945 @tab @code{qAttached}
18946 @tab Querying remote process attach state.
18948 @item @code{trace-buffer-size}
18949 @tab @code{QTBuffer:size}
18950 @tab @code{set trace-buffer-size}
18952 @item @code{trace-status}
18953 @tab @code{qTStatus}
18954 @tab @code{tstatus}
18956 @item @code{traceframe-info}
18957 @tab @code{qXfer:traceframe-info:read}
18958 @tab Traceframe info
18960 @item @code{install-in-trace}
18961 @tab @code{InstallInTrace}
18962 @tab Install tracepoint in tracing
18964 @item @code{disable-randomization}
18965 @tab @code{QDisableRandomization}
18966 @tab @code{set disable-randomization}
18968 @item @code{conditional-breakpoints-packet}
18969 @tab @code{Z0 and Z1}
18970 @tab @code{Support for target-side breakpoint condition evaluation}
18974 @section Implementing a Remote Stub
18976 @cindex debugging stub, example
18977 @cindex remote stub, example
18978 @cindex stub example, remote debugging
18979 The stub files provided with @value{GDBN} implement the target side of the
18980 communication protocol, and the @value{GDBN} side is implemented in the
18981 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
18982 these subroutines to communicate, and ignore the details. (If you're
18983 implementing your own stub file, you can still ignore the details: start
18984 with one of the existing stub files. @file{sparc-stub.c} is the best
18985 organized, and therefore the easiest to read.)
18987 @cindex remote serial debugging, overview
18988 To debug a program running on another machine (the debugging
18989 @dfn{target} machine), you must first arrange for all the usual
18990 prerequisites for the program to run by itself. For example, for a C
18995 A startup routine to set up the C runtime environment; these usually
18996 have a name like @file{crt0}. The startup routine may be supplied by
18997 your hardware supplier, or you may have to write your own.
19000 A C subroutine library to support your program's
19001 subroutine calls, notably managing input and output.
19004 A way of getting your program to the other machine---for example, a
19005 download program. These are often supplied by the hardware
19006 manufacturer, but you may have to write your own from hardware
19010 The next step is to arrange for your program to use a serial port to
19011 communicate with the machine where @value{GDBN} is running (the @dfn{host}
19012 machine). In general terms, the scheme looks like this:
19016 @value{GDBN} already understands how to use this protocol; when everything
19017 else is set up, you can simply use the @samp{target remote} command
19018 (@pxref{Targets,,Specifying a Debugging Target}).
19020 @item On the target,
19021 you must link with your program a few special-purpose subroutines that
19022 implement the @value{GDBN} remote serial protocol. The file containing these
19023 subroutines is called a @dfn{debugging stub}.
19025 On certain remote targets, you can use an auxiliary program
19026 @code{gdbserver} instead of linking a stub into your program.
19027 @xref{Server,,Using the @code{gdbserver} Program}, for details.
19030 The debugging stub is specific to the architecture of the remote
19031 machine; for example, use @file{sparc-stub.c} to debug programs on
19034 @cindex remote serial stub list
19035 These working remote stubs are distributed with @value{GDBN}:
19040 @cindex @file{i386-stub.c}
19043 For Intel 386 and compatible architectures.
19046 @cindex @file{m68k-stub.c}
19047 @cindex Motorola 680x0
19049 For Motorola 680x0 architectures.
19052 @cindex @file{sh-stub.c}
19055 For Renesas SH architectures.
19058 @cindex @file{sparc-stub.c}
19060 For @sc{sparc} architectures.
19062 @item sparcl-stub.c
19063 @cindex @file{sparcl-stub.c}
19066 For Fujitsu @sc{sparclite} architectures.
19070 The @file{README} file in the @value{GDBN} distribution may list other
19071 recently added stubs.
19074 * Stub Contents:: What the stub can do for you
19075 * Bootstrapping:: What you must do for the stub
19076 * Debug Session:: Putting it all together
19079 @node Stub Contents
19080 @subsection What the Stub Can Do for You
19082 @cindex remote serial stub
19083 The debugging stub for your architecture supplies these three
19087 @item set_debug_traps
19088 @findex set_debug_traps
19089 @cindex remote serial stub, initialization
19090 This routine arranges for @code{handle_exception} to run when your
19091 program stops. You must call this subroutine explicitly in your
19092 program's startup code.
19094 @item handle_exception
19095 @findex handle_exception
19096 @cindex remote serial stub, main routine
19097 This is the central workhorse, but your program never calls it
19098 explicitly---the setup code arranges for @code{handle_exception} to
19099 run when a trap is triggered.
19101 @code{handle_exception} takes control when your program stops during
19102 execution (for example, on a breakpoint), and mediates communications
19103 with @value{GDBN} on the host machine. This is where the communications
19104 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
19105 representative on the target machine. It begins by sending summary
19106 information on the state of your program, then continues to execute,
19107 retrieving and transmitting any information @value{GDBN} needs, until you
19108 execute a @value{GDBN} command that makes your program resume; at that point,
19109 @code{handle_exception} returns control to your own code on the target
19113 @cindex @code{breakpoint} subroutine, remote
19114 Use this auxiliary subroutine to make your program contain a
19115 breakpoint. Depending on the particular situation, this may be the only
19116 way for @value{GDBN} to get control. For instance, if your target
19117 machine has some sort of interrupt button, you won't need to call this;
19118 pressing the interrupt button transfers control to
19119 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
19120 simply receiving characters on the serial port may also trigger a trap;
19121 again, in that situation, you don't need to call @code{breakpoint} from
19122 your own program---simply running @samp{target remote} from the host
19123 @value{GDBN} session gets control.
19125 Call @code{breakpoint} if none of these is true, or if you simply want
19126 to make certain your program stops at a predetermined point for the
19127 start of your debugging session.
19130 @node Bootstrapping
19131 @subsection What You Must Do for the Stub
19133 @cindex remote stub, support routines
19134 The debugging stubs that come with @value{GDBN} are set up for a particular
19135 chip architecture, but they have no information about the rest of your
19136 debugging target machine.
19138 First of all you need to tell the stub how to communicate with the
19142 @item int getDebugChar()
19143 @findex getDebugChar
19144 Write this subroutine to read a single character from the serial port.
19145 It may be identical to @code{getchar} for your target system; a
19146 different name is used to allow you to distinguish the two if you wish.
19148 @item void putDebugChar(int)
19149 @findex putDebugChar
19150 Write this subroutine to write a single character to the serial port.
19151 It may be identical to @code{putchar} for your target system; a
19152 different name is used to allow you to distinguish the two if you wish.
19155 @cindex control C, and remote debugging
19156 @cindex interrupting remote targets
19157 If you want @value{GDBN} to be able to stop your program while it is
19158 running, you need to use an interrupt-driven serial driver, and arrange
19159 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
19160 character). That is the character which @value{GDBN} uses to tell the
19161 remote system to stop.
19163 Getting the debugging target to return the proper status to @value{GDBN}
19164 probably requires changes to the standard stub; one quick and dirty way
19165 is to just execute a breakpoint instruction (the ``dirty'' part is that
19166 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
19168 Other routines you need to supply are:
19171 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
19172 @findex exceptionHandler
19173 Write this function to install @var{exception_address} in the exception
19174 handling tables. You need to do this because the stub does not have any
19175 way of knowing what the exception handling tables on your target system
19176 are like (for example, the processor's table might be in @sc{rom},
19177 containing entries which point to a table in @sc{ram}).
19178 @var{exception_number} is the exception number which should be changed;
19179 its meaning is architecture-dependent (for example, different numbers
19180 might represent divide by zero, misaligned access, etc). When this
19181 exception occurs, control should be transferred directly to
19182 @var{exception_address}, and the processor state (stack, registers,
19183 and so on) should be just as it is when a processor exception occurs. So if
19184 you want to use a jump instruction to reach @var{exception_address}, it
19185 should be a simple jump, not a jump to subroutine.
19187 For the 386, @var{exception_address} should be installed as an interrupt
19188 gate so that interrupts are masked while the handler runs. The gate
19189 should be at privilege level 0 (the most privileged level). The
19190 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
19191 help from @code{exceptionHandler}.
19193 @item void flush_i_cache()
19194 @findex flush_i_cache
19195 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
19196 instruction cache, if any, on your target machine. If there is no
19197 instruction cache, this subroutine may be a no-op.
19199 On target machines that have instruction caches, @value{GDBN} requires this
19200 function to make certain that the state of your program is stable.
19204 You must also make sure this library routine is available:
19207 @item void *memset(void *, int, int)
19209 This is the standard library function @code{memset} that sets an area of
19210 memory to a known value. If you have one of the free versions of
19211 @code{libc.a}, @code{memset} can be found there; otherwise, you must
19212 either obtain it from your hardware manufacturer, or write your own.
19215 If you do not use the GNU C compiler, you may need other standard
19216 library subroutines as well; this varies from one stub to another,
19217 but in general the stubs are likely to use any of the common library
19218 subroutines which @code{@value{NGCC}} generates as inline code.
19221 @node Debug Session
19222 @subsection Putting it All Together
19224 @cindex remote serial debugging summary
19225 In summary, when your program is ready to debug, you must follow these
19230 Make sure you have defined the supporting low-level routines
19231 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
19233 @code{getDebugChar}, @code{putDebugChar},
19234 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
19238 Insert these lines in your program's startup code, before the main
19239 procedure is called:
19246 On some machines, when a breakpoint trap is raised, the hardware
19247 automatically makes the PC point to the instruction after the
19248 breakpoint. If your machine doesn't do that, you may need to adjust
19249 @code{handle_exception} to arrange for it to return to the instruction
19250 after the breakpoint on this first invocation, so that your program
19251 doesn't keep hitting the initial breakpoint instead of making
19255 For the 680x0 stub only, you need to provide a variable called
19256 @code{exceptionHook}. Normally you just use:
19259 void (*exceptionHook)() = 0;
19263 but if before calling @code{set_debug_traps}, you set it to point to a
19264 function in your program, that function is called when
19265 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
19266 error). The function indicated by @code{exceptionHook} is called with
19267 one parameter: an @code{int} which is the exception number.
19270 Compile and link together: your program, the @value{GDBN} debugging stub for
19271 your target architecture, and the supporting subroutines.
19274 Make sure you have a serial connection between your target machine and
19275 the @value{GDBN} host, and identify the serial port on the host.
19278 @c The "remote" target now provides a `load' command, so we should
19279 @c document that. FIXME.
19280 Download your program to your target machine (or get it there by
19281 whatever means the manufacturer provides), and start it.
19284 Start @value{GDBN} on the host, and connect to the target
19285 (@pxref{Connecting,,Connecting to a Remote Target}).
19289 @node Configurations
19290 @chapter Configuration-Specific Information
19292 While nearly all @value{GDBN} commands are available for all native and
19293 cross versions of the debugger, there are some exceptions. This chapter
19294 describes things that are only available in certain configurations.
19296 There are three major categories of configurations: native
19297 configurations, where the host and target are the same, embedded
19298 operating system configurations, which are usually the same for several
19299 different processor architectures, and bare embedded processors, which
19300 are quite different from each other.
19305 * Embedded Processors::
19312 This section describes details specific to particular native
19317 * BSD libkvm Interface:: Debugging BSD kernel memory images
19318 * SVR4 Process Information:: SVR4 process information
19319 * DJGPP Native:: Features specific to the DJGPP port
19320 * Cygwin Native:: Features specific to the Cygwin port
19321 * Hurd Native:: Features specific to @sc{gnu} Hurd
19322 * Darwin:: Features specific to Darwin
19328 On HP-UX systems, if you refer to a function or variable name that
19329 begins with a dollar sign, @value{GDBN} searches for a user or system
19330 name first, before it searches for a convenience variable.
19333 @node BSD libkvm Interface
19334 @subsection BSD libkvm Interface
19337 @cindex kernel memory image
19338 @cindex kernel crash dump
19340 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
19341 interface that provides a uniform interface for accessing kernel virtual
19342 memory images, including live systems and crash dumps. @value{GDBN}
19343 uses this interface to allow you to debug live kernels and kernel crash
19344 dumps on many native BSD configurations. This is implemented as a
19345 special @code{kvm} debugging target. For debugging a live system, load
19346 the currently running kernel into @value{GDBN} and connect to the
19350 (@value{GDBP}) @b{target kvm}
19353 For debugging crash dumps, provide the file name of the crash dump as an
19357 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
19360 Once connected to the @code{kvm} target, the following commands are
19366 Set current context from the @dfn{Process Control Block} (PCB) address.
19369 Set current context from proc address. This command isn't available on
19370 modern FreeBSD systems.
19373 @node SVR4 Process Information
19374 @subsection SVR4 Process Information
19376 @cindex examine process image
19377 @cindex process info via @file{/proc}
19379 Many versions of SVR4 and compatible systems provide a facility called
19380 @samp{/proc} that can be used to examine the image of a running
19381 process using file-system subroutines.
19383 If @value{GDBN} is configured for an operating system with this
19384 facility, the command @code{info proc} is available to report
19385 information about the process running your program, or about any
19386 process running on your system. This includes, as of this writing,
19387 @sc{gnu}/Linux, OSF/1 (Digital Unix), Solaris, and Irix, but
19388 not HP-UX, for example.
19390 This command may also work on core files that were created on a system
19391 that has the @samp{/proc} facility.
19397 @itemx info proc @var{process-id}
19398 Summarize available information about any running process. If a
19399 process ID is specified by @var{process-id}, display information about
19400 that process; otherwise display information about the program being
19401 debugged. The summary includes the debugged process ID, the command
19402 line used to invoke it, its current working directory, and its
19403 executable file's absolute file name.
19405 On some systems, @var{process-id} can be of the form
19406 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
19407 within a process. If the optional @var{pid} part is missing, it means
19408 a thread from the process being debugged (the leading @samp{/} still
19409 needs to be present, or else @value{GDBN} will interpret the number as
19410 a process ID rather than a thread ID).
19412 @item info proc cmdline
19413 @cindex info proc cmdline
19414 Show the original command line of the process. This command is
19415 specific to @sc{gnu}/Linux.
19417 @item info proc cwd
19418 @cindex info proc cwd
19419 Show the current working directory of the process. This command is
19420 specific to @sc{gnu}/Linux.
19422 @item info proc exe
19423 @cindex info proc exe
19424 Show the name of executable of the process. This command is specific
19427 @item info proc mappings
19428 @cindex memory address space mappings
19429 Report the memory address space ranges accessible in the program, with
19430 information on whether the process has read, write, or execute access
19431 rights to each range. On @sc{gnu}/Linux systems, each memory range
19432 includes the object file which is mapped to that range, instead of the
19433 memory access rights to that range.
19435 @item info proc stat
19436 @itemx info proc status
19437 @cindex process detailed status information
19438 These subcommands are specific to @sc{gnu}/Linux systems. They show
19439 the process-related information, including the user ID and group ID;
19440 how many threads are there in the process; its virtual memory usage;
19441 the signals that are pending, blocked, and ignored; its TTY; its
19442 consumption of system and user time; its stack size; its @samp{nice}
19443 value; etc. For more information, see the @samp{proc} man page
19444 (type @kbd{man 5 proc} from your shell prompt).
19446 @item info proc all
19447 Show all the information about the process described under all of the
19448 above @code{info proc} subcommands.
19451 @comment These sub-options of 'info proc' were not included when
19452 @comment procfs.c was re-written. Keep their descriptions around
19453 @comment against the day when someone finds the time to put them back in.
19454 @kindex info proc times
19455 @item info proc times
19456 Starting time, user CPU time, and system CPU time for your program and
19459 @kindex info proc id
19461 Report on the process IDs related to your program: its own process ID,
19462 the ID of its parent, the process group ID, and the session ID.
19465 @item set procfs-trace
19466 @kindex set procfs-trace
19467 @cindex @code{procfs} API calls
19468 This command enables and disables tracing of @code{procfs} API calls.
19470 @item show procfs-trace
19471 @kindex show procfs-trace
19472 Show the current state of @code{procfs} API call tracing.
19474 @item set procfs-file @var{file}
19475 @kindex set procfs-file
19476 Tell @value{GDBN} to write @code{procfs} API trace to the named
19477 @var{file}. @value{GDBN} appends the trace info to the previous
19478 contents of the file. The default is to display the trace on the
19481 @item show procfs-file
19482 @kindex show procfs-file
19483 Show the file to which @code{procfs} API trace is written.
19485 @item proc-trace-entry
19486 @itemx proc-trace-exit
19487 @itemx proc-untrace-entry
19488 @itemx proc-untrace-exit
19489 @kindex proc-trace-entry
19490 @kindex proc-trace-exit
19491 @kindex proc-untrace-entry
19492 @kindex proc-untrace-exit
19493 These commands enable and disable tracing of entries into and exits
19494 from the @code{syscall} interface.
19497 @kindex info pidlist
19498 @cindex process list, QNX Neutrino
19499 For QNX Neutrino only, this command displays the list of all the
19500 processes and all the threads within each process.
19503 @kindex info meminfo
19504 @cindex mapinfo list, QNX Neutrino
19505 For QNX Neutrino only, this command displays the list of all mapinfos.
19509 @subsection Features for Debugging @sc{djgpp} Programs
19510 @cindex @sc{djgpp} debugging
19511 @cindex native @sc{djgpp} debugging
19512 @cindex MS-DOS-specific commands
19515 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
19516 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
19517 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
19518 top of real-mode DOS systems and their emulations.
19520 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
19521 defines a few commands specific to the @sc{djgpp} port. This
19522 subsection describes those commands.
19527 This is a prefix of @sc{djgpp}-specific commands which print
19528 information about the target system and important OS structures.
19531 @cindex MS-DOS system info
19532 @cindex free memory information (MS-DOS)
19533 @item info dos sysinfo
19534 This command displays assorted information about the underlying
19535 platform: the CPU type and features, the OS version and flavor, the
19536 DPMI version, and the available conventional and DPMI memory.
19541 @cindex segment descriptor tables
19542 @cindex descriptor tables display
19544 @itemx info dos ldt
19545 @itemx info dos idt
19546 These 3 commands display entries from, respectively, Global, Local,
19547 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
19548 tables are data structures which store a descriptor for each segment
19549 that is currently in use. The segment's selector is an index into a
19550 descriptor table; the table entry for that index holds the
19551 descriptor's base address and limit, and its attributes and access
19554 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
19555 segment (used for both data and the stack), and a DOS segment (which
19556 allows access to DOS/BIOS data structures and absolute addresses in
19557 conventional memory). However, the DPMI host will usually define
19558 additional segments in order to support the DPMI environment.
19560 @cindex garbled pointers
19561 These commands allow to display entries from the descriptor tables.
19562 Without an argument, all entries from the specified table are
19563 displayed. An argument, which should be an integer expression, means
19564 display a single entry whose index is given by the argument. For
19565 example, here's a convenient way to display information about the
19566 debugged program's data segment:
19569 @exdent @code{(@value{GDBP}) info dos ldt $ds}
19570 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
19574 This comes in handy when you want to see whether a pointer is outside
19575 the data segment's limit (i.e.@: @dfn{garbled}).
19577 @cindex page tables display (MS-DOS)
19579 @itemx info dos pte
19580 These two commands display entries from, respectively, the Page
19581 Directory and the Page Tables. Page Directories and Page Tables are
19582 data structures which control how virtual memory addresses are mapped
19583 into physical addresses. A Page Table includes an entry for every
19584 page of memory that is mapped into the program's address space; there
19585 may be several Page Tables, each one holding up to 4096 entries. A
19586 Page Directory has up to 4096 entries, one each for every Page Table
19587 that is currently in use.
19589 Without an argument, @kbd{info dos pde} displays the entire Page
19590 Directory, and @kbd{info dos pte} displays all the entries in all of
19591 the Page Tables. An argument, an integer expression, given to the
19592 @kbd{info dos pde} command means display only that entry from the Page
19593 Directory table. An argument given to the @kbd{info dos pte} command
19594 means display entries from a single Page Table, the one pointed to by
19595 the specified entry in the Page Directory.
19597 @cindex direct memory access (DMA) on MS-DOS
19598 These commands are useful when your program uses @dfn{DMA} (Direct
19599 Memory Access), which needs physical addresses to program the DMA
19602 These commands are supported only with some DPMI servers.
19604 @cindex physical address from linear address
19605 @item info dos address-pte @var{addr}
19606 This command displays the Page Table entry for a specified linear
19607 address. The argument @var{addr} is a linear address which should
19608 already have the appropriate segment's base address added to it,
19609 because this command accepts addresses which may belong to @emph{any}
19610 segment. For example, here's how to display the Page Table entry for
19611 the page where a variable @code{i} is stored:
19614 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
19615 @exdent @code{Page Table entry for address 0x11a00d30:}
19616 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
19620 This says that @code{i} is stored at offset @code{0xd30} from the page
19621 whose physical base address is @code{0x02698000}, and shows all the
19622 attributes of that page.
19624 Note that you must cast the addresses of variables to a @code{char *},
19625 since otherwise the value of @code{__djgpp_base_address}, the base
19626 address of all variables and functions in a @sc{djgpp} program, will
19627 be added using the rules of C pointer arithmetics: if @code{i} is
19628 declared an @code{int}, @value{GDBN} will add 4 times the value of
19629 @code{__djgpp_base_address} to the address of @code{i}.
19631 Here's another example, it displays the Page Table entry for the
19635 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
19636 @exdent @code{Page Table entry for address 0x29110:}
19637 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
19641 (The @code{+ 3} offset is because the transfer buffer's address is the
19642 3rd member of the @code{_go32_info_block} structure.) The output
19643 clearly shows that this DPMI server maps the addresses in conventional
19644 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
19645 linear (@code{0x29110}) addresses are identical.
19647 This command is supported only with some DPMI servers.
19650 @cindex DOS serial data link, remote debugging
19651 In addition to native debugging, the DJGPP port supports remote
19652 debugging via a serial data link. The following commands are specific
19653 to remote serial debugging in the DJGPP port of @value{GDBN}.
19656 @kindex set com1base
19657 @kindex set com1irq
19658 @kindex set com2base
19659 @kindex set com2irq
19660 @kindex set com3base
19661 @kindex set com3irq
19662 @kindex set com4base
19663 @kindex set com4irq
19664 @item set com1base @var{addr}
19665 This command sets the base I/O port address of the @file{COM1} serial
19668 @item set com1irq @var{irq}
19669 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
19670 for the @file{COM1} serial port.
19672 There are similar commands @samp{set com2base}, @samp{set com3irq},
19673 etc.@: for setting the port address and the @code{IRQ} lines for the
19676 @kindex show com1base
19677 @kindex show com1irq
19678 @kindex show com2base
19679 @kindex show com2irq
19680 @kindex show com3base
19681 @kindex show com3irq
19682 @kindex show com4base
19683 @kindex show com4irq
19684 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
19685 display the current settings of the base address and the @code{IRQ}
19686 lines used by the COM ports.
19689 @kindex info serial
19690 @cindex DOS serial port status
19691 This command prints the status of the 4 DOS serial ports. For each
19692 port, it prints whether it's active or not, its I/O base address and
19693 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
19694 counts of various errors encountered so far.
19698 @node Cygwin Native
19699 @subsection Features for Debugging MS Windows PE Executables
19700 @cindex MS Windows debugging
19701 @cindex native Cygwin debugging
19702 @cindex Cygwin-specific commands
19704 @value{GDBN} supports native debugging of MS Windows programs, including
19705 DLLs with and without symbolic debugging information.
19707 @cindex Ctrl-BREAK, MS-Windows
19708 @cindex interrupt debuggee on MS-Windows
19709 MS-Windows programs that call @code{SetConsoleMode} to switch off the
19710 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
19711 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
19712 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
19713 sequence, which can be used to interrupt the debuggee even if it
19716 There are various additional Cygwin-specific commands, described in
19717 this section. Working with DLLs that have no debugging symbols is
19718 described in @ref{Non-debug DLL Symbols}.
19723 This is a prefix of MS Windows-specific commands which print
19724 information about the target system and important OS structures.
19726 @item info w32 selector
19727 This command displays information returned by
19728 the Win32 API @code{GetThreadSelectorEntry} function.
19729 It takes an optional argument that is evaluated to
19730 a long value to give the information about this given selector.
19731 Without argument, this command displays information
19732 about the six segment registers.
19734 @item info w32 thread-information-block
19735 This command displays thread specific information stored in the
19736 Thread Information Block (readable on the X86 CPU family using @code{$fs}
19737 selector for 32-bit programs and @code{$gs} for 64-bit programs).
19741 This is a Cygwin-specific alias of @code{info shared}.
19743 @kindex dll-symbols
19745 This command loads symbols from a dll similarly to
19746 add-sym command but without the need to specify a base address.
19748 @kindex set cygwin-exceptions
19749 @cindex debugging the Cygwin DLL
19750 @cindex Cygwin DLL, debugging
19751 @item set cygwin-exceptions @var{mode}
19752 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
19753 happen inside the Cygwin DLL. If @var{mode} is @code{off},
19754 @value{GDBN} will delay recognition of exceptions, and may ignore some
19755 exceptions which seem to be caused by internal Cygwin DLL
19756 ``bookkeeping''. This option is meant primarily for debugging the
19757 Cygwin DLL itself; the default value is @code{off} to avoid annoying
19758 @value{GDBN} users with false @code{SIGSEGV} signals.
19760 @kindex show cygwin-exceptions
19761 @item show cygwin-exceptions
19762 Displays whether @value{GDBN} will break on exceptions that happen
19763 inside the Cygwin DLL itself.
19765 @kindex set new-console
19766 @item set new-console @var{mode}
19767 If @var{mode} is @code{on} the debuggee will
19768 be started in a new console on next start.
19769 If @var{mode} is @code{off}, the debuggee will
19770 be started in the same console as the debugger.
19772 @kindex show new-console
19773 @item show new-console
19774 Displays whether a new console is used
19775 when the debuggee is started.
19777 @kindex set new-group
19778 @item set new-group @var{mode}
19779 This boolean value controls whether the debuggee should
19780 start a new group or stay in the same group as the debugger.
19781 This affects the way the Windows OS handles
19784 @kindex show new-group
19785 @item show new-group
19786 Displays current value of new-group boolean.
19788 @kindex set debugevents
19789 @item set debugevents
19790 This boolean value adds debug output concerning kernel events related
19791 to the debuggee seen by the debugger. This includes events that
19792 signal thread and process creation and exit, DLL loading and
19793 unloading, console interrupts, and debugging messages produced by the
19794 Windows @code{OutputDebugString} API call.
19796 @kindex set debugexec
19797 @item set debugexec
19798 This boolean value adds debug output concerning execute events
19799 (such as resume thread) seen by the debugger.
19801 @kindex set debugexceptions
19802 @item set debugexceptions
19803 This boolean value adds debug output concerning exceptions in the
19804 debuggee seen by the debugger.
19806 @kindex set debugmemory
19807 @item set debugmemory
19808 This boolean value adds debug output concerning debuggee memory reads
19809 and writes by the debugger.
19813 This boolean values specifies whether the debuggee is called
19814 via a shell or directly (default value is on).
19818 Displays if the debuggee will be started with a shell.
19823 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
19826 @node Non-debug DLL Symbols
19827 @subsubsection Support for DLLs without Debugging Symbols
19828 @cindex DLLs with no debugging symbols
19829 @cindex Minimal symbols and DLLs
19831 Very often on windows, some of the DLLs that your program relies on do
19832 not include symbolic debugging information (for example,
19833 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
19834 symbols in a DLL, it relies on the minimal amount of symbolic
19835 information contained in the DLL's export table. This section
19836 describes working with such symbols, known internally to @value{GDBN} as
19837 ``minimal symbols''.
19839 Note that before the debugged program has started execution, no DLLs
19840 will have been loaded. The easiest way around this problem is simply to
19841 start the program --- either by setting a breakpoint or letting the
19842 program run once to completion. It is also possible to force
19843 @value{GDBN} to load a particular DLL before starting the executable ---
19844 see the shared library information in @ref{Files}, or the
19845 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
19846 explicitly loading symbols from a DLL with no debugging information will
19847 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
19848 which may adversely affect symbol lookup performance.
19850 @subsubsection DLL Name Prefixes
19852 In keeping with the naming conventions used by the Microsoft debugging
19853 tools, DLL export symbols are made available with a prefix based on the
19854 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
19855 also entered into the symbol table, so @code{CreateFileA} is often
19856 sufficient. In some cases there will be name clashes within a program
19857 (particularly if the executable itself includes full debugging symbols)
19858 necessitating the use of the fully qualified name when referring to the
19859 contents of the DLL. Use single-quotes around the name to avoid the
19860 exclamation mark (``!'') being interpreted as a language operator.
19862 Note that the internal name of the DLL may be all upper-case, even
19863 though the file name of the DLL is lower-case, or vice-versa. Since
19864 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
19865 some confusion. If in doubt, try the @code{info functions} and
19866 @code{info variables} commands or even @code{maint print msymbols}
19867 (@pxref{Symbols}). Here's an example:
19870 (@value{GDBP}) info function CreateFileA
19871 All functions matching regular expression "CreateFileA":
19873 Non-debugging symbols:
19874 0x77e885f4 CreateFileA
19875 0x77e885f4 KERNEL32!CreateFileA
19879 (@value{GDBP}) info function !
19880 All functions matching regular expression "!":
19882 Non-debugging symbols:
19883 0x6100114c cygwin1!__assert
19884 0x61004034 cygwin1!_dll_crt0@@0
19885 0x61004240 cygwin1!dll_crt0(per_process *)
19889 @subsubsection Working with Minimal Symbols
19891 Symbols extracted from a DLL's export table do not contain very much
19892 type information. All that @value{GDBN} can do is guess whether a symbol
19893 refers to a function or variable depending on the linker section that
19894 contains the symbol. Also note that the actual contents of the memory
19895 contained in a DLL are not available unless the program is running. This
19896 means that you cannot examine the contents of a variable or disassemble
19897 a function within a DLL without a running program.
19899 Variables are generally treated as pointers and dereferenced
19900 automatically. For this reason, it is often necessary to prefix a
19901 variable name with the address-of operator (``&'') and provide explicit
19902 type information in the command. Here's an example of the type of
19906 (@value{GDBP}) print 'cygwin1!__argv'
19911 (@value{GDBP}) x 'cygwin1!__argv'
19912 0x10021610: "\230y\""
19915 And two possible solutions:
19918 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
19919 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
19923 (@value{GDBP}) x/2x &'cygwin1!__argv'
19924 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
19925 (@value{GDBP}) x/x 0x10021608
19926 0x10021608: 0x0022fd98
19927 (@value{GDBP}) x/s 0x0022fd98
19928 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
19931 Setting a break point within a DLL is possible even before the program
19932 starts execution. However, under these circumstances, @value{GDBN} can't
19933 examine the initial instructions of the function in order to skip the
19934 function's frame set-up code. You can work around this by using ``*&''
19935 to set the breakpoint at a raw memory address:
19938 (@value{GDBP}) break *&'python22!PyOS_Readline'
19939 Breakpoint 1 at 0x1e04eff0
19942 The author of these extensions is not entirely convinced that setting a
19943 break point within a shared DLL like @file{kernel32.dll} is completely
19947 @subsection Commands Specific to @sc{gnu} Hurd Systems
19948 @cindex @sc{gnu} Hurd debugging
19950 This subsection describes @value{GDBN} commands specific to the
19951 @sc{gnu} Hurd native debugging.
19956 @kindex set signals@r{, Hurd command}
19957 @kindex set sigs@r{, Hurd command}
19958 This command toggles the state of inferior signal interception by
19959 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
19960 affected by this command. @code{sigs} is a shorthand alias for
19965 @kindex show signals@r{, Hurd command}
19966 @kindex show sigs@r{, Hurd command}
19967 Show the current state of intercepting inferior's signals.
19969 @item set signal-thread
19970 @itemx set sigthread
19971 @kindex set signal-thread
19972 @kindex set sigthread
19973 This command tells @value{GDBN} which thread is the @code{libc} signal
19974 thread. That thread is run when a signal is delivered to a running
19975 process. @code{set sigthread} is the shorthand alias of @code{set
19978 @item show signal-thread
19979 @itemx show sigthread
19980 @kindex show signal-thread
19981 @kindex show sigthread
19982 These two commands show which thread will run when the inferior is
19983 delivered a signal.
19986 @kindex set stopped@r{, Hurd command}
19987 This commands tells @value{GDBN} that the inferior process is stopped,
19988 as with the @code{SIGSTOP} signal. The stopped process can be
19989 continued by delivering a signal to it.
19992 @kindex show stopped@r{, Hurd command}
19993 This command shows whether @value{GDBN} thinks the debuggee is
19996 @item set exceptions
19997 @kindex set exceptions@r{, Hurd command}
19998 Use this command to turn off trapping of exceptions in the inferior.
19999 When exception trapping is off, neither breakpoints nor
20000 single-stepping will work. To restore the default, set exception
20003 @item show exceptions
20004 @kindex show exceptions@r{, Hurd command}
20005 Show the current state of trapping exceptions in the inferior.
20007 @item set task pause
20008 @kindex set task@r{, Hurd commands}
20009 @cindex task attributes (@sc{gnu} Hurd)
20010 @cindex pause current task (@sc{gnu} Hurd)
20011 This command toggles task suspension when @value{GDBN} has control.
20012 Setting it to on takes effect immediately, and the task is suspended
20013 whenever @value{GDBN} gets control. Setting it to off will take
20014 effect the next time the inferior is continued. If this option is set
20015 to off, you can use @code{set thread default pause on} or @code{set
20016 thread pause on} (see below) to pause individual threads.
20018 @item show task pause
20019 @kindex show task@r{, Hurd commands}
20020 Show the current state of task suspension.
20022 @item set task detach-suspend-count
20023 @cindex task suspend count
20024 @cindex detach from task, @sc{gnu} Hurd
20025 This command sets the suspend count the task will be left with when
20026 @value{GDBN} detaches from it.
20028 @item show task detach-suspend-count
20029 Show the suspend count the task will be left with when detaching.
20031 @item set task exception-port
20032 @itemx set task excp
20033 @cindex task exception port, @sc{gnu} Hurd
20034 This command sets the task exception port to which @value{GDBN} will
20035 forward exceptions. The argument should be the value of the @dfn{send
20036 rights} of the task. @code{set task excp} is a shorthand alias.
20038 @item set noninvasive
20039 @cindex noninvasive task options
20040 This command switches @value{GDBN} to a mode that is the least
20041 invasive as far as interfering with the inferior is concerned. This
20042 is the same as using @code{set task pause}, @code{set exceptions}, and
20043 @code{set signals} to values opposite to the defaults.
20045 @item info send-rights
20046 @itemx info receive-rights
20047 @itemx info port-rights
20048 @itemx info port-sets
20049 @itemx info dead-names
20052 @cindex send rights, @sc{gnu} Hurd
20053 @cindex receive rights, @sc{gnu} Hurd
20054 @cindex port rights, @sc{gnu} Hurd
20055 @cindex port sets, @sc{gnu} Hurd
20056 @cindex dead names, @sc{gnu} Hurd
20057 These commands display information about, respectively, send rights,
20058 receive rights, port rights, port sets, and dead names of a task.
20059 There are also shorthand aliases: @code{info ports} for @code{info
20060 port-rights} and @code{info psets} for @code{info port-sets}.
20062 @item set thread pause
20063 @kindex set thread@r{, Hurd command}
20064 @cindex thread properties, @sc{gnu} Hurd
20065 @cindex pause current thread (@sc{gnu} Hurd)
20066 This command toggles current thread suspension when @value{GDBN} has
20067 control. Setting it to on takes effect immediately, and the current
20068 thread is suspended whenever @value{GDBN} gets control. Setting it to
20069 off will take effect the next time the inferior is continued.
20070 Normally, this command has no effect, since when @value{GDBN} has
20071 control, the whole task is suspended. However, if you used @code{set
20072 task pause off} (see above), this command comes in handy to suspend
20073 only the current thread.
20075 @item show thread pause
20076 @kindex show thread@r{, Hurd command}
20077 This command shows the state of current thread suspension.
20079 @item set thread run
20080 This command sets whether the current thread is allowed to run.
20082 @item show thread run
20083 Show whether the current thread is allowed to run.
20085 @item set thread detach-suspend-count
20086 @cindex thread suspend count, @sc{gnu} Hurd
20087 @cindex detach from thread, @sc{gnu} Hurd
20088 This command sets the suspend count @value{GDBN} will leave on a
20089 thread when detaching. This number is relative to the suspend count
20090 found by @value{GDBN} when it notices the thread; use @code{set thread
20091 takeover-suspend-count} to force it to an absolute value.
20093 @item show thread detach-suspend-count
20094 Show the suspend count @value{GDBN} will leave on the thread when
20097 @item set thread exception-port
20098 @itemx set thread excp
20099 Set the thread exception port to which to forward exceptions. This
20100 overrides the port set by @code{set task exception-port} (see above).
20101 @code{set thread excp} is the shorthand alias.
20103 @item set thread takeover-suspend-count
20104 Normally, @value{GDBN}'s thread suspend counts are relative to the
20105 value @value{GDBN} finds when it notices each thread. This command
20106 changes the suspend counts to be absolute instead.
20108 @item set thread default
20109 @itemx show thread default
20110 @cindex thread default settings, @sc{gnu} Hurd
20111 Each of the above @code{set thread} commands has a @code{set thread
20112 default} counterpart (e.g., @code{set thread default pause}, @code{set
20113 thread default exception-port}, etc.). The @code{thread default}
20114 variety of commands sets the default thread properties for all
20115 threads; you can then change the properties of individual threads with
20116 the non-default commands.
20123 @value{GDBN} provides the following commands specific to the Darwin target:
20126 @item set debug darwin @var{num}
20127 @kindex set debug darwin
20128 When set to a non zero value, enables debugging messages specific to
20129 the Darwin support. Higher values produce more verbose output.
20131 @item show debug darwin
20132 @kindex show debug darwin
20133 Show the current state of Darwin messages.
20135 @item set debug mach-o @var{num}
20136 @kindex set debug mach-o
20137 When set to a non zero value, enables debugging messages while
20138 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
20139 file format used on Darwin for object and executable files.) Higher
20140 values produce more verbose output. This is a command to diagnose
20141 problems internal to @value{GDBN} and should not be needed in normal
20144 @item show debug mach-o
20145 @kindex show debug mach-o
20146 Show the current state of Mach-O file messages.
20148 @item set mach-exceptions on
20149 @itemx set mach-exceptions off
20150 @kindex set mach-exceptions
20151 On Darwin, faults are first reported as a Mach exception and are then
20152 mapped to a Posix signal. Use this command to turn on trapping of
20153 Mach exceptions in the inferior. This might be sometimes useful to
20154 better understand the cause of a fault. The default is off.
20156 @item show mach-exceptions
20157 @kindex show mach-exceptions
20158 Show the current state of exceptions trapping.
20163 @section Embedded Operating Systems
20165 This section describes configurations involving the debugging of
20166 embedded operating systems that are available for several different
20170 * VxWorks:: Using @value{GDBN} with VxWorks
20173 @value{GDBN} includes the ability to debug programs running on
20174 various real-time operating systems.
20177 @subsection Using @value{GDBN} with VxWorks
20183 @kindex target vxworks
20184 @item target vxworks @var{machinename}
20185 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
20186 is the target system's machine name or IP address.
20190 On VxWorks, @code{load} links @var{filename} dynamically on the
20191 current target system as well as adding its symbols in @value{GDBN}.
20193 @value{GDBN} enables developers to spawn and debug tasks running on networked
20194 VxWorks targets from a Unix host. Already-running tasks spawned from
20195 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
20196 both the Unix host and on the VxWorks target. The program
20197 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
20198 installed with the name @code{vxgdb}, to distinguish it from a
20199 @value{GDBN} for debugging programs on the host itself.)
20202 @item VxWorks-timeout @var{args}
20203 @kindex vxworks-timeout
20204 All VxWorks-based targets now support the option @code{vxworks-timeout}.
20205 This option is set by the user, and @var{args} represents the number of
20206 seconds @value{GDBN} waits for responses to rpc's. You might use this if
20207 your VxWorks target is a slow software simulator or is on the far side
20208 of a thin network line.
20211 The following information on connecting to VxWorks was current when
20212 this manual was produced; newer releases of VxWorks may use revised
20215 @findex INCLUDE_RDB
20216 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
20217 to include the remote debugging interface routines in the VxWorks
20218 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
20219 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
20220 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
20221 source debugging task @code{tRdbTask} when VxWorks is booted. For more
20222 information on configuring and remaking VxWorks, see the manufacturer's
20224 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
20226 Once you have included @file{rdb.a} in your VxWorks system image and set
20227 your Unix execution search path to find @value{GDBN}, you are ready to
20228 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
20229 @code{vxgdb}, depending on your installation).
20231 @value{GDBN} comes up showing the prompt:
20238 * VxWorks Connection:: Connecting to VxWorks
20239 * VxWorks Download:: VxWorks download
20240 * VxWorks Attach:: Running tasks
20243 @node VxWorks Connection
20244 @subsubsection Connecting to VxWorks
20246 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
20247 network. To connect to a target whose host name is ``@code{tt}'', type:
20250 (vxgdb) target vxworks tt
20254 @value{GDBN} displays messages like these:
20257 Attaching remote machine across net...
20262 @value{GDBN} then attempts to read the symbol tables of any object modules
20263 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
20264 these files by searching the directories listed in the command search
20265 path (@pxref{Environment, ,Your Program's Environment}); if it fails
20266 to find an object file, it displays a message such as:
20269 prog.o: No such file or directory.
20272 When this happens, add the appropriate directory to the search path with
20273 the @value{GDBN} command @code{path}, and execute the @code{target}
20276 @node VxWorks Download
20277 @subsubsection VxWorks Download
20279 @cindex download to VxWorks
20280 If you have connected to the VxWorks target and you want to debug an
20281 object that has not yet been loaded, you can use the @value{GDBN}
20282 @code{load} command to download a file from Unix to VxWorks
20283 incrementally. The object file given as an argument to the @code{load}
20284 command is actually opened twice: first by the VxWorks target in order
20285 to download the code, then by @value{GDBN} in order to read the symbol
20286 table. This can lead to problems if the current working directories on
20287 the two systems differ. If both systems have NFS mounted the same
20288 filesystems, you can avoid these problems by using absolute paths.
20289 Otherwise, it is simplest to set the working directory on both systems
20290 to the directory in which the object file resides, and then to reference
20291 the file by its name, without any path. For instance, a program
20292 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
20293 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
20294 program, type this on VxWorks:
20297 -> cd "@var{vxpath}/vw/demo/rdb"
20301 Then, in @value{GDBN}, type:
20304 (vxgdb) cd @var{hostpath}/vw/demo/rdb
20305 (vxgdb) load prog.o
20308 @value{GDBN} displays a response similar to this:
20311 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
20314 You can also use the @code{load} command to reload an object module
20315 after editing and recompiling the corresponding source file. Note that
20316 this makes @value{GDBN} delete all currently-defined breakpoints,
20317 auto-displays, and convenience variables, and to clear the value
20318 history. (This is necessary in order to preserve the integrity of
20319 debugger's data structures that reference the target system's symbol
20322 @node VxWorks Attach
20323 @subsubsection Running Tasks
20325 @cindex running VxWorks tasks
20326 You can also attach to an existing task using the @code{attach} command as
20330 (vxgdb) attach @var{task}
20334 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
20335 or suspended when you attach to it. Running tasks are suspended at
20336 the time of attachment.
20338 @node Embedded Processors
20339 @section Embedded Processors
20341 This section goes into details specific to particular embedded
20344 @cindex send command to simulator
20345 Whenever a specific embedded processor has a simulator, @value{GDBN}
20346 allows to send an arbitrary command to the simulator.
20349 @item sim @var{command}
20350 @kindex sim@r{, a command}
20351 Send an arbitrary @var{command} string to the simulator. Consult the
20352 documentation for the specific simulator in use for information about
20353 acceptable commands.
20359 * M32R/D:: Renesas M32R/D
20360 * M68K:: Motorola M68K
20361 * MicroBlaze:: Xilinx MicroBlaze
20362 * MIPS Embedded:: MIPS Embedded
20363 * PowerPC Embedded:: PowerPC Embedded
20364 * PA:: HP PA Embedded
20365 * Sparclet:: Tsqware Sparclet
20366 * Sparclite:: Fujitsu Sparclite
20367 * Z8000:: Zilog Z8000
20370 * Super-H:: Renesas Super-H
20379 @item target rdi @var{dev}
20380 ARM Angel monitor, via RDI library interface to ADP protocol. You may
20381 use this target to communicate with both boards running the Angel
20382 monitor, or with the EmbeddedICE JTAG debug device.
20385 @item target rdp @var{dev}
20390 @value{GDBN} provides the following ARM-specific commands:
20393 @item set arm disassembler
20395 This commands selects from a list of disassembly styles. The
20396 @code{"std"} style is the standard style.
20398 @item show arm disassembler
20400 Show the current disassembly style.
20402 @item set arm apcs32
20403 @cindex ARM 32-bit mode
20404 This command toggles ARM operation mode between 32-bit and 26-bit.
20406 @item show arm apcs32
20407 Display the current usage of the ARM 32-bit mode.
20409 @item set arm fpu @var{fputype}
20410 This command sets the ARM floating-point unit (FPU) type. The
20411 argument @var{fputype} can be one of these:
20415 Determine the FPU type by querying the OS ABI.
20417 Software FPU, with mixed-endian doubles on little-endian ARM
20420 GCC-compiled FPA co-processor.
20422 Software FPU with pure-endian doubles.
20428 Show the current type of the FPU.
20431 This command forces @value{GDBN} to use the specified ABI.
20434 Show the currently used ABI.
20436 @item set arm fallback-mode (arm|thumb|auto)
20437 @value{GDBN} uses the symbol table, when available, to determine
20438 whether instructions are ARM or Thumb. This command controls
20439 @value{GDBN}'s default behavior when the symbol table is not
20440 available. The default is @samp{auto}, which causes @value{GDBN} to
20441 use the current execution mode (from the @code{T} bit in the @code{CPSR}
20444 @item show arm fallback-mode
20445 Show the current fallback instruction mode.
20447 @item set arm force-mode (arm|thumb|auto)
20448 This command overrides use of the symbol table to determine whether
20449 instructions are ARM or Thumb. The default is @samp{auto}, which
20450 causes @value{GDBN} to use the symbol table and then the setting
20451 of @samp{set arm fallback-mode}.
20453 @item show arm force-mode
20454 Show the current forced instruction mode.
20456 @item set debug arm
20457 Toggle whether to display ARM-specific debugging messages from the ARM
20458 target support subsystem.
20460 @item show debug arm
20461 Show whether ARM-specific debugging messages are enabled.
20464 The following commands are available when an ARM target is debugged
20465 using the RDI interface:
20468 @item rdilogfile @r{[}@var{file}@r{]}
20470 @cindex ADP (Angel Debugger Protocol) logging
20471 Set the filename for the ADP (Angel Debugger Protocol) packet log.
20472 With an argument, sets the log file to the specified @var{file}. With
20473 no argument, show the current log file name. The default log file is
20476 @item rdilogenable @r{[}@var{arg}@r{]}
20477 @kindex rdilogenable
20478 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
20479 enables logging, with an argument 0 or @code{"no"} disables it. With
20480 no arguments displays the current setting. When logging is enabled,
20481 ADP packets exchanged between @value{GDBN} and the RDI target device
20482 are logged to a file.
20484 @item set rdiromatzero
20485 @kindex set rdiromatzero
20486 @cindex ROM at zero address, RDI
20487 Tell @value{GDBN} whether the target has ROM at address 0. If on,
20488 vector catching is disabled, so that zero address can be used. If off
20489 (the default), vector catching is enabled. For this command to take
20490 effect, it needs to be invoked prior to the @code{target rdi} command.
20492 @item show rdiromatzero
20493 @kindex show rdiromatzero
20494 Show the current setting of ROM at zero address.
20496 @item set rdiheartbeat
20497 @kindex set rdiheartbeat
20498 @cindex RDI heartbeat
20499 Enable or disable RDI heartbeat packets. It is not recommended to
20500 turn on this option, since it confuses ARM and EPI JTAG interface, as
20501 well as the Angel monitor.
20503 @item show rdiheartbeat
20504 @kindex show rdiheartbeat
20505 Show the setting of RDI heartbeat packets.
20509 @item target sim @r{[}@var{simargs}@r{]} @dots{}
20510 The @value{GDBN} ARM simulator accepts the following optional arguments.
20513 @item --swi-support=@var{type}
20514 Tell the simulator which SWI interfaces to support.
20515 @var{type} may be a comma separated list of the following values.
20516 The default value is @code{all}.
20529 @subsection Renesas M32R/D and M32R/SDI
20532 @kindex target m32r
20533 @item target m32r @var{dev}
20534 Renesas M32R/D ROM monitor.
20536 @kindex target m32rsdi
20537 @item target m32rsdi @var{dev}
20538 Renesas M32R SDI server, connected via parallel port to the board.
20541 The following @value{GDBN} commands are specific to the M32R monitor:
20544 @item set download-path @var{path}
20545 @kindex set download-path
20546 @cindex find downloadable @sc{srec} files (M32R)
20547 Set the default path for finding downloadable @sc{srec} files.
20549 @item show download-path
20550 @kindex show download-path
20551 Show the default path for downloadable @sc{srec} files.
20553 @item set board-address @var{addr}
20554 @kindex set board-address
20555 @cindex M32-EVA target board address
20556 Set the IP address for the M32R-EVA target board.
20558 @item show board-address
20559 @kindex show board-address
20560 Show the current IP address of the target board.
20562 @item set server-address @var{addr}
20563 @kindex set server-address
20564 @cindex download server address (M32R)
20565 Set the IP address for the download server, which is the @value{GDBN}'s
20568 @item show server-address
20569 @kindex show server-address
20570 Display the IP address of the download server.
20572 @item upload @r{[}@var{file}@r{]}
20573 @kindex upload@r{, M32R}
20574 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
20575 upload capability. If no @var{file} argument is given, the current
20576 executable file is uploaded.
20578 @item tload @r{[}@var{file}@r{]}
20579 @kindex tload@r{, M32R}
20580 Test the @code{upload} command.
20583 The following commands are available for M32R/SDI:
20588 @cindex reset SDI connection, M32R
20589 This command resets the SDI connection.
20593 This command shows the SDI connection status.
20596 @kindex debug_chaos
20597 @cindex M32R/Chaos debugging
20598 Instructs the remote that M32R/Chaos debugging is to be used.
20600 @item use_debug_dma
20601 @kindex use_debug_dma
20602 Instructs the remote to use the DEBUG_DMA method of accessing memory.
20605 @kindex use_mon_code
20606 Instructs the remote to use the MON_CODE method of accessing memory.
20609 @kindex use_ib_break
20610 Instructs the remote to set breakpoints by IB break.
20612 @item use_dbt_break
20613 @kindex use_dbt_break
20614 Instructs the remote to set breakpoints by DBT.
20620 The Motorola m68k configuration includes ColdFire support, and a
20621 target command for the following ROM monitor.
20625 @kindex target dbug
20626 @item target dbug @var{dev}
20627 dBUG ROM monitor for Motorola ColdFire.
20632 @subsection MicroBlaze
20633 @cindex Xilinx MicroBlaze
20634 @cindex XMD, Xilinx Microprocessor Debugger
20636 The MicroBlaze is a soft-core processor supported on various Xilinx
20637 FPGAs, such as Spartan or Virtex series. Boards with these processors
20638 usually have JTAG ports which connect to a host system running the Xilinx
20639 Embedded Development Kit (EDK) or Software Development Kit (SDK).
20640 This host system is used to download the configuration bitstream to
20641 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
20642 communicates with the target board using the JTAG interface and
20643 presents a @code{gdbserver} interface to the board. By default
20644 @code{xmd} uses port @code{1234}. (While it is possible to change
20645 this default port, it requires the use of undocumented @code{xmd}
20646 commands. Contact Xilinx support if you need to do this.)
20648 Use these GDB commands to connect to the MicroBlaze target processor.
20651 @item target remote :1234
20652 Use this command to connect to the target if you are running @value{GDBN}
20653 on the same system as @code{xmd}.
20655 @item target remote @var{xmd-host}:1234
20656 Use this command to connect to the target if it is connected to @code{xmd}
20657 running on a different system named @var{xmd-host}.
20660 Use this command to download a program to the MicroBlaze target.
20662 @item set debug microblaze @var{n}
20663 Enable MicroBlaze-specific debugging messages if non-zero.
20665 @item show debug microblaze @var{n}
20666 Show MicroBlaze-specific debugging level.
20669 @node MIPS Embedded
20670 @subsection @acronym{MIPS} Embedded
20672 @cindex @acronym{MIPS} boards
20673 @value{GDBN} can use the @acronym{MIPS} remote debugging protocol to talk to a
20674 @acronym{MIPS} board attached to a serial line. This is available when
20675 you configure @value{GDBN} with @samp{--target=mips-elf}.
20678 Use these @value{GDBN} commands to specify the connection to your target board:
20681 @item target mips @var{port}
20682 @kindex target mips @var{port}
20683 To run a program on the board, start up @code{@value{GDBP}} with the
20684 name of your program as the argument. To connect to the board, use the
20685 command @samp{target mips @var{port}}, where @var{port} is the name of
20686 the serial port connected to the board. If the program has not already
20687 been downloaded to the board, you may use the @code{load} command to
20688 download it. You can then use all the usual @value{GDBN} commands.
20690 For example, this sequence connects to the target board through a serial
20691 port, and loads and runs a program called @var{prog} through the
20695 host$ @value{GDBP} @var{prog}
20696 @value{GDBN} is free software and @dots{}
20697 (@value{GDBP}) target mips /dev/ttyb
20698 (@value{GDBP}) load @var{prog}
20702 @item target mips @var{hostname}:@var{portnumber}
20703 On some @value{GDBN} host configurations, you can specify a TCP
20704 connection (for instance, to a serial line managed by a terminal
20705 concentrator) instead of a serial port, using the syntax
20706 @samp{@var{hostname}:@var{portnumber}}.
20708 @item target pmon @var{port}
20709 @kindex target pmon @var{port}
20712 @item target ddb @var{port}
20713 @kindex target ddb @var{port}
20714 NEC's DDB variant of PMON for Vr4300.
20716 @item target lsi @var{port}
20717 @kindex target lsi @var{port}
20718 LSI variant of PMON.
20720 @kindex target r3900
20721 @item target r3900 @var{dev}
20722 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
20724 @kindex target array
20725 @item target array @var{dev}
20726 Array Tech LSI33K RAID controller board.
20732 @value{GDBN} also supports these special commands for @acronym{MIPS} targets:
20735 @item set mipsfpu double
20736 @itemx set mipsfpu single
20737 @itemx set mipsfpu none
20738 @itemx set mipsfpu auto
20739 @itemx show mipsfpu
20740 @kindex set mipsfpu
20741 @kindex show mipsfpu
20742 @cindex @acronym{MIPS} remote floating point
20743 @cindex floating point, @acronym{MIPS} remote
20744 If your target board does not support the @acronym{MIPS} floating point
20745 coprocessor, you should use the command @samp{set mipsfpu none} (if you
20746 need this, you may wish to put the command in your @value{GDBN} init
20747 file). This tells @value{GDBN} how to find the return value of
20748 functions which return floating point values. It also allows
20749 @value{GDBN} to avoid saving the floating point registers when calling
20750 functions on the board. If you are using a floating point coprocessor
20751 with only single precision floating point support, as on the @sc{r4650}
20752 processor, use the command @samp{set mipsfpu single}. The default
20753 double precision floating point coprocessor may be selected using
20754 @samp{set mipsfpu double}.
20756 In previous versions the only choices were double precision or no
20757 floating point, so @samp{set mipsfpu on} will select double precision
20758 and @samp{set mipsfpu off} will select no floating point.
20760 As usual, you can inquire about the @code{mipsfpu} variable with
20761 @samp{show mipsfpu}.
20763 @item set timeout @var{seconds}
20764 @itemx set retransmit-timeout @var{seconds}
20765 @itemx show timeout
20766 @itemx show retransmit-timeout
20767 @cindex @code{timeout}, @acronym{MIPS} protocol
20768 @cindex @code{retransmit-timeout}, @acronym{MIPS} protocol
20769 @kindex set timeout
20770 @kindex show timeout
20771 @kindex set retransmit-timeout
20772 @kindex show retransmit-timeout
20773 You can control the timeout used while waiting for a packet, in the @acronym{MIPS}
20774 remote protocol, with the @code{set timeout @var{seconds}} command. The
20775 default is 5 seconds. Similarly, you can control the timeout used while
20776 waiting for an acknowledgment of a packet with the @code{set
20777 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
20778 You can inspect both values with @code{show timeout} and @code{show
20779 retransmit-timeout}. (These commands are @emph{only} available when
20780 @value{GDBN} is configured for @samp{--target=mips-elf}.)
20782 The timeout set by @code{set timeout} does not apply when @value{GDBN}
20783 is waiting for your program to stop. In that case, @value{GDBN} waits
20784 forever because it has no way of knowing how long the program is going
20785 to run before stopping.
20787 @item set syn-garbage-limit @var{num}
20788 @kindex set syn-garbage-limit@r{, @acronym{MIPS} remote}
20789 @cindex synchronize with remote @acronym{MIPS} target
20790 Limit the maximum number of characters @value{GDBN} should ignore when
20791 it tries to synchronize with the remote target. The default is 10
20792 characters. Setting the limit to -1 means there's no limit.
20794 @item show syn-garbage-limit
20795 @kindex show syn-garbage-limit@r{, @acronym{MIPS} remote}
20796 Show the current limit on the number of characters to ignore when
20797 trying to synchronize with the remote system.
20799 @item set monitor-prompt @var{prompt}
20800 @kindex set monitor-prompt@r{, @acronym{MIPS} remote}
20801 @cindex remote monitor prompt
20802 Tell @value{GDBN} to expect the specified @var{prompt} string from the
20803 remote monitor. The default depends on the target:
20813 @item show monitor-prompt
20814 @kindex show monitor-prompt@r{, @acronym{MIPS} remote}
20815 Show the current strings @value{GDBN} expects as the prompt from the
20818 @item set monitor-warnings
20819 @kindex set monitor-warnings@r{, @acronym{MIPS} remote}
20820 Enable or disable monitor warnings about hardware breakpoints. This
20821 has effect only for the @code{lsi} target. When on, @value{GDBN} will
20822 display warning messages whose codes are returned by the @code{lsi}
20823 PMON monitor for breakpoint commands.
20825 @item show monitor-warnings
20826 @kindex show monitor-warnings@r{, @acronym{MIPS} remote}
20827 Show the current setting of printing monitor warnings.
20829 @item pmon @var{command}
20830 @kindex pmon@r{, @acronym{MIPS} remote}
20831 @cindex send PMON command
20832 This command allows sending an arbitrary @var{command} string to the
20833 monitor. The monitor must be in debug mode for this to work.
20836 @node PowerPC Embedded
20837 @subsection PowerPC Embedded
20839 @cindex DVC register
20840 @value{GDBN} supports using the DVC (Data Value Compare) register to
20841 implement in hardware simple hardware watchpoint conditions of the form:
20844 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
20845 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
20848 The DVC register will be automatically used when @value{GDBN} detects
20849 such pattern in a condition expression, and the created watchpoint uses one
20850 debug register (either the @code{exact-watchpoints} option is on and the
20851 variable is scalar, or the variable has a length of one byte). This feature
20852 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
20855 When running on PowerPC embedded processors, @value{GDBN} automatically uses
20856 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
20857 in which case watchpoints using only one debug register are created when
20858 watching variables of scalar types.
20860 You can create an artificial array to watch an arbitrary memory
20861 region using one of the following commands (@pxref{Expressions}):
20864 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
20865 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
20868 PowerPC embedded processors support masked watchpoints. See the discussion
20869 about the @code{mask} argument in @ref{Set Watchpoints}.
20871 @cindex ranged breakpoint
20872 PowerPC embedded processors support hardware accelerated
20873 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
20874 the inferior whenever it executes an instruction at any address within
20875 the range it specifies. To set a ranged breakpoint in @value{GDBN},
20876 use the @code{break-range} command.
20878 @value{GDBN} provides the following PowerPC-specific commands:
20881 @kindex break-range
20882 @item break-range @var{start-location}, @var{end-location}
20883 Set a breakpoint for an address range.
20884 @var{start-location} and @var{end-location} can specify a function name,
20885 a line number, an offset of lines from the current line or from the start
20886 location, or an address of an instruction (see @ref{Specify Location},
20887 for a list of all the possible ways to specify a @var{location}.)
20888 The breakpoint will stop execution of the inferior whenever it
20889 executes an instruction at any address within the specified range,
20890 (including @var{start-location} and @var{end-location}.)
20892 @kindex set powerpc
20893 @item set powerpc soft-float
20894 @itemx show powerpc soft-float
20895 Force @value{GDBN} to use (or not use) a software floating point calling
20896 convention. By default, @value{GDBN} selects the calling convention based
20897 on the selected architecture and the provided executable file.
20899 @item set powerpc vector-abi
20900 @itemx show powerpc vector-abi
20901 Force @value{GDBN} to use the specified calling convention for vector
20902 arguments and return values. The valid options are @samp{auto};
20903 @samp{generic}, to avoid vector registers even if they are present;
20904 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
20905 registers. By default, @value{GDBN} selects the calling convention
20906 based on the selected architecture and the provided executable file.
20908 @item set powerpc exact-watchpoints
20909 @itemx show powerpc exact-watchpoints
20910 Allow @value{GDBN} to use only one debug register when watching a variable
20911 of scalar type, thus assuming that the variable is accessed through the
20912 address of its first byte.
20914 @kindex target dink32
20915 @item target dink32 @var{dev}
20916 DINK32 ROM monitor.
20918 @kindex target ppcbug
20919 @item target ppcbug @var{dev}
20920 @kindex target ppcbug1
20921 @item target ppcbug1 @var{dev}
20922 PPCBUG ROM monitor for PowerPC.
20925 @item target sds @var{dev}
20926 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
20929 @cindex SDS protocol
20930 The following commands specific to the SDS protocol are supported
20934 @item set sdstimeout @var{nsec}
20935 @kindex set sdstimeout
20936 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
20937 default is 2 seconds.
20939 @item show sdstimeout
20940 @kindex show sdstimeout
20941 Show the current value of the SDS timeout.
20943 @item sds @var{command}
20944 @kindex sds@r{, a command}
20945 Send the specified @var{command} string to the SDS monitor.
20950 @subsection HP PA Embedded
20954 @kindex target op50n
20955 @item target op50n @var{dev}
20956 OP50N monitor, running on an OKI HPPA board.
20958 @kindex target w89k
20959 @item target w89k @var{dev}
20960 W89K monitor, running on a Winbond HPPA board.
20965 @subsection Tsqware Sparclet
20969 @value{GDBN} enables developers to debug tasks running on
20970 Sparclet targets from a Unix host.
20971 @value{GDBN} uses code that runs on
20972 both the Unix host and on the Sparclet target. The program
20973 @code{@value{GDBP}} is installed and executed on the Unix host.
20976 @item remotetimeout @var{args}
20977 @kindex remotetimeout
20978 @value{GDBN} supports the option @code{remotetimeout}.
20979 This option is set by the user, and @var{args} represents the number of
20980 seconds @value{GDBN} waits for responses.
20983 @cindex compiling, on Sparclet
20984 When compiling for debugging, include the options @samp{-g} to get debug
20985 information and @samp{-Ttext} to relocate the program to where you wish to
20986 load it on the target. You may also want to add the options @samp{-n} or
20987 @samp{-N} in order to reduce the size of the sections. Example:
20990 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
20993 You can use @code{objdump} to verify that the addresses are what you intended:
20996 sparclet-aout-objdump --headers --syms prog
20999 @cindex running, on Sparclet
21001 your Unix execution search path to find @value{GDBN}, you are ready to
21002 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
21003 (or @code{sparclet-aout-gdb}, depending on your installation).
21005 @value{GDBN} comes up showing the prompt:
21012 * Sparclet File:: Setting the file to debug
21013 * Sparclet Connection:: Connecting to Sparclet
21014 * Sparclet Download:: Sparclet download
21015 * Sparclet Execution:: Running and debugging
21018 @node Sparclet File
21019 @subsubsection Setting File to Debug
21021 The @value{GDBN} command @code{file} lets you choose with program to debug.
21024 (gdbslet) file prog
21028 @value{GDBN} then attempts to read the symbol table of @file{prog}.
21029 @value{GDBN} locates
21030 the file by searching the directories listed in the command search
21032 If the file was compiled with debug information (option @samp{-g}), source
21033 files will be searched as well.
21034 @value{GDBN} locates
21035 the source files by searching the directories listed in the directory search
21036 path (@pxref{Environment, ,Your Program's Environment}).
21038 to find a file, it displays a message such as:
21041 prog: No such file or directory.
21044 When this happens, add the appropriate directories to the search paths with
21045 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
21046 @code{target} command again.
21048 @node Sparclet Connection
21049 @subsubsection Connecting to Sparclet
21051 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
21052 To connect to a target on serial port ``@code{ttya}'', type:
21055 (gdbslet) target sparclet /dev/ttya
21056 Remote target sparclet connected to /dev/ttya
21057 main () at ../prog.c:3
21061 @value{GDBN} displays messages like these:
21067 @node Sparclet Download
21068 @subsubsection Sparclet Download
21070 @cindex download to Sparclet
21071 Once connected to the Sparclet target,
21072 you can use the @value{GDBN}
21073 @code{load} command to download the file from the host to the target.
21074 The file name and load offset should be given as arguments to the @code{load}
21076 Since the file format is aout, the program must be loaded to the starting
21077 address. You can use @code{objdump} to find out what this value is. The load
21078 offset is an offset which is added to the VMA (virtual memory address)
21079 of each of the file's sections.
21080 For instance, if the program
21081 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
21082 and bss at 0x12010170, in @value{GDBN}, type:
21085 (gdbslet) load prog 0x12010000
21086 Loading section .text, size 0xdb0 vma 0x12010000
21089 If the code is loaded at a different address then what the program was linked
21090 to, you may need to use the @code{section} and @code{add-symbol-file} commands
21091 to tell @value{GDBN} where to map the symbol table.
21093 @node Sparclet Execution
21094 @subsubsection Running and Debugging
21096 @cindex running and debugging Sparclet programs
21097 You can now begin debugging the task using @value{GDBN}'s execution control
21098 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
21099 manual for the list of commands.
21103 Breakpoint 1 at 0x12010000: file prog.c, line 3.
21105 Starting program: prog
21106 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
21107 3 char *symarg = 0;
21109 4 char *execarg = "hello!";
21114 @subsection Fujitsu Sparclite
21118 @kindex target sparclite
21119 @item target sparclite @var{dev}
21120 Fujitsu sparclite boards, used only for the purpose of loading.
21121 You must use an additional command to debug the program.
21122 For example: target remote @var{dev} using @value{GDBN} standard
21128 @subsection Zilog Z8000
21131 @cindex simulator, Z8000
21132 @cindex Zilog Z8000 simulator
21134 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
21137 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
21138 unsegmented variant of the Z8000 architecture) or the Z8001 (the
21139 segmented variant). The simulator recognizes which architecture is
21140 appropriate by inspecting the object code.
21143 @item target sim @var{args}
21145 @kindex target sim@r{, with Z8000}
21146 Debug programs on a simulated CPU. If the simulator supports setup
21147 options, specify them via @var{args}.
21151 After specifying this target, you can debug programs for the simulated
21152 CPU in the same style as programs for your host computer; use the
21153 @code{file} command to load a new program image, the @code{run} command
21154 to run your program, and so on.
21156 As well as making available all the usual machine registers
21157 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
21158 additional items of information as specially named registers:
21163 Counts clock-ticks in the simulator.
21166 Counts instructions run in the simulator.
21169 Execution time in 60ths of a second.
21173 You can refer to these values in @value{GDBN} expressions with the usual
21174 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
21175 conditional breakpoint that suspends only after at least 5000
21176 simulated clock ticks.
21179 @subsection Atmel AVR
21182 When configured for debugging the Atmel AVR, @value{GDBN} supports the
21183 following AVR-specific commands:
21186 @item info io_registers
21187 @kindex info io_registers@r{, AVR}
21188 @cindex I/O registers (Atmel AVR)
21189 This command displays information about the AVR I/O registers. For
21190 each register, @value{GDBN} prints its number and value.
21197 When configured for debugging CRIS, @value{GDBN} provides the
21198 following CRIS-specific commands:
21201 @item set cris-version @var{ver}
21202 @cindex CRIS version
21203 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
21204 The CRIS version affects register names and sizes. This command is useful in
21205 case autodetection of the CRIS version fails.
21207 @item show cris-version
21208 Show the current CRIS version.
21210 @item set cris-dwarf2-cfi
21211 @cindex DWARF-2 CFI and CRIS
21212 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
21213 Change to @samp{off} when using @code{gcc-cris} whose version is below
21216 @item show cris-dwarf2-cfi
21217 Show the current state of using DWARF-2 CFI.
21219 @item set cris-mode @var{mode}
21221 Set the current CRIS mode to @var{mode}. It should only be changed when
21222 debugging in guru mode, in which case it should be set to
21223 @samp{guru} (the default is @samp{normal}).
21225 @item show cris-mode
21226 Show the current CRIS mode.
21230 @subsection Renesas Super-H
21233 For the Renesas Super-H processor, @value{GDBN} provides these
21237 @item set sh calling-convention @var{convention}
21238 @kindex set sh calling-convention
21239 Set the calling-convention used when calling functions from @value{GDBN}.
21240 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
21241 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
21242 convention. If the DWARF-2 information of the called function specifies
21243 that the function follows the Renesas calling convention, the function
21244 is called using the Renesas calling convention. If the calling convention
21245 is set to @samp{renesas}, the Renesas calling convention is always used,
21246 regardless of the DWARF-2 information. This can be used to override the
21247 default of @samp{gcc} if debug information is missing, or the compiler
21248 does not emit the DWARF-2 calling convention entry for a function.
21250 @item show sh calling-convention
21251 @kindex show sh calling-convention
21252 Show the current calling convention setting.
21257 @node Architectures
21258 @section Architectures
21260 This section describes characteristics of architectures that affect
21261 all uses of @value{GDBN} with the architecture, both native and cross.
21268 * HPPA:: HP PA architecture
21269 * SPU:: Cell Broadband Engine SPU architecture
21275 @subsection AArch64
21276 @cindex AArch64 support
21278 When @value{GDBN} is debugging the AArch64 architecture, it provides the
21279 following special commands:
21282 @item set debug aarch64
21283 @kindex set debug aarch64
21284 This command determines whether AArch64 architecture-specific debugging
21285 messages are to be displayed.
21287 @item show debug aarch64
21288 Show whether AArch64 debugging messages are displayed.
21293 @subsection x86 Architecture-specific Issues
21296 @item set struct-convention @var{mode}
21297 @kindex set struct-convention
21298 @cindex struct return convention
21299 @cindex struct/union returned in registers
21300 Set the convention used by the inferior to return @code{struct}s and
21301 @code{union}s from functions to @var{mode}. Possible values of
21302 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
21303 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
21304 are returned on the stack, while @code{"reg"} means that a
21305 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
21306 be returned in a register.
21308 @item show struct-convention
21309 @kindex show struct-convention
21310 Show the current setting of the convention to return @code{struct}s
21317 See the following section.
21320 @subsection @acronym{MIPS}
21322 @cindex stack on Alpha
21323 @cindex stack on @acronym{MIPS}
21324 @cindex Alpha stack
21325 @cindex @acronym{MIPS} stack
21326 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
21327 sometimes requires @value{GDBN} to search backward in the object code to
21328 find the beginning of a function.
21330 @cindex response time, @acronym{MIPS} debugging
21331 To improve response time (especially for embedded applications, where
21332 @value{GDBN} may be restricted to a slow serial line for this search)
21333 you may want to limit the size of this search, using one of these
21337 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
21338 @item set heuristic-fence-post @var{limit}
21339 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
21340 search for the beginning of a function. A value of @var{0} (the
21341 default) means there is no limit. However, except for @var{0}, the
21342 larger the limit the more bytes @code{heuristic-fence-post} must search
21343 and therefore the longer it takes to run. You should only need to use
21344 this command when debugging a stripped executable.
21346 @item show heuristic-fence-post
21347 Display the current limit.
21351 These commands are available @emph{only} when @value{GDBN} is configured
21352 for debugging programs on Alpha or @acronym{MIPS} processors.
21354 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
21358 @item set mips abi @var{arg}
21359 @kindex set mips abi
21360 @cindex set ABI for @acronym{MIPS}
21361 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
21362 values of @var{arg} are:
21366 The default ABI associated with the current binary (this is the
21376 @item show mips abi
21377 @kindex show mips abi
21378 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
21380 @item set mips compression @var{arg}
21381 @kindex set mips compression
21382 @cindex code compression, @acronym{MIPS}
21383 Tell @value{GDBN} which @acronym{MIPS} compressed
21384 @acronym{ISA, Instruction Set Architecture} encoding is used by the
21385 inferior. @value{GDBN} uses this for code disassembly and other
21386 internal interpretation purposes. This setting is only referred to
21387 when no executable has been associated with the debugging session or
21388 the executable does not provide information about the encoding it uses.
21389 Otherwise this setting is automatically updated from information
21390 provided by the executable.
21392 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
21393 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
21394 executables containing @acronym{MIPS16} code frequently are not
21395 identified as such.
21397 This setting is ``sticky''; that is, it retains its value across
21398 debugging sessions until reset either explicitly with this command or
21399 implicitly from an executable.
21401 The compiler and/or assembler typically add symbol table annotations to
21402 identify functions compiled for the @acronym{MIPS16} or
21403 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
21404 are present, @value{GDBN} uses them in preference to the global
21405 compressed @acronym{ISA} encoding setting.
21407 @item show mips compression
21408 @kindex show mips compression
21409 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
21410 @value{GDBN} to debug the inferior.
21413 @itemx show mipsfpu
21414 @xref{MIPS Embedded, set mipsfpu}.
21416 @item set mips mask-address @var{arg}
21417 @kindex set mips mask-address
21418 @cindex @acronym{MIPS} addresses, masking
21419 This command determines whether the most-significant 32 bits of 64-bit
21420 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
21421 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
21422 setting, which lets @value{GDBN} determine the correct value.
21424 @item show mips mask-address
21425 @kindex show mips mask-address
21426 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
21429 @item set remote-mips64-transfers-32bit-regs
21430 @kindex set remote-mips64-transfers-32bit-regs
21431 This command controls compatibility with 64-bit @acronym{MIPS} targets that
21432 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
21433 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
21434 and 64 bits for other registers, set this option to @samp{on}.
21436 @item show remote-mips64-transfers-32bit-regs
21437 @kindex show remote-mips64-transfers-32bit-regs
21438 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
21440 @item set debug mips
21441 @kindex set debug mips
21442 This command turns on and off debugging messages for the @acronym{MIPS}-specific
21443 target code in @value{GDBN}.
21445 @item show debug mips
21446 @kindex show debug mips
21447 Show the current setting of @acronym{MIPS} debugging messages.
21453 @cindex HPPA support
21455 When @value{GDBN} is debugging the HP PA architecture, it provides the
21456 following special commands:
21459 @item set debug hppa
21460 @kindex set debug hppa
21461 This command determines whether HPPA architecture-specific debugging
21462 messages are to be displayed.
21464 @item show debug hppa
21465 Show whether HPPA debugging messages are displayed.
21467 @item maint print unwind @var{address}
21468 @kindex maint print unwind@r{, HPPA}
21469 This command displays the contents of the unwind table entry at the
21470 given @var{address}.
21476 @subsection Cell Broadband Engine SPU architecture
21477 @cindex Cell Broadband Engine
21480 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
21481 it provides the following special commands:
21484 @item info spu event
21486 Display SPU event facility status. Shows current event mask
21487 and pending event status.
21489 @item info spu signal
21490 Display SPU signal notification facility status. Shows pending
21491 signal-control word and signal notification mode of both signal
21492 notification channels.
21494 @item info spu mailbox
21495 Display SPU mailbox facility status. Shows all pending entries,
21496 in order of processing, in each of the SPU Write Outbound,
21497 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
21500 Display MFC DMA status. Shows all pending commands in the MFC
21501 DMA queue. For each entry, opcode, tag, class IDs, effective
21502 and local store addresses and transfer size are shown.
21504 @item info spu proxydma
21505 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
21506 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
21507 and local store addresses and transfer size are shown.
21511 When @value{GDBN} is debugging a combined PowerPC/SPU application
21512 on the Cell Broadband Engine, it provides in addition the following
21516 @item set spu stop-on-load @var{arg}
21518 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
21519 will give control to the user when a new SPE thread enters its @code{main}
21520 function. The default is @code{off}.
21522 @item show spu stop-on-load
21524 Show whether to stop for new SPE threads.
21526 @item set spu auto-flush-cache @var{arg}
21527 Set whether to automatically flush the software-managed cache. When set to
21528 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
21529 cache to be flushed whenever SPE execution stops. This provides a consistent
21530 view of PowerPC memory that is accessed via the cache. If an application
21531 does not use the software-managed cache, this option has no effect.
21533 @item show spu auto-flush-cache
21534 Show whether to automatically flush the software-managed cache.
21539 @subsection PowerPC
21540 @cindex PowerPC architecture
21542 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
21543 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
21544 numbers stored in the floating point registers. These values must be stored
21545 in two consecutive registers, always starting at an even register like
21546 @code{f0} or @code{f2}.
21548 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
21549 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
21550 @code{f2} and @code{f3} for @code{$dl1} and so on.
21552 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
21553 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
21556 @subsection Nios II
21557 @cindex Nios II architecture
21559 When @value{GDBN} is debugging the Nios II architecture,
21560 it provides the following special commands:
21564 @item set debug nios2
21565 @kindex set debug nios2
21566 This command turns on and off debugging messages for the Nios II
21567 target code in @value{GDBN}.
21569 @item show debug nios2
21570 @kindex show debug nios2
21571 Show the current setting of Nios II debugging messages.
21574 @node Controlling GDB
21575 @chapter Controlling @value{GDBN}
21577 You can alter the way @value{GDBN} interacts with you by using the
21578 @code{set} command. For commands controlling how @value{GDBN} displays
21579 data, see @ref{Print Settings, ,Print Settings}. Other settings are
21584 * Editing:: Command editing
21585 * Command History:: Command history
21586 * Screen Size:: Screen size
21587 * Numbers:: Numbers
21588 * ABI:: Configuring the current ABI
21589 * Auto-loading:: Automatically loading associated files
21590 * Messages/Warnings:: Optional warnings and messages
21591 * Debugging Output:: Optional messages about internal happenings
21592 * Other Misc Settings:: Other Miscellaneous Settings
21600 @value{GDBN} indicates its readiness to read a command by printing a string
21601 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
21602 can change the prompt string with the @code{set prompt} command. For
21603 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
21604 the prompt in one of the @value{GDBN} sessions so that you can always tell
21605 which one you are talking to.
21607 @emph{Note:} @code{set prompt} does not add a space for you after the
21608 prompt you set. This allows you to set a prompt which ends in a space
21609 or a prompt that does not.
21613 @item set prompt @var{newprompt}
21614 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
21616 @kindex show prompt
21618 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
21621 Versions of @value{GDBN} that ship with Python scripting enabled have
21622 prompt extensions. The commands for interacting with these extensions
21626 @kindex set extended-prompt
21627 @item set extended-prompt @var{prompt}
21628 Set an extended prompt that allows for substitutions.
21629 @xref{gdb.prompt}, for a list of escape sequences that can be used for
21630 substitution. Any escape sequences specified as part of the prompt
21631 string are replaced with the corresponding strings each time the prompt
21637 set extended-prompt Current working directory: \w (gdb)
21640 Note that when an extended-prompt is set, it takes control of the
21641 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
21643 @kindex show extended-prompt
21644 @item show extended-prompt
21645 Prints the extended prompt. Any escape sequences specified as part of
21646 the prompt string with @code{set extended-prompt}, are replaced with the
21647 corresponding strings each time the prompt is displayed.
21651 @section Command Editing
21653 @cindex command line editing
21655 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
21656 @sc{gnu} library provides consistent behavior for programs which provide a
21657 command line interface to the user. Advantages are @sc{gnu} Emacs-style
21658 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
21659 substitution, and a storage and recall of command history across
21660 debugging sessions.
21662 You may control the behavior of command line editing in @value{GDBN} with the
21663 command @code{set}.
21666 @kindex set editing
21669 @itemx set editing on
21670 Enable command line editing (enabled by default).
21672 @item set editing off
21673 Disable command line editing.
21675 @kindex show editing
21677 Show whether command line editing is enabled.
21680 @ifset SYSTEM_READLINE
21681 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
21683 @ifclear SYSTEM_READLINE
21684 @xref{Command Line Editing},
21686 for more details about the Readline
21687 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
21688 encouraged to read that chapter.
21690 @node Command History
21691 @section Command History
21692 @cindex command history
21694 @value{GDBN} can keep track of the commands you type during your
21695 debugging sessions, so that you can be certain of precisely what
21696 happened. Use these commands to manage the @value{GDBN} command
21699 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
21700 package, to provide the history facility.
21701 @ifset SYSTEM_READLINE
21702 @xref{Using History Interactively, , , history, GNU History Library},
21704 @ifclear SYSTEM_READLINE
21705 @xref{Using History Interactively},
21707 for the detailed description of the History library.
21709 To issue a command to @value{GDBN} without affecting certain aspects of
21710 the state which is seen by users, prefix it with @samp{server }
21711 (@pxref{Server Prefix}). This
21712 means that this command will not affect the command history, nor will it
21713 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
21714 pressed on a line by itself.
21716 @cindex @code{server}, command prefix
21717 The server prefix does not affect the recording of values into the value
21718 history; to print a value without recording it into the value history,
21719 use the @code{output} command instead of the @code{print} command.
21721 Here is the description of @value{GDBN} commands related to command
21725 @cindex history substitution
21726 @cindex history file
21727 @kindex set history filename
21728 @cindex @env{GDBHISTFILE}, environment variable
21729 @item set history filename @var{fname}
21730 Set the name of the @value{GDBN} command history file to @var{fname}.
21731 This is the file where @value{GDBN} reads an initial command history
21732 list, and where it writes the command history from this session when it
21733 exits. You can access this list through history expansion or through
21734 the history command editing characters listed below. This file defaults
21735 to the value of the environment variable @code{GDBHISTFILE}, or to
21736 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
21739 @cindex save command history
21740 @kindex set history save
21741 @item set history save
21742 @itemx set history save on
21743 Record command history in a file, whose name may be specified with the
21744 @code{set history filename} command. By default, this option is disabled.
21746 @item set history save off
21747 Stop recording command history in a file.
21749 @cindex history size
21750 @kindex set history size
21751 @cindex @env{HISTSIZE}, environment variable
21752 @item set history size @var{size}
21753 @itemx set history size unlimited
21754 Set the number of commands which @value{GDBN} keeps in its history list.
21755 This defaults to the value of the environment variable
21756 @code{HISTSIZE}, or to 256 if this variable is not set. If @var{size}
21757 is @code{unlimited}, the number of commands @value{GDBN} keeps in the
21758 history list is unlimited.
21761 History expansion assigns special meaning to the character @kbd{!}.
21762 @ifset SYSTEM_READLINE
21763 @xref{Event Designators, , , history, GNU History Library},
21765 @ifclear SYSTEM_READLINE
21766 @xref{Event Designators},
21770 @cindex history expansion, turn on/off
21771 Since @kbd{!} is also the logical not operator in C, history expansion
21772 is off by default. If you decide to enable history expansion with the
21773 @code{set history expansion on} command, you may sometimes need to
21774 follow @kbd{!} (when it is used as logical not, in an expression) with
21775 a space or a tab to prevent it from being expanded. The readline
21776 history facilities do not attempt substitution on the strings
21777 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
21779 The commands to control history expansion are:
21782 @item set history expansion on
21783 @itemx set history expansion
21784 @kindex set history expansion
21785 Enable history expansion. History expansion is off by default.
21787 @item set history expansion off
21788 Disable history expansion.
21791 @kindex show history
21793 @itemx show history filename
21794 @itemx show history save
21795 @itemx show history size
21796 @itemx show history expansion
21797 These commands display the state of the @value{GDBN} history parameters.
21798 @code{show history} by itself displays all four states.
21803 @kindex show commands
21804 @cindex show last commands
21805 @cindex display command history
21806 @item show commands
21807 Display the last ten commands in the command history.
21809 @item show commands @var{n}
21810 Print ten commands centered on command number @var{n}.
21812 @item show commands +
21813 Print ten commands just after the commands last printed.
21817 @section Screen Size
21818 @cindex size of screen
21819 @cindex pauses in output
21821 Certain commands to @value{GDBN} may produce large amounts of
21822 information output to the screen. To help you read all of it,
21823 @value{GDBN} pauses and asks you for input at the end of each page of
21824 output. Type @key{RET} when you want to continue the output, or @kbd{q}
21825 to discard the remaining output. Also, the screen width setting
21826 determines when to wrap lines of output. Depending on what is being
21827 printed, @value{GDBN} tries to break the line at a readable place,
21828 rather than simply letting it overflow onto the following line.
21830 Normally @value{GDBN} knows the size of the screen from the terminal
21831 driver software. For example, on Unix @value{GDBN} uses the termcap data base
21832 together with the value of the @code{TERM} environment variable and the
21833 @code{stty rows} and @code{stty cols} settings. If this is not correct,
21834 you can override it with the @code{set height} and @code{set
21841 @kindex show height
21842 @item set height @var{lpp}
21843 @itemx set height unlimited
21845 @itemx set width @var{cpl}
21846 @itemx set width unlimited
21848 These @code{set} commands specify a screen height of @var{lpp} lines and
21849 a screen width of @var{cpl} characters. The associated @code{show}
21850 commands display the current settings.
21852 If you specify a height of either @code{unlimited} or zero lines,
21853 @value{GDBN} does not pause during output no matter how long the
21854 output is. This is useful if output is to a file or to an editor
21857 Likewise, you can specify @samp{set width unlimited} or @samp{set
21858 width 0} to prevent @value{GDBN} from wrapping its output.
21860 @item set pagination on
21861 @itemx set pagination off
21862 @kindex set pagination
21863 Turn the output pagination on or off; the default is on. Turning
21864 pagination off is the alternative to @code{set height unlimited}. Note that
21865 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
21866 Options, -batch}) also automatically disables pagination.
21868 @item show pagination
21869 @kindex show pagination
21870 Show the current pagination mode.
21875 @cindex number representation
21876 @cindex entering numbers
21878 You can always enter numbers in octal, decimal, or hexadecimal in
21879 @value{GDBN} by the usual conventions: octal numbers begin with
21880 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
21881 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
21882 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
21883 10; likewise, the default display for numbers---when no particular
21884 format is specified---is base 10. You can change the default base for
21885 both input and output with the commands described below.
21888 @kindex set input-radix
21889 @item set input-radix @var{base}
21890 Set the default base for numeric input. Supported choices
21891 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
21892 specified either unambiguously or using the current input radix; for
21896 set input-radix 012
21897 set input-radix 10.
21898 set input-radix 0xa
21902 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
21903 leaves the input radix unchanged, no matter what it was, since
21904 @samp{10}, being without any leading or trailing signs of its base, is
21905 interpreted in the current radix. Thus, if the current radix is 16,
21906 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
21909 @kindex set output-radix
21910 @item set output-radix @var{base}
21911 Set the default base for numeric display. Supported choices
21912 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
21913 specified either unambiguously or using the current input radix.
21915 @kindex show input-radix
21916 @item show input-radix
21917 Display the current default base for numeric input.
21919 @kindex show output-radix
21920 @item show output-radix
21921 Display the current default base for numeric display.
21923 @item set radix @r{[}@var{base}@r{]}
21927 These commands set and show the default base for both input and output
21928 of numbers. @code{set radix} sets the radix of input and output to
21929 the same base; without an argument, it resets the radix back to its
21930 default value of 10.
21935 @section Configuring the Current ABI
21937 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
21938 application automatically. However, sometimes you need to override its
21939 conclusions. Use these commands to manage @value{GDBN}'s view of the
21945 @cindex Newlib OS ABI and its influence on the longjmp handling
21947 One @value{GDBN} configuration can debug binaries for multiple operating
21948 system targets, either via remote debugging or native emulation.
21949 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
21950 but you can override its conclusion using the @code{set osabi} command.
21951 One example where this is useful is in debugging of binaries which use
21952 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
21953 not have the same identifying marks that the standard C library for your
21956 When @value{GDBN} is debugging the AArch64 architecture, it provides a
21957 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
21958 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
21959 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
21963 Show the OS ABI currently in use.
21966 With no argument, show the list of registered available OS ABI's.
21968 @item set osabi @var{abi}
21969 Set the current OS ABI to @var{abi}.
21972 @cindex float promotion
21974 Generally, the way that an argument of type @code{float} is passed to a
21975 function depends on whether the function is prototyped. For a prototyped
21976 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
21977 according to the architecture's convention for @code{float}. For unprototyped
21978 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
21979 @code{double} and then passed.
21981 Unfortunately, some forms of debug information do not reliably indicate whether
21982 a function is prototyped. If @value{GDBN} calls a function that is not marked
21983 as prototyped, it consults @kbd{set coerce-float-to-double}.
21986 @kindex set coerce-float-to-double
21987 @item set coerce-float-to-double
21988 @itemx set coerce-float-to-double on
21989 Arguments of type @code{float} will be promoted to @code{double} when passed
21990 to an unprototyped function. This is the default setting.
21992 @item set coerce-float-to-double off
21993 Arguments of type @code{float} will be passed directly to unprototyped
21996 @kindex show coerce-float-to-double
21997 @item show coerce-float-to-double
21998 Show the current setting of promoting @code{float} to @code{double}.
22002 @kindex show cp-abi
22003 @value{GDBN} needs to know the ABI used for your program's C@t{++}
22004 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
22005 used to build your application. @value{GDBN} only fully supports
22006 programs with a single C@t{++} ABI; if your program contains code using
22007 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
22008 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
22009 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
22010 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
22011 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
22012 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
22017 Show the C@t{++} ABI currently in use.
22020 With no argument, show the list of supported C@t{++} ABI's.
22022 @item set cp-abi @var{abi}
22023 @itemx set cp-abi auto
22024 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
22028 @section Automatically loading associated files
22029 @cindex auto-loading
22031 @value{GDBN} sometimes reads files with commands and settings automatically,
22032 without being explicitly told so by the user. We call this feature
22033 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
22034 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
22035 results or introduce security risks (e.g., if the file comes from untrusted
22038 Note that loading of these associated files (including the local @file{.gdbinit}
22039 file) requires accordingly configured @code{auto-load safe-path}
22040 (@pxref{Auto-loading safe path}).
22042 For these reasons, @value{GDBN} includes commands and options to let you
22043 control when to auto-load files and which files should be auto-loaded.
22046 @anchor{set auto-load off}
22047 @kindex set auto-load off
22048 @item set auto-load off
22049 Globally disable loading of all auto-loaded files.
22050 You may want to use this command with the @samp{-iex} option
22051 (@pxref{Option -init-eval-command}) such as:
22053 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
22056 Be aware that system init file (@pxref{System-wide configuration})
22057 and init files from your home directory (@pxref{Home Directory Init File})
22058 still get read (as they come from generally trusted directories).
22059 To prevent @value{GDBN} from auto-loading even those init files, use the
22060 @option{-nx} option (@pxref{Mode Options}), in addition to
22061 @code{set auto-load no}.
22063 @anchor{show auto-load}
22064 @kindex show auto-load
22065 @item show auto-load
22066 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
22070 (gdb) show auto-load
22071 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
22072 libthread-db: Auto-loading of inferior specific libthread_db is on.
22073 local-gdbinit: Auto-loading of .gdbinit script from current directory
22075 python-scripts: Auto-loading of Python scripts is on.
22076 safe-path: List of directories from which it is safe to auto-load files
22077 is $debugdir:$datadir/auto-load.
22078 scripts-directory: List of directories from which to load auto-loaded scripts
22079 is $debugdir:$datadir/auto-load.
22082 @anchor{info auto-load}
22083 @kindex info auto-load
22084 @item info auto-load
22085 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
22089 (gdb) info auto-load
22092 Yes /home/user/gdb/gdb-gdb.gdb
22093 libthread-db: No auto-loaded libthread-db.
22094 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
22098 Yes /home/user/gdb/gdb-gdb.py
22102 These are various kinds of files @value{GDBN} can automatically load:
22106 @xref{objfile-gdb.py file}, controlled by @ref{set auto-load python-scripts}.
22108 @xref{objfile-gdb.gdb file}, controlled by @ref{set auto-load gdb-scripts}.
22110 @xref{dotdebug_gdb_scripts section},
22111 controlled by @ref{set auto-load python-scripts}.
22113 @xref{Init File in the Current Directory},
22114 controlled by @ref{set auto-load local-gdbinit}.
22116 @xref{libthread_db.so.1 file}, controlled by @ref{set auto-load libthread-db}.
22119 These are @value{GDBN} control commands for the auto-loading:
22121 @multitable @columnfractions .5 .5
22122 @item @xref{set auto-load off}.
22123 @tab Disable auto-loading globally.
22124 @item @xref{show auto-load}.
22125 @tab Show setting of all kinds of files.
22126 @item @xref{info auto-load}.
22127 @tab Show state of all kinds of files.
22128 @item @xref{set auto-load gdb-scripts}.
22129 @tab Control for @value{GDBN} command scripts.
22130 @item @xref{show auto-load gdb-scripts}.
22131 @tab Show setting of @value{GDBN} command scripts.
22132 @item @xref{info auto-load gdb-scripts}.
22133 @tab Show state of @value{GDBN} command scripts.
22134 @item @xref{set auto-load python-scripts}.
22135 @tab Control for @value{GDBN} Python scripts.
22136 @item @xref{show auto-load python-scripts}.
22137 @tab Show setting of @value{GDBN} Python scripts.
22138 @item @xref{info auto-load python-scripts}.
22139 @tab Show state of @value{GDBN} Python scripts.
22140 @item @xref{set auto-load scripts-directory}.
22141 @tab Control for @value{GDBN} auto-loaded scripts location.
22142 @item @xref{show auto-load scripts-directory}.
22143 @tab Show @value{GDBN} auto-loaded scripts location.
22144 @item @xref{set auto-load local-gdbinit}.
22145 @tab Control for init file in the current directory.
22146 @item @xref{show auto-load local-gdbinit}.
22147 @tab Show setting of init file in the current directory.
22148 @item @xref{info auto-load local-gdbinit}.
22149 @tab Show state of init file in the current directory.
22150 @item @xref{set auto-load libthread-db}.
22151 @tab Control for thread debugging library.
22152 @item @xref{show auto-load libthread-db}.
22153 @tab Show setting of thread debugging library.
22154 @item @xref{info auto-load libthread-db}.
22155 @tab Show state of thread debugging library.
22156 @item @xref{set auto-load safe-path}.
22157 @tab Control directories trusted for automatic loading.
22158 @item @xref{show auto-load safe-path}.
22159 @tab Show directories trusted for automatic loading.
22160 @item @xref{add-auto-load-safe-path}.
22161 @tab Add directory trusted for automatic loading.
22165 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
22166 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
22167 * objfile-gdb.gdb file:: @samp{set/show/info auto-load gdb-script}
22168 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
22169 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
22170 @xref{Python Auto-loading}.
22173 @node Init File in the Current Directory
22174 @subsection Automatically loading init file in the current directory
22175 @cindex auto-loading init file in the current directory
22177 By default, @value{GDBN} reads and executes the canned sequences of commands
22178 from init file (if any) in the current working directory,
22179 see @ref{Init File in the Current Directory during Startup}.
22181 Note that loading of this local @file{.gdbinit} file also requires accordingly
22182 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
22185 @anchor{set auto-load local-gdbinit}
22186 @kindex set auto-load local-gdbinit
22187 @item set auto-load local-gdbinit [on|off]
22188 Enable or disable the auto-loading of canned sequences of commands
22189 (@pxref{Sequences}) found in init file in the current directory.
22191 @anchor{show auto-load local-gdbinit}
22192 @kindex show auto-load local-gdbinit
22193 @item show auto-load local-gdbinit
22194 Show whether auto-loading of canned sequences of commands from init file in the
22195 current directory is enabled or disabled.
22197 @anchor{info auto-load local-gdbinit}
22198 @kindex info auto-load local-gdbinit
22199 @item info auto-load local-gdbinit
22200 Print whether canned sequences of commands from init file in the
22201 current directory have been auto-loaded.
22204 @node libthread_db.so.1 file
22205 @subsection Automatically loading thread debugging library
22206 @cindex auto-loading libthread_db.so.1
22208 This feature is currently present only on @sc{gnu}/Linux native hosts.
22210 @value{GDBN} reads in some cases thread debugging library from places specific
22211 to the inferior (@pxref{set libthread-db-search-path}).
22213 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
22214 without checking this @samp{set auto-load libthread-db} switch as system
22215 libraries have to be trusted in general. In all other cases of
22216 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
22217 auto-load libthread-db} is enabled before trying to open such thread debugging
22220 Note that loading of this debugging library also requires accordingly configured
22221 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
22224 @anchor{set auto-load libthread-db}
22225 @kindex set auto-load libthread-db
22226 @item set auto-load libthread-db [on|off]
22227 Enable or disable the auto-loading of inferior specific thread debugging library.
22229 @anchor{show auto-load libthread-db}
22230 @kindex show auto-load libthread-db
22231 @item show auto-load libthread-db
22232 Show whether auto-loading of inferior specific thread debugging library is
22233 enabled or disabled.
22235 @anchor{info auto-load libthread-db}
22236 @kindex info auto-load libthread-db
22237 @item info auto-load libthread-db
22238 Print the list of all loaded inferior specific thread debugging libraries and
22239 for each such library print list of inferior @var{pid}s using it.
22242 @node objfile-gdb.gdb file
22243 @subsection The @file{@var{objfile}-gdb.gdb} file
22244 @cindex auto-loading @file{@var{objfile}-gdb.gdb}
22246 @value{GDBN} tries to load an @file{@var{objfile}-gdb.gdb} file containing
22247 canned sequences of commands (@pxref{Sequences}), as long as @samp{set
22248 auto-load gdb-scripts} is set to @samp{on}.
22250 Note that loading of this script file also requires accordingly configured
22251 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
22253 For more background refer to the similar Python scripts auto-loading
22254 description (@pxref{objfile-gdb.py file}).
22257 @anchor{set auto-load gdb-scripts}
22258 @kindex set auto-load gdb-scripts
22259 @item set auto-load gdb-scripts [on|off]
22260 Enable or disable the auto-loading of canned sequences of commands scripts.
22262 @anchor{show auto-load gdb-scripts}
22263 @kindex show auto-load gdb-scripts
22264 @item show auto-load gdb-scripts
22265 Show whether auto-loading of canned sequences of commands scripts is enabled or
22268 @anchor{info auto-load gdb-scripts}
22269 @kindex info auto-load gdb-scripts
22270 @cindex print list of auto-loaded canned sequences of commands scripts
22271 @item info auto-load gdb-scripts [@var{regexp}]
22272 Print the list of all canned sequences of commands scripts that @value{GDBN}
22276 If @var{regexp} is supplied only canned sequences of commands scripts with
22277 matching names are printed.
22279 @node Auto-loading safe path
22280 @subsection Security restriction for auto-loading
22281 @cindex auto-loading safe-path
22283 As the files of inferior can come from untrusted source (such as submitted by
22284 an application user) @value{GDBN} does not always load any files automatically.
22285 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
22286 directories trusted for loading files not explicitly requested by user.
22287 Each directory can also be a shell wildcard pattern.
22289 If the path is not set properly you will see a warning and the file will not
22294 Reading symbols from /home/user/gdb/gdb...done.
22295 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
22296 declined by your `auto-load safe-path' set
22297 to "$debugdir:$datadir/auto-load".
22298 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
22299 declined by your `auto-load safe-path' set
22300 to "$debugdir:$datadir/auto-load".
22304 To instruct @value{GDBN} to go ahead and use the init files anyway,
22305 invoke @value{GDBN} like this:
22308 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
22311 The list of trusted directories is controlled by the following commands:
22314 @anchor{set auto-load safe-path}
22315 @kindex set auto-load safe-path
22316 @item set auto-load safe-path @r{[}@var{directories}@r{]}
22317 Set the list of directories (and their subdirectories) trusted for automatic
22318 loading and execution of scripts. You can also enter a specific trusted file.
22319 Each directory can also be a shell wildcard pattern; wildcards do not match
22320 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
22321 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
22322 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
22323 its default value as specified during @value{GDBN} compilation.
22325 The list of directories uses path separator (@samp{:} on GNU and Unix
22326 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
22327 to the @env{PATH} environment variable.
22329 @anchor{show auto-load safe-path}
22330 @kindex show auto-load safe-path
22331 @item show auto-load safe-path
22332 Show the list of directories trusted for automatic loading and execution of
22335 @anchor{add-auto-load-safe-path}
22336 @kindex add-auto-load-safe-path
22337 @item add-auto-load-safe-path
22338 Add an entry (or list of entries) the list of directories trusted for automatic
22339 loading and execution of scripts. Multiple entries may be delimited by the
22340 host platform path separator in use.
22343 This variable defaults to what @code{--with-auto-load-dir} has been configured
22344 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
22345 substitution applies the same as for @ref{set auto-load scripts-directory}.
22346 The default @code{set auto-load safe-path} value can be also overriden by
22347 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
22349 Setting this variable to @file{/} disables this security protection,
22350 corresponding @value{GDBN} configuration option is
22351 @option{--without-auto-load-safe-path}.
22352 This variable is supposed to be set to the system directories writable by the
22353 system superuser only. Users can add their source directories in init files in
22354 their home directories (@pxref{Home Directory Init File}). See also deprecated
22355 init file in the current directory
22356 (@pxref{Init File in the Current Directory during Startup}).
22358 To force @value{GDBN} to load the files it declined to load in the previous
22359 example, you could use one of the following ways:
22362 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
22363 Specify this trusted directory (or a file) as additional component of the list.
22364 You have to specify also any existing directories displayed by
22365 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
22367 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
22368 Specify this directory as in the previous case but just for a single
22369 @value{GDBN} session.
22371 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
22372 Disable auto-loading safety for a single @value{GDBN} session.
22373 This assumes all the files you debug during this @value{GDBN} session will come
22374 from trusted sources.
22376 @item @kbd{./configure --without-auto-load-safe-path}
22377 During compilation of @value{GDBN} you may disable any auto-loading safety.
22378 This assumes all the files you will ever debug with this @value{GDBN} come from
22382 On the other hand you can also explicitly forbid automatic files loading which
22383 also suppresses any such warning messages:
22386 @item @kbd{gdb -iex "set auto-load no" @dots{}}
22387 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
22389 @item @file{~/.gdbinit}: @samp{set auto-load no}
22390 Disable auto-loading globally for the user
22391 (@pxref{Home Directory Init File}). While it is improbable, you could also
22392 use system init file instead (@pxref{System-wide configuration}).
22395 This setting applies to the file names as entered by user. If no entry matches
22396 @value{GDBN} tries as a last resort to also resolve all the file names into
22397 their canonical form (typically resolving symbolic links) and compare the
22398 entries again. @value{GDBN} already canonicalizes most of the filenames on its
22399 own before starting the comparison so a canonical form of directories is
22400 recommended to be entered.
22402 @node Auto-loading verbose mode
22403 @subsection Displaying files tried for auto-load
22404 @cindex auto-loading verbose mode
22406 For better visibility of all the file locations where you can place scripts to
22407 be auto-loaded with inferior --- or to protect yourself against accidental
22408 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
22409 all the files attempted to be loaded. Both existing and non-existing files may
22412 For example the list of directories from which it is safe to auto-load files
22413 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
22414 may not be too obvious while setting it up.
22417 (gdb) set debug auto-load on
22418 (gdb) file ~/src/t/true
22419 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
22420 for objfile "/tmp/true".
22421 auto-load: Updating directories of "/usr:/opt".
22422 auto-load: Using directory "/usr".
22423 auto-load: Using directory "/opt".
22424 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
22425 by your `auto-load safe-path' set to "/usr:/opt".
22429 @anchor{set debug auto-load}
22430 @kindex set debug auto-load
22431 @item set debug auto-load [on|off]
22432 Set whether to print the filenames attempted to be auto-loaded.
22434 @anchor{show debug auto-load}
22435 @kindex show debug auto-load
22436 @item show debug auto-load
22437 Show whether printing of the filenames attempted to be auto-loaded is turned
22441 @node Messages/Warnings
22442 @section Optional Warnings and Messages
22444 @cindex verbose operation
22445 @cindex optional warnings
22446 By default, @value{GDBN} is silent about its inner workings. If you are
22447 running on a slow machine, you may want to use the @code{set verbose}
22448 command. This makes @value{GDBN} tell you when it does a lengthy
22449 internal operation, so you will not think it has crashed.
22451 Currently, the messages controlled by @code{set verbose} are those
22452 which announce that the symbol table for a source file is being read;
22453 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
22456 @kindex set verbose
22457 @item set verbose on
22458 Enables @value{GDBN} output of certain informational messages.
22460 @item set verbose off
22461 Disables @value{GDBN} output of certain informational messages.
22463 @kindex show verbose
22465 Displays whether @code{set verbose} is on or off.
22468 By default, if @value{GDBN} encounters bugs in the symbol table of an
22469 object file, it is silent; but if you are debugging a compiler, you may
22470 find this information useful (@pxref{Symbol Errors, ,Errors Reading
22475 @kindex set complaints
22476 @item set complaints @var{limit}
22477 Permits @value{GDBN} to output @var{limit} complaints about each type of
22478 unusual symbols before becoming silent about the problem. Set
22479 @var{limit} to zero to suppress all complaints; set it to a large number
22480 to prevent complaints from being suppressed.
22482 @kindex show complaints
22483 @item show complaints
22484 Displays how many symbol complaints @value{GDBN} is permitted to produce.
22488 @anchor{confirmation requests}
22489 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
22490 lot of stupid questions to confirm certain commands. For example, if
22491 you try to run a program which is already running:
22495 The program being debugged has been started already.
22496 Start it from the beginning? (y or n)
22499 If you are willing to unflinchingly face the consequences of your own
22500 commands, you can disable this ``feature'':
22504 @kindex set confirm
22506 @cindex confirmation
22507 @cindex stupid questions
22508 @item set confirm off
22509 Disables confirmation requests. Note that running @value{GDBN} with
22510 the @option{--batch} option (@pxref{Mode Options, -batch}) also
22511 automatically disables confirmation requests.
22513 @item set confirm on
22514 Enables confirmation requests (the default).
22516 @kindex show confirm
22518 Displays state of confirmation requests.
22522 @cindex command tracing
22523 If you need to debug user-defined commands or sourced files you may find it
22524 useful to enable @dfn{command tracing}. In this mode each command will be
22525 printed as it is executed, prefixed with one or more @samp{+} symbols, the
22526 quantity denoting the call depth of each command.
22529 @kindex set trace-commands
22530 @cindex command scripts, debugging
22531 @item set trace-commands on
22532 Enable command tracing.
22533 @item set trace-commands off
22534 Disable command tracing.
22535 @item show trace-commands
22536 Display the current state of command tracing.
22539 @node Debugging Output
22540 @section Optional Messages about Internal Happenings
22541 @cindex optional debugging messages
22543 @value{GDBN} has commands that enable optional debugging messages from
22544 various @value{GDBN} subsystems; normally these commands are of
22545 interest to @value{GDBN} maintainers, or when reporting a bug. This
22546 section documents those commands.
22549 @kindex set exec-done-display
22550 @item set exec-done-display
22551 Turns on or off the notification of asynchronous commands'
22552 completion. When on, @value{GDBN} will print a message when an
22553 asynchronous command finishes its execution. The default is off.
22554 @kindex show exec-done-display
22555 @item show exec-done-display
22556 Displays the current setting of asynchronous command completion
22559 @cindex ARM AArch64
22560 @item set debug aarch64
22561 Turns on or off display of debugging messages related to ARM AArch64.
22562 The default is off.
22564 @item show debug aarch64
22565 Displays the current state of displaying debugging messages related to
22567 @cindex gdbarch debugging info
22568 @cindex architecture debugging info
22569 @item set debug arch
22570 Turns on or off display of gdbarch debugging info. The default is off
22571 @item show debug arch
22572 Displays the current state of displaying gdbarch debugging info.
22573 @item set debug aix-solib
22574 @cindex AIX shared library debugging
22575 Control display of debugging messages from the AIX shared library
22576 support module. The default is off.
22577 @item show debug aix-thread
22578 Show the current state of displaying AIX shared library debugging messages.
22579 @item set debug aix-thread
22580 @cindex AIX threads
22581 Display debugging messages about inner workings of the AIX thread
22583 @item show debug aix-thread
22584 Show the current state of AIX thread debugging info display.
22585 @item set debug check-physname
22587 Check the results of the ``physname'' computation. When reading DWARF
22588 debugging information for C@t{++}, @value{GDBN} attempts to compute
22589 each entity's name. @value{GDBN} can do this computation in two
22590 different ways, depending on exactly what information is present.
22591 When enabled, this setting causes @value{GDBN} to compute the names
22592 both ways and display any discrepancies.
22593 @item show debug check-physname
22594 Show the current state of ``physname'' checking.
22595 @item set debug coff-pe-read
22596 @cindex COFF/PE exported symbols
22597 Control display of debugging messages related to reading of COFF/PE
22598 exported symbols. The default is off.
22599 @item show debug coff-pe-read
22600 Displays the current state of displaying debugging messages related to
22601 reading of COFF/PE exported symbols.
22602 @item set debug dwarf2-die
22603 @cindex DWARF2 DIEs
22604 Dump DWARF2 DIEs after they are read in.
22605 The value is the number of nesting levels to print.
22606 A value of zero turns off the display.
22607 @item show debug dwarf2-die
22608 Show the current state of DWARF2 DIE debugging.
22609 @item set debug dwarf2-read
22610 @cindex DWARF2 Reading
22611 Turns on or off display of debugging messages related to reading
22612 DWARF debug info. The default is off.
22613 @item show debug dwarf2-read
22614 Show the current state of DWARF2 reader debugging.
22615 @item set debug displaced
22616 @cindex displaced stepping debugging info
22617 Turns on or off display of @value{GDBN} debugging info for the
22618 displaced stepping support. The default is off.
22619 @item show debug displaced
22620 Displays the current state of displaying @value{GDBN} debugging info
22621 related to displaced stepping.
22622 @item set debug event
22623 @cindex event debugging info
22624 Turns on or off display of @value{GDBN} event debugging info. The
22626 @item show debug event
22627 Displays the current state of displaying @value{GDBN} event debugging
22629 @item set debug expression
22630 @cindex expression debugging info
22631 Turns on or off display of debugging info about @value{GDBN}
22632 expression parsing. The default is off.
22633 @item show debug expression
22634 Displays the current state of displaying debugging info about
22635 @value{GDBN} expression parsing.
22636 @item set debug frame
22637 @cindex frame debugging info
22638 Turns on or off display of @value{GDBN} frame debugging info. The
22640 @item show debug frame
22641 Displays the current state of displaying @value{GDBN} frame debugging
22643 @item set debug gnu-nat
22644 @cindex @sc{gnu}/Hurd debug messages
22645 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
22646 @item show debug gnu-nat
22647 Show the current state of @sc{gnu}/Hurd debugging messages.
22648 @item set debug infrun
22649 @cindex inferior debugging info
22650 Turns on or off display of @value{GDBN} debugging info for running the inferior.
22651 The default is off. @file{infrun.c} contains GDB's runtime state machine used
22652 for implementing operations such as single-stepping the inferior.
22653 @item show debug infrun
22654 Displays the current state of @value{GDBN} inferior debugging.
22655 @item set debug jit
22656 @cindex just-in-time compilation, debugging messages
22657 Turns on or off debugging messages from JIT debug support.
22658 @item show debug jit
22659 Displays the current state of @value{GDBN} JIT debugging.
22660 @item set debug lin-lwp
22661 @cindex @sc{gnu}/Linux LWP debug messages
22662 @cindex Linux lightweight processes
22663 Turns on or off debugging messages from the Linux LWP debug support.
22664 @item show debug lin-lwp
22665 Show the current state of Linux LWP debugging messages.
22666 @item set debug mach-o
22667 @cindex Mach-O symbols processing
22668 Control display of debugging messages related to Mach-O symbols
22669 processing. The default is off.
22670 @item show debug mach-o
22671 Displays the current state of displaying debugging messages related to
22672 reading of COFF/PE exported symbols.
22673 @item set debug notification
22674 @cindex remote async notification debugging info
22675 Turns on or off debugging messages about remote async notification.
22676 The default is off.
22677 @item show debug notification
22678 Displays the current state of remote async notification debugging messages.
22679 @item set debug observer
22680 @cindex observer debugging info
22681 Turns on or off display of @value{GDBN} observer debugging. This
22682 includes info such as the notification of observable events.
22683 @item show debug observer
22684 Displays the current state of observer debugging.
22685 @item set debug overload
22686 @cindex C@t{++} overload debugging info
22687 Turns on or off display of @value{GDBN} C@t{++} overload debugging
22688 info. This includes info such as ranking of functions, etc. The default
22690 @item show debug overload
22691 Displays the current state of displaying @value{GDBN} C@t{++} overload
22693 @cindex expression parser, debugging info
22694 @cindex debug expression parser
22695 @item set debug parser
22696 Turns on or off the display of expression parser debugging output.
22697 Internally, this sets the @code{yydebug} variable in the expression
22698 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
22699 details. The default is off.
22700 @item show debug parser
22701 Show the current state of expression parser debugging.
22702 @cindex packets, reporting on stdout
22703 @cindex serial connections, debugging
22704 @cindex debug remote protocol
22705 @cindex remote protocol debugging
22706 @cindex display remote packets
22707 @item set debug remote
22708 Turns on or off display of reports on all packets sent back and forth across
22709 the serial line to the remote machine. The info is printed on the
22710 @value{GDBN} standard output stream. The default is off.
22711 @item show debug remote
22712 Displays the state of display of remote packets.
22713 @item set debug serial
22714 Turns on or off display of @value{GDBN} serial debugging info. The
22716 @item show debug serial
22717 Displays the current state of displaying @value{GDBN} serial debugging
22719 @item set debug solib-frv
22720 @cindex FR-V shared-library debugging
22721 Turns on or off debugging messages for FR-V shared-library code.
22722 @item show debug solib-frv
22723 Display the current state of FR-V shared-library code debugging
22725 @item set debug symfile
22726 @cindex symbol file functions
22727 Turns on or off display of debugging messages related to symbol file functions.
22728 The default is off. @xref{Files}.
22729 @item show debug symfile
22730 Show the current state of symbol file debugging messages.
22731 @item set debug symtab-create
22732 @cindex symbol table creation
22733 Turns on or off display of debugging messages related to symbol table creation.
22734 The default is 0 (off).
22735 A value of 1 provides basic information.
22736 A value greater than 1 provides more verbose information.
22737 @item show debug symtab-create
22738 Show the current state of symbol table creation debugging.
22739 @item set debug target
22740 @cindex target debugging info
22741 Turns on or off display of @value{GDBN} target debugging info. This info
22742 includes what is going on at the target level of GDB, as it happens. The
22743 default is 0. Set it to 1 to track events, and to 2 to also track the
22744 value of large memory transfers. Changes to this flag do not take effect
22745 until the next time you connect to a target or use the @code{run} command.
22746 @item show debug target
22747 Displays the current state of displaying @value{GDBN} target debugging
22749 @item set debug timestamp
22750 @cindex timestampping debugging info
22751 Turns on or off display of timestamps with @value{GDBN} debugging info.
22752 When enabled, seconds and microseconds are displayed before each debugging
22754 @item show debug timestamp
22755 Displays the current state of displaying timestamps with @value{GDBN}
22757 @item set debugvarobj
22758 @cindex variable object debugging info
22759 Turns on or off display of @value{GDBN} variable object debugging
22760 info. The default is off.
22761 @item show debugvarobj
22762 Displays the current state of displaying @value{GDBN} variable object
22764 @item set debug xml
22765 @cindex XML parser debugging
22766 Turns on or off debugging messages for built-in XML parsers.
22767 @item show debug xml
22768 Displays the current state of XML debugging messages.
22771 @node Other Misc Settings
22772 @section Other Miscellaneous Settings
22773 @cindex miscellaneous settings
22776 @kindex set interactive-mode
22777 @item set interactive-mode
22778 If @code{on}, forces @value{GDBN} to assume that GDB was started
22779 in a terminal. In practice, this means that @value{GDBN} should wait
22780 for the user to answer queries generated by commands entered at
22781 the command prompt. If @code{off}, forces @value{GDBN} to operate
22782 in the opposite mode, and it uses the default answers to all queries.
22783 If @code{auto} (the default), @value{GDBN} tries to determine whether
22784 its standard input is a terminal, and works in interactive-mode if it
22785 is, non-interactively otherwise.
22787 In the vast majority of cases, the debugger should be able to guess
22788 correctly which mode should be used. But this setting can be useful
22789 in certain specific cases, such as running a MinGW @value{GDBN}
22790 inside a cygwin window.
22792 @kindex show interactive-mode
22793 @item show interactive-mode
22794 Displays whether the debugger is operating in interactive mode or not.
22797 @node Extending GDB
22798 @chapter Extending @value{GDBN}
22799 @cindex extending GDB
22801 @value{GDBN} provides three mechanisms for extension. The first is based
22802 on composition of @value{GDBN} commands, the second is based on the
22803 Python scripting language, and the third is for defining new aliases of
22806 To facilitate the use of the first two extensions, @value{GDBN} is capable
22807 of evaluating the contents of a file. When doing so, @value{GDBN}
22808 can recognize which scripting language is being used by looking at
22809 the filename extension. Files with an unrecognized filename extension
22810 are always treated as a @value{GDBN} Command Files.
22811 @xref{Command Files,, Command files}.
22813 You can control how @value{GDBN} evaluates these files with the following
22817 @kindex set script-extension
22818 @kindex show script-extension
22819 @item set script-extension off
22820 All scripts are always evaluated as @value{GDBN} Command Files.
22822 @item set script-extension soft
22823 The debugger determines the scripting language based on filename
22824 extension. If this scripting language is supported, @value{GDBN}
22825 evaluates the script using that language. Otherwise, it evaluates
22826 the file as a @value{GDBN} Command File.
22828 @item set script-extension strict
22829 The debugger determines the scripting language based on filename
22830 extension, and evaluates the script using that language. If the
22831 language is not supported, then the evaluation fails.
22833 @item show script-extension
22834 Display the current value of the @code{script-extension} option.
22839 * Sequences:: Canned Sequences of Commands
22840 * Python:: Scripting @value{GDBN} using Python
22841 * Aliases:: Creating new spellings of existing commands
22845 @section Canned Sequences of Commands
22847 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
22848 Command Lists}), @value{GDBN} provides two ways to store sequences of
22849 commands for execution as a unit: user-defined commands and command
22853 * Define:: How to define your own commands
22854 * Hooks:: Hooks for user-defined commands
22855 * Command Files:: How to write scripts of commands to be stored in a file
22856 * Output:: Commands for controlled output
22860 @subsection User-defined Commands
22862 @cindex user-defined command
22863 @cindex arguments, to user-defined commands
22864 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
22865 which you assign a new name as a command. This is done with the
22866 @code{define} command. User commands may accept up to 10 arguments
22867 separated by whitespace. Arguments are accessed within the user command
22868 via @code{$arg0@dots{}$arg9}. A trivial example:
22872 print $arg0 + $arg1 + $arg2
22877 To execute the command use:
22884 This defines the command @code{adder}, which prints the sum of
22885 its three arguments. Note the arguments are text substitutions, so they may
22886 reference variables, use complex expressions, or even perform inferior
22889 @cindex argument count in user-defined commands
22890 @cindex how many arguments (user-defined commands)
22891 In addition, @code{$argc} may be used to find out how many arguments have
22892 been passed. This expands to a number in the range 0@dots{}10.
22897 print $arg0 + $arg1
22900 print $arg0 + $arg1 + $arg2
22908 @item define @var{commandname}
22909 Define a command named @var{commandname}. If there is already a command
22910 by that name, you are asked to confirm that you want to redefine it.
22911 @var{commandname} may be a bare command name consisting of letters,
22912 numbers, dashes, and underscores. It may also start with any predefined
22913 prefix command. For example, @samp{define target my-target} creates
22914 a user-defined @samp{target my-target} command.
22916 The definition of the command is made up of other @value{GDBN} command lines,
22917 which are given following the @code{define} command. The end of these
22918 commands is marked by a line containing @code{end}.
22921 @kindex end@r{ (user-defined commands)}
22922 @item document @var{commandname}
22923 Document the user-defined command @var{commandname}, so that it can be
22924 accessed by @code{help}. The command @var{commandname} must already be
22925 defined. This command reads lines of documentation just as @code{define}
22926 reads the lines of the command definition, ending with @code{end}.
22927 After the @code{document} command is finished, @code{help} on command
22928 @var{commandname} displays the documentation you have written.
22930 You may use the @code{document} command again to change the
22931 documentation of a command. Redefining the command with @code{define}
22932 does not change the documentation.
22934 @kindex dont-repeat
22935 @cindex don't repeat command
22937 Used inside a user-defined command, this tells @value{GDBN} that this
22938 command should not be repeated when the user hits @key{RET}
22939 (@pxref{Command Syntax, repeat last command}).
22941 @kindex help user-defined
22942 @item help user-defined
22943 List all user-defined commands and all python commands defined in class
22944 COMAND_USER. The first line of the documentation or docstring is
22949 @itemx show user @var{commandname}
22950 Display the @value{GDBN} commands used to define @var{commandname} (but
22951 not its documentation). If no @var{commandname} is given, display the
22952 definitions for all user-defined commands.
22953 This does not work for user-defined python commands.
22955 @cindex infinite recursion in user-defined commands
22956 @kindex show max-user-call-depth
22957 @kindex set max-user-call-depth
22958 @item show max-user-call-depth
22959 @itemx set max-user-call-depth
22960 The value of @code{max-user-call-depth} controls how many recursion
22961 levels are allowed in user-defined commands before @value{GDBN} suspects an
22962 infinite recursion and aborts the command.
22963 This does not apply to user-defined python commands.
22966 In addition to the above commands, user-defined commands frequently
22967 use control flow commands, described in @ref{Command Files}.
22969 When user-defined commands are executed, the
22970 commands of the definition are not printed. An error in any command
22971 stops execution of the user-defined command.
22973 If used interactively, commands that would ask for confirmation proceed
22974 without asking when used inside a user-defined command. Many @value{GDBN}
22975 commands that normally print messages to say what they are doing omit the
22976 messages when used in a user-defined command.
22979 @subsection User-defined Command Hooks
22980 @cindex command hooks
22981 @cindex hooks, for commands
22982 @cindex hooks, pre-command
22985 You may define @dfn{hooks}, which are a special kind of user-defined
22986 command. Whenever you run the command @samp{foo}, if the user-defined
22987 command @samp{hook-foo} exists, it is executed (with no arguments)
22988 before that command.
22990 @cindex hooks, post-command
22992 A hook may also be defined which is run after the command you executed.
22993 Whenever you run the command @samp{foo}, if the user-defined command
22994 @samp{hookpost-foo} exists, it is executed (with no arguments) after
22995 that command. Post-execution hooks may exist simultaneously with
22996 pre-execution hooks, for the same command.
22998 It is valid for a hook to call the command which it hooks. If this
22999 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
23001 @c It would be nice if hookpost could be passed a parameter indicating
23002 @c if the command it hooks executed properly or not. FIXME!
23004 @kindex stop@r{, a pseudo-command}
23005 In addition, a pseudo-command, @samp{stop} exists. Defining
23006 (@samp{hook-stop}) makes the associated commands execute every time
23007 execution stops in your program: before breakpoint commands are run,
23008 displays are printed, or the stack frame is printed.
23010 For example, to ignore @code{SIGALRM} signals while
23011 single-stepping, but treat them normally during normal execution,
23016 handle SIGALRM nopass
23020 handle SIGALRM pass
23023 define hook-continue
23024 handle SIGALRM pass
23028 As a further example, to hook at the beginning and end of the @code{echo}
23029 command, and to add extra text to the beginning and end of the message,
23037 define hookpost-echo
23041 (@value{GDBP}) echo Hello World
23042 <<<---Hello World--->>>
23047 You can define a hook for any single-word command in @value{GDBN}, but
23048 not for command aliases; you should define a hook for the basic command
23049 name, e.g.@: @code{backtrace} rather than @code{bt}.
23050 @c FIXME! So how does Joe User discover whether a command is an alias
23052 You can hook a multi-word command by adding @code{hook-} or
23053 @code{hookpost-} to the last word of the command, e.g.@:
23054 @samp{define target hook-remote} to add a hook to @samp{target remote}.
23056 If an error occurs during the execution of your hook, execution of
23057 @value{GDBN} commands stops and @value{GDBN} issues a prompt
23058 (before the command that you actually typed had a chance to run).
23060 If you try to define a hook which does not match any known command, you
23061 get a warning from the @code{define} command.
23063 @node Command Files
23064 @subsection Command Files
23066 @cindex command files
23067 @cindex scripting commands
23068 A command file for @value{GDBN} is a text file made of lines that are
23069 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
23070 also be included. An empty line in a command file does nothing; it
23071 does not mean to repeat the last command, as it would from the
23074 You can request the execution of a command file with the @code{source}
23075 command. Note that the @code{source} command is also used to evaluate
23076 scripts that are not Command Files. The exact behavior can be configured
23077 using the @code{script-extension} setting.
23078 @xref{Extending GDB,, Extending GDB}.
23082 @cindex execute commands from a file
23083 @item source [-s] [-v] @var{filename}
23084 Execute the command file @var{filename}.
23087 The lines in a command file are generally executed sequentially,
23088 unless the order of execution is changed by one of the
23089 @emph{flow-control commands} described below. The commands are not
23090 printed as they are executed. An error in any command terminates
23091 execution of the command file and control is returned to the console.
23093 @value{GDBN} first searches for @var{filename} in the current directory.
23094 If the file is not found there, and @var{filename} does not specify a
23095 directory, then @value{GDBN} also looks for the file on the source search path
23096 (specified with the @samp{directory} command);
23097 except that @file{$cdir} is not searched because the compilation directory
23098 is not relevant to scripts.
23100 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
23101 on the search path even if @var{filename} specifies a directory.
23102 The search is done by appending @var{filename} to each element of the
23103 search path. So, for example, if @var{filename} is @file{mylib/myscript}
23104 and the search path contains @file{/home/user} then @value{GDBN} will
23105 look for the script @file{/home/user/mylib/myscript}.
23106 The search is also done if @var{filename} is an absolute path.
23107 For example, if @var{filename} is @file{/tmp/myscript} and
23108 the search path contains @file{/home/user} then @value{GDBN} will
23109 look for the script @file{/home/user/tmp/myscript}.
23110 For DOS-like systems, if @var{filename} contains a drive specification,
23111 it is stripped before concatenation. For example, if @var{filename} is
23112 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
23113 will look for the script @file{c:/tmp/myscript}.
23115 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
23116 each command as it is executed. The option must be given before
23117 @var{filename}, and is interpreted as part of the filename anywhere else.
23119 Commands that would ask for confirmation if used interactively proceed
23120 without asking when used in a command file. Many @value{GDBN} commands that
23121 normally print messages to say what they are doing omit the messages
23122 when called from command files.
23124 @value{GDBN} also accepts command input from standard input. In this
23125 mode, normal output goes to standard output and error output goes to
23126 standard error. Errors in a command file supplied on standard input do
23127 not terminate execution of the command file---execution continues with
23131 gdb < cmds > log 2>&1
23134 (The syntax above will vary depending on the shell used.) This example
23135 will execute commands from the file @file{cmds}. All output and errors
23136 would be directed to @file{log}.
23138 Since commands stored on command files tend to be more general than
23139 commands typed interactively, they frequently need to deal with
23140 complicated situations, such as different or unexpected values of
23141 variables and symbols, changes in how the program being debugged is
23142 built, etc. @value{GDBN} provides a set of flow-control commands to
23143 deal with these complexities. Using these commands, you can write
23144 complex scripts that loop over data structures, execute commands
23145 conditionally, etc.
23152 This command allows to include in your script conditionally executed
23153 commands. The @code{if} command takes a single argument, which is an
23154 expression to evaluate. It is followed by a series of commands that
23155 are executed only if the expression is true (its value is nonzero).
23156 There can then optionally be an @code{else} line, followed by a series
23157 of commands that are only executed if the expression was false. The
23158 end of the list is marked by a line containing @code{end}.
23162 This command allows to write loops. Its syntax is similar to
23163 @code{if}: the command takes a single argument, which is an expression
23164 to evaluate, and must be followed by the commands to execute, one per
23165 line, terminated by an @code{end}. These commands are called the
23166 @dfn{body} of the loop. The commands in the body of @code{while} are
23167 executed repeatedly as long as the expression evaluates to true.
23171 This command exits the @code{while} loop in whose body it is included.
23172 Execution of the script continues after that @code{while}s @code{end}
23175 @kindex loop_continue
23176 @item loop_continue
23177 This command skips the execution of the rest of the body of commands
23178 in the @code{while} loop in whose body it is included. Execution
23179 branches to the beginning of the @code{while} loop, where it evaluates
23180 the controlling expression.
23182 @kindex end@r{ (if/else/while commands)}
23184 Terminate the block of commands that are the body of @code{if},
23185 @code{else}, or @code{while} flow-control commands.
23190 @subsection Commands for Controlled Output
23192 During the execution of a command file or a user-defined command, normal
23193 @value{GDBN} output is suppressed; the only output that appears is what is
23194 explicitly printed by the commands in the definition. This section
23195 describes three commands useful for generating exactly the output you
23200 @item echo @var{text}
23201 @c I do not consider backslash-space a standard C escape sequence
23202 @c because it is not in ANSI.
23203 Print @var{text}. Nonprinting characters can be included in
23204 @var{text} using C escape sequences, such as @samp{\n} to print a
23205 newline. @strong{No newline is printed unless you specify one.}
23206 In addition to the standard C escape sequences, a backslash followed
23207 by a space stands for a space. This is useful for displaying a
23208 string with spaces at the beginning or the end, since leading and
23209 trailing spaces are otherwise trimmed from all arguments.
23210 To print @samp{@w{ }and foo =@w{ }}, use the command
23211 @samp{echo \@w{ }and foo = \@w{ }}.
23213 A backslash at the end of @var{text} can be used, as in C, to continue
23214 the command onto subsequent lines. For example,
23217 echo This is some text\n\
23218 which is continued\n\
23219 onto several lines.\n
23222 produces the same output as
23225 echo This is some text\n
23226 echo which is continued\n
23227 echo onto several lines.\n
23231 @item output @var{expression}
23232 Print the value of @var{expression} and nothing but that value: no
23233 newlines, no @samp{$@var{nn} = }. The value is not entered in the
23234 value history either. @xref{Expressions, ,Expressions}, for more information
23237 @item output/@var{fmt} @var{expression}
23238 Print the value of @var{expression} in format @var{fmt}. You can use
23239 the same formats as for @code{print}. @xref{Output Formats,,Output
23240 Formats}, for more information.
23243 @item printf @var{template}, @var{expressions}@dots{}
23244 Print the values of one or more @var{expressions} under the control of
23245 the string @var{template}. To print several values, make
23246 @var{expressions} be a comma-separated list of individual expressions,
23247 which may be either numbers or pointers. Their values are printed as
23248 specified by @var{template}, exactly as a C program would do by
23249 executing the code below:
23252 printf (@var{template}, @var{expressions}@dots{});
23255 As in @code{C} @code{printf}, ordinary characters in @var{template}
23256 are printed verbatim, while @dfn{conversion specification} introduced
23257 by the @samp{%} character cause subsequent @var{expressions} to be
23258 evaluated, their values converted and formatted according to type and
23259 style information encoded in the conversion specifications, and then
23262 For example, you can print two values in hex like this:
23265 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
23268 @code{printf} supports all the standard @code{C} conversion
23269 specifications, including the flags and modifiers between the @samp{%}
23270 character and the conversion letter, with the following exceptions:
23274 The argument-ordering modifiers, such as @samp{2$}, are not supported.
23277 The modifier @samp{*} is not supported for specifying precision or
23281 The @samp{'} flag (for separation of digits into groups according to
23282 @code{LC_NUMERIC'}) is not supported.
23285 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
23289 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
23292 The conversion letters @samp{a} and @samp{A} are not supported.
23296 Note that the @samp{ll} type modifier is supported only if the
23297 underlying @code{C} implementation used to build @value{GDBN} supports
23298 the @code{long long int} type, and the @samp{L} type modifier is
23299 supported only if @code{long double} type is available.
23301 As in @code{C}, @code{printf} supports simple backslash-escape
23302 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
23303 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
23304 single character. Octal and hexadecimal escape sequences are not
23307 Additionally, @code{printf} supports conversion specifications for DFP
23308 (@dfn{Decimal Floating Point}) types using the following length modifiers
23309 together with a floating point specifier.
23314 @samp{H} for printing @code{Decimal32} types.
23317 @samp{D} for printing @code{Decimal64} types.
23320 @samp{DD} for printing @code{Decimal128} types.
23323 If the underlying @code{C} implementation used to build @value{GDBN} has
23324 support for the three length modifiers for DFP types, other modifiers
23325 such as width and precision will also be available for @value{GDBN} to use.
23327 In case there is no such @code{C} support, no additional modifiers will be
23328 available and the value will be printed in the standard way.
23330 Here's an example of printing DFP types using the above conversion letters:
23332 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
23336 @item eval @var{template}, @var{expressions}@dots{}
23337 Convert the values of one or more @var{expressions} under the control of
23338 the string @var{template} to a command line, and call it.
23343 @section Scripting @value{GDBN} using Python
23344 @cindex python scripting
23345 @cindex scripting with python
23347 You can script @value{GDBN} using the @uref{http://www.python.org/,
23348 Python programming language}. This feature is available only if
23349 @value{GDBN} was configured using @option{--with-python}.
23351 @cindex python directory
23352 Python scripts used by @value{GDBN} should be installed in
23353 @file{@var{data-directory}/python}, where @var{data-directory} is
23354 the data directory as determined at @value{GDBN} startup (@pxref{Data Files}).
23355 This directory, known as the @dfn{python directory},
23356 is automatically added to the Python Search Path in order to allow
23357 the Python interpreter to locate all scripts installed at this location.
23359 Additionally, @value{GDBN} commands and convenience functions which
23360 are written in Python and are located in the
23361 @file{@var{data-directory}/python/gdb/command} or
23362 @file{@var{data-directory}/python/gdb/function} directories are
23363 automatically imported when @value{GDBN} starts.
23366 * Python Commands:: Accessing Python from @value{GDBN}.
23367 * Python API:: Accessing @value{GDBN} from Python.
23368 * Python Auto-loading:: Automatically loading Python code.
23369 * Python modules:: Python modules provided by @value{GDBN}.
23372 @node Python Commands
23373 @subsection Python Commands
23374 @cindex python commands
23375 @cindex commands to access python
23377 @value{GDBN} provides two commands for accessing the Python interpreter,
23378 and one related setting:
23381 @kindex python-interactive
23383 @item python-interactive @r{[}@var{command}@r{]}
23384 @itemx pi @r{[}@var{command}@r{]}
23385 Without an argument, the @code{python-interactive} command can be used
23386 to start an interactive Python prompt. To return to @value{GDBN},
23387 type the @code{EOF} character (e.g., @kbd{Ctrl-D} on an empty prompt).
23389 Alternatively, a single-line Python command can be given as an
23390 argument and evaluated. If the command is an expression, the result
23391 will be printed; otherwise, nothing will be printed. For example:
23394 (@value{GDBP}) python-interactive 2 + 3
23400 @item python @r{[}@var{command}@r{]}
23401 @itemx py @r{[}@var{command}@r{]}
23402 The @code{python} command can be used to evaluate Python code.
23404 If given an argument, the @code{python} command will evaluate the
23405 argument as a Python command. For example:
23408 (@value{GDBP}) python print 23
23412 If you do not provide an argument to @code{python}, it will act as a
23413 multi-line command, like @code{define}. In this case, the Python
23414 script is made up of subsequent command lines, given after the
23415 @code{python} command. This command list is terminated using a line
23416 containing @code{end}. For example:
23419 (@value{GDBP}) python
23421 End with a line saying just "end".
23427 @kindex set python print-stack
23428 @item set python print-stack
23429 By default, @value{GDBN} will print only the message component of a
23430 Python exception when an error occurs in a Python script. This can be
23431 controlled using @code{set python print-stack}: if @code{full}, then
23432 full Python stack printing is enabled; if @code{none}, then Python stack
23433 and message printing is disabled; if @code{message}, the default, only
23434 the message component of the error is printed.
23437 It is also possible to execute a Python script from the @value{GDBN}
23441 @item source @file{script-name}
23442 The script name must end with @samp{.py} and @value{GDBN} must be configured
23443 to recognize the script language based on filename extension using
23444 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
23446 @item python execfile ("script-name")
23447 This method is based on the @code{execfile} Python built-in function,
23448 and thus is always available.
23452 @subsection Python API
23454 @cindex programming in python
23456 You can get quick online help for @value{GDBN}'s Python API by issuing
23457 the command @w{@kbd{python help (gdb)}}.
23459 Functions and methods which have two or more optional arguments allow
23460 them to be specified using keyword syntax. This allows passing some
23461 optional arguments while skipping others. Example:
23462 @w{@code{gdb.some_function ('foo', bar = 1, baz = 2)}}.
23465 * Basic Python:: Basic Python Functions.
23466 * Exception Handling:: How Python exceptions are translated.
23467 * Values From Inferior:: Python representation of values.
23468 * Types In Python:: Python representation of types.
23469 * Pretty Printing API:: Pretty-printing values.
23470 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
23471 * Writing a Pretty-Printer:: Writing a Pretty-Printer.
23472 * Type Printing API:: Pretty-printing types.
23473 * Frame Filter API:: Filtering Frames.
23474 * Frame Decorator API:: Decorating Frames.
23475 * Writing a Frame Filter:: Writing a Frame Filter.
23476 * Inferiors In Python:: Python representation of inferiors (processes)
23477 * Events In Python:: Listening for events from @value{GDBN}.
23478 * Threads In Python:: Accessing inferior threads from Python.
23479 * Commands In Python:: Implementing new commands in Python.
23480 * Parameters In Python:: Adding new @value{GDBN} parameters.
23481 * Functions In Python:: Writing new convenience functions.
23482 * Progspaces In Python:: Program spaces.
23483 * Objfiles In Python:: Object files.
23484 * Frames In Python:: Accessing inferior stack frames from Python.
23485 * Blocks In Python:: Accessing blocks from Python.
23486 * Symbols In Python:: Python representation of symbols.
23487 * Symbol Tables In Python:: Python representation of symbol tables.
23488 * Breakpoints In Python:: Manipulating breakpoints using Python.
23489 * Finish Breakpoints in Python:: Setting Breakpoints on function return
23491 * Lazy Strings In Python:: Python representation of lazy strings.
23492 * Architectures In Python:: Python representation of architectures.
23496 @subsubsection Basic Python
23498 @cindex python stdout
23499 @cindex python pagination
23500 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
23501 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
23502 A Python program which outputs to one of these streams may have its
23503 output interrupted by the user (@pxref{Screen Size}). In this
23504 situation, a Python @code{KeyboardInterrupt} exception is thrown.
23506 Some care must be taken when writing Python code to run in
23507 @value{GDBN}. Two things worth noting in particular:
23511 @value{GDBN} install handlers for @code{SIGCHLD} and @code{SIGINT}.
23512 Python code must not override these, or even change the options using
23513 @code{sigaction}. If your program changes the handling of these
23514 signals, @value{GDBN} will most likely stop working correctly. Note
23515 that it is unfortunately common for GUI toolkits to install a
23516 @code{SIGCHLD} handler.
23519 @value{GDBN} takes care to mark its internal file descriptors as
23520 close-on-exec. However, this cannot be done in a thread-safe way on
23521 all platforms. Your Python programs should be aware of this and
23522 should both create new file descriptors with the close-on-exec flag
23523 set and arrange to close unneeded file descriptors before starting a
23527 @cindex python functions
23528 @cindex python module
23530 @value{GDBN} introduces a new Python module, named @code{gdb}. All
23531 methods and classes added by @value{GDBN} are placed in this module.
23532 @value{GDBN} automatically @code{import}s the @code{gdb} module for
23533 use in all scripts evaluated by the @code{python} command.
23535 @findex gdb.PYTHONDIR
23536 @defvar gdb.PYTHONDIR
23537 A string containing the python directory (@pxref{Python}).
23540 @findex gdb.execute
23541 @defun gdb.execute (command @r{[}, from_tty @r{[}, to_string@r{]]})
23542 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
23543 If a GDB exception happens while @var{command} runs, it is
23544 translated as described in @ref{Exception Handling,,Exception Handling}.
23546 @var{from_tty} specifies whether @value{GDBN} ought to consider this
23547 command as having originated from the user invoking it interactively.
23548 It must be a boolean value. If omitted, it defaults to @code{False}.
23550 By default, any output produced by @var{command} is sent to
23551 @value{GDBN}'s standard output. If the @var{to_string} parameter is
23552 @code{True}, then output will be collected by @code{gdb.execute} and
23553 returned as a string. The default is @code{False}, in which case the
23554 return value is @code{None}. If @var{to_string} is @code{True}, the
23555 @value{GDBN} virtual terminal will be temporarily set to unlimited width
23556 and height, and its pagination will be disabled; @pxref{Screen Size}.
23559 @findex gdb.breakpoints
23560 @defun gdb.breakpoints ()
23561 Return a sequence holding all of @value{GDBN}'s breakpoints.
23562 @xref{Breakpoints In Python}, for more information.
23565 @findex gdb.parameter
23566 @defun gdb.parameter (parameter)
23567 Return the value of a @value{GDBN} parameter. @var{parameter} is a
23568 string naming the parameter to look up; @var{parameter} may contain
23569 spaces if the parameter has a multi-part name. For example,
23570 @samp{print object} is a valid parameter name.
23572 If the named parameter does not exist, this function throws a
23573 @code{gdb.error} (@pxref{Exception Handling}). Otherwise, the
23574 parameter's value is converted to a Python value of the appropriate
23575 type, and returned.
23578 @findex gdb.history
23579 @defun gdb.history (number)
23580 Return a value from @value{GDBN}'s value history (@pxref{Value
23581 History}). @var{number} indicates which history element to return.
23582 If @var{number} is negative, then @value{GDBN} will take its absolute value
23583 and count backward from the last element (i.e., the most recent element) to
23584 find the value to return. If @var{number} is zero, then @value{GDBN} will
23585 return the most recent element. If the element specified by @var{number}
23586 doesn't exist in the value history, a @code{gdb.error} exception will be
23589 If no exception is raised, the return value is always an instance of
23590 @code{gdb.Value} (@pxref{Values From Inferior}).
23593 @findex gdb.parse_and_eval
23594 @defun gdb.parse_and_eval (expression)
23595 Parse @var{expression} as an expression in the current language,
23596 evaluate it, and return the result as a @code{gdb.Value}.
23597 @var{expression} must be a string.
23599 This function can be useful when implementing a new command
23600 (@pxref{Commands In Python}), as it provides a way to parse the
23601 command's argument as an expression. It is also useful simply to
23602 compute values, for example, it is the only way to get the value of a
23603 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
23606 @findex gdb.find_pc_line
23607 @defun gdb.find_pc_line (pc)
23608 Return the @code{gdb.Symtab_and_line} object corresponding to the
23609 @var{pc} value. @xref{Symbol Tables In Python}. If an invalid
23610 value of @var{pc} is passed as an argument, then the @code{symtab} and
23611 @code{line} attributes of the returned @code{gdb.Symtab_and_line} object
23612 will be @code{None} and 0 respectively.
23615 @findex gdb.post_event
23616 @defun gdb.post_event (event)
23617 Put @var{event}, a callable object taking no arguments, into
23618 @value{GDBN}'s internal event queue. This callable will be invoked at
23619 some later point, during @value{GDBN}'s event processing. Events
23620 posted using @code{post_event} will be run in the order in which they
23621 were posted; however, there is no way to know when they will be
23622 processed relative to other events inside @value{GDBN}.
23624 @value{GDBN} is not thread-safe. If your Python program uses multiple
23625 threads, you must be careful to only call @value{GDBN}-specific
23626 functions in the main @value{GDBN} thread. @code{post_event} ensures
23630 (@value{GDBP}) python
23634 > def __init__(self, message):
23635 > self.message = message;
23636 > def __call__(self):
23637 > gdb.write(self.message)
23639 >class MyThread1 (threading.Thread):
23641 > gdb.post_event(Writer("Hello "))
23643 >class MyThread2 (threading.Thread):
23645 > gdb.post_event(Writer("World\n"))
23647 >MyThread1().start()
23648 >MyThread2().start()
23650 (@value{GDBP}) Hello World
23655 @defun gdb.write (string @r{[}, stream{]})
23656 Print a string to @value{GDBN}'s paginated output stream. The
23657 optional @var{stream} determines the stream to print to. The default
23658 stream is @value{GDBN}'s standard output stream. Possible stream
23665 @value{GDBN}'s standard output stream.
23670 @value{GDBN}'s standard error stream.
23675 @value{GDBN}'s log stream (@pxref{Logging Output}).
23678 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
23679 call this function and will automatically direct the output to the
23684 @defun gdb.flush ()
23685 Flush the buffer of a @value{GDBN} paginated stream so that the
23686 contents are displayed immediately. @value{GDBN} will flush the
23687 contents of a stream automatically when it encounters a newline in the
23688 buffer. The optional @var{stream} determines the stream to flush. The
23689 default stream is @value{GDBN}'s standard output stream. Possible
23696 @value{GDBN}'s standard output stream.
23701 @value{GDBN}'s standard error stream.
23706 @value{GDBN}'s log stream (@pxref{Logging Output}).
23710 Flushing @code{sys.stdout} or @code{sys.stderr} will automatically
23711 call this function for the relevant stream.
23714 @findex gdb.target_charset
23715 @defun gdb.target_charset ()
23716 Return the name of the current target character set (@pxref{Character
23717 Sets}). This differs from @code{gdb.parameter('target-charset')} in
23718 that @samp{auto} is never returned.
23721 @findex gdb.target_wide_charset
23722 @defun gdb.target_wide_charset ()
23723 Return the name of the current target wide character set
23724 (@pxref{Character Sets}). This differs from
23725 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
23729 @findex gdb.solib_name
23730 @defun gdb.solib_name (address)
23731 Return the name of the shared library holding the given @var{address}
23732 as a string, or @code{None}.
23735 @findex gdb.decode_line
23736 @defun gdb.decode_line @r{[}expression@r{]}
23737 Return locations of the line specified by @var{expression}, or of the
23738 current line if no argument was given. This function returns a Python
23739 tuple containing two elements. The first element contains a string
23740 holding any unparsed section of @var{expression} (or @code{None} if
23741 the expression has been fully parsed). The second element contains
23742 either @code{None} or another tuple that contains all the locations
23743 that match the expression represented as @code{gdb.Symtab_and_line}
23744 objects (@pxref{Symbol Tables In Python}). If @var{expression} is
23745 provided, it is decoded the way that @value{GDBN}'s inbuilt
23746 @code{break} or @code{edit} commands do (@pxref{Specify Location}).
23749 @defun gdb.prompt_hook (current_prompt)
23750 @anchor{prompt_hook}
23752 If @var{prompt_hook} is callable, @value{GDBN} will call the method
23753 assigned to this operation before a prompt is displayed by
23756 The parameter @code{current_prompt} contains the current @value{GDBN}
23757 prompt. This method must return a Python string, or @code{None}. If
23758 a string is returned, the @value{GDBN} prompt will be set to that
23759 string. If @code{None} is returned, @value{GDBN} will continue to use
23760 the current prompt.
23762 Some prompts cannot be substituted in @value{GDBN}. Secondary prompts
23763 such as those used by readline for command input, and annotation
23764 related prompts are prohibited from being changed.
23767 @node Exception Handling
23768 @subsubsection Exception Handling
23769 @cindex python exceptions
23770 @cindex exceptions, python
23772 When executing the @code{python} command, Python exceptions
23773 uncaught within the Python code are translated to calls to
23774 @value{GDBN} error-reporting mechanism. If the command that called
23775 @code{python} does not handle the error, @value{GDBN} will
23776 terminate it and print an error message containing the Python
23777 exception name, the associated value, and the Python call stack
23778 backtrace at the point where the exception was raised. Example:
23781 (@value{GDBP}) python print foo
23782 Traceback (most recent call last):
23783 File "<string>", line 1, in <module>
23784 NameError: name 'foo' is not defined
23787 @value{GDBN} errors that happen in @value{GDBN} commands invoked by
23788 Python code are converted to Python exceptions. The type of the
23789 Python exception depends on the error.
23793 This is the base class for most exceptions generated by @value{GDBN}.
23794 It is derived from @code{RuntimeError}, for compatibility with earlier
23795 versions of @value{GDBN}.
23797 If an error occurring in @value{GDBN} does not fit into some more
23798 specific category, then the generated exception will have this type.
23800 @item gdb.MemoryError
23801 This is a subclass of @code{gdb.error} which is thrown when an
23802 operation tried to access invalid memory in the inferior.
23804 @item KeyboardInterrupt
23805 User interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
23806 prompt) is translated to a Python @code{KeyboardInterrupt} exception.
23809 In all cases, your exception handler will see the @value{GDBN} error
23810 message as its value and the Python call stack backtrace at the Python
23811 statement closest to where the @value{GDBN} error occured as the
23814 @findex gdb.GdbError
23815 When implementing @value{GDBN} commands in Python via @code{gdb.Command},
23816 it is useful to be able to throw an exception that doesn't cause a
23817 traceback to be printed. For example, the user may have invoked the
23818 command incorrectly. Use the @code{gdb.GdbError} exception
23819 to handle this case. Example:
23823 >class HelloWorld (gdb.Command):
23824 > """Greet the whole world."""
23825 > def __init__ (self):
23826 > super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
23827 > def invoke (self, args, from_tty):
23828 > argv = gdb.string_to_argv (args)
23829 > if len (argv) != 0:
23830 > raise gdb.GdbError ("hello-world takes no arguments")
23831 > print "Hello, World!"
23834 (gdb) hello-world 42
23835 hello-world takes no arguments
23838 @node Values From Inferior
23839 @subsubsection Values From Inferior
23840 @cindex values from inferior, with Python
23841 @cindex python, working with values from inferior
23843 @cindex @code{gdb.Value}
23844 @value{GDBN} provides values it obtains from the inferior program in
23845 an object of type @code{gdb.Value}. @value{GDBN} uses this object
23846 for its internal bookkeeping of the inferior's values, and for
23847 fetching values when necessary.
23849 Inferior values that are simple scalars can be used directly in
23850 Python expressions that are valid for the value's data type. Here's
23851 an example for an integer or floating-point value @code{some_val}:
23858 As result of this, @code{bar} will also be a @code{gdb.Value} object
23859 whose values are of the same type as those of @code{some_val}.
23861 Inferior values that are structures or instances of some class can
23862 be accessed using the Python @dfn{dictionary syntax}. For example, if
23863 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
23864 can access its @code{foo} element with:
23867 bar = some_val['foo']
23870 Again, @code{bar} will also be a @code{gdb.Value} object.
23872 A @code{gdb.Value} that represents a function can be executed via
23873 inferior function call. Any arguments provided to the call must match
23874 the function's prototype, and must be provided in the order specified
23877 For example, @code{some_val} is a @code{gdb.Value} instance
23878 representing a function that takes two integers as arguments. To
23879 execute this function, call it like so:
23882 result = some_val (10,20)
23885 Any values returned from a function call will be stored as a
23888 The following attributes are provided:
23890 @defvar Value.address
23891 If this object is addressable, this read-only attribute holds a
23892 @code{gdb.Value} object representing the address. Otherwise,
23893 this attribute holds @code{None}.
23896 @cindex optimized out value in Python
23897 @defvar Value.is_optimized_out
23898 This read-only boolean attribute is true if the compiler optimized out
23899 this value, thus it is not available for fetching from the inferior.
23903 The type of this @code{gdb.Value}. The value of this attribute is a
23904 @code{gdb.Type} object (@pxref{Types In Python}).
23907 @defvar Value.dynamic_type
23908 The dynamic type of this @code{gdb.Value}. This uses C@t{++} run-time
23909 type information (@acronym{RTTI}) to determine the dynamic type of the
23910 value. If this value is of class type, it will return the class in
23911 which the value is embedded, if any. If this value is of pointer or
23912 reference to a class type, it will compute the dynamic type of the
23913 referenced object, and return a pointer or reference to that type,
23914 respectively. In all other cases, it will return the value's static
23917 Note that this feature will only work when debugging a C@t{++} program
23918 that includes @acronym{RTTI} for the object in question. Otherwise,
23919 it will just return the static type of the value as in @kbd{ptype foo}
23920 (@pxref{Symbols, ptype}).
23923 @defvar Value.is_lazy
23924 The value of this read-only boolean attribute is @code{True} if this
23925 @code{gdb.Value} has not yet been fetched from the inferior.
23926 @value{GDBN} does not fetch values until necessary, for efficiency.
23930 myval = gdb.parse_and_eval ('somevar')
23933 The value of @code{somevar} is not fetched at this time. It will be
23934 fetched when the value is needed, or when the @code{fetch_lazy}
23938 The following methods are provided:
23940 @defun Value.__init__ (@var{val})
23941 Many Python values can be converted directly to a @code{gdb.Value} via
23942 this object initializer. Specifically:
23945 @item Python boolean
23946 A Python boolean is converted to the boolean type from the current
23949 @item Python integer
23950 A Python integer is converted to the C @code{long} type for the
23951 current architecture.
23954 A Python long is converted to the C @code{long long} type for the
23955 current architecture.
23958 A Python float is converted to the C @code{double} type for the
23959 current architecture.
23961 @item Python string
23962 A Python string is converted to a target string, using the current
23965 @item @code{gdb.Value}
23966 If @code{val} is a @code{gdb.Value}, then a copy of the value is made.
23968 @item @code{gdb.LazyString}
23969 If @code{val} is a @code{gdb.LazyString} (@pxref{Lazy Strings In
23970 Python}), then the lazy string's @code{value} method is called, and
23971 its result is used.
23975 @defun Value.cast (type)
23976 Return a new instance of @code{gdb.Value} that is the result of
23977 casting this instance to the type described by @var{type}, which must
23978 be a @code{gdb.Type} object. If the cast cannot be performed for some
23979 reason, this method throws an exception.
23982 @defun Value.dereference ()
23983 For pointer data types, this method returns a new @code{gdb.Value} object
23984 whose contents is the object pointed to by the pointer. For example, if
23985 @code{foo} is a C pointer to an @code{int}, declared in your C program as
23992 then you can use the corresponding @code{gdb.Value} to access what
23993 @code{foo} points to like this:
23996 bar = foo.dereference ()
23999 The result @code{bar} will be a @code{gdb.Value} object holding the
24000 value pointed to by @code{foo}.
24002 A similar function @code{Value.referenced_value} exists which also
24003 returns @code{gdb.Value} objects corresonding to the values pointed to
24004 by pointer values (and additionally, values referenced by reference
24005 values). However, the behavior of @code{Value.dereference}
24006 differs from @code{Value.referenced_value} by the fact that the
24007 behavior of @code{Value.dereference} is identical to applying the C
24008 unary operator @code{*} on a given value. For example, consider a
24009 reference to a pointer @code{ptrref}, declared in your C@t{++} program
24013 typedef int *intptr;
24017 intptr &ptrref = ptr;
24020 Though @code{ptrref} is a reference value, one can apply the method
24021 @code{Value.dereference} to the @code{gdb.Value} object corresponding
24022 to it and obtain a @code{gdb.Value} which is identical to that
24023 corresponding to @code{val}. However, if you apply the method
24024 @code{Value.referenced_value}, the result would be a @code{gdb.Value}
24025 object identical to that corresponding to @code{ptr}.
24028 py_ptrref = gdb.parse_and_eval ("ptrref")
24029 py_val = py_ptrref.dereference ()
24030 py_ptr = py_ptrref.referenced_value ()
24033 The @code{gdb.Value} object @code{py_val} is identical to that
24034 corresponding to @code{val}, and @code{py_ptr} is identical to that
24035 corresponding to @code{ptr}. In general, @code{Value.dereference} can
24036 be applied whenever the C unary operator @code{*} can be applied
24037 to the corresponding C value. For those cases where applying both
24038 @code{Value.dereference} and @code{Value.referenced_value} is allowed,
24039 the results obtained need not be identical (as we have seen in the above
24040 example). The results are however identical when applied on
24041 @code{gdb.Value} objects corresponding to pointers (@code{gdb.Value}
24042 objects with type code @code{TYPE_CODE_PTR}) in a C/C@t{++} program.
24045 @defun Value.referenced_value ()
24046 For pointer or reference data types, this method returns a new
24047 @code{gdb.Value} object corresponding to the value referenced by the
24048 pointer/reference value. For pointer data types,
24049 @code{Value.dereference} and @code{Value.referenced_value} produce
24050 identical results. The difference between these methods is that
24051 @code{Value.dereference} cannot get the values referenced by reference
24052 values. For example, consider a reference to an @code{int}, declared
24053 in your C@t{++} program as
24061 then applying @code{Value.dereference} to the @code{gdb.Value} object
24062 corresponding to @code{ref} will result in an error, while applying
24063 @code{Value.referenced_value} will result in a @code{gdb.Value} object
24064 identical to that corresponding to @code{val}.
24067 py_ref = gdb.parse_and_eval ("ref")
24068 er_ref = py_ref.dereference () # Results in error
24069 py_val = py_ref.referenced_value () # Returns the referenced value
24072 The @code{gdb.Value} object @code{py_val} is identical to that
24073 corresponding to @code{val}.
24076 @defun Value.dynamic_cast (type)
24077 Like @code{Value.cast}, but works as if the C@t{++} @code{dynamic_cast}
24078 operator were used. Consult a C@t{++} reference for details.
24081 @defun Value.reinterpret_cast (type)
24082 Like @code{Value.cast}, but works as if the C@t{++} @code{reinterpret_cast}
24083 operator were used. Consult a C@t{++} reference for details.
24086 @defun Value.string (@r{[}encoding@r{[}, errors@r{[}, length@r{]]]})
24087 If this @code{gdb.Value} represents a string, then this method
24088 converts the contents to a Python string. Otherwise, this method will
24089 throw an exception.
24091 Strings are recognized in a language-specific way; whether a given
24092 @code{gdb.Value} represents a string is determined by the current
24095 For C-like languages, a value is a string if it is a pointer to or an
24096 array of characters or ints. The string is assumed to be terminated
24097 by a zero of the appropriate width. However if the optional length
24098 argument is given, the string will be converted to that given length,
24099 ignoring any embedded zeros that the string may contain.
24101 If the optional @var{encoding} argument is given, it must be a string
24102 naming the encoding of the string in the @code{gdb.Value}, such as
24103 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
24104 the same encodings as the corresponding argument to Python's
24105 @code{string.decode} method, and the Python codec machinery will be used
24106 to convert the string. If @var{encoding} is not given, or if
24107 @var{encoding} is the empty string, then either the @code{target-charset}
24108 (@pxref{Character Sets}) will be used, or a language-specific encoding
24109 will be used, if the current language is able to supply one.
24111 The optional @var{errors} argument is the same as the corresponding
24112 argument to Python's @code{string.decode} method.
24114 If the optional @var{length} argument is given, the string will be
24115 fetched and converted to the given length.
24118 @defun Value.lazy_string (@r{[}encoding @r{[}, length@r{]]})
24119 If this @code{gdb.Value} represents a string, then this method
24120 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
24121 In Python}). Otherwise, this method will throw an exception.
24123 If the optional @var{encoding} argument is given, it must be a string
24124 naming the encoding of the @code{gdb.LazyString}. Some examples are:
24125 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
24126 @var{encoding} argument is an encoding that @value{GDBN} does
24127 recognize, @value{GDBN} will raise an error.
24129 When a lazy string is printed, the @value{GDBN} encoding machinery is
24130 used to convert the string during printing. If the optional
24131 @var{encoding} argument is not provided, or is an empty string,
24132 @value{GDBN} will automatically select the encoding most suitable for
24133 the string type. For further information on encoding in @value{GDBN}
24134 please see @ref{Character Sets}.
24136 If the optional @var{length} argument is given, the string will be
24137 fetched and encoded to the length of characters specified. If
24138 the @var{length} argument is not provided, the string will be fetched
24139 and encoded until a null of appropriate width is found.
24142 @defun Value.fetch_lazy ()
24143 If the @code{gdb.Value} object is currently a lazy value
24144 (@code{gdb.Value.is_lazy} is @code{True}), then the value is
24145 fetched from the inferior. Any errors that occur in the process
24146 will produce a Python exception.
24148 If the @code{gdb.Value} object is not a lazy value, this method
24151 This method does not return a value.
24155 @node Types In Python
24156 @subsubsection Types In Python
24157 @cindex types in Python
24158 @cindex Python, working with types
24161 @value{GDBN} represents types from the inferior using the class
24164 The following type-related functions are available in the @code{gdb}
24167 @findex gdb.lookup_type
24168 @defun gdb.lookup_type (name @r{[}, block@r{]})
24169 This function looks up a type by name. @var{name} is the name of the
24170 type to look up. It must be a string.
24172 If @var{block} is given, then @var{name} is looked up in that scope.
24173 Otherwise, it is searched for globally.
24175 Ordinarily, this function will return an instance of @code{gdb.Type}.
24176 If the named type cannot be found, it will throw an exception.
24179 If the type is a structure or class type, or an enum type, the fields
24180 of that type can be accessed using the Python @dfn{dictionary syntax}.
24181 For example, if @code{some_type} is a @code{gdb.Type} instance holding
24182 a structure type, you can access its @code{foo} field with:
24185 bar = some_type['foo']
24188 @code{bar} will be a @code{gdb.Field} object; see below under the
24189 description of the @code{Type.fields} method for a description of the
24190 @code{gdb.Field} class.
24192 An instance of @code{Type} has the following attributes:
24195 The type code for this type. The type code will be one of the
24196 @code{TYPE_CODE_} constants defined below.
24199 @defvar Type.sizeof
24200 The size of this type, in target @code{char} units. Usually, a
24201 target's @code{char} type will be an 8-bit byte. However, on some
24202 unusual platforms, this type may have a different size.
24206 The tag name for this type. The tag name is the name after
24207 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
24208 languages have this concept. If this type has no tag name, then
24209 @code{None} is returned.
24212 The following methods are provided:
24214 @defun Type.fields ()
24215 For structure and union types, this method returns the fields. Range
24216 types have two fields, the minimum and maximum values. Enum types
24217 have one field per enum constant. Function and method types have one
24218 field per parameter. The base types of C@t{++} classes are also
24219 represented as fields. If the type has no fields, or does not fit
24220 into one of these categories, an empty sequence will be returned.
24222 Each field is a @code{gdb.Field} object, with some pre-defined attributes:
24225 This attribute is not available for @code{static} fields (as in
24226 C@t{++} or Java). For non-@code{static} fields, the value is the bit
24227 position of the field. For @code{enum} fields, the value is the
24228 enumeration member's integer representation.
24231 The name of the field, or @code{None} for anonymous fields.
24234 This is @code{True} if the field is artificial, usually meaning that
24235 it was provided by the compiler and not the user. This attribute is
24236 always provided, and is @code{False} if the field is not artificial.
24238 @item is_base_class
24239 This is @code{True} if the field represents a base class of a C@t{++}
24240 structure. This attribute is always provided, and is @code{False}
24241 if the field is not a base class of the type that is the argument of
24242 @code{fields}, or if that type was not a C@t{++} class.
24245 If the field is packed, or is a bitfield, then this will have a
24246 non-zero value, which is the size of the field in bits. Otherwise,
24247 this will be zero; in this case the field's size is given by its type.
24250 The type of the field. This is usually an instance of @code{Type},
24251 but it can be @code{None} in some situations.
24255 @defun Type.array (@var{n1} @r{[}, @var{n2}@r{]})
24256 Return a new @code{gdb.Type} object which represents an array of this
24257 type. If one argument is given, it is the inclusive upper bound of
24258 the array; in this case the lower bound is zero. If two arguments are
24259 given, the first argument is the lower bound of the array, and the
24260 second argument is the upper bound of the array. An array's length
24261 must not be negative, but the bounds can be.
24264 @defun Type.vector (@var{n1} @r{[}, @var{n2}@r{]})
24265 Return a new @code{gdb.Type} object which represents a vector of this
24266 type. If one argument is given, it is the inclusive upper bound of
24267 the vector; in this case the lower bound is zero. If two arguments are
24268 given, the first argument is the lower bound of the vector, and the
24269 second argument is the upper bound of the vector. A vector's length
24270 must not be negative, but the bounds can be.
24272 The difference between an @code{array} and a @code{vector} is that
24273 arrays behave like in C: when used in expressions they decay to a pointer
24274 to the first element whereas vectors are treated as first class values.
24277 @defun Type.const ()
24278 Return a new @code{gdb.Type} object which represents a
24279 @code{const}-qualified variant of this type.
24282 @defun Type.volatile ()
24283 Return a new @code{gdb.Type} object which represents a
24284 @code{volatile}-qualified variant of this type.
24287 @defun Type.unqualified ()
24288 Return a new @code{gdb.Type} object which represents an unqualified
24289 variant of this type. That is, the result is neither @code{const} nor
24293 @defun Type.range ()
24294 Return a Python @code{Tuple} object that contains two elements: the
24295 low bound of the argument type and the high bound of that type. If
24296 the type does not have a range, @value{GDBN} will raise a
24297 @code{gdb.error} exception (@pxref{Exception Handling}).
24300 @defun Type.reference ()
24301 Return a new @code{gdb.Type} object which represents a reference to this
24305 @defun Type.pointer ()
24306 Return a new @code{gdb.Type} object which represents a pointer to this
24310 @defun Type.strip_typedefs ()
24311 Return a new @code{gdb.Type} that represents the real type,
24312 after removing all layers of typedefs.
24315 @defun Type.target ()
24316 Return a new @code{gdb.Type} object which represents the target type
24319 For a pointer type, the target type is the type of the pointed-to
24320 object. For an array type (meaning C-like arrays), the target type is
24321 the type of the elements of the array. For a function or method type,
24322 the target type is the type of the return value. For a complex type,
24323 the target type is the type of the elements. For a typedef, the
24324 target type is the aliased type.
24326 If the type does not have a target, this method will throw an
24330 @defun Type.template_argument (n @r{[}, block@r{]})
24331 If this @code{gdb.Type} is an instantiation of a template, this will
24332 return a new @code{gdb.Type} which represents the type of the
24333 @var{n}th template argument.
24335 If this @code{gdb.Type} is not a template type, this will throw an
24336 exception. Ordinarily, only C@t{++} code will have template types.
24338 If @var{block} is given, then @var{name} is looked up in that scope.
24339 Otherwise, it is searched for globally.
24343 Each type has a code, which indicates what category this type falls
24344 into. The available type categories are represented by constants
24345 defined in the @code{gdb} module:
24348 @findex TYPE_CODE_PTR
24349 @findex gdb.TYPE_CODE_PTR
24350 @item gdb.TYPE_CODE_PTR
24351 The type is a pointer.
24353 @findex TYPE_CODE_ARRAY
24354 @findex gdb.TYPE_CODE_ARRAY
24355 @item gdb.TYPE_CODE_ARRAY
24356 The type is an array.
24358 @findex TYPE_CODE_STRUCT
24359 @findex gdb.TYPE_CODE_STRUCT
24360 @item gdb.TYPE_CODE_STRUCT
24361 The type is a structure.
24363 @findex TYPE_CODE_UNION
24364 @findex gdb.TYPE_CODE_UNION
24365 @item gdb.TYPE_CODE_UNION
24366 The type is a union.
24368 @findex TYPE_CODE_ENUM
24369 @findex gdb.TYPE_CODE_ENUM
24370 @item gdb.TYPE_CODE_ENUM
24371 The type is an enum.
24373 @findex TYPE_CODE_FLAGS
24374 @findex gdb.TYPE_CODE_FLAGS
24375 @item gdb.TYPE_CODE_FLAGS
24376 A bit flags type, used for things such as status registers.
24378 @findex TYPE_CODE_FUNC
24379 @findex gdb.TYPE_CODE_FUNC
24380 @item gdb.TYPE_CODE_FUNC
24381 The type is a function.
24383 @findex TYPE_CODE_INT
24384 @findex gdb.TYPE_CODE_INT
24385 @item gdb.TYPE_CODE_INT
24386 The type is an integer type.
24388 @findex TYPE_CODE_FLT
24389 @findex gdb.TYPE_CODE_FLT
24390 @item gdb.TYPE_CODE_FLT
24391 A floating point type.
24393 @findex TYPE_CODE_VOID
24394 @findex gdb.TYPE_CODE_VOID
24395 @item gdb.TYPE_CODE_VOID
24396 The special type @code{void}.
24398 @findex TYPE_CODE_SET
24399 @findex gdb.TYPE_CODE_SET
24400 @item gdb.TYPE_CODE_SET
24403 @findex TYPE_CODE_RANGE
24404 @findex gdb.TYPE_CODE_RANGE
24405 @item gdb.TYPE_CODE_RANGE
24406 A range type, that is, an integer type with bounds.
24408 @findex TYPE_CODE_STRING
24409 @findex gdb.TYPE_CODE_STRING
24410 @item gdb.TYPE_CODE_STRING
24411 A string type. Note that this is only used for certain languages with
24412 language-defined string types; C strings are not represented this way.
24414 @findex TYPE_CODE_BITSTRING
24415 @findex gdb.TYPE_CODE_BITSTRING
24416 @item gdb.TYPE_CODE_BITSTRING
24417 A string of bits. It is deprecated.
24419 @findex TYPE_CODE_ERROR
24420 @findex gdb.TYPE_CODE_ERROR
24421 @item gdb.TYPE_CODE_ERROR
24422 An unknown or erroneous type.
24424 @findex TYPE_CODE_METHOD
24425 @findex gdb.TYPE_CODE_METHOD
24426 @item gdb.TYPE_CODE_METHOD
24427 A method type, as found in C@t{++} or Java.
24429 @findex TYPE_CODE_METHODPTR
24430 @findex gdb.TYPE_CODE_METHODPTR
24431 @item gdb.TYPE_CODE_METHODPTR
24432 A pointer-to-member-function.
24434 @findex TYPE_CODE_MEMBERPTR
24435 @findex gdb.TYPE_CODE_MEMBERPTR
24436 @item gdb.TYPE_CODE_MEMBERPTR
24437 A pointer-to-member.
24439 @findex TYPE_CODE_REF
24440 @findex gdb.TYPE_CODE_REF
24441 @item gdb.TYPE_CODE_REF
24444 @findex TYPE_CODE_CHAR
24445 @findex gdb.TYPE_CODE_CHAR
24446 @item gdb.TYPE_CODE_CHAR
24449 @findex TYPE_CODE_BOOL
24450 @findex gdb.TYPE_CODE_BOOL
24451 @item gdb.TYPE_CODE_BOOL
24454 @findex TYPE_CODE_COMPLEX
24455 @findex gdb.TYPE_CODE_COMPLEX
24456 @item gdb.TYPE_CODE_COMPLEX
24457 A complex float type.
24459 @findex TYPE_CODE_TYPEDEF
24460 @findex gdb.TYPE_CODE_TYPEDEF
24461 @item gdb.TYPE_CODE_TYPEDEF
24462 A typedef to some other type.
24464 @findex TYPE_CODE_NAMESPACE
24465 @findex gdb.TYPE_CODE_NAMESPACE
24466 @item gdb.TYPE_CODE_NAMESPACE
24467 A C@t{++} namespace.
24469 @findex TYPE_CODE_DECFLOAT
24470 @findex gdb.TYPE_CODE_DECFLOAT
24471 @item gdb.TYPE_CODE_DECFLOAT
24472 A decimal floating point type.
24474 @findex TYPE_CODE_INTERNAL_FUNCTION
24475 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
24476 @item gdb.TYPE_CODE_INTERNAL_FUNCTION
24477 A function internal to @value{GDBN}. This is the type used to represent
24478 convenience functions.
24481 Further support for types is provided in the @code{gdb.types}
24482 Python module (@pxref{gdb.types}).
24484 @node Pretty Printing API
24485 @subsubsection Pretty Printing API
24487 An example output is provided (@pxref{Pretty Printing}).
24489 A pretty-printer is just an object that holds a value and implements a
24490 specific interface, defined here.
24492 @defun pretty_printer.children (self)
24493 @value{GDBN} will call this method on a pretty-printer to compute the
24494 children of the pretty-printer's value.
24496 This method must return an object conforming to the Python iterator
24497 protocol. Each item returned by the iterator must be a tuple holding
24498 two elements. The first element is the ``name'' of the child; the
24499 second element is the child's value. The value can be any Python
24500 object which is convertible to a @value{GDBN} value.
24502 This method is optional. If it does not exist, @value{GDBN} will act
24503 as though the value has no children.
24506 @defun pretty_printer.display_hint (self)
24507 The CLI may call this method and use its result to change the
24508 formatting of a value. The result will also be supplied to an MI
24509 consumer as a @samp{displayhint} attribute of the variable being
24512 This method is optional. If it does exist, this method must return a
24515 Some display hints are predefined by @value{GDBN}:
24519 Indicate that the object being printed is ``array-like''. The CLI
24520 uses this to respect parameters such as @code{set print elements} and
24521 @code{set print array}.
24524 Indicate that the object being printed is ``map-like'', and that the
24525 children of this value can be assumed to alternate between keys and
24529 Indicate that the object being printed is ``string-like''. If the
24530 printer's @code{to_string} method returns a Python string of some
24531 kind, then @value{GDBN} will call its internal language-specific
24532 string-printing function to format the string. For the CLI this means
24533 adding quotation marks, possibly escaping some characters, respecting
24534 @code{set print elements}, and the like.
24538 @defun pretty_printer.to_string (self)
24539 @value{GDBN} will call this method to display the string
24540 representation of the value passed to the object's constructor.
24542 When printing from the CLI, if the @code{to_string} method exists,
24543 then @value{GDBN} will prepend its result to the values returned by
24544 @code{children}. Exactly how this formatting is done is dependent on
24545 the display hint, and may change as more hints are added. Also,
24546 depending on the print settings (@pxref{Print Settings}), the CLI may
24547 print just the result of @code{to_string} in a stack trace, omitting
24548 the result of @code{children}.
24550 If this method returns a string, it is printed verbatim.
24552 Otherwise, if this method returns an instance of @code{gdb.Value},
24553 then @value{GDBN} prints this value. This may result in a call to
24554 another pretty-printer.
24556 If instead the method returns a Python value which is convertible to a
24557 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
24558 the resulting value. Again, this may result in a call to another
24559 pretty-printer. Python scalars (integers, floats, and booleans) and
24560 strings are convertible to @code{gdb.Value}; other types are not.
24562 Finally, if this method returns @code{None} then no further operations
24563 are peformed in this method and nothing is printed.
24565 If the result is not one of these types, an exception is raised.
24568 @value{GDBN} provides a function which can be used to look up the
24569 default pretty-printer for a @code{gdb.Value}:
24571 @findex gdb.default_visualizer
24572 @defun gdb.default_visualizer (value)
24573 This function takes a @code{gdb.Value} object as an argument. If a
24574 pretty-printer for this value exists, then it is returned. If no such
24575 printer exists, then this returns @code{None}.
24578 @node Selecting Pretty-Printers
24579 @subsubsection Selecting Pretty-Printers
24581 The Python list @code{gdb.pretty_printers} contains an array of
24582 functions or callable objects that have been registered via addition
24583 as a pretty-printer. Printers in this list are called @code{global}
24584 printers, they're available when debugging all inferiors.
24585 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
24586 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
24589 Each function on these lists is passed a single @code{gdb.Value}
24590 argument and should return a pretty-printer object conforming to the
24591 interface definition above (@pxref{Pretty Printing API}). If a function
24592 cannot create a pretty-printer for the value, it should return
24595 @value{GDBN} first checks the @code{pretty_printers} attribute of each
24596 @code{gdb.Objfile} in the current program space and iteratively calls
24597 each enabled lookup routine in the list for that @code{gdb.Objfile}
24598 until it receives a pretty-printer object.
24599 If no pretty-printer is found in the objfile lists, @value{GDBN} then
24600 searches the pretty-printer list of the current program space,
24601 calling each enabled function until an object is returned.
24602 After these lists have been exhausted, it tries the global
24603 @code{gdb.pretty_printers} list, again calling each enabled function until an
24604 object is returned.
24606 The order in which the objfiles are searched is not specified. For a
24607 given list, functions are always invoked from the head of the list,
24608 and iterated over sequentially until the end of the list, or a printer
24609 object is returned.
24611 For various reasons a pretty-printer may not work.
24612 For example, the underlying data structure may have changed and
24613 the pretty-printer is out of date.
24615 The consequences of a broken pretty-printer are severe enough that
24616 @value{GDBN} provides support for enabling and disabling individual
24617 printers. For example, if @code{print frame-arguments} is on,
24618 a backtrace can become highly illegible if any argument is printed
24619 with a broken printer.
24621 Pretty-printers are enabled and disabled by attaching an @code{enabled}
24622 attribute to the registered function or callable object. If this attribute
24623 is present and its value is @code{False}, the printer is disabled, otherwise
24624 the printer is enabled.
24626 @node Writing a Pretty-Printer
24627 @subsubsection Writing a Pretty-Printer
24628 @cindex writing a pretty-printer
24630 A pretty-printer consists of two parts: a lookup function to detect
24631 if the type is supported, and the printer itself.
24633 Here is an example showing how a @code{std::string} printer might be
24634 written. @xref{Pretty Printing API}, for details on the API this class
24638 class StdStringPrinter(object):
24639 "Print a std::string"
24641 def __init__(self, val):
24644 def to_string(self):
24645 return self.val['_M_dataplus']['_M_p']
24647 def display_hint(self):
24651 And here is an example showing how a lookup function for the printer
24652 example above might be written.
24655 def str_lookup_function(val):
24656 lookup_tag = val.type.tag
24657 if lookup_tag == None:
24659 regex = re.compile("^std::basic_string<char,.*>$")
24660 if regex.match(lookup_tag):
24661 return StdStringPrinter(val)
24665 The example lookup function extracts the value's type, and attempts to
24666 match it to a type that it can pretty-print. If it is a type the
24667 printer can pretty-print, it will return a printer object. If not, it
24668 returns @code{None}.
24670 We recommend that you put your core pretty-printers into a Python
24671 package. If your pretty-printers are for use with a library, we
24672 further recommend embedding a version number into the package name.
24673 This practice will enable @value{GDBN} to load multiple versions of
24674 your pretty-printers at the same time, because they will have
24677 You should write auto-loaded code (@pxref{Python Auto-loading}) such that it
24678 can be evaluated multiple times without changing its meaning. An
24679 ideal auto-load file will consist solely of @code{import}s of your
24680 printer modules, followed by a call to a register pretty-printers with
24681 the current objfile.
24683 Taken as a whole, this approach will scale nicely to multiple
24684 inferiors, each potentially using a different library version.
24685 Embedding a version number in the Python package name will ensure that
24686 @value{GDBN} is able to load both sets of printers simultaneously.
24687 Then, because the search for pretty-printers is done by objfile, and
24688 because your auto-loaded code took care to register your library's
24689 printers with a specific objfile, @value{GDBN} will find the correct
24690 printers for the specific version of the library used by each
24693 To continue the @code{std::string} example (@pxref{Pretty Printing API}),
24694 this code might appear in @code{gdb.libstdcxx.v6}:
24697 def register_printers(objfile):
24698 objfile.pretty_printers.append(str_lookup_function)
24702 And then the corresponding contents of the auto-load file would be:
24705 import gdb.libstdcxx.v6
24706 gdb.libstdcxx.v6.register_printers(gdb.current_objfile())
24709 The previous example illustrates a basic pretty-printer.
24710 There are a few things that can be improved on.
24711 The printer doesn't have a name, making it hard to identify in a
24712 list of installed printers. The lookup function has a name, but
24713 lookup functions can have arbitrary, even identical, names.
24715 Second, the printer only handles one type, whereas a library typically has
24716 several types. One could install a lookup function for each desired type
24717 in the library, but one could also have a single lookup function recognize
24718 several types. The latter is the conventional way this is handled.
24719 If a pretty-printer can handle multiple data types, then its
24720 @dfn{subprinters} are the printers for the individual data types.
24722 The @code{gdb.printing} module provides a formal way of solving these
24723 problems (@pxref{gdb.printing}).
24724 Here is another example that handles multiple types.
24726 These are the types we are going to pretty-print:
24729 struct foo @{ int a, b; @};
24730 struct bar @{ struct foo x, y; @};
24733 Here are the printers:
24737 """Print a foo object."""
24739 def __init__(self, val):
24742 def to_string(self):
24743 return ("a=<" + str(self.val["a"]) +
24744 "> b=<" + str(self.val["b"]) + ">")
24747 """Print a bar object."""
24749 def __init__(self, val):
24752 def to_string(self):
24753 return ("x=<" + str(self.val["x"]) +
24754 "> y=<" + str(self.val["y"]) + ">")
24757 This example doesn't need a lookup function, that is handled by the
24758 @code{gdb.printing} module. Instead a function is provided to build up
24759 the object that handles the lookup.
24762 import gdb.printing
24764 def build_pretty_printer():
24765 pp = gdb.printing.RegexpCollectionPrettyPrinter(
24767 pp.add_printer('foo', '^foo$', fooPrinter)
24768 pp.add_printer('bar', '^bar$', barPrinter)
24772 And here is the autoload support:
24775 import gdb.printing
24777 gdb.printing.register_pretty_printer(
24778 gdb.current_objfile(),
24779 my_library.build_pretty_printer())
24782 Finally, when this printer is loaded into @value{GDBN}, here is the
24783 corresponding output of @samp{info pretty-printer}:
24786 (gdb) info pretty-printer
24793 @node Type Printing API
24794 @subsubsection Type Printing API
24795 @cindex type printing API for Python
24797 @value{GDBN} provides a way for Python code to customize type display.
24798 This is mainly useful for substituting canonical typedef names for
24801 @cindex type printer
24802 A @dfn{type printer} is just a Python object conforming to a certain
24803 protocol. A simple base class implementing the protocol is provided;
24804 see @ref{gdb.types}. A type printer must supply at least:
24806 @defivar type_printer enabled
24807 A boolean which is True if the printer is enabled, and False
24808 otherwise. This is manipulated by the @code{enable type-printer}
24809 and @code{disable type-printer} commands.
24812 @defivar type_printer name
24813 The name of the type printer. This must be a string. This is used by
24814 the @code{enable type-printer} and @code{disable type-printer}
24818 @defmethod type_printer instantiate (self)
24819 This is called by @value{GDBN} at the start of type-printing. It is
24820 only called if the type printer is enabled. This method must return a
24821 new object that supplies a @code{recognize} method, as described below.
24825 When displaying a type, say via the @code{ptype} command, @value{GDBN}
24826 will compute a list of type recognizers. This is done by iterating
24827 first over the per-objfile type printers (@pxref{Objfiles In Python}),
24828 followed by the per-progspace type printers (@pxref{Progspaces In
24829 Python}), and finally the global type printers.
24831 @value{GDBN} will call the @code{instantiate} method of each enabled
24832 type printer. If this method returns @code{None}, then the result is
24833 ignored; otherwise, it is appended to the list of recognizers.
24835 Then, when @value{GDBN} is going to display a type name, it iterates
24836 over the list of recognizers. For each one, it calls the recognition
24837 function, stopping if the function returns a non-@code{None} value.
24838 The recognition function is defined as:
24840 @defmethod type_recognizer recognize (self, type)
24841 If @var{type} is not recognized, return @code{None}. Otherwise,
24842 return a string which is to be printed as the name of @var{type}.
24843 @var{type} will be an instance of @code{gdb.Type} (@pxref{Types In
24847 @value{GDBN} uses this two-pass approach so that type printers can
24848 efficiently cache information without holding on to it too long. For
24849 example, it can be convenient to look up type information in a type
24850 printer and hold it for a recognizer's lifetime; if a single pass were
24851 done then type printers would have to make use of the event system in
24852 order to avoid holding information that could become stale as the
24855 @node Frame Filter API
24856 @subsubsection Filtering Frames.
24857 @cindex frame filters api
24859 Frame filters are Python objects that manipulate the visibility of a
24860 frame or frames when a backtrace (@pxref{Backtrace}) is printed by
24863 Only commands that print a backtrace, or, in the case of @sc{gdb/mi}
24864 commands (@pxref{GDB/MI}), those that return a collection of frames
24865 are affected. The commands that work with frame filters are:
24867 @code{backtrace} (@pxref{backtrace-command,, The backtrace command}),
24868 @code{-stack-list-frames}
24869 (@pxref{-stack-list-frames,, The -stack-list-frames command}),
24870 @code{-stack-list-variables} (@pxref{-stack-list-variables,, The
24871 -stack-list-variables command}), @code{-stack-list-arguments}
24872 @pxref{-stack-list-arguments,, The -stack-list-arguments command}) and
24873 @code{-stack-list-locals} (@pxref{-stack-list-locals,, The
24874 -stack-list-locals command}).
24876 A frame filter works by taking an iterator as an argument, applying
24877 actions to the contents of that iterator, and returning another
24878 iterator (or, possibly, the same iterator it was provided in the case
24879 where the filter does not perform any operations). Typically, frame
24880 filters utilize tools such as the Python's @code{itertools} module to
24881 work with and create new iterators from the source iterator.
24882 Regardless of how a filter chooses to apply actions, it must not alter
24883 the underlying @value{GDBN} frame or frames, or attempt to alter the
24884 call-stack within @value{GDBN}. This preserves data integrity within
24885 @value{GDBN}. Frame filters are executed on a priority basis and care
24886 should be taken that some frame filters may have been executed before,
24887 and that some frame filters will be executed after.
24889 An important consideration when designing frame filters, and well
24890 worth reflecting upon, is that frame filters should avoid unwinding
24891 the call stack if possible. Some stacks can run very deep, into the
24892 tens of thousands in some cases. To search every frame when a frame
24893 filter executes may be too expensive at that step. The frame filter
24894 cannot know how many frames it has to iterate over, and it may have to
24895 iterate through them all. This ends up duplicating effort as
24896 @value{GDBN} performs this iteration when it prints the frames. If
24897 the filter can defer unwinding frames until frame decorators are
24898 executed, after the last filter has executed, it should. @xref{Frame
24899 Decorator API}, for more information on decorators. Also, there are
24900 examples for both frame decorators and filters in later chapters.
24901 @xref{Writing a Frame Filter}, for more information.
24903 The Python dictionary @code{gdb.frame_filters} contains key/object
24904 pairings that comprise a frame filter. Frame filters in this
24905 dictionary are called @code{global} frame filters, and they are
24906 available when debugging all inferiors. These frame filters must
24907 register with the dictionary directly. In addition to the
24908 @code{global} dictionary, there are other dictionaries that are loaded
24909 with different inferiors via auto-loading (@pxref{Python
24910 Auto-loading}). The two other areas where frame filter dictionaries
24911 can be found are: @code{gdb.Progspace} which contains a
24912 @code{frame_filters} dictionary attribute, and each @code{gdb.Objfile}
24913 object which also contains a @code{frame_filters} dictionary
24916 When a command is executed from @value{GDBN} that is compatible with
24917 frame filters, @value{GDBN} combines the @code{global},
24918 @code{gdb.Progspace} and all @code{gdb.Objfile} dictionaries currently
24919 loaded. All of the @code{gdb.Objfile} dictionaries are combined, as
24920 several frames, and thus several object files, might be in use.
24921 @value{GDBN} then prunes any frame filter whose @code{enabled}
24922 attribute is @code{False}. This pruned list is then sorted according
24923 to the @code{priority} attribute in each filter.
24925 Once the dictionaries are combined, pruned and sorted, @value{GDBN}
24926 creates an iterator which wraps each frame in the call stack in a
24927 @code{FrameDecorator} object, and calls each filter in order. The
24928 output from the previous filter will always be the input to the next
24931 Frame filters have a mandatory interface which each frame filter must
24932 implement, defined here:
24934 @defun FrameFilter.filter (iterator)
24935 @value{GDBN} will call this method on a frame filter when it has
24936 reached the order in the priority list for that filter.
24938 For example, if there are four frame filters:
24949 The order that the frame filters will be called is:
24952 Filter3 -> Filter2 -> Filter1 -> Filter4
24955 Note that the output from @code{Filter3} is passed to the input of
24956 @code{Filter2}, and so on.
24958 This @code{filter} method is passed a Python iterator. This iterator
24959 contains a sequence of frame decorators that wrap each
24960 @code{gdb.Frame}, or a frame decorator that wraps another frame
24961 decorator. The first filter that is executed in the sequence of frame
24962 filters will receive an iterator entirely comprised of default
24963 @code{FrameDecorator} objects. However, after each frame filter is
24964 executed, the previous frame filter may have wrapped some or all of
24965 the frame decorators with their own frame decorator. As frame
24966 decorators must also conform to a mandatory interface, these
24967 decorators can be assumed to act in a uniform manner (@pxref{Frame
24970 This method must return an object conforming to the Python iterator
24971 protocol. Each item in the iterator must be an object conforming to
24972 the frame decorator interface. If a frame filter does not wish to
24973 perform any operations on this iterator, it should return that
24974 iterator untouched.
24976 This method is not optional. If it does not exist, @value{GDBN} will
24977 raise and print an error.
24980 @defvar FrameFilter.name
24981 The @code{name} attribute must be Python string which contains the
24982 name of the filter displayed by @value{GDBN} (@pxref{Frame Filter
24983 Management}). This attribute may contain any combination of letters
24984 or numbers. Care should be taken to ensure that it is unique. This
24985 attribute is mandatory.
24988 @defvar FrameFilter.enabled
24989 The @code{enabled} attribute must be Python boolean. This attribute
24990 indicates to @value{GDBN} whether the frame filter is enabled, and
24991 should be considered when frame filters are executed. If
24992 @code{enabled} is @code{True}, then the frame filter will be executed
24993 when any of the backtrace commands detailed earlier in this chapter
24994 are executed. If @code{enabled} is @code{False}, then the frame
24995 filter will not be executed. This attribute is mandatory.
24998 @defvar FrameFilter.priority
24999 The @code{priority} attribute must be Python integer. This attribute
25000 controls the order of execution in relation to other frame filters.
25001 There are no imposed limits on the range of @code{priority} other than
25002 it must be a valid integer. The higher the @code{priority} attribute,
25003 the sooner the frame filter will be executed in relation to other
25004 frame filters. Although @code{priority} can be negative, it is
25005 recommended practice to assume zero is the lowest priority that a
25006 frame filter can be assigned. Frame filters that have the same
25007 priority are executed in unsorted order in that priority slot. This
25008 attribute is mandatory.
25011 @node Frame Decorator API
25012 @subsubsection Decorating Frames.
25013 @cindex frame decorator api
25015 Frame decorators are sister objects to frame filters (@pxref{Frame
25016 Filter API}). Frame decorators are applied by a frame filter and can
25017 only be used in conjunction with frame filters.
25019 The purpose of a frame decorator is to customize the printed content
25020 of each @code{gdb.Frame} in commands where frame filters are executed.
25021 This concept is called decorating a frame. Frame decorators decorate
25022 a @code{gdb.Frame} with Python code contained within each API call.
25023 This separates the actual data contained in a @code{gdb.Frame} from
25024 the decorated data produced by a frame decorator. This abstraction is
25025 necessary to maintain integrity of the data contained in each
25028 Frame decorators have a mandatory interface, defined below.
25030 @value{GDBN} already contains a frame decorator called
25031 @code{FrameDecorator}. This contains substantial amounts of
25032 boilerplate code to decorate the content of a @code{gdb.Frame}. It is
25033 recommended that other frame decorators inherit and extend this
25034 object, and only to override the methods needed.
25036 @defun FrameDecorator.elided (self)
25038 The @code{elided} method groups frames together in a hierarchical
25039 system. An example would be an interpreter, where multiple low-level
25040 frames make up a single call in the interpreted language. In this
25041 example, the frame filter would elide the low-level frames and present
25042 a single high-level frame, representing the call in the interpreted
25043 language, to the user.
25045 The @code{elided} function must return an iterable and this iterable
25046 must contain the frames that are being elided wrapped in a suitable
25047 frame decorator. If no frames are being elided this function may
25048 return an empty iterable, or @code{None}. Elided frames are indented
25049 from normal frames in a @code{CLI} backtrace, or in the case of
25050 @code{GDB/MI}, are placed in the @code{children} field of the eliding
25053 It is the frame filter's task to also filter out the elided frames from
25054 the source iterator. This will avoid printing the frame twice.
25057 @defun FrameDecorator.function (self)
25059 This method returns the name of the function in the frame that is to
25062 This method must return a Python string describing the function, or
25065 If this function returns @code{None}, @value{GDBN} will not print any
25066 data for this field.
25069 @defun FrameDecorator.address (self)
25071 This method returns the address of the frame that is to be printed.
25073 This method must return a Python numeric integer type of sufficient
25074 size to describe the address of the frame, or @code{None}.
25076 If this function returns a @code{None}, @value{GDBN} will not print
25077 any data for this field.
25080 @defun FrameDecorator.filename (self)
25082 This method returns the filename and path associated with this frame.
25084 This method must return a Python string containing the filename and
25085 the path to the object file backing the frame, or @code{None}.
25087 If this function returns a @code{None}, @value{GDBN} will not print
25088 any data for this field.
25091 @defun FrameDecorator.line (self):
25093 This method returns the line number associated with the current
25094 position within the function addressed by this frame.
25096 This method must return a Python integer type, or @code{None}.
25098 If this function returns a @code{None}, @value{GDBN} will not print
25099 any data for this field.
25102 @defun FrameDecorator.frame_args (self)
25103 @anchor{frame_args}
25105 This method must return an iterable, or @code{None}. Returning an
25106 empty iterable, or @code{None} means frame arguments will not be
25107 printed for this frame. This iterable must contain objects that
25108 implement two methods, described here.
25110 This object must implement a @code{argument} method which takes a
25111 single @code{self} parameter and must return a @code{gdb.Symbol}
25112 (@pxref{Symbols In Python}), or a Python string. The object must also
25113 implement a @code{value} method which takes a single @code{self}
25114 parameter and must return a @code{gdb.Value} (@pxref{Values From
25115 Inferior}), a Python value, or @code{None}. If the @code{value}
25116 method returns @code{None}, and the @code{argument} method returns a
25117 @code{gdb.Symbol}, @value{GDBN} will look-up and print the value of
25118 the @code{gdb.Symbol} automatically.
25123 class SymValueWrapper():
25125 def __init__(self, symbol, value):
25135 class SomeFrameDecorator()
25138 def frame_args(self):
25141 block = self.inferior_frame.block()
25145 # Iterate over all symbols in a block. Only add
25146 # symbols that are arguments.
25148 if not sym.is_argument:
25150 args.append(SymValueWrapper(sym,None))
25152 # Add example synthetic argument.
25153 args.append(SymValueWrapper(``foo'', 42))
25159 @defun FrameDecorator.frame_locals (self)
25161 This method must return an iterable or @code{None}. Returning an
25162 empty iterable, or @code{None} means frame local arguments will not be
25163 printed for this frame.
25165 The object interface, the description of the various strategies for
25166 reading frame locals, and the example are largely similar to those
25167 described in the @code{frame_args} function, (@pxref{frame_args,,The
25168 frame filter frame_args function}). Below is a modified example:
25171 class SomeFrameDecorator()
25174 def frame_locals(self):
25177 block = self.inferior_frame.block()
25181 # Iterate over all symbols in a block. Add all
25182 # symbols, except arguments.
25184 if sym.is_argument:
25186 vars.append(SymValueWrapper(sym,None))
25188 # Add an example of a synthetic local variable.
25189 vars.append(SymValueWrapper(``bar'', 99))
25195 @defun FrameDecorator.inferior_frame (self):
25197 This method must return the underlying @code{gdb.Frame} that this
25198 frame decorator is decorating. @value{GDBN} requires the underlying
25199 frame for internal frame information to determine how to print certain
25200 values when printing a frame.
25203 @node Writing a Frame Filter
25204 @subsubsection Writing a Frame Filter
25205 @cindex writing a frame filter
25207 There are three basic elements that a frame filter must implement: it
25208 must correctly implement the documented interface (@pxref{Frame Filter
25209 API}), it must register itself with @value{GDBN}, and finally, it must
25210 decide if it is to work on the data provided by @value{GDBN}. In all
25211 cases, whether it works on the iterator or not, each frame filter must
25212 return an iterator. A bare-bones frame filter follows the pattern in
25213 the following example.
25218 class FrameFilter():
25220 def __init__(self):
25221 # Frame filter attribute creation.
25223 # 'name' is the name of the filter that GDB will display.
25225 # 'priority' is the priority of the filter relative to other
25228 # 'enabled' is a boolean that indicates whether this filter is
25229 # enabled and should be executed.
25232 self.priority = 100
25233 self.enabled = True
25235 # Register this frame filter with the global frame_filters
25237 gdb.frame_filters[self.name] = self
25239 def filter(self, frame_iter):
25240 # Just return the iterator.
25244 The frame filter in the example above implements the three
25245 requirements for all frame filters. It implements the API, self
25246 registers, and makes a decision on the iterator (in this case, it just
25247 returns the iterator untouched).
25249 The first step is attribute creation and assignment, and as shown in
25250 the comments the filter assigns the following attributes: @code{name},
25251 @code{priority} and whether the filter should be enabled with the
25252 @code{enabled} attribute.
25254 The second step is registering the frame filter with the dictionary or
25255 dictionaries that the frame filter has interest in. As shown in the
25256 comments, this filter just registers itself with the global dictionary
25257 @code{gdb.frame_filters}. As noted earlier, @code{gdb.frame_filters}
25258 is a dictionary that is initialized in the @code{gdb} module when
25259 @value{GDBN} starts. What dictionary a filter registers with is an
25260 important consideration. Generally, if a filter is specific to a set
25261 of code, it should be registered either in the @code{objfile} or
25262 @code{progspace} dictionaries as they are specific to the program
25263 currently loaded in @value{GDBN}. The global dictionary is always
25264 present in @value{GDBN} and is never unloaded. Any filters registered
25265 with the global dictionary will exist until @value{GDBN} exits. To
25266 avoid filters that may conflict, it is generally better to register
25267 frame filters against the dictionaries that more closely align with
25268 the usage of the filter currently in question. @xref{Python
25269 Auto-loading}, for further information on auto-loading Python scripts.
25271 @value{GDBN} takes a hands-off approach to frame filter registration,
25272 therefore it is the frame filter's responsibility to ensure
25273 registration has occurred, and that any exceptions are handled
25274 appropriately. In particular, you may wish to handle exceptions
25275 relating to Python dictionary key uniqueness. It is mandatory that
25276 the dictionary key is the same as frame filter's @code{name}
25277 attribute. When a user manages frame filters (@pxref{Frame Filter
25278 Management}), the names @value{GDBN} will display are those contained
25279 in the @code{name} attribute.
25281 The final step of this example is the implementation of the
25282 @code{filter} method. As shown in the example comments, we define the
25283 @code{filter} method and note that the method must take an iterator,
25284 and also must return an iterator. In this bare-bones example, the
25285 frame filter is not very useful as it just returns the iterator
25286 untouched. However this is a valid operation for frame filters that
25287 have the @code{enabled} attribute set, but decide not to operate on
25290 In the next example, the frame filter operates on all frames and
25291 utilizes a frame decorator to perform some work on the frames.
25292 @xref{Frame Decorator API}, for further information on the frame
25293 decorator interface.
25295 This example works on inlined frames. It highlights frames which are
25296 inlined by tagging them with an ``[inlined]'' tag. By applying a
25297 frame decorator to all frames with the Python @code{itertools imap}
25298 method, the example defers actions to the frame decorator. Frame
25299 decorators are only processed when @value{GDBN} prints the backtrace.
25301 This introduces a new decision making topic: whether to perform
25302 decision making operations at the filtering step, or at the printing
25303 step. In this example's approach, it does not perform any filtering
25304 decisions at the filtering step beyond mapping a frame decorator to
25305 each frame. This allows the actual decision making to be performed
25306 when each frame is printed. This is an important consideration, and
25307 well worth reflecting upon when designing a frame filter. An issue
25308 that frame filters should avoid is unwinding the stack if possible.
25309 Some stacks can run very deep, into the tens of thousands in some
25310 cases. To search every frame to determine if it is inlined ahead of
25311 time may be too expensive at the filtering step. The frame filter
25312 cannot know how many frames it has to iterate over, and it would have
25313 to iterate through them all. This ends up duplicating effort as
25314 @value{GDBN} performs this iteration when it prints the frames.
25316 In this example decision making can be deferred to the printing step.
25317 As each frame is printed, the frame decorator can examine each frame
25318 in turn when @value{GDBN} iterates. From a performance viewpoint,
25319 this is the most appropriate decision to make as it avoids duplicating
25320 the effort that the printing step would undertake anyway. Also, if
25321 there are many frame filters unwinding the stack during filtering, it
25322 can substantially delay the printing of the backtrace which will
25323 result in large memory usage, and a poor user experience.
25326 class InlineFilter():
25328 def __init__(self):
25329 self.name = "InlinedFrameFilter"
25330 self.priority = 100
25331 self.enabled = True
25332 gdb.frame_filters[self.name] = self
25334 def filter(self, frame_iter):
25335 frame_iter = itertools.imap(InlinedFrameDecorator,
25340 This frame filter is somewhat similar to the earlier example, except
25341 that the @code{filter} method applies a frame decorator object called
25342 @code{InlinedFrameDecorator} to each element in the iterator. The
25343 @code{imap} Python method is light-weight. It does not proactively
25344 iterate over the iterator, but rather creates a new iterator which
25345 wraps the existing one.
25347 Below is the frame decorator for this example.
25350 class InlinedFrameDecorator(FrameDecorator):
25352 def __init__(self, fobj):
25353 super(InlinedFrameDecorator, self).__init__(fobj)
25355 def function(self):
25356 frame = fobj.inferior_frame()
25357 name = str(frame.name())
25359 if frame.type() == gdb.INLINE_FRAME:
25360 name = name + " [inlined]"
25365 This frame decorator only defines and overrides the @code{function}
25366 method. It lets the supplied @code{FrameDecorator}, which is shipped
25367 with @value{GDBN}, perform the other work associated with printing
25370 The combination of these two objects create this output from a
25374 #0 0x004004e0 in bar () at inline.c:11
25375 #1 0x00400566 in max [inlined] (b=6, a=12) at inline.c:21
25376 #2 0x00400566 in main () at inline.c:31
25379 So in the case of this example, a frame decorator is applied to all
25380 frames, regardless of whether they may be inlined or not. As
25381 @value{GDBN} iterates over the iterator produced by the frame filters,
25382 @value{GDBN} executes each frame decorator which then makes a decision
25383 on what to print in the @code{function} callback. Using a strategy
25384 like this is a way to defer decisions on the frame content to printing
25387 @subheading Eliding Frames
25389 It might be that the above example is not desirable for representing
25390 inlined frames, and a hierarchical approach may be preferred. If we
25391 want to hierarchically represent frames, the @code{elided} frame
25392 decorator interface might be preferable.
25394 This example approaches the issue with the @code{elided} method. This
25395 example is quite long, but very simplistic. It is out-of-scope for
25396 this section to write a complete example that comprehensively covers
25397 all approaches of finding and printing inlined frames. However, this
25398 example illustrates the approach an author might use.
25400 This example comprises of three sections.
25403 class InlineFrameFilter():
25405 def __init__(self):
25406 self.name = "InlinedFrameFilter"
25407 self.priority = 100
25408 self.enabled = True
25409 gdb.frame_filters[self.name] = self
25411 def filter(self, frame_iter):
25412 return ElidingInlineIterator(frame_iter)
25415 This frame filter is very similar to the other examples. The only
25416 difference is this frame filter is wrapping the iterator provided to
25417 it (@code{frame_iter}) with a custom iterator called
25418 @code{ElidingInlineIterator}. This again defers actions to when
25419 @value{GDBN} prints the backtrace, as the iterator is not traversed
25422 The iterator for this example is as follows. It is in this section of
25423 the example where decisions are made on the content of the backtrace.
25426 class ElidingInlineIterator:
25427 def __init__(self, ii):
25428 self.input_iterator = ii
25430 def __iter__(self):
25434 frame = next(self.input_iterator)
25436 if frame.inferior_frame().type() != gdb.INLINE_FRAME:
25440 eliding_frame = next(self.input_iterator)
25441 except StopIteration:
25443 return ElidingFrameDecorator(eliding_frame, [frame])
25446 This iterator implements the Python iterator protocol. When the
25447 @code{next} function is called (when @value{GDBN} prints each frame),
25448 the iterator checks if this frame decorator, @code{frame}, is wrapping
25449 an inlined frame. If it is not, it returns the existing frame decorator
25450 untouched. If it is wrapping an inlined frame, it assumes that the
25451 inlined frame was contained within the next oldest frame,
25452 @code{eliding_frame}, which it fetches. It then creates and returns a
25453 frame decorator, @code{ElidingFrameDecorator}, which contains both the
25454 elided frame, and the eliding frame.
25457 class ElidingInlineDecorator(FrameDecorator):
25459 def __init__(self, frame, elided_frames):
25460 super(ElidingInlineDecorator, self).__init__(frame)
25462 self.elided_frames = elided_frames
25465 return iter(self.elided_frames)
25468 This frame decorator overrides one function and returns the inlined
25469 frame in the @code{elided} method. As before it lets
25470 @code{FrameDecorator} do the rest of the work involved in printing
25471 this frame. This produces the following output.
25474 #0 0x004004e0 in bar () at inline.c:11
25475 #2 0x00400529 in main () at inline.c:25
25476 #1 0x00400529 in max (b=6, a=12) at inline.c:15
25479 In that output, @code{max} which has been inlined into @code{main} is
25480 printed hierarchically. Another approach would be to combine the
25481 @code{function} method, and the @code{elided} method to both print a
25482 marker in the inlined frame, and also show the hierarchical
25485 @node Inferiors In Python
25486 @subsubsection Inferiors In Python
25487 @cindex inferiors in Python
25489 @findex gdb.Inferior
25490 Programs which are being run under @value{GDBN} are called inferiors
25491 (@pxref{Inferiors and Programs}). Python scripts can access
25492 information about and manipulate inferiors controlled by @value{GDBN}
25493 via objects of the @code{gdb.Inferior} class.
25495 The following inferior-related functions are available in the @code{gdb}
25498 @defun gdb.inferiors ()
25499 Return a tuple containing all inferior objects.
25502 @defun gdb.selected_inferior ()
25503 Return an object representing the current inferior.
25506 A @code{gdb.Inferior} object has the following attributes:
25508 @defvar Inferior.num
25509 ID of inferior, as assigned by GDB.
25512 @defvar Inferior.pid
25513 Process ID of the inferior, as assigned by the underlying operating
25517 @defvar Inferior.was_attached
25518 Boolean signaling whether the inferior was created using `attach', or
25519 started by @value{GDBN} itself.
25522 A @code{gdb.Inferior} object has the following methods:
25524 @defun Inferior.is_valid ()
25525 Returns @code{True} if the @code{gdb.Inferior} object is valid,
25526 @code{False} if not. A @code{gdb.Inferior} object will become invalid
25527 if the inferior no longer exists within @value{GDBN}. All other
25528 @code{gdb.Inferior} methods will throw an exception if it is invalid
25529 at the time the method is called.
25532 @defun Inferior.threads ()
25533 This method returns a tuple holding all the threads which are valid
25534 when it is called. If there are no valid threads, the method will
25535 return an empty tuple.
25538 @findex Inferior.read_memory
25539 @defun Inferior.read_memory (address, length)
25540 Read @var{length} bytes of memory from the inferior, starting at
25541 @var{address}. Returns a buffer object, which behaves much like an array
25542 or a string. It can be modified and given to the
25543 @code{Inferior.write_memory} function. In @code{Python} 3, the return
25544 value is a @code{memoryview} object.
25547 @findex Inferior.write_memory
25548 @defun Inferior.write_memory (address, buffer @r{[}, length@r{]})
25549 Write the contents of @var{buffer} to the inferior, starting at
25550 @var{address}. The @var{buffer} parameter must be a Python object
25551 which supports the buffer protocol, i.e., a string, an array or the
25552 object returned from @code{Inferior.read_memory}. If given, @var{length}
25553 determines the number of bytes from @var{buffer} to be written.
25556 @findex gdb.search_memory
25557 @defun Inferior.search_memory (address, length, pattern)
25558 Search a region of the inferior memory starting at @var{address} with
25559 the given @var{length} using the search pattern supplied in
25560 @var{pattern}. The @var{pattern} parameter must be a Python object
25561 which supports the buffer protocol, i.e., a string, an array or the
25562 object returned from @code{gdb.read_memory}. Returns a Python @code{Long}
25563 containing the address where the pattern was found, or @code{None} if
25564 the pattern could not be found.
25567 @node Events In Python
25568 @subsubsection Events In Python
25569 @cindex inferior events in Python
25571 @value{GDBN} provides a general event facility so that Python code can be
25572 notified of various state changes, particularly changes that occur in
25575 An @dfn{event} is just an object that describes some state change. The
25576 type of the object and its attributes will vary depending on the details
25577 of the change. All the existing events are described below.
25579 In order to be notified of an event, you must register an event handler
25580 with an @dfn{event registry}. An event registry is an object in the
25581 @code{gdb.events} module which dispatches particular events. A registry
25582 provides methods to register and unregister event handlers:
25584 @defun EventRegistry.connect (object)
25585 Add the given callable @var{object} to the registry. This object will be
25586 called when an event corresponding to this registry occurs.
25589 @defun EventRegistry.disconnect (object)
25590 Remove the given @var{object} from the registry. Once removed, the object
25591 will no longer receive notifications of events.
25594 Here is an example:
25597 def exit_handler (event):
25598 print "event type: exit"
25599 print "exit code: %d" % (event.exit_code)
25601 gdb.events.exited.connect (exit_handler)
25604 In the above example we connect our handler @code{exit_handler} to the
25605 registry @code{events.exited}. Once connected, @code{exit_handler} gets
25606 called when the inferior exits. The argument @dfn{event} in this example is
25607 of type @code{gdb.ExitedEvent}. As you can see in the example the
25608 @code{ExitedEvent} object has an attribute which indicates the exit code of
25611 The following is a listing of the event registries that are available and
25612 details of the events they emit:
25617 Emits @code{gdb.ThreadEvent}.
25619 Some events can be thread specific when @value{GDBN} is running in non-stop
25620 mode. When represented in Python, these events all extend
25621 @code{gdb.ThreadEvent}. Note, this event is not emitted directly; instead,
25622 events which are emitted by this or other modules might extend this event.
25623 Examples of these events are @code{gdb.BreakpointEvent} and
25624 @code{gdb.ContinueEvent}.
25626 @defvar ThreadEvent.inferior_thread
25627 In non-stop mode this attribute will be set to the specific thread which was
25628 involved in the emitted event. Otherwise, it will be set to @code{None}.
25631 Emits @code{gdb.ContinueEvent} which extends @code{gdb.ThreadEvent}.
25633 This event indicates that the inferior has been continued after a stop. For
25634 inherited attribute refer to @code{gdb.ThreadEvent} above.
25636 @item events.exited
25637 Emits @code{events.ExitedEvent} which indicates that the inferior has exited.
25638 @code{events.ExitedEvent} has two attributes:
25639 @defvar ExitedEvent.exit_code
25640 An integer representing the exit code, if available, which the inferior
25641 has returned. (The exit code could be unavailable if, for example,
25642 @value{GDBN} detaches from the inferior.) If the exit code is unavailable,
25643 the attribute does not exist.
25645 @defvar ExitedEvent inferior
25646 A reference to the inferior which triggered the @code{exited} event.
25650 Emits @code{gdb.StopEvent} which extends @code{gdb.ThreadEvent}.
25652 Indicates that the inferior has stopped. All events emitted by this registry
25653 extend StopEvent. As a child of @code{gdb.ThreadEvent}, @code{gdb.StopEvent}
25654 will indicate the stopped thread when @value{GDBN} is running in non-stop
25655 mode. Refer to @code{gdb.ThreadEvent} above for more details.
25657 Emits @code{gdb.SignalEvent} which extends @code{gdb.StopEvent}.
25659 This event indicates that the inferior or one of its threads has received as
25660 signal. @code{gdb.SignalEvent} has the following attributes:
25662 @defvar SignalEvent.stop_signal
25663 A string representing the signal received by the inferior. A list of possible
25664 signal values can be obtained by running the command @code{info signals} in
25665 the @value{GDBN} command prompt.
25668 Also emits @code{gdb.BreakpointEvent} which extends @code{gdb.StopEvent}.
25670 @code{gdb.BreakpointEvent} event indicates that one or more breakpoints have
25671 been hit, and has the following attributes:
25673 @defvar BreakpointEvent.breakpoints
25674 A sequence containing references to all the breakpoints (type
25675 @code{gdb.Breakpoint}) that were hit.
25676 @xref{Breakpoints In Python}, for details of the @code{gdb.Breakpoint} object.
25678 @defvar BreakpointEvent.breakpoint
25679 A reference to the first breakpoint that was hit.
25680 This function is maintained for backward compatibility and is now deprecated
25681 in favor of the @code{gdb.BreakpointEvent.breakpoints} attribute.
25684 @item events.new_objfile
25685 Emits @code{gdb.NewObjFileEvent} which indicates that a new object file has
25686 been loaded by @value{GDBN}. @code{gdb.NewObjFileEvent} has one attribute:
25688 @defvar NewObjFileEvent.new_objfile
25689 A reference to the object file (@code{gdb.Objfile}) which has been loaded.
25690 @xref{Objfiles In Python}, for details of the @code{gdb.Objfile} object.
25695 @node Threads In Python
25696 @subsubsection Threads In Python
25697 @cindex threads in python
25699 @findex gdb.InferiorThread
25700 Python scripts can access information about, and manipulate inferior threads
25701 controlled by @value{GDBN}, via objects of the @code{gdb.InferiorThread} class.
25703 The following thread-related functions are available in the @code{gdb}
25706 @findex gdb.selected_thread
25707 @defun gdb.selected_thread ()
25708 This function returns the thread object for the selected thread. If there
25709 is no selected thread, this will return @code{None}.
25712 A @code{gdb.InferiorThread} object has the following attributes:
25714 @defvar InferiorThread.name
25715 The name of the thread. If the user specified a name using
25716 @code{thread name}, then this returns that name. Otherwise, if an
25717 OS-supplied name is available, then it is returned. Otherwise, this
25718 returns @code{None}.
25720 This attribute can be assigned to. The new value must be a string
25721 object, which sets the new name, or @code{None}, which removes any
25722 user-specified thread name.
25725 @defvar InferiorThread.num
25726 ID of the thread, as assigned by GDB.
25729 @defvar InferiorThread.ptid
25730 ID of the thread, as assigned by the operating system. This attribute is a
25731 tuple containing three integers. The first is the Process ID (PID); the second
25732 is the Lightweight Process ID (LWPID), and the third is the Thread ID (TID).
25733 Either the LWPID or TID may be 0, which indicates that the operating system
25734 does not use that identifier.
25737 A @code{gdb.InferiorThread} object has the following methods:
25739 @defun InferiorThread.is_valid ()
25740 Returns @code{True} if the @code{gdb.InferiorThread} object is valid,
25741 @code{False} if not. A @code{gdb.InferiorThread} object will become
25742 invalid if the thread exits, or the inferior that the thread belongs
25743 is deleted. All other @code{gdb.InferiorThread} methods will throw an
25744 exception if it is invalid at the time the method is called.
25747 @defun InferiorThread.switch ()
25748 This changes @value{GDBN}'s currently selected thread to the one represented
25752 @defun InferiorThread.is_stopped ()
25753 Return a Boolean indicating whether the thread is stopped.
25756 @defun InferiorThread.is_running ()
25757 Return a Boolean indicating whether the thread is running.
25760 @defun InferiorThread.is_exited ()
25761 Return a Boolean indicating whether the thread is exited.
25764 @node Commands In Python
25765 @subsubsection Commands In Python
25767 @cindex commands in python
25768 @cindex python commands
25769 You can implement new @value{GDBN} CLI commands in Python. A CLI
25770 command is implemented using an instance of the @code{gdb.Command}
25771 class, most commonly using a subclass.
25773 @defun Command.__init__ (name, @var{command_class} @r{[}, @var{completer_class} @r{[}, @var{prefix}@r{]]})
25774 The object initializer for @code{Command} registers the new command
25775 with @value{GDBN}. This initializer is normally invoked from the
25776 subclass' own @code{__init__} method.
25778 @var{name} is the name of the command. If @var{name} consists of
25779 multiple words, then the initial words are looked for as prefix
25780 commands. In this case, if one of the prefix commands does not exist,
25781 an exception is raised.
25783 There is no support for multi-line commands.
25785 @var{command_class} should be one of the @samp{COMMAND_} constants
25786 defined below. This argument tells @value{GDBN} how to categorize the
25787 new command in the help system.
25789 @var{completer_class} is an optional argument. If given, it should be
25790 one of the @samp{COMPLETE_} constants defined below. This argument
25791 tells @value{GDBN} how to perform completion for this command. If not
25792 given, @value{GDBN} will attempt to complete using the object's
25793 @code{complete} method (see below); if no such method is found, an
25794 error will occur when completion is attempted.
25796 @var{prefix} is an optional argument. If @code{True}, then the new
25797 command is a prefix command; sub-commands of this command may be
25800 The help text for the new command is taken from the Python
25801 documentation string for the command's class, if there is one. If no
25802 documentation string is provided, the default value ``This command is
25803 not documented.'' is used.
25806 @cindex don't repeat Python command
25807 @defun Command.dont_repeat ()
25808 By default, a @value{GDBN} command is repeated when the user enters a
25809 blank line at the command prompt. A command can suppress this
25810 behavior by invoking the @code{dont_repeat} method. This is similar
25811 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
25814 @defun Command.invoke (argument, from_tty)
25815 This method is called by @value{GDBN} when this command is invoked.
25817 @var{argument} is a string. It is the argument to the command, after
25818 leading and trailing whitespace has been stripped.
25820 @var{from_tty} is a boolean argument. When true, this means that the
25821 command was entered by the user at the terminal; when false it means
25822 that the command came from elsewhere.
25824 If this method throws an exception, it is turned into a @value{GDBN}
25825 @code{error} call. Otherwise, the return value is ignored.
25827 @findex gdb.string_to_argv
25828 To break @var{argument} up into an argv-like string use
25829 @code{gdb.string_to_argv}. This function behaves identically to
25830 @value{GDBN}'s internal argument lexer @code{buildargv}.
25831 It is recommended to use this for consistency.
25832 Arguments are separated by spaces and may be quoted.
25836 print gdb.string_to_argv ("1 2\ \\\"3 '4 \"5' \"6 '7\"")
25837 ['1', '2 "3', '4 "5', "6 '7"]
25842 @cindex completion of Python commands
25843 @defun Command.complete (text, word)
25844 This method is called by @value{GDBN} when the user attempts
25845 completion on this command. All forms of completion are handled by
25846 this method, that is, the @key{TAB} and @key{M-?} key bindings
25847 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
25850 The arguments @var{text} and @var{word} are both strings. @var{text}
25851 holds the complete command line up to the cursor's location.
25852 @var{word} holds the last word of the command line; this is computed
25853 using a word-breaking heuristic.
25855 The @code{complete} method can return several values:
25858 If the return value is a sequence, the contents of the sequence are
25859 used as the completions. It is up to @code{complete} to ensure that the
25860 contents actually do complete the word. A zero-length sequence is
25861 allowed, it means that there were no completions available. Only
25862 string elements of the sequence are used; other elements in the
25863 sequence are ignored.
25866 If the return value is one of the @samp{COMPLETE_} constants defined
25867 below, then the corresponding @value{GDBN}-internal completion
25868 function is invoked, and its result is used.
25871 All other results are treated as though there were no available
25876 When a new command is registered, it must be declared as a member of
25877 some general class of commands. This is used to classify top-level
25878 commands in the on-line help system; note that prefix commands are not
25879 listed under their own category but rather that of their top-level
25880 command. The available classifications are represented by constants
25881 defined in the @code{gdb} module:
25884 @findex COMMAND_NONE
25885 @findex gdb.COMMAND_NONE
25886 @item gdb.COMMAND_NONE
25887 The command does not belong to any particular class. A command in
25888 this category will not be displayed in any of the help categories.
25890 @findex COMMAND_RUNNING
25891 @findex gdb.COMMAND_RUNNING
25892 @item gdb.COMMAND_RUNNING
25893 The command is related to running the inferior. For example,
25894 @code{start}, @code{step}, and @code{continue} are in this category.
25895 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
25896 commands in this category.
25898 @findex COMMAND_DATA
25899 @findex gdb.COMMAND_DATA
25900 @item gdb.COMMAND_DATA
25901 The command is related to data or variables. For example,
25902 @code{call}, @code{find}, and @code{print} are in this category. Type
25903 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
25906 @findex COMMAND_STACK
25907 @findex gdb.COMMAND_STACK
25908 @item gdb.COMMAND_STACK
25909 The command has to do with manipulation of the stack. For example,
25910 @code{backtrace}, @code{frame}, and @code{return} are in this
25911 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
25912 list of commands in this category.
25914 @findex COMMAND_FILES
25915 @findex gdb.COMMAND_FILES
25916 @item gdb.COMMAND_FILES
25917 This class is used for file-related commands. For example,
25918 @code{file}, @code{list} and @code{section} are in this category.
25919 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
25920 commands in this category.
25922 @findex COMMAND_SUPPORT
25923 @findex gdb.COMMAND_SUPPORT
25924 @item gdb.COMMAND_SUPPORT
25925 This should be used for ``support facilities'', generally meaning
25926 things that are useful to the user when interacting with @value{GDBN},
25927 but not related to the state of the inferior. For example,
25928 @code{help}, @code{make}, and @code{shell} are in this category. Type
25929 @kbd{help support} at the @value{GDBN} prompt to see a list of
25930 commands in this category.
25932 @findex COMMAND_STATUS
25933 @findex gdb.COMMAND_STATUS
25934 @item gdb.COMMAND_STATUS
25935 The command is an @samp{info}-related command, that is, related to the
25936 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
25937 and @code{show} are in this category. Type @kbd{help status} at the
25938 @value{GDBN} prompt to see a list of commands in this category.
25940 @findex COMMAND_BREAKPOINTS
25941 @findex gdb.COMMAND_BREAKPOINTS
25942 @item gdb.COMMAND_BREAKPOINTS
25943 The command has to do with breakpoints. For example, @code{break},
25944 @code{clear}, and @code{delete} are in this category. Type @kbd{help
25945 breakpoints} at the @value{GDBN} prompt to see a list of commands in
25948 @findex COMMAND_TRACEPOINTS
25949 @findex gdb.COMMAND_TRACEPOINTS
25950 @item gdb.COMMAND_TRACEPOINTS
25951 The command has to do with tracepoints. For example, @code{trace},
25952 @code{actions}, and @code{tfind} are in this category. Type
25953 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
25954 commands in this category.
25956 @findex COMMAND_USER
25957 @findex gdb.COMMAND_USER
25958 @item gdb.COMMAND_USER
25959 The command is a general purpose command for the user, and typically
25960 does not fit in one of the other categories.
25961 Type @kbd{help user-defined} at the @value{GDBN} prompt to see
25962 a list of commands in this category, as well as the list of gdb macros
25963 (@pxref{Sequences}).
25965 @findex COMMAND_OBSCURE
25966 @findex gdb.COMMAND_OBSCURE
25967 @item gdb.COMMAND_OBSCURE
25968 The command is only used in unusual circumstances, or is not of
25969 general interest to users. For example, @code{checkpoint},
25970 @code{fork}, and @code{stop} are in this category. Type @kbd{help
25971 obscure} at the @value{GDBN} prompt to see a list of commands in this
25974 @findex COMMAND_MAINTENANCE
25975 @findex gdb.COMMAND_MAINTENANCE
25976 @item gdb.COMMAND_MAINTENANCE
25977 The command is only useful to @value{GDBN} maintainers. The
25978 @code{maintenance} and @code{flushregs} commands are in this category.
25979 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
25980 commands in this category.
25983 A new command can use a predefined completion function, either by
25984 specifying it via an argument at initialization, or by returning it
25985 from the @code{complete} method. These predefined completion
25986 constants are all defined in the @code{gdb} module:
25989 @findex COMPLETE_NONE
25990 @findex gdb.COMPLETE_NONE
25991 @item gdb.COMPLETE_NONE
25992 This constant means that no completion should be done.
25994 @findex COMPLETE_FILENAME
25995 @findex gdb.COMPLETE_FILENAME
25996 @item gdb.COMPLETE_FILENAME
25997 This constant means that filename completion should be performed.
25999 @findex COMPLETE_LOCATION
26000 @findex gdb.COMPLETE_LOCATION
26001 @item gdb.COMPLETE_LOCATION
26002 This constant means that location completion should be done.
26003 @xref{Specify Location}.
26005 @findex COMPLETE_COMMAND
26006 @findex gdb.COMPLETE_COMMAND
26007 @item gdb.COMPLETE_COMMAND
26008 This constant means that completion should examine @value{GDBN}
26011 @findex COMPLETE_SYMBOL
26012 @findex gdb.COMPLETE_SYMBOL
26013 @item gdb.COMPLETE_SYMBOL
26014 This constant means that completion should be done using symbol names
26017 @findex COMPLETE_EXPRESSION
26018 @findex gdb.COMPLETE_EXPRESSION
26019 @item gdb.COMPLETE_EXPRESSION
26020 This constant means that completion should be done on expressions.
26021 Often this means completing on symbol names, but some language
26022 parsers also have support for completing on field names.
26025 The following code snippet shows how a trivial CLI command can be
26026 implemented in Python:
26029 class HelloWorld (gdb.Command):
26030 """Greet the whole world."""
26032 def __init__ (self):
26033 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
26035 def invoke (self, arg, from_tty):
26036 print "Hello, World!"
26041 The last line instantiates the class, and is necessary to trigger the
26042 registration of the command with @value{GDBN}. Depending on how the
26043 Python code is read into @value{GDBN}, you may need to import the
26044 @code{gdb} module explicitly.
26046 @node Parameters In Python
26047 @subsubsection Parameters In Python
26049 @cindex parameters in python
26050 @cindex python parameters
26051 @tindex gdb.Parameter
26053 You can implement new @value{GDBN} parameters using Python. A new
26054 parameter is implemented as an instance of the @code{gdb.Parameter}
26057 Parameters are exposed to the user via the @code{set} and
26058 @code{show} commands. @xref{Help}.
26060 There are many parameters that already exist and can be set in
26061 @value{GDBN}. Two examples are: @code{set follow fork} and
26062 @code{set charset}. Setting these parameters influences certain
26063 behavior in @value{GDBN}. Similarly, you can define parameters that
26064 can be used to influence behavior in custom Python scripts and commands.
26066 @defun Parameter.__init__ (name, @var{command-class}, @var{parameter-class} @r{[}, @var{enum-sequence}@r{]})
26067 The object initializer for @code{Parameter} registers the new
26068 parameter with @value{GDBN}. This initializer is normally invoked
26069 from the subclass' own @code{__init__} method.
26071 @var{name} is the name of the new parameter. If @var{name} consists
26072 of multiple words, then the initial words are looked for as prefix
26073 parameters. An example of this can be illustrated with the
26074 @code{set print} set of parameters. If @var{name} is
26075 @code{print foo}, then @code{print} will be searched as the prefix
26076 parameter. In this case the parameter can subsequently be accessed in
26077 @value{GDBN} as @code{set print foo}.
26079 If @var{name} consists of multiple words, and no prefix parameter group
26080 can be found, an exception is raised.
26082 @var{command-class} should be one of the @samp{COMMAND_} constants
26083 (@pxref{Commands In Python}). This argument tells @value{GDBN} how to
26084 categorize the new parameter in the help system.
26086 @var{parameter-class} should be one of the @samp{PARAM_} constants
26087 defined below. This argument tells @value{GDBN} the type of the new
26088 parameter; this information is used for input validation and
26091 If @var{parameter-class} is @code{PARAM_ENUM}, then
26092 @var{enum-sequence} must be a sequence of strings. These strings
26093 represent the possible values for the parameter.
26095 If @var{parameter-class} is not @code{PARAM_ENUM}, then the presence
26096 of a fourth argument will cause an exception to be thrown.
26098 The help text for the new parameter is taken from the Python
26099 documentation string for the parameter's class, if there is one. If
26100 there is no documentation string, a default value is used.
26103 @defvar Parameter.set_doc
26104 If this attribute exists, and is a string, then its value is used as
26105 the help text for this parameter's @code{set} command. The value is
26106 examined when @code{Parameter.__init__} is invoked; subsequent changes
26110 @defvar Parameter.show_doc
26111 If this attribute exists, and is a string, then its value is used as
26112 the help text for this parameter's @code{show} command. The value is
26113 examined when @code{Parameter.__init__} is invoked; subsequent changes
26117 @defvar Parameter.value
26118 The @code{value} attribute holds the underlying value of the
26119 parameter. It can be read and assigned to just as any other
26120 attribute. @value{GDBN} does validation when assignments are made.
26123 There are two methods that should be implemented in any
26124 @code{Parameter} class. These are:
26126 @defun Parameter.get_set_string (self)
26127 @value{GDBN} will call this method when a @var{parameter}'s value has
26128 been changed via the @code{set} API (for example, @kbd{set foo off}).
26129 The @code{value} attribute has already been populated with the new
26130 value and may be used in output. This method must return a string.
26133 @defun Parameter.get_show_string (self, svalue)
26134 @value{GDBN} will call this method when a @var{parameter}'s
26135 @code{show} API has been invoked (for example, @kbd{show foo}). The
26136 argument @code{svalue} receives the string representation of the
26137 current value. This method must return a string.
26140 When a new parameter is defined, its type must be specified. The
26141 available types are represented by constants defined in the @code{gdb}
26145 @findex PARAM_BOOLEAN
26146 @findex gdb.PARAM_BOOLEAN
26147 @item gdb.PARAM_BOOLEAN
26148 The value is a plain boolean. The Python boolean values, @code{True}
26149 and @code{False} are the only valid values.
26151 @findex PARAM_AUTO_BOOLEAN
26152 @findex gdb.PARAM_AUTO_BOOLEAN
26153 @item gdb.PARAM_AUTO_BOOLEAN
26154 The value has three possible states: true, false, and @samp{auto}. In
26155 Python, true and false are represented using boolean constants, and
26156 @samp{auto} is represented using @code{None}.
26158 @findex PARAM_UINTEGER
26159 @findex gdb.PARAM_UINTEGER
26160 @item gdb.PARAM_UINTEGER
26161 The value is an unsigned integer. The value of 0 should be
26162 interpreted to mean ``unlimited''.
26164 @findex PARAM_INTEGER
26165 @findex gdb.PARAM_INTEGER
26166 @item gdb.PARAM_INTEGER
26167 The value is a signed integer. The value of 0 should be interpreted
26168 to mean ``unlimited''.
26170 @findex PARAM_STRING
26171 @findex gdb.PARAM_STRING
26172 @item gdb.PARAM_STRING
26173 The value is a string. When the user modifies the string, any escape
26174 sequences, such as @samp{\t}, @samp{\f}, and octal escapes, are
26175 translated into corresponding characters and encoded into the current
26178 @findex PARAM_STRING_NOESCAPE
26179 @findex gdb.PARAM_STRING_NOESCAPE
26180 @item gdb.PARAM_STRING_NOESCAPE
26181 The value is a string. When the user modifies the string, escapes are
26182 passed through untranslated.
26184 @findex PARAM_OPTIONAL_FILENAME
26185 @findex gdb.PARAM_OPTIONAL_FILENAME
26186 @item gdb.PARAM_OPTIONAL_FILENAME
26187 The value is a either a filename (a string), or @code{None}.
26189 @findex PARAM_FILENAME
26190 @findex gdb.PARAM_FILENAME
26191 @item gdb.PARAM_FILENAME
26192 The value is a filename. This is just like
26193 @code{PARAM_STRING_NOESCAPE}, but uses file names for completion.
26195 @findex PARAM_ZINTEGER
26196 @findex gdb.PARAM_ZINTEGER
26197 @item gdb.PARAM_ZINTEGER
26198 The value is an integer. This is like @code{PARAM_INTEGER}, except 0
26199 is interpreted as itself.
26202 @findex gdb.PARAM_ENUM
26203 @item gdb.PARAM_ENUM
26204 The value is a string, which must be one of a collection string
26205 constants provided when the parameter is created.
26208 @node Functions In Python
26209 @subsubsection Writing new convenience functions
26211 @cindex writing convenience functions
26212 @cindex convenience functions in python
26213 @cindex python convenience functions
26214 @tindex gdb.Function
26216 You can implement new convenience functions (@pxref{Convenience Vars})
26217 in Python. A convenience function is an instance of a subclass of the
26218 class @code{gdb.Function}.
26220 @defun Function.__init__ (name)
26221 The initializer for @code{Function} registers the new function with
26222 @value{GDBN}. The argument @var{name} is the name of the function,
26223 a string. The function will be visible to the user as a convenience
26224 variable of type @code{internal function}, whose name is the same as
26225 the given @var{name}.
26227 The documentation for the new function is taken from the documentation
26228 string for the new class.
26231 @defun Function.invoke (@var{*args})
26232 When a convenience function is evaluated, its arguments are converted
26233 to instances of @code{gdb.Value}, and then the function's
26234 @code{invoke} method is called. Note that @value{GDBN} does not
26235 predetermine the arity of convenience functions. Instead, all
26236 available arguments are passed to @code{invoke}, following the
26237 standard Python calling convention. In particular, a convenience
26238 function can have default values for parameters without ill effect.
26240 The return value of this method is used as its value in the enclosing
26241 expression. If an ordinary Python value is returned, it is converted
26242 to a @code{gdb.Value} following the usual rules.
26245 The following code snippet shows how a trivial convenience function can
26246 be implemented in Python:
26249 class Greet (gdb.Function):
26250 """Return string to greet someone.
26251 Takes a name as argument."""
26253 def __init__ (self):
26254 super (Greet, self).__init__ ("greet")
26256 def invoke (self, name):
26257 return "Hello, %s!" % name.string ()
26262 The last line instantiates the class, and is necessary to trigger the
26263 registration of the function with @value{GDBN}. Depending on how the
26264 Python code is read into @value{GDBN}, you may need to import the
26265 @code{gdb} module explicitly.
26267 Now you can use the function in an expression:
26270 (gdb) print $greet("Bob")
26274 @node Progspaces In Python
26275 @subsubsection Program Spaces In Python
26277 @cindex progspaces in python
26278 @tindex gdb.Progspace
26280 A program space, or @dfn{progspace}, represents a symbolic view
26281 of an address space.
26282 It consists of all of the objfiles of the program.
26283 @xref{Objfiles In Python}.
26284 @xref{Inferiors and Programs, program spaces}, for more details
26285 about program spaces.
26287 The following progspace-related functions are available in the
26290 @findex gdb.current_progspace
26291 @defun gdb.current_progspace ()
26292 This function returns the program space of the currently selected inferior.
26293 @xref{Inferiors and Programs}.
26296 @findex gdb.progspaces
26297 @defun gdb.progspaces ()
26298 Return a sequence of all the progspaces currently known to @value{GDBN}.
26301 Each progspace is represented by an instance of the @code{gdb.Progspace}
26304 @defvar Progspace.filename
26305 The file name of the progspace as a string.
26308 @defvar Progspace.pretty_printers
26309 The @code{pretty_printers} attribute is a list of functions. It is
26310 used to look up pretty-printers. A @code{Value} is passed to each
26311 function in order; if the function returns @code{None}, then the
26312 search continues. Otherwise, the return value should be an object
26313 which is used to format the value. @xref{Pretty Printing API}, for more
26317 @defvar Progspace.type_printers
26318 The @code{type_printers} attribute is a list of type printer objects.
26319 @xref{Type Printing API}, for more information.
26322 @defvar Progspace.frame_filters
26323 The @code{frame_filters} attribute is a dictionary of frame filter
26324 objects. @xref{Frame Filter API}, for more information.
26327 @node Objfiles In Python
26328 @subsubsection Objfiles In Python
26330 @cindex objfiles in python
26331 @tindex gdb.Objfile
26333 @value{GDBN} loads symbols for an inferior from various
26334 symbol-containing files (@pxref{Files}). These include the primary
26335 executable file, any shared libraries used by the inferior, and any
26336 separate debug info files (@pxref{Separate Debug Files}).
26337 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
26339 The following objfile-related functions are available in the
26342 @findex gdb.current_objfile
26343 @defun gdb.current_objfile ()
26344 When auto-loading a Python script (@pxref{Python Auto-loading}), @value{GDBN}
26345 sets the ``current objfile'' to the corresponding objfile. This
26346 function returns the current objfile. If there is no current objfile,
26347 this function returns @code{None}.
26350 @findex gdb.objfiles
26351 @defun gdb.objfiles ()
26352 Return a sequence of all the objfiles current known to @value{GDBN}.
26353 @xref{Objfiles In Python}.
26356 Each objfile is represented by an instance of the @code{gdb.Objfile}
26359 @defvar Objfile.filename
26360 The file name of the objfile as a string.
26363 @defvar Objfile.pretty_printers
26364 The @code{pretty_printers} attribute is a list of functions. It is
26365 used to look up pretty-printers. A @code{Value} is passed to each
26366 function in order; if the function returns @code{None}, then the
26367 search continues. Otherwise, the return value should be an object
26368 which is used to format the value. @xref{Pretty Printing API}, for more
26372 @defvar Objfile.type_printers
26373 The @code{type_printers} attribute is a list of type printer objects.
26374 @xref{Type Printing API}, for more information.
26377 @defvar Objfile.frame_filters
26378 The @code{frame_filters} attribute is a dictionary of frame filter
26379 objects. @xref{Frame Filter API}, for more information.
26382 A @code{gdb.Objfile} object has the following methods:
26384 @defun Objfile.is_valid ()
26385 Returns @code{True} if the @code{gdb.Objfile} object is valid,
26386 @code{False} if not. A @code{gdb.Objfile} object can become invalid
26387 if the object file it refers to is not loaded in @value{GDBN} any
26388 longer. All other @code{gdb.Objfile} methods will throw an exception
26389 if it is invalid at the time the method is called.
26392 @node Frames In Python
26393 @subsubsection Accessing inferior stack frames from Python.
26395 @cindex frames in python
26396 When the debugged program stops, @value{GDBN} is able to analyze its call
26397 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
26398 represents a frame in the stack. A @code{gdb.Frame} object is only valid
26399 while its corresponding frame exists in the inferior's stack. If you try
26400 to use an invalid frame object, @value{GDBN} will throw a @code{gdb.error}
26401 exception (@pxref{Exception Handling}).
26403 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
26407 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
26411 The following frame-related functions are available in the @code{gdb} module:
26413 @findex gdb.selected_frame
26414 @defun gdb.selected_frame ()
26415 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
26418 @findex gdb.newest_frame
26419 @defun gdb.newest_frame ()
26420 Return the newest frame object for the selected thread.
26423 @defun gdb.frame_stop_reason_string (reason)
26424 Return a string explaining the reason why @value{GDBN} stopped unwinding
26425 frames, as expressed by the given @var{reason} code (an integer, see the
26426 @code{unwind_stop_reason} method further down in this section).
26429 A @code{gdb.Frame} object has the following methods:
26431 @defun Frame.is_valid ()
26432 Returns true if the @code{gdb.Frame} object is valid, false if not.
26433 A frame object can become invalid if the frame it refers to doesn't
26434 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
26435 an exception if it is invalid at the time the method is called.
26438 @defun Frame.name ()
26439 Returns the function name of the frame, or @code{None} if it can't be
26443 @defun Frame.architecture ()
26444 Returns the @code{gdb.Architecture} object corresponding to the frame's
26445 architecture. @xref{Architectures In Python}.
26448 @defun Frame.type ()
26449 Returns the type of the frame. The value can be one of:
26451 @item gdb.NORMAL_FRAME
26452 An ordinary stack frame.
26454 @item gdb.DUMMY_FRAME
26455 A fake stack frame that was created by @value{GDBN} when performing an
26456 inferior function call.
26458 @item gdb.INLINE_FRAME
26459 A frame representing an inlined function. The function was inlined
26460 into a @code{gdb.NORMAL_FRAME} that is older than this one.
26462 @item gdb.TAILCALL_FRAME
26463 A frame representing a tail call. @xref{Tail Call Frames}.
26465 @item gdb.SIGTRAMP_FRAME
26466 A signal trampoline frame. This is the frame created by the OS when
26467 it calls into a signal handler.
26469 @item gdb.ARCH_FRAME
26470 A fake stack frame representing a cross-architecture call.
26472 @item gdb.SENTINEL_FRAME
26473 This is like @code{gdb.NORMAL_FRAME}, but it is only used for the
26478 @defun Frame.unwind_stop_reason ()
26479 Return an integer representing the reason why it's not possible to find
26480 more frames toward the outermost frame. Use
26481 @code{gdb.frame_stop_reason_string} to convert the value returned by this
26482 function to a string. The value can be one of:
26485 @item gdb.FRAME_UNWIND_NO_REASON
26486 No particular reason (older frames should be available).
26488 @item gdb.FRAME_UNWIND_NULL_ID
26489 The previous frame's analyzer returns an invalid result.
26491 @item gdb.FRAME_UNWIND_OUTERMOST
26492 This frame is the outermost.
26494 @item gdb.FRAME_UNWIND_UNAVAILABLE
26495 Cannot unwind further, because that would require knowing the
26496 values of registers or memory that have not been collected.
26498 @item gdb.FRAME_UNWIND_INNER_ID
26499 This frame ID looks like it ought to belong to a NEXT frame,
26500 but we got it for a PREV frame. Normally, this is a sign of
26501 unwinder failure. It could also indicate stack corruption.
26503 @item gdb.FRAME_UNWIND_SAME_ID
26504 This frame has the same ID as the previous one. That means
26505 that unwinding further would almost certainly give us another
26506 frame with exactly the same ID, so break the chain. Normally,
26507 this is a sign of unwinder failure. It could also indicate
26510 @item gdb.FRAME_UNWIND_NO_SAVED_PC
26511 The frame unwinder did not find any saved PC, but we needed
26512 one to unwind further.
26514 @item gdb.FRAME_UNWIND_FIRST_ERROR
26515 Any stop reason greater or equal to this value indicates some kind
26516 of error. This special value facilitates writing code that tests
26517 for errors in unwinding in a way that will work correctly even if
26518 the list of the other values is modified in future @value{GDBN}
26519 versions. Using it, you could write:
26521 reason = gdb.selected_frame().unwind_stop_reason ()
26522 reason_str = gdb.frame_stop_reason_string (reason)
26523 if reason >= gdb.FRAME_UNWIND_FIRST_ERROR:
26524 print "An error occured: %s" % reason_str
26531 Returns the frame's resume address.
26534 @defun Frame.block ()
26535 Return the frame's code block. @xref{Blocks In Python}.
26538 @defun Frame.function ()
26539 Return the symbol for the function corresponding to this frame.
26540 @xref{Symbols In Python}.
26543 @defun Frame.older ()
26544 Return the frame that called this frame.
26547 @defun Frame.newer ()
26548 Return the frame called by this frame.
26551 @defun Frame.find_sal ()
26552 Return the frame's symtab and line object.
26553 @xref{Symbol Tables In Python}.
26556 @defun Frame.read_var (variable @r{[}, block@r{]})
26557 Return the value of @var{variable} in this frame. If the optional
26558 argument @var{block} is provided, search for the variable from that
26559 block; otherwise start at the frame's current block (which is
26560 determined by the frame's current program counter). @var{variable}
26561 must be a string or a @code{gdb.Symbol} object. @var{block} must be a
26562 @code{gdb.Block} object.
26565 @defun Frame.select ()
26566 Set this frame to be the selected frame. @xref{Stack, ,Examining the
26570 @node Blocks In Python
26571 @subsubsection Accessing blocks from Python.
26573 @cindex blocks in python
26576 In @value{GDBN}, symbols are stored in blocks. A block corresponds
26577 roughly to a scope in the source code. Blocks are organized
26578 hierarchically, and are represented individually in Python as a
26579 @code{gdb.Block}. Blocks rely on debugging information being
26582 A frame has a block. Please see @ref{Frames In Python}, for a more
26583 in-depth discussion of frames.
26585 The outermost block is known as the @dfn{global block}. The global
26586 block typically holds public global variables and functions.
26588 The block nested just inside the global block is the @dfn{static
26589 block}. The static block typically holds file-scoped variables and
26592 @value{GDBN} provides a method to get a block's superblock, but there
26593 is currently no way to examine the sub-blocks of a block, or to
26594 iterate over all the blocks in a symbol table (@pxref{Symbol Tables In
26597 Here is a short example that should help explain blocks:
26600 /* This is in the global block. */
26603 /* This is in the static block. */
26604 static int file_scope;
26606 /* 'function' is in the global block, and 'argument' is
26607 in a block nested inside of 'function'. */
26608 int function (int argument)
26610 /* 'local' is in a block inside 'function'. It may or may
26611 not be in the same block as 'argument'. */
26615 /* 'inner' is in a block whose superblock is the one holding
26619 /* If this call is expanded by the compiler, you may see
26620 a nested block here whose function is 'inline_function'
26621 and whose superblock is the one holding 'inner'. */
26622 inline_function ();
26627 A @code{gdb.Block} is iterable. The iterator returns the symbols
26628 (@pxref{Symbols In Python}) local to the block. Python programs
26629 should not assume that a specific block object will always contain a
26630 given symbol, since changes in @value{GDBN} features and
26631 infrastructure may cause symbols move across blocks in a symbol
26634 The following block-related functions are available in the @code{gdb}
26637 @findex gdb.block_for_pc
26638 @defun gdb.block_for_pc (pc)
26639 Return the innermost @code{gdb.Block} containing the given @var{pc}
26640 value. If the block cannot be found for the @var{pc} value specified,
26641 the function will return @code{None}.
26644 A @code{gdb.Block} object has the following methods:
26646 @defun Block.is_valid ()
26647 Returns @code{True} if the @code{gdb.Block} object is valid,
26648 @code{False} if not. A block object can become invalid if the block it
26649 refers to doesn't exist anymore in the inferior. All other
26650 @code{gdb.Block} methods will throw an exception if it is invalid at
26651 the time the method is called. The block's validity is also checked
26652 during iteration over symbols of the block.
26655 A @code{gdb.Block} object has the following attributes:
26657 @defvar Block.start
26658 The start address of the block. This attribute is not writable.
26662 The end address of the block. This attribute is not writable.
26665 @defvar Block.function
26666 The name of the block represented as a @code{gdb.Symbol}. If the
26667 block is not named, then this attribute holds @code{None}. This
26668 attribute is not writable.
26670 For ordinary function blocks, the superblock is the static block.
26671 However, you should note that it is possible for a function block to
26672 have a superblock that is not the static block -- for instance this
26673 happens for an inlined function.
26676 @defvar Block.superblock
26677 The block containing this block. If this parent block does not exist,
26678 this attribute holds @code{None}. This attribute is not writable.
26681 @defvar Block.global_block
26682 The global block associated with this block. This attribute is not
26686 @defvar Block.static_block
26687 The static block associated with this block. This attribute is not
26691 @defvar Block.is_global
26692 @code{True} if the @code{gdb.Block} object is a global block,
26693 @code{False} if not. This attribute is not
26697 @defvar Block.is_static
26698 @code{True} if the @code{gdb.Block} object is a static block,
26699 @code{False} if not. This attribute is not writable.
26702 @node Symbols In Python
26703 @subsubsection Python representation of Symbols.
26705 @cindex symbols in python
26708 @value{GDBN} represents every variable, function and type as an
26709 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
26710 Similarly, Python represents these symbols in @value{GDBN} with the
26711 @code{gdb.Symbol} object.
26713 The following symbol-related functions are available in the @code{gdb}
26716 @findex gdb.lookup_symbol
26717 @defun gdb.lookup_symbol (name @r{[}, block @r{[}, domain@r{]]})
26718 This function searches for a symbol by name. The search scope can be
26719 restricted to the parameters defined in the optional domain and block
26722 @var{name} is the name of the symbol. It must be a string. The
26723 optional @var{block} argument restricts the search to symbols visible
26724 in that @var{block}. The @var{block} argument must be a
26725 @code{gdb.Block} object. If omitted, the block for the current frame
26726 is used. The optional @var{domain} argument restricts
26727 the search to the domain type. The @var{domain} argument must be a
26728 domain constant defined in the @code{gdb} module and described later
26731 The result is a tuple of two elements.
26732 The first element is a @code{gdb.Symbol} object or @code{None} if the symbol
26734 If the symbol is found, the second element is @code{True} if the symbol
26735 is a field of a method's object (e.g., @code{this} in C@t{++}),
26736 otherwise it is @code{False}.
26737 If the symbol is not found, the second element is @code{False}.
26740 @findex gdb.lookup_global_symbol
26741 @defun gdb.lookup_global_symbol (name @r{[}, domain@r{]})
26742 This function searches for a global symbol by name.
26743 The search scope can be restricted to by the domain argument.
26745 @var{name} is the name of the symbol. It must be a string.
26746 The optional @var{domain} argument restricts the search to the domain type.
26747 The @var{domain} argument must be a domain constant defined in the @code{gdb}
26748 module and described later in this chapter.
26750 The result is a @code{gdb.Symbol} object or @code{None} if the symbol
26754 A @code{gdb.Symbol} object has the following attributes:
26756 @defvar Symbol.type
26757 The type of the symbol or @code{None} if no type is recorded.
26758 This attribute is represented as a @code{gdb.Type} object.
26759 @xref{Types In Python}. This attribute is not writable.
26762 @defvar Symbol.symtab
26763 The symbol table in which the symbol appears. This attribute is
26764 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
26765 Python}. This attribute is not writable.
26768 @defvar Symbol.line
26769 The line number in the source code at which the symbol was defined.
26770 This is an integer.
26773 @defvar Symbol.name
26774 The name of the symbol as a string. This attribute is not writable.
26777 @defvar Symbol.linkage_name
26778 The name of the symbol, as used by the linker (i.e., may be mangled).
26779 This attribute is not writable.
26782 @defvar Symbol.print_name
26783 The name of the symbol in a form suitable for output. This is either
26784 @code{name} or @code{linkage_name}, depending on whether the user
26785 asked @value{GDBN} to display demangled or mangled names.
26788 @defvar Symbol.addr_class
26789 The address class of the symbol. This classifies how to find the value
26790 of a symbol. Each address class is a constant defined in the
26791 @code{gdb} module and described later in this chapter.
26794 @defvar Symbol.needs_frame
26795 This is @code{True} if evaluating this symbol's value requires a frame
26796 (@pxref{Frames In Python}) and @code{False} otherwise. Typically,
26797 local variables will require a frame, but other symbols will not.
26800 @defvar Symbol.is_argument
26801 @code{True} if the symbol is an argument of a function.
26804 @defvar Symbol.is_constant
26805 @code{True} if the symbol is a constant.
26808 @defvar Symbol.is_function
26809 @code{True} if the symbol is a function or a method.
26812 @defvar Symbol.is_variable
26813 @code{True} if the symbol is a variable.
26816 A @code{gdb.Symbol} object has the following methods:
26818 @defun Symbol.is_valid ()
26819 Returns @code{True} if the @code{gdb.Symbol} object is valid,
26820 @code{False} if not. A @code{gdb.Symbol} object can become invalid if
26821 the symbol it refers to does not exist in @value{GDBN} any longer.
26822 All other @code{gdb.Symbol} methods will throw an exception if it is
26823 invalid at the time the method is called.
26826 @defun Symbol.value (@r{[}frame@r{]})
26827 Compute the value of the symbol, as a @code{gdb.Value}. For
26828 functions, this computes the address of the function, cast to the
26829 appropriate type. If the symbol requires a frame in order to compute
26830 its value, then @var{frame} must be given. If @var{frame} is not
26831 given, or if @var{frame} is invalid, then this method will throw an
26835 The available domain categories in @code{gdb.Symbol} are represented
26836 as constants in the @code{gdb} module:
26839 @findex SYMBOL_UNDEF_DOMAIN
26840 @findex gdb.SYMBOL_UNDEF_DOMAIN
26841 @item gdb.SYMBOL_UNDEF_DOMAIN
26842 This is used when a domain has not been discovered or none of the
26843 following domains apply. This usually indicates an error either
26844 in the symbol information or in @value{GDBN}'s handling of symbols.
26845 @findex SYMBOL_VAR_DOMAIN
26846 @findex gdb.SYMBOL_VAR_DOMAIN
26847 @item gdb.SYMBOL_VAR_DOMAIN
26848 This domain contains variables, function names, typedef names and enum
26850 @findex SYMBOL_STRUCT_DOMAIN
26851 @findex gdb.SYMBOL_STRUCT_DOMAIN
26852 @item gdb.SYMBOL_STRUCT_DOMAIN
26853 This domain holds struct, union and enum type names.
26854 @findex SYMBOL_LABEL_DOMAIN
26855 @findex gdb.SYMBOL_LABEL_DOMAIN
26856 @item gdb.SYMBOL_LABEL_DOMAIN
26857 This domain contains names of labels (for gotos).
26858 @findex SYMBOL_VARIABLES_DOMAIN
26859 @findex gdb.SYMBOL_VARIABLES_DOMAIN
26860 @item gdb.SYMBOL_VARIABLES_DOMAIN
26861 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
26862 contains everything minus functions and types.
26863 @findex SYMBOL_FUNCTIONS_DOMAIN
26864 @findex gdb.SYMBOL_FUNCTIONS_DOMAIN
26865 @item gdb.SYMBOL_FUNCTION_DOMAIN
26866 This domain contains all functions.
26867 @findex SYMBOL_TYPES_DOMAIN
26868 @findex gdb.SYMBOL_TYPES_DOMAIN
26869 @item gdb.SYMBOL_TYPES_DOMAIN
26870 This domain contains all types.
26873 The available address class categories in @code{gdb.Symbol} are represented
26874 as constants in the @code{gdb} module:
26877 @findex SYMBOL_LOC_UNDEF
26878 @findex gdb.SYMBOL_LOC_UNDEF
26879 @item gdb.SYMBOL_LOC_UNDEF
26880 If this is returned by address class, it indicates an error either in
26881 the symbol information or in @value{GDBN}'s handling of symbols.
26882 @findex SYMBOL_LOC_CONST
26883 @findex gdb.SYMBOL_LOC_CONST
26884 @item gdb.SYMBOL_LOC_CONST
26885 Value is constant int.
26886 @findex SYMBOL_LOC_STATIC
26887 @findex gdb.SYMBOL_LOC_STATIC
26888 @item gdb.SYMBOL_LOC_STATIC
26889 Value is at a fixed address.
26890 @findex SYMBOL_LOC_REGISTER
26891 @findex gdb.SYMBOL_LOC_REGISTER
26892 @item gdb.SYMBOL_LOC_REGISTER
26893 Value is in a register.
26894 @findex SYMBOL_LOC_ARG
26895 @findex gdb.SYMBOL_LOC_ARG
26896 @item gdb.SYMBOL_LOC_ARG
26897 Value is an argument. This value is at the offset stored within the
26898 symbol inside the frame's argument list.
26899 @findex SYMBOL_LOC_REF_ARG
26900 @findex gdb.SYMBOL_LOC_REF_ARG
26901 @item gdb.SYMBOL_LOC_REF_ARG
26902 Value address is stored in the frame's argument list. Just like
26903 @code{LOC_ARG} except that the value's address is stored at the
26904 offset, not the value itself.
26905 @findex SYMBOL_LOC_REGPARM_ADDR
26906 @findex gdb.SYMBOL_LOC_REGPARM_ADDR
26907 @item gdb.SYMBOL_LOC_REGPARM_ADDR
26908 Value is a specified register. Just like @code{LOC_REGISTER} except
26909 the register holds the address of the argument instead of the argument
26911 @findex SYMBOL_LOC_LOCAL
26912 @findex gdb.SYMBOL_LOC_LOCAL
26913 @item gdb.SYMBOL_LOC_LOCAL
26914 Value is a local variable.
26915 @findex SYMBOL_LOC_TYPEDEF
26916 @findex gdb.SYMBOL_LOC_TYPEDEF
26917 @item gdb.SYMBOL_LOC_TYPEDEF
26918 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
26920 @findex SYMBOL_LOC_BLOCK
26921 @findex gdb.SYMBOL_LOC_BLOCK
26922 @item gdb.SYMBOL_LOC_BLOCK
26924 @findex SYMBOL_LOC_CONST_BYTES
26925 @findex gdb.SYMBOL_LOC_CONST_BYTES
26926 @item gdb.SYMBOL_LOC_CONST_BYTES
26927 Value is a byte-sequence.
26928 @findex SYMBOL_LOC_UNRESOLVED
26929 @findex gdb.SYMBOL_LOC_UNRESOLVED
26930 @item gdb.SYMBOL_LOC_UNRESOLVED
26931 Value is at a fixed address, but the address of the variable has to be
26932 determined from the minimal symbol table whenever the variable is
26934 @findex SYMBOL_LOC_OPTIMIZED_OUT
26935 @findex gdb.SYMBOL_LOC_OPTIMIZED_OUT
26936 @item gdb.SYMBOL_LOC_OPTIMIZED_OUT
26937 The value does not actually exist in the program.
26938 @findex SYMBOL_LOC_COMPUTED
26939 @findex gdb.SYMBOL_LOC_COMPUTED
26940 @item gdb.SYMBOL_LOC_COMPUTED
26941 The value's address is a computed location.
26944 @node Symbol Tables In Python
26945 @subsubsection Symbol table representation in Python.
26947 @cindex symbol tables in python
26949 @tindex gdb.Symtab_and_line
26951 Access to symbol table data maintained by @value{GDBN} on the inferior
26952 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
26953 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
26954 from the @code{find_sal} method in @code{gdb.Frame} object.
26955 @xref{Frames In Python}.
26957 For more information on @value{GDBN}'s symbol table management, see
26958 @ref{Symbols, ,Examining the Symbol Table}, for more information.
26960 A @code{gdb.Symtab_and_line} object has the following attributes:
26962 @defvar Symtab_and_line.symtab
26963 The symbol table object (@code{gdb.Symtab}) for this frame.
26964 This attribute is not writable.
26967 @defvar Symtab_and_line.pc
26968 Indicates the start of the address range occupied by code for the
26969 current source line. This attribute is not writable.
26972 @defvar Symtab_and_line.last
26973 Indicates the end of the address range occupied by code for the current
26974 source line. This attribute is not writable.
26977 @defvar Symtab_and_line.line
26978 Indicates the current line number for this object. This
26979 attribute is not writable.
26982 A @code{gdb.Symtab_and_line} object has the following methods:
26984 @defun Symtab_and_line.is_valid ()
26985 Returns @code{True} if the @code{gdb.Symtab_and_line} object is valid,
26986 @code{False} if not. A @code{gdb.Symtab_and_line} object can become
26987 invalid if the Symbol table and line object it refers to does not
26988 exist in @value{GDBN} any longer. All other
26989 @code{gdb.Symtab_and_line} methods will throw an exception if it is
26990 invalid at the time the method is called.
26993 A @code{gdb.Symtab} object has the following attributes:
26995 @defvar Symtab.filename
26996 The symbol table's source filename. This attribute is not writable.
26999 @defvar Symtab.objfile
27000 The symbol table's backing object file. @xref{Objfiles In Python}.
27001 This attribute is not writable.
27004 A @code{gdb.Symtab} object has the following methods:
27006 @defun Symtab.is_valid ()
27007 Returns @code{True} if the @code{gdb.Symtab} object is valid,
27008 @code{False} if not. A @code{gdb.Symtab} object can become invalid if
27009 the symbol table it refers to does not exist in @value{GDBN} any
27010 longer. All other @code{gdb.Symtab} methods will throw an exception
27011 if it is invalid at the time the method is called.
27014 @defun Symtab.fullname ()
27015 Return the symbol table's source absolute file name.
27018 @defun Symtab.global_block ()
27019 Return the global block of the underlying symbol table.
27020 @xref{Blocks In Python}.
27023 @defun Symtab.static_block ()
27024 Return the static block of the underlying symbol table.
27025 @xref{Blocks In Python}.
27028 @node Breakpoints In Python
27029 @subsubsection Manipulating breakpoints using Python
27031 @cindex breakpoints in python
27032 @tindex gdb.Breakpoint
27034 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
27037 @defun Breakpoint.__init__ (spec @r{[}, type @r{[}, wp_class @r{[},internal @r{[},temporary@r{]]]]})
27038 Create a new breakpoint. @var{spec} is a string naming the location
27039 of the breakpoint, or an expression that defines a watchpoint. The
27040 contents can be any location recognized by the @code{break} command,
27041 or in the case of a watchpoint, by the @code{watch} command. The
27042 optional @var{type} denotes the breakpoint to create from the types
27043 defined later in this chapter. This argument can be either:
27044 @code{gdb.BP_BREAKPOINT} or @code{gdb.BP_WATCHPOINT}. @var{type}
27045 defaults to @code{gdb.BP_BREAKPOINT}. The optional @var{internal}
27046 argument allows the breakpoint to become invisible to the user. The
27047 breakpoint will neither be reported when created, nor will it be
27048 listed in the output from @code{info breakpoints} (but will be listed
27049 with the @code{maint info breakpoints} command). The optional
27050 @var{temporary} argument makes the breakpoint a temporary breakpoint.
27051 Temporary breakpoints are deleted after they have been hit. Any
27052 further access to the Python breakpoint after it has been hit will
27053 result in a runtime error (as that breakpoint has now been
27054 automatically deleted). The optional @var{wp_class} argument defines
27055 the class of watchpoint to create, if @var{type} is
27056 @code{gdb.BP_WATCHPOINT}. If a watchpoint class is not provided, it
27057 is assumed to be a @code{gdb.WP_WRITE} class.
27060 @defun Breakpoint.stop (self)
27061 The @code{gdb.Breakpoint} class can be sub-classed and, in
27062 particular, you may choose to implement the @code{stop} method.
27063 If this method is defined as a sub-class of @code{gdb.Breakpoint},
27064 it will be called when the inferior reaches any location of a
27065 breakpoint which instantiates that sub-class. If the method returns
27066 @code{True}, the inferior will be stopped at the location of the
27067 breakpoint, otherwise the inferior will continue.
27069 If there are multiple breakpoints at the same location with a
27070 @code{stop} method, each one will be called regardless of the
27071 return status of the previous. This ensures that all @code{stop}
27072 methods have a chance to execute at that location. In this scenario
27073 if one of the methods returns @code{True} but the others return
27074 @code{False}, the inferior will still be stopped.
27076 You should not alter the execution state of the inferior (i.e.@:, step,
27077 next, etc.), alter the current frame context (i.e.@:, change the current
27078 active frame), or alter, add or delete any breakpoint. As a general
27079 rule, you should not alter any data within @value{GDBN} or the inferior
27082 Example @code{stop} implementation:
27085 class MyBreakpoint (gdb.Breakpoint):
27087 inf_val = gdb.parse_and_eval("foo")
27094 The available watchpoint types represented by constants are defined in the
27099 @findex gdb.WP_READ
27101 Read only watchpoint.
27104 @findex gdb.WP_WRITE
27106 Write only watchpoint.
27109 @findex gdb.WP_ACCESS
27110 @item gdb.WP_ACCESS
27111 Read/Write watchpoint.
27114 @defun Breakpoint.is_valid ()
27115 Return @code{True} if this @code{Breakpoint} object is valid,
27116 @code{False} otherwise. A @code{Breakpoint} object can become invalid
27117 if the user deletes the breakpoint. In this case, the object still
27118 exists, but the underlying breakpoint does not. In the cases of
27119 watchpoint scope, the watchpoint remains valid even if execution of the
27120 inferior leaves the scope of that watchpoint.
27123 @defun Breakpoint.delete
27124 Permanently deletes the @value{GDBN} breakpoint. This also
27125 invalidates the Python @code{Breakpoint} object. Any further access
27126 to this object's attributes or methods will raise an error.
27129 @defvar Breakpoint.enabled
27130 This attribute is @code{True} if the breakpoint is enabled, and
27131 @code{False} otherwise. This attribute is writable.
27134 @defvar Breakpoint.silent
27135 This attribute is @code{True} if the breakpoint is silent, and
27136 @code{False} otherwise. This attribute is writable.
27138 Note that a breakpoint can also be silent if it has commands and the
27139 first command is @code{silent}. This is not reported by the
27140 @code{silent} attribute.
27143 @defvar Breakpoint.thread
27144 If the breakpoint is thread-specific, this attribute holds the thread
27145 id. If the breakpoint is not thread-specific, this attribute is
27146 @code{None}. This attribute is writable.
27149 @defvar Breakpoint.task
27150 If the breakpoint is Ada task-specific, this attribute holds the Ada task
27151 id. If the breakpoint is not task-specific (or the underlying
27152 language is not Ada), this attribute is @code{None}. This attribute
27156 @defvar Breakpoint.ignore_count
27157 This attribute holds the ignore count for the breakpoint, an integer.
27158 This attribute is writable.
27161 @defvar Breakpoint.number
27162 This attribute holds the breakpoint's number --- the identifier used by
27163 the user to manipulate the breakpoint. This attribute is not writable.
27166 @defvar Breakpoint.type
27167 This attribute holds the breakpoint's type --- the identifier used to
27168 determine the actual breakpoint type or use-case. This attribute is not
27172 @defvar Breakpoint.visible
27173 This attribute tells whether the breakpoint is visible to the user
27174 when set, or when the @samp{info breakpoints} command is run. This
27175 attribute is not writable.
27178 @defvar Breakpoint.temporary
27179 This attribute indicates whether the breakpoint was created as a
27180 temporary breakpoint. Temporary breakpoints are automatically deleted
27181 after that breakpoint has been hit. Access to this attribute, and all
27182 other attributes and functions other than the @code{is_valid}
27183 function, will result in an error after the breakpoint has been hit
27184 (as it has been automatically deleted). This attribute is not
27188 The available types are represented by constants defined in the @code{gdb}
27192 @findex BP_BREAKPOINT
27193 @findex gdb.BP_BREAKPOINT
27194 @item gdb.BP_BREAKPOINT
27195 Normal code breakpoint.
27197 @findex BP_WATCHPOINT
27198 @findex gdb.BP_WATCHPOINT
27199 @item gdb.BP_WATCHPOINT
27200 Watchpoint breakpoint.
27202 @findex BP_HARDWARE_WATCHPOINT
27203 @findex gdb.BP_HARDWARE_WATCHPOINT
27204 @item gdb.BP_HARDWARE_WATCHPOINT
27205 Hardware assisted watchpoint.
27207 @findex BP_READ_WATCHPOINT
27208 @findex gdb.BP_READ_WATCHPOINT
27209 @item gdb.BP_READ_WATCHPOINT
27210 Hardware assisted read watchpoint.
27212 @findex BP_ACCESS_WATCHPOINT
27213 @findex gdb.BP_ACCESS_WATCHPOINT
27214 @item gdb.BP_ACCESS_WATCHPOINT
27215 Hardware assisted access watchpoint.
27218 @defvar Breakpoint.hit_count
27219 This attribute holds the hit count for the breakpoint, an integer.
27220 This attribute is writable, but currently it can only be set to zero.
27223 @defvar Breakpoint.location
27224 This attribute holds the location of the breakpoint, as specified by
27225 the user. It is a string. If the breakpoint does not have a location
27226 (that is, it is a watchpoint) the attribute's value is @code{None}. This
27227 attribute is not writable.
27230 @defvar Breakpoint.expression
27231 This attribute holds a breakpoint expression, as specified by
27232 the user. It is a string. If the breakpoint does not have an
27233 expression (the breakpoint is not a watchpoint) the attribute's value
27234 is @code{None}. This attribute is not writable.
27237 @defvar Breakpoint.condition
27238 This attribute holds the condition of the breakpoint, as specified by
27239 the user. It is a string. If there is no condition, this attribute's
27240 value is @code{None}. This attribute is writable.
27243 @defvar Breakpoint.commands
27244 This attribute holds the commands attached to the breakpoint. If
27245 there are commands, this attribute's value is a string holding all the
27246 commands, separated by newlines. If there are no commands, this
27247 attribute is @code{None}. This attribute is not writable.
27250 @node Finish Breakpoints in Python
27251 @subsubsection Finish Breakpoints
27253 @cindex python finish breakpoints
27254 @tindex gdb.FinishBreakpoint
27256 A finish breakpoint is a temporary breakpoint set at the return address of
27257 a frame, based on the @code{finish} command. @code{gdb.FinishBreakpoint}
27258 extends @code{gdb.Breakpoint}. The underlying breakpoint will be disabled
27259 and deleted when the execution will run out of the breakpoint scope (i.e.@:
27260 @code{Breakpoint.stop} or @code{FinishBreakpoint.out_of_scope} triggered).
27261 Finish breakpoints are thread specific and must be create with the right
27264 @defun FinishBreakpoint.__init__ (@r{[}frame@r{]} @r{[}, internal@r{]})
27265 Create a finish breakpoint at the return address of the @code{gdb.Frame}
27266 object @var{frame}. If @var{frame} is not provided, this defaults to the
27267 newest frame. The optional @var{internal} argument allows the breakpoint to
27268 become invisible to the user. @xref{Breakpoints In Python}, for further
27269 details about this argument.
27272 @defun FinishBreakpoint.out_of_scope (self)
27273 In some circumstances (e.g.@: @code{longjmp}, C@t{++} exceptions, @value{GDBN}
27274 @code{return} command, @dots{}), a function may not properly terminate, and
27275 thus never hit the finish breakpoint. When @value{GDBN} notices such a
27276 situation, the @code{out_of_scope} callback will be triggered.
27278 You may want to sub-class @code{gdb.FinishBreakpoint} and override this
27282 class MyFinishBreakpoint (gdb.FinishBreakpoint)
27284 print "normal finish"
27287 def out_of_scope ():
27288 print "abnormal finish"
27292 @defvar FinishBreakpoint.return_value
27293 When @value{GDBN} is stopped at a finish breakpoint and the frame
27294 used to build the @code{gdb.FinishBreakpoint} object had debug symbols, this
27295 attribute will contain a @code{gdb.Value} object corresponding to the return
27296 value of the function. The value will be @code{None} if the function return
27297 type is @code{void} or if the return value was not computable. This attribute
27301 @node Lazy Strings In Python
27302 @subsubsection Python representation of lazy strings.
27304 @cindex lazy strings in python
27305 @tindex gdb.LazyString
27307 A @dfn{lazy string} is a string whose contents is not retrieved or
27308 encoded until it is needed.
27310 A @code{gdb.LazyString} is represented in @value{GDBN} as an
27311 @code{address} that points to a region of memory, an @code{encoding}
27312 that will be used to encode that region of memory, and a @code{length}
27313 to delimit the region of memory that represents the string. The
27314 difference between a @code{gdb.LazyString} and a string wrapped within
27315 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
27316 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
27317 retrieved and encoded during printing, while a @code{gdb.Value}
27318 wrapping a string is immediately retrieved and encoded on creation.
27320 A @code{gdb.LazyString} object has the following functions:
27322 @defun LazyString.value ()
27323 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
27324 will point to the string in memory, but will lose all the delayed
27325 retrieval, encoding and handling that @value{GDBN} applies to a
27326 @code{gdb.LazyString}.
27329 @defvar LazyString.address
27330 This attribute holds the address of the string. This attribute is not
27334 @defvar LazyString.length
27335 This attribute holds the length of the string in characters. If the
27336 length is -1, then the string will be fetched and encoded up to the
27337 first null of appropriate width. This attribute is not writable.
27340 @defvar LazyString.encoding
27341 This attribute holds the encoding that will be applied to the string
27342 when the string is printed by @value{GDBN}. If the encoding is not
27343 set, or contains an empty string, then @value{GDBN} will select the
27344 most appropriate encoding when the string is printed. This attribute
27348 @defvar LazyString.type
27349 This attribute holds the type that is represented by the lazy string's
27350 type. For a lazy string this will always be a pointer type. To
27351 resolve this to the lazy string's character type, use the type's
27352 @code{target} method. @xref{Types In Python}. This attribute is not
27356 @node Architectures In Python
27357 @subsubsection Python representation of architectures
27358 @cindex Python architectures
27360 @value{GDBN} uses architecture specific parameters and artifacts in a
27361 number of its various computations. An architecture is represented
27362 by an instance of the @code{gdb.Architecture} class.
27364 A @code{gdb.Architecture} class has the following methods:
27366 @defun Architecture.name ()
27367 Return the name (string value) of the architecture.
27370 @defun Architecture.disassemble (@var{start_pc} @r{[}, @var{end_pc} @r{[}, @var{count}@r{]]})
27371 Return a list of disassembled instructions starting from the memory
27372 address @var{start_pc}. The optional arguments @var{end_pc} and
27373 @var{count} determine the number of instructions in the returned list.
27374 If both the optional arguments @var{end_pc} and @var{count} are
27375 specified, then a list of at most @var{count} disassembled instructions
27376 whose start address falls in the closed memory address interval from
27377 @var{start_pc} to @var{end_pc} are returned. If @var{end_pc} is not
27378 specified, but @var{count} is specified, then @var{count} number of
27379 instructions starting from the address @var{start_pc} are returned. If
27380 @var{count} is not specified but @var{end_pc} is specified, then all
27381 instructions whose start address falls in the closed memory address
27382 interval from @var{start_pc} to @var{end_pc} are returned. If neither
27383 @var{end_pc} nor @var{count} are specified, then a single instruction at
27384 @var{start_pc} is returned. For all of these cases, each element of the
27385 returned list is a Python @code{dict} with the following string keys:
27390 The value corresponding to this key is a Python long integer capturing
27391 the memory address of the instruction.
27394 The value corresponding to this key is a string value which represents
27395 the instruction with assembly language mnemonics. The assembly
27396 language flavor used is the same as that specified by the current CLI
27397 variable @code{disassembly-flavor}. @xref{Machine Code}.
27400 The value corresponding to this key is the length (integer value) of the
27401 instruction in bytes.
27406 @node Python Auto-loading
27407 @subsection Python Auto-loading
27408 @cindex Python auto-loading
27410 When a new object file is read (for example, due to the @code{file}
27411 command, or because the inferior has loaded a shared library),
27412 @value{GDBN} will look for Python support scripts in several ways:
27413 @file{@var{objfile}-gdb.py} (@pxref{objfile-gdb.py file})
27414 and @code{.debug_gdb_scripts} section
27415 (@pxref{dotdebug_gdb_scripts section}).
27417 The auto-loading feature is useful for supplying application-specific
27418 debugging commands and scripts.
27420 Auto-loading can be enabled or disabled,
27421 and the list of auto-loaded scripts can be printed.
27424 @anchor{set auto-load python-scripts}
27425 @kindex set auto-load python-scripts
27426 @item set auto-load python-scripts [on|off]
27427 Enable or disable the auto-loading of Python scripts.
27429 @anchor{show auto-load python-scripts}
27430 @kindex show auto-load python-scripts
27431 @item show auto-load python-scripts
27432 Show whether auto-loading of Python scripts is enabled or disabled.
27434 @anchor{info auto-load python-scripts}
27435 @kindex info auto-load python-scripts
27436 @cindex print list of auto-loaded Python scripts
27437 @item info auto-load python-scripts [@var{regexp}]
27438 Print the list of all Python scripts that @value{GDBN} auto-loaded.
27440 Also printed is the list of Python scripts that were mentioned in
27441 the @code{.debug_gdb_scripts} section and were not found
27442 (@pxref{dotdebug_gdb_scripts section}).
27443 This is useful because their names are not printed when @value{GDBN}
27444 tries to load them and fails. There may be many of them, and printing
27445 an error message for each one is problematic.
27447 If @var{regexp} is supplied only Python scripts with matching names are printed.
27452 (gdb) info auto-load python-scripts
27454 Yes py-section-script.py
27455 full name: /tmp/py-section-script.py
27456 No my-foo-pretty-printers.py
27460 When reading an auto-loaded file, @value{GDBN} sets the
27461 @dfn{current objfile}. This is available via the @code{gdb.current_objfile}
27462 function (@pxref{Objfiles In Python}). This can be useful for
27463 registering objfile-specific pretty-printers and frame-filters.
27466 * objfile-gdb.py file:: The @file{@var{objfile}-gdb.py} file
27467 * dotdebug_gdb_scripts section:: The @code{.debug_gdb_scripts} section
27468 * Which flavor to choose?::
27471 @node objfile-gdb.py file
27472 @subsubsection The @file{@var{objfile}-gdb.py} file
27473 @cindex @file{@var{objfile}-gdb.py}
27475 When a new object file is read, @value{GDBN} looks for
27476 a file named @file{@var{objfile}-gdb.py} (we call it @var{script-name} below),
27477 where @var{objfile} is the object file's real name, formed by ensuring
27478 that the file name is absolute, following all symlinks, and resolving
27479 @code{.} and @code{..} components. If this file exists and is
27480 readable, @value{GDBN} will evaluate it as a Python script.
27482 If this file does not exist, then @value{GDBN} will look for
27483 @var{script-name} file in all of the directories as specified below.
27485 Note that loading of this script file also requires accordingly configured
27486 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27488 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
27489 scripts normally according to its @file{.exe} filename. But if no scripts are
27490 found @value{GDBN} also tries script filenames matching the object file without
27491 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
27492 is attempted on any platform. This makes the script filenames compatible
27493 between Unix and MS-Windows hosts.
27496 @anchor{set auto-load scripts-directory}
27497 @kindex set auto-load scripts-directory
27498 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
27499 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
27500 may be delimited by the host platform path separator in use
27501 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
27503 Each entry here needs to be covered also by the security setting
27504 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
27506 @anchor{with-auto-load-dir}
27507 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
27508 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
27509 configuration option @option{--with-auto-load-dir}.
27511 Any reference to @file{$debugdir} will get replaced by
27512 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
27513 reference to @file{$datadir} will get replaced by @var{data-directory} which is
27514 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
27515 @file{$datadir} must be placed as a directory component --- either alone or
27516 delimited by @file{/} or @file{\} directory separators, depending on the host
27519 The list of directories uses path separator (@samp{:} on GNU and Unix
27520 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
27521 to the @env{PATH} environment variable.
27523 @anchor{show auto-load scripts-directory}
27524 @kindex show auto-load scripts-directory
27525 @item show auto-load scripts-directory
27526 Show @value{GDBN} auto-loaded scripts location.
27529 @value{GDBN} does not track which files it has already auto-loaded this way.
27530 @value{GDBN} will load the associated script every time the corresponding
27531 @var{objfile} is opened.
27532 So your @file{-gdb.py} file should be careful to avoid errors if it
27533 is evaluated more than once.
27535 @node dotdebug_gdb_scripts section
27536 @subsubsection The @code{.debug_gdb_scripts} section
27537 @cindex @code{.debug_gdb_scripts} section
27539 For systems using file formats like ELF and COFF,
27540 when @value{GDBN} loads a new object file
27541 it will look for a special section named @samp{.debug_gdb_scripts}.
27542 If this section exists, its contents is a list of names of scripts to load.
27544 @value{GDBN} will look for each specified script file first in the
27545 current directory and then along the source search path
27546 (@pxref{Source Path, ,Specifying Source Directories}),
27547 except that @file{$cdir} is not searched, since the compilation
27548 directory is not relevant to scripts.
27550 Entries can be placed in section @code{.debug_gdb_scripts} with,
27551 for example, this GCC macro:
27554 /* Note: The "MS" section flags are to remove duplicates. */
27555 #define DEFINE_GDB_SCRIPT(script_name) \
27557 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
27559 .asciz \"" script_name "\"\n\
27565 Then one can reference the macro in a header or source file like this:
27568 DEFINE_GDB_SCRIPT ("my-app-scripts.py")
27571 The script name may include directories if desired.
27573 Note that loading of this script file also requires accordingly configured
27574 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27576 If the macro is put in a header, any application or library
27577 using this header will get a reference to the specified script.
27579 @node Which flavor to choose?
27580 @subsubsection Which flavor to choose?
27582 Given the multiple ways of auto-loading Python scripts, it might not always
27583 be clear which one to choose. This section provides some guidance.
27585 Benefits of the @file{-gdb.py} way:
27589 Can be used with file formats that don't support multiple sections.
27592 Ease of finding scripts for public libraries.
27594 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
27595 in the source search path.
27596 For publicly installed libraries, e.g., @file{libstdc++}, there typically
27597 isn't a source directory in which to find the script.
27600 Doesn't require source code additions.
27603 Benefits of the @code{.debug_gdb_scripts} way:
27607 Works with static linking.
27609 Scripts for libraries done the @file{-gdb.py} way require an objfile to
27610 trigger their loading. When an application is statically linked the only
27611 objfile available is the executable, and it is cumbersome to attach all the
27612 scripts from all the input libraries to the executable's @file{-gdb.py} script.
27615 Works with classes that are entirely inlined.
27617 Some classes can be entirely inlined, and thus there may not be an associated
27618 shared library to attach a @file{-gdb.py} script to.
27621 Scripts needn't be copied out of the source tree.
27623 In some circumstances, apps can be built out of large collections of internal
27624 libraries, and the build infrastructure necessary to install the
27625 @file{-gdb.py} scripts in a place where @value{GDBN} can find them is
27626 cumbersome. It may be easier to specify the scripts in the
27627 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
27628 top of the source tree to the source search path.
27631 @node Python modules
27632 @subsection Python modules
27633 @cindex python modules
27635 @value{GDBN} comes with several modules to assist writing Python code.
27638 * gdb.printing:: Building and registering pretty-printers.
27639 * gdb.types:: Utilities for working with types.
27640 * gdb.prompt:: Utilities for prompt value substitution.
27644 @subsubsection gdb.printing
27645 @cindex gdb.printing
27647 This module provides a collection of utilities for working with
27651 @item PrettyPrinter (@var{name}, @var{subprinters}=None)
27652 This class specifies the API that makes @samp{info pretty-printer},
27653 @samp{enable pretty-printer} and @samp{disable pretty-printer} work.
27654 Pretty-printers should generally inherit from this class.
27656 @item SubPrettyPrinter (@var{name})
27657 For printers that handle multiple types, this class specifies the
27658 corresponding API for the subprinters.
27660 @item RegexpCollectionPrettyPrinter (@var{name})
27661 Utility class for handling multiple printers, all recognized via
27662 regular expressions.
27663 @xref{Writing a Pretty-Printer}, for an example.
27665 @item FlagEnumerationPrinter (@var{name})
27666 A pretty-printer which handles printing of @code{enum} values. Unlike
27667 @value{GDBN}'s built-in @code{enum} printing, this printer attempts to
27668 work properly when there is some overlap between the enumeration
27669 constants. @var{name} is the name of the printer and also the name of
27670 the @code{enum} type to look up.
27672 @item register_pretty_printer (@var{obj}, @var{printer}, @var{replace}=False)
27673 Register @var{printer} with the pretty-printer list of @var{obj}.
27674 If @var{replace} is @code{True} then any existing copy of the printer
27675 is replaced. Otherwise a @code{RuntimeError} exception is raised
27676 if a printer with the same name already exists.
27680 @subsubsection gdb.types
27683 This module provides a collection of utilities for working with
27684 @code{gdb.Type} objects.
27687 @item get_basic_type (@var{type})
27688 Return @var{type} with const and volatile qualifiers stripped,
27689 and with typedefs and C@t{++} references converted to the underlying type.
27694 typedef const int const_int;
27696 const_int& foo_ref (foo);
27697 int main () @{ return 0; @}
27704 (gdb) python import gdb.types
27705 (gdb) python foo_ref = gdb.parse_and_eval("foo_ref")
27706 (gdb) python print gdb.types.get_basic_type(foo_ref.type)
27710 @item has_field (@var{type}, @var{field})
27711 Return @code{True} if @var{type}, assumed to be a type with fields
27712 (e.g., a structure or union), has field @var{field}.
27714 @item make_enum_dict (@var{enum_type})
27715 Return a Python @code{dictionary} type produced from @var{enum_type}.
27717 @item deep_items (@var{type})
27718 Returns a Python iterator similar to the standard
27719 @code{gdb.Type.iteritems} method, except that the iterator returned
27720 by @code{deep_items} will recursively traverse anonymous struct or
27721 union fields. For example:
27735 Then in @value{GDBN}:
27737 (@value{GDBP}) python import gdb.types
27738 (@value{GDBP}) python struct_a = gdb.lookup_type("struct A")
27739 (@value{GDBP}) python print struct_a.keys ()
27741 (@value{GDBP}) python print [k for k,v in gdb.types.deep_items(struct_a)]
27742 @{['a', 'b0', 'b1']@}
27745 @item get_type_recognizers ()
27746 Return a list of the enabled type recognizers for the current context.
27747 This is called by @value{GDBN} during the type-printing process
27748 (@pxref{Type Printing API}).
27750 @item apply_type_recognizers (recognizers, type_obj)
27751 Apply the type recognizers, @var{recognizers}, to the type object
27752 @var{type_obj}. If any recognizer returns a string, return that
27753 string. Otherwise, return @code{None}. This is called by
27754 @value{GDBN} during the type-printing process (@pxref{Type Printing
27757 @item register_type_printer (locus, printer)
27758 This is a convenience function to register a type printer.
27759 @var{printer} is the type printer to register. It must implement the
27760 type printer protocol. @var{locus} is either a @code{gdb.Objfile}, in
27761 which case the printer is registered with that objfile; a
27762 @code{gdb.Progspace}, in which case the printer is registered with
27763 that progspace; or @code{None}, in which case the printer is
27764 registered globally.
27767 This is a base class that implements the type printer protocol. Type
27768 printers are encouraged, but not required, to derive from this class.
27769 It defines a constructor:
27771 @defmethod TypePrinter __init__ (self, name)
27772 Initialize the type printer with the given name. The new printer
27773 starts in the enabled state.
27779 @subsubsection gdb.prompt
27782 This module provides a method for prompt value-substitution.
27785 @item substitute_prompt (@var{string})
27786 Return @var{string} with escape sequences substituted by values. Some
27787 escape sequences take arguments. You can specify arguments inside
27788 ``@{@}'' immediately following the escape sequence.
27790 The escape sequences you can pass to this function are:
27794 Substitute a backslash.
27796 Substitute an ESC character.
27798 Substitute the selected frame; an argument names a frame parameter.
27800 Substitute a newline.
27802 Substitute a parameter's value; the argument names the parameter.
27804 Substitute a carriage return.
27806 Substitute the selected thread; an argument names a thread parameter.
27808 Substitute the version of GDB.
27810 Substitute the current working directory.
27812 Begin a sequence of non-printing characters. These sequences are
27813 typically used with the ESC character, and are not counted in the string
27814 length. Example: ``\[\e[0;34m\](gdb)\[\e[0m\]'' will return a
27815 blue-colored ``(gdb)'' prompt where the length is five.
27817 End a sequence of non-printing characters.
27823 substitute_prompt (``frame: \f,
27824 print arguments: \p@{print frame-arguments@}'')
27827 @exdent will return the string:
27830 "frame: main, print arguments: scalars"
27835 @section Creating new spellings of existing commands
27836 @cindex aliases for commands
27838 It is often useful to define alternate spellings of existing commands.
27839 For example, if a new @value{GDBN} command defined in Python has
27840 a long name to type, it is handy to have an abbreviated version of it
27841 that involves less typing.
27843 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
27844 of the @samp{step} command even though it is otherwise an ambiguous
27845 abbreviation of other commands like @samp{set} and @samp{show}.
27847 Aliases are also used to provide shortened or more common versions
27848 of multi-word commands. For example, @value{GDBN} provides the
27849 @samp{tty} alias of the @samp{set inferior-tty} command.
27851 You can define a new alias with the @samp{alias} command.
27856 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
27860 @var{ALIAS} specifies the name of the new alias.
27861 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
27864 @var{COMMAND} specifies the name of an existing command
27865 that is being aliased.
27867 The @samp{-a} option specifies that the new alias is an abbreviation
27868 of the command. Abbreviations are not shown in command
27869 lists displayed by the @samp{help} command.
27871 The @samp{--} option specifies the end of options,
27872 and is useful when @var{ALIAS} begins with a dash.
27874 Here is a simple example showing how to make an abbreviation
27875 of a command so that there is less to type.
27876 Suppose you were tired of typing @samp{disas}, the current
27877 shortest unambiguous abbreviation of the @samp{disassemble} command
27878 and you wanted an even shorter version named @samp{di}.
27879 The following will accomplish this.
27882 (gdb) alias -a di = disas
27885 Note that aliases are different from user-defined commands.
27886 With a user-defined command, you also need to write documentation
27887 for it with the @samp{document} command.
27888 An alias automatically picks up the documentation of the existing command.
27890 Here is an example where we make @samp{elms} an abbreviation of
27891 @samp{elements} in the @samp{set print elements} command.
27892 This is to show that you can make an abbreviation of any part
27896 (gdb) alias -a set print elms = set print elements
27897 (gdb) alias -a show print elms = show print elements
27898 (gdb) set p elms 20
27900 Limit on string chars or array elements to print is 200.
27903 Note that if you are defining an alias of a @samp{set} command,
27904 and you want to have an alias for the corresponding @samp{show}
27905 command, then you need to define the latter separately.
27907 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
27908 @var{ALIAS}, just as they are normally.
27911 (gdb) alias -a set pr elms = set p ele
27914 Finally, here is an example showing the creation of a one word
27915 alias for a more complex command.
27916 This creates alias @samp{spe} of the command @samp{set print elements}.
27919 (gdb) alias spe = set print elements
27924 @chapter Command Interpreters
27925 @cindex command interpreters
27927 @value{GDBN} supports multiple command interpreters, and some command
27928 infrastructure to allow users or user interface writers to switch
27929 between interpreters or run commands in other interpreters.
27931 @value{GDBN} currently supports two command interpreters, the console
27932 interpreter (sometimes called the command-line interpreter or @sc{cli})
27933 and the machine interface interpreter (or @sc{gdb/mi}). This manual
27934 describes both of these interfaces in great detail.
27936 By default, @value{GDBN} will start with the console interpreter.
27937 However, the user may choose to start @value{GDBN} with another
27938 interpreter by specifying the @option{-i} or @option{--interpreter}
27939 startup options. Defined interpreters include:
27943 @cindex console interpreter
27944 The traditional console or command-line interpreter. This is the most often
27945 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
27946 @value{GDBN} will use this interpreter.
27949 @cindex mi interpreter
27950 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
27951 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
27952 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
27956 @cindex mi2 interpreter
27957 The current @sc{gdb/mi} interface.
27960 @cindex mi1 interpreter
27961 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
27965 @cindex invoke another interpreter
27966 The interpreter being used by @value{GDBN} may not be dynamically
27967 switched at runtime. Although possible, this could lead to a very
27968 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
27969 enters the command "interpreter-set console" in a console view,
27970 @value{GDBN} would switch to using the console interpreter, rendering
27971 the IDE inoperable!
27973 @kindex interpreter-exec
27974 Although you may only choose a single interpreter at startup, you may execute
27975 commands in any interpreter from the current interpreter using the appropriate
27976 command. If you are running the console interpreter, simply use the
27977 @code{interpreter-exec} command:
27980 interpreter-exec mi "-data-list-register-names"
27983 @sc{gdb/mi} has a similar command, although it is only available in versions of
27984 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
27987 @chapter @value{GDBN} Text User Interface
27989 @cindex Text User Interface
27992 * TUI Overview:: TUI overview
27993 * TUI Keys:: TUI key bindings
27994 * TUI Single Key Mode:: TUI single key mode
27995 * TUI Commands:: TUI-specific commands
27996 * TUI Configuration:: TUI configuration variables
27999 The @value{GDBN} Text User Interface (TUI) is a terminal
28000 interface which uses the @code{curses} library to show the source
28001 file, the assembly output, the program registers and @value{GDBN}
28002 commands in separate text windows. The TUI mode is supported only
28003 on platforms where a suitable version of the @code{curses} library
28006 The TUI mode is enabled by default when you invoke @value{GDBN} as
28007 @samp{@value{GDBP} -tui}.
28008 You can also switch in and out of TUI mode while @value{GDBN} runs by
28009 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
28010 @xref{TUI Keys, ,TUI Key Bindings}.
28013 @section TUI Overview
28015 In TUI mode, @value{GDBN} can display several text windows:
28019 This window is the @value{GDBN} command window with the @value{GDBN}
28020 prompt and the @value{GDBN} output. The @value{GDBN} input is still
28021 managed using readline.
28024 The source window shows the source file of the program. The current
28025 line and active breakpoints are displayed in this window.
28028 The assembly window shows the disassembly output of the program.
28031 This window shows the processor registers. Registers are highlighted
28032 when their values change.
28035 The source and assembly windows show the current program position
28036 by highlighting the current line and marking it with a @samp{>} marker.
28037 Breakpoints are indicated with two markers. The first marker
28038 indicates the breakpoint type:
28042 Breakpoint which was hit at least once.
28045 Breakpoint which was never hit.
28048 Hardware breakpoint which was hit at least once.
28051 Hardware breakpoint which was never hit.
28054 The second marker indicates whether the breakpoint is enabled or not:
28058 Breakpoint is enabled.
28061 Breakpoint is disabled.
28064 The source, assembly and register windows are updated when the current
28065 thread changes, when the frame changes, or when the program counter
28068 These windows are not all visible at the same time. The command
28069 window is always visible. The others can be arranged in several
28080 source and assembly,
28083 source and registers, or
28086 assembly and registers.
28089 A status line above the command window shows the following information:
28093 Indicates the current @value{GDBN} target.
28094 (@pxref{Targets, ,Specifying a Debugging Target}).
28097 Gives the current process or thread number.
28098 When no process is being debugged, this field is set to @code{No process}.
28101 Gives the current function name for the selected frame.
28102 The name is demangled if demangling is turned on (@pxref{Print Settings}).
28103 When there is no symbol corresponding to the current program counter,
28104 the string @code{??} is displayed.
28107 Indicates the current line number for the selected frame.
28108 When the current line number is not known, the string @code{??} is displayed.
28111 Indicates the current program counter address.
28115 @section TUI Key Bindings
28116 @cindex TUI key bindings
28118 The TUI installs several key bindings in the readline keymaps
28119 @ifset SYSTEM_READLINE
28120 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
28122 @ifclear SYSTEM_READLINE
28123 (@pxref{Command Line Editing}).
28125 The following key bindings are installed for both TUI mode and the
28126 @value{GDBN} standard mode.
28135 Enter or leave the TUI mode. When leaving the TUI mode,
28136 the curses window management stops and @value{GDBN} operates using
28137 its standard mode, writing on the terminal directly. When reentering
28138 the TUI mode, control is given back to the curses windows.
28139 The screen is then refreshed.
28143 Use a TUI layout with only one window. The layout will
28144 either be @samp{source} or @samp{assembly}. When the TUI mode
28145 is not active, it will switch to the TUI mode.
28147 Think of this key binding as the Emacs @kbd{C-x 1} binding.
28151 Use a TUI layout with at least two windows. When the current
28152 layout already has two windows, the next layout with two windows is used.
28153 When a new layout is chosen, one window will always be common to the
28154 previous layout and the new one.
28156 Think of it as the Emacs @kbd{C-x 2} binding.
28160 Change the active window. The TUI associates several key bindings
28161 (like scrolling and arrow keys) with the active window. This command
28162 gives the focus to the next TUI window.
28164 Think of it as the Emacs @kbd{C-x o} binding.
28168 Switch in and out of the TUI SingleKey mode that binds single
28169 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
28172 The following key bindings only work in the TUI mode:
28177 Scroll the active window one page up.
28181 Scroll the active window one page down.
28185 Scroll the active window one line up.
28189 Scroll the active window one line down.
28193 Scroll the active window one column left.
28197 Scroll the active window one column right.
28201 Refresh the screen.
28204 Because the arrow keys scroll the active window in the TUI mode, they
28205 are not available for their normal use by readline unless the command
28206 window has the focus. When another window is active, you must use
28207 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
28208 and @kbd{C-f} to control the command window.
28210 @node TUI Single Key Mode
28211 @section TUI Single Key Mode
28212 @cindex TUI single key mode
28214 The TUI also provides a @dfn{SingleKey} mode, which binds several
28215 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
28216 switch into this mode, where the following key bindings are used:
28219 @kindex c @r{(SingleKey TUI key)}
28223 @kindex d @r{(SingleKey TUI key)}
28227 @kindex f @r{(SingleKey TUI key)}
28231 @kindex n @r{(SingleKey TUI key)}
28235 @kindex q @r{(SingleKey TUI key)}
28237 exit the SingleKey mode.
28239 @kindex r @r{(SingleKey TUI key)}
28243 @kindex s @r{(SingleKey TUI key)}
28247 @kindex u @r{(SingleKey TUI key)}
28251 @kindex v @r{(SingleKey TUI key)}
28255 @kindex w @r{(SingleKey TUI key)}
28260 Other keys temporarily switch to the @value{GDBN} command prompt.
28261 The key that was pressed is inserted in the editing buffer so that
28262 it is possible to type most @value{GDBN} commands without interaction
28263 with the TUI SingleKey mode. Once the command is entered the TUI
28264 SingleKey mode is restored. The only way to permanently leave
28265 this mode is by typing @kbd{q} or @kbd{C-x s}.
28269 @section TUI-specific Commands
28270 @cindex TUI commands
28272 The TUI has specific commands to control the text windows.
28273 These commands are always available, even when @value{GDBN} is not in
28274 the TUI mode. When @value{GDBN} is in the standard mode, most
28275 of these commands will automatically switch to the TUI mode.
28277 Note that if @value{GDBN}'s @code{stdout} is not connected to a
28278 terminal, or @value{GDBN} has been started with the machine interface
28279 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
28280 these commands will fail with an error, because it would not be
28281 possible or desirable to enable curses window management.
28286 List and give the size of all displayed windows.
28290 Display the next layout.
28293 Display the previous layout.
28296 Display the source window only.
28299 Display the assembly window only.
28302 Display the source and assembly window.
28305 Display the register window together with the source or assembly window.
28309 Make the next window active for scrolling.
28312 Make the previous window active for scrolling.
28315 Make the source window active for scrolling.
28318 Make the assembly window active for scrolling.
28321 Make the register window active for scrolling.
28324 Make the command window active for scrolling.
28328 Refresh the screen. This is similar to typing @kbd{C-L}.
28330 @item tui reg float
28332 Show the floating point registers in the register window.
28334 @item tui reg general
28335 Show the general registers in the register window.
28338 Show the next register group. The list of register groups as well as
28339 their order is target specific. The predefined register groups are the
28340 following: @code{general}, @code{float}, @code{system}, @code{vector},
28341 @code{all}, @code{save}, @code{restore}.
28343 @item tui reg system
28344 Show the system registers in the register window.
28348 Update the source window and the current execution point.
28350 @item winheight @var{name} +@var{count}
28351 @itemx winheight @var{name} -@var{count}
28353 Change the height of the window @var{name} by @var{count}
28354 lines. Positive counts increase the height, while negative counts
28357 @item tabset @var{nchars}
28359 Set the width of tab stops to be @var{nchars} characters.
28362 @node TUI Configuration
28363 @section TUI Configuration Variables
28364 @cindex TUI configuration variables
28366 Several configuration variables control the appearance of TUI windows.
28369 @item set tui border-kind @var{kind}
28370 @kindex set tui border-kind
28371 Select the border appearance for the source, assembly and register windows.
28372 The possible values are the following:
28375 Use a space character to draw the border.
28378 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
28381 Use the Alternate Character Set to draw the border. The border is
28382 drawn using character line graphics if the terminal supports them.
28385 @item set tui border-mode @var{mode}
28386 @kindex set tui border-mode
28387 @itemx set tui active-border-mode @var{mode}
28388 @kindex set tui active-border-mode
28389 Select the display attributes for the borders of the inactive windows
28390 or the active window. The @var{mode} can be one of the following:
28393 Use normal attributes to display the border.
28399 Use reverse video mode.
28402 Use half bright mode.
28404 @item half-standout
28405 Use half bright and standout mode.
28408 Use extra bright or bold mode.
28410 @item bold-standout
28411 Use extra bright or bold and standout mode.
28416 @chapter Using @value{GDBN} under @sc{gnu} Emacs
28419 @cindex @sc{gnu} Emacs
28420 A special interface allows you to use @sc{gnu} Emacs to view (and
28421 edit) the source files for the program you are debugging with
28424 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
28425 executable file you want to debug as an argument. This command starts
28426 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
28427 created Emacs buffer.
28428 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
28430 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
28435 All ``terminal'' input and output goes through an Emacs buffer, called
28438 This applies both to @value{GDBN} commands and their output, and to the input
28439 and output done by the program you are debugging.
28441 This is useful because it means that you can copy the text of previous
28442 commands and input them again; you can even use parts of the output
28445 All the facilities of Emacs' Shell mode are available for interacting
28446 with your program. In particular, you can send signals the usual
28447 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
28451 @value{GDBN} displays source code through Emacs.
28453 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
28454 source file for that frame and puts an arrow (@samp{=>}) at the
28455 left margin of the current line. Emacs uses a separate buffer for
28456 source display, and splits the screen to show both your @value{GDBN} session
28459 Explicit @value{GDBN} @code{list} or search commands still produce output as
28460 usual, but you probably have no reason to use them from Emacs.
28463 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
28464 a graphical mode, enabled by default, which provides further buffers
28465 that can control the execution and describe the state of your program.
28466 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
28468 If you specify an absolute file name when prompted for the @kbd{M-x
28469 gdb} argument, then Emacs sets your current working directory to where
28470 your program resides. If you only specify the file name, then Emacs
28471 sets your current working directory to the directory associated
28472 with the previous buffer. In this case, @value{GDBN} may find your
28473 program by searching your environment's @code{PATH} variable, but on
28474 some operating systems it might not find the source. So, although the
28475 @value{GDBN} input and output session proceeds normally, the auxiliary
28476 buffer does not display the current source and line of execution.
28478 The initial working directory of @value{GDBN} is printed on the top
28479 line of the GUD buffer and this serves as a default for the commands
28480 that specify files for @value{GDBN} to operate on. @xref{Files,
28481 ,Commands to Specify Files}.
28483 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
28484 need to call @value{GDBN} by a different name (for example, if you
28485 keep several configurations around, with different names) you can
28486 customize the Emacs variable @code{gud-gdb-command-name} to run the
28489 In the GUD buffer, you can use these special Emacs commands in
28490 addition to the standard Shell mode commands:
28494 Describe the features of Emacs' GUD Mode.
28497 Execute to another source line, like the @value{GDBN} @code{step} command; also
28498 update the display window to show the current file and location.
28501 Execute to next source line in this function, skipping all function
28502 calls, like the @value{GDBN} @code{next} command. Then update the display window
28503 to show the current file and location.
28506 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
28507 display window accordingly.
28510 Execute until exit from the selected stack frame, like the @value{GDBN}
28511 @code{finish} command.
28514 Continue execution of your program, like the @value{GDBN} @code{continue}
28518 Go up the number of frames indicated by the numeric argument
28519 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
28520 like the @value{GDBN} @code{up} command.
28523 Go down the number of frames indicated by the numeric argument, like the
28524 @value{GDBN} @code{down} command.
28527 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
28528 tells @value{GDBN} to set a breakpoint on the source line point is on.
28530 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
28531 separate frame which shows a backtrace when the GUD buffer is current.
28532 Move point to any frame in the stack and type @key{RET} to make it
28533 become the current frame and display the associated source in the
28534 source buffer. Alternatively, click @kbd{Mouse-2} to make the
28535 selected frame become the current one. In graphical mode, the
28536 speedbar displays watch expressions.
28538 If you accidentally delete the source-display buffer, an easy way to get
28539 it back is to type the command @code{f} in the @value{GDBN} buffer, to
28540 request a frame display; when you run under Emacs, this recreates
28541 the source buffer if necessary to show you the context of the current
28544 The source files displayed in Emacs are in ordinary Emacs buffers
28545 which are visiting the source files in the usual way. You can edit
28546 the files with these buffers if you wish; but keep in mind that @value{GDBN}
28547 communicates with Emacs in terms of line numbers. If you add or
28548 delete lines from the text, the line numbers that @value{GDBN} knows cease
28549 to correspond properly with the code.
28551 A more detailed description of Emacs' interaction with @value{GDBN} is
28552 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
28556 @chapter The @sc{gdb/mi} Interface
28558 @unnumberedsec Function and Purpose
28560 @cindex @sc{gdb/mi}, its purpose
28561 @sc{gdb/mi} is a line based machine oriented text interface to
28562 @value{GDBN} and is activated by specifying using the
28563 @option{--interpreter} command line option (@pxref{Mode Options}). It
28564 is specifically intended to support the development of systems which
28565 use the debugger as just one small component of a larger system.
28567 This chapter is a specification of the @sc{gdb/mi} interface. It is written
28568 in the form of a reference manual.
28570 Note that @sc{gdb/mi} is still under construction, so some of the
28571 features described below are incomplete and subject to change
28572 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
28574 @unnumberedsec Notation and Terminology
28576 @cindex notational conventions, for @sc{gdb/mi}
28577 This chapter uses the following notation:
28581 @code{|} separates two alternatives.
28584 @code{[ @var{something} ]} indicates that @var{something} is optional:
28585 it may or may not be given.
28588 @code{( @var{group} )*} means that @var{group} inside the parentheses
28589 may repeat zero or more times.
28592 @code{( @var{group} )+} means that @var{group} inside the parentheses
28593 may repeat one or more times.
28596 @code{"@var{string}"} means a literal @var{string}.
28600 @heading Dependencies
28604 * GDB/MI General Design::
28605 * GDB/MI Command Syntax::
28606 * GDB/MI Compatibility with CLI::
28607 * GDB/MI Development and Front Ends::
28608 * GDB/MI Output Records::
28609 * GDB/MI Simple Examples::
28610 * GDB/MI Command Description Format::
28611 * GDB/MI Breakpoint Commands::
28612 * GDB/MI Catchpoint Commands::
28613 * GDB/MI Program Context::
28614 * GDB/MI Thread Commands::
28615 * GDB/MI Ada Tasking Commands::
28616 * GDB/MI Program Execution::
28617 * GDB/MI Stack Manipulation::
28618 * GDB/MI Variable Objects::
28619 * GDB/MI Data Manipulation::
28620 * GDB/MI Tracepoint Commands::
28621 * GDB/MI Symbol Query::
28622 * GDB/MI File Commands::
28624 * GDB/MI Kod Commands::
28625 * GDB/MI Memory Overlay Commands::
28626 * GDB/MI Signal Handling Commands::
28628 * GDB/MI Target Manipulation::
28629 * GDB/MI File Transfer Commands::
28630 * GDB/MI Miscellaneous Commands::
28633 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28634 @node GDB/MI General Design
28635 @section @sc{gdb/mi} General Design
28636 @cindex GDB/MI General Design
28638 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
28639 parts---commands sent to @value{GDBN}, responses to those commands
28640 and notifications. Each command results in exactly one response,
28641 indicating either successful completion of the command, or an error.
28642 For the commands that do not resume the target, the response contains the
28643 requested information. For the commands that resume the target, the
28644 response only indicates whether the target was successfully resumed.
28645 Notifications is the mechanism for reporting changes in the state of the
28646 target, or in @value{GDBN} state, that cannot conveniently be associated with
28647 a command and reported as part of that command response.
28649 The important examples of notifications are:
28653 Exec notifications. These are used to report changes in
28654 target state---when a target is resumed, or stopped. It would not
28655 be feasible to include this information in response of resuming
28656 commands, because one resume commands can result in multiple events in
28657 different threads. Also, quite some time may pass before any event
28658 happens in the target, while a frontend needs to know whether the resuming
28659 command itself was successfully executed.
28662 Console output, and status notifications. Console output
28663 notifications are used to report output of CLI commands, as well as
28664 diagnostics for other commands. Status notifications are used to
28665 report the progress of a long-running operation. Naturally, including
28666 this information in command response would mean no output is produced
28667 until the command is finished, which is undesirable.
28670 General notifications. Commands may have various side effects on
28671 the @value{GDBN} or target state beyond their official purpose. For example,
28672 a command may change the selected thread. Although such changes can
28673 be included in command response, using notification allows for more
28674 orthogonal frontend design.
28678 There's no guarantee that whenever an MI command reports an error,
28679 @value{GDBN} or the target are in any specific state, and especially,
28680 the state is not reverted to the state before the MI command was
28681 processed. Therefore, whenever an MI command results in an error,
28682 we recommend that the frontend refreshes all the information shown in
28683 the user interface.
28687 * Context management::
28688 * Asynchronous and non-stop modes::
28692 @node Context management
28693 @subsection Context management
28695 In most cases when @value{GDBN} accesses the target, this access is
28696 done in context of a specific thread and frame (@pxref{Frames}).
28697 Often, even when accessing global data, the target requires that a thread
28698 be specified. The CLI interface maintains the selected thread and frame,
28699 and supplies them to target on each command. This is convenient,
28700 because a command line user would not want to specify that information
28701 explicitly on each command, and because user interacts with
28702 @value{GDBN} via a single terminal, so no confusion is possible as
28703 to what thread and frame are the current ones.
28705 In the case of MI, the concept of selected thread and frame is less
28706 useful. First, a frontend can easily remember this information
28707 itself. Second, a graphical frontend can have more than one window,
28708 each one used for debugging a different thread, and the frontend might
28709 want to access additional threads for internal purposes. This
28710 increases the risk that by relying on implicitly selected thread, the
28711 frontend may be operating on a wrong one. Therefore, each MI command
28712 should explicitly specify which thread and frame to operate on. To
28713 make it possible, each MI command accepts the @samp{--thread} and
28714 @samp{--frame} options, the value to each is @value{GDBN} identifier
28715 for thread and frame to operate on.
28717 Usually, each top-level window in a frontend allows the user to select
28718 a thread and a frame, and remembers the user selection for further
28719 operations. However, in some cases @value{GDBN} may suggest that the
28720 current thread be changed. For example, when stopping on a breakpoint
28721 it is reasonable to switch to the thread where breakpoint is hit. For
28722 another example, if the user issues the CLI @samp{thread} command via
28723 the frontend, it is desirable to change the frontend's selected thread to the
28724 one specified by user. @value{GDBN} communicates the suggestion to
28725 change current thread using the @samp{=thread-selected} notification.
28726 No such notification is available for the selected frame at the moment.
28728 Note that historically, MI shares the selected thread with CLI, so
28729 frontends used the @code{-thread-select} to execute commands in the
28730 right context. However, getting this to work right is cumbersome. The
28731 simplest way is for frontend to emit @code{-thread-select} command
28732 before every command. This doubles the number of commands that need
28733 to be sent. The alternative approach is to suppress @code{-thread-select}
28734 if the selected thread in @value{GDBN} is supposed to be identical to the
28735 thread the frontend wants to operate on. However, getting this
28736 optimization right can be tricky. In particular, if the frontend
28737 sends several commands to @value{GDBN}, and one of the commands changes the
28738 selected thread, then the behaviour of subsequent commands will
28739 change. So, a frontend should either wait for response from such
28740 problematic commands, or explicitly add @code{-thread-select} for
28741 all subsequent commands. No frontend is known to do this exactly
28742 right, so it is suggested to just always pass the @samp{--thread} and
28743 @samp{--frame} options.
28745 @node Asynchronous and non-stop modes
28746 @subsection Asynchronous command execution and non-stop mode
28748 On some targets, @value{GDBN} is capable of processing MI commands
28749 even while the target is running. This is called @dfn{asynchronous
28750 command execution} (@pxref{Background Execution}). The frontend may
28751 specify a preferrence for asynchronous execution using the
28752 @code{-gdb-set target-async 1} command, which should be emitted before
28753 either running the executable or attaching to the target. After the
28754 frontend has started the executable or attached to the target, it can
28755 find if asynchronous execution is enabled using the
28756 @code{-list-target-features} command.
28758 Even if @value{GDBN} can accept a command while target is running,
28759 many commands that access the target do not work when the target is
28760 running. Therefore, asynchronous command execution is most useful
28761 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
28762 it is possible to examine the state of one thread, while other threads
28765 When a given thread is running, MI commands that try to access the
28766 target in the context of that thread may not work, or may work only on
28767 some targets. In particular, commands that try to operate on thread's
28768 stack will not work, on any target. Commands that read memory, or
28769 modify breakpoints, may work or not work, depending on the target. Note
28770 that even commands that operate on global state, such as @code{print},
28771 @code{set}, and breakpoint commands, still access the target in the
28772 context of a specific thread, so frontend should try to find a
28773 stopped thread and perform the operation on that thread (using the
28774 @samp{--thread} option).
28776 Which commands will work in the context of a running thread is
28777 highly target dependent. However, the two commands
28778 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
28779 to find the state of a thread, will always work.
28781 @node Thread groups
28782 @subsection Thread groups
28783 @value{GDBN} may be used to debug several processes at the same time.
28784 On some platfroms, @value{GDBN} may support debugging of several
28785 hardware systems, each one having several cores with several different
28786 processes running on each core. This section describes the MI
28787 mechanism to support such debugging scenarios.
28789 The key observation is that regardless of the structure of the
28790 target, MI can have a global list of threads, because most commands that
28791 accept the @samp{--thread} option do not need to know what process that
28792 thread belongs to. Therefore, it is not necessary to introduce
28793 neither additional @samp{--process} option, nor an notion of the
28794 current process in the MI interface. The only strictly new feature
28795 that is required is the ability to find how the threads are grouped
28798 To allow the user to discover such grouping, and to support arbitrary
28799 hierarchy of machines/cores/processes, MI introduces the concept of a
28800 @dfn{thread group}. Thread group is a collection of threads and other
28801 thread groups. A thread group always has a string identifier, a type,
28802 and may have additional attributes specific to the type. A new
28803 command, @code{-list-thread-groups}, returns the list of top-level
28804 thread groups, which correspond to processes that @value{GDBN} is
28805 debugging at the moment. By passing an identifier of a thread group
28806 to the @code{-list-thread-groups} command, it is possible to obtain
28807 the members of specific thread group.
28809 To allow the user to easily discover processes, and other objects, he
28810 wishes to debug, a concept of @dfn{available thread group} is
28811 introduced. Available thread group is an thread group that
28812 @value{GDBN} is not debugging, but that can be attached to, using the
28813 @code{-target-attach} command. The list of available top-level thread
28814 groups can be obtained using @samp{-list-thread-groups --available}.
28815 In general, the content of a thread group may be only retrieved only
28816 after attaching to that thread group.
28818 Thread groups are related to inferiors (@pxref{Inferiors and
28819 Programs}). Each inferior corresponds to a thread group of a special
28820 type @samp{process}, and some additional operations are permitted on
28821 such thread groups.
28823 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28824 @node GDB/MI Command Syntax
28825 @section @sc{gdb/mi} Command Syntax
28828 * GDB/MI Input Syntax::
28829 * GDB/MI Output Syntax::
28832 @node GDB/MI Input Syntax
28833 @subsection @sc{gdb/mi} Input Syntax
28835 @cindex input syntax for @sc{gdb/mi}
28836 @cindex @sc{gdb/mi}, input syntax
28838 @item @var{command} @expansion{}
28839 @code{@var{cli-command} | @var{mi-command}}
28841 @item @var{cli-command} @expansion{}
28842 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
28843 @var{cli-command} is any existing @value{GDBN} CLI command.
28845 @item @var{mi-command} @expansion{}
28846 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
28847 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
28849 @item @var{token} @expansion{}
28850 "any sequence of digits"
28852 @item @var{option} @expansion{}
28853 @code{"-" @var{parameter} [ " " @var{parameter} ]}
28855 @item @var{parameter} @expansion{}
28856 @code{@var{non-blank-sequence} | @var{c-string}}
28858 @item @var{operation} @expansion{}
28859 @emph{any of the operations described in this chapter}
28861 @item @var{non-blank-sequence} @expansion{}
28862 @emph{anything, provided it doesn't contain special characters such as
28863 "-", @var{nl}, """ and of course " "}
28865 @item @var{c-string} @expansion{}
28866 @code{""" @var{seven-bit-iso-c-string-content} """}
28868 @item @var{nl} @expansion{}
28877 The CLI commands are still handled by the @sc{mi} interpreter; their
28878 output is described below.
28881 The @code{@var{token}}, when present, is passed back when the command
28885 Some @sc{mi} commands accept optional arguments as part of the parameter
28886 list. Each option is identified by a leading @samp{-} (dash) and may be
28887 followed by an optional argument parameter. Options occur first in the
28888 parameter list and can be delimited from normal parameters using
28889 @samp{--} (this is useful when some parameters begin with a dash).
28896 We want easy access to the existing CLI syntax (for debugging).
28899 We want it to be easy to spot a @sc{mi} operation.
28902 @node GDB/MI Output Syntax
28903 @subsection @sc{gdb/mi} Output Syntax
28905 @cindex output syntax of @sc{gdb/mi}
28906 @cindex @sc{gdb/mi}, output syntax
28907 The output from @sc{gdb/mi} consists of zero or more out-of-band records
28908 followed, optionally, by a single result record. This result record
28909 is for the most recent command. The sequence of output records is
28910 terminated by @samp{(gdb)}.
28912 If an input command was prefixed with a @code{@var{token}} then the
28913 corresponding output for that command will also be prefixed by that same
28917 @item @var{output} @expansion{}
28918 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
28920 @item @var{result-record} @expansion{}
28921 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
28923 @item @var{out-of-band-record} @expansion{}
28924 @code{@var{async-record} | @var{stream-record}}
28926 @item @var{async-record} @expansion{}
28927 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
28929 @item @var{exec-async-output} @expansion{}
28930 @code{[ @var{token} ] "*" @var{async-output}}
28932 @item @var{status-async-output} @expansion{}
28933 @code{[ @var{token} ] "+" @var{async-output}}
28935 @item @var{notify-async-output} @expansion{}
28936 @code{[ @var{token} ] "=" @var{async-output}}
28938 @item @var{async-output} @expansion{}
28939 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
28941 @item @var{result-class} @expansion{}
28942 @code{"done" | "running" | "connected" | "error" | "exit"}
28944 @item @var{async-class} @expansion{}
28945 @code{"stopped" | @var{others}} (where @var{others} will be added
28946 depending on the needs---this is still in development).
28948 @item @var{result} @expansion{}
28949 @code{ @var{variable} "=" @var{value}}
28951 @item @var{variable} @expansion{}
28952 @code{ @var{string} }
28954 @item @var{value} @expansion{}
28955 @code{ @var{const} | @var{tuple} | @var{list} }
28957 @item @var{const} @expansion{}
28958 @code{@var{c-string}}
28960 @item @var{tuple} @expansion{}
28961 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
28963 @item @var{list} @expansion{}
28964 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
28965 @var{result} ( "," @var{result} )* "]" }
28967 @item @var{stream-record} @expansion{}
28968 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
28970 @item @var{console-stream-output} @expansion{}
28971 @code{"~" @var{c-string}}
28973 @item @var{target-stream-output} @expansion{}
28974 @code{"@@" @var{c-string}}
28976 @item @var{log-stream-output} @expansion{}
28977 @code{"&" @var{c-string}}
28979 @item @var{nl} @expansion{}
28982 @item @var{token} @expansion{}
28983 @emph{any sequence of digits}.
28991 All output sequences end in a single line containing a period.
28994 The @code{@var{token}} is from the corresponding request. Note that
28995 for all async output, while the token is allowed by the grammar and
28996 may be output by future versions of @value{GDBN} for select async
28997 output messages, it is generally omitted. Frontends should treat
28998 all async output as reporting general changes in the state of the
28999 target and there should be no need to associate async output to any
29003 @cindex status output in @sc{gdb/mi}
29004 @var{status-async-output} contains on-going status information about the
29005 progress of a slow operation. It can be discarded. All status output is
29006 prefixed by @samp{+}.
29009 @cindex async output in @sc{gdb/mi}
29010 @var{exec-async-output} contains asynchronous state change on the target
29011 (stopped, started, disappeared). All async output is prefixed by
29015 @cindex notify output in @sc{gdb/mi}
29016 @var{notify-async-output} contains supplementary information that the
29017 client should handle (e.g., a new breakpoint information). All notify
29018 output is prefixed by @samp{=}.
29021 @cindex console output in @sc{gdb/mi}
29022 @var{console-stream-output} is output that should be displayed as is in the
29023 console. It is the textual response to a CLI command. All the console
29024 output is prefixed by @samp{~}.
29027 @cindex target output in @sc{gdb/mi}
29028 @var{target-stream-output} is the output produced by the target program.
29029 All the target output is prefixed by @samp{@@}.
29032 @cindex log output in @sc{gdb/mi}
29033 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
29034 instance messages that should be displayed as part of an error log. All
29035 the log output is prefixed by @samp{&}.
29038 @cindex list output in @sc{gdb/mi}
29039 New @sc{gdb/mi} commands should only output @var{lists} containing
29045 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
29046 details about the various output records.
29048 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29049 @node GDB/MI Compatibility with CLI
29050 @section @sc{gdb/mi} Compatibility with CLI
29052 @cindex compatibility, @sc{gdb/mi} and CLI
29053 @cindex @sc{gdb/mi}, compatibility with CLI
29055 For the developers convenience CLI commands can be entered directly,
29056 but there may be some unexpected behaviour. For example, commands
29057 that query the user will behave as if the user replied yes, breakpoint
29058 command lists are not executed and some CLI commands, such as
29059 @code{if}, @code{when} and @code{define}, prompt for further input with
29060 @samp{>}, which is not valid MI output.
29062 This feature may be removed at some stage in the future and it is
29063 recommended that front ends use the @code{-interpreter-exec} command
29064 (@pxref{-interpreter-exec}).
29066 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29067 @node GDB/MI Development and Front Ends
29068 @section @sc{gdb/mi} Development and Front Ends
29069 @cindex @sc{gdb/mi} development
29071 The application which takes the MI output and presents the state of the
29072 program being debugged to the user is called a @dfn{front end}.
29074 Although @sc{gdb/mi} is still incomplete, it is currently being used
29075 by a variety of front ends to @value{GDBN}. This makes it difficult
29076 to introduce new functionality without breaking existing usage. This
29077 section tries to minimize the problems by describing how the protocol
29080 Some changes in MI need not break a carefully designed front end, and
29081 for these the MI version will remain unchanged. The following is a
29082 list of changes that may occur within one level, so front ends should
29083 parse MI output in a way that can handle them:
29087 New MI commands may be added.
29090 New fields may be added to the output of any MI command.
29093 The range of values for fields with specified values, e.g.,
29094 @code{in_scope} (@pxref{-var-update}) may be extended.
29096 @c The format of field's content e.g type prefix, may change so parse it
29097 @c at your own risk. Yes, in general?
29099 @c The order of fields may change? Shouldn't really matter but it might
29100 @c resolve inconsistencies.
29103 If the changes are likely to break front ends, the MI version level
29104 will be increased by one. This will allow the front end to parse the
29105 output according to the MI version. Apart from mi0, new versions of
29106 @value{GDBN} will not support old versions of MI and it will be the
29107 responsibility of the front end to work with the new one.
29109 @c Starting with mi3, add a new command -mi-version that prints the MI
29112 The best way to avoid unexpected changes in MI that might break your front
29113 end is to make your project known to @value{GDBN} developers and
29114 follow development on @email{gdb@@sourceware.org} and
29115 @email{gdb-patches@@sourceware.org}.
29116 @cindex mailing lists
29118 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29119 @node GDB/MI Output Records
29120 @section @sc{gdb/mi} Output Records
29123 * GDB/MI Result Records::
29124 * GDB/MI Stream Records::
29125 * GDB/MI Async Records::
29126 * GDB/MI Breakpoint Information::
29127 * GDB/MI Frame Information::
29128 * GDB/MI Thread Information::
29129 * GDB/MI Ada Exception Information::
29132 @node GDB/MI Result Records
29133 @subsection @sc{gdb/mi} Result Records
29135 @cindex result records in @sc{gdb/mi}
29136 @cindex @sc{gdb/mi}, result records
29137 In addition to a number of out-of-band notifications, the response to a
29138 @sc{gdb/mi} command includes one of the following result indications:
29142 @item "^done" [ "," @var{results} ]
29143 The synchronous operation was successful, @code{@var{results}} are the return
29148 This result record is equivalent to @samp{^done}. Historically, it
29149 was output instead of @samp{^done} if the command has resumed the
29150 target. This behaviour is maintained for backward compatibility, but
29151 all frontends should treat @samp{^done} and @samp{^running}
29152 identically and rely on the @samp{*running} output record to determine
29153 which threads are resumed.
29157 @value{GDBN} has connected to a remote target.
29159 @item "^error" "," @var{c-string}
29161 The operation failed. The @code{@var{c-string}} contains the corresponding
29166 @value{GDBN} has terminated.
29170 @node GDB/MI Stream Records
29171 @subsection @sc{gdb/mi} Stream Records
29173 @cindex @sc{gdb/mi}, stream records
29174 @cindex stream records in @sc{gdb/mi}
29175 @value{GDBN} internally maintains a number of output streams: the console, the
29176 target, and the log. The output intended for each of these streams is
29177 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
29179 Each stream record begins with a unique @dfn{prefix character} which
29180 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
29181 Syntax}). In addition to the prefix, each stream record contains a
29182 @code{@var{string-output}}. This is either raw text (with an implicit new
29183 line) or a quoted C string (which does not contain an implicit newline).
29186 @item "~" @var{string-output}
29187 The console output stream contains text that should be displayed in the
29188 CLI console window. It contains the textual responses to CLI commands.
29190 @item "@@" @var{string-output}
29191 The target output stream contains any textual output from the running
29192 target. This is only present when GDB's event loop is truly
29193 asynchronous, which is currently only the case for remote targets.
29195 @item "&" @var{string-output}
29196 The log stream contains debugging messages being produced by @value{GDBN}'s
29200 @node GDB/MI Async Records
29201 @subsection @sc{gdb/mi} Async Records
29203 @cindex async records in @sc{gdb/mi}
29204 @cindex @sc{gdb/mi}, async records
29205 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
29206 additional changes that have occurred. Those changes can either be a
29207 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
29208 target activity (e.g., target stopped).
29210 The following is the list of possible async records:
29214 @item *running,thread-id="@var{thread}"
29215 The target is now running. The @var{thread} field tells which
29216 specific thread is now running, and can be @samp{all} if all threads
29217 are running. The frontend should assume that no interaction with a
29218 running thread is possible after this notification is produced.
29219 The frontend should not assume that this notification is output
29220 only once for any command. @value{GDBN} may emit this notification
29221 several times, either for different threads, because it cannot resume
29222 all threads together, or even for a single thread, if the thread must
29223 be stepped though some code before letting it run freely.
29225 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
29226 The target has stopped. The @var{reason} field can have one of the
29230 @item breakpoint-hit
29231 A breakpoint was reached.
29232 @item watchpoint-trigger
29233 A watchpoint was triggered.
29234 @item read-watchpoint-trigger
29235 A read watchpoint was triggered.
29236 @item access-watchpoint-trigger
29237 An access watchpoint was triggered.
29238 @item function-finished
29239 An -exec-finish or similar CLI command was accomplished.
29240 @item location-reached
29241 An -exec-until or similar CLI command was accomplished.
29242 @item watchpoint-scope
29243 A watchpoint has gone out of scope.
29244 @item end-stepping-range
29245 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
29246 similar CLI command was accomplished.
29247 @item exited-signalled
29248 The inferior exited because of a signal.
29250 The inferior exited.
29251 @item exited-normally
29252 The inferior exited normally.
29253 @item signal-received
29254 A signal was received by the inferior.
29256 The inferior has stopped due to a library being loaded or unloaded.
29257 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
29258 set or when a @code{catch load} or @code{catch unload} catchpoint is
29259 in use (@pxref{Set Catchpoints}).
29261 The inferior has forked. This is reported when @code{catch fork}
29262 (@pxref{Set Catchpoints}) has been used.
29264 The inferior has vforked. This is reported in when @code{catch vfork}
29265 (@pxref{Set Catchpoints}) has been used.
29266 @item syscall-entry
29267 The inferior entered a system call. This is reported when @code{catch
29268 syscall} (@pxref{Set Catchpoints}) has been used.
29269 @item syscall-entry
29270 The inferior returned from a system call. This is reported when
29271 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
29273 The inferior called @code{exec}. This is reported when @code{catch exec}
29274 (@pxref{Set Catchpoints}) has been used.
29277 The @var{id} field identifies the thread that directly caused the stop
29278 -- for example by hitting a breakpoint. Depending on whether all-stop
29279 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
29280 stop all threads, or only the thread that directly triggered the stop.
29281 If all threads are stopped, the @var{stopped} field will have the
29282 value of @code{"all"}. Otherwise, the value of the @var{stopped}
29283 field will be a list of thread identifiers. Presently, this list will
29284 always include a single thread, but frontend should be prepared to see
29285 several threads in the list. The @var{core} field reports the
29286 processor core on which the stop event has happened. This field may be absent
29287 if such information is not available.
29289 @item =thread-group-added,id="@var{id}"
29290 @itemx =thread-group-removed,id="@var{id}"
29291 A thread group was either added or removed. The @var{id} field
29292 contains the @value{GDBN} identifier of the thread group. When a thread
29293 group is added, it generally might not be associated with a running
29294 process. When a thread group is removed, its id becomes invalid and
29295 cannot be used in any way.
29297 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
29298 A thread group became associated with a running program,
29299 either because the program was just started or the thread group
29300 was attached to a program. The @var{id} field contains the
29301 @value{GDBN} identifier of the thread group. The @var{pid} field
29302 contains process identifier, specific to the operating system.
29304 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
29305 A thread group is no longer associated with a running program,
29306 either because the program has exited, or because it was detached
29307 from. The @var{id} field contains the @value{GDBN} identifier of the
29308 thread group. @var{code} is the exit code of the inferior; it exists
29309 only when the inferior exited with some code.
29311 @item =thread-created,id="@var{id}",group-id="@var{gid}"
29312 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
29313 A thread either was created, or has exited. The @var{id} field
29314 contains the @value{GDBN} identifier of the thread. The @var{gid}
29315 field identifies the thread group this thread belongs to.
29317 @item =thread-selected,id="@var{id}"
29318 Informs that the selected thread was changed as result of the last
29319 command. This notification is not emitted as result of @code{-thread-select}
29320 command but is emitted whenever an MI command that is not documented
29321 to change the selected thread actually changes it. In particular,
29322 invoking, directly or indirectly (via user-defined command), the CLI
29323 @code{thread} command, will generate this notification.
29325 We suggest that in response to this notification, front ends
29326 highlight the selected thread and cause subsequent commands to apply to
29329 @item =library-loaded,...
29330 Reports that a new library file was loaded by the program. This
29331 notification has 4 fields---@var{id}, @var{target-name},
29332 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
29333 opaque identifier of the library. For remote debugging case,
29334 @var{target-name} and @var{host-name} fields give the name of the
29335 library file on the target, and on the host respectively. For native
29336 debugging, both those fields have the same value. The
29337 @var{symbols-loaded} field is emitted only for backward compatibility
29338 and should not be relied on to convey any useful information. The
29339 @var{thread-group} field, if present, specifies the id of the thread
29340 group in whose context the library was loaded. If the field is
29341 absent, it means the library was loaded in the context of all present
29344 @item =library-unloaded,...
29345 Reports that a library was unloaded by the program. This notification
29346 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
29347 the same meaning as for the @code{=library-loaded} notification.
29348 The @var{thread-group} field, if present, specifies the id of the
29349 thread group in whose context the library was unloaded. If the field is
29350 absent, it means the library was unloaded in the context of all present
29353 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
29354 @itemx =traceframe-changed,end
29355 Reports that the trace frame was changed and its new number is
29356 @var{tfnum}. The number of the tracepoint associated with this trace
29357 frame is @var{tpnum}.
29359 @item =tsv-created,name=@var{name},initial=@var{initial}
29360 Reports that the new trace state variable @var{name} is created with
29361 initial value @var{initial}.
29363 @item =tsv-deleted,name=@var{name}
29364 @itemx =tsv-deleted
29365 Reports that the trace state variable @var{name} is deleted or all
29366 trace state variables are deleted.
29368 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
29369 Reports that the trace state variable @var{name} is modified with
29370 the initial value @var{initial}. The current value @var{current} of
29371 trace state variable is optional and is reported if the current
29372 value of trace state variable is known.
29374 @item =breakpoint-created,bkpt=@{...@}
29375 @itemx =breakpoint-modified,bkpt=@{...@}
29376 @itemx =breakpoint-deleted,id=@var{number}
29377 Reports that a breakpoint was created, modified, or deleted,
29378 respectively. Only user-visible breakpoints are reported to the MI
29381 The @var{bkpt} argument is of the same form as returned by the various
29382 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
29383 @var{number} is the ordinal number of the breakpoint.
29385 Note that if a breakpoint is emitted in the result record of a
29386 command, then it will not also be emitted in an async record.
29388 @item =record-started,thread-group="@var{id}"
29389 @itemx =record-stopped,thread-group="@var{id}"
29390 Execution log recording was either started or stopped on an
29391 inferior. The @var{id} is the @value{GDBN} identifier of the thread
29392 group corresponding to the affected inferior.
29394 @item =cmd-param-changed,param=@var{param},value=@var{value}
29395 Reports that a parameter of the command @code{set @var{param}} is
29396 changed to @var{value}. In the multi-word @code{set} command,
29397 the @var{param} is the whole parameter list to @code{set} command.
29398 For example, In command @code{set check type on}, @var{param}
29399 is @code{check type} and @var{value} is @code{on}.
29401 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
29402 Reports that bytes from @var{addr} to @var{data} + @var{len} were
29403 written in an inferior. The @var{id} is the identifier of the
29404 thread group corresponding to the affected inferior. The optional
29405 @code{type="code"} part is reported if the memory written to holds
29409 @node GDB/MI Breakpoint Information
29410 @subsection @sc{gdb/mi} Breakpoint Information
29412 When @value{GDBN} reports information about a breakpoint, a
29413 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
29418 The breakpoint number. For a breakpoint that represents one location
29419 of a multi-location breakpoint, this will be a dotted pair, like
29423 The type of the breakpoint. For ordinary breakpoints this will be
29424 @samp{breakpoint}, but many values are possible.
29427 If the type of the breakpoint is @samp{catchpoint}, then this
29428 indicates the exact type of catchpoint.
29431 This is the breakpoint disposition---either @samp{del}, meaning that
29432 the breakpoint will be deleted at the next stop, or @samp{keep},
29433 meaning that the breakpoint will not be deleted.
29436 This indicates whether the breakpoint is enabled, in which case the
29437 value is @samp{y}, or disabled, in which case the value is @samp{n}.
29438 Note that this is not the same as the field @code{enable}.
29441 The address of the breakpoint. This may be a hexidecimal number,
29442 giving the address; or the string @samp{<PENDING>}, for a pending
29443 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
29444 multiple locations. This field will not be present if no address can
29445 be determined. For example, a watchpoint does not have an address.
29448 If known, the function in which the breakpoint appears.
29449 If not known, this field is not present.
29452 The name of the source file which contains this function, if known.
29453 If not known, this field is not present.
29456 The full file name of the source file which contains this function, if
29457 known. If not known, this field is not present.
29460 The line number at which this breakpoint appears, if known.
29461 If not known, this field is not present.
29464 If the source file is not known, this field may be provided. If
29465 provided, this holds the address of the breakpoint, possibly followed
29469 If this breakpoint is pending, this field is present and holds the
29470 text used to set the breakpoint, as entered by the user.
29473 Where this breakpoint's condition is evaluated, either @samp{host} or
29477 If this is a thread-specific breakpoint, then this identifies the
29478 thread in which the breakpoint can trigger.
29481 If this breakpoint is restricted to a particular Ada task, then this
29482 field will hold the task identifier.
29485 If the breakpoint is conditional, this is the condition expression.
29488 The ignore count of the breakpoint.
29491 The enable count of the breakpoint.
29493 @item traceframe-usage
29496 @item static-tracepoint-marker-string-id
29497 For a static tracepoint, the name of the static tracepoint marker.
29500 For a masked watchpoint, this is the mask.
29503 A tracepoint's pass count.
29505 @item original-location
29506 The location of the breakpoint as originally specified by the user.
29507 This field is optional.
29510 The number of times the breakpoint has been hit.
29513 This field is only given for tracepoints. This is either @samp{y},
29514 meaning that the tracepoint is installed, or @samp{n}, meaning that it
29518 Some extra data, the exact contents of which are type-dependent.
29522 For example, here is what the output of @code{-break-insert}
29523 (@pxref{GDB/MI Breakpoint Commands}) might be:
29526 -> -break-insert main
29527 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29528 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29529 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29534 @node GDB/MI Frame Information
29535 @subsection @sc{gdb/mi} Frame Information
29537 Response from many MI commands includes an information about stack
29538 frame. This information is a tuple that may have the following
29543 The level of the stack frame. The innermost frame has the level of
29544 zero. This field is always present.
29547 The name of the function corresponding to the frame. This field may
29548 be absent if @value{GDBN} is unable to determine the function name.
29551 The code address for the frame. This field is always present.
29554 The name of the source files that correspond to the frame's code
29555 address. This field may be absent.
29558 The source line corresponding to the frames' code address. This field
29562 The name of the binary file (either executable or shared library) the
29563 corresponds to the frame's code address. This field may be absent.
29567 @node GDB/MI Thread Information
29568 @subsection @sc{gdb/mi} Thread Information
29570 Whenever @value{GDBN} has to report an information about a thread, it
29571 uses a tuple with the following fields:
29575 The numeric id assigned to the thread by @value{GDBN}. This field is
29579 Target-specific string identifying the thread. This field is always present.
29582 Additional information about the thread provided by the target.
29583 It is supposed to be human-readable and not interpreted by the
29584 frontend. This field is optional.
29587 Either @samp{stopped} or @samp{running}, depending on whether the
29588 thread is presently running. This field is always present.
29591 The value of this field is an integer number of the processor core the
29592 thread was last seen on. This field is optional.
29595 @node GDB/MI Ada Exception Information
29596 @subsection @sc{gdb/mi} Ada Exception Information
29598 Whenever a @code{*stopped} record is emitted because the program
29599 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
29600 @value{GDBN} provides the name of the exception that was raised via
29601 the @code{exception-name} field.
29603 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29604 @node GDB/MI Simple Examples
29605 @section Simple Examples of @sc{gdb/mi} Interaction
29606 @cindex @sc{gdb/mi}, simple examples
29608 This subsection presents several simple examples of interaction using
29609 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
29610 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
29611 the output received from @sc{gdb/mi}.
29613 Note the line breaks shown in the examples are here only for
29614 readability, they don't appear in the real output.
29616 @subheading Setting a Breakpoint
29618 Setting a breakpoint generates synchronous output which contains detailed
29619 information of the breakpoint.
29622 -> -break-insert main
29623 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29624 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29625 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29630 @subheading Program Execution
29632 Program execution generates asynchronous records and MI gives the
29633 reason that execution stopped.
29639 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
29640 frame=@{addr="0x08048564",func="main",
29641 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
29642 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
29647 <- *stopped,reason="exited-normally"
29651 @subheading Quitting @value{GDBN}
29653 Quitting @value{GDBN} just prints the result class @samp{^exit}.
29661 Please note that @samp{^exit} is printed immediately, but it might
29662 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
29663 performs necessary cleanups, including killing programs being debugged
29664 or disconnecting from debug hardware, so the frontend should wait till
29665 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
29666 fails to exit in reasonable time.
29668 @subheading A Bad Command
29670 Here's what happens if you pass a non-existent command:
29674 <- ^error,msg="Undefined MI command: rubbish"
29679 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29680 @node GDB/MI Command Description Format
29681 @section @sc{gdb/mi} Command Description Format
29683 The remaining sections describe blocks of commands. Each block of
29684 commands is laid out in a fashion similar to this section.
29686 @subheading Motivation
29688 The motivation for this collection of commands.
29690 @subheading Introduction
29692 A brief introduction to this collection of commands as a whole.
29694 @subheading Commands
29696 For each command in the block, the following is described:
29698 @subsubheading Synopsis
29701 -command @var{args}@dots{}
29704 @subsubheading Result
29706 @subsubheading @value{GDBN} Command
29708 The corresponding @value{GDBN} CLI command(s), if any.
29710 @subsubheading Example
29712 Example(s) formatted for readability. Some of the described commands have
29713 not been implemented yet and these are labeled N.A.@: (not available).
29716 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29717 @node GDB/MI Breakpoint Commands
29718 @section @sc{gdb/mi} Breakpoint Commands
29720 @cindex breakpoint commands for @sc{gdb/mi}
29721 @cindex @sc{gdb/mi}, breakpoint commands
29722 This section documents @sc{gdb/mi} commands for manipulating
29725 @subheading The @code{-break-after} Command
29726 @findex -break-after
29728 @subsubheading Synopsis
29731 -break-after @var{number} @var{count}
29734 The breakpoint number @var{number} is not in effect until it has been
29735 hit @var{count} times. To see how this is reflected in the output of
29736 the @samp{-break-list} command, see the description of the
29737 @samp{-break-list} command below.
29739 @subsubheading @value{GDBN} Command
29741 The corresponding @value{GDBN} command is @samp{ignore}.
29743 @subsubheading Example
29748 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29749 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29750 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29758 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29759 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29760 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29761 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29762 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29763 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29764 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29765 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29766 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29767 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
29772 @subheading The @code{-break-catch} Command
29773 @findex -break-catch
29776 @subheading The @code{-break-commands} Command
29777 @findex -break-commands
29779 @subsubheading Synopsis
29782 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
29785 Specifies the CLI commands that should be executed when breakpoint
29786 @var{number} is hit. The parameters @var{command1} to @var{commandN}
29787 are the commands. If no command is specified, any previously-set
29788 commands are cleared. @xref{Break Commands}. Typical use of this
29789 functionality is tracing a program, that is, printing of values of
29790 some variables whenever breakpoint is hit and then continuing.
29792 @subsubheading @value{GDBN} Command
29794 The corresponding @value{GDBN} command is @samp{commands}.
29796 @subsubheading Example
29801 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29802 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29803 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29806 -break-commands 1 "print v" "continue"
29811 @subheading The @code{-break-condition} Command
29812 @findex -break-condition
29814 @subsubheading Synopsis
29817 -break-condition @var{number} @var{expr}
29820 Breakpoint @var{number} will stop the program only if the condition in
29821 @var{expr} is true. The condition becomes part of the
29822 @samp{-break-list} output (see the description of the @samp{-break-list}
29825 @subsubheading @value{GDBN} Command
29827 The corresponding @value{GDBN} command is @samp{condition}.
29829 @subsubheading Example
29833 -break-condition 1 1
29837 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29838 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29839 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29840 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29841 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29842 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29843 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29844 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29845 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29846 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
29850 @subheading The @code{-break-delete} Command
29851 @findex -break-delete
29853 @subsubheading Synopsis
29856 -break-delete ( @var{breakpoint} )+
29859 Delete the breakpoint(s) whose number(s) are specified in the argument
29860 list. This is obviously reflected in the breakpoint list.
29862 @subsubheading @value{GDBN} Command
29864 The corresponding @value{GDBN} command is @samp{delete}.
29866 @subsubheading Example
29874 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29875 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29876 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29877 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29878 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29879 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29880 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29885 @subheading The @code{-break-disable} Command
29886 @findex -break-disable
29888 @subsubheading Synopsis
29891 -break-disable ( @var{breakpoint} )+
29894 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
29895 break list is now set to @samp{n} for the named @var{breakpoint}(s).
29897 @subsubheading @value{GDBN} Command
29899 The corresponding @value{GDBN} command is @samp{disable}.
29901 @subsubheading Example
29909 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29910 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29911 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29912 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29913 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29914 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29915 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29916 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
29917 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29918 line="5",thread-groups=["i1"],times="0"@}]@}
29922 @subheading The @code{-break-enable} Command
29923 @findex -break-enable
29925 @subsubheading Synopsis
29928 -break-enable ( @var{breakpoint} )+
29931 Enable (previously disabled) @var{breakpoint}(s).
29933 @subsubheading @value{GDBN} Command
29935 The corresponding @value{GDBN} command is @samp{enable}.
29937 @subsubheading Example
29945 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29946 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29947 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29948 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29949 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29950 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29951 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29952 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29953 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29954 line="5",thread-groups=["i1"],times="0"@}]@}
29958 @subheading The @code{-break-info} Command
29959 @findex -break-info
29961 @subsubheading Synopsis
29964 -break-info @var{breakpoint}
29968 Get information about a single breakpoint.
29970 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
29971 Information}, for details on the format of each breakpoint in the
29974 @subsubheading @value{GDBN} Command
29976 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
29978 @subsubheading Example
29981 @subheading The @code{-break-insert} Command
29982 @findex -break-insert
29984 @subsubheading Synopsis
29987 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
29988 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29989 [ -p @var{thread-id} ] [ @var{location} ]
29993 If specified, @var{location}, can be one of:
30000 @item filename:linenum
30001 @item filename:function
30005 The possible optional parameters of this command are:
30009 Insert a temporary breakpoint.
30011 Insert a hardware breakpoint.
30013 If @var{location} cannot be parsed (for example if it
30014 refers to unknown files or functions), create a pending
30015 breakpoint. Without this flag, @value{GDBN} will report
30016 an error, and won't create a breakpoint, if @var{location}
30019 Create a disabled breakpoint.
30021 Create a tracepoint. @xref{Tracepoints}. When this parameter
30022 is used together with @samp{-h}, a fast tracepoint is created.
30023 @item -c @var{condition}
30024 Make the breakpoint conditional on @var{condition}.
30025 @item -i @var{ignore-count}
30026 Initialize the @var{ignore-count}.
30027 @item -p @var{thread-id}
30028 Restrict the breakpoint to the specified @var{thread-id}.
30031 @subsubheading Result
30033 @xref{GDB/MI Breakpoint Information}, for details on the format of the
30034 resulting breakpoint.
30036 Note: this format is open to change.
30037 @c An out-of-band breakpoint instead of part of the result?
30039 @subsubheading @value{GDBN} Command
30041 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
30042 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
30044 @subsubheading Example
30049 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
30050 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
30053 -break-insert -t foo
30054 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
30055 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
30059 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
30060 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30061 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30062 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30063 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30064 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30065 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30066 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30067 addr="0x0001072c", func="main",file="recursive2.c",
30068 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
30070 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
30071 addr="0x00010774",func="foo",file="recursive2.c",
30072 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
30075 @c -break-insert -r foo.*
30076 @c ~int foo(int, int);
30077 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
30078 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
30083 @subheading The @code{-dprintf-insert} Command
30084 @findex -dprintf-insert
30086 @subsubheading Synopsis
30089 -dprintf-insert [ -t ] [ -f ] [ -d ]
30090 [ -c @var{condition} ] [ -i @var{ignore-count} ]
30091 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
30096 If specified, @var{location}, can be one of:
30099 @item @var{function}
30102 @c @item @var{linenum}
30103 @item @var{filename}:@var{linenum}
30104 @item @var{filename}:function
30105 @item *@var{address}
30108 The possible optional parameters of this command are:
30112 Insert a temporary breakpoint.
30114 If @var{location} cannot be parsed (for example, if it
30115 refers to unknown files or functions), create a pending
30116 breakpoint. Without this flag, @value{GDBN} will report
30117 an error, and won't create a breakpoint, if @var{location}
30120 Create a disabled breakpoint.
30121 @item -c @var{condition}
30122 Make the breakpoint conditional on @var{condition}.
30123 @item -i @var{ignore-count}
30124 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
30125 to @var{ignore-count}.
30126 @item -p @var{thread-id}
30127 Restrict the breakpoint to the specified @var{thread-id}.
30130 @subsubheading Result
30132 @xref{GDB/MI Breakpoint Information}, for details on the format of the
30133 resulting breakpoint.
30135 @c An out-of-band breakpoint instead of part of the result?
30137 @subsubheading @value{GDBN} Command
30139 The corresponding @value{GDBN} command is @samp{dprintf}.
30141 @subsubheading Example
30145 4-dprintf-insert foo "At foo entry\n"
30146 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
30147 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
30148 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
30149 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
30150 original-location="foo"@}
30152 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
30153 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
30154 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
30155 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
30156 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
30157 original-location="mi-dprintf.c:26"@}
30161 @subheading The @code{-break-list} Command
30162 @findex -break-list
30164 @subsubheading Synopsis
30170 Displays the list of inserted breakpoints, showing the following fields:
30174 number of the breakpoint
30176 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
30178 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
30181 is the breakpoint enabled or no: @samp{y} or @samp{n}
30183 memory location at which the breakpoint is set
30185 logical location of the breakpoint, expressed by function name, file
30187 @item Thread-groups
30188 list of thread groups to which this breakpoint applies
30190 number of times the breakpoint has been hit
30193 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
30194 @code{body} field is an empty list.
30196 @subsubheading @value{GDBN} Command
30198 The corresponding @value{GDBN} command is @samp{info break}.
30200 @subsubheading Example
30205 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
30206 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30207 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30208 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30209 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30210 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30211 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30212 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30213 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
30215 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
30216 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
30217 line="13",thread-groups=["i1"],times="0"@}]@}
30221 Here's an example of the result when there are no breakpoints:
30226 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
30227 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30228 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30229 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30230 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30231 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30232 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30237 @subheading The @code{-break-passcount} Command
30238 @findex -break-passcount
30240 @subsubheading Synopsis
30243 -break-passcount @var{tracepoint-number} @var{passcount}
30246 Set the passcount for tracepoint @var{tracepoint-number} to
30247 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
30248 is not a tracepoint, error is emitted. This corresponds to CLI
30249 command @samp{passcount}.
30251 @subheading The @code{-break-watch} Command
30252 @findex -break-watch
30254 @subsubheading Synopsis
30257 -break-watch [ -a | -r ]
30260 Create a watchpoint. With the @samp{-a} option it will create an
30261 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
30262 read from or on a write to the memory location. With the @samp{-r}
30263 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
30264 trigger only when the memory location is accessed for reading. Without
30265 either of the options, the watchpoint created is a regular watchpoint,
30266 i.e., it will trigger when the memory location is accessed for writing.
30267 @xref{Set Watchpoints, , Setting Watchpoints}.
30269 Note that @samp{-break-list} will report a single list of watchpoints and
30270 breakpoints inserted.
30272 @subsubheading @value{GDBN} Command
30274 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
30277 @subsubheading Example
30279 Setting a watchpoint on a variable in the @code{main} function:
30284 ^done,wpt=@{number="2",exp="x"@}
30289 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
30290 value=@{old="-268439212",new="55"@},
30291 frame=@{func="main",args=[],file="recursive2.c",
30292 fullname="/home/foo/bar/recursive2.c",line="5"@}
30296 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
30297 the program execution twice: first for the variable changing value, then
30298 for the watchpoint going out of scope.
30303 ^done,wpt=@{number="5",exp="C"@}
30308 *stopped,reason="watchpoint-trigger",
30309 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
30310 frame=@{func="callee4",args=[],
30311 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30312 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
30317 *stopped,reason="watchpoint-scope",wpnum="5",
30318 frame=@{func="callee3",args=[@{name="strarg",
30319 value="0x11940 \"A string argument.\""@}],
30320 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30321 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
30325 Listing breakpoints and watchpoints, at different points in the program
30326 execution. Note that once the watchpoint goes out of scope, it is
30332 ^done,wpt=@{number="2",exp="C"@}
30335 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
30336 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30337 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30338 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30339 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30340 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30341 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30342 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30343 addr="0x00010734",func="callee4",
30344 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30345 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
30347 bkpt=@{number="2",type="watchpoint",disp="keep",
30348 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
30353 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
30354 value=@{old="-276895068",new="3"@},
30355 frame=@{func="callee4",args=[],
30356 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30357 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
30360 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
30361 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30362 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30363 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30364 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30365 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30366 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30367 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30368 addr="0x00010734",func="callee4",
30369 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30370 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
30372 bkpt=@{number="2",type="watchpoint",disp="keep",
30373 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
30377 ^done,reason="watchpoint-scope",wpnum="2",
30378 frame=@{func="callee3",args=[@{name="strarg",
30379 value="0x11940 \"A string argument.\""@}],
30380 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30381 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
30384 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
30385 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30386 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30387 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30388 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30389 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30390 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30391 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30392 addr="0x00010734",func="callee4",
30393 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30394 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30395 thread-groups=["i1"],times="1"@}]@}
30400 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30401 @node GDB/MI Catchpoint Commands
30402 @section @sc{gdb/mi} Catchpoint Commands
30404 This section documents @sc{gdb/mi} commands for manipulating
30408 * Shared Library GDB/MI Catchpoint Commands::
30409 * Ada Exception GDB/MI Catchpoint Commands::
30412 @node Shared Library GDB/MI Catchpoint Commands
30413 @subsection Shared Library @sc{gdb/mi} Catchpoints
30415 @subheading The @code{-catch-load} Command
30416 @findex -catch-load
30418 @subsubheading Synopsis
30421 -catch-load [ -t ] [ -d ] @var{regexp}
30424 Add a catchpoint for library load events. If the @samp{-t} option is used,
30425 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
30426 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
30427 in a disabled state. The @samp{regexp} argument is a regular
30428 expression used to match the name of the loaded library.
30431 @subsubheading @value{GDBN} Command
30433 The corresponding @value{GDBN} command is @samp{catch load}.
30435 @subsubheading Example
30438 -catch-load -t foo.so
30439 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
30440 what="load of library matching foo.so",catch-type="load",times="0"@}
30445 @subheading The @code{-catch-unload} Command
30446 @findex -catch-unload
30448 @subsubheading Synopsis
30451 -catch-unload [ -t ] [ -d ] @var{regexp}
30454 Add a catchpoint for library unload events. If the @samp{-t} option is
30455 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
30456 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
30457 created in a disabled state. The @samp{regexp} argument is a regular
30458 expression used to match the name of the unloaded library.
30460 @subsubheading @value{GDBN} Command
30462 The corresponding @value{GDBN} command is @samp{catch unload}.
30464 @subsubheading Example
30467 -catch-unload -d bar.so
30468 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
30469 what="load of library matching bar.so",catch-type="unload",times="0"@}
30473 @node Ada Exception GDB/MI Catchpoint Commands
30474 @subsection Ada Exception @sc{gdb/mi} Catchpoints
30476 The following @sc{gdb/mi} commands can be used to create catchpoints
30477 that stop the execution when Ada exceptions are being raised.
30479 @subheading The @code{-catch-assert} Command
30480 @findex -catch-assert
30482 @subsubheading Synopsis
30485 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
30488 Add a catchpoint for failed Ada assertions.
30490 The possible optional parameters for this command are:
30493 @item -c @var{condition}
30494 Make the catchpoint conditional on @var{condition}.
30496 Create a disabled catchpoint.
30498 Create a temporary catchpoint.
30501 @subsubheading @value{GDBN} Command
30503 The corresponding @value{GDBN} command is @samp{catch assert}.
30505 @subsubheading Example
30509 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
30510 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
30511 thread-groups=["i1"],times="0",
30512 original-location="__gnat_debug_raise_assert_failure"@}
30516 @subheading The @code{-catch-exception} Command
30517 @findex -catch-exception
30519 @subsubheading Synopsis
30522 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
30526 Add a catchpoint stopping when Ada exceptions are raised.
30527 By default, the command stops the program when any Ada exception
30528 gets raised. But it is also possible, by using some of the
30529 optional parameters described below, to create more selective
30532 The possible optional parameters for this command are:
30535 @item -c @var{condition}
30536 Make the catchpoint conditional on @var{condition}.
30538 Create a disabled catchpoint.
30539 @item -e @var{exception-name}
30540 Only stop when @var{exception-name} is raised. This option cannot
30541 be used combined with @samp{-u}.
30543 Create a temporary catchpoint.
30545 Stop only when an unhandled exception gets raised. This option
30546 cannot be used combined with @samp{-e}.
30549 @subsubheading @value{GDBN} Command
30551 The corresponding @value{GDBN} commands are @samp{catch exception}
30552 and @samp{catch exception unhandled}.
30554 @subsubheading Example
30557 -catch-exception -e Program_Error
30558 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
30559 enabled="y",addr="0x0000000000404874",
30560 what="`Program_Error' Ada exception", thread-groups=["i1"],
30561 times="0",original-location="__gnat_debug_raise_exception"@}
30565 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30566 @node GDB/MI Program Context
30567 @section @sc{gdb/mi} Program Context
30569 @subheading The @code{-exec-arguments} Command
30570 @findex -exec-arguments
30573 @subsubheading Synopsis
30576 -exec-arguments @var{args}
30579 Set the inferior program arguments, to be used in the next
30582 @subsubheading @value{GDBN} Command
30584 The corresponding @value{GDBN} command is @samp{set args}.
30586 @subsubheading Example
30590 -exec-arguments -v word
30597 @subheading The @code{-exec-show-arguments} Command
30598 @findex -exec-show-arguments
30600 @subsubheading Synopsis
30603 -exec-show-arguments
30606 Print the arguments of the program.
30608 @subsubheading @value{GDBN} Command
30610 The corresponding @value{GDBN} command is @samp{show args}.
30612 @subsubheading Example
30617 @subheading The @code{-environment-cd} Command
30618 @findex -environment-cd
30620 @subsubheading Synopsis
30623 -environment-cd @var{pathdir}
30626 Set @value{GDBN}'s working directory.
30628 @subsubheading @value{GDBN} Command
30630 The corresponding @value{GDBN} command is @samp{cd}.
30632 @subsubheading Example
30636 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30642 @subheading The @code{-environment-directory} Command
30643 @findex -environment-directory
30645 @subsubheading Synopsis
30648 -environment-directory [ -r ] [ @var{pathdir} ]+
30651 Add directories @var{pathdir} to beginning of search path for source files.
30652 If the @samp{-r} option is used, the search path is reset to the default
30653 search path. If directories @var{pathdir} are supplied in addition to the
30654 @samp{-r} option, the search path is first reset and then addition
30656 Multiple directories may be specified, separated by blanks. Specifying
30657 multiple directories in a single command
30658 results in the directories added to the beginning of the
30659 search path in the same order they were presented in the command.
30660 If blanks are needed as
30661 part of a directory name, double-quotes should be used around
30662 the name. In the command output, the path will show up separated
30663 by the system directory-separator character. The directory-separator
30664 character must not be used
30665 in any directory name.
30666 If no directories are specified, the current search path is displayed.
30668 @subsubheading @value{GDBN} Command
30670 The corresponding @value{GDBN} command is @samp{dir}.
30672 @subsubheading Example
30676 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30677 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30679 -environment-directory ""
30680 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30682 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
30683 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
30685 -environment-directory -r
30686 ^done,source-path="$cdir:$cwd"
30691 @subheading The @code{-environment-path} Command
30692 @findex -environment-path
30694 @subsubheading Synopsis
30697 -environment-path [ -r ] [ @var{pathdir} ]+
30700 Add directories @var{pathdir} to beginning of search path for object files.
30701 If the @samp{-r} option is used, the search path is reset to the original
30702 search path that existed at gdb start-up. If directories @var{pathdir} are
30703 supplied in addition to the
30704 @samp{-r} option, the search path is first reset and then addition
30706 Multiple directories may be specified, separated by blanks. Specifying
30707 multiple directories in a single command
30708 results in the directories added to the beginning of the
30709 search path in the same order they were presented in the command.
30710 If blanks are needed as
30711 part of a directory name, double-quotes should be used around
30712 the name. In the command output, the path will show up separated
30713 by the system directory-separator character. The directory-separator
30714 character must not be used
30715 in any directory name.
30716 If no directories are specified, the current path is displayed.
30719 @subsubheading @value{GDBN} Command
30721 The corresponding @value{GDBN} command is @samp{path}.
30723 @subsubheading Example
30728 ^done,path="/usr/bin"
30730 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
30731 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
30733 -environment-path -r /usr/local/bin
30734 ^done,path="/usr/local/bin:/usr/bin"
30739 @subheading The @code{-environment-pwd} Command
30740 @findex -environment-pwd
30742 @subsubheading Synopsis
30748 Show the current working directory.
30750 @subsubheading @value{GDBN} Command
30752 The corresponding @value{GDBN} command is @samp{pwd}.
30754 @subsubheading Example
30759 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
30763 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30764 @node GDB/MI Thread Commands
30765 @section @sc{gdb/mi} Thread Commands
30768 @subheading The @code{-thread-info} Command
30769 @findex -thread-info
30771 @subsubheading Synopsis
30774 -thread-info [ @var{thread-id} ]
30777 Reports information about either a specific thread, if
30778 the @var{thread-id} parameter is present, or about all
30779 threads. When printing information about all threads,
30780 also reports the current thread.
30782 @subsubheading @value{GDBN} Command
30784 The @samp{info thread} command prints the same information
30787 @subsubheading Result
30789 The result is a list of threads. The following attributes are
30790 defined for a given thread:
30794 This field exists only for the current thread. It has the value @samp{*}.
30797 The identifier that @value{GDBN} uses to refer to the thread.
30800 The identifier that the target uses to refer to the thread.
30803 Extra information about the thread, in a target-specific format. This
30807 The name of the thread. If the user specified a name using the
30808 @code{thread name} command, then this name is given. Otherwise, if
30809 @value{GDBN} can extract the thread name from the target, then that
30810 name is given. If @value{GDBN} cannot find the thread name, then this
30814 The stack frame currently executing in the thread.
30817 The thread's state. The @samp{state} field may have the following
30822 The thread is stopped. Frame information is available for stopped
30826 The thread is running. There's no frame information for running
30832 If @value{GDBN} can find the CPU core on which this thread is running,
30833 then this field is the core identifier. This field is optional.
30837 @subsubheading Example
30842 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
30843 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
30844 args=[]@},state="running"@},
30845 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
30846 frame=@{level="0",addr="0x0804891f",func="foo",
30847 args=[@{name="i",value="10"@}],
30848 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
30849 state="running"@}],
30850 current-thread-id="1"
30854 @subheading The @code{-thread-list-ids} Command
30855 @findex -thread-list-ids
30857 @subsubheading Synopsis
30863 Produces a list of the currently known @value{GDBN} thread ids. At the
30864 end of the list it also prints the total number of such threads.
30866 This command is retained for historical reasons, the
30867 @code{-thread-info} command should be used instead.
30869 @subsubheading @value{GDBN} Command
30871 Part of @samp{info threads} supplies the same information.
30873 @subsubheading Example
30878 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30879 current-thread-id="1",number-of-threads="3"
30884 @subheading The @code{-thread-select} Command
30885 @findex -thread-select
30887 @subsubheading Synopsis
30890 -thread-select @var{threadnum}
30893 Make @var{threadnum} the current thread. It prints the number of the new
30894 current thread, and the topmost frame for that thread.
30896 This command is deprecated in favor of explicitly using the
30897 @samp{--thread} option to each command.
30899 @subsubheading @value{GDBN} Command
30901 The corresponding @value{GDBN} command is @samp{thread}.
30903 @subsubheading Example
30910 *stopped,reason="end-stepping-range",thread-id="2",line="187",
30911 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
30915 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30916 number-of-threads="3"
30919 ^done,new-thread-id="3",
30920 frame=@{level="0",func="vprintf",
30921 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
30922 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
30926 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30927 @node GDB/MI Ada Tasking Commands
30928 @section @sc{gdb/mi} Ada Tasking Commands
30930 @subheading The @code{-ada-task-info} Command
30931 @findex -ada-task-info
30933 @subsubheading Synopsis
30936 -ada-task-info [ @var{task-id} ]
30939 Reports information about either a specific Ada task, if the
30940 @var{task-id} parameter is present, or about all Ada tasks.
30942 @subsubheading @value{GDBN} Command
30944 The @samp{info tasks} command prints the same information
30945 about all Ada tasks (@pxref{Ada Tasks}).
30947 @subsubheading Result
30949 The result is a table of Ada tasks. The following columns are
30950 defined for each Ada task:
30954 This field exists only for the current thread. It has the value @samp{*}.
30957 The identifier that @value{GDBN} uses to refer to the Ada task.
30960 The identifier that the target uses to refer to the Ada task.
30963 The identifier of the thread corresponding to the Ada task.
30965 This field should always exist, as Ada tasks are always implemented
30966 on top of a thread. But if @value{GDBN} cannot find this corresponding
30967 thread for any reason, the field is omitted.
30970 This field exists only when the task was created by another task.
30971 In this case, it provides the ID of the parent task.
30974 The base priority of the task.
30977 The current state of the task. For a detailed description of the
30978 possible states, see @ref{Ada Tasks}.
30981 The name of the task.
30985 @subsubheading Example
30989 ^done,tasks=@{nr_rows="3",nr_cols="8",
30990 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
30991 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
30992 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
30993 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
30994 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
30995 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
30996 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
30997 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
30998 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
30999 state="Child Termination Wait",name="main_task"@}]@}
31003 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31004 @node GDB/MI Program Execution
31005 @section @sc{gdb/mi} Program Execution
31007 These are the asynchronous commands which generate the out-of-band
31008 record @samp{*stopped}. Currently @value{GDBN} only really executes
31009 asynchronously with remote targets and this interaction is mimicked in
31012 @subheading The @code{-exec-continue} Command
31013 @findex -exec-continue
31015 @subsubheading Synopsis
31018 -exec-continue [--reverse] [--all|--thread-group N]
31021 Resumes the execution of the inferior program, which will continue
31022 to execute until it reaches a debugger stop event. If the
31023 @samp{--reverse} option is specified, execution resumes in reverse until
31024 it reaches a stop event. Stop events may include
31027 breakpoints or watchpoints
31029 signals or exceptions
31031 the end of the process (or its beginning under @samp{--reverse})
31033 the end or beginning of a replay log if one is being used.
31035 In all-stop mode (@pxref{All-Stop
31036 Mode}), may resume only one thread, or all threads, depending on the
31037 value of the @samp{scheduler-locking} variable. If @samp{--all} is
31038 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
31039 ignored in all-stop mode. If the @samp{--thread-group} options is
31040 specified, then all threads in that thread group are resumed.
31042 @subsubheading @value{GDBN} Command
31044 The corresponding @value{GDBN} corresponding is @samp{continue}.
31046 @subsubheading Example
31053 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
31054 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
31060 @subheading The @code{-exec-finish} Command
31061 @findex -exec-finish
31063 @subsubheading Synopsis
31066 -exec-finish [--reverse]
31069 Resumes the execution of the inferior program until the current
31070 function is exited. Displays the results returned by the function.
31071 If the @samp{--reverse} option is specified, resumes the reverse
31072 execution of the inferior program until the point where current
31073 function was called.
31075 @subsubheading @value{GDBN} Command
31077 The corresponding @value{GDBN} command is @samp{finish}.
31079 @subsubheading Example
31081 Function returning @code{void}.
31088 *stopped,reason="function-finished",frame=@{func="main",args=[],
31089 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
31093 Function returning other than @code{void}. The name of the internal
31094 @value{GDBN} variable storing the result is printed, together with the
31101 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
31102 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
31103 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31104 gdb-result-var="$1",return-value="0"
31109 @subheading The @code{-exec-interrupt} Command
31110 @findex -exec-interrupt
31112 @subsubheading Synopsis
31115 -exec-interrupt [--all|--thread-group N]
31118 Interrupts the background execution of the target. Note how the token
31119 associated with the stop message is the one for the execution command
31120 that has been interrupted. The token for the interrupt itself only
31121 appears in the @samp{^done} output. If the user is trying to
31122 interrupt a non-running program, an error message will be printed.
31124 Note that when asynchronous execution is enabled, this command is
31125 asynchronous just like other execution commands. That is, first the
31126 @samp{^done} response will be printed, and the target stop will be
31127 reported after that using the @samp{*stopped} notification.
31129 In non-stop mode, only the context thread is interrupted by default.
31130 All threads (in all inferiors) will be interrupted if the
31131 @samp{--all} option is specified. If the @samp{--thread-group}
31132 option is specified, all threads in that group will be interrupted.
31134 @subsubheading @value{GDBN} Command
31136 The corresponding @value{GDBN} command is @samp{interrupt}.
31138 @subsubheading Example
31149 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
31150 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
31151 fullname="/home/foo/bar/try.c",line="13"@}
31156 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
31160 @subheading The @code{-exec-jump} Command
31163 @subsubheading Synopsis
31166 -exec-jump @var{location}
31169 Resumes execution of the inferior program at the location specified by
31170 parameter. @xref{Specify Location}, for a description of the
31171 different forms of @var{location}.
31173 @subsubheading @value{GDBN} Command
31175 The corresponding @value{GDBN} command is @samp{jump}.
31177 @subsubheading Example
31180 -exec-jump foo.c:10
31181 *running,thread-id="all"
31186 @subheading The @code{-exec-next} Command
31189 @subsubheading Synopsis
31192 -exec-next [--reverse]
31195 Resumes execution of the inferior program, stopping when the beginning
31196 of the next source line is reached.
31198 If the @samp{--reverse} option is specified, resumes reverse execution
31199 of the inferior program, stopping at the beginning of the previous
31200 source line. If you issue this command on the first line of a
31201 function, it will take you back to the caller of that function, to the
31202 source line where the function was called.
31205 @subsubheading @value{GDBN} Command
31207 The corresponding @value{GDBN} command is @samp{next}.
31209 @subsubheading Example
31215 *stopped,reason="end-stepping-range",line="8",file="hello.c"
31220 @subheading The @code{-exec-next-instruction} Command
31221 @findex -exec-next-instruction
31223 @subsubheading Synopsis
31226 -exec-next-instruction [--reverse]
31229 Executes one machine instruction. If the instruction is a function
31230 call, continues until the function returns. If the program stops at an
31231 instruction in the middle of a source line, the address will be
31234 If the @samp{--reverse} option is specified, resumes reverse execution
31235 of the inferior program, stopping at the previous instruction. If the
31236 previously executed instruction was a return from another function,
31237 it will continue to execute in reverse until the call to that function
31238 (from the current stack frame) is reached.
31240 @subsubheading @value{GDBN} Command
31242 The corresponding @value{GDBN} command is @samp{nexti}.
31244 @subsubheading Example
31248 -exec-next-instruction
31252 *stopped,reason="end-stepping-range",
31253 addr="0x000100d4",line="5",file="hello.c"
31258 @subheading The @code{-exec-return} Command
31259 @findex -exec-return
31261 @subsubheading Synopsis
31267 Makes current function return immediately. Doesn't execute the inferior.
31268 Displays the new current frame.
31270 @subsubheading @value{GDBN} Command
31272 The corresponding @value{GDBN} command is @samp{return}.
31274 @subsubheading Example
31278 200-break-insert callee4
31279 200^done,bkpt=@{number="1",addr="0x00010734",
31280 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
31285 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
31286 frame=@{func="callee4",args=[],
31287 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31288 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
31294 111^done,frame=@{level="0",func="callee3",
31295 args=[@{name="strarg",
31296 value="0x11940 \"A string argument.\""@}],
31297 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31298 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
31303 @subheading The @code{-exec-run} Command
31306 @subsubheading Synopsis
31309 -exec-run [ --all | --thread-group N ] [ --start ]
31312 Starts execution of the inferior from the beginning. The inferior
31313 executes until either a breakpoint is encountered or the program
31314 exits. In the latter case the output will include an exit code, if
31315 the program has exited exceptionally.
31317 When neither the @samp{--all} nor the @samp{--thread-group} option
31318 is specified, the current inferior is started. If the
31319 @samp{--thread-group} option is specified, it should refer to a thread
31320 group of type @samp{process}, and that thread group will be started.
31321 If the @samp{--all} option is specified, then all inferiors will be started.
31323 Using the @samp{--start} option instructs the debugger to stop
31324 the execution at the start of the inferior's main subprogram,
31325 following the same behavior as the @code{start} command
31326 (@pxref{Starting}).
31328 @subsubheading @value{GDBN} Command
31330 The corresponding @value{GDBN} command is @samp{run}.
31332 @subsubheading Examples
31337 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
31342 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
31343 frame=@{func="main",args=[],file="recursive2.c",
31344 fullname="/home/foo/bar/recursive2.c",line="4"@}
31349 Program exited normally:
31357 *stopped,reason="exited-normally"
31362 Program exited exceptionally:
31370 *stopped,reason="exited",exit-code="01"
31374 Another way the program can terminate is if it receives a signal such as
31375 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
31379 *stopped,reason="exited-signalled",signal-name="SIGINT",
31380 signal-meaning="Interrupt"
31384 @c @subheading -exec-signal
31387 @subheading The @code{-exec-step} Command
31390 @subsubheading Synopsis
31393 -exec-step [--reverse]
31396 Resumes execution of the inferior program, stopping when the beginning
31397 of the next source line is reached, if the next source line is not a
31398 function call. If it is, stop at the first instruction of the called
31399 function. If the @samp{--reverse} option is specified, resumes reverse
31400 execution of the inferior program, stopping at the beginning of the
31401 previously executed source line.
31403 @subsubheading @value{GDBN} Command
31405 The corresponding @value{GDBN} command is @samp{step}.
31407 @subsubheading Example
31409 Stepping into a function:
31415 *stopped,reason="end-stepping-range",
31416 frame=@{func="foo",args=[@{name="a",value="10"@},
31417 @{name="b",value="0"@}],file="recursive2.c",
31418 fullname="/home/foo/bar/recursive2.c",line="11"@}
31428 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
31433 @subheading The @code{-exec-step-instruction} Command
31434 @findex -exec-step-instruction
31436 @subsubheading Synopsis
31439 -exec-step-instruction [--reverse]
31442 Resumes the inferior which executes one machine instruction. If the
31443 @samp{--reverse} option is specified, resumes reverse execution of the
31444 inferior program, stopping at the previously executed instruction.
31445 The output, once @value{GDBN} has stopped, will vary depending on
31446 whether we have stopped in the middle of a source line or not. In the
31447 former case, the address at which the program stopped will be printed
31450 @subsubheading @value{GDBN} Command
31452 The corresponding @value{GDBN} command is @samp{stepi}.
31454 @subsubheading Example
31458 -exec-step-instruction
31462 *stopped,reason="end-stepping-range",
31463 frame=@{func="foo",args=[],file="try.c",
31464 fullname="/home/foo/bar/try.c",line="10"@}
31466 -exec-step-instruction
31470 *stopped,reason="end-stepping-range",
31471 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
31472 fullname="/home/foo/bar/try.c",line="10"@}
31477 @subheading The @code{-exec-until} Command
31478 @findex -exec-until
31480 @subsubheading Synopsis
31483 -exec-until [ @var{location} ]
31486 Executes the inferior until the @var{location} specified in the
31487 argument is reached. If there is no argument, the inferior executes
31488 until a source line greater than the current one is reached. The
31489 reason for stopping in this case will be @samp{location-reached}.
31491 @subsubheading @value{GDBN} Command
31493 The corresponding @value{GDBN} command is @samp{until}.
31495 @subsubheading Example
31499 -exec-until recursive2.c:6
31503 *stopped,reason="location-reached",frame=@{func="main",args=[],
31504 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
31509 @subheading -file-clear
31510 Is this going away????
31513 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31514 @node GDB/MI Stack Manipulation
31515 @section @sc{gdb/mi} Stack Manipulation Commands
31517 @subheading The @code{-enable-frame-filters} Command
31518 @findex -enable-frame-filters
31521 -enable-frame-filters
31524 @value{GDBN} allows Python-based frame filters to affect the output of
31525 the MI commands relating to stack traces. As there is no way to
31526 implement this in a fully backward-compatible way, a front end must
31527 request that this functionality be enabled.
31529 Once enabled, this feature cannot be disabled.
31531 Note that if Python support has not been compiled into @value{GDBN},
31532 this command will still succeed (and do nothing).
31534 @subheading The @code{-stack-info-frame} Command
31535 @findex -stack-info-frame
31537 @subsubheading Synopsis
31543 Get info on the selected frame.
31545 @subsubheading @value{GDBN} Command
31547 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
31548 (without arguments).
31550 @subsubheading Example
31555 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
31556 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31557 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
31561 @subheading The @code{-stack-info-depth} Command
31562 @findex -stack-info-depth
31564 @subsubheading Synopsis
31567 -stack-info-depth [ @var{max-depth} ]
31570 Return the depth of the stack. If the integer argument @var{max-depth}
31571 is specified, do not count beyond @var{max-depth} frames.
31573 @subsubheading @value{GDBN} Command
31575 There's no equivalent @value{GDBN} command.
31577 @subsubheading Example
31579 For a stack with frame levels 0 through 11:
31586 -stack-info-depth 4
31589 -stack-info-depth 12
31592 -stack-info-depth 11
31595 -stack-info-depth 13
31600 @anchor{-stack-list-arguments}
31601 @subheading The @code{-stack-list-arguments} Command
31602 @findex -stack-list-arguments
31604 @subsubheading Synopsis
31607 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31608 [ @var{low-frame} @var{high-frame} ]
31611 Display a list of the arguments for the frames between @var{low-frame}
31612 and @var{high-frame} (inclusive). If @var{low-frame} and
31613 @var{high-frame} are not provided, list the arguments for the whole
31614 call stack. If the two arguments are equal, show the single frame
31615 at the corresponding level. It is an error if @var{low-frame} is
31616 larger than the actual number of frames. On the other hand,
31617 @var{high-frame} may be larger than the actual number of frames, in
31618 which case only existing frames will be returned.
31620 If @var{print-values} is 0 or @code{--no-values}, print only the names of
31621 the variables; if it is 1 or @code{--all-values}, print also their
31622 values; and if it is 2 or @code{--simple-values}, print the name,
31623 type and value for simple data types, and the name and type for arrays,
31624 structures and unions. If the option @code{--no-frame-filters} is
31625 supplied, then Python frame filters will not be executed.
31627 If the @code{--skip-unavailable} option is specified, arguments that
31628 are not available are not listed. Partially available arguments
31629 are still displayed, however.
31631 Use of this command to obtain arguments in a single frame is
31632 deprecated in favor of the @samp{-stack-list-variables} command.
31634 @subsubheading @value{GDBN} Command
31636 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
31637 @samp{gdb_get_args} command which partially overlaps with the
31638 functionality of @samp{-stack-list-arguments}.
31640 @subsubheading Example
31647 frame=@{level="0",addr="0x00010734",func="callee4",
31648 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31649 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
31650 frame=@{level="1",addr="0x0001076c",func="callee3",
31651 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31652 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
31653 frame=@{level="2",addr="0x0001078c",func="callee2",
31654 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31655 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
31656 frame=@{level="3",addr="0x000107b4",func="callee1",
31657 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31658 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
31659 frame=@{level="4",addr="0x000107e0",func="main",
31660 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31661 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
31663 -stack-list-arguments 0
31666 frame=@{level="0",args=[]@},
31667 frame=@{level="1",args=[name="strarg"]@},
31668 frame=@{level="2",args=[name="intarg",name="strarg"]@},
31669 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
31670 frame=@{level="4",args=[]@}]
31672 -stack-list-arguments 1
31675 frame=@{level="0",args=[]@},
31677 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31678 frame=@{level="2",args=[
31679 @{name="intarg",value="2"@},
31680 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31681 @{frame=@{level="3",args=[
31682 @{name="intarg",value="2"@},
31683 @{name="strarg",value="0x11940 \"A string argument.\""@},
31684 @{name="fltarg",value="3.5"@}]@},
31685 frame=@{level="4",args=[]@}]
31687 -stack-list-arguments 0 2 2
31688 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
31690 -stack-list-arguments 1 2 2
31691 ^done,stack-args=[frame=@{level="2",
31692 args=[@{name="intarg",value="2"@},
31693 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
31697 @c @subheading -stack-list-exception-handlers
31700 @anchor{-stack-list-frames}
31701 @subheading The @code{-stack-list-frames} Command
31702 @findex -stack-list-frames
31704 @subsubheading Synopsis
31707 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
31710 List the frames currently on the stack. For each frame it displays the
31715 The frame number, 0 being the topmost frame, i.e., the innermost function.
31717 The @code{$pc} value for that frame.
31721 File name of the source file where the function lives.
31722 @item @var{fullname}
31723 The full file name of the source file where the function lives.
31725 Line number corresponding to the @code{$pc}.
31727 The shared library where this function is defined. This is only given
31728 if the frame's function is not known.
31731 If invoked without arguments, this command prints a backtrace for the
31732 whole stack. If given two integer arguments, it shows the frames whose
31733 levels are between the two arguments (inclusive). If the two arguments
31734 are equal, it shows the single frame at the corresponding level. It is
31735 an error if @var{low-frame} is larger than the actual number of
31736 frames. On the other hand, @var{high-frame} may be larger than the
31737 actual number of frames, in which case only existing frames will be
31738 returned. If the option @code{--no-frame-filters} is supplied, then
31739 Python frame filters will not be executed.
31741 @subsubheading @value{GDBN} Command
31743 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
31745 @subsubheading Example
31747 Full stack backtrace:
31753 [frame=@{level="0",addr="0x0001076c",func="foo",
31754 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
31755 frame=@{level="1",addr="0x000107a4",func="foo",
31756 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31757 frame=@{level="2",addr="0x000107a4",func="foo",
31758 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31759 frame=@{level="3",addr="0x000107a4",func="foo",
31760 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31761 frame=@{level="4",addr="0x000107a4",func="foo",
31762 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31763 frame=@{level="5",addr="0x000107a4",func="foo",
31764 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31765 frame=@{level="6",addr="0x000107a4",func="foo",
31766 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31767 frame=@{level="7",addr="0x000107a4",func="foo",
31768 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31769 frame=@{level="8",addr="0x000107a4",func="foo",
31770 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31771 frame=@{level="9",addr="0x000107a4",func="foo",
31772 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31773 frame=@{level="10",addr="0x000107a4",func="foo",
31774 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31775 frame=@{level="11",addr="0x00010738",func="main",
31776 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
31780 Show frames between @var{low_frame} and @var{high_frame}:
31784 -stack-list-frames 3 5
31786 [frame=@{level="3",addr="0x000107a4",func="foo",
31787 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31788 frame=@{level="4",addr="0x000107a4",func="foo",
31789 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31790 frame=@{level="5",addr="0x000107a4",func="foo",
31791 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
31795 Show a single frame:
31799 -stack-list-frames 3 3
31801 [frame=@{level="3",addr="0x000107a4",func="foo",
31802 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
31807 @subheading The @code{-stack-list-locals} Command
31808 @findex -stack-list-locals
31809 @anchor{-stack-list-locals}
31811 @subsubheading Synopsis
31814 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31817 Display the local variable names for the selected frame. If
31818 @var{print-values} is 0 or @code{--no-values}, print only the names of
31819 the variables; if it is 1 or @code{--all-values}, print also their
31820 values; and if it is 2 or @code{--simple-values}, print the name,
31821 type and value for simple data types, and the name and type for arrays,
31822 structures and unions. In this last case, a frontend can immediately
31823 display the value of simple data types and create variable objects for
31824 other data types when the user wishes to explore their values in
31825 more detail. If the option @code{--no-frame-filters} is supplied, then
31826 Python frame filters will not be executed.
31828 If the @code{--skip-unavailable} option is specified, local variables
31829 that are not available are not listed. Partially available local
31830 variables are still displayed, however.
31832 This command is deprecated in favor of the
31833 @samp{-stack-list-variables} command.
31835 @subsubheading @value{GDBN} Command
31837 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
31839 @subsubheading Example
31843 -stack-list-locals 0
31844 ^done,locals=[name="A",name="B",name="C"]
31846 -stack-list-locals --all-values
31847 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
31848 @{name="C",value="@{1, 2, 3@}"@}]
31849 -stack-list-locals --simple-values
31850 ^done,locals=[@{name="A",type="int",value="1"@},
31851 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
31855 @anchor{-stack-list-variables}
31856 @subheading The @code{-stack-list-variables} Command
31857 @findex -stack-list-variables
31859 @subsubheading Synopsis
31862 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31865 Display the names of local variables and function arguments for the selected frame. If
31866 @var{print-values} is 0 or @code{--no-values}, print only the names of
31867 the variables; if it is 1 or @code{--all-values}, print also their
31868 values; and if it is 2 or @code{--simple-values}, print the name,
31869 type and value for simple data types, and the name and type for arrays,
31870 structures and unions. If the option @code{--no-frame-filters} is
31871 supplied, then Python frame filters will not be executed.
31873 If the @code{--skip-unavailable} option is specified, local variables
31874 and arguments that are not available are not listed. Partially
31875 available arguments and local variables are still displayed, however.
31877 @subsubheading Example
31881 -stack-list-variables --thread 1 --frame 0 --all-values
31882 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
31887 @subheading The @code{-stack-select-frame} Command
31888 @findex -stack-select-frame
31890 @subsubheading Synopsis
31893 -stack-select-frame @var{framenum}
31896 Change the selected frame. Select a different frame @var{framenum} on
31899 This command in deprecated in favor of passing the @samp{--frame}
31900 option to every command.
31902 @subsubheading @value{GDBN} Command
31904 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
31905 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
31907 @subsubheading Example
31911 -stack-select-frame 2
31916 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31917 @node GDB/MI Variable Objects
31918 @section @sc{gdb/mi} Variable Objects
31922 @subheading Motivation for Variable Objects in @sc{gdb/mi}
31924 For the implementation of a variable debugger window (locals, watched
31925 expressions, etc.), we are proposing the adaptation of the existing code
31926 used by @code{Insight}.
31928 The two main reasons for that are:
31932 It has been proven in practice (it is already on its second generation).
31935 It will shorten development time (needless to say how important it is
31939 The original interface was designed to be used by Tcl code, so it was
31940 slightly changed so it could be used through @sc{gdb/mi}. This section
31941 describes the @sc{gdb/mi} operations that will be available and gives some
31942 hints about their use.
31944 @emph{Note}: In addition to the set of operations described here, we
31945 expect the @sc{gui} implementation of a variable window to require, at
31946 least, the following operations:
31949 @item @code{-gdb-show} @code{output-radix}
31950 @item @code{-stack-list-arguments}
31951 @item @code{-stack-list-locals}
31952 @item @code{-stack-select-frame}
31957 @subheading Introduction to Variable Objects
31959 @cindex variable objects in @sc{gdb/mi}
31961 Variable objects are "object-oriented" MI interface for examining and
31962 changing values of expressions. Unlike some other MI interfaces that
31963 work with expressions, variable objects are specifically designed for
31964 simple and efficient presentation in the frontend. A variable object
31965 is identified by string name. When a variable object is created, the
31966 frontend specifies the expression for that variable object. The
31967 expression can be a simple variable, or it can be an arbitrary complex
31968 expression, and can even involve CPU registers. After creating a
31969 variable object, the frontend can invoke other variable object
31970 operations---for example to obtain or change the value of a variable
31971 object, or to change display format.
31973 Variable objects have hierarchical tree structure. Any variable object
31974 that corresponds to a composite type, such as structure in C, has
31975 a number of child variable objects, for example corresponding to each
31976 element of a structure. A child variable object can itself have
31977 children, recursively. Recursion ends when we reach
31978 leaf variable objects, which always have built-in types. Child variable
31979 objects are created only by explicit request, so if a frontend
31980 is not interested in the children of a particular variable object, no
31981 child will be created.
31983 For a leaf variable object it is possible to obtain its value as a
31984 string, or set the value from a string. String value can be also
31985 obtained for a non-leaf variable object, but it's generally a string
31986 that only indicates the type of the object, and does not list its
31987 contents. Assignment to a non-leaf variable object is not allowed.
31989 A frontend does not need to read the values of all variable objects each time
31990 the program stops. Instead, MI provides an update command that lists all
31991 variable objects whose values has changed since the last update
31992 operation. This considerably reduces the amount of data that must
31993 be transferred to the frontend. As noted above, children variable
31994 objects are created on demand, and only leaf variable objects have a
31995 real value. As result, gdb will read target memory only for leaf
31996 variables that frontend has created.
31998 The automatic update is not always desirable. For example, a frontend
31999 might want to keep a value of some expression for future reference,
32000 and never update it. For another example, fetching memory is
32001 relatively slow for embedded targets, so a frontend might want
32002 to disable automatic update for the variables that are either not
32003 visible on the screen, or ``closed''. This is possible using so
32004 called ``frozen variable objects''. Such variable objects are never
32005 implicitly updated.
32007 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
32008 fixed variable object, the expression is parsed when the variable
32009 object is created, including associating identifiers to specific
32010 variables. The meaning of expression never changes. For a floating
32011 variable object the values of variables whose names appear in the
32012 expressions are re-evaluated every time in the context of the current
32013 frame. Consider this example:
32018 struct work_state state;
32025 If a fixed variable object for the @code{state} variable is created in
32026 this function, and we enter the recursive call, the variable
32027 object will report the value of @code{state} in the top-level
32028 @code{do_work} invocation. On the other hand, a floating variable
32029 object will report the value of @code{state} in the current frame.
32031 If an expression specified when creating a fixed variable object
32032 refers to a local variable, the variable object becomes bound to the
32033 thread and frame in which the variable object is created. When such
32034 variable object is updated, @value{GDBN} makes sure that the
32035 thread/frame combination the variable object is bound to still exists,
32036 and re-evaluates the variable object in context of that thread/frame.
32038 The following is the complete set of @sc{gdb/mi} operations defined to
32039 access this functionality:
32041 @multitable @columnfractions .4 .6
32042 @item @strong{Operation}
32043 @tab @strong{Description}
32045 @item @code{-enable-pretty-printing}
32046 @tab enable Python-based pretty-printing
32047 @item @code{-var-create}
32048 @tab create a variable object
32049 @item @code{-var-delete}
32050 @tab delete the variable object and/or its children
32051 @item @code{-var-set-format}
32052 @tab set the display format of this variable
32053 @item @code{-var-show-format}
32054 @tab show the display format of this variable
32055 @item @code{-var-info-num-children}
32056 @tab tells how many children this object has
32057 @item @code{-var-list-children}
32058 @tab return a list of the object's children
32059 @item @code{-var-info-type}
32060 @tab show the type of this variable object
32061 @item @code{-var-info-expression}
32062 @tab print parent-relative expression that this variable object represents
32063 @item @code{-var-info-path-expression}
32064 @tab print full expression that this variable object represents
32065 @item @code{-var-show-attributes}
32066 @tab is this variable editable? does it exist here?
32067 @item @code{-var-evaluate-expression}
32068 @tab get the value of this variable
32069 @item @code{-var-assign}
32070 @tab set the value of this variable
32071 @item @code{-var-update}
32072 @tab update the variable and its children
32073 @item @code{-var-set-frozen}
32074 @tab set frozeness attribute
32075 @item @code{-var-set-update-range}
32076 @tab set range of children to display on update
32079 In the next subsection we describe each operation in detail and suggest
32080 how it can be used.
32082 @subheading Description And Use of Operations on Variable Objects
32084 @subheading The @code{-enable-pretty-printing} Command
32085 @findex -enable-pretty-printing
32088 -enable-pretty-printing
32091 @value{GDBN} allows Python-based visualizers to affect the output of the
32092 MI variable object commands. However, because there was no way to
32093 implement this in a fully backward-compatible way, a front end must
32094 request that this functionality be enabled.
32096 Once enabled, this feature cannot be disabled.
32098 Note that if Python support has not been compiled into @value{GDBN},
32099 this command will still succeed (and do nothing).
32101 This feature is currently (as of @value{GDBN} 7.0) experimental, and
32102 may work differently in future versions of @value{GDBN}.
32104 @subheading The @code{-var-create} Command
32105 @findex -var-create
32107 @subsubheading Synopsis
32110 -var-create @{@var{name} | "-"@}
32111 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
32114 This operation creates a variable object, which allows the monitoring of
32115 a variable, the result of an expression, a memory cell or a CPU
32118 The @var{name} parameter is the string by which the object can be
32119 referenced. It must be unique. If @samp{-} is specified, the varobj
32120 system will generate a string ``varNNNNNN'' automatically. It will be
32121 unique provided that one does not specify @var{name} of that format.
32122 The command fails if a duplicate name is found.
32124 The frame under which the expression should be evaluated can be
32125 specified by @var{frame-addr}. A @samp{*} indicates that the current
32126 frame should be used. A @samp{@@} indicates that a floating variable
32127 object must be created.
32129 @var{expression} is any expression valid on the current language set (must not
32130 begin with a @samp{*}), or one of the following:
32134 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
32137 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
32140 @samp{$@var{regname}} --- a CPU register name
32143 @cindex dynamic varobj
32144 A varobj's contents may be provided by a Python-based pretty-printer. In this
32145 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
32146 have slightly different semantics in some cases. If the
32147 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
32148 will never create a dynamic varobj. This ensures backward
32149 compatibility for existing clients.
32151 @subsubheading Result
32153 This operation returns attributes of the newly-created varobj. These
32158 The name of the varobj.
32161 The number of children of the varobj. This number is not necessarily
32162 reliable for a dynamic varobj. Instead, you must examine the
32163 @samp{has_more} attribute.
32166 The varobj's scalar value. For a varobj whose type is some sort of
32167 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
32168 will not be interesting.
32171 The varobj's type. This is a string representation of the type, as
32172 would be printed by the @value{GDBN} CLI. If @samp{print object}
32173 (@pxref{Print Settings, set print object}) is set to @code{on}, the
32174 @emph{actual} (derived) type of the object is shown rather than the
32175 @emph{declared} one.
32178 If a variable object is bound to a specific thread, then this is the
32179 thread's identifier.
32182 For a dynamic varobj, this indicates whether there appear to be any
32183 children available. For a non-dynamic varobj, this will be 0.
32186 This attribute will be present and have the value @samp{1} if the
32187 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32188 then this attribute will not be present.
32191 A dynamic varobj can supply a display hint to the front end. The
32192 value comes directly from the Python pretty-printer object's
32193 @code{display_hint} method. @xref{Pretty Printing API}.
32196 Typical output will look like this:
32199 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
32200 has_more="@var{has_more}"
32204 @subheading The @code{-var-delete} Command
32205 @findex -var-delete
32207 @subsubheading Synopsis
32210 -var-delete [ -c ] @var{name}
32213 Deletes a previously created variable object and all of its children.
32214 With the @samp{-c} option, just deletes the children.
32216 Returns an error if the object @var{name} is not found.
32219 @subheading The @code{-var-set-format} Command
32220 @findex -var-set-format
32222 @subsubheading Synopsis
32225 -var-set-format @var{name} @var{format-spec}
32228 Sets the output format for the value of the object @var{name} to be
32231 @anchor{-var-set-format}
32232 The syntax for the @var{format-spec} is as follows:
32235 @var{format-spec} @expansion{}
32236 @{binary | decimal | hexadecimal | octal | natural@}
32239 The natural format is the default format choosen automatically
32240 based on the variable type (like decimal for an @code{int}, hex
32241 for pointers, etc.).
32243 For a variable with children, the format is set only on the
32244 variable itself, and the children are not affected.
32246 @subheading The @code{-var-show-format} Command
32247 @findex -var-show-format
32249 @subsubheading Synopsis
32252 -var-show-format @var{name}
32255 Returns the format used to display the value of the object @var{name}.
32258 @var{format} @expansion{}
32263 @subheading The @code{-var-info-num-children} Command
32264 @findex -var-info-num-children
32266 @subsubheading Synopsis
32269 -var-info-num-children @var{name}
32272 Returns the number of children of a variable object @var{name}:
32278 Note that this number is not completely reliable for a dynamic varobj.
32279 It will return the current number of children, but more children may
32283 @subheading The @code{-var-list-children} Command
32284 @findex -var-list-children
32286 @subsubheading Synopsis
32289 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
32291 @anchor{-var-list-children}
32293 Return a list of the children of the specified variable object and
32294 create variable objects for them, if they do not already exist. With
32295 a single argument or if @var{print-values} has a value of 0 or
32296 @code{--no-values}, print only the names of the variables; if
32297 @var{print-values} is 1 or @code{--all-values}, also print their
32298 values; and if it is 2 or @code{--simple-values} print the name and
32299 value for simple data types and just the name for arrays, structures
32302 @var{from} and @var{to}, if specified, indicate the range of children
32303 to report. If @var{from} or @var{to} is less than zero, the range is
32304 reset and all children will be reported. Otherwise, children starting
32305 at @var{from} (zero-based) and up to and excluding @var{to} will be
32308 If a child range is requested, it will only affect the current call to
32309 @code{-var-list-children}, but not future calls to @code{-var-update}.
32310 For this, you must instead use @code{-var-set-update-range}. The
32311 intent of this approach is to enable a front end to implement any
32312 update approach it likes; for example, scrolling a view may cause the
32313 front end to request more children with @code{-var-list-children}, and
32314 then the front end could call @code{-var-set-update-range} with a
32315 different range to ensure that future updates are restricted to just
32318 For each child the following results are returned:
32323 Name of the variable object created for this child.
32326 The expression to be shown to the user by the front end to designate this child.
32327 For example this may be the name of a structure member.
32329 For a dynamic varobj, this value cannot be used to form an
32330 expression. There is no way to do this at all with a dynamic varobj.
32332 For C/C@t{++} structures there are several pseudo children returned to
32333 designate access qualifiers. For these pseudo children @var{exp} is
32334 @samp{public}, @samp{private}, or @samp{protected}. In this case the
32335 type and value are not present.
32337 A dynamic varobj will not report the access qualifying
32338 pseudo-children, regardless of the language. This information is not
32339 available at all with a dynamic varobj.
32342 Number of children this child has. For a dynamic varobj, this will be
32346 The type of the child. If @samp{print object}
32347 (@pxref{Print Settings, set print object}) is set to @code{on}, the
32348 @emph{actual} (derived) type of the object is shown rather than the
32349 @emph{declared} one.
32352 If values were requested, this is the value.
32355 If this variable object is associated with a thread, this is the thread id.
32356 Otherwise this result is not present.
32359 If the variable object is frozen, this variable will be present with a value of 1.
32362 The result may have its own attributes:
32366 A dynamic varobj can supply a display hint to the front end. The
32367 value comes directly from the Python pretty-printer object's
32368 @code{display_hint} method. @xref{Pretty Printing API}.
32371 This is an integer attribute which is nonzero if there are children
32372 remaining after the end of the selected range.
32375 @subsubheading Example
32379 -var-list-children n
32380 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
32381 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
32383 -var-list-children --all-values n
32384 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
32385 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
32389 @subheading The @code{-var-info-type} Command
32390 @findex -var-info-type
32392 @subsubheading Synopsis
32395 -var-info-type @var{name}
32398 Returns the type of the specified variable @var{name}. The type is
32399 returned as a string in the same format as it is output by the
32403 type=@var{typename}
32407 @subheading The @code{-var-info-expression} Command
32408 @findex -var-info-expression
32410 @subsubheading Synopsis
32413 -var-info-expression @var{name}
32416 Returns a string that is suitable for presenting this
32417 variable object in user interface. The string is generally
32418 not valid expression in the current language, and cannot be evaluated.
32420 For example, if @code{a} is an array, and variable object
32421 @code{A} was created for @code{a}, then we'll get this output:
32424 (gdb) -var-info-expression A.1
32425 ^done,lang="C",exp="1"
32429 Here, the value of @code{lang} is the language name, which can be
32430 found in @ref{Supported Languages}.
32432 Note that the output of the @code{-var-list-children} command also
32433 includes those expressions, so the @code{-var-info-expression} command
32436 @subheading The @code{-var-info-path-expression} Command
32437 @findex -var-info-path-expression
32439 @subsubheading Synopsis
32442 -var-info-path-expression @var{name}
32445 Returns an expression that can be evaluated in the current
32446 context and will yield the same value that a variable object has.
32447 Compare this with the @code{-var-info-expression} command, which
32448 result can be used only for UI presentation. Typical use of
32449 the @code{-var-info-path-expression} command is creating a
32450 watchpoint from a variable object.
32452 This command is currently not valid for children of a dynamic varobj,
32453 and will give an error when invoked on one.
32455 For example, suppose @code{C} is a C@t{++} class, derived from class
32456 @code{Base}, and that the @code{Base} class has a member called
32457 @code{m_size}. Assume a variable @code{c} is has the type of
32458 @code{C} and a variable object @code{C} was created for variable
32459 @code{c}. Then, we'll get this output:
32461 (gdb) -var-info-path-expression C.Base.public.m_size
32462 ^done,path_expr=((Base)c).m_size)
32465 @subheading The @code{-var-show-attributes} Command
32466 @findex -var-show-attributes
32468 @subsubheading Synopsis
32471 -var-show-attributes @var{name}
32474 List attributes of the specified variable object @var{name}:
32477 status=@var{attr} [ ( ,@var{attr} )* ]
32481 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
32483 @subheading The @code{-var-evaluate-expression} Command
32484 @findex -var-evaluate-expression
32486 @subsubheading Synopsis
32489 -var-evaluate-expression [-f @var{format-spec}] @var{name}
32492 Evaluates the expression that is represented by the specified variable
32493 object and returns its value as a string. The format of the string
32494 can be specified with the @samp{-f} option. The possible values of
32495 this option are the same as for @code{-var-set-format}
32496 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
32497 the current display format will be used. The current display format
32498 can be changed using the @code{-var-set-format} command.
32504 Note that one must invoke @code{-var-list-children} for a variable
32505 before the value of a child variable can be evaluated.
32507 @subheading The @code{-var-assign} Command
32508 @findex -var-assign
32510 @subsubheading Synopsis
32513 -var-assign @var{name} @var{expression}
32516 Assigns the value of @var{expression} to the variable object specified
32517 by @var{name}. The object must be @samp{editable}. If the variable's
32518 value is altered by the assign, the variable will show up in any
32519 subsequent @code{-var-update} list.
32521 @subsubheading Example
32529 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
32533 @subheading The @code{-var-update} Command
32534 @findex -var-update
32536 @subsubheading Synopsis
32539 -var-update [@var{print-values}] @{@var{name} | "*"@}
32542 Reevaluate the expressions corresponding to the variable object
32543 @var{name} and all its direct and indirect children, and return the
32544 list of variable objects whose values have changed; @var{name} must
32545 be a root variable object. Here, ``changed'' means that the result of
32546 @code{-var-evaluate-expression} before and after the
32547 @code{-var-update} is different. If @samp{*} is used as the variable
32548 object names, all existing variable objects are updated, except
32549 for frozen ones (@pxref{-var-set-frozen}). The option
32550 @var{print-values} determines whether both names and values, or just
32551 names are printed. The possible values of this option are the same
32552 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
32553 recommended to use the @samp{--all-values} option, to reduce the
32554 number of MI commands needed on each program stop.
32556 With the @samp{*} parameter, if a variable object is bound to a
32557 currently running thread, it will not be updated, without any
32560 If @code{-var-set-update-range} was previously used on a varobj, then
32561 only the selected range of children will be reported.
32563 @code{-var-update} reports all the changed varobjs in a tuple named
32566 Each item in the change list is itself a tuple holding:
32570 The name of the varobj.
32573 If values were requested for this update, then this field will be
32574 present and will hold the value of the varobj.
32577 @anchor{-var-update}
32578 This field is a string which may take one of three values:
32582 The variable object's current value is valid.
32585 The variable object does not currently hold a valid value but it may
32586 hold one in the future if its associated expression comes back into
32590 The variable object no longer holds a valid value.
32591 This can occur when the executable file being debugged has changed,
32592 either through recompilation or by using the @value{GDBN} @code{file}
32593 command. The front end should normally choose to delete these variable
32597 In the future new values may be added to this list so the front should
32598 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
32601 This is only present if the varobj is still valid. If the type
32602 changed, then this will be the string @samp{true}; otherwise it will
32605 When a varobj's type changes, its children are also likely to have
32606 become incorrect. Therefore, the varobj's children are automatically
32607 deleted when this attribute is @samp{true}. Also, the varobj's update
32608 range, when set using the @code{-var-set-update-range} command, is
32612 If the varobj's type changed, then this field will be present and will
32615 @item new_num_children
32616 For a dynamic varobj, if the number of children changed, or if the
32617 type changed, this will be the new number of children.
32619 The @samp{numchild} field in other varobj responses is generally not
32620 valid for a dynamic varobj -- it will show the number of children that
32621 @value{GDBN} knows about, but because dynamic varobjs lazily
32622 instantiate their children, this will not reflect the number of
32623 children which may be available.
32625 The @samp{new_num_children} attribute only reports changes to the
32626 number of children known by @value{GDBN}. This is the only way to
32627 detect whether an update has removed children (which necessarily can
32628 only happen at the end of the update range).
32631 The display hint, if any.
32634 This is an integer value, which will be 1 if there are more children
32635 available outside the varobj's update range.
32638 This attribute will be present and have the value @samp{1} if the
32639 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32640 then this attribute will not be present.
32643 If new children were added to a dynamic varobj within the selected
32644 update range (as set by @code{-var-set-update-range}), then they will
32645 be listed in this attribute.
32648 @subsubheading Example
32655 -var-update --all-values var1
32656 ^done,changelist=[@{name="var1",value="3",in_scope="true",
32657 type_changed="false"@}]
32661 @subheading The @code{-var-set-frozen} Command
32662 @findex -var-set-frozen
32663 @anchor{-var-set-frozen}
32665 @subsubheading Synopsis
32668 -var-set-frozen @var{name} @var{flag}
32671 Set the frozenness flag on the variable object @var{name}. The
32672 @var{flag} parameter should be either @samp{1} to make the variable
32673 frozen or @samp{0} to make it unfrozen. If a variable object is
32674 frozen, then neither itself, nor any of its children, are
32675 implicitly updated by @code{-var-update} of
32676 a parent variable or by @code{-var-update *}. Only
32677 @code{-var-update} of the variable itself will update its value and
32678 values of its children. After a variable object is unfrozen, it is
32679 implicitly updated by all subsequent @code{-var-update} operations.
32680 Unfreezing a variable does not update it, only subsequent
32681 @code{-var-update} does.
32683 @subsubheading Example
32687 -var-set-frozen V 1
32692 @subheading The @code{-var-set-update-range} command
32693 @findex -var-set-update-range
32694 @anchor{-var-set-update-range}
32696 @subsubheading Synopsis
32699 -var-set-update-range @var{name} @var{from} @var{to}
32702 Set the range of children to be returned by future invocations of
32703 @code{-var-update}.
32705 @var{from} and @var{to} indicate the range of children to report. If
32706 @var{from} or @var{to} is less than zero, the range is reset and all
32707 children will be reported. Otherwise, children starting at @var{from}
32708 (zero-based) and up to and excluding @var{to} will be reported.
32710 @subsubheading Example
32714 -var-set-update-range V 1 2
32718 @subheading The @code{-var-set-visualizer} command
32719 @findex -var-set-visualizer
32720 @anchor{-var-set-visualizer}
32722 @subsubheading Synopsis
32725 -var-set-visualizer @var{name} @var{visualizer}
32728 Set a visualizer for the variable object @var{name}.
32730 @var{visualizer} is the visualizer to use. The special value
32731 @samp{None} means to disable any visualizer in use.
32733 If not @samp{None}, @var{visualizer} must be a Python expression.
32734 This expression must evaluate to a callable object which accepts a
32735 single argument. @value{GDBN} will call this object with the value of
32736 the varobj @var{name} as an argument (this is done so that the same
32737 Python pretty-printing code can be used for both the CLI and MI).
32738 When called, this object must return an object which conforms to the
32739 pretty-printing interface (@pxref{Pretty Printing API}).
32741 The pre-defined function @code{gdb.default_visualizer} may be used to
32742 select a visualizer by following the built-in process
32743 (@pxref{Selecting Pretty-Printers}). This is done automatically when
32744 a varobj is created, and so ordinarily is not needed.
32746 This feature is only available if Python support is enabled. The MI
32747 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
32748 can be used to check this.
32750 @subsubheading Example
32752 Resetting the visualizer:
32756 -var-set-visualizer V None
32760 Reselecting the default (type-based) visualizer:
32764 -var-set-visualizer V gdb.default_visualizer
32768 Suppose @code{SomeClass} is a visualizer class. A lambda expression
32769 can be used to instantiate this class for a varobj:
32773 -var-set-visualizer V "lambda val: SomeClass()"
32777 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32778 @node GDB/MI Data Manipulation
32779 @section @sc{gdb/mi} Data Manipulation
32781 @cindex data manipulation, in @sc{gdb/mi}
32782 @cindex @sc{gdb/mi}, data manipulation
32783 This section describes the @sc{gdb/mi} commands that manipulate data:
32784 examine memory and registers, evaluate expressions, etc.
32786 @c REMOVED FROM THE INTERFACE.
32787 @c @subheading -data-assign
32788 @c Change the value of a program variable. Plenty of side effects.
32789 @c @subsubheading GDB Command
32791 @c @subsubheading Example
32794 @subheading The @code{-data-disassemble} Command
32795 @findex -data-disassemble
32797 @subsubheading Synopsis
32801 [ -s @var{start-addr} -e @var{end-addr} ]
32802 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
32810 @item @var{start-addr}
32811 is the beginning address (or @code{$pc})
32812 @item @var{end-addr}
32814 @item @var{filename}
32815 is the name of the file to disassemble
32816 @item @var{linenum}
32817 is the line number to disassemble around
32819 is the number of disassembly lines to be produced. If it is -1,
32820 the whole function will be disassembled, in case no @var{end-addr} is
32821 specified. If @var{end-addr} is specified as a non-zero value, and
32822 @var{lines} is lower than the number of disassembly lines between
32823 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
32824 displayed; if @var{lines} is higher than the number of lines between
32825 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
32828 is either 0 (meaning only disassembly), 1 (meaning mixed source and
32829 disassembly), 2 (meaning disassembly with raw opcodes), or 3 (meaning
32830 mixed source and disassembly with raw opcodes).
32833 @subsubheading Result
32835 The result of the @code{-data-disassemble} command will be a list named
32836 @samp{asm_insns}, the contents of this list depend on the @var{mode}
32837 used with the @code{-data-disassemble} command.
32839 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
32844 The address at which this instruction was disassembled.
32847 The name of the function this instruction is within.
32850 The decimal offset in bytes from the start of @samp{func-name}.
32853 The text disassembly for this @samp{address}.
32856 This field is only present for mode 2. This contains the raw opcode
32857 bytes for the @samp{inst} field.
32861 For modes 1 and 3 the @samp{asm_insns} list contains tuples named
32862 @samp{src_and_asm_line}, each of which has the following fields:
32866 The line number within @samp{file}.
32869 The file name from the compilation unit. This might be an absolute
32870 file name or a relative file name depending on the compile command
32874 Absolute file name of @samp{file}. It is converted to a canonical form
32875 using the source file search path
32876 (@pxref{Source Path, ,Specifying Source Directories})
32877 and after resolving all the symbolic links.
32879 If the source file is not found this field will contain the path as
32880 present in the debug information.
32882 @item line_asm_insn
32883 This is a list of tuples containing the disassembly for @samp{line} in
32884 @samp{file}. The fields of each tuple are the same as for
32885 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
32886 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
32891 Note that whatever included in the @samp{inst} field, is not
32892 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
32895 @subsubheading @value{GDBN} Command
32897 The corresponding @value{GDBN} command is @samp{disassemble}.
32899 @subsubheading Example
32901 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
32905 -data-disassemble -s $pc -e "$pc + 20" -- 0
32908 @{address="0x000107c0",func-name="main",offset="4",
32909 inst="mov 2, %o0"@},
32910 @{address="0x000107c4",func-name="main",offset="8",
32911 inst="sethi %hi(0x11800), %o2"@},
32912 @{address="0x000107c8",func-name="main",offset="12",
32913 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
32914 @{address="0x000107cc",func-name="main",offset="16",
32915 inst="sethi %hi(0x11800), %o2"@},
32916 @{address="0x000107d0",func-name="main",offset="20",
32917 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
32921 Disassemble the whole @code{main} function. Line 32 is part of
32925 -data-disassemble -f basics.c -l 32 -- 0
32927 @{address="0x000107bc",func-name="main",offset="0",
32928 inst="save %sp, -112, %sp"@},
32929 @{address="0x000107c0",func-name="main",offset="4",
32930 inst="mov 2, %o0"@},
32931 @{address="0x000107c4",func-name="main",offset="8",
32932 inst="sethi %hi(0x11800), %o2"@},
32934 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
32935 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
32939 Disassemble 3 instructions from the start of @code{main}:
32943 -data-disassemble -f basics.c -l 32 -n 3 -- 0
32945 @{address="0x000107bc",func-name="main",offset="0",
32946 inst="save %sp, -112, %sp"@},
32947 @{address="0x000107c0",func-name="main",offset="4",
32948 inst="mov 2, %o0"@},
32949 @{address="0x000107c4",func-name="main",offset="8",
32950 inst="sethi %hi(0x11800), %o2"@}]
32954 Disassemble 3 instructions from the start of @code{main} in mixed mode:
32958 -data-disassemble -f basics.c -l 32 -n 3 -- 1
32960 src_and_asm_line=@{line="31",
32961 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32962 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32963 line_asm_insn=[@{address="0x000107bc",
32964 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
32965 src_and_asm_line=@{line="32",
32966 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32967 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32968 line_asm_insn=[@{address="0x000107c0",
32969 func-name="main",offset="4",inst="mov 2, %o0"@},
32970 @{address="0x000107c4",func-name="main",offset="8",
32971 inst="sethi %hi(0x11800), %o2"@}]@}]
32976 @subheading The @code{-data-evaluate-expression} Command
32977 @findex -data-evaluate-expression
32979 @subsubheading Synopsis
32982 -data-evaluate-expression @var{expr}
32985 Evaluate @var{expr} as an expression. The expression could contain an
32986 inferior function call. The function call will execute synchronously.
32987 If the expression contains spaces, it must be enclosed in double quotes.
32989 @subsubheading @value{GDBN} Command
32991 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
32992 @samp{call}. In @code{gdbtk} only, there's a corresponding
32993 @samp{gdb_eval} command.
32995 @subsubheading Example
32997 In the following example, the numbers that precede the commands are the
32998 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
32999 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
33003 211-data-evaluate-expression A
33006 311-data-evaluate-expression &A
33007 311^done,value="0xefffeb7c"
33009 411-data-evaluate-expression A+3
33012 511-data-evaluate-expression "A + 3"
33018 @subheading The @code{-data-list-changed-registers} Command
33019 @findex -data-list-changed-registers
33021 @subsubheading Synopsis
33024 -data-list-changed-registers
33027 Display a list of the registers that have changed.
33029 @subsubheading @value{GDBN} Command
33031 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
33032 has the corresponding command @samp{gdb_changed_register_list}.
33034 @subsubheading Example
33036 On a PPC MBX board:
33044 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
33045 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
33048 -data-list-changed-registers
33049 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
33050 "10","11","13","14","15","16","17","18","19","20","21","22","23",
33051 "24","25","26","27","28","30","31","64","65","66","67","69"]
33056 @subheading The @code{-data-list-register-names} Command
33057 @findex -data-list-register-names
33059 @subsubheading Synopsis
33062 -data-list-register-names [ ( @var{regno} )+ ]
33065 Show a list of register names for the current target. If no arguments
33066 are given, it shows a list of the names of all the registers. If
33067 integer numbers are given as arguments, it will print a list of the
33068 names of the registers corresponding to the arguments. To ensure
33069 consistency between a register name and its number, the output list may
33070 include empty register names.
33072 @subsubheading @value{GDBN} Command
33074 @value{GDBN} does not have a command which corresponds to
33075 @samp{-data-list-register-names}. In @code{gdbtk} there is a
33076 corresponding command @samp{gdb_regnames}.
33078 @subsubheading Example
33080 For the PPC MBX board:
33083 -data-list-register-names
33084 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
33085 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
33086 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
33087 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
33088 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
33089 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
33090 "", "pc","ps","cr","lr","ctr","xer"]
33092 -data-list-register-names 1 2 3
33093 ^done,register-names=["r1","r2","r3"]
33097 @subheading The @code{-data-list-register-values} Command
33098 @findex -data-list-register-values
33100 @subsubheading Synopsis
33103 -data-list-register-values
33104 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
33107 Display the registers' contents. @var{fmt} is the format according to
33108 which the registers' contents are to be returned, followed by an optional
33109 list of numbers specifying the registers to display. A missing list of
33110 numbers indicates that the contents of all the registers must be
33111 returned. The @code{--skip-unavailable} option indicates that only
33112 the available registers are to be returned.
33114 Allowed formats for @var{fmt} are:
33131 @subsubheading @value{GDBN} Command
33133 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
33134 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
33136 @subsubheading Example
33138 For a PPC MBX board (note: line breaks are for readability only, they
33139 don't appear in the actual output):
33143 -data-list-register-values r 64 65
33144 ^done,register-values=[@{number="64",value="0xfe00a300"@},
33145 @{number="65",value="0x00029002"@}]
33147 -data-list-register-values x
33148 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
33149 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
33150 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
33151 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
33152 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
33153 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
33154 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
33155 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
33156 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
33157 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
33158 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
33159 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
33160 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
33161 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
33162 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
33163 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
33164 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
33165 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
33166 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
33167 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
33168 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
33169 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
33170 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
33171 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
33172 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
33173 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
33174 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
33175 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
33176 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
33177 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
33178 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
33179 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
33180 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
33181 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
33182 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
33183 @{number="69",value="0x20002b03"@}]
33188 @subheading The @code{-data-read-memory} Command
33189 @findex -data-read-memory
33191 This command is deprecated, use @code{-data-read-memory-bytes} instead.
33193 @subsubheading Synopsis
33196 -data-read-memory [ -o @var{byte-offset} ]
33197 @var{address} @var{word-format} @var{word-size}
33198 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
33205 @item @var{address}
33206 An expression specifying the address of the first memory word to be
33207 read. Complex expressions containing embedded white space should be
33208 quoted using the C convention.
33210 @item @var{word-format}
33211 The format to be used to print the memory words. The notation is the
33212 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
33215 @item @var{word-size}
33216 The size of each memory word in bytes.
33218 @item @var{nr-rows}
33219 The number of rows in the output table.
33221 @item @var{nr-cols}
33222 The number of columns in the output table.
33225 If present, indicates that each row should include an @sc{ascii} dump. The
33226 value of @var{aschar} is used as a padding character when a byte is not a
33227 member of the printable @sc{ascii} character set (printable @sc{ascii}
33228 characters are those whose code is between 32 and 126, inclusively).
33230 @item @var{byte-offset}
33231 An offset to add to the @var{address} before fetching memory.
33234 This command displays memory contents as a table of @var{nr-rows} by
33235 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
33236 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
33237 (returned as @samp{total-bytes}). Should less than the requested number
33238 of bytes be returned by the target, the missing words are identified
33239 using @samp{N/A}. The number of bytes read from the target is returned
33240 in @samp{nr-bytes} and the starting address used to read memory in
33243 The address of the next/previous row or page is available in
33244 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
33247 @subsubheading @value{GDBN} Command
33249 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
33250 @samp{gdb_get_mem} memory read command.
33252 @subsubheading Example
33254 Read six bytes of memory starting at @code{bytes+6} but then offset by
33255 @code{-6} bytes. Format as three rows of two columns. One byte per
33256 word. Display each word in hex.
33260 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
33261 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
33262 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
33263 prev-page="0x0000138a",memory=[
33264 @{addr="0x00001390",data=["0x00","0x01"]@},
33265 @{addr="0x00001392",data=["0x02","0x03"]@},
33266 @{addr="0x00001394",data=["0x04","0x05"]@}]
33270 Read two bytes of memory starting at address @code{shorts + 64} and
33271 display as a single word formatted in decimal.
33275 5-data-read-memory shorts+64 d 2 1 1
33276 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
33277 next-row="0x00001512",prev-row="0x0000150e",
33278 next-page="0x00001512",prev-page="0x0000150e",memory=[
33279 @{addr="0x00001510",data=["128"]@}]
33283 Read thirty two bytes of memory starting at @code{bytes+16} and format
33284 as eight rows of four columns. Include a string encoding with @samp{x}
33285 used as the non-printable character.
33289 4-data-read-memory bytes+16 x 1 8 4 x
33290 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
33291 next-row="0x000013c0",prev-row="0x0000139c",
33292 next-page="0x000013c0",prev-page="0x00001380",memory=[
33293 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
33294 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
33295 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
33296 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
33297 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
33298 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
33299 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
33300 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
33304 @subheading The @code{-data-read-memory-bytes} Command
33305 @findex -data-read-memory-bytes
33307 @subsubheading Synopsis
33310 -data-read-memory-bytes [ -o @var{byte-offset} ]
33311 @var{address} @var{count}
33318 @item @var{address}
33319 An expression specifying the address of the first memory word to be
33320 read. Complex expressions containing embedded white space should be
33321 quoted using the C convention.
33324 The number of bytes to read. This should be an integer literal.
33326 @item @var{byte-offset}
33327 The offsets in bytes relative to @var{address} at which to start
33328 reading. This should be an integer literal. This option is provided
33329 so that a frontend is not required to first evaluate address and then
33330 perform address arithmetics itself.
33334 This command attempts to read all accessible memory regions in the
33335 specified range. First, all regions marked as unreadable in the memory
33336 map (if one is defined) will be skipped. @xref{Memory Region
33337 Attributes}. Second, @value{GDBN} will attempt to read the remaining
33338 regions. For each one, if reading full region results in an errors,
33339 @value{GDBN} will try to read a subset of the region.
33341 In general, every single byte in the region may be readable or not,
33342 and the only way to read every readable byte is to try a read at
33343 every address, which is not practical. Therefore, @value{GDBN} will
33344 attempt to read all accessible bytes at either beginning or the end
33345 of the region, using a binary division scheme. This heuristic works
33346 well for reading accross a memory map boundary. Note that if a region
33347 has a readable range that is neither at the beginning or the end,
33348 @value{GDBN} will not read it.
33350 The result record (@pxref{GDB/MI Result Records}) that is output of
33351 the command includes a field named @samp{memory} whose content is a
33352 list of tuples. Each tuple represent a successfully read memory block
33353 and has the following fields:
33357 The start address of the memory block, as hexadecimal literal.
33360 The end address of the memory block, as hexadecimal literal.
33363 The offset of the memory block, as hexadecimal literal, relative to
33364 the start address passed to @code{-data-read-memory-bytes}.
33367 The contents of the memory block, in hex.
33373 @subsubheading @value{GDBN} Command
33375 The corresponding @value{GDBN} command is @samp{x}.
33377 @subsubheading Example
33381 -data-read-memory-bytes &a 10
33382 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
33384 contents="01000000020000000300"@}]
33389 @subheading The @code{-data-write-memory-bytes} Command
33390 @findex -data-write-memory-bytes
33392 @subsubheading Synopsis
33395 -data-write-memory-bytes @var{address} @var{contents}
33396 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
33403 @item @var{address}
33404 An expression specifying the address of the first memory word to be
33405 read. Complex expressions containing embedded white space should be
33406 quoted using the C convention.
33408 @item @var{contents}
33409 The hex-encoded bytes to write.
33412 Optional argument indicating the number of bytes to be written. If @var{count}
33413 is greater than @var{contents}' length, @value{GDBN} will repeatedly
33414 write @var{contents} until it fills @var{count} bytes.
33418 @subsubheading @value{GDBN} Command
33420 There's no corresponding @value{GDBN} command.
33422 @subsubheading Example
33426 -data-write-memory-bytes &a "aabbccdd"
33433 -data-write-memory-bytes &a "aabbccdd" 16e
33438 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33439 @node GDB/MI Tracepoint Commands
33440 @section @sc{gdb/mi} Tracepoint Commands
33442 The commands defined in this section implement MI support for
33443 tracepoints. For detailed introduction, see @ref{Tracepoints}.
33445 @subheading The @code{-trace-find} Command
33446 @findex -trace-find
33448 @subsubheading Synopsis
33451 -trace-find @var{mode} [@var{parameters}@dots{}]
33454 Find a trace frame using criteria defined by @var{mode} and
33455 @var{parameters}. The following table lists permissible
33456 modes and their parameters. For details of operation, see @ref{tfind}.
33461 No parameters are required. Stops examining trace frames.
33464 An integer is required as parameter. Selects tracepoint frame with
33467 @item tracepoint-number
33468 An integer is required as parameter. Finds next
33469 trace frame that corresponds to tracepoint with the specified number.
33472 An address is required as parameter. Finds
33473 next trace frame that corresponds to any tracepoint at the specified
33476 @item pc-inside-range
33477 Two addresses are required as parameters. Finds next trace
33478 frame that corresponds to a tracepoint at an address inside the
33479 specified range. Both bounds are considered to be inside the range.
33481 @item pc-outside-range
33482 Two addresses are required as parameters. Finds
33483 next trace frame that corresponds to a tracepoint at an address outside
33484 the specified range. Both bounds are considered to be inside the range.
33487 Line specification is required as parameter. @xref{Specify Location}.
33488 Finds next trace frame that corresponds to a tracepoint at
33489 the specified location.
33493 If @samp{none} was passed as @var{mode}, the response does not
33494 have fields. Otherwise, the response may have the following fields:
33498 This field has either @samp{0} or @samp{1} as the value, depending
33499 on whether a matching tracepoint was found.
33502 The index of the found traceframe. This field is present iff
33503 the @samp{found} field has value of @samp{1}.
33506 The index of the found tracepoint. This field is present iff
33507 the @samp{found} field has value of @samp{1}.
33510 The information about the frame corresponding to the found trace
33511 frame. This field is present only if a trace frame was found.
33512 @xref{GDB/MI Frame Information}, for description of this field.
33516 @subsubheading @value{GDBN} Command
33518 The corresponding @value{GDBN} command is @samp{tfind}.
33520 @subheading -trace-define-variable
33521 @findex -trace-define-variable
33523 @subsubheading Synopsis
33526 -trace-define-variable @var{name} [ @var{value} ]
33529 Create trace variable @var{name} if it does not exist. If
33530 @var{value} is specified, sets the initial value of the specified
33531 trace variable to that value. Note that the @var{name} should start
33532 with the @samp{$} character.
33534 @subsubheading @value{GDBN} Command
33536 The corresponding @value{GDBN} command is @samp{tvariable}.
33538 @subheading The @code{-trace-frame-collected} Command
33539 @findex -trace-frame-collected
33541 @subsubheading Synopsis
33544 -trace-frame-collected
33545 [--var-print-values @var{var_pval}]
33546 [--comp-print-values @var{comp_pval}]
33547 [--registers-format @var{regformat}]
33548 [--memory-contents]
33551 This command returns the set of collected objects, register names,
33552 trace state variable names, memory ranges and computed expressions
33553 that have been collected at a particular trace frame. The optional
33554 parameters to the command affect the output format in different ways.
33555 See the output description table below for more details.
33557 The reported names can be used in the normal manner to create
33558 varobjs and inspect the objects themselves. The items returned by
33559 this command are categorized so that it is clear which is a variable,
33560 which is a register, which is a trace state variable, which is a
33561 memory range and which is a computed expression.
33563 For instance, if the actions were
33565 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
33566 collect *(int*)0xaf02bef0@@40
33570 the object collected in its entirety would be @code{myVar}. The
33571 object @code{myArray} would be partially collected, because only the
33572 element at index @code{myIndex} would be collected. The remaining
33573 objects would be computed expressions.
33575 An example output would be:
33579 -trace-frame-collected
33581 explicit-variables=[@{name="myVar",value="1"@}],
33582 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
33583 @{name="myObj.field",value="0"@},
33584 @{name="myPtr->field",value="1"@},
33585 @{name="myCount + 2",value="3"@},
33586 @{name="$tvar1 + 1",value="43970027"@}],
33587 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
33588 @{number="1",value="0x0"@},
33589 @{number="2",value="0x4"@},
33591 @{number="125",value="0x0"@}],
33592 tvars=[@{name="$tvar1",current="43970026"@}],
33593 memory=[@{address="0x0000000000602264",length="4"@},
33594 @{address="0x0000000000615bc0",length="4"@}]
33601 @item explicit-variables
33602 The set of objects that have been collected in their entirety (as
33603 opposed to collecting just a few elements of an array or a few struct
33604 members). For each object, its name and value are printed.
33605 The @code{--var-print-values} option affects how or whether the value
33606 field is output. If @var{var_pval} is 0, then print only the names;
33607 if it is 1, print also their values; and if it is 2, print the name,
33608 type and value for simple data types, and the name and type for
33609 arrays, structures and unions.
33611 @item computed-expressions
33612 The set of computed expressions that have been collected at the
33613 current trace frame. The @code{--comp-print-values} option affects
33614 this set like the @code{--var-print-values} option affects the
33615 @code{explicit-variables} set. See above.
33618 The registers that have been collected at the current trace frame.
33619 For each register collected, the name and current value are returned.
33620 The value is formatted according to the @code{--registers-format}
33621 option. See the @command{-data-list-register-values} command for a
33622 list of the allowed formats. The default is @samp{x}.
33625 The trace state variables that have been collected at the current
33626 trace frame. For each trace state variable collected, the name and
33627 current value are returned.
33630 The set of memory ranges that have been collected at the current trace
33631 frame. Its content is a list of tuples. Each tuple represents a
33632 collected memory range and has the following fields:
33636 The start address of the memory range, as hexadecimal literal.
33639 The length of the memory range, as decimal literal.
33642 The contents of the memory block, in hex. This field is only present
33643 if the @code{--memory-contents} option is specified.
33649 @subsubheading @value{GDBN} Command
33651 There is no corresponding @value{GDBN} command.
33653 @subsubheading Example
33655 @subheading -trace-list-variables
33656 @findex -trace-list-variables
33658 @subsubheading Synopsis
33661 -trace-list-variables
33664 Return a table of all defined trace variables. Each element of the
33665 table has the following fields:
33669 The name of the trace variable. This field is always present.
33672 The initial value. This is a 64-bit signed integer. This
33673 field is always present.
33676 The value the trace variable has at the moment. This is a 64-bit
33677 signed integer. This field is absent iff current value is
33678 not defined, for example if the trace was never run, or is
33683 @subsubheading @value{GDBN} Command
33685 The corresponding @value{GDBN} command is @samp{tvariables}.
33687 @subsubheading Example
33691 -trace-list-variables
33692 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
33693 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
33694 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
33695 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
33696 body=[variable=@{name="$trace_timestamp",initial="0"@}
33697 variable=@{name="$foo",initial="10",current="15"@}]@}
33701 @subheading -trace-save
33702 @findex -trace-save
33704 @subsubheading Synopsis
33707 -trace-save [-r ] @var{filename}
33710 Saves the collected trace data to @var{filename}. Without the
33711 @samp{-r} option, the data is downloaded from the target and saved
33712 in a local file. With the @samp{-r} option the target is asked
33713 to perform the save.
33715 @subsubheading @value{GDBN} Command
33717 The corresponding @value{GDBN} command is @samp{tsave}.
33720 @subheading -trace-start
33721 @findex -trace-start
33723 @subsubheading Synopsis
33729 Starts a tracing experiments. The result of this command does not
33732 @subsubheading @value{GDBN} Command
33734 The corresponding @value{GDBN} command is @samp{tstart}.
33736 @subheading -trace-status
33737 @findex -trace-status
33739 @subsubheading Synopsis
33745 Obtains the status of a tracing experiment. The result may include
33746 the following fields:
33751 May have a value of either @samp{0}, when no tracing operations are
33752 supported, @samp{1}, when all tracing operations are supported, or
33753 @samp{file} when examining trace file. In the latter case, examining
33754 of trace frame is possible but new tracing experiement cannot be
33755 started. This field is always present.
33758 May have a value of either @samp{0} or @samp{1} depending on whether
33759 tracing experiement is in progress on target. This field is present
33760 if @samp{supported} field is not @samp{0}.
33763 Report the reason why the tracing was stopped last time. This field
33764 may be absent iff tracing was never stopped on target yet. The
33765 value of @samp{request} means the tracing was stopped as result of
33766 the @code{-trace-stop} command. The value of @samp{overflow} means
33767 the tracing buffer is full. The value of @samp{disconnection} means
33768 tracing was automatically stopped when @value{GDBN} has disconnected.
33769 The value of @samp{passcount} means tracing was stopped when a
33770 tracepoint was passed a maximal number of times for that tracepoint.
33771 This field is present if @samp{supported} field is not @samp{0}.
33773 @item stopping-tracepoint
33774 The number of tracepoint whose passcount as exceeded. This field is
33775 present iff the @samp{stop-reason} field has the value of
33779 @itemx frames-created
33780 The @samp{frames} field is a count of the total number of trace frames
33781 in the trace buffer, while @samp{frames-created} is the total created
33782 during the run, including ones that were discarded, such as when a
33783 circular trace buffer filled up. Both fields are optional.
33787 These fields tell the current size of the tracing buffer and the
33788 remaining space. These fields are optional.
33791 The value of the circular trace buffer flag. @code{1} means that the
33792 trace buffer is circular and old trace frames will be discarded if
33793 necessary to make room, @code{0} means that the trace buffer is linear
33797 The value of the disconnected tracing flag. @code{1} means that
33798 tracing will continue after @value{GDBN} disconnects, @code{0} means
33799 that the trace run will stop.
33802 The filename of the trace file being examined. This field is
33803 optional, and only present when examining a trace file.
33807 @subsubheading @value{GDBN} Command
33809 The corresponding @value{GDBN} command is @samp{tstatus}.
33811 @subheading -trace-stop
33812 @findex -trace-stop
33814 @subsubheading Synopsis
33820 Stops a tracing experiment. The result of this command has the same
33821 fields as @code{-trace-status}, except that the @samp{supported} and
33822 @samp{running} fields are not output.
33824 @subsubheading @value{GDBN} Command
33826 The corresponding @value{GDBN} command is @samp{tstop}.
33829 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33830 @node GDB/MI Symbol Query
33831 @section @sc{gdb/mi} Symbol Query Commands
33835 @subheading The @code{-symbol-info-address} Command
33836 @findex -symbol-info-address
33838 @subsubheading Synopsis
33841 -symbol-info-address @var{symbol}
33844 Describe where @var{symbol} is stored.
33846 @subsubheading @value{GDBN} Command
33848 The corresponding @value{GDBN} command is @samp{info address}.
33850 @subsubheading Example
33854 @subheading The @code{-symbol-info-file} Command
33855 @findex -symbol-info-file
33857 @subsubheading Synopsis
33863 Show the file for the symbol.
33865 @subsubheading @value{GDBN} Command
33867 There's no equivalent @value{GDBN} command. @code{gdbtk} has
33868 @samp{gdb_find_file}.
33870 @subsubheading Example
33874 @subheading The @code{-symbol-info-function} Command
33875 @findex -symbol-info-function
33877 @subsubheading Synopsis
33880 -symbol-info-function
33883 Show which function the symbol lives in.
33885 @subsubheading @value{GDBN} Command
33887 @samp{gdb_get_function} in @code{gdbtk}.
33889 @subsubheading Example
33893 @subheading The @code{-symbol-info-line} Command
33894 @findex -symbol-info-line
33896 @subsubheading Synopsis
33902 Show the core addresses of the code for a source line.
33904 @subsubheading @value{GDBN} Command
33906 The corresponding @value{GDBN} command is @samp{info line}.
33907 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
33909 @subsubheading Example
33913 @subheading The @code{-symbol-info-symbol} Command
33914 @findex -symbol-info-symbol
33916 @subsubheading Synopsis
33919 -symbol-info-symbol @var{addr}
33922 Describe what symbol is at location @var{addr}.
33924 @subsubheading @value{GDBN} Command
33926 The corresponding @value{GDBN} command is @samp{info symbol}.
33928 @subsubheading Example
33932 @subheading The @code{-symbol-list-functions} Command
33933 @findex -symbol-list-functions
33935 @subsubheading Synopsis
33938 -symbol-list-functions
33941 List the functions in the executable.
33943 @subsubheading @value{GDBN} Command
33945 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
33946 @samp{gdb_search} in @code{gdbtk}.
33948 @subsubheading Example
33953 @subheading The @code{-symbol-list-lines} Command
33954 @findex -symbol-list-lines
33956 @subsubheading Synopsis
33959 -symbol-list-lines @var{filename}
33962 Print the list of lines that contain code and their associated program
33963 addresses for the given source filename. The entries are sorted in
33964 ascending PC order.
33966 @subsubheading @value{GDBN} Command
33968 There is no corresponding @value{GDBN} command.
33970 @subsubheading Example
33973 -symbol-list-lines basics.c
33974 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
33980 @subheading The @code{-symbol-list-types} Command
33981 @findex -symbol-list-types
33983 @subsubheading Synopsis
33989 List all the type names.
33991 @subsubheading @value{GDBN} Command
33993 The corresponding commands are @samp{info types} in @value{GDBN},
33994 @samp{gdb_search} in @code{gdbtk}.
33996 @subsubheading Example
34000 @subheading The @code{-symbol-list-variables} Command
34001 @findex -symbol-list-variables
34003 @subsubheading Synopsis
34006 -symbol-list-variables
34009 List all the global and static variable names.
34011 @subsubheading @value{GDBN} Command
34013 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
34015 @subsubheading Example
34019 @subheading The @code{-symbol-locate} Command
34020 @findex -symbol-locate
34022 @subsubheading Synopsis
34028 @subsubheading @value{GDBN} Command
34030 @samp{gdb_loc} in @code{gdbtk}.
34032 @subsubheading Example
34036 @subheading The @code{-symbol-type} Command
34037 @findex -symbol-type
34039 @subsubheading Synopsis
34042 -symbol-type @var{variable}
34045 Show type of @var{variable}.
34047 @subsubheading @value{GDBN} Command
34049 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
34050 @samp{gdb_obj_variable}.
34052 @subsubheading Example
34057 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34058 @node GDB/MI File Commands
34059 @section @sc{gdb/mi} File Commands
34061 This section describes the GDB/MI commands to specify executable file names
34062 and to read in and obtain symbol table information.
34064 @subheading The @code{-file-exec-and-symbols} Command
34065 @findex -file-exec-and-symbols
34067 @subsubheading Synopsis
34070 -file-exec-and-symbols @var{file}
34073 Specify the executable file to be debugged. This file is the one from
34074 which the symbol table is also read. If no file is specified, the
34075 command clears the executable and symbol information. If breakpoints
34076 are set when using this command with no arguments, @value{GDBN} will produce
34077 error messages. Otherwise, no output is produced, except a completion
34080 @subsubheading @value{GDBN} Command
34082 The corresponding @value{GDBN} command is @samp{file}.
34084 @subsubheading Example
34088 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
34094 @subheading The @code{-file-exec-file} Command
34095 @findex -file-exec-file
34097 @subsubheading Synopsis
34100 -file-exec-file @var{file}
34103 Specify the executable file to be debugged. Unlike
34104 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
34105 from this file. If used without argument, @value{GDBN} clears the information
34106 about the executable file. No output is produced, except a completion
34109 @subsubheading @value{GDBN} Command
34111 The corresponding @value{GDBN} command is @samp{exec-file}.
34113 @subsubheading Example
34117 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
34124 @subheading The @code{-file-list-exec-sections} Command
34125 @findex -file-list-exec-sections
34127 @subsubheading Synopsis
34130 -file-list-exec-sections
34133 List the sections of the current executable file.
34135 @subsubheading @value{GDBN} Command
34137 The @value{GDBN} command @samp{info file} shows, among the rest, the same
34138 information as this command. @code{gdbtk} has a corresponding command
34139 @samp{gdb_load_info}.
34141 @subsubheading Example
34146 @subheading The @code{-file-list-exec-source-file} Command
34147 @findex -file-list-exec-source-file
34149 @subsubheading Synopsis
34152 -file-list-exec-source-file
34155 List the line number, the current source file, and the absolute path
34156 to the current source file for the current executable. The macro
34157 information field has a value of @samp{1} or @samp{0} depending on
34158 whether or not the file includes preprocessor macro information.
34160 @subsubheading @value{GDBN} Command
34162 The @value{GDBN} equivalent is @samp{info source}
34164 @subsubheading Example
34168 123-file-list-exec-source-file
34169 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
34174 @subheading The @code{-file-list-exec-source-files} Command
34175 @findex -file-list-exec-source-files
34177 @subsubheading Synopsis
34180 -file-list-exec-source-files
34183 List the source files for the current executable.
34185 It will always output both the filename and fullname (absolute file
34186 name) of a source file.
34188 @subsubheading @value{GDBN} Command
34190 The @value{GDBN} equivalent is @samp{info sources}.
34191 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
34193 @subsubheading Example
34196 -file-list-exec-source-files
34198 @{file=foo.c,fullname=/home/foo.c@},
34199 @{file=/home/bar.c,fullname=/home/bar.c@},
34200 @{file=gdb_could_not_find_fullpath.c@}]
34205 @subheading The @code{-file-list-shared-libraries} Command
34206 @findex -file-list-shared-libraries
34208 @subsubheading Synopsis
34211 -file-list-shared-libraries
34214 List the shared libraries in the program.
34216 @subsubheading @value{GDBN} Command
34218 The corresponding @value{GDBN} command is @samp{info shared}.
34220 @subsubheading Example
34224 @subheading The @code{-file-list-symbol-files} Command
34225 @findex -file-list-symbol-files
34227 @subsubheading Synopsis
34230 -file-list-symbol-files
34235 @subsubheading @value{GDBN} Command
34237 The corresponding @value{GDBN} command is @samp{info file} (part of it).
34239 @subsubheading Example
34244 @subheading The @code{-file-symbol-file} Command
34245 @findex -file-symbol-file
34247 @subsubheading Synopsis
34250 -file-symbol-file @var{file}
34253 Read symbol table info from the specified @var{file} argument. When
34254 used without arguments, clears @value{GDBN}'s symbol table info. No output is
34255 produced, except for a completion notification.
34257 @subsubheading @value{GDBN} Command
34259 The corresponding @value{GDBN} command is @samp{symbol-file}.
34261 @subsubheading Example
34265 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
34271 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34272 @node GDB/MI Memory Overlay Commands
34273 @section @sc{gdb/mi} Memory Overlay Commands
34275 The memory overlay commands are not implemented.
34277 @c @subheading -overlay-auto
34279 @c @subheading -overlay-list-mapping-state
34281 @c @subheading -overlay-list-overlays
34283 @c @subheading -overlay-map
34285 @c @subheading -overlay-off
34287 @c @subheading -overlay-on
34289 @c @subheading -overlay-unmap
34291 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34292 @node GDB/MI Signal Handling Commands
34293 @section @sc{gdb/mi} Signal Handling Commands
34295 Signal handling commands are not implemented.
34297 @c @subheading -signal-handle
34299 @c @subheading -signal-list-handle-actions
34301 @c @subheading -signal-list-signal-types
34305 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34306 @node GDB/MI Target Manipulation
34307 @section @sc{gdb/mi} Target Manipulation Commands
34310 @subheading The @code{-target-attach} Command
34311 @findex -target-attach
34313 @subsubheading Synopsis
34316 -target-attach @var{pid} | @var{gid} | @var{file}
34319 Attach to a process @var{pid} or a file @var{file} outside of
34320 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
34321 group, the id previously returned by
34322 @samp{-list-thread-groups --available} must be used.
34324 @subsubheading @value{GDBN} Command
34326 The corresponding @value{GDBN} command is @samp{attach}.
34328 @subsubheading Example
34332 =thread-created,id="1"
34333 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
34339 @subheading The @code{-target-compare-sections} Command
34340 @findex -target-compare-sections
34342 @subsubheading Synopsis
34345 -target-compare-sections [ @var{section} ]
34348 Compare data of section @var{section} on target to the exec file.
34349 Without the argument, all sections are compared.
34351 @subsubheading @value{GDBN} Command
34353 The @value{GDBN} equivalent is @samp{compare-sections}.
34355 @subsubheading Example
34360 @subheading The @code{-target-detach} Command
34361 @findex -target-detach
34363 @subsubheading Synopsis
34366 -target-detach [ @var{pid} | @var{gid} ]
34369 Detach from the remote target which normally resumes its execution.
34370 If either @var{pid} or @var{gid} is specified, detaches from either
34371 the specified process, or specified thread group. There's no output.
34373 @subsubheading @value{GDBN} Command
34375 The corresponding @value{GDBN} command is @samp{detach}.
34377 @subsubheading Example
34387 @subheading The @code{-target-disconnect} Command
34388 @findex -target-disconnect
34390 @subsubheading Synopsis
34396 Disconnect from the remote target. There's no output and the target is
34397 generally not resumed.
34399 @subsubheading @value{GDBN} Command
34401 The corresponding @value{GDBN} command is @samp{disconnect}.
34403 @subsubheading Example
34413 @subheading The @code{-target-download} Command
34414 @findex -target-download
34416 @subsubheading Synopsis
34422 Loads the executable onto the remote target.
34423 It prints out an update message every half second, which includes the fields:
34427 The name of the section.
34429 The size of what has been sent so far for that section.
34431 The size of the section.
34433 The total size of what was sent so far (the current and the previous sections).
34435 The size of the overall executable to download.
34439 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
34440 @sc{gdb/mi} Output Syntax}).
34442 In addition, it prints the name and size of the sections, as they are
34443 downloaded. These messages include the following fields:
34447 The name of the section.
34449 The size of the section.
34451 The size of the overall executable to download.
34455 At the end, a summary is printed.
34457 @subsubheading @value{GDBN} Command
34459 The corresponding @value{GDBN} command is @samp{load}.
34461 @subsubheading Example
34463 Note: each status message appears on a single line. Here the messages
34464 have been broken down so that they can fit onto a page.
34469 +download,@{section=".text",section-size="6668",total-size="9880"@}
34470 +download,@{section=".text",section-sent="512",section-size="6668",
34471 total-sent="512",total-size="9880"@}
34472 +download,@{section=".text",section-sent="1024",section-size="6668",
34473 total-sent="1024",total-size="9880"@}
34474 +download,@{section=".text",section-sent="1536",section-size="6668",
34475 total-sent="1536",total-size="9880"@}
34476 +download,@{section=".text",section-sent="2048",section-size="6668",
34477 total-sent="2048",total-size="9880"@}
34478 +download,@{section=".text",section-sent="2560",section-size="6668",
34479 total-sent="2560",total-size="9880"@}
34480 +download,@{section=".text",section-sent="3072",section-size="6668",
34481 total-sent="3072",total-size="9880"@}
34482 +download,@{section=".text",section-sent="3584",section-size="6668",
34483 total-sent="3584",total-size="9880"@}
34484 +download,@{section=".text",section-sent="4096",section-size="6668",
34485 total-sent="4096",total-size="9880"@}
34486 +download,@{section=".text",section-sent="4608",section-size="6668",
34487 total-sent="4608",total-size="9880"@}
34488 +download,@{section=".text",section-sent="5120",section-size="6668",
34489 total-sent="5120",total-size="9880"@}
34490 +download,@{section=".text",section-sent="5632",section-size="6668",
34491 total-sent="5632",total-size="9880"@}
34492 +download,@{section=".text",section-sent="6144",section-size="6668",
34493 total-sent="6144",total-size="9880"@}
34494 +download,@{section=".text",section-sent="6656",section-size="6668",
34495 total-sent="6656",total-size="9880"@}
34496 +download,@{section=".init",section-size="28",total-size="9880"@}
34497 +download,@{section=".fini",section-size="28",total-size="9880"@}
34498 +download,@{section=".data",section-size="3156",total-size="9880"@}
34499 +download,@{section=".data",section-sent="512",section-size="3156",
34500 total-sent="7236",total-size="9880"@}
34501 +download,@{section=".data",section-sent="1024",section-size="3156",
34502 total-sent="7748",total-size="9880"@}
34503 +download,@{section=".data",section-sent="1536",section-size="3156",
34504 total-sent="8260",total-size="9880"@}
34505 +download,@{section=".data",section-sent="2048",section-size="3156",
34506 total-sent="8772",total-size="9880"@}
34507 +download,@{section=".data",section-sent="2560",section-size="3156",
34508 total-sent="9284",total-size="9880"@}
34509 +download,@{section=".data",section-sent="3072",section-size="3156",
34510 total-sent="9796",total-size="9880"@}
34511 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
34518 @subheading The @code{-target-exec-status} Command
34519 @findex -target-exec-status
34521 @subsubheading Synopsis
34524 -target-exec-status
34527 Provide information on the state of the target (whether it is running or
34528 not, for instance).
34530 @subsubheading @value{GDBN} Command
34532 There's no equivalent @value{GDBN} command.
34534 @subsubheading Example
34538 @subheading The @code{-target-list-available-targets} Command
34539 @findex -target-list-available-targets
34541 @subsubheading Synopsis
34544 -target-list-available-targets
34547 List the possible targets to connect to.
34549 @subsubheading @value{GDBN} Command
34551 The corresponding @value{GDBN} command is @samp{help target}.
34553 @subsubheading Example
34557 @subheading The @code{-target-list-current-targets} Command
34558 @findex -target-list-current-targets
34560 @subsubheading Synopsis
34563 -target-list-current-targets
34566 Describe the current target.
34568 @subsubheading @value{GDBN} Command
34570 The corresponding information is printed by @samp{info file} (among
34573 @subsubheading Example
34577 @subheading The @code{-target-list-parameters} Command
34578 @findex -target-list-parameters
34580 @subsubheading Synopsis
34583 -target-list-parameters
34589 @subsubheading @value{GDBN} Command
34593 @subsubheading Example
34597 @subheading The @code{-target-select} Command
34598 @findex -target-select
34600 @subsubheading Synopsis
34603 -target-select @var{type} @var{parameters @dots{}}
34606 Connect @value{GDBN} to the remote target. This command takes two args:
34610 The type of target, for instance @samp{remote}, etc.
34611 @item @var{parameters}
34612 Device names, host names and the like. @xref{Target Commands, ,
34613 Commands for Managing Targets}, for more details.
34616 The output is a connection notification, followed by the address at
34617 which the target program is, in the following form:
34620 ^connected,addr="@var{address}",func="@var{function name}",
34621 args=[@var{arg list}]
34624 @subsubheading @value{GDBN} Command
34626 The corresponding @value{GDBN} command is @samp{target}.
34628 @subsubheading Example
34632 -target-select remote /dev/ttya
34633 ^connected,addr="0xfe00a300",func="??",args=[]
34637 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34638 @node GDB/MI File Transfer Commands
34639 @section @sc{gdb/mi} File Transfer Commands
34642 @subheading The @code{-target-file-put} Command
34643 @findex -target-file-put
34645 @subsubheading Synopsis
34648 -target-file-put @var{hostfile} @var{targetfile}
34651 Copy file @var{hostfile} from the host system (the machine running
34652 @value{GDBN}) to @var{targetfile} on the target system.
34654 @subsubheading @value{GDBN} Command
34656 The corresponding @value{GDBN} command is @samp{remote put}.
34658 @subsubheading Example
34662 -target-file-put localfile remotefile
34668 @subheading The @code{-target-file-get} Command
34669 @findex -target-file-get
34671 @subsubheading Synopsis
34674 -target-file-get @var{targetfile} @var{hostfile}
34677 Copy file @var{targetfile} from the target system to @var{hostfile}
34678 on the host system.
34680 @subsubheading @value{GDBN} Command
34682 The corresponding @value{GDBN} command is @samp{remote get}.
34684 @subsubheading Example
34688 -target-file-get remotefile localfile
34694 @subheading The @code{-target-file-delete} Command
34695 @findex -target-file-delete
34697 @subsubheading Synopsis
34700 -target-file-delete @var{targetfile}
34703 Delete @var{targetfile} from the target system.
34705 @subsubheading @value{GDBN} Command
34707 The corresponding @value{GDBN} command is @samp{remote delete}.
34709 @subsubheading Example
34713 -target-file-delete remotefile
34719 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34720 @node GDB/MI Miscellaneous Commands
34721 @section Miscellaneous @sc{gdb/mi} Commands
34723 @c @subheading -gdb-complete
34725 @subheading The @code{-gdb-exit} Command
34728 @subsubheading Synopsis
34734 Exit @value{GDBN} immediately.
34736 @subsubheading @value{GDBN} Command
34738 Approximately corresponds to @samp{quit}.
34740 @subsubheading Example
34750 @subheading The @code{-exec-abort} Command
34751 @findex -exec-abort
34753 @subsubheading Synopsis
34759 Kill the inferior running program.
34761 @subsubheading @value{GDBN} Command
34763 The corresponding @value{GDBN} command is @samp{kill}.
34765 @subsubheading Example
34770 @subheading The @code{-gdb-set} Command
34773 @subsubheading Synopsis
34779 Set an internal @value{GDBN} variable.
34780 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
34782 @subsubheading @value{GDBN} Command
34784 The corresponding @value{GDBN} command is @samp{set}.
34786 @subsubheading Example
34796 @subheading The @code{-gdb-show} Command
34799 @subsubheading Synopsis
34805 Show the current value of a @value{GDBN} variable.
34807 @subsubheading @value{GDBN} Command
34809 The corresponding @value{GDBN} command is @samp{show}.
34811 @subsubheading Example
34820 @c @subheading -gdb-source
34823 @subheading The @code{-gdb-version} Command
34824 @findex -gdb-version
34826 @subsubheading Synopsis
34832 Show version information for @value{GDBN}. Used mostly in testing.
34834 @subsubheading @value{GDBN} Command
34836 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
34837 default shows this information when you start an interactive session.
34839 @subsubheading Example
34841 @c This example modifies the actual output from GDB to avoid overfull
34847 ~Copyright 2000 Free Software Foundation, Inc.
34848 ~GDB is free software, covered by the GNU General Public License, and
34849 ~you are welcome to change it and/or distribute copies of it under
34850 ~ certain conditions.
34851 ~Type "show copying" to see the conditions.
34852 ~There is absolutely no warranty for GDB. Type "show warranty" for
34854 ~This GDB was configured as
34855 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
34860 @subheading The @code{-list-features} Command
34861 @findex -list-features
34863 Returns a list of particular features of the MI protocol that
34864 this version of gdb implements. A feature can be a command,
34865 or a new field in an output of some command, or even an
34866 important bugfix. While a frontend can sometimes detect presence
34867 of a feature at runtime, it is easier to perform detection at debugger
34870 The command returns a list of strings, with each string naming an
34871 available feature. Each returned string is just a name, it does not
34872 have any internal structure. The list of possible feature names
34878 (gdb) -list-features
34879 ^done,result=["feature1","feature2"]
34882 The current list of features is:
34885 @item frozen-varobjs
34886 Indicates support for the @code{-var-set-frozen} command, as well
34887 as possible presense of the @code{frozen} field in the output
34888 of @code{-varobj-create}.
34889 @item pending-breakpoints
34890 Indicates support for the @option{-f} option to the @code{-break-insert}
34893 Indicates Python scripting support, Python-based
34894 pretty-printing commands, and possible presence of the
34895 @samp{display_hint} field in the output of @code{-var-list-children}
34897 Indicates support for the @code{-thread-info} command.
34898 @item data-read-memory-bytes
34899 Indicates support for the @code{-data-read-memory-bytes} and the
34900 @code{-data-write-memory-bytes} commands.
34901 @item breakpoint-notifications
34902 Indicates that changes to breakpoints and breakpoints created via the
34903 CLI will be announced via async records.
34904 @item ada-task-info
34905 Indicates support for the @code{-ada-task-info} command.
34908 @subheading The @code{-list-target-features} Command
34909 @findex -list-target-features
34911 Returns a list of particular features that are supported by the
34912 target. Those features affect the permitted MI commands, but
34913 unlike the features reported by the @code{-list-features} command, the
34914 features depend on which target GDB is using at the moment. Whenever
34915 a target can change, due to commands such as @code{-target-select},
34916 @code{-target-attach} or @code{-exec-run}, the list of target features
34917 may change, and the frontend should obtain it again.
34921 (gdb) -list-target-features
34922 ^done,result=["async"]
34925 The current list of features is:
34929 Indicates that the target is capable of asynchronous command
34930 execution, which means that @value{GDBN} will accept further commands
34931 while the target is running.
34934 Indicates that the target is capable of reverse execution.
34935 @xref{Reverse Execution}, for more information.
34939 @subheading The @code{-list-thread-groups} Command
34940 @findex -list-thread-groups
34942 @subheading Synopsis
34945 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
34948 Lists thread groups (@pxref{Thread groups}). When a single thread
34949 group is passed as the argument, lists the children of that group.
34950 When several thread group are passed, lists information about those
34951 thread groups. Without any parameters, lists information about all
34952 top-level thread groups.
34954 Normally, thread groups that are being debugged are reported.
34955 With the @samp{--available} option, @value{GDBN} reports thread groups
34956 available on the target.
34958 The output of this command may have either a @samp{threads} result or
34959 a @samp{groups} result. The @samp{thread} result has a list of tuples
34960 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
34961 Information}). The @samp{groups} result has a list of tuples as value,
34962 each tuple describing a thread group. If top-level groups are
34963 requested (that is, no parameter is passed), or when several groups
34964 are passed, the output always has a @samp{groups} result. The format
34965 of the @samp{group} result is described below.
34967 To reduce the number of roundtrips it's possible to list thread groups
34968 together with their children, by passing the @samp{--recurse} option
34969 and the recursion depth. Presently, only recursion depth of 1 is
34970 permitted. If this option is present, then every reported thread group
34971 will also include its children, either as @samp{group} or
34972 @samp{threads} field.
34974 In general, any combination of option and parameters is permitted, with
34975 the following caveats:
34979 When a single thread group is passed, the output will typically
34980 be the @samp{threads} result. Because threads may not contain
34981 anything, the @samp{recurse} option will be ignored.
34984 When the @samp{--available} option is passed, limited information may
34985 be available. In particular, the list of threads of a process might
34986 be inaccessible. Further, specifying specific thread groups might
34987 not give any performance advantage over listing all thread groups.
34988 The frontend should assume that @samp{-list-thread-groups --available}
34989 is always an expensive operation and cache the results.
34993 The @samp{groups} result is a list of tuples, where each tuple may
34994 have the following fields:
34998 Identifier of the thread group. This field is always present.
34999 The identifier is an opaque string; frontends should not try to
35000 convert it to an integer, even though it might look like one.
35003 The type of the thread group. At present, only @samp{process} is a
35007 The target-specific process identifier. This field is only present
35008 for thread groups of type @samp{process} and only if the process exists.
35011 The number of children this thread group has. This field may be
35012 absent for an available thread group.
35015 This field has a list of tuples as value, each tuple describing a
35016 thread. It may be present if the @samp{--recurse} option is
35017 specified, and it's actually possible to obtain the threads.
35020 This field is a list of integers, each identifying a core that one
35021 thread of the group is running on. This field may be absent if
35022 such information is not available.
35025 The name of the executable file that corresponds to this thread group.
35026 The field is only present for thread groups of type @samp{process},
35027 and only if there is a corresponding executable file.
35031 @subheading Example
35035 -list-thread-groups
35036 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
35037 -list-thread-groups 17
35038 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
35039 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
35040 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
35041 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
35042 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
35043 -list-thread-groups --available
35044 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
35045 -list-thread-groups --available --recurse 1
35046 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
35047 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
35048 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
35049 -list-thread-groups --available --recurse 1 17 18
35050 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
35051 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
35052 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
35055 @subheading The @code{-info-os} Command
35058 @subsubheading Synopsis
35061 -info-os [ @var{type} ]
35064 If no argument is supplied, the command returns a table of available
35065 operating-system-specific information types. If one of these types is
35066 supplied as an argument @var{type}, then the command returns a table
35067 of data of that type.
35069 The types of information available depend on the target operating
35072 @subsubheading @value{GDBN} Command
35074 The corresponding @value{GDBN} command is @samp{info os}.
35076 @subsubheading Example
35078 When run on a @sc{gnu}/Linux system, the output will look something
35084 ^done,OSDataTable=@{nr_rows="9",nr_cols="3",
35085 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
35086 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
35087 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
35088 body=[item=@{col0="processes",col1="Listing of all processes",
35089 col2="Processes"@},
35090 item=@{col0="procgroups",col1="Listing of all process groups",
35091 col2="Process groups"@},
35092 item=@{col0="threads",col1="Listing of all threads",
35094 item=@{col0="files",col1="Listing of all file descriptors",
35095 col2="File descriptors"@},
35096 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
35098 item=@{col0="shm",col1="Listing of all shared-memory regions",
35099 col2="Shared-memory regions"@},
35100 item=@{col0="semaphores",col1="Listing of all semaphores",
35101 col2="Semaphores"@},
35102 item=@{col0="msg",col1="Listing of all message queues",
35103 col2="Message queues"@},
35104 item=@{col0="modules",col1="Listing of all loaded kernel modules",
35105 col2="Kernel modules"@}]@}
35108 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
35109 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
35110 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
35111 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
35112 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
35113 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
35114 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
35115 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
35117 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
35118 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
35122 (Note that the MI output here includes a @code{"Title"} column that
35123 does not appear in command-line @code{info os}; this column is useful
35124 for MI clients that want to enumerate the types of data, such as in a
35125 popup menu, but is needless clutter on the command line, and
35126 @code{info os} omits it.)
35128 @subheading The @code{-add-inferior} Command
35129 @findex -add-inferior
35131 @subheading Synopsis
35137 Creates a new inferior (@pxref{Inferiors and Programs}). The created
35138 inferior is not associated with any executable. Such association may
35139 be established with the @samp{-file-exec-and-symbols} command
35140 (@pxref{GDB/MI File Commands}). The command response has a single
35141 field, @samp{inferior}, whose value is the identifier of the
35142 thread group corresponding to the new inferior.
35144 @subheading Example
35149 ^done,inferior="i3"
35152 @subheading The @code{-interpreter-exec} Command
35153 @findex -interpreter-exec
35155 @subheading Synopsis
35158 -interpreter-exec @var{interpreter} @var{command}
35160 @anchor{-interpreter-exec}
35162 Execute the specified @var{command} in the given @var{interpreter}.
35164 @subheading @value{GDBN} Command
35166 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
35168 @subheading Example
35172 -interpreter-exec console "break main"
35173 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
35174 &"During symbol reading, bad structure-type format.\n"
35175 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
35180 @subheading The @code{-inferior-tty-set} Command
35181 @findex -inferior-tty-set
35183 @subheading Synopsis
35186 -inferior-tty-set /dev/pts/1
35189 Set terminal for future runs of the program being debugged.
35191 @subheading @value{GDBN} Command
35193 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
35195 @subheading Example
35199 -inferior-tty-set /dev/pts/1
35204 @subheading The @code{-inferior-tty-show} Command
35205 @findex -inferior-tty-show
35207 @subheading Synopsis
35213 Show terminal for future runs of program being debugged.
35215 @subheading @value{GDBN} Command
35217 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
35219 @subheading Example
35223 -inferior-tty-set /dev/pts/1
35227 ^done,inferior_tty_terminal="/dev/pts/1"
35231 @subheading The @code{-enable-timings} Command
35232 @findex -enable-timings
35234 @subheading Synopsis
35237 -enable-timings [yes | no]
35240 Toggle the printing of the wallclock, user and system times for an MI
35241 command as a field in its output. This command is to help frontend
35242 developers optimize the performance of their code. No argument is
35243 equivalent to @samp{yes}.
35245 @subheading @value{GDBN} Command
35249 @subheading Example
35257 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
35258 addr="0x080484ed",func="main",file="myprog.c",
35259 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
35261 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
35269 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
35270 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
35271 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
35272 fullname="/home/nickrob/myprog.c",line="73"@}
35277 @chapter @value{GDBN} Annotations
35279 This chapter describes annotations in @value{GDBN}. Annotations were
35280 designed to interface @value{GDBN} to graphical user interfaces or other
35281 similar programs which want to interact with @value{GDBN} at a
35282 relatively high level.
35284 The annotation mechanism has largely been superseded by @sc{gdb/mi}
35288 This is Edition @value{EDITION}, @value{DATE}.
35292 * Annotations Overview:: What annotations are; the general syntax.
35293 * Server Prefix:: Issuing a command without affecting user state.
35294 * Prompting:: Annotations marking @value{GDBN}'s need for input.
35295 * Errors:: Annotations for error messages.
35296 * Invalidation:: Some annotations describe things now invalid.
35297 * Annotations for Running::
35298 Whether the program is running, how it stopped, etc.
35299 * Source Annotations:: Annotations describing source code.
35302 @node Annotations Overview
35303 @section What is an Annotation?
35304 @cindex annotations
35306 Annotations start with a newline character, two @samp{control-z}
35307 characters, and the name of the annotation. If there is no additional
35308 information associated with this annotation, the name of the annotation
35309 is followed immediately by a newline. If there is additional
35310 information, the name of the annotation is followed by a space, the
35311 additional information, and a newline. The additional information
35312 cannot contain newline characters.
35314 Any output not beginning with a newline and two @samp{control-z}
35315 characters denotes literal output from @value{GDBN}. Currently there is
35316 no need for @value{GDBN} to output a newline followed by two
35317 @samp{control-z} characters, but if there was such a need, the
35318 annotations could be extended with an @samp{escape} annotation which
35319 means those three characters as output.
35321 The annotation @var{level}, which is specified using the
35322 @option{--annotate} command line option (@pxref{Mode Options}), controls
35323 how much information @value{GDBN} prints together with its prompt,
35324 values of expressions, source lines, and other types of output. Level 0
35325 is for no annotations, level 1 is for use when @value{GDBN} is run as a
35326 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
35327 for programs that control @value{GDBN}, and level 2 annotations have
35328 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
35329 Interface, annotate, GDB's Obsolete Annotations}).
35332 @kindex set annotate
35333 @item set annotate @var{level}
35334 The @value{GDBN} command @code{set annotate} sets the level of
35335 annotations to the specified @var{level}.
35337 @item show annotate
35338 @kindex show annotate
35339 Show the current annotation level.
35342 This chapter describes level 3 annotations.
35344 A simple example of starting up @value{GDBN} with annotations is:
35347 $ @kbd{gdb --annotate=3}
35349 Copyright 2003 Free Software Foundation, Inc.
35350 GDB is free software, covered by the GNU General Public License,
35351 and you are welcome to change it and/or distribute copies of it
35352 under certain conditions.
35353 Type "show copying" to see the conditions.
35354 There is absolutely no warranty for GDB. Type "show warranty"
35356 This GDB was configured as "i386-pc-linux-gnu"
35367 Here @samp{quit} is input to @value{GDBN}; the rest is output from
35368 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
35369 denotes a @samp{control-z} character) are annotations; the rest is
35370 output from @value{GDBN}.
35372 @node Server Prefix
35373 @section The Server Prefix
35374 @cindex server prefix
35376 If you prefix a command with @samp{server } then it will not affect
35377 the command history, nor will it affect @value{GDBN}'s notion of which
35378 command to repeat if @key{RET} is pressed on a line by itself. This
35379 means that commands can be run behind a user's back by a front-end in
35380 a transparent manner.
35382 The @code{server } prefix does not affect the recording of values into
35383 the value history; to print a value without recording it into the
35384 value history, use the @code{output} command instead of the
35385 @code{print} command.
35387 Using this prefix also disables confirmation requests
35388 (@pxref{confirmation requests}).
35391 @section Annotation for @value{GDBN} Input
35393 @cindex annotations for prompts
35394 When @value{GDBN} prompts for input, it annotates this fact so it is possible
35395 to know when to send output, when the output from a given command is
35398 Different kinds of input each have a different @dfn{input type}. Each
35399 input type has three annotations: a @code{pre-} annotation, which
35400 denotes the beginning of any prompt which is being output, a plain
35401 annotation, which denotes the end of the prompt, and then a @code{post-}
35402 annotation which denotes the end of any echo which may (or may not) be
35403 associated with the input. For example, the @code{prompt} input type
35404 features the following annotations:
35412 The input types are
35415 @findex pre-prompt annotation
35416 @findex prompt annotation
35417 @findex post-prompt annotation
35419 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
35421 @findex pre-commands annotation
35422 @findex commands annotation
35423 @findex post-commands annotation
35425 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
35426 command. The annotations are repeated for each command which is input.
35428 @findex pre-overload-choice annotation
35429 @findex overload-choice annotation
35430 @findex post-overload-choice annotation
35431 @item overload-choice
35432 When @value{GDBN} wants the user to select between various overloaded functions.
35434 @findex pre-query annotation
35435 @findex query annotation
35436 @findex post-query annotation
35438 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
35440 @findex pre-prompt-for-continue annotation
35441 @findex prompt-for-continue annotation
35442 @findex post-prompt-for-continue annotation
35443 @item prompt-for-continue
35444 When @value{GDBN} is asking the user to press return to continue. Note: Don't
35445 expect this to work well; instead use @code{set height 0} to disable
35446 prompting. This is because the counting of lines is buggy in the
35447 presence of annotations.
35452 @cindex annotations for errors, warnings and interrupts
35454 @findex quit annotation
35459 This annotation occurs right before @value{GDBN} responds to an interrupt.
35461 @findex error annotation
35466 This annotation occurs right before @value{GDBN} responds to an error.
35468 Quit and error annotations indicate that any annotations which @value{GDBN} was
35469 in the middle of may end abruptly. For example, if a
35470 @code{value-history-begin} annotation is followed by a @code{error}, one
35471 cannot expect to receive the matching @code{value-history-end}. One
35472 cannot expect not to receive it either, however; an error annotation
35473 does not necessarily mean that @value{GDBN} is immediately returning all the way
35476 @findex error-begin annotation
35477 A quit or error annotation may be preceded by
35483 Any output between that and the quit or error annotation is the error
35486 Warning messages are not yet annotated.
35487 @c If we want to change that, need to fix warning(), type_error(),
35488 @c range_error(), and possibly other places.
35491 @section Invalidation Notices
35493 @cindex annotations for invalidation messages
35494 The following annotations say that certain pieces of state may have
35498 @findex frames-invalid annotation
35499 @item ^Z^Zframes-invalid
35501 The frames (for example, output from the @code{backtrace} command) may
35504 @findex breakpoints-invalid annotation
35505 @item ^Z^Zbreakpoints-invalid
35507 The breakpoints may have changed. For example, the user just added or
35508 deleted a breakpoint.
35511 @node Annotations for Running
35512 @section Running the Program
35513 @cindex annotations for running programs
35515 @findex starting annotation
35516 @findex stopping annotation
35517 When the program starts executing due to a @value{GDBN} command such as
35518 @code{step} or @code{continue},
35524 is output. When the program stops,
35530 is output. Before the @code{stopped} annotation, a variety of
35531 annotations describe how the program stopped.
35534 @findex exited annotation
35535 @item ^Z^Zexited @var{exit-status}
35536 The program exited, and @var{exit-status} is the exit status (zero for
35537 successful exit, otherwise nonzero).
35539 @findex signalled annotation
35540 @findex signal-name annotation
35541 @findex signal-name-end annotation
35542 @findex signal-string annotation
35543 @findex signal-string-end annotation
35544 @item ^Z^Zsignalled
35545 The program exited with a signal. After the @code{^Z^Zsignalled}, the
35546 annotation continues:
35552 ^Z^Zsignal-name-end
35556 ^Z^Zsignal-string-end
35561 where @var{name} is the name of the signal, such as @code{SIGILL} or
35562 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
35563 as @code{Illegal Instruction} or @code{Segmentation fault}.
35564 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
35565 user's benefit and have no particular format.
35567 @findex signal annotation
35569 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
35570 just saying that the program received the signal, not that it was
35571 terminated with it.
35573 @findex breakpoint annotation
35574 @item ^Z^Zbreakpoint @var{number}
35575 The program hit breakpoint number @var{number}.
35577 @findex watchpoint annotation
35578 @item ^Z^Zwatchpoint @var{number}
35579 The program hit watchpoint number @var{number}.
35582 @node Source Annotations
35583 @section Displaying Source
35584 @cindex annotations for source display
35586 @findex source annotation
35587 The following annotation is used instead of displaying source code:
35590 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
35593 where @var{filename} is an absolute file name indicating which source
35594 file, @var{line} is the line number within that file (where 1 is the
35595 first line in the file), @var{character} is the character position
35596 within the file (where 0 is the first character in the file) (for most
35597 debug formats this will necessarily point to the beginning of a line),
35598 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
35599 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
35600 @var{addr} is the address in the target program associated with the
35601 source which is being displayed. @var{addr} is in the form @samp{0x}
35602 followed by one or more lowercase hex digits (note that this does not
35603 depend on the language).
35605 @node JIT Interface
35606 @chapter JIT Compilation Interface
35607 @cindex just-in-time compilation
35608 @cindex JIT compilation interface
35610 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
35611 interface. A JIT compiler is a program or library that generates native
35612 executable code at runtime and executes it, usually in order to achieve good
35613 performance while maintaining platform independence.
35615 Programs that use JIT compilation are normally difficult to debug because
35616 portions of their code are generated at runtime, instead of being loaded from
35617 object files, which is where @value{GDBN} normally finds the program's symbols
35618 and debug information. In order to debug programs that use JIT compilation,
35619 @value{GDBN} has an interface that allows the program to register in-memory
35620 symbol files with @value{GDBN} at runtime.
35622 If you are using @value{GDBN} to debug a program that uses this interface, then
35623 it should work transparently so long as you have not stripped the binary. If
35624 you are developing a JIT compiler, then the interface is documented in the rest
35625 of this chapter. At this time, the only known client of this interface is the
35628 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
35629 JIT compiler communicates with @value{GDBN} by writing data into a global
35630 variable and calling a fuction at a well-known symbol. When @value{GDBN}
35631 attaches, it reads a linked list of symbol files from the global variable to
35632 find existing code, and puts a breakpoint in the function so that it can find
35633 out about additional code.
35636 * Declarations:: Relevant C struct declarations
35637 * Registering Code:: Steps to register code
35638 * Unregistering Code:: Steps to unregister code
35639 * Custom Debug Info:: Emit debug information in a custom format
35643 @section JIT Declarations
35645 These are the relevant struct declarations that a C program should include to
35646 implement the interface:
35656 struct jit_code_entry
35658 struct jit_code_entry *next_entry;
35659 struct jit_code_entry *prev_entry;
35660 const char *symfile_addr;
35661 uint64_t symfile_size;
35664 struct jit_descriptor
35667 /* This type should be jit_actions_t, but we use uint32_t
35668 to be explicit about the bitwidth. */
35669 uint32_t action_flag;
35670 struct jit_code_entry *relevant_entry;
35671 struct jit_code_entry *first_entry;
35674 /* GDB puts a breakpoint in this function. */
35675 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
35677 /* Make sure to specify the version statically, because the
35678 debugger may check the version before we can set it. */
35679 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
35682 If the JIT is multi-threaded, then it is important that the JIT synchronize any
35683 modifications to this global data properly, which can easily be done by putting
35684 a global mutex around modifications to these structures.
35686 @node Registering Code
35687 @section Registering Code
35689 To register code with @value{GDBN}, the JIT should follow this protocol:
35693 Generate an object file in memory with symbols and other desired debug
35694 information. The file must include the virtual addresses of the sections.
35697 Create a code entry for the file, which gives the start and size of the symbol
35701 Add it to the linked list in the JIT descriptor.
35704 Point the relevant_entry field of the descriptor at the entry.
35707 Set @code{action_flag} to @code{JIT_REGISTER} and call
35708 @code{__jit_debug_register_code}.
35711 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
35712 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
35713 new code. However, the linked list must still be maintained in order to allow
35714 @value{GDBN} to attach to a running process and still find the symbol files.
35716 @node Unregistering Code
35717 @section Unregistering Code
35719 If code is freed, then the JIT should use the following protocol:
35723 Remove the code entry corresponding to the code from the linked list.
35726 Point the @code{relevant_entry} field of the descriptor at the code entry.
35729 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
35730 @code{__jit_debug_register_code}.
35733 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
35734 and the JIT will leak the memory used for the associated symbol files.
35736 @node Custom Debug Info
35737 @section Custom Debug Info
35738 @cindex custom JIT debug info
35739 @cindex JIT debug info reader
35741 Generating debug information in platform-native file formats (like ELF
35742 or COFF) may be an overkill for JIT compilers; especially if all the
35743 debug info is used for is displaying a meaningful backtrace. The
35744 issue can be resolved by having the JIT writers decide on a debug info
35745 format and also provide a reader that parses the debug info generated
35746 by the JIT compiler. This section gives a brief overview on writing
35747 such a parser. More specific details can be found in the source file
35748 @file{gdb/jit-reader.in}, which is also installed as a header at
35749 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
35751 The reader is implemented as a shared object (so this functionality is
35752 not available on platforms which don't allow loading shared objects at
35753 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
35754 @code{jit-reader-unload} are provided, to be used to load and unload
35755 the readers from a preconfigured directory. Once loaded, the shared
35756 object is used the parse the debug information emitted by the JIT
35760 * Using JIT Debug Info Readers:: How to use supplied readers correctly
35761 * Writing JIT Debug Info Readers:: Creating a debug-info reader
35764 @node Using JIT Debug Info Readers
35765 @subsection Using JIT Debug Info Readers
35766 @kindex jit-reader-load
35767 @kindex jit-reader-unload
35769 Readers can be loaded and unloaded using the @code{jit-reader-load}
35770 and @code{jit-reader-unload} commands.
35773 @item jit-reader-load @var{reader}
35774 Load the JIT reader named @var{reader}. @var{reader} is a shared
35775 object specified as either an absolute or a relative file name. In
35776 the latter case, @value{GDBN} will try to load the reader from a
35777 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
35778 system (here @var{libdir} is the system library directory, often
35779 @file{/usr/local/lib}).
35781 Only one reader can be active at a time; trying to load a second
35782 reader when one is already loaded will result in @value{GDBN}
35783 reporting an error. A new JIT reader can be loaded by first unloading
35784 the current one using @code{jit-reader-unload} and then invoking
35785 @code{jit-reader-load}.
35787 @item jit-reader-unload
35788 Unload the currently loaded JIT reader.
35792 @node Writing JIT Debug Info Readers
35793 @subsection Writing JIT Debug Info Readers
35794 @cindex writing JIT debug info readers
35796 As mentioned, a reader is essentially a shared object conforming to a
35797 certain ABI. This ABI is described in @file{jit-reader.h}.
35799 @file{jit-reader.h} defines the structures, macros and functions
35800 required to write a reader. It is installed (along with
35801 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
35802 the system include directory.
35804 Readers need to be released under a GPL compatible license. A reader
35805 can be declared as released under such a license by placing the macro
35806 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
35808 The entry point for readers is the symbol @code{gdb_init_reader},
35809 which is expected to be a function with the prototype
35811 @findex gdb_init_reader
35813 extern struct gdb_reader_funcs *gdb_init_reader (void);
35816 @cindex @code{struct gdb_reader_funcs}
35818 @code{struct gdb_reader_funcs} contains a set of pointers to callback
35819 functions. These functions are executed to read the debug info
35820 generated by the JIT compiler (@code{read}), to unwind stack frames
35821 (@code{unwind}) and to create canonical frame IDs
35822 (@code{get_Frame_id}). It also has a callback that is called when the
35823 reader is being unloaded (@code{destroy}). The struct looks like this
35826 struct gdb_reader_funcs
35828 /* Must be set to GDB_READER_INTERFACE_VERSION. */
35829 int reader_version;
35831 /* For use by the reader. */
35834 gdb_read_debug_info *read;
35835 gdb_unwind_frame *unwind;
35836 gdb_get_frame_id *get_frame_id;
35837 gdb_destroy_reader *destroy;
35841 @cindex @code{struct gdb_symbol_callbacks}
35842 @cindex @code{struct gdb_unwind_callbacks}
35844 The callbacks are provided with another set of callbacks by
35845 @value{GDBN} to do their job. For @code{read}, these callbacks are
35846 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
35847 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
35848 @code{struct gdb_symbol_callbacks} has callbacks to create new object
35849 files and new symbol tables inside those object files. @code{struct
35850 gdb_unwind_callbacks} has callbacks to read registers off the current
35851 frame and to write out the values of the registers in the previous
35852 frame. Both have a callback (@code{target_read}) to read bytes off the
35853 target's address space.
35855 @node In-Process Agent
35856 @chapter In-Process Agent
35857 @cindex debugging agent
35858 The traditional debugging model is conceptually low-speed, but works fine,
35859 because most bugs can be reproduced in debugging-mode execution. However,
35860 as multi-core or many-core processors are becoming mainstream, and
35861 multi-threaded programs become more and more popular, there should be more
35862 and more bugs that only manifest themselves at normal-mode execution, for
35863 example, thread races, because debugger's interference with the program's
35864 timing may conceal the bugs. On the other hand, in some applications,
35865 it is not feasible for the debugger to interrupt the program's execution
35866 long enough for the developer to learn anything helpful about its behavior.
35867 If the program's correctness depends on its real-time behavior, delays
35868 introduced by a debugger might cause the program to fail, even when the
35869 code itself is correct. It is useful to be able to observe the program's
35870 behavior without interrupting it.
35872 Therefore, traditional debugging model is too intrusive to reproduce
35873 some bugs. In order to reduce the interference with the program, we can
35874 reduce the number of operations performed by debugger. The
35875 @dfn{In-Process Agent}, a shared library, is running within the same
35876 process with inferior, and is able to perform some debugging operations
35877 itself. As a result, debugger is only involved when necessary, and
35878 performance of debugging can be improved accordingly. Note that
35879 interference with program can be reduced but can't be removed completely,
35880 because the in-process agent will still stop or slow down the program.
35882 The in-process agent can interpret and execute Agent Expressions
35883 (@pxref{Agent Expressions}) during performing debugging operations. The
35884 agent expressions can be used for different purposes, such as collecting
35885 data in tracepoints, and condition evaluation in breakpoints.
35887 @anchor{Control Agent}
35888 You can control whether the in-process agent is used as an aid for
35889 debugging with the following commands:
35892 @kindex set agent on
35894 Causes the in-process agent to perform some operations on behalf of the
35895 debugger. Just which operations requested by the user will be done
35896 by the in-process agent depends on the its capabilities. For example,
35897 if you request to evaluate breakpoint conditions in the in-process agent,
35898 and the in-process agent has such capability as well, then breakpoint
35899 conditions will be evaluated in the in-process agent.
35901 @kindex set agent off
35902 @item set agent off
35903 Disables execution of debugging operations by the in-process agent. All
35904 of the operations will be performed by @value{GDBN}.
35908 Display the current setting of execution of debugging operations by
35909 the in-process agent.
35913 * In-Process Agent Protocol::
35916 @node In-Process Agent Protocol
35917 @section In-Process Agent Protocol
35918 @cindex in-process agent protocol
35920 The in-process agent is able to communicate with both @value{GDBN} and
35921 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
35922 used for communications between @value{GDBN} or GDBserver and the IPA.
35923 In general, @value{GDBN} or GDBserver sends commands
35924 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
35925 in-process agent replies back with the return result of the command, or
35926 some other information. The data sent to in-process agent is composed
35927 of primitive data types, such as 4-byte or 8-byte type, and composite
35928 types, which are called objects (@pxref{IPA Protocol Objects}).
35931 * IPA Protocol Objects::
35932 * IPA Protocol Commands::
35935 @node IPA Protocol Objects
35936 @subsection IPA Protocol Objects
35937 @cindex ipa protocol objects
35939 The commands sent to and results received from agent may contain some
35940 complex data types called @dfn{objects}.
35942 The in-process agent is running on the same machine with @value{GDBN}
35943 or GDBserver, so it doesn't have to handle as much differences between
35944 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
35945 However, there are still some differences of two ends in two processes:
35949 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
35950 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
35952 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
35953 GDBserver is compiled with one, and in-process agent is compiled with
35957 Here are the IPA Protocol Objects:
35961 agent expression object. It represents an agent expression
35962 (@pxref{Agent Expressions}).
35963 @anchor{agent expression object}
35965 tracepoint action object. It represents a tracepoint action
35966 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
35967 memory, static trace data and to evaluate expression.
35968 @anchor{tracepoint action object}
35970 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
35971 @anchor{tracepoint object}
35975 The following table describes important attributes of each IPA protocol
35978 @multitable @columnfractions .30 .20 .50
35979 @headitem Name @tab Size @tab Description
35980 @item @emph{agent expression object} @tab @tab
35981 @item length @tab 4 @tab length of bytes code
35982 @item byte code @tab @var{length} @tab contents of byte code
35983 @item @emph{tracepoint action for collecting memory} @tab @tab
35984 @item 'M' @tab 1 @tab type of tracepoint action
35985 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
35986 address of the lowest byte to collect, otherwise @var{addr} is the offset
35987 of @var{basereg} for memory collecting.
35988 @item len @tab 8 @tab length of memory for collecting
35989 @item basereg @tab 4 @tab the register number containing the starting
35990 memory address for collecting.
35991 @item @emph{tracepoint action for collecting registers} @tab @tab
35992 @item 'R' @tab 1 @tab type of tracepoint action
35993 @item @emph{tracepoint action for collecting static trace data} @tab @tab
35994 @item 'L' @tab 1 @tab type of tracepoint action
35995 @item @emph{tracepoint action for expression evaluation} @tab @tab
35996 @item 'X' @tab 1 @tab type of tracepoint action
35997 @item agent expression @tab length of @tab @ref{agent expression object}
35998 @item @emph{tracepoint object} @tab @tab
35999 @item number @tab 4 @tab number of tracepoint
36000 @item address @tab 8 @tab address of tracepoint inserted on
36001 @item type @tab 4 @tab type of tracepoint
36002 @item enabled @tab 1 @tab enable or disable of tracepoint
36003 @item step_count @tab 8 @tab step
36004 @item pass_count @tab 8 @tab pass
36005 @item numactions @tab 4 @tab number of tracepoint actions
36006 @item hit count @tab 8 @tab hit count
36007 @item trace frame usage @tab 8 @tab trace frame usage
36008 @item compiled_cond @tab 8 @tab compiled condition
36009 @item orig_size @tab 8 @tab orig size
36010 @item condition @tab 4 if condition is NULL otherwise length of
36011 @ref{agent expression object}
36012 @tab zero if condition is NULL, otherwise is
36013 @ref{agent expression object}
36014 @item actions @tab variable
36015 @tab numactions number of @ref{tracepoint action object}
36018 @node IPA Protocol Commands
36019 @subsection IPA Protocol Commands
36020 @cindex ipa protocol commands
36022 The spaces in each command are delimiters to ease reading this commands
36023 specification. They don't exist in real commands.
36027 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
36028 Installs a new fast tracepoint described by @var{tracepoint_object}
36029 (@pxref{tracepoint object}). @var{gdb_jump_pad_head}, 8-byte long, is the
36030 head of @dfn{jumppad}, which is used to jump to data collection routine
36035 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
36036 @var{target_address} is address of tracepoint in the inferior.
36037 @var{gdb_jump_pad_head} is updated head of jumppad. Both of
36038 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
36039 @var{fjump} contains a sequence of instructions jump to jumppad entry.
36040 @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
36047 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
36048 is about to kill inferiors.
36056 @item probe_marker_at:@var{address}
36057 Asks in-process agent to probe the marker at @var{address}.
36064 @item unprobe_marker_at:@var{address}
36065 Asks in-process agent to unprobe the marker at @var{address}.
36069 @chapter Reporting Bugs in @value{GDBN}
36070 @cindex bugs in @value{GDBN}
36071 @cindex reporting bugs in @value{GDBN}
36073 Your bug reports play an essential role in making @value{GDBN} reliable.
36075 Reporting a bug may help you by bringing a solution to your problem, or it
36076 may not. But in any case the principal function of a bug report is to help
36077 the entire community by making the next version of @value{GDBN} work better. Bug
36078 reports are your contribution to the maintenance of @value{GDBN}.
36080 In order for a bug report to serve its purpose, you must include the
36081 information that enables us to fix the bug.
36084 * Bug Criteria:: Have you found a bug?
36085 * Bug Reporting:: How to report bugs
36089 @section Have You Found a Bug?
36090 @cindex bug criteria
36092 If you are not sure whether you have found a bug, here are some guidelines:
36095 @cindex fatal signal
36096 @cindex debugger crash
36097 @cindex crash of debugger
36099 If the debugger gets a fatal signal, for any input whatever, that is a
36100 @value{GDBN} bug. Reliable debuggers never crash.
36102 @cindex error on valid input
36104 If @value{GDBN} produces an error message for valid input, that is a
36105 bug. (Note that if you're cross debugging, the problem may also be
36106 somewhere in the connection to the target.)
36108 @cindex invalid input
36110 If @value{GDBN} does not produce an error message for invalid input,
36111 that is a bug. However, you should note that your idea of
36112 ``invalid input'' might be our idea of ``an extension'' or ``support
36113 for traditional practice''.
36116 If you are an experienced user of debugging tools, your suggestions
36117 for improvement of @value{GDBN} are welcome in any case.
36120 @node Bug Reporting
36121 @section How to Report Bugs
36122 @cindex bug reports
36123 @cindex @value{GDBN} bugs, reporting
36125 A number of companies and individuals offer support for @sc{gnu} products.
36126 If you obtained @value{GDBN} from a support organization, we recommend you
36127 contact that organization first.
36129 You can find contact information for many support companies and
36130 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
36132 @c should add a web page ref...
36135 @ifset BUGURL_DEFAULT
36136 In any event, we also recommend that you submit bug reports for
36137 @value{GDBN}. The preferred method is to submit them directly using
36138 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
36139 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
36142 @strong{Do not send bug reports to @samp{info-gdb}, or to
36143 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
36144 not want to receive bug reports. Those that do have arranged to receive
36147 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
36148 serves as a repeater. The mailing list and the newsgroup carry exactly
36149 the same messages. Often people think of posting bug reports to the
36150 newsgroup instead of mailing them. This appears to work, but it has one
36151 problem which can be crucial: a newsgroup posting often lacks a mail
36152 path back to the sender. Thus, if we need to ask for more information,
36153 we may be unable to reach you. For this reason, it is better to send
36154 bug reports to the mailing list.
36156 @ifclear BUGURL_DEFAULT
36157 In any event, we also recommend that you submit bug reports for
36158 @value{GDBN} to @value{BUGURL}.
36162 The fundamental principle of reporting bugs usefully is this:
36163 @strong{report all the facts}. If you are not sure whether to state a
36164 fact or leave it out, state it!
36166 Often people omit facts because they think they know what causes the
36167 problem and assume that some details do not matter. Thus, you might
36168 assume that the name of the variable you use in an example does not matter.
36169 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
36170 stray memory reference which happens to fetch from the location where that
36171 name is stored in memory; perhaps, if the name were different, the contents
36172 of that location would fool the debugger into doing the right thing despite
36173 the bug. Play it safe and give a specific, complete example. That is the
36174 easiest thing for you to do, and the most helpful.
36176 Keep in mind that the purpose of a bug report is to enable us to fix the
36177 bug. It may be that the bug has been reported previously, but neither
36178 you nor we can know that unless your bug report is complete and
36181 Sometimes people give a few sketchy facts and ask, ``Does this ring a
36182 bell?'' Those bug reports are useless, and we urge everyone to
36183 @emph{refuse to respond to them} except to chide the sender to report
36186 To enable us to fix the bug, you should include all these things:
36190 The version of @value{GDBN}. @value{GDBN} announces it if you start
36191 with no arguments; you can also print it at any time using @code{show
36194 Without this, we will not know whether there is any point in looking for
36195 the bug in the current version of @value{GDBN}.
36198 The type of machine you are using, and the operating system name and
36202 The details of the @value{GDBN} build-time configuration.
36203 @value{GDBN} shows these details if you invoke it with the
36204 @option{--configuration} command-line option, or if you type
36205 @code{show configuration} at @value{GDBN}'s prompt.
36208 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
36209 ``@value{GCC}--2.8.1''.
36212 What compiler (and its version) was used to compile the program you are
36213 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
36214 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
36215 to get this information; for other compilers, see the documentation for
36219 The command arguments you gave the compiler to compile your example and
36220 observe the bug. For example, did you use @samp{-O}? To guarantee
36221 you will not omit something important, list them all. A copy of the
36222 Makefile (or the output from make) is sufficient.
36224 If we were to try to guess the arguments, we would probably guess wrong
36225 and then we might not encounter the bug.
36228 A complete input script, and all necessary source files, that will
36232 A description of what behavior you observe that you believe is
36233 incorrect. For example, ``It gets a fatal signal.''
36235 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
36236 will certainly notice it. But if the bug is incorrect output, we might
36237 not notice unless it is glaringly wrong. You might as well not give us
36238 a chance to make a mistake.
36240 Even if the problem you experience is a fatal signal, you should still
36241 say so explicitly. Suppose something strange is going on, such as, your
36242 copy of @value{GDBN} is out of synch, or you have encountered a bug in
36243 the C library on your system. (This has happened!) Your copy might
36244 crash and ours would not. If you told us to expect a crash, then when
36245 ours fails to crash, we would know that the bug was not happening for
36246 us. If you had not told us to expect a crash, then we would not be able
36247 to draw any conclusion from our observations.
36250 @cindex recording a session script
36251 To collect all this information, you can use a session recording program
36252 such as @command{script}, which is available on many Unix systems.
36253 Just run your @value{GDBN} session inside @command{script} and then
36254 include the @file{typescript} file with your bug report.
36256 Another way to record a @value{GDBN} session is to run @value{GDBN}
36257 inside Emacs and then save the entire buffer to a file.
36260 If you wish to suggest changes to the @value{GDBN} source, send us context
36261 diffs. If you even discuss something in the @value{GDBN} source, refer to
36262 it by context, not by line number.
36264 The line numbers in our development sources will not match those in your
36265 sources. Your line numbers would convey no useful information to us.
36269 Here are some things that are not necessary:
36273 A description of the envelope of the bug.
36275 Often people who encounter a bug spend a lot of time investigating
36276 which changes to the input file will make the bug go away and which
36277 changes will not affect it.
36279 This is often time consuming and not very useful, because the way we
36280 will find the bug is by running a single example under the debugger
36281 with breakpoints, not by pure deduction from a series of examples.
36282 We recommend that you save your time for something else.
36284 Of course, if you can find a simpler example to report @emph{instead}
36285 of the original one, that is a convenience for us. Errors in the
36286 output will be easier to spot, running under the debugger will take
36287 less time, and so on.
36289 However, simplification is not vital; if you do not want to do this,
36290 report the bug anyway and send us the entire test case you used.
36293 A patch for the bug.
36295 A patch for the bug does help us if it is a good one. But do not omit
36296 the necessary information, such as the test case, on the assumption that
36297 a patch is all we need. We might see problems with your patch and decide
36298 to fix the problem another way, or we might not understand it at all.
36300 Sometimes with a program as complicated as @value{GDBN} it is very hard to
36301 construct an example that will make the program follow a certain path
36302 through the code. If you do not send us the example, we will not be able
36303 to construct one, so we will not be able to verify that the bug is fixed.
36305 And if we cannot understand what bug you are trying to fix, or why your
36306 patch should be an improvement, we will not install it. A test case will
36307 help us to understand.
36310 A guess about what the bug is or what it depends on.
36312 Such guesses are usually wrong. Even we cannot guess right about such
36313 things without first using the debugger to find the facts.
36316 @c The readline documentation is distributed with the readline code
36317 @c and consists of the two following files:
36320 @c Use -I with makeinfo to point to the appropriate directory,
36321 @c environment var TEXINPUTS with TeX.
36322 @ifclear SYSTEM_READLINE
36323 @include rluser.texi
36324 @include hsuser.texi
36328 @appendix In Memoriam
36330 The @value{GDBN} project mourns the loss of the following long-time
36335 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
36336 to Free Software in general. Outside of @value{GDBN}, he was known in
36337 the Amiga world for his series of Fish Disks, and the GeekGadget project.
36339 @item Michael Snyder
36340 Michael was one of the Global Maintainers of the @value{GDBN} project,
36341 with contributions recorded as early as 1996, until 2011. In addition
36342 to his day to day participation, he was a large driving force behind
36343 adding Reverse Debugging to @value{GDBN}.
36346 Beyond their technical contributions to the project, they were also
36347 enjoyable members of the Free Software Community. We will miss them.
36349 @node Formatting Documentation
36350 @appendix Formatting Documentation
36352 @cindex @value{GDBN} reference card
36353 @cindex reference card
36354 The @value{GDBN} 4 release includes an already-formatted reference card, ready
36355 for printing with PostScript or Ghostscript, in the @file{gdb}
36356 subdirectory of the main source directory@footnote{In
36357 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
36358 release.}. If you can use PostScript or Ghostscript with your printer,
36359 you can print the reference card immediately with @file{refcard.ps}.
36361 The release also includes the source for the reference card. You
36362 can format it, using @TeX{}, by typing:
36368 The @value{GDBN} reference card is designed to print in @dfn{landscape}
36369 mode on US ``letter'' size paper;
36370 that is, on a sheet 11 inches wide by 8.5 inches
36371 high. You will need to specify this form of printing as an option to
36372 your @sc{dvi} output program.
36374 @cindex documentation
36376 All the documentation for @value{GDBN} comes as part of the machine-readable
36377 distribution. The documentation is written in Texinfo format, which is
36378 a documentation system that uses a single source file to produce both
36379 on-line information and a printed manual. You can use one of the Info
36380 formatting commands to create the on-line version of the documentation
36381 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
36383 @value{GDBN} includes an already formatted copy of the on-line Info
36384 version of this manual in the @file{gdb} subdirectory. The main Info
36385 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
36386 subordinate files matching @samp{gdb.info*} in the same directory. If
36387 necessary, you can print out these files, or read them with any editor;
36388 but they are easier to read using the @code{info} subsystem in @sc{gnu}
36389 Emacs or the standalone @code{info} program, available as part of the
36390 @sc{gnu} Texinfo distribution.
36392 If you want to format these Info files yourself, you need one of the
36393 Info formatting programs, such as @code{texinfo-format-buffer} or
36396 If you have @code{makeinfo} installed, and are in the top level
36397 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
36398 version @value{GDBVN}), you can make the Info file by typing:
36405 If you want to typeset and print copies of this manual, you need @TeX{},
36406 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
36407 Texinfo definitions file.
36409 @TeX{} is a typesetting program; it does not print files directly, but
36410 produces output files called @sc{dvi} files. To print a typeset
36411 document, you need a program to print @sc{dvi} files. If your system
36412 has @TeX{} installed, chances are it has such a program. The precise
36413 command to use depends on your system; @kbd{lpr -d} is common; another
36414 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
36415 require a file name without any extension or a @samp{.dvi} extension.
36417 @TeX{} also requires a macro definitions file called
36418 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
36419 written in Texinfo format. On its own, @TeX{} cannot either read or
36420 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
36421 and is located in the @file{gdb-@var{version-number}/texinfo}
36424 If you have @TeX{} and a @sc{dvi} printer program installed, you can
36425 typeset and print this manual. First switch to the @file{gdb}
36426 subdirectory of the main source directory (for example, to
36427 @file{gdb-@value{GDBVN}/gdb}) and type:
36433 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
36435 @node Installing GDB
36436 @appendix Installing @value{GDBN}
36437 @cindex installation
36440 * Requirements:: Requirements for building @value{GDBN}
36441 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
36442 * Separate Objdir:: Compiling @value{GDBN} in another directory
36443 * Config Names:: Specifying names for hosts and targets
36444 * Configure Options:: Summary of options for configure
36445 * System-wide configuration:: Having a system-wide init file
36449 @section Requirements for Building @value{GDBN}
36450 @cindex building @value{GDBN}, requirements for
36452 Building @value{GDBN} requires various tools and packages to be available.
36453 Other packages will be used only if they are found.
36455 @heading Tools/Packages Necessary for Building @value{GDBN}
36457 @item ISO C90 compiler
36458 @value{GDBN} is written in ISO C90. It should be buildable with any
36459 working C90 compiler, e.g.@: GCC.
36463 @heading Tools/Packages Optional for Building @value{GDBN}
36467 @value{GDBN} can use the Expat XML parsing library. This library may be
36468 included with your operating system distribution; if it is not, you
36469 can get the latest version from @url{http://expat.sourceforge.net}.
36470 The @file{configure} script will search for this library in several
36471 standard locations; if it is installed in an unusual path, you can
36472 use the @option{--with-libexpat-prefix} option to specify its location.
36478 Remote protocol memory maps (@pxref{Memory Map Format})
36480 Target descriptions (@pxref{Target Descriptions})
36482 Remote shared library lists (@xref{Library List Format},
36483 or alternatively @pxref{Library List Format for SVR4 Targets})
36485 MS-Windows shared libraries (@pxref{Shared Libraries})
36487 Traceframe info (@pxref{Traceframe Info Format})
36489 Branch trace (@pxref{Branch Trace Format})
36493 @cindex compressed debug sections
36494 @value{GDBN} will use the @samp{zlib} library, if available, to read
36495 compressed debug sections. Some linkers, such as GNU gold, are capable
36496 of producing binaries with compressed debug sections. If @value{GDBN}
36497 is compiled with @samp{zlib}, it will be able to read the debug
36498 information in such binaries.
36500 The @samp{zlib} library is likely included with your operating system
36501 distribution; if it is not, you can get the latest version from
36502 @url{http://zlib.net}.
36505 @value{GDBN}'s features related to character sets (@pxref{Character
36506 Sets}) require a functioning @code{iconv} implementation. If you are
36507 on a GNU system, then this is provided by the GNU C Library. Some
36508 other systems also provide a working @code{iconv}.
36510 If @value{GDBN} is using the @code{iconv} program which is installed
36511 in a non-standard place, you will need to tell @value{GDBN} where to find it.
36512 This is done with @option{--with-iconv-bin} which specifies the
36513 directory that contains the @code{iconv} program.
36515 On systems without @code{iconv}, you can install GNU Libiconv. If you
36516 have previously installed Libiconv, you can use the
36517 @option{--with-libiconv-prefix} option to configure.
36519 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
36520 arrange to build Libiconv if a directory named @file{libiconv} appears
36521 in the top-most source directory. If Libiconv is built this way, and
36522 if the operating system does not provide a suitable @code{iconv}
36523 implementation, then the just-built library will automatically be used
36524 by @value{GDBN}. One easy way to set this up is to download GNU
36525 Libiconv, unpack it, and then rename the directory holding the
36526 Libiconv source code to @samp{libiconv}.
36529 @node Running Configure
36530 @section Invoking the @value{GDBN} @file{configure} Script
36531 @cindex configuring @value{GDBN}
36532 @value{GDBN} comes with a @file{configure} script that automates the process
36533 of preparing @value{GDBN} for installation; you can then use @code{make} to
36534 build the @code{gdb} program.
36536 @c irrelevant in info file; it's as current as the code it lives with.
36537 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
36538 look at the @file{README} file in the sources; we may have improved the
36539 installation procedures since publishing this manual.}
36542 The @value{GDBN} distribution includes all the source code you need for
36543 @value{GDBN} in a single directory, whose name is usually composed by
36544 appending the version number to @samp{gdb}.
36546 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
36547 @file{gdb-@value{GDBVN}} directory. That directory contains:
36550 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
36551 script for configuring @value{GDBN} and all its supporting libraries
36553 @item gdb-@value{GDBVN}/gdb
36554 the source specific to @value{GDBN} itself
36556 @item gdb-@value{GDBVN}/bfd
36557 source for the Binary File Descriptor library
36559 @item gdb-@value{GDBVN}/include
36560 @sc{gnu} include files
36562 @item gdb-@value{GDBVN}/libiberty
36563 source for the @samp{-liberty} free software library
36565 @item gdb-@value{GDBVN}/opcodes
36566 source for the library of opcode tables and disassemblers
36568 @item gdb-@value{GDBVN}/readline
36569 source for the @sc{gnu} command-line interface
36571 @item gdb-@value{GDBVN}/glob
36572 source for the @sc{gnu} filename pattern-matching subroutine
36574 @item gdb-@value{GDBVN}/mmalloc
36575 source for the @sc{gnu} memory-mapped malloc package
36578 The simplest way to configure and build @value{GDBN} is to run @file{configure}
36579 from the @file{gdb-@var{version-number}} source directory, which in
36580 this example is the @file{gdb-@value{GDBVN}} directory.
36582 First switch to the @file{gdb-@var{version-number}} source directory
36583 if you are not already in it; then run @file{configure}. Pass the
36584 identifier for the platform on which @value{GDBN} will run as an
36590 cd gdb-@value{GDBVN}
36591 ./configure @var{host}
36596 where @var{host} is an identifier such as @samp{sun4} or
36597 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
36598 (You can often leave off @var{host}; @file{configure} tries to guess the
36599 correct value by examining your system.)
36601 Running @samp{configure @var{host}} and then running @code{make} builds the
36602 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
36603 libraries, then @code{gdb} itself. The configured source files, and the
36604 binaries, are left in the corresponding source directories.
36607 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
36608 system does not recognize this automatically when you run a different
36609 shell, you may need to run @code{sh} on it explicitly:
36612 sh configure @var{host}
36615 If you run @file{configure} from a directory that contains source
36616 directories for multiple libraries or programs, such as the
36617 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
36619 creates configuration files for every directory level underneath (unless
36620 you tell it not to, with the @samp{--norecursion} option).
36622 You should run the @file{configure} script from the top directory in the
36623 source tree, the @file{gdb-@var{version-number}} directory. If you run
36624 @file{configure} from one of the subdirectories, you will configure only
36625 that subdirectory. That is usually not what you want. In particular,
36626 if you run the first @file{configure} from the @file{gdb} subdirectory
36627 of the @file{gdb-@var{version-number}} directory, you will omit the
36628 configuration of @file{bfd}, @file{readline}, and other sibling
36629 directories of the @file{gdb} subdirectory. This leads to build errors
36630 about missing include files such as @file{bfd/bfd.h}.
36632 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
36633 However, you should make sure that the shell on your path (named by
36634 the @samp{SHELL} environment variable) is publicly readable. Remember
36635 that @value{GDBN} uses the shell to start your program---some systems refuse to
36636 let @value{GDBN} debug child processes whose programs are not readable.
36638 @node Separate Objdir
36639 @section Compiling @value{GDBN} in Another Directory
36641 If you want to run @value{GDBN} versions for several host or target machines,
36642 you need a different @code{gdb} compiled for each combination of
36643 host and target. @file{configure} is designed to make this easy by
36644 allowing you to generate each configuration in a separate subdirectory,
36645 rather than in the source directory. If your @code{make} program
36646 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
36647 @code{make} in each of these directories builds the @code{gdb}
36648 program specified there.
36650 To build @code{gdb} in a separate directory, run @file{configure}
36651 with the @samp{--srcdir} option to specify where to find the source.
36652 (You also need to specify a path to find @file{configure}
36653 itself from your working directory. If the path to @file{configure}
36654 would be the same as the argument to @samp{--srcdir}, you can leave out
36655 the @samp{--srcdir} option; it is assumed.)
36657 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
36658 separate directory for a Sun 4 like this:
36662 cd gdb-@value{GDBVN}
36665 ../gdb-@value{GDBVN}/configure sun4
36670 When @file{configure} builds a configuration using a remote source
36671 directory, it creates a tree for the binaries with the same structure
36672 (and using the same names) as the tree under the source directory. In
36673 the example, you'd find the Sun 4 library @file{libiberty.a} in the
36674 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
36675 @file{gdb-sun4/gdb}.
36677 Make sure that your path to the @file{configure} script has just one
36678 instance of @file{gdb} in it. If your path to @file{configure} looks
36679 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
36680 one subdirectory of @value{GDBN}, not the whole package. This leads to
36681 build errors about missing include files such as @file{bfd/bfd.h}.
36683 One popular reason to build several @value{GDBN} configurations in separate
36684 directories is to configure @value{GDBN} for cross-compiling (where
36685 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
36686 programs that run on another machine---the @dfn{target}).
36687 You specify a cross-debugging target by
36688 giving the @samp{--target=@var{target}} option to @file{configure}.
36690 When you run @code{make} to build a program or library, you must run
36691 it in a configured directory---whatever directory you were in when you
36692 called @file{configure} (or one of its subdirectories).
36694 The @code{Makefile} that @file{configure} generates in each source
36695 directory also runs recursively. If you type @code{make} in a source
36696 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
36697 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
36698 will build all the required libraries, and then build GDB.
36700 When you have multiple hosts or targets configured in separate
36701 directories, you can run @code{make} on them in parallel (for example,
36702 if they are NFS-mounted on each of the hosts); they will not interfere
36706 @section Specifying Names for Hosts and Targets
36708 The specifications used for hosts and targets in the @file{configure}
36709 script are based on a three-part naming scheme, but some short predefined
36710 aliases are also supported. The full naming scheme encodes three pieces
36711 of information in the following pattern:
36714 @var{architecture}-@var{vendor}-@var{os}
36717 For example, you can use the alias @code{sun4} as a @var{host} argument,
36718 or as the value for @var{target} in a @code{--target=@var{target}}
36719 option. The equivalent full name is @samp{sparc-sun-sunos4}.
36721 The @file{configure} script accompanying @value{GDBN} does not provide
36722 any query facility to list all supported host and target names or
36723 aliases. @file{configure} calls the Bourne shell script
36724 @code{config.sub} to map abbreviations to full names; you can read the
36725 script, if you wish, or you can use it to test your guesses on
36726 abbreviations---for example:
36729 % sh config.sub i386-linux
36731 % sh config.sub alpha-linux
36732 alpha-unknown-linux-gnu
36733 % sh config.sub hp9k700
36735 % sh config.sub sun4
36736 sparc-sun-sunos4.1.1
36737 % sh config.sub sun3
36738 m68k-sun-sunos4.1.1
36739 % sh config.sub i986v
36740 Invalid configuration `i986v': machine `i986v' not recognized
36744 @code{config.sub} is also distributed in the @value{GDBN} source
36745 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
36747 @node Configure Options
36748 @section @file{configure} Options
36750 Here is a summary of the @file{configure} options and arguments that
36751 are most often useful for building @value{GDBN}. @file{configure} also has
36752 several other options not listed here. @inforef{What Configure
36753 Does,,configure.info}, for a full explanation of @file{configure}.
36756 configure @r{[}--help@r{]}
36757 @r{[}--prefix=@var{dir}@r{]}
36758 @r{[}--exec-prefix=@var{dir}@r{]}
36759 @r{[}--srcdir=@var{dirname}@r{]}
36760 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
36761 @r{[}--target=@var{target}@r{]}
36766 You may introduce options with a single @samp{-} rather than
36767 @samp{--} if you prefer; but you may abbreviate option names if you use
36772 Display a quick summary of how to invoke @file{configure}.
36774 @item --prefix=@var{dir}
36775 Configure the source to install programs and files under directory
36778 @item --exec-prefix=@var{dir}
36779 Configure the source to install programs under directory
36782 @c avoid splitting the warning from the explanation:
36784 @item --srcdir=@var{dirname}
36785 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
36786 @code{make} that implements the @code{VPATH} feature.}@*
36787 Use this option to make configurations in directories separate from the
36788 @value{GDBN} source directories. Among other things, you can use this to
36789 build (or maintain) several configurations simultaneously, in separate
36790 directories. @file{configure} writes configuration-specific files in
36791 the current directory, but arranges for them to use the source in the
36792 directory @var{dirname}. @file{configure} creates directories under
36793 the working directory in parallel to the source directories below
36796 @item --norecursion
36797 Configure only the directory level where @file{configure} is executed; do not
36798 propagate configuration to subdirectories.
36800 @item --target=@var{target}
36801 Configure @value{GDBN} for cross-debugging programs running on the specified
36802 @var{target}. Without this option, @value{GDBN} is configured to debug
36803 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
36805 There is no convenient way to generate a list of all available targets.
36807 @item @var{host} @dots{}
36808 Configure @value{GDBN} to run on the specified @var{host}.
36810 There is no convenient way to generate a list of all available hosts.
36813 There are many other options available as well, but they are generally
36814 needed for special purposes only.
36816 @node System-wide configuration
36817 @section System-wide configuration and settings
36818 @cindex system-wide init file
36820 @value{GDBN} can be configured to have a system-wide init file;
36821 this file will be read and executed at startup (@pxref{Startup, , What
36822 @value{GDBN} does during startup}).
36824 Here is the corresponding configure option:
36827 @item --with-system-gdbinit=@var{file}
36828 Specify that the default location of the system-wide init file is
36832 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
36833 it may be subject to relocation. Two possible cases:
36837 If the default location of this init file contains @file{$prefix},
36838 it will be subject to relocation. Suppose that the configure options
36839 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
36840 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
36841 init file is looked for as @file{$install/etc/gdbinit} instead of
36842 @file{$prefix/etc/gdbinit}.
36845 By contrast, if the default location does not contain the prefix,
36846 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
36847 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
36848 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
36849 wherever @value{GDBN} is installed.
36852 If the configured location of the system-wide init file (as given by the
36853 @option{--with-system-gdbinit} option at configure time) is in the
36854 data-directory (as specified by @option{--with-gdb-datadir} at configure
36855 time) or in one of its subdirectories, then @value{GDBN} will look for the
36856 system-wide init file in the directory specified by the
36857 @option{--data-directory} command-line option.
36858 Note that the system-wide init file is only read once, during @value{GDBN}
36859 initialization. If the data-directory is changed after @value{GDBN} has
36860 started with the @code{set data-directory} command, the file will not be
36864 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
36867 @node System-wide Configuration Scripts
36868 @subsection Installed System-wide Configuration Scripts
36869 @cindex system-wide configuration scripts
36871 The @file{system-gdbinit} directory, located inside the data-directory
36872 (as specified by @option{--with-gdb-datadir} at configure time) contains
36873 a number of scripts which can be used as system-wide init files. To
36874 automatically source those scripts at startup, @value{GDBN} should be
36875 configured with @option{--with-system-gdbinit}. Otherwise, any user
36876 should be able to source them by hand as needed.
36878 The following scripts are currently available:
36881 @item @file{elinos.py}
36883 @cindex ELinOS system-wide configuration script
36884 This script is useful when debugging a program on an ELinOS target.
36885 It takes advantage of the environment variables defined in a standard
36886 ELinOS environment in order to determine the location of the system
36887 shared libraries, and then sets the @samp{solib-absolute-prefix}
36888 and @samp{solib-search-path} variables appropriately.
36890 @item @file{wrs-linux.py}
36891 @pindex wrs-linux.py
36892 @cindex Wind River Linux system-wide configuration script
36893 This script is useful when debugging a program on a target running
36894 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
36895 the host-side sysroot used by the target system.
36899 @node Maintenance Commands
36900 @appendix Maintenance Commands
36901 @cindex maintenance commands
36902 @cindex internal commands
36904 In addition to commands intended for @value{GDBN} users, @value{GDBN}
36905 includes a number of commands intended for @value{GDBN} developers,
36906 that are not documented elsewhere in this manual. These commands are
36907 provided here for reference. (For commands that turn on debugging
36908 messages, see @ref{Debugging Output}.)
36911 @kindex maint agent
36912 @kindex maint agent-eval
36913 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36914 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36915 Translate the given @var{expression} into remote agent bytecodes.
36916 This command is useful for debugging the Agent Expression mechanism
36917 (@pxref{Agent Expressions}). The @samp{agent} version produces an
36918 expression useful for data collection, such as by tracepoints, while
36919 @samp{maint agent-eval} produces an expression that evaluates directly
36920 to a result. For instance, a collection expression for @code{globa +
36921 globb} will include bytecodes to record four bytes of memory at each
36922 of the addresses of @code{globa} and @code{globb}, while discarding
36923 the result of the addition, while an evaluation expression will do the
36924 addition and return the sum.
36925 If @code{-at} is given, generate remote agent bytecode for @var{location}.
36926 If not, generate remote agent bytecode for current frame PC address.
36928 @kindex maint agent-printf
36929 @item maint agent-printf @var{format},@var{expr},...
36930 Translate the given format string and list of argument expressions
36931 into remote agent bytecodes and display them as a disassembled list.
36932 This command is useful for debugging the agent version of dynamic
36933 printf (@pxref{Dynamic Printf}).
36935 @kindex maint info breakpoints
36936 @item @anchor{maint info breakpoints}maint info breakpoints
36937 Using the same format as @samp{info breakpoints}, display both the
36938 breakpoints you've set explicitly, and those @value{GDBN} is using for
36939 internal purposes. Internal breakpoints are shown with negative
36940 breakpoint numbers. The type column identifies what kind of breakpoint
36945 Normal, explicitly set breakpoint.
36948 Normal, explicitly set watchpoint.
36951 Internal breakpoint, used to handle correctly stepping through
36952 @code{longjmp} calls.
36954 @item longjmp resume
36955 Internal breakpoint at the target of a @code{longjmp}.
36958 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
36961 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
36964 Shared library events.
36968 @kindex maint info bfds
36969 @item maint info bfds
36970 This prints information about each @code{bfd} object that is known to
36971 @value{GDBN}. @xref{Top, , BFD, bfd, The Binary File Descriptor Library}.
36973 @kindex set displaced-stepping
36974 @kindex show displaced-stepping
36975 @cindex displaced stepping support
36976 @cindex out-of-line single-stepping
36977 @item set displaced-stepping
36978 @itemx show displaced-stepping
36979 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
36980 if the target supports it. Displaced stepping is a way to single-step
36981 over breakpoints without removing them from the inferior, by executing
36982 an out-of-line copy of the instruction that was originally at the
36983 breakpoint location. It is also known as out-of-line single-stepping.
36986 @item set displaced-stepping on
36987 If the target architecture supports it, @value{GDBN} will use
36988 displaced stepping to step over breakpoints.
36990 @item set displaced-stepping off
36991 @value{GDBN} will not use displaced stepping to step over breakpoints,
36992 even if such is supported by the target architecture.
36994 @cindex non-stop mode, and @samp{set displaced-stepping}
36995 @item set displaced-stepping auto
36996 This is the default mode. @value{GDBN} will use displaced stepping
36997 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
36998 architecture supports displaced stepping.
37001 @kindex maint check-psymtabs
37002 @item maint check-psymtabs
37003 Check the consistency of currently expanded psymtabs versus symtabs.
37004 Use this to check, for example, whether a symbol is in one but not the other.
37006 @kindex maint check-symtabs
37007 @item maint check-symtabs
37008 Check the consistency of currently expanded symtabs.
37010 @kindex maint expand-symtabs
37011 @item maint expand-symtabs [@var{regexp}]
37012 Expand symbol tables.
37013 If @var{regexp} is specified, only expand symbol tables for file
37014 names matching @var{regexp}.
37016 @kindex maint cplus first_component
37017 @item maint cplus first_component @var{name}
37018 Print the first C@t{++} class/namespace component of @var{name}.
37020 @kindex maint cplus namespace
37021 @item maint cplus namespace
37022 Print the list of possible C@t{++} namespaces.
37024 @kindex maint demangle
37025 @item maint demangle @var{name}
37026 Demangle a C@t{++} or Objective-C mangled @var{name}.
37028 @kindex maint deprecate
37029 @kindex maint undeprecate
37030 @cindex deprecated commands
37031 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
37032 @itemx maint undeprecate @var{command}
37033 Deprecate or undeprecate the named @var{command}. Deprecated commands
37034 cause @value{GDBN} to issue a warning when you use them. The optional
37035 argument @var{replacement} says which newer command should be used in
37036 favor of the deprecated one; if it is given, @value{GDBN} will mention
37037 the replacement as part of the warning.
37039 @kindex maint dump-me
37040 @item maint dump-me
37041 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
37042 Cause a fatal signal in the debugger and force it to dump its core.
37043 This is supported only on systems which support aborting a program
37044 with the @code{SIGQUIT} signal.
37046 @kindex maint internal-error
37047 @kindex maint internal-warning
37048 @item maint internal-error @r{[}@var{message-text}@r{]}
37049 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
37050 Cause @value{GDBN} to call the internal function @code{internal_error}
37051 or @code{internal_warning} and hence behave as though an internal error
37052 or internal warning has been detected. In addition to reporting the
37053 internal problem, these functions give the user the opportunity to
37054 either quit @value{GDBN} or create a core file of the current
37055 @value{GDBN} session.
37057 These commands take an optional parameter @var{message-text} that is
37058 used as the text of the error or warning message.
37060 Here's an example of using @code{internal-error}:
37063 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
37064 @dots{}/maint.c:121: internal-error: testing, 1, 2
37065 A problem internal to GDB has been detected. Further
37066 debugging may prove unreliable.
37067 Quit this debugging session? (y or n) @kbd{n}
37068 Create a core file? (y or n) @kbd{n}
37072 @cindex @value{GDBN} internal error
37073 @cindex internal errors, control of @value{GDBN} behavior
37075 @kindex maint set internal-error
37076 @kindex maint show internal-error
37077 @kindex maint set internal-warning
37078 @kindex maint show internal-warning
37079 @item maint set internal-error @var{action} [ask|yes|no]
37080 @itemx maint show internal-error @var{action}
37081 @itemx maint set internal-warning @var{action} [ask|yes|no]
37082 @itemx maint show internal-warning @var{action}
37083 When @value{GDBN} reports an internal problem (error or warning) it
37084 gives the user the opportunity to both quit @value{GDBN} and create a
37085 core file of the current @value{GDBN} session. These commands let you
37086 override the default behaviour for each particular @var{action},
37087 described in the table below.
37091 You can specify that @value{GDBN} should always (yes) or never (no)
37092 quit. The default is to ask the user what to do.
37095 You can specify that @value{GDBN} should always (yes) or never (no)
37096 create a core file. The default is to ask the user what to do.
37099 @kindex maint packet
37100 @item maint packet @var{text}
37101 If @value{GDBN} is talking to an inferior via the serial protocol,
37102 then this command sends the string @var{text} to the inferior, and
37103 displays the response packet. @value{GDBN} supplies the initial
37104 @samp{$} character, the terminating @samp{#} character, and the
37107 @kindex maint print architecture
37108 @item maint print architecture @r{[}@var{file}@r{]}
37109 Print the entire architecture configuration. The optional argument
37110 @var{file} names the file where the output goes.
37112 @kindex maint print c-tdesc
37113 @item maint print c-tdesc
37114 Print the current target description (@pxref{Target Descriptions}) as
37115 a C source file. The created source file can be used in @value{GDBN}
37116 when an XML parser is not available to parse the description.
37118 @kindex maint print dummy-frames
37119 @item maint print dummy-frames
37120 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
37123 (@value{GDBP}) @kbd{b add}
37125 (@value{GDBP}) @kbd{print add(2,3)}
37126 Breakpoint 2, add (a=2, b=3) at @dots{}
37128 The program being debugged stopped while in a function called from GDB.
37130 (@value{GDBP}) @kbd{maint print dummy-frames}
37131 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
37132 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
37133 call_lo=0x01014000 call_hi=0x01014001
37137 Takes an optional file parameter.
37139 @kindex maint print registers
37140 @kindex maint print raw-registers
37141 @kindex maint print cooked-registers
37142 @kindex maint print register-groups
37143 @kindex maint print remote-registers
37144 @item maint print registers @r{[}@var{file}@r{]}
37145 @itemx maint print raw-registers @r{[}@var{file}@r{]}
37146 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
37147 @itemx maint print register-groups @r{[}@var{file}@r{]}
37148 @itemx maint print remote-registers @r{[}@var{file}@r{]}
37149 Print @value{GDBN}'s internal register data structures.
37151 The command @code{maint print raw-registers} includes the contents of
37152 the raw register cache; the command @code{maint print
37153 cooked-registers} includes the (cooked) value of all registers,
37154 including registers which aren't available on the target nor visible
37155 to user; the command @code{maint print register-groups} includes the
37156 groups that each register is a member of; and the command @code{maint
37157 print remote-registers} includes the remote target's register numbers
37158 and offsets in the `G' packets.
37160 These commands take an optional parameter, a file name to which to
37161 write the information.
37163 @kindex maint print reggroups
37164 @item maint print reggroups @r{[}@var{file}@r{]}
37165 Print @value{GDBN}'s internal register group data structures. The
37166 optional argument @var{file} tells to what file to write the
37169 The register groups info looks like this:
37172 (@value{GDBP}) @kbd{maint print reggroups}
37185 This command forces @value{GDBN} to flush its internal register cache.
37187 @kindex maint print objfiles
37188 @cindex info for known object files
37189 @item maint print objfiles @r{[}@var{regexp}@r{]}
37190 Print a dump of all known object files.
37191 If @var{regexp} is specified, only print object files whose names
37192 match @var{regexp}. For each object file, this command prints its name,
37193 address in memory, and all of its psymtabs and symtabs.
37195 @kindex maint print section-scripts
37196 @cindex info for known .debug_gdb_scripts-loaded scripts
37197 @item maint print section-scripts [@var{regexp}]
37198 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
37199 If @var{regexp} is specified, only print scripts loaded by object files
37200 matching @var{regexp}.
37201 For each script, this command prints its name as specified in the objfile,
37202 and the full path if known.
37203 @xref{dotdebug_gdb_scripts section}.
37205 @kindex maint print statistics
37206 @cindex bcache statistics
37207 @item maint print statistics
37208 This command prints, for each object file in the program, various data
37209 about that object file followed by the byte cache (@dfn{bcache})
37210 statistics for the object file. The objfile data includes the number
37211 of minimal, partial, full, and stabs symbols, the number of types
37212 defined by the objfile, the number of as yet unexpanded psym tables,
37213 the number of line tables and string tables, and the amount of memory
37214 used by the various tables. The bcache statistics include the counts,
37215 sizes, and counts of duplicates of all and unique objects, max,
37216 average, and median entry size, total memory used and its overhead and
37217 savings, and various measures of the hash table size and chain
37220 @kindex maint print target-stack
37221 @cindex target stack description
37222 @item maint print target-stack
37223 A @dfn{target} is an interface between the debugger and a particular
37224 kind of file or process. Targets can be stacked in @dfn{strata},
37225 so that more than one target can potentially respond to a request.
37226 In particular, memory accesses will walk down the stack of targets
37227 until they find a target that is interested in handling that particular
37230 This command prints a short description of each layer that was pushed on
37231 the @dfn{target stack}, starting from the top layer down to the bottom one.
37233 @kindex maint print type
37234 @cindex type chain of a data type
37235 @item maint print type @var{expr}
37236 Print the type chain for a type specified by @var{expr}. The argument
37237 can be either a type name or a symbol. If it is a symbol, the type of
37238 that symbol is described. The type chain produced by this command is
37239 a recursive definition of the data type as stored in @value{GDBN}'s
37240 data structures, including its flags and contained types.
37242 @kindex maint set dwarf2 always-disassemble
37243 @kindex maint show dwarf2 always-disassemble
37244 @item maint set dwarf2 always-disassemble
37245 @item maint show dwarf2 always-disassemble
37246 Control the behavior of @code{info address} when using DWARF debugging
37249 The default is @code{off}, which means that @value{GDBN} should try to
37250 describe a variable's location in an easily readable format. When
37251 @code{on}, @value{GDBN} will instead display the DWARF location
37252 expression in an assembly-like format. Note that some locations are
37253 too complex for @value{GDBN} to describe simply; in this case you will
37254 always see the disassembly form.
37256 Here is an example of the resulting disassembly:
37259 (gdb) info addr argc
37260 Symbol "argc" is a complex DWARF expression:
37264 For more information on these expressions, see
37265 @uref{http://www.dwarfstd.org/, the DWARF standard}.
37267 @kindex maint set dwarf2 max-cache-age
37268 @kindex maint show dwarf2 max-cache-age
37269 @item maint set dwarf2 max-cache-age
37270 @itemx maint show dwarf2 max-cache-age
37271 Control the DWARF 2 compilation unit cache.
37273 @cindex DWARF 2 compilation units cache
37274 In object files with inter-compilation-unit references, such as those
37275 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
37276 reader needs to frequently refer to previously read compilation units.
37277 This setting controls how long a compilation unit will remain in the
37278 cache if it is not referenced. A higher limit means that cached
37279 compilation units will be stored in memory longer, and more total
37280 memory will be used. Setting it to zero disables caching, which will
37281 slow down @value{GDBN} startup, but reduce memory consumption.
37283 @kindex maint set profile
37284 @kindex maint show profile
37285 @cindex profiling GDB
37286 @item maint set profile
37287 @itemx maint show profile
37288 Control profiling of @value{GDBN}.
37290 Profiling will be disabled until you use the @samp{maint set profile}
37291 command to enable it. When you enable profiling, the system will begin
37292 collecting timing and execution count data; when you disable profiling or
37293 exit @value{GDBN}, the results will be written to a log file. Remember that
37294 if you use profiling, @value{GDBN} will overwrite the profiling log file
37295 (often called @file{gmon.out}). If you have a record of important profiling
37296 data in a @file{gmon.out} file, be sure to move it to a safe location.
37298 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
37299 compiled with the @samp{-pg} compiler option.
37301 @kindex maint set show-debug-regs
37302 @kindex maint show show-debug-regs
37303 @cindex hardware debug registers
37304 @item maint set show-debug-regs
37305 @itemx maint show show-debug-regs
37306 Control whether to show variables that mirror the hardware debug
37307 registers. Use @code{ON} to enable, @code{OFF} to disable. If
37308 enabled, the debug registers values are shown when @value{GDBN} inserts or
37309 removes a hardware breakpoint or watchpoint, and when the inferior
37310 triggers a hardware-assisted breakpoint or watchpoint.
37312 @kindex maint set show-all-tib
37313 @kindex maint show show-all-tib
37314 @item maint set show-all-tib
37315 @itemx maint show show-all-tib
37316 Control whether to show all non zero areas within a 1k block starting
37317 at thread local base, when using the @samp{info w32 thread-information-block}
37320 @kindex maint set per-command
37321 @kindex maint show per-command
37322 @item maint set per-command
37323 @itemx maint show per-command
37324 @cindex resources used by commands
37326 @value{GDBN} can display the resources used by each command.
37327 This is useful in debugging performance problems.
37330 @item maint set per-command space [on|off]
37331 @itemx maint show per-command space
37332 Enable or disable the printing of the memory used by GDB for each command.
37333 If enabled, @value{GDBN} will display how much memory each command
37334 took, following the command's own output.
37335 This can also be requested by invoking @value{GDBN} with the
37336 @option{--statistics} command-line switch (@pxref{Mode Options}).
37338 @item maint set per-command time [on|off]
37339 @itemx maint show per-command time
37340 Enable or disable the printing of the execution time of @value{GDBN}
37342 If enabled, @value{GDBN} will display how much time it
37343 took to execute each command, following the command's own output.
37344 Both CPU time and wallclock time are printed.
37345 Printing both is useful when trying to determine whether the cost is
37346 CPU or, e.g., disk/network latency.
37347 Note that the CPU time printed is for @value{GDBN} only, it does not include
37348 the execution time of the inferior because there's no mechanism currently
37349 to compute how much time was spent by @value{GDBN} and how much time was
37350 spent by the program been debugged.
37351 This can also be requested by invoking @value{GDBN} with the
37352 @option{--statistics} command-line switch (@pxref{Mode Options}).
37354 @item maint set per-command symtab [on|off]
37355 @itemx maint show per-command symtab
37356 Enable or disable the printing of basic symbol table statistics
37358 If enabled, @value{GDBN} will display the following information:
37362 number of symbol tables
37364 number of primary symbol tables
37366 number of blocks in the blockvector
37370 @kindex maint space
37371 @cindex memory used by commands
37372 @item maint space @var{value}
37373 An alias for @code{maint set per-command space}.
37374 A non-zero value enables it, zero disables it.
37377 @cindex time of command execution
37378 @item maint time @var{value}
37379 An alias for @code{maint set per-command time}.
37380 A non-zero value enables it, zero disables it.
37382 @kindex maint translate-address
37383 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
37384 Find the symbol stored at the location specified by the address
37385 @var{addr} and an optional section name @var{section}. If found,
37386 @value{GDBN} prints the name of the closest symbol and an offset from
37387 the symbol's location to the specified address. This is similar to
37388 the @code{info address} command (@pxref{Symbols}), except that this
37389 command also allows to find symbols in other sections.
37391 If section was not specified, the section in which the symbol was found
37392 is also printed. For dynamically linked executables, the name of
37393 executable or shared library containing the symbol is printed as well.
37397 The following command is useful for non-interactive invocations of
37398 @value{GDBN}, such as in the test suite.
37401 @item set watchdog @var{nsec}
37402 @kindex set watchdog
37403 @cindex watchdog timer
37404 @cindex timeout for commands
37405 Set the maximum number of seconds @value{GDBN} will wait for the
37406 target operation to finish. If this time expires, @value{GDBN}
37407 reports and error and the command is aborted.
37409 @item show watchdog
37410 Show the current setting of the target wait timeout.
37413 @node Remote Protocol
37414 @appendix @value{GDBN} Remote Serial Protocol
37419 * Stop Reply Packets::
37420 * General Query Packets::
37421 * Architecture-Specific Protocol Details::
37422 * Tracepoint Packets::
37423 * Host I/O Packets::
37425 * Notification Packets::
37426 * Remote Non-Stop::
37427 * Packet Acknowledgment::
37429 * File-I/O Remote Protocol Extension::
37430 * Library List Format::
37431 * Library List Format for SVR4 Targets::
37432 * Memory Map Format::
37433 * Thread List Format::
37434 * Traceframe Info Format::
37435 * Branch Trace Format::
37441 There may be occasions when you need to know something about the
37442 protocol---for example, if there is only one serial port to your target
37443 machine, you might want your program to do something special if it
37444 recognizes a packet meant for @value{GDBN}.
37446 In the examples below, @samp{->} and @samp{<-} are used to indicate
37447 transmitted and received data, respectively.
37449 @cindex protocol, @value{GDBN} remote serial
37450 @cindex serial protocol, @value{GDBN} remote
37451 @cindex remote serial protocol
37452 All @value{GDBN} commands and responses (other than acknowledgments
37453 and notifications, see @ref{Notification Packets}) are sent as a
37454 @var{packet}. A @var{packet} is introduced with the character
37455 @samp{$}, the actual @var{packet-data}, and the terminating character
37456 @samp{#} followed by a two-digit @var{checksum}:
37459 @code{$}@var{packet-data}@code{#}@var{checksum}
37463 @cindex checksum, for @value{GDBN} remote
37465 The two-digit @var{checksum} is computed as the modulo 256 sum of all
37466 characters between the leading @samp{$} and the trailing @samp{#} (an
37467 eight bit unsigned checksum).
37469 Implementors should note that prior to @value{GDBN} 5.0 the protocol
37470 specification also included an optional two-digit @var{sequence-id}:
37473 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
37476 @cindex sequence-id, for @value{GDBN} remote
37478 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
37479 has never output @var{sequence-id}s. Stubs that handle packets added
37480 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
37482 When either the host or the target machine receives a packet, the first
37483 response expected is an acknowledgment: either @samp{+} (to indicate
37484 the package was received correctly) or @samp{-} (to request
37488 -> @code{$}@var{packet-data}@code{#}@var{checksum}
37493 The @samp{+}/@samp{-} acknowledgments can be disabled
37494 once a connection is established.
37495 @xref{Packet Acknowledgment}, for details.
37497 The host (@value{GDBN}) sends @var{command}s, and the target (the
37498 debugging stub incorporated in your program) sends a @var{response}. In
37499 the case of step and continue @var{command}s, the response is only sent
37500 when the operation has completed, and the target has again stopped all
37501 threads in all attached processes. This is the default all-stop mode
37502 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
37503 execution mode; see @ref{Remote Non-Stop}, for details.
37505 @var{packet-data} consists of a sequence of characters with the
37506 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
37509 @cindex remote protocol, field separator
37510 Fields within the packet should be separated using @samp{,} @samp{;} or
37511 @samp{:}. Except where otherwise noted all numbers are represented in
37512 @sc{hex} with leading zeros suppressed.
37514 Implementors should note that prior to @value{GDBN} 5.0, the character
37515 @samp{:} could not appear as the third character in a packet (as it
37516 would potentially conflict with the @var{sequence-id}).
37518 @cindex remote protocol, binary data
37519 @anchor{Binary Data}
37520 Binary data in most packets is encoded either as two hexadecimal
37521 digits per byte of binary data. This allowed the traditional remote
37522 protocol to work over connections which were only seven-bit clean.
37523 Some packets designed more recently assume an eight-bit clean
37524 connection, and use a more efficient encoding to send and receive
37527 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
37528 as an escape character. Any escaped byte is transmitted as the escape
37529 character followed by the original character XORed with @code{0x20}.
37530 For example, the byte @code{0x7d} would be transmitted as the two
37531 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
37532 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
37533 @samp{@}}) must always be escaped. Responses sent by the stub
37534 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
37535 is not interpreted as the start of a run-length encoded sequence
37538 Response @var{data} can be run-length encoded to save space.
37539 Run-length encoding replaces runs of identical characters with one
37540 instance of the repeated character, followed by a @samp{*} and a
37541 repeat count. The repeat count is itself sent encoded, to avoid
37542 binary characters in @var{data}: a value of @var{n} is sent as
37543 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
37544 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
37545 code 32) for a repeat count of 3. (This is because run-length
37546 encoding starts to win for counts 3 or more.) Thus, for example,
37547 @samp{0* } is a run-length encoding of ``0000'': the space character
37548 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
37551 The printable characters @samp{#} and @samp{$} or with a numeric value
37552 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
37553 seven repeats (@samp{$}) can be expanded using a repeat count of only
37554 five (@samp{"}). For example, @samp{00000000} can be encoded as
37557 The error response returned for some packets includes a two character
37558 error number. That number is not well defined.
37560 @cindex empty response, for unsupported packets
37561 For any @var{command} not supported by the stub, an empty response
37562 (@samp{$#00}) should be returned. That way it is possible to extend the
37563 protocol. A newer @value{GDBN} can tell if a packet is supported based
37566 At a minimum, a stub is required to support the @samp{g} and @samp{G}
37567 commands for register access, and the @samp{m} and @samp{M} commands
37568 for memory access. Stubs that only control single-threaded targets
37569 can implement run control with the @samp{c} (continue), and @samp{s}
37570 (step) commands. Stubs that support multi-threading targets should
37571 support the @samp{vCont} command. All other commands are optional.
37576 The following table provides a complete list of all currently defined
37577 @var{command}s and their corresponding response @var{data}.
37578 @xref{File-I/O Remote Protocol Extension}, for details about the File
37579 I/O extension of the remote protocol.
37581 Each packet's description has a template showing the packet's overall
37582 syntax, followed by an explanation of the packet's meaning. We
37583 include spaces in some of the templates for clarity; these are not
37584 part of the packet's syntax. No @value{GDBN} packet uses spaces to
37585 separate its components. For example, a template like @samp{foo
37586 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
37587 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
37588 @var{baz}. @value{GDBN} does not transmit a space character between the
37589 @samp{foo} and the @var{bar}, or between the @var{bar} and the
37592 @cindex @var{thread-id}, in remote protocol
37593 @anchor{thread-id syntax}
37594 Several packets and replies include a @var{thread-id} field to identify
37595 a thread. Normally these are positive numbers with a target-specific
37596 interpretation, formatted as big-endian hex strings. A @var{thread-id}
37597 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
37600 In addition, the remote protocol supports a multiprocess feature in
37601 which the @var{thread-id} syntax is extended to optionally include both
37602 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
37603 The @var{pid} (process) and @var{tid} (thread) components each have the
37604 format described above: a positive number with target-specific
37605 interpretation formatted as a big-endian hex string, literal @samp{-1}
37606 to indicate all processes or threads (respectively), or @samp{0} to
37607 indicate an arbitrary process or thread. Specifying just a process, as
37608 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
37609 error to specify all processes but a specific thread, such as
37610 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
37611 for those packets and replies explicitly documented to include a process
37612 ID, rather than a @var{thread-id}.
37614 The multiprocess @var{thread-id} syntax extensions are only used if both
37615 @value{GDBN} and the stub report support for the @samp{multiprocess}
37616 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
37619 Note that all packet forms beginning with an upper- or lower-case
37620 letter, other than those described here, are reserved for future use.
37622 Here are the packet descriptions.
37627 @cindex @samp{!} packet
37628 @anchor{extended mode}
37629 Enable extended mode. In extended mode, the remote server is made
37630 persistent. The @samp{R} packet is used to restart the program being
37636 The remote target both supports and has enabled extended mode.
37640 @cindex @samp{?} packet
37641 Indicate the reason the target halted. The reply is the same as for
37642 step and continue. This packet has a special interpretation when the
37643 target is in non-stop mode; see @ref{Remote Non-Stop}.
37646 @xref{Stop Reply Packets}, for the reply specifications.
37648 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
37649 @cindex @samp{A} packet
37650 Initialized @code{argv[]} array passed into program. @var{arglen}
37651 specifies the number of bytes in the hex encoded byte stream
37652 @var{arg}. See @code{gdbserver} for more details.
37657 The arguments were set.
37663 @cindex @samp{b} packet
37664 (Don't use this packet; its behavior is not well-defined.)
37665 Change the serial line speed to @var{baud}.
37667 JTC: @emph{When does the transport layer state change? When it's
37668 received, or after the ACK is transmitted. In either case, there are
37669 problems if the command or the acknowledgment packet is dropped.}
37671 Stan: @emph{If people really wanted to add something like this, and get
37672 it working for the first time, they ought to modify ser-unix.c to send
37673 some kind of out-of-band message to a specially-setup stub and have the
37674 switch happen "in between" packets, so that from remote protocol's point
37675 of view, nothing actually happened.}
37677 @item B @var{addr},@var{mode}
37678 @cindex @samp{B} packet
37679 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
37680 breakpoint at @var{addr}.
37682 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
37683 (@pxref{insert breakpoint or watchpoint packet}).
37685 @cindex @samp{bc} packet
37688 Backward continue. Execute the target system in reverse. No parameter.
37689 @xref{Reverse Execution}, for more information.
37692 @xref{Stop Reply Packets}, for the reply specifications.
37694 @cindex @samp{bs} packet
37697 Backward single step. Execute one instruction in reverse. No parameter.
37698 @xref{Reverse Execution}, for more information.
37701 @xref{Stop Reply Packets}, for the reply specifications.
37703 @item c @r{[}@var{addr}@r{]}
37704 @cindex @samp{c} packet
37705 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
37706 resume at current address.
37708 This packet is deprecated for multi-threading support. @xref{vCont
37712 @xref{Stop Reply Packets}, for the reply specifications.
37714 @item C @var{sig}@r{[};@var{addr}@r{]}
37715 @cindex @samp{C} packet
37716 Continue with signal @var{sig} (hex signal number). If
37717 @samp{;@var{addr}} is omitted, resume at same address.
37719 This packet is deprecated for multi-threading support. @xref{vCont
37723 @xref{Stop Reply Packets}, for the reply specifications.
37726 @cindex @samp{d} packet
37729 Don't use this packet; instead, define a general set packet
37730 (@pxref{General Query Packets}).
37734 @cindex @samp{D} packet
37735 The first form of the packet is used to detach @value{GDBN} from the
37736 remote system. It is sent to the remote target
37737 before @value{GDBN} disconnects via the @code{detach} command.
37739 The second form, including a process ID, is used when multiprocess
37740 protocol extensions are enabled (@pxref{multiprocess extensions}), to
37741 detach only a specific process. The @var{pid} is specified as a
37742 big-endian hex string.
37752 @item F @var{RC},@var{EE},@var{CF};@var{XX}
37753 @cindex @samp{F} packet
37754 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
37755 This is part of the File-I/O protocol extension. @xref{File-I/O
37756 Remote Protocol Extension}, for the specification.
37759 @anchor{read registers packet}
37760 @cindex @samp{g} packet
37761 Read general registers.
37765 @item @var{XX@dots{}}
37766 Each byte of register data is described by two hex digits. The bytes
37767 with the register are transmitted in target byte order. The size of
37768 each register and their position within the @samp{g} packet are
37769 determined by the @value{GDBN} internal gdbarch functions
37770 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
37771 specification of several standard @samp{g} packets is specified below.
37773 When reading registers from a trace frame (@pxref{Analyze Collected
37774 Data,,Using the Collected Data}), the stub may also return a string of
37775 literal @samp{x}'s in place of the register data digits, to indicate
37776 that the corresponding register has not been collected, thus its value
37777 is unavailable. For example, for an architecture with 4 registers of
37778 4 bytes each, the following reply indicates to @value{GDBN} that
37779 registers 0 and 2 have not been collected, while registers 1 and 3
37780 have been collected, and both have zero value:
37784 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
37791 @item G @var{XX@dots{}}
37792 @cindex @samp{G} packet
37793 Write general registers. @xref{read registers packet}, for a
37794 description of the @var{XX@dots{}} data.
37804 @item H @var{op} @var{thread-id}
37805 @cindex @samp{H} packet
37806 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
37807 @samp{G}, et.al.). @var{op} depends on the operation to be performed:
37808 it should be @samp{c} for step and continue operations (note that this
37809 is deprecated, supporting the @samp{vCont} command is a better
37810 option), @samp{g} for other operations. The thread designator
37811 @var{thread-id} has the format and interpretation described in
37812 @ref{thread-id syntax}.
37823 @c 'H': How restrictive (or permissive) is the thread model. If a
37824 @c thread is selected and stopped, are other threads allowed
37825 @c to continue to execute? As I mentioned above, I think the
37826 @c semantics of each command when a thread is selected must be
37827 @c described. For example:
37829 @c 'g': If the stub supports threads and a specific thread is
37830 @c selected, returns the register block from that thread;
37831 @c otherwise returns current registers.
37833 @c 'G' If the stub supports threads and a specific thread is
37834 @c selected, sets the registers of the register block of
37835 @c that thread; otherwise sets current registers.
37837 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
37838 @anchor{cycle step packet}
37839 @cindex @samp{i} packet
37840 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
37841 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
37842 step starting at that address.
37845 @cindex @samp{I} packet
37846 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
37850 @cindex @samp{k} packet
37853 FIXME: @emph{There is no description of how to operate when a specific
37854 thread context has been selected (i.e.@: does 'k' kill only that
37857 @item m @var{addr},@var{length}
37858 @cindex @samp{m} packet
37859 Read @var{length} bytes of memory starting at address @var{addr}.
37860 Note that @var{addr} may not be aligned to any particular boundary.
37862 The stub need not use any particular size or alignment when gathering
37863 data from memory for the response; even if @var{addr} is word-aligned
37864 and @var{length} is a multiple of the word size, the stub is free to
37865 use byte accesses, or not. For this reason, this packet may not be
37866 suitable for accessing memory-mapped I/O devices.
37867 @cindex alignment of remote memory accesses
37868 @cindex size of remote memory accesses
37869 @cindex memory, alignment and size of remote accesses
37873 @item @var{XX@dots{}}
37874 Memory contents; each byte is transmitted as a two-digit hexadecimal
37875 number. The reply may contain fewer bytes than requested if the
37876 server was able to read only part of the region of memory.
37881 @item M @var{addr},@var{length}:@var{XX@dots{}}
37882 @cindex @samp{M} packet
37883 Write @var{length} bytes of memory starting at address @var{addr}.
37884 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
37885 hexadecimal number.
37892 for an error (this includes the case where only part of the data was
37897 @cindex @samp{p} packet
37898 Read the value of register @var{n}; @var{n} is in hex.
37899 @xref{read registers packet}, for a description of how the returned
37900 register value is encoded.
37904 @item @var{XX@dots{}}
37905 the register's value
37909 Indicating an unrecognized @var{query}.
37912 @item P @var{n@dots{}}=@var{r@dots{}}
37913 @anchor{write register packet}
37914 @cindex @samp{P} packet
37915 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
37916 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
37917 digits for each byte in the register (target byte order).
37927 @item q @var{name} @var{params}@dots{}
37928 @itemx Q @var{name} @var{params}@dots{}
37929 @cindex @samp{q} packet
37930 @cindex @samp{Q} packet
37931 General query (@samp{q}) and set (@samp{Q}). These packets are
37932 described fully in @ref{General Query Packets}.
37935 @cindex @samp{r} packet
37936 Reset the entire system.
37938 Don't use this packet; use the @samp{R} packet instead.
37941 @cindex @samp{R} packet
37942 Restart the program being debugged. @var{XX}, while needed, is ignored.
37943 This packet is only available in extended mode (@pxref{extended mode}).
37945 The @samp{R} packet has no reply.
37947 @item s @r{[}@var{addr}@r{]}
37948 @cindex @samp{s} packet
37949 Single step. @var{addr} is the address at which to resume. If
37950 @var{addr} is omitted, resume at same address.
37952 This packet is deprecated for multi-threading support. @xref{vCont
37956 @xref{Stop Reply Packets}, for the reply specifications.
37958 @item S @var{sig}@r{[};@var{addr}@r{]}
37959 @anchor{step with signal packet}
37960 @cindex @samp{S} packet
37961 Step with signal. This is analogous to the @samp{C} packet, but
37962 requests a single-step, rather than a normal resumption of execution.
37964 This packet is deprecated for multi-threading support. @xref{vCont
37968 @xref{Stop Reply Packets}, for the reply specifications.
37970 @item t @var{addr}:@var{PP},@var{MM}
37971 @cindex @samp{t} packet
37972 Search backwards starting at address @var{addr} for a match with pattern
37973 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
37974 @var{addr} must be at least 3 digits.
37976 @item T @var{thread-id}
37977 @cindex @samp{T} packet
37978 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
37983 thread is still alive
37989 Packets starting with @samp{v} are identified by a multi-letter name,
37990 up to the first @samp{;} or @samp{?} (or the end of the packet).
37992 @item vAttach;@var{pid}
37993 @cindex @samp{vAttach} packet
37994 Attach to a new process with the specified process ID @var{pid}.
37995 The process ID is a
37996 hexadecimal integer identifying the process. In all-stop mode, all
37997 threads in the attached process are stopped; in non-stop mode, it may be
37998 attached without being stopped if that is supported by the target.
38000 @c In non-stop mode, on a successful vAttach, the stub should set the
38001 @c current thread to a thread of the newly-attached process. After
38002 @c attaching, GDB queries for the attached process's thread ID with qC.
38003 @c Also note that, from a user perspective, whether or not the
38004 @c target is stopped on attach in non-stop mode depends on whether you
38005 @c use the foreground or background version of the attach command, not
38006 @c on what vAttach does; GDB does the right thing with respect to either
38007 @c stopping or restarting threads.
38009 This packet is only available in extended mode (@pxref{extended mode}).
38015 @item @r{Any stop packet}
38016 for success in all-stop mode (@pxref{Stop Reply Packets})
38018 for success in non-stop mode (@pxref{Remote Non-Stop})
38021 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
38022 @cindex @samp{vCont} packet
38023 @anchor{vCont packet}
38024 Resume the inferior, specifying different actions for each thread.
38025 If an action is specified with no @var{thread-id}, then it is applied to any
38026 threads that don't have a specific action specified; if no default action is
38027 specified then other threads should remain stopped in all-stop mode and
38028 in their current state in non-stop mode.
38029 Specifying multiple
38030 default actions is an error; specifying no actions is also an error.
38031 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
38033 Currently supported actions are:
38039 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
38043 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
38046 @item r @var{start},@var{end}
38047 Step once, and then keep stepping as long as the thread stops at
38048 addresses between @var{start} (inclusive) and @var{end} (exclusive).
38049 The remote stub reports a stop reply when either the thread goes out
38050 of the range or is stopped due to an unrelated reason, such as hitting
38051 a breakpoint. @xref{range stepping}.
38053 If the range is empty (@var{start} == @var{end}), then the action
38054 becomes equivalent to the @samp{s} action. In other words,
38055 single-step once, and report the stop (even if the stepped instruction
38056 jumps to @var{start}).
38058 (A stop reply may be sent at any point even if the PC is still within
38059 the stepping range; for example, it is valid to implement this packet
38060 in a degenerate way as a single instruction step operation.)
38064 The optional argument @var{addr} normally associated with the
38065 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
38066 not supported in @samp{vCont}.
38068 The @samp{t} action is only relevant in non-stop mode
38069 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
38070 A stop reply should be generated for any affected thread not already stopped.
38071 When a thread is stopped by means of a @samp{t} action,
38072 the corresponding stop reply should indicate that the thread has stopped with
38073 signal @samp{0}, regardless of whether the target uses some other signal
38074 as an implementation detail.
38076 The stub must support @samp{vCont} if it reports support for
38077 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
38078 this case @samp{vCont} actions can be specified to apply to all threads
38079 in a process by using the @samp{p@var{pid}.-1} form of the
38083 @xref{Stop Reply Packets}, for the reply specifications.
38086 @cindex @samp{vCont?} packet
38087 Request a list of actions supported by the @samp{vCont} packet.
38091 @item vCont@r{[};@var{action}@dots{}@r{]}
38092 The @samp{vCont} packet is supported. Each @var{action} is a supported
38093 command in the @samp{vCont} packet.
38095 The @samp{vCont} packet is not supported.
38098 @item vFile:@var{operation}:@var{parameter}@dots{}
38099 @cindex @samp{vFile} packet
38100 Perform a file operation on the target system. For details,
38101 see @ref{Host I/O Packets}.
38103 @item vFlashErase:@var{addr},@var{length}
38104 @cindex @samp{vFlashErase} packet
38105 Direct the stub to erase @var{length} bytes of flash starting at
38106 @var{addr}. The region may enclose any number of flash blocks, but
38107 its start and end must fall on block boundaries, as indicated by the
38108 flash block size appearing in the memory map (@pxref{Memory Map
38109 Format}). @value{GDBN} groups flash memory programming operations
38110 together, and sends a @samp{vFlashDone} request after each group; the
38111 stub is allowed to delay erase operation until the @samp{vFlashDone}
38112 packet is received.
38122 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
38123 @cindex @samp{vFlashWrite} packet
38124 Direct the stub to write data to flash address @var{addr}. The data
38125 is passed in binary form using the same encoding as for the @samp{X}
38126 packet (@pxref{Binary Data}). The memory ranges specified by
38127 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
38128 not overlap, and must appear in order of increasing addresses
38129 (although @samp{vFlashErase} packets for higher addresses may already
38130 have been received; the ordering is guaranteed only between
38131 @samp{vFlashWrite} packets). If a packet writes to an address that was
38132 neither erased by a preceding @samp{vFlashErase} packet nor by some other
38133 target-specific method, the results are unpredictable.
38141 for vFlashWrite addressing non-flash memory
38147 @cindex @samp{vFlashDone} packet
38148 Indicate to the stub that flash programming operation is finished.
38149 The stub is permitted to delay or batch the effects of a group of
38150 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
38151 @samp{vFlashDone} packet is received. The contents of the affected
38152 regions of flash memory are unpredictable until the @samp{vFlashDone}
38153 request is completed.
38155 @item vKill;@var{pid}
38156 @cindex @samp{vKill} packet
38157 Kill the process with the specified process ID. @var{pid} is a
38158 hexadecimal integer identifying the process. This packet is used in
38159 preference to @samp{k} when multiprocess protocol extensions are
38160 supported; see @ref{multiprocess extensions}.
38170 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
38171 @cindex @samp{vRun} packet
38172 Run the program @var{filename}, passing it each @var{argument} on its
38173 command line. The file and arguments are hex-encoded strings. If
38174 @var{filename} is an empty string, the stub may use a default program
38175 (e.g.@: the last program run). The program is created in the stopped
38178 @c FIXME: What about non-stop mode?
38180 This packet is only available in extended mode (@pxref{extended mode}).
38186 @item @r{Any stop packet}
38187 for success (@pxref{Stop Reply Packets})
38191 @cindex @samp{vStopped} packet
38192 @xref{Notification Packets}.
38194 @item X @var{addr},@var{length}:@var{XX@dots{}}
38196 @cindex @samp{X} packet
38197 Write data to memory, where the data is transmitted in binary.
38198 @var{addr} is address, @var{length} is number of bytes,
38199 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
38209 @item z @var{type},@var{addr},@var{kind}
38210 @itemx Z @var{type},@var{addr},@var{kind}
38211 @anchor{insert breakpoint or watchpoint packet}
38212 @cindex @samp{z} packet
38213 @cindex @samp{Z} packets
38214 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
38215 watchpoint starting at address @var{address} of kind @var{kind}.
38217 Each breakpoint and watchpoint packet @var{type} is documented
38220 @emph{Implementation notes: A remote target shall return an empty string
38221 for an unrecognized breakpoint or watchpoint packet @var{type}. A
38222 remote target shall support either both or neither of a given
38223 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
38224 avoid potential problems with duplicate packets, the operations should
38225 be implemented in an idempotent way.}
38227 @item z0,@var{addr},@var{kind}
38228 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38229 @cindex @samp{z0} packet
38230 @cindex @samp{Z0} packet
38231 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
38232 @var{addr} of type @var{kind}.
38234 A memory breakpoint is implemented by replacing the instruction at
38235 @var{addr} with a software breakpoint or trap instruction. The
38236 @var{kind} is target-specific and typically indicates the size of
38237 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
38238 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
38239 architectures have additional meanings for @var{kind};
38240 @var{cond_list} is an optional list of conditional expressions in bytecode
38241 form that should be evaluated on the target's side. These are the
38242 conditions that should be taken into consideration when deciding if
38243 the breakpoint trigger should be reported back to @var{GDBN}.
38245 The @var{cond_list} parameter is comprised of a series of expressions,
38246 concatenated without separators. Each expression has the following form:
38250 @item X @var{len},@var{expr}
38251 @var{len} is the length of the bytecode expression and @var{expr} is the
38252 actual conditional expression in bytecode form.
38256 The optional @var{cmd_list} parameter introduces commands that may be
38257 run on the target, rather than being reported back to @value{GDBN}.
38258 The parameter starts with a numeric flag @var{persist}; if the flag is
38259 nonzero, then the breakpoint may remain active and the commands
38260 continue to be run even when @value{GDBN} disconnects from the target.
38261 Following this flag is a series of expressions concatenated with no
38262 separators. Each expression has the following form:
38266 @item X @var{len},@var{expr}
38267 @var{len} is the length of the bytecode expression and @var{expr} is the
38268 actual conditional expression in bytecode form.
38272 see @ref{Architecture-Specific Protocol Details}.
38274 @emph{Implementation note: It is possible for a target to copy or move
38275 code that contains memory breakpoints (e.g., when implementing
38276 overlays). The behavior of this packet, in the presence of such a
38277 target, is not defined.}
38289 @item z1,@var{addr},@var{kind}
38290 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}
38291 @cindex @samp{z1} packet
38292 @cindex @samp{Z1} packet
38293 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
38294 address @var{addr}.
38296 A hardware breakpoint is implemented using a mechanism that is not
38297 dependant on being able to modify the target's memory. @var{kind}
38298 and @var{cond_list} have the same meaning as in @samp{Z0} packets.
38300 @emph{Implementation note: A hardware breakpoint is not affected by code
38313 @item z2,@var{addr},@var{kind}
38314 @itemx Z2,@var{addr},@var{kind}
38315 @cindex @samp{z2} packet
38316 @cindex @samp{Z2} packet
38317 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
38318 @var{kind} is interpreted as the number of bytes to watch.
38330 @item z3,@var{addr},@var{kind}
38331 @itemx Z3,@var{addr},@var{kind}
38332 @cindex @samp{z3} packet
38333 @cindex @samp{Z3} packet
38334 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
38335 @var{kind} is interpreted as the number of bytes to watch.
38347 @item z4,@var{addr},@var{kind}
38348 @itemx Z4,@var{addr},@var{kind}
38349 @cindex @samp{z4} packet
38350 @cindex @samp{Z4} packet
38351 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
38352 @var{kind} is interpreted as the number of bytes to watch.
38366 @node Stop Reply Packets
38367 @section Stop Reply Packets
38368 @cindex stop reply packets
38370 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
38371 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
38372 receive any of the below as a reply. Except for @samp{?}
38373 and @samp{vStopped}, that reply is only returned
38374 when the target halts. In the below the exact meaning of @dfn{signal
38375 number} is defined by the header @file{include/gdb/signals.h} in the
38376 @value{GDBN} source code.
38378 As in the description of request packets, we include spaces in the
38379 reply templates for clarity; these are not part of the reply packet's
38380 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
38386 The program received signal number @var{AA} (a two-digit hexadecimal
38387 number). This is equivalent to a @samp{T} response with no
38388 @var{n}:@var{r} pairs.
38390 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
38391 @cindex @samp{T} packet reply
38392 The program received signal number @var{AA} (a two-digit hexadecimal
38393 number). This is equivalent to an @samp{S} response, except that the
38394 @samp{@var{n}:@var{r}} pairs can carry values of important registers
38395 and other information directly in the stop reply packet, reducing
38396 round-trip latency. Single-step and breakpoint traps are reported
38397 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
38401 If @var{n} is a hexadecimal number, it is a register number, and the
38402 corresponding @var{r} gives that register's value. @var{r} is a
38403 series of bytes in target byte order, with each byte given by a
38404 two-digit hex number.
38407 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
38408 the stopped thread, as specified in @ref{thread-id syntax}.
38411 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
38412 the core on which the stop event was detected.
38415 If @var{n} is a recognized @dfn{stop reason}, it describes a more
38416 specific event that stopped the target. The currently defined stop
38417 reasons are listed below. @var{aa} should be @samp{05}, the trap
38418 signal. At most one stop reason should be present.
38421 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
38422 and go on to the next; this allows us to extend the protocol in the
38426 The currently defined stop reasons are:
38432 The packet indicates a watchpoint hit, and @var{r} is the data address, in
38435 @cindex shared library events, remote reply
38437 The packet indicates that the loaded libraries have changed.
38438 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
38439 list of loaded libraries. @var{r} is ignored.
38441 @cindex replay log events, remote reply
38443 The packet indicates that the target cannot continue replaying
38444 logged execution events, because it has reached the end (or the
38445 beginning when executing backward) of the log. The value of @var{r}
38446 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
38447 for more information.
38451 @itemx W @var{AA} ; process:@var{pid}
38452 The process exited, and @var{AA} is the exit status. This is only
38453 applicable to certain targets.
38455 The second form of the response, including the process ID of the exited
38456 process, can be used only when @value{GDBN} has reported support for
38457 multiprocess protocol extensions; see @ref{multiprocess extensions}.
38458 The @var{pid} is formatted as a big-endian hex string.
38461 @itemx X @var{AA} ; process:@var{pid}
38462 The process terminated with signal @var{AA}.
38464 The second form of the response, including the process ID of the
38465 terminated process, can be used only when @value{GDBN} has reported
38466 support for multiprocess protocol extensions; see @ref{multiprocess
38467 extensions}. The @var{pid} is formatted as a big-endian hex string.
38469 @item O @var{XX}@dots{}
38470 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
38471 written as the program's console output. This can happen at any time
38472 while the program is running and the debugger should continue to wait
38473 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
38475 @item F @var{call-id},@var{parameter}@dots{}
38476 @var{call-id} is the identifier which says which host system call should
38477 be called. This is just the name of the function. Translation into the
38478 correct system call is only applicable as it's defined in @value{GDBN}.
38479 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
38482 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
38483 this very system call.
38485 The target replies with this packet when it expects @value{GDBN} to
38486 call a host system call on behalf of the target. @value{GDBN} replies
38487 with an appropriate @samp{F} packet and keeps up waiting for the next
38488 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
38489 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
38490 Protocol Extension}, for more details.
38494 @node General Query Packets
38495 @section General Query Packets
38496 @cindex remote query requests
38498 Packets starting with @samp{q} are @dfn{general query packets};
38499 packets starting with @samp{Q} are @dfn{general set packets}. General
38500 query and set packets are a semi-unified form for retrieving and
38501 sending information to and from the stub.
38503 The initial letter of a query or set packet is followed by a name
38504 indicating what sort of thing the packet applies to. For example,
38505 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
38506 definitions with the stub. These packet names follow some
38511 The name must not contain commas, colons or semicolons.
38513 Most @value{GDBN} query and set packets have a leading upper case
38516 The names of custom vendor packets should use a company prefix, in
38517 lower case, followed by a period. For example, packets designed at
38518 the Acme Corporation might begin with @samp{qacme.foo} (for querying
38519 foos) or @samp{Qacme.bar} (for setting bars).
38522 The name of a query or set packet should be separated from any
38523 parameters by a @samp{:}; the parameters themselves should be
38524 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
38525 full packet name, and check for a separator or the end of the packet,
38526 in case two packet names share a common prefix. New packets should not begin
38527 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
38528 packets predate these conventions, and have arguments without any terminator
38529 for the packet name; we suspect they are in widespread use in places that
38530 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
38531 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
38534 Like the descriptions of the other packets, each description here
38535 has a template showing the packet's overall syntax, followed by an
38536 explanation of the packet's meaning. We include spaces in some of the
38537 templates for clarity; these are not part of the packet's syntax. No
38538 @value{GDBN} packet uses spaces to separate its components.
38540 Here are the currently defined query and set packets:
38546 Turn on or off the agent as a helper to perform some debugging operations
38547 delegated from @value{GDBN} (@pxref{Control Agent}).
38549 @item QAllow:@var{op}:@var{val}@dots{}
38550 @cindex @samp{QAllow} packet
38551 Specify which operations @value{GDBN} expects to request of the
38552 target, as a semicolon-separated list of operation name and value
38553 pairs. Possible values for @var{op} include @samp{WriteReg},
38554 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
38555 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
38556 indicating that @value{GDBN} will not request the operation, or 1,
38557 indicating that it may. (The target can then use this to set up its
38558 own internals optimally, for instance if the debugger never expects to
38559 insert breakpoints, it may not need to install its own trap handler.)
38562 @cindex current thread, remote request
38563 @cindex @samp{qC} packet
38564 Return the current thread ID.
38568 @item QC @var{thread-id}
38569 Where @var{thread-id} is a thread ID as documented in
38570 @ref{thread-id syntax}.
38571 @item @r{(anything else)}
38572 Any other reply implies the old thread ID.
38575 @item qCRC:@var{addr},@var{length}
38576 @cindex CRC of memory block, remote request
38577 @cindex @samp{qCRC} packet
38578 Compute the CRC checksum of a block of memory using CRC-32 defined in
38579 IEEE 802.3. The CRC is computed byte at a time, taking the most
38580 significant bit of each byte first. The initial pattern code
38581 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
38583 @emph{Note:} This is the same CRC used in validating separate debug
38584 files (@pxref{Separate Debug Files, , Debugging Information in Separate
38585 Files}). However the algorithm is slightly different. When validating
38586 separate debug files, the CRC is computed taking the @emph{least}
38587 significant bit of each byte first, and the final result is inverted to
38588 detect trailing zeros.
38593 An error (such as memory fault)
38594 @item C @var{crc32}
38595 The specified memory region's checksum is @var{crc32}.
38598 @item QDisableRandomization:@var{value}
38599 @cindex disable address space randomization, remote request
38600 @cindex @samp{QDisableRandomization} packet
38601 Some target operating systems will randomize the virtual address space
38602 of the inferior process as a security feature, but provide a feature
38603 to disable such randomization, e.g.@: to allow for a more deterministic
38604 debugging experience. On such systems, this packet with a @var{value}
38605 of 1 directs the target to disable address space randomization for
38606 processes subsequently started via @samp{vRun} packets, while a packet
38607 with a @var{value} of 0 tells the target to enable address space
38610 This packet is only available in extended mode (@pxref{extended mode}).
38615 The request succeeded.
38618 An error occurred. @var{nn} are hex digits.
38621 An empty reply indicates that @samp{QDisableRandomization} is not supported
38625 This packet is not probed by default; the remote stub must request it,
38626 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38627 This should only be done on targets that actually support disabling
38628 address space randomization.
38631 @itemx qsThreadInfo
38632 @cindex list active threads, remote request
38633 @cindex @samp{qfThreadInfo} packet
38634 @cindex @samp{qsThreadInfo} packet
38635 Obtain a list of all active thread IDs from the target (OS). Since there
38636 may be too many active threads to fit into one reply packet, this query
38637 works iteratively: it may require more than one query/reply sequence to
38638 obtain the entire list of threads. The first query of the sequence will
38639 be the @samp{qfThreadInfo} query; subsequent queries in the
38640 sequence will be the @samp{qsThreadInfo} query.
38642 NOTE: This packet replaces the @samp{qL} query (see below).
38646 @item m @var{thread-id}
38648 @item m @var{thread-id},@var{thread-id}@dots{}
38649 a comma-separated list of thread IDs
38651 (lower case letter @samp{L}) denotes end of list.
38654 In response to each query, the target will reply with a list of one or
38655 more thread IDs, separated by commas.
38656 @value{GDBN} will respond to each reply with a request for more thread
38657 ids (using the @samp{qs} form of the query), until the target responds
38658 with @samp{l} (lower-case ell, for @dfn{last}).
38659 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
38662 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
38663 @cindex get thread-local storage address, remote request
38664 @cindex @samp{qGetTLSAddr} packet
38665 Fetch the address associated with thread local storage specified
38666 by @var{thread-id}, @var{offset}, and @var{lm}.
38668 @var{thread-id} is the thread ID associated with the
38669 thread for which to fetch the TLS address. @xref{thread-id syntax}.
38671 @var{offset} is the (big endian, hex encoded) offset associated with the
38672 thread local variable. (This offset is obtained from the debug
38673 information associated with the variable.)
38675 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
38676 load module associated with the thread local storage. For example,
38677 a @sc{gnu}/Linux system will pass the link map address of the shared
38678 object associated with the thread local storage under consideration.
38679 Other operating environments may choose to represent the load module
38680 differently, so the precise meaning of this parameter will vary.
38684 @item @var{XX}@dots{}
38685 Hex encoded (big endian) bytes representing the address of the thread
38686 local storage requested.
38689 An error occurred. @var{nn} are hex digits.
38692 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
38695 @item qGetTIBAddr:@var{thread-id}
38696 @cindex get thread information block address
38697 @cindex @samp{qGetTIBAddr} packet
38698 Fetch address of the Windows OS specific Thread Information Block.
38700 @var{thread-id} is the thread ID associated with the thread.
38704 @item @var{XX}@dots{}
38705 Hex encoded (big endian) bytes representing the linear address of the
38706 thread information block.
38709 An error occured. This means that either the thread was not found, or the
38710 address could not be retrieved.
38713 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
38716 @item qL @var{startflag} @var{threadcount} @var{nextthread}
38717 Obtain thread information from RTOS. Where: @var{startflag} (one hex
38718 digit) is one to indicate the first query and zero to indicate a
38719 subsequent query; @var{threadcount} (two hex digits) is the maximum
38720 number of threads the response packet can contain; and @var{nextthread}
38721 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
38722 returned in the response as @var{argthread}.
38724 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
38728 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
38729 Where: @var{count} (two hex digits) is the number of threads being
38730 returned; @var{done} (one hex digit) is zero to indicate more threads
38731 and one indicates no further threads; @var{argthreadid} (eight hex
38732 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
38733 is a sequence of thread IDs from the target. @var{threadid} (eight hex
38734 digits). See @code{remote.c:parse_threadlist_response()}.
38738 @cindex section offsets, remote request
38739 @cindex @samp{qOffsets} packet
38740 Get section offsets that the target used when relocating the downloaded
38745 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
38746 Relocate the @code{Text} section by @var{xxx} from its original address.
38747 Relocate the @code{Data} section by @var{yyy} from its original address.
38748 If the object file format provides segment information (e.g.@: @sc{elf}
38749 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
38750 segments by the supplied offsets.
38752 @emph{Note: while a @code{Bss} offset may be included in the response,
38753 @value{GDBN} ignores this and instead applies the @code{Data} offset
38754 to the @code{Bss} section.}
38756 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
38757 Relocate the first segment of the object file, which conventionally
38758 contains program code, to a starting address of @var{xxx}. If
38759 @samp{DataSeg} is specified, relocate the second segment, which
38760 conventionally contains modifiable data, to a starting address of
38761 @var{yyy}. @value{GDBN} will report an error if the object file
38762 does not contain segment information, or does not contain at least
38763 as many segments as mentioned in the reply. Extra segments are
38764 kept at fixed offsets relative to the last relocated segment.
38767 @item qP @var{mode} @var{thread-id}
38768 @cindex thread information, remote request
38769 @cindex @samp{qP} packet
38770 Returns information on @var{thread-id}. Where: @var{mode} is a hex
38771 encoded 32 bit mode; @var{thread-id} is a thread ID
38772 (@pxref{thread-id syntax}).
38774 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
38777 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
38781 @cindex non-stop mode, remote request
38782 @cindex @samp{QNonStop} packet
38784 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
38785 @xref{Remote Non-Stop}, for more information.
38790 The request succeeded.
38793 An error occurred. @var{nn} are hex digits.
38796 An empty reply indicates that @samp{QNonStop} is not supported by
38800 This packet is not probed by default; the remote stub must request it,
38801 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38802 Use of this packet is controlled by the @code{set non-stop} command;
38803 @pxref{Non-Stop Mode}.
38805 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
38806 @cindex pass signals to inferior, remote request
38807 @cindex @samp{QPassSignals} packet
38808 @anchor{QPassSignals}
38809 Each listed @var{signal} should be passed directly to the inferior process.
38810 Signals are numbered identically to continue packets and stop replies
38811 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
38812 strictly greater than the previous item. These signals do not need to stop
38813 the inferior, or be reported to @value{GDBN}. All other signals should be
38814 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
38815 combine; any earlier @samp{QPassSignals} list is completely replaced by the
38816 new list. This packet improves performance when using @samp{handle
38817 @var{signal} nostop noprint pass}.
38822 The request succeeded.
38825 An error occurred. @var{nn} are hex digits.
38828 An empty reply indicates that @samp{QPassSignals} is not supported by
38832 Use of this packet is controlled by the @code{set remote pass-signals}
38833 command (@pxref{Remote Configuration, set remote pass-signals}).
38834 This packet is not probed by default; the remote stub must request it,
38835 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38837 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
38838 @cindex signals the inferior may see, remote request
38839 @cindex @samp{QProgramSignals} packet
38840 @anchor{QProgramSignals}
38841 Each listed @var{signal} may be delivered to the inferior process.
38842 Others should be silently discarded.
38844 In some cases, the remote stub may need to decide whether to deliver a
38845 signal to the program or not without @value{GDBN} involvement. One
38846 example of that is while detaching --- the program's threads may have
38847 stopped for signals that haven't yet had a chance of being reported to
38848 @value{GDBN}, and so the remote stub can use the signal list specified
38849 by this packet to know whether to deliver or ignore those pending
38852 This does not influence whether to deliver a signal as requested by a
38853 resumption packet (@pxref{vCont packet}).
38855 Signals are numbered identically to continue packets and stop replies
38856 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
38857 strictly greater than the previous item. Multiple
38858 @samp{QProgramSignals} packets do not combine; any earlier
38859 @samp{QProgramSignals} list is completely replaced by the new list.
38864 The request succeeded.
38867 An error occurred. @var{nn} are hex digits.
38870 An empty reply indicates that @samp{QProgramSignals} is not supported
38874 Use of this packet is controlled by the @code{set remote program-signals}
38875 command (@pxref{Remote Configuration, set remote program-signals}).
38876 This packet is not probed by default; the remote stub must request it,
38877 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38879 @item qRcmd,@var{command}
38880 @cindex execute remote command, remote request
38881 @cindex @samp{qRcmd} packet
38882 @var{command} (hex encoded) is passed to the local interpreter for
38883 execution. Invalid commands should be reported using the output
38884 string. Before the final result packet, the target may also respond
38885 with a number of intermediate @samp{O@var{output}} console output
38886 packets. @emph{Implementors should note that providing access to a
38887 stubs's interpreter may have security implications}.
38892 A command response with no output.
38894 A command response with the hex encoded output string @var{OUTPUT}.
38896 Indicate a badly formed request.
38898 An empty reply indicates that @samp{qRcmd} is not recognized.
38901 (Note that the @code{qRcmd} packet's name is separated from the
38902 command by a @samp{,}, not a @samp{:}, contrary to the naming
38903 conventions above. Please don't use this packet as a model for new
38906 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
38907 @cindex searching memory, in remote debugging
38909 @cindex @samp{qSearch:memory} packet
38911 @cindex @samp{qSearch memory} packet
38912 @anchor{qSearch memory}
38913 Search @var{length} bytes at @var{address} for @var{search-pattern}.
38914 @var{address} and @var{length} are encoded in hex.
38915 @var{search-pattern} is a sequence of bytes, hex encoded.
38920 The pattern was not found.
38922 The pattern was found at @var{address}.
38924 A badly formed request or an error was encountered while searching memory.
38926 An empty reply indicates that @samp{qSearch:memory} is not recognized.
38929 @item QStartNoAckMode
38930 @cindex @samp{QStartNoAckMode} packet
38931 @anchor{QStartNoAckMode}
38932 Request that the remote stub disable the normal @samp{+}/@samp{-}
38933 protocol acknowledgments (@pxref{Packet Acknowledgment}).
38938 The stub has switched to no-acknowledgment mode.
38939 @value{GDBN} acknowledges this reponse,
38940 but neither the stub nor @value{GDBN} shall send or expect further
38941 @samp{+}/@samp{-} acknowledgments in the current connection.
38943 An empty reply indicates that the stub does not support no-acknowledgment mode.
38946 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
38947 @cindex supported packets, remote query
38948 @cindex features of the remote protocol
38949 @cindex @samp{qSupported} packet
38950 @anchor{qSupported}
38951 Tell the remote stub about features supported by @value{GDBN}, and
38952 query the stub for features it supports. This packet allows
38953 @value{GDBN} and the remote stub to take advantage of each others'
38954 features. @samp{qSupported} also consolidates multiple feature probes
38955 at startup, to improve @value{GDBN} performance---a single larger
38956 packet performs better than multiple smaller probe packets on
38957 high-latency links. Some features may enable behavior which must not
38958 be on by default, e.g.@: because it would confuse older clients or
38959 stubs. Other features may describe packets which could be
38960 automatically probed for, but are not. These features must be
38961 reported before @value{GDBN} will use them. This ``default
38962 unsupported'' behavior is not appropriate for all packets, but it
38963 helps to keep the initial connection time under control with new
38964 versions of @value{GDBN} which support increasing numbers of packets.
38968 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
38969 The stub supports or does not support each returned @var{stubfeature},
38970 depending on the form of each @var{stubfeature} (see below for the
38973 An empty reply indicates that @samp{qSupported} is not recognized,
38974 or that no features needed to be reported to @value{GDBN}.
38977 The allowed forms for each feature (either a @var{gdbfeature} in the
38978 @samp{qSupported} packet, or a @var{stubfeature} in the response)
38982 @item @var{name}=@var{value}
38983 The remote protocol feature @var{name} is supported, and associated
38984 with the specified @var{value}. The format of @var{value} depends
38985 on the feature, but it must not include a semicolon.
38987 The remote protocol feature @var{name} is supported, and does not
38988 need an associated value.
38990 The remote protocol feature @var{name} is not supported.
38992 The remote protocol feature @var{name} may be supported, and
38993 @value{GDBN} should auto-detect support in some other way when it is
38994 needed. This form will not be used for @var{gdbfeature} notifications,
38995 but may be used for @var{stubfeature} responses.
38998 Whenever the stub receives a @samp{qSupported} request, the
38999 supplied set of @value{GDBN} features should override any previous
39000 request. This allows @value{GDBN} to put the stub in a known
39001 state, even if the stub had previously been communicating with
39002 a different version of @value{GDBN}.
39004 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
39009 This feature indicates whether @value{GDBN} supports multiprocess
39010 extensions to the remote protocol. @value{GDBN} does not use such
39011 extensions unless the stub also reports that it supports them by
39012 including @samp{multiprocess+} in its @samp{qSupported} reply.
39013 @xref{multiprocess extensions}, for details.
39016 This feature indicates that @value{GDBN} supports the XML target
39017 description. If the stub sees @samp{xmlRegisters=} with target
39018 specific strings separated by a comma, it will report register
39022 This feature indicates whether @value{GDBN} supports the
39023 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
39024 instruction reply packet}).
39027 Stubs should ignore any unknown values for
39028 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
39029 packet supports receiving packets of unlimited length (earlier
39030 versions of @value{GDBN} may reject overly long responses). Additional values
39031 for @var{gdbfeature} may be defined in the future to let the stub take
39032 advantage of new features in @value{GDBN}, e.g.@: incompatible
39033 improvements in the remote protocol---the @samp{multiprocess} feature is
39034 an example of such a feature. The stub's reply should be independent
39035 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
39036 describes all the features it supports, and then the stub replies with
39037 all the features it supports.
39039 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
39040 responses, as long as each response uses one of the standard forms.
39042 Some features are flags. A stub which supports a flag feature
39043 should respond with a @samp{+} form response. Other features
39044 require values, and the stub should respond with an @samp{=}
39047 Each feature has a default value, which @value{GDBN} will use if
39048 @samp{qSupported} is not available or if the feature is not mentioned
39049 in the @samp{qSupported} response. The default values are fixed; a
39050 stub is free to omit any feature responses that match the defaults.
39052 Not all features can be probed, but for those which can, the probing
39053 mechanism is useful: in some cases, a stub's internal
39054 architecture may not allow the protocol layer to know some information
39055 about the underlying target in advance. This is especially common in
39056 stubs which may be configured for multiple targets.
39058 These are the currently defined stub features and their properties:
39060 @multitable @columnfractions 0.35 0.2 0.12 0.2
39061 @c NOTE: The first row should be @headitem, but we do not yet require
39062 @c a new enough version of Texinfo (4.7) to use @headitem.
39064 @tab Value Required
39068 @item @samp{PacketSize}
39073 @item @samp{qXfer:auxv:read}
39078 @item @samp{qXfer:btrace:read}
39083 @item @samp{qXfer:features:read}
39088 @item @samp{qXfer:libraries:read}
39093 @item @samp{qXfer:libraries-svr4:read}
39098 @item @samp{augmented-libraries-svr4-read}
39103 @item @samp{qXfer:memory-map:read}
39108 @item @samp{qXfer:sdata:read}
39113 @item @samp{qXfer:spu:read}
39118 @item @samp{qXfer:spu:write}
39123 @item @samp{qXfer:siginfo:read}
39128 @item @samp{qXfer:siginfo:write}
39133 @item @samp{qXfer:threads:read}
39138 @item @samp{qXfer:traceframe-info:read}
39143 @item @samp{qXfer:uib:read}
39148 @item @samp{qXfer:fdpic:read}
39153 @item @samp{Qbtrace:off}
39158 @item @samp{Qbtrace:bts}
39163 @item @samp{QNonStop}
39168 @item @samp{QPassSignals}
39173 @item @samp{QStartNoAckMode}
39178 @item @samp{multiprocess}
39183 @item @samp{ConditionalBreakpoints}
39188 @item @samp{ConditionalTracepoints}
39193 @item @samp{ReverseContinue}
39198 @item @samp{ReverseStep}
39203 @item @samp{TracepointSource}
39208 @item @samp{QAgent}
39213 @item @samp{QAllow}
39218 @item @samp{QDisableRandomization}
39223 @item @samp{EnableDisableTracepoints}
39228 @item @samp{QTBuffer:size}
39233 @item @samp{tracenz}
39238 @item @samp{BreakpointCommands}
39245 These are the currently defined stub features, in more detail:
39248 @cindex packet size, remote protocol
39249 @item PacketSize=@var{bytes}
39250 The remote stub can accept packets up to at least @var{bytes} in
39251 length. @value{GDBN} will send packets up to this size for bulk
39252 transfers, and will never send larger packets. This is a limit on the
39253 data characters in the packet, including the frame and checksum.
39254 There is no trailing NUL byte in a remote protocol packet; if the stub
39255 stores packets in a NUL-terminated format, it should allow an extra
39256 byte in its buffer for the NUL. If this stub feature is not supported,
39257 @value{GDBN} guesses based on the size of the @samp{g} packet response.
39259 @item qXfer:auxv:read
39260 The remote stub understands the @samp{qXfer:auxv:read} packet
39261 (@pxref{qXfer auxiliary vector read}).
39263 @item qXfer:btrace:read
39264 The remote stub understands the @samp{qXfer:btrace:read}
39265 packet (@pxref{qXfer btrace read}).
39267 @item qXfer:features:read
39268 The remote stub understands the @samp{qXfer:features:read} packet
39269 (@pxref{qXfer target description read}).
39271 @item qXfer:libraries:read
39272 The remote stub understands the @samp{qXfer:libraries:read} packet
39273 (@pxref{qXfer library list read}).
39275 @item qXfer:libraries-svr4:read
39276 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
39277 (@pxref{qXfer svr4 library list read}).
39279 @item augmented-libraries-svr4-read
39280 The remote stub understands the augmented form of the
39281 @samp{qXfer:libraries-svr4:read} packet
39282 (@pxref{qXfer svr4 library list read}).
39284 @item qXfer:memory-map:read
39285 The remote stub understands the @samp{qXfer:memory-map:read} packet
39286 (@pxref{qXfer memory map read}).
39288 @item qXfer:sdata:read
39289 The remote stub understands the @samp{qXfer:sdata:read} packet
39290 (@pxref{qXfer sdata read}).
39292 @item qXfer:spu:read
39293 The remote stub understands the @samp{qXfer:spu:read} packet
39294 (@pxref{qXfer spu read}).
39296 @item qXfer:spu:write
39297 The remote stub understands the @samp{qXfer:spu:write} packet
39298 (@pxref{qXfer spu write}).
39300 @item qXfer:siginfo:read
39301 The remote stub understands the @samp{qXfer:siginfo:read} packet
39302 (@pxref{qXfer siginfo read}).
39304 @item qXfer:siginfo:write
39305 The remote stub understands the @samp{qXfer:siginfo:write} packet
39306 (@pxref{qXfer siginfo write}).
39308 @item qXfer:threads:read
39309 The remote stub understands the @samp{qXfer:threads:read} packet
39310 (@pxref{qXfer threads read}).
39312 @item qXfer:traceframe-info:read
39313 The remote stub understands the @samp{qXfer:traceframe-info:read}
39314 packet (@pxref{qXfer traceframe info read}).
39316 @item qXfer:uib:read
39317 The remote stub understands the @samp{qXfer:uib:read}
39318 packet (@pxref{qXfer unwind info block}).
39320 @item qXfer:fdpic:read
39321 The remote stub understands the @samp{qXfer:fdpic:read}
39322 packet (@pxref{qXfer fdpic loadmap read}).
39325 The remote stub understands the @samp{QNonStop} packet
39326 (@pxref{QNonStop}).
39329 The remote stub understands the @samp{QPassSignals} packet
39330 (@pxref{QPassSignals}).
39332 @item QStartNoAckMode
39333 The remote stub understands the @samp{QStartNoAckMode} packet and
39334 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
39337 @anchor{multiprocess extensions}
39338 @cindex multiprocess extensions, in remote protocol
39339 The remote stub understands the multiprocess extensions to the remote
39340 protocol syntax. The multiprocess extensions affect the syntax of
39341 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
39342 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
39343 replies. Note that reporting this feature indicates support for the
39344 syntactic extensions only, not that the stub necessarily supports
39345 debugging of more than one process at a time. The stub must not use
39346 multiprocess extensions in packet replies unless @value{GDBN} has also
39347 indicated it supports them in its @samp{qSupported} request.
39349 @item qXfer:osdata:read
39350 The remote stub understands the @samp{qXfer:osdata:read} packet
39351 ((@pxref{qXfer osdata read}).
39353 @item ConditionalBreakpoints
39354 The target accepts and implements evaluation of conditional expressions
39355 defined for breakpoints. The target will only report breakpoint triggers
39356 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
39358 @item ConditionalTracepoints
39359 The remote stub accepts and implements conditional expressions defined
39360 for tracepoints (@pxref{Tracepoint Conditions}).
39362 @item ReverseContinue
39363 The remote stub accepts and implements the reverse continue packet
39367 The remote stub accepts and implements the reverse step packet
39370 @item TracepointSource
39371 The remote stub understands the @samp{QTDPsrc} packet that supplies
39372 the source form of tracepoint definitions.
39375 The remote stub understands the @samp{QAgent} packet.
39378 The remote stub understands the @samp{QAllow} packet.
39380 @item QDisableRandomization
39381 The remote stub understands the @samp{QDisableRandomization} packet.
39383 @item StaticTracepoint
39384 @cindex static tracepoints, in remote protocol
39385 The remote stub supports static tracepoints.
39387 @item InstallInTrace
39388 @anchor{install tracepoint in tracing}
39389 The remote stub supports installing tracepoint in tracing.
39391 @item EnableDisableTracepoints
39392 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
39393 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
39394 to be enabled and disabled while a trace experiment is running.
39396 @item QTBuffer:size
39397 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
39398 packet that allows to change the size of the trace buffer.
39401 @cindex string tracing, in remote protocol
39402 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
39403 See @ref{Bytecode Descriptions} for details about the bytecode.
39405 @item BreakpointCommands
39406 @cindex breakpoint commands, in remote protocol
39407 The remote stub supports running a breakpoint's command list itself,
39408 rather than reporting the hit to @value{GDBN}.
39411 The remote stub understands the @samp{Qbtrace:off} packet.
39414 The remote stub understands the @samp{Qbtrace:bts} packet.
39419 @cindex symbol lookup, remote request
39420 @cindex @samp{qSymbol} packet
39421 Notify the target that @value{GDBN} is prepared to serve symbol lookup
39422 requests. Accept requests from the target for the values of symbols.
39427 The target does not need to look up any (more) symbols.
39428 @item qSymbol:@var{sym_name}
39429 The target requests the value of symbol @var{sym_name} (hex encoded).
39430 @value{GDBN} may provide the value by using the
39431 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
39435 @item qSymbol:@var{sym_value}:@var{sym_name}
39436 Set the value of @var{sym_name} to @var{sym_value}.
39438 @var{sym_name} (hex encoded) is the name of a symbol whose value the
39439 target has previously requested.
39441 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
39442 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
39448 The target does not need to look up any (more) symbols.
39449 @item qSymbol:@var{sym_name}
39450 The target requests the value of a new symbol @var{sym_name} (hex
39451 encoded). @value{GDBN} will continue to supply the values of symbols
39452 (if available), until the target ceases to request them.
39457 @itemx QTDisconnected
39464 @itemx qTMinFTPILen
39466 @xref{Tracepoint Packets}.
39468 @item qThreadExtraInfo,@var{thread-id}
39469 @cindex thread attributes info, remote request
39470 @cindex @samp{qThreadExtraInfo} packet
39471 Obtain a printable string description of a thread's attributes from
39472 the target OS. @var{thread-id} is a thread ID;
39473 see @ref{thread-id syntax}. This
39474 string may contain anything that the target OS thinks is interesting
39475 for @value{GDBN} to tell the user about the thread. The string is
39476 displayed in @value{GDBN}'s @code{info threads} display. Some
39477 examples of possible thread extra info strings are @samp{Runnable}, or
39478 @samp{Blocked on Mutex}.
39482 @item @var{XX}@dots{}
39483 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
39484 comprising the printable string containing the extra information about
39485 the thread's attributes.
39488 (Note that the @code{qThreadExtraInfo} packet's name is separated from
39489 the command by a @samp{,}, not a @samp{:}, contrary to the naming
39490 conventions above. Please don't use this packet as a model for new
39509 @xref{Tracepoint Packets}.
39511 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
39512 @cindex read special object, remote request
39513 @cindex @samp{qXfer} packet
39514 @anchor{qXfer read}
39515 Read uninterpreted bytes from the target's special data area
39516 identified by the keyword @var{object}. Request @var{length} bytes
39517 starting at @var{offset} bytes into the data. The content and
39518 encoding of @var{annex} is specific to @var{object}; it can supply
39519 additional details about what data to access.
39521 Here are the specific requests of this form defined so far. All
39522 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
39523 formats, listed below.
39526 @item qXfer:auxv:read::@var{offset},@var{length}
39527 @anchor{qXfer auxiliary vector read}
39528 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
39529 auxiliary vector}. Note @var{annex} must be empty.
39531 This packet is not probed by default; the remote stub must request it,
39532 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39534 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
39535 @anchor{qXfer btrace read}
39537 Return a description of the current branch trace.
39538 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
39539 packet may have one of the following values:
39543 Returns all available branch trace.
39546 Returns all available branch trace if the branch trace changed since
39547 the last read request.
39550 This packet is not probed by default; the remote stub must request it
39551 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39553 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
39554 @anchor{qXfer target description read}
39555 Access the @dfn{target description}. @xref{Target Descriptions}. The
39556 annex specifies which XML document to access. The main description is
39557 always loaded from the @samp{target.xml} annex.
39559 This packet is not probed by default; the remote stub must request it,
39560 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39562 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
39563 @anchor{qXfer library list read}
39564 Access the target's list of loaded libraries. @xref{Library List Format}.
39565 The annex part of the generic @samp{qXfer} packet must be empty
39566 (@pxref{qXfer read}).
39568 Targets which maintain a list of libraries in the program's memory do
39569 not need to implement this packet; it is designed for platforms where
39570 the operating system manages the list of loaded libraries.
39572 This packet is not probed by default; the remote stub must request it,
39573 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39575 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
39576 @anchor{qXfer svr4 library list read}
39577 Access the target's list of loaded libraries when the target is an SVR4
39578 platform. @xref{Library List Format for SVR4 Targets}. The annex part
39579 of the generic @samp{qXfer} packet must be empty unless the remote
39580 stub indicated it supports the augmented form of this packet
39581 by supplying an appropriate @samp{qSupported} response
39582 (@pxref{qXfer read}, @ref{qSupported}).
39584 This packet is optional for better performance on SVR4 targets.
39585 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
39587 This packet is not probed by default; the remote stub must request it,
39588 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39590 If the remote stub indicates it supports the augmented form of this
39591 packet then the annex part of the generic @samp{qXfer} packet may
39592 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
39593 arguments. The currently supported arguments are:
39596 @item start=@var{address}
39597 A hexadecimal number specifying the address of the @samp{struct
39598 link_map} to start reading the library list from. If unset or zero
39599 then the first @samp{struct link_map} in the library list will be
39600 chosen as the starting point.
39602 @item prev=@var{address}
39603 A hexadecimal number specifying the address of the @samp{struct
39604 link_map} immediately preceding the @samp{struct link_map}
39605 specified by the @samp{start} argument. If unset or zero then
39606 the remote stub will expect that no @samp{struct link_map}
39607 exists prior to the starting point.
39611 Arguments that are not understood by the remote stub will be silently
39614 @item qXfer:memory-map:read::@var{offset},@var{length}
39615 @anchor{qXfer memory map read}
39616 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
39617 annex part of the generic @samp{qXfer} packet must be empty
39618 (@pxref{qXfer read}).
39620 This packet is not probed by default; the remote stub must request it,
39621 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39623 @item qXfer:sdata:read::@var{offset},@var{length}
39624 @anchor{qXfer sdata read}
39626 Read contents of the extra collected static tracepoint marker
39627 information. The annex part of the generic @samp{qXfer} packet must
39628 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
39631 This packet is not probed by default; the remote stub must request it,
39632 by supplying an appropriate @samp{qSupported} response
39633 (@pxref{qSupported}).
39635 @item qXfer:siginfo:read::@var{offset},@var{length}
39636 @anchor{qXfer siginfo read}
39637 Read contents of the extra signal information on the target
39638 system. The annex part of the generic @samp{qXfer} packet must be
39639 empty (@pxref{qXfer read}).
39641 This packet is not probed by default; the remote stub must request it,
39642 by supplying an appropriate @samp{qSupported} response
39643 (@pxref{qSupported}).
39645 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
39646 @anchor{qXfer spu read}
39647 Read contents of an @code{spufs} file on the target system. The
39648 annex specifies which file to read; it must be of the form
39649 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
39650 in the target process, and @var{name} identifes the @code{spufs} file
39651 in that context to be accessed.
39653 This packet is not probed by default; the remote stub must request it,
39654 by supplying an appropriate @samp{qSupported} response
39655 (@pxref{qSupported}).
39657 @item qXfer:threads:read::@var{offset},@var{length}
39658 @anchor{qXfer threads read}
39659 Access the list of threads on target. @xref{Thread List Format}. The
39660 annex part of the generic @samp{qXfer} packet must be empty
39661 (@pxref{qXfer read}).
39663 This packet is not probed by default; the remote stub must request it,
39664 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39666 @item qXfer:traceframe-info:read::@var{offset},@var{length}
39667 @anchor{qXfer traceframe info read}
39669 Return a description of the current traceframe's contents.
39670 @xref{Traceframe Info Format}. The annex part of the generic
39671 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
39673 This packet is not probed by default; the remote stub must request it,
39674 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39676 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
39677 @anchor{qXfer unwind info block}
39679 Return the unwind information block for @var{pc}. This packet is used
39680 on OpenVMS/ia64 to ask the kernel unwind information.
39682 This packet is not probed by default.
39684 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
39685 @anchor{qXfer fdpic loadmap read}
39686 Read contents of @code{loadmap}s on the target system. The
39687 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
39688 executable @code{loadmap} or interpreter @code{loadmap} to read.
39690 This packet is not probed by default; the remote stub must request it,
39691 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39693 @item qXfer:osdata:read::@var{offset},@var{length}
39694 @anchor{qXfer osdata read}
39695 Access the target's @dfn{operating system information}.
39696 @xref{Operating System Information}.
39703 Data @var{data} (@pxref{Binary Data}) has been read from the
39704 target. There may be more data at a higher address (although
39705 it is permitted to return @samp{m} even for the last valid
39706 block of data, as long as at least one byte of data was read).
39707 @var{data} may have fewer bytes than the @var{length} in the
39711 Data @var{data} (@pxref{Binary Data}) has been read from the target.
39712 There is no more data to be read. @var{data} may have fewer bytes
39713 than the @var{length} in the request.
39716 The @var{offset} in the request is at the end of the data.
39717 There is no more data to be read.
39720 The request was malformed, or @var{annex} was invalid.
39723 The offset was invalid, or there was an error encountered reading the data.
39724 @var{nn} is a hex-encoded @code{errno} value.
39727 An empty reply indicates the @var{object} string was not recognized by
39728 the stub, or that the object does not support reading.
39731 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
39732 @cindex write data into object, remote request
39733 @anchor{qXfer write}
39734 Write uninterpreted bytes into the target's special data area
39735 identified by the keyword @var{object}, starting at @var{offset} bytes
39736 into the data. @var{data}@dots{} is the binary-encoded data
39737 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
39738 is specific to @var{object}; it can supply additional details about what data
39741 Here are the specific requests of this form defined so far. All
39742 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
39743 formats, listed below.
39746 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
39747 @anchor{qXfer siginfo write}
39748 Write @var{data} to the extra signal information on the target system.
39749 The annex part of the generic @samp{qXfer} packet must be
39750 empty (@pxref{qXfer write}).
39752 This packet is not probed by default; the remote stub must request it,
39753 by supplying an appropriate @samp{qSupported} response
39754 (@pxref{qSupported}).
39756 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
39757 @anchor{qXfer spu write}
39758 Write @var{data} to an @code{spufs} file on the target system. The
39759 annex specifies which file to write; it must be of the form
39760 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
39761 in the target process, and @var{name} identifes the @code{spufs} file
39762 in that context to be accessed.
39764 This packet is not probed by default; the remote stub must request it,
39765 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39771 @var{nn} (hex encoded) is the number of bytes written.
39772 This may be fewer bytes than supplied in the request.
39775 The request was malformed, or @var{annex} was invalid.
39778 The offset was invalid, or there was an error encountered writing the data.
39779 @var{nn} is a hex-encoded @code{errno} value.
39782 An empty reply indicates the @var{object} string was not
39783 recognized by the stub, or that the object does not support writing.
39786 @item qXfer:@var{object}:@var{operation}:@dots{}
39787 Requests of this form may be added in the future. When a stub does
39788 not recognize the @var{object} keyword, or its support for
39789 @var{object} does not recognize the @var{operation} keyword, the stub
39790 must respond with an empty packet.
39792 @item qAttached:@var{pid}
39793 @cindex query attached, remote request
39794 @cindex @samp{qAttached} packet
39795 Return an indication of whether the remote server attached to an
39796 existing process or created a new process. When the multiprocess
39797 protocol extensions are supported (@pxref{multiprocess extensions}),
39798 @var{pid} is an integer in hexadecimal format identifying the target
39799 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
39800 the query packet will be simplified as @samp{qAttached}.
39802 This query is used, for example, to know whether the remote process
39803 should be detached or killed when a @value{GDBN} session is ended with
39804 the @code{quit} command.
39809 The remote server attached to an existing process.
39811 The remote server created a new process.
39813 A badly formed request or an error was encountered.
39817 Enable branch tracing for the current thread using bts tracing.
39822 Branch tracing has been enabled.
39824 A badly formed request or an error was encountered.
39828 Disable branch tracing for the current thread.
39833 Branch tracing has been disabled.
39835 A badly formed request or an error was encountered.
39840 @node Architecture-Specific Protocol Details
39841 @section Architecture-Specific Protocol Details
39843 This section describes how the remote protocol is applied to specific
39844 target architectures. Also see @ref{Standard Target Features}, for
39845 details of XML target descriptions for each architecture.
39848 * ARM-Specific Protocol Details::
39849 * MIPS-Specific Protocol Details::
39852 @node ARM-Specific Protocol Details
39853 @subsection @acronym{ARM}-specific Protocol Details
39856 * ARM Breakpoint Kinds::
39859 @node ARM Breakpoint Kinds
39860 @subsubsection @acronym{ARM} Breakpoint Kinds
39861 @cindex breakpoint kinds, @acronym{ARM}
39863 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
39868 16-bit Thumb mode breakpoint.
39871 32-bit Thumb mode (Thumb-2) breakpoint.
39874 32-bit @acronym{ARM} mode breakpoint.
39878 @node MIPS-Specific Protocol Details
39879 @subsection @acronym{MIPS}-specific Protocol Details
39882 * MIPS Register packet Format::
39883 * MIPS Breakpoint Kinds::
39886 @node MIPS Register packet Format
39887 @subsubsection @acronym{MIPS} Register Packet Format
39888 @cindex register packet format, @acronym{MIPS}
39890 The following @code{g}/@code{G} packets have previously been defined.
39891 In the below, some thirty-two bit registers are transferred as
39892 sixty-four bits. Those registers should be zero/sign extended (which?)
39893 to fill the space allocated. Register bytes are transferred in target
39894 byte order. The two nibbles within a register byte are transferred
39895 most-significant -- least-significant.
39900 All registers are transferred as thirty-two bit quantities in the order:
39901 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
39902 registers; fsr; fir; fp.
39905 All registers are transferred as sixty-four bit quantities (including
39906 thirty-two bit registers such as @code{sr}). The ordering is the same
39911 @node MIPS Breakpoint Kinds
39912 @subsubsection @acronym{MIPS} Breakpoint Kinds
39913 @cindex breakpoint kinds, @acronym{MIPS}
39915 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
39920 16-bit @acronym{MIPS16} mode breakpoint.
39923 16-bit @acronym{microMIPS} mode breakpoint.
39926 32-bit standard @acronym{MIPS} mode breakpoint.
39929 32-bit @acronym{microMIPS} mode breakpoint.
39933 @node Tracepoint Packets
39934 @section Tracepoint Packets
39935 @cindex tracepoint packets
39936 @cindex packets, tracepoint
39938 Here we describe the packets @value{GDBN} uses to implement
39939 tracepoints (@pxref{Tracepoints}).
39943 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
39944 @cindex @samp{QTDP} packet
39945 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
39946 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
39947 the tracepoint is disabled. @var{step} is the tracepoint's step
39948 count, and @var{pass} is its pass count. If an @samp{F} is present,
39949 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
39950 the number of bytes that the target should copy elsewhere to make room
39951 for the tracepoint. If an @samp{X} is present, it introduces a
39952 tracepoint condition, which consists of a hexadecimal length, followed
39953 by a comma and hex-encoded bytes, in a manner similar to action
39954 encodings as described below. If the trailing @samp{-} is present,
39955 further @samp{QTDP} packets will follow to specify this tracepoint's
39961 The packet was understood and carried out.
39963 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
39965 The packet was not recognized.
39968 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
39969 Define actions to be taken when a tracepoint is hit. @var{n} and
39970 @var{addr} must be the same as in the initial @samp{QTDP} packet for
39971 this tracepoint. This packet may only be sent immediately after
39972 another @samp{QTDP} packet that ended with a @samp{-}. If the
39973 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
39974 specifying more actions for this tracepoint.
39976 In the series of action packets for a given tracepoint, at most one
39977 can have an @samp{S} before its first @var{action}. If such a packet
39978 is sent, it and the following packets define ``while-stepping''
39979 actions. Any prior packets define ordinary actions --- that is, those
39980 taken when the tracepoint is first hit. If no action packet has an
39981 @samp{S}, then all the packets in the series specify ordinary
39982 tracepoint actions.
39984 The @samp{@var{action}@dots{}} portion of the packet is a series of
39985 actions, concatenated without separators. Each action has one of the
39991 Collect the registers whose bits are set in @var{mask}. @var{mask} is
39992 a hexadecimal number whose @var{i}'th bit is set if register number
39993 @var{i} should be collected. (The least significant bit is numbered
39994 zero.) Note that @var{mask} may be any number of digits long; it may
39995 not fit in a 32-bit word.
39997 @item M @var{basereg},@var{offset},@var{len}
39998 Collect @var{len} bytes of memory starting at the address in register
39999 number @var{basereg}, plus @var{offset}. If @var{basereg} is
40000 @samp{-1}, then the range has a fixed address: @var{offset} is the
40001 address of the lowest byte to collect. The @var{basereg},
40002 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
40003 values (the @samp{-1} value for @var{basereg} is a special case).
40005 @item X @var{len},@var{expr}
40006 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
40007 it directs. @var{expr} is an agent expression, as described in
40008 @ref{Agent Expressions}. Each byte of the expression is encoded as a
40009 two-digit hex number in the packet; @var{len} is the number of bytes
40010 in the expression (and thus one-half the number of hex digits in the
40015 Any number of actions may be packed together in a single @samp{QTDP}
40016 packet, as long as the packet does not exceed the maximum packet
40017 length (400 bytes, for many stubs). There may be only one @samp{R}
40018 action per tracepoint, and it must precede any @samp{M} or @samp{X}
40019 actions. Any registers referred to by @samp{M} and @samp{X} actions
40020 must be collected by a preceding @samp{R} action. (The
40021 ``while-stepping'' actions are treated as if they were attached to a
40022 separate tracepoint, as far as these restrictions are concerned.)
40027 The packet was understood and carried out.
40029 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
40031 The packet was not recognized.
40034 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
40035 @cindex @samp{QTDPsrc} packet
40036 Specify a source string of tracepoint @var{n} at address @var{addr}.
40037 This is useful to get accurate reproduction of the tracepoints
40038 originally downloaded at the beginning of the trace run. @var{type}
40039 is the name of the tracepoint part, such as @samp{cond} for the
40040 tracepoint's conditional expression (see below for a list of types), while
40041 @var{bytes} is the string, encoded in hexadecimal.
40043 @var{start} is the offset of the @var{bytes} within the overall source
40044 string, while @var{slen} is the total length of the source string.
40045 This is intended for handling source strings that are longer than will
40046 fit in a single packet.
40047 @c Add detailed example when this info is moved into a dedicated
40048 @c tracepoint descriptions section.
40050 The available string types are @samp{at} for the location,
40051 @samp{cond} for the conditional, and @samp{cmd} for an action command.
40052 @value{GDBN} sends a separate packet for each command in the action
40053 list, in the same order in which the commands are stored in the list.
40055 The target does not need to do anything with source strings except
40056 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
40059 Although this packet is optional, and @value{GDBN} will only send it
40060 if the target replies with @samp{TracepointSource} @xref{General
40061 Query Packets}, it makes both disconnected tracing and trace files
40062 much easier to use. Otherwise the user must be careful that the
40063 tracepoints in effect while looking at trace frames are identical to
40064 the ones in effect during the trace run; even a small discrepancy
40065 could cause @samp{tdump} not to work, or a particular trace frame not
40068 @item QTDV:@var{n}:@var{value}
40069 @cindex define trace state variable, remote request
40070 @cindex @samp{QTDV} packet
40071 Create a new trace state variable, number @var{n}, with an initial
40072 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
40073 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
40074 the option of not using this packet for initial values of zero; the
40075 target should simply create the trace state variables as they are
40076 mentioned in expressions.
40078 @item QTFrame:@var{n}
40079 @cindex @samp{QTFrame} packet
40080 Select the @var{n}'th tracepoint frame from the buffer, and use the
40081 register and memory contents recorded there to answer subsequent
40082 request packets from @value{GDBN}.
40084 A successful reply from the stub indicates that the stub has found the
40085 requested frame. The response is a series of parts, concatenated
40086 without separators, describing the frame we selected. Each part has
40087 one of the following forms:
40091 The selected frame is number @var{n} in the trace frame buffer;
40092 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
40093 was no frame matching the criteria in the request packet.
40096 The selected trace frame records a hit of tracepoint number @var{t};
40097 @var{t} is a hexadecimal number.
40101 @item QTFrame:pc:@var{addr}
40102 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40103 currently selected frame whose PC is @var{addr};
40104 @var{addr} is a hexadecimal number.
40106 @item QTFrame:tdp:@var{t}
40107 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40108 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
40109 is a hexadecimal number.
40111 @item QTFrame:range:@var{start}:@var{end}
40112 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40113 currently selected frame whose PC is between @var{start} (inclusive)
40114 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
40117 @item QTFrame:outside:@var{start}:@var{end}
40118 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
40119 frame @emph{outside} the given range of addresses (exclusive).
40122 @cindex @samp{qTMinFTPILen} packet
40123 This packet requests the minimum length of instruction at which a fast
40124 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
40125 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
40126 it depends on the target system being able to create trampolines in
40127 the first 64K of memory, which might or might not be possible for that
40128 system. So the reply to this packet will be 4 if it is able to
40135 The minimum instruction length is currently unknown.
40137 The minimum instruction length is @var{length}, where @var{length} is greater
40138 or equal to 1. @var{length} is a hexadecimal number. A reply of 1 means
40139 that a fast tracepoint may be placed on any instruction regardless of size.
40141 An error has occurred.
40143 An empty reply indicates that the request is not supported by the stub.
40147 @cindex @samp{QTStart} packet
40148 Begin the tracepoint experiment. Begin collecting data from
40149 tracepoint hits in the trace frame buffer. This packet supports the
40150 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
40151 instruction reply packet}).
40154 @cindex @samp{QTStop} packet
40155 End the tracepoint experiment. Stop collecting trace frames.
40157 @item QTEnable:@var{n}:@var{addr}
40159 @cindex @samp{QTEnable} packet
40160 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
40161 experiment. If the tracepoint was previously disabled, then collection
40162 of data from it will resume.
40164 @item QTDisable:@var{n}:@var{addr}
40166 @cindex @samp{QTDisable} packet
40167 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
40168 experiment. No more data will be collected from the tracepoint unless
40169 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
40172 @cindex @samp{QTinit} packet
40173 Clear the table of tracepoints, and empty the trace frame buffer.
40175 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
40176 @cindex @samp{QTro} packet
40177 Establish the given ranges of memory as ``transparent''. The stub
40178 will answer requests for these ranges from memory's current contents,
40179 if they were not collected as part of the tracepoint hit.
40181 @value{GDBN} uses this to mark read-only regions of memory, like those
40182 containing program code. Since these areas never change, they should
40183 still have the same contents they did when the tracepoint was hit, so
40184 there's no reason for the stub to refuse to provide their contents.
40186 @item QTDisconnected:@var{value}
40187 @cindex @samp{QTDisconnected} packet
40188 Set the choice to what to do with the tracing run when @value{GDBN}
40189 disconnects from the target. A @var{value} of 1 directs the target to
40190 continue the tracing run, while 0 tells the target to stop tracing if
40191 @value{GDBN} is no longer in the picture.
40194 @cindex @samp{qTStatus} packet
40195 Ask the stub if there is a trace experiment running right now.
40197 The reply has the form:
40201 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
40202 @var{running} is a single digit @code{1} if the trace is presently
40203 running, or @code{0} if not. It is followed by semicolon-separated
40204 optional fields that an agent may use to report additional status.
40208 If the trace is not running, the agent may report any of several
40209 explanations as one of the optional fields:
40214 No trace has been run yet.
40216 @item tstop[:@var{text}]:0
40217 The trace was stopped by a user-originated stop command. The optional
40218 @var{text} field is a user-supplied string supplied as part of the
40219 stop command (for instance, an explanation of why the trace was
40220 stopped manually). It is hex-encoded.
40223 The trace stopped because the trace buffer filled up.
40225 @item tdisconnected:0
40226 The trace stopped because @value{GDBN} disconnected from the target.
40228 @item tpasscount:@var{tpnum}
40229 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
40231 @item terror:@var{text}:@var{tpnum}
40232 The trace stopped because tracepoint @var{tpnum} had an error. The
40233 string @var{text} is available to describe the nature of the error
40234 (for instance, a divide by zero in the condition expression).
40235 @var{text} is hex encoded.
40238 The trace stopped for some other reason.
40242 Additional optional fields supply statistical and other information.
40243 Although not required, they are extremely useful for users monitoring
40244 the progress of a trace run. If a trace has stopped, and these
40245 numbers are reported, they must reflect the state of the just-stopped
40250 @item tframes:@var{n}
40251 The number of trace frames in the buffer.
40253 @item tcreated:@var{n}
40254 The total number of trace frames created during the run. This may
40255 be larger than the trace frame count, if the buffer is circular.
40257 @item tsize:@var{n}
40258 The total size of the trace buffer, in bytes.
40260 @item tfree:@var{n}
40261 The number of bytes still unused in the buffer.
40263 @item circular:@var{n}
40264 The value of the circular trace buffer flag. @code{1} means that the
40265 trace buffer is circular and old trace frames will be discarded if
40266 necessary to make room, @code{0} means that the trace buffer is linear
40269 @item disconn:@var{n}
40270 The value of the disconnected tracing flag. @code{1} means that
40271 tracing will continue after @value{GDBN} disconnects, @code{0} means
40272 that the trace run will stop.
40276 @item qTP:@var{tp}:@var{addr}
40277 @cindex tracepoint status, remote request
40278 @cindex @samp{qTP} packet
40279 Ask the stub for the current state of tracepoint number @var{tp} at
40280 address @var{addr}.
40284 @item V@var{hits}:@var{usage}
40285 The tracepoint has been hit @var{hits} times so far during the trace
40286 run, and accounts for @var{usage} in the trace buffer. Note that
40287 @code{while-stepping} steps are not counted as separate hits, but the
40288 steps' space consumption is added into the usage number.
40292 @item qTV:@var{var}
40293 @cindex trace state variable value, remote request
40294 @cindex @samp{qTV} packet
40295 Ask the stub for the value of the trace state variable number @var{var}.
40300 The value of the variable is @var{value}. This will be the current
40301 value of the variable if the user is examining a running target, or a
40302 saved value if the variable was collected in the trace frame that the
40303 user is looking at. Note that multiple requests may result in
40304 different reply values, such as when requesting values while the
40305 program is running.
40308 The value of the variable is unknown. This would occur, for example,
40309 if the user is examining a trace frame in which the requested variable
40314 @cindex @samp{qTfP} packet
40316 @cindex @samp{qTsP} packet
40317 These packets request data about tracepoints that are being used by
40318 the target. @value{GDBN} sends @code{qTfP} to get the first piece
40319 of data, and multiple @code{qTsP} to get additional pieces. Replies
40320 to these packets generally take the form of the @code{QTDP} packets
40321 that define tracepoints. (FIXME add detailed syntax)
40324 @cindex @samp{qTfV} packet
40326 @cindex @samp{qTsV} packet
40327 These packets request data about trace state variables that are on the
40328 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
40329 and multiple @code{qTsV} to get additional variables. Replies to
40330 these packets follow the syntax of the @code{QTDV} packets that define
40331 trace state variables.
40337 @cindex @samp{qTfSTM} packet
40338 @cindex @samp{qTsSTM} packet
40339 These packets request data about static tracepoint markers that exist
40340 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
40341 first piece of data, and multiple @code{qTsSTM} to get additional
40342 pieces. Replies to these packets take the following form:
40346 @item m @var{address}:@var{id}:@var{extra}
40348 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
40349 a comma-separated list of markers
40351 (lower case letter @samp{L}) denotes end of list.
40353 An error occurred. @var{nn} are hex digits.
40355 An empty reply indicates that the request is not supported by the
40359 @var{address} is encoded in hex.
40360 @var{id} and @var{extra} are strings encoded in hex.
40362 In response to each query, the target will reply with a list of one or
40363 more markers, separated by commas. @value{GDBN} will respond to each
40364 reply with a request for more markers (using the @samp{qs} form of the
40365 query), until the target responds with @samp{l} (lower-case ell, for
40368 @item qTSTMat:@var{address}
40370 @cindex @samp{qTSTMat} packet
40371 This packets requests data about static tracepoint markers in the
40372 target program at @var{address}. Replies to this packet follow the
40373 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
40374 tracepoint markers.
40376 @item QTSave:@var{filename}
40377 @cindex @samp{QTSave} packet
40378 This packet directs the target to save trace data to the file name
40379 @var{filename} in the target's filesystem. @var{filename} is encoded
40380 as a hex string; the interpretation of the file name (relative vs
40381 absolute, wild cards, etc) is up to the target.
40383 @item qTBuffer:@var{offset},@var{len}
40384 @cindex @samp{qTBuffer} packet
40385 Return up to @var{len} bytes of the current contents of trace buffer,
40386 starting at @var{offset}. The trace buffer is treated as if it were
40387 a contiguous collection of traceframes, as per the trace file format.
40388 The reply consists as many hex-encoded bytes as the target can deliver
40389 in a packet; it is not an error to return fewer than were asked for.
40390 A reply consisting of just @code{l} indicates that no bytes are
40393 @item QTBuffer:circular:@var{value}
40394 This packet directs the target to use a circular trace buffer if
40395 @var{value} is 1, or a linear buffer if the value is 0.
40397 @item QTBuffer:size:@var{size}
40398 @anchor{QTBuffer-size}
40399 @cindex @samp{QTBuffer size} packet
40400 This packet directs the target to make the trace buffer be of size
40401 @var{size} if possible. A value of @code{-1} tells the target to
40402 use whatever size it prefers.
40404 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
40405 @cindex @samp{QTNotes} packet
40406 This packet adds optional textual notes to the trace run. Allowable
40407 types include @code{user}, @code{notes}, and @code{tstop}, the
40408 @var{text} fields are arbitrary strings, hex-encoded.
40412 @subsection Relocate instruction reply packet
40413 When installing fast tracepoints in memory, the target may need to
40414 relocate the instruction currently at the tracepoint address to a
40415 different address in memory. For most instructions, a simple copy is
40416 enough, but, for example, call instructions that implicitly push the
40417 return address on the stack, and relative branches or other
40418 PC-relative instructions require offset adjustment, so that the effect
40419 of executing the instruction at a different address is the same as if
40420 it had executed in the original location.
40422 In response to several of the tracepoint packets, the target may also
40423 respond with a number of intermediate @samp{qRelocInsn} request
40424 packets before the final result packet, to have @value{GDBN} handle
40425 this relocation operation. If a packet supports this mechanism, its
40426 documentation will explicitly say so. See for example the above
40427 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
40428 format of the request is:
40431 @item qRelocInsn:@var{from};@var{to}
40433 This requests @value{GDBN} to copy instruction at address @var{from}
40434 to address @var{to}, possibly adjusted so that executing the
40435 instruction at @var{to} has the same effect as executing it at
40436 @var{from}. @value{GDBN} writes the adjusted instruction to target
40437 memory starting at @var{to}.
40442 @item qRelocInsn:@var{adjusted_size}
40443 Informs the stub the relocation is complete. @var{adjusted_size} is
40444 the length in bytes of resulting relocated instruction sequence.
40446 A badly formed request was detected, or an error was encountered while
40447 relocating the instruction.
40450 @node Host I/O Packets
40451 @section Host I/O Packets
40452 @cindex Host I/O, remote protocol
40453 @cindex file transfer, remote protocol
40455 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
40456 operations on the far side of a remote link. For example, Host I/O is
40457 used to upload and download files to a remote target with its own
40458 filesystem. Host I/O uses the same constant values and data structure
40459 layout as the target-initiated File-I/O protocol. However, the
40460 Host I/O packets are structured differently. The target-initiated
40461 protocol relies on target memory to store parameters and buffers.
40462 Host I/O requests are initiated by @value{GDBN}, and the
40463 target's memory is not involved. @xref{File-I/O Remote Protocol
40464 Extension}, for more details on the target-initiated protocol.
40466 The Host I/O request packets all encode a single operation along with
40467 its arguments. They have this format:
40471 @item vFile:@var{operation}: @var{parameter}@dots{}
40472 @var{operation} is the name of the particular request; the target
40473 should compare the entire packet name up to the second colon when checking
40474 for a supported operation. The format of @var{parameter} depends on
40475 the operation. Numbers are always passed in hexadecimal. Negative
40476 numbers have an explicit minus sign (i.e.@: two's complement is not
40477 used). Strings (e.g.@: filenames) are encoded as a series of
40478 hexadecimal bytes. The last argument to a system call may be a
40479 buffer of escaped binary data (@pxref{Binary Data}).
40483 The valid responses to Host I/O packets are:
40487 @item F @var{result} [, @var{errno}] [; @var{attachment}]
40488 @var{result} is the integer value returned by this operation, usually
40489 non-negative for success and -1 for errors. If an error has occured,
40490 @var{errno} will be included in the result. @var{errno} will have a
40491 value defined by the File-I/O protocol (@pxref{Errno Values}). For
40492 operations which return data, @var{attachment} supplies the data as a
40493 binary buffer. Binary buffers in response packets are escaped in the
40494 normal way (@pxref{Binary Data}). See the individual packet
40495 documentation for the interpretation of @var{result} and
40499 An empty response indicates that this operation is not recognized.
40503 These are the supported Host I/O operations:
40506 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
40507 Open a file at @var{pathname} and return a file descriptor for it, or
40508 return -1 if an error occurs. @var{pathname} is a string,
40509 @var{flags} is an integer indicating a mask of open flags
40510 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
40511 of mode bits to use if the file is created (@pxref{mode_t Values}).
40512 @xref{open}, for details of the open flags and mode values.
40514 @item vFile:close: @var{fd}
40515 Close the open file corresponding to @var{fd} and return 0, or
40516 -1 if an error occurs.
40518 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
40519 Read data from the open file corresponding to @var{fd}. Up to
40520 @var{count} bytes will be read from the file, starting at @var{offset}
40521 relative to the start of the file. The target may read fewer bytes;
40522 common reasons include packet size limits and an end-of-file
40523 condition. The number of bytes read is returned. Zero should only be
40524 returned for a successful read at the end of the file, or if
40525 @var{count} was zero.
40527 The data read should be returned as a binary attachment on success.
40528 If zero bytes were read, the response should include an empty binary
40529 attachment (i.e.@: a trailing semicolon). The return value is the
40530 number of target bytes read; the binary attachment may be longer if
40531 some characters were escaped.
40533 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
40534 Write @var{data} (a binary buffer) to the open file corresponding
40535 to @var{fd}. Start the write at @var{offset} from the start of the
40536 file. Unlike many @code{write} system calls, there is no
40537 separate @var{count} argument; the length of @var{data} in the
40538 packet is used. @samp{vFile:write} returns the number of bytes written,
40539 which may be shorter than the length of @var{data}, or -1 if an
40542 @item vFile:unlink: @var{pathname}
40543 Delete the file at @var{pathname} on the target. Return 0,
40544 or -1 if an error occurs. @var{pathname} is a string.
40546 @item vFile:readlink: @var{filename}
40547 Read value of symbolic link @var{filename} on the target. Return
40548 the number of bytes read, or -1 if an error occurs.
40550 The data read should be returned as a binary attachment on success.
40551 If zero bytes were read, the response should include an empty binary
40552 attachment (i.e.@: a trailing semicolon). The return value is the
40553 number of target bytes read; the binary attachment may be longer if
40554 some characters were escaped.
40559 @section Interrupts
40560 @cindex interrupts (remote protocol)
40562 When a program on the remote target is running, @value{GDBN} may
40563 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
40564 a @code{BREAK} followed by @code{g},
40565 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
40567 The precise meaning of @code{BREAK} is defined by the transport
40568 mechanism and may, in fact, be undefined. @value{GDBN} does not
40569 currently define a @code{BREAK} mechanism for any of the network
40570 interfaces except for TCP, in which case @value{GDBN} sends the
40571 @code{telnet} BREAK sequence.
40573 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
40574 transport mechanisms. It is represented by sending the single byte
40575 @code{0x03} without any of the usual packet overhead described in
40576 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
40577 transmitted as part of a packet, it is considered to be packet data
40578 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
40579 (@pxref{X packet}), used for binary downloads, may include an unescaped
40580 @code{0x03} as part of its packet.
40582 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
40583 When Linux kernel receives this sequence from serial port,
40584 it stops execution and connects to gdb.
40586 Stubs are not required to recognize these interrupt mechanisms and the
40587 precise meaning associated with receipt of the interrupt is
40588 implementation defined. If the target supports debugging of multiple
40589 threads and/or processes, it should attempt to interrupt all
40590 currently-executing threads and processes.
40591 If the stub is successful at interrupting the
40592 running program, it should send one of the stop
40593 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
40594 of successfully stopping the program in all-stop mode, and a stop reply
40595 for each stopped thread in non-stop mode.
40596 Interrupts received while the
40597 program is stopped are discarded.
40599 @node Notification Packets
40600 @section Notification Packets
40601 @cindex notification packets
40602 @cindex packets, notification
40604 The @value{GDBN} remote serial protocol includes @dfn{notifications},
40605 packets that require no acknowledgment. Both the GDB and the stub
40606 may send notifications (although the only notifications defined at
40607 present are sent by the stub). Notifications carry information
40608 without incurring the round-trip latency of an acknowledgment, and so
40609 are useful for low-impact communications where occasional packet loss
40612 A notification packet has the form @samp{% @var{data} #
40613 @var{checksum}}, where @var{data} is the content of the notification,
40614 and @var{checksum} is a checksum of @var{data}, computed and formatted
40615 as for ordinary @value{GDBN} packets. A notification's @var{data}
40616 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
40617 receiving a notification, the recipient sends no @samp{+} or @samp{-}
40618 to acknowledge the notification's receipt or to report its corruption.
40620 Every notification's @var{data} begins with a name, which contains no
40621 colon characters, followed by a colon character.
40623 Recipients should silently ignore corrupted notifications and
40624 notifications they do not understand. Recipients should restart
40625 timeout periods on receipt of a well-formed notification, whether or
40626 not they understand it.
40628 Senders should only send the notifications described here when this
40629 protocol description specifies that they are permitted. In the
40630 future, we may extend the protocol to permit existing notifications in
40631 new contexts; this rule helps older senders avoid confusing newer
40634 (Older versions of @value{GDBN} ignore bytes received until they see
40635 the @samp{$} byte that begins an ordinary packet, so new stubs may
40636 transmit notifications without fear of confusing older clients. There
40637 are no notifications defined for @value{GDBN} to send at the moment, but we
40638 assume that most older stubs would ignore them, as well.)
40640 Each notification is comprised of three parts:
40642 @item @var{name}:@var{event}
40643 The notification packet is sent by the side that initiates the
40644 exchange (currently, only the stub does that), with @var{event}
40645 carrying the specific information about the notification.
40646 @var{name} is the name of the notification.
40648 The acknowledge sent by the other side, usually @value{GDBN}, to
40649 acknowledge the exchange and request the event.
40652 The purpose of an asynchronous notification mechanism is to report to
40653 @value{GDBN} that something interesting happened in the remote stub.
40655 The remote stub may send notification @var{name}:@var{event}
40656 at any time, but @value{GDBN} acknowledges the notification when
40657 appropriate. The notification event is pending before @value{GDBN}
40658 acknowledges. Only one notification at a time may be pending; if
40659 additional events occur before @value{GDBN} has acknowledged the
40660 previous notification, they must be queued by the stub for later
40661 synchronous transmission in response to @var{ack} packets from
40662 @value{GDBN}. Because the notification mechanism is unreliable,
40663 the stub is permitted to resend a notification if it believes
40664 @value{GDBN} may not have received it.
40666 Specifically, notifications may appear when @value{GDBN} is not
40667 otherwise reading input from the stub, or when @value{GDBN} is
40668 expecting to read a normal synchronous response or a
40669 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
40670 Notification packets are distinct from any other communication from
40671 the stub so there is no ambiguity.
40673 After receiving a notification, @value{GDBN} shall acknowledge it by
40674 sending a @var{ack} packet as a regular, synchronous request to the
40675 stub. Such acknowledgment is not required to happen immediately, as
40676 @value{GDBN} is permitted to send other, unrelated packets to the
40677 stub first, which the stub should process normally.
40679 Upon receiving a @var{ack} packet, if the stub has other queued
40680 events to report to @value{GDBN}, it shall respond by sending a
40681 normal @var{event}. @value{GDBN} shall then send another @var{ack}
40682 packet to solicit further responses; again, it is permitted to send
40683 other, unrelated packets as well which the stub should process
40686 If the stub receives a @var{ack} packet and there are no additional
40687 @var{event} to report, the stub shall return an @samp{OK} response.
40688 At this point, @value{GDBN} has finished processing a notification
40689 and the stub has completed sending any queued events. @value{GDBN}
40690 won't accept any new notifications until the final @samp{OK} is
40691 received . If further notification events occur, the stub shall send
40692 a new notification, @value{GDBN} shall accept the notification, and
40693 the process shall be repeated.
40695 The process of asynchronous notification can be illustrated by the
40698 <- @code{%%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
40701 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
40703 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
40708 The following notifications are defined:
40709 @multitable @columnfractions 0.12 0.12 0.38 0.38
40718 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
40719 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
40720 for information on how these notifications are acknowledged by
40722 @tab Report an asynchronous stop event in non-stop mode.
40726 @node Remote Non-Stop
40727 @section Remote Protocol Support for Non-Stop Mode
40729 @value{GDBN}'s remote protocol supports non-stop debugging of
40730 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
40731 supports non-stop mode, it should report that to @value{GDBN} by including
40732 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
40734 @value{GDBN} typically sends a @samp{QNonStop} packet only when
40735 establishing a new connection with the stub. Entering non-stop mode
40736 does not alter the state of any currently-running threads, but targets
40737 must stop all threads in any already-attached processes when entering
40738 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
40739 probe the target state after a mode change.
40741 In non-stop mode, when an attached process encounters an event that
40742 would otherwise be reported with a stop reply, it uses the
40743 asynchronous notification mechanism (@pxref{Notification Packets}) to
40744 inform @value{GDBN}. In contrast to all-stop mode, where all threads
40745 in all processes are stopped when a stop reply is sent, in non-stop
40746 mode only the thread reporting the stop event is stopped. That is,
40747 when reporting a @samp{S} or @samp{T} response to indicate completion
40748 of a step operation, hitting a breakpoint, or a fault, only the
40749 affected thread is stopped; any other still-running threads continue
40750 to run. When reporting a @samp{W} or @samp{X} response, all running
40751 threads belonging to other attached processes continue to run.
40753 In non-stop mode, the target shall respond to the @samp{?} packet as
40754 follows. First, any incomplete stop reply notification/@samp{vStopped}
40755 sequence in progress is abandoned. The target must begin a new
40756 sequence reporting stop events for all stopped threads, whether or not
40757 it has previously reported those events to @value{GDBN}. The first
40758 stop reply is sent as a synchronous reply to the @samp{?} packet, and
40759 subsequent stop replies are sent as responses to @samp{vStopped} packets
40760 using the mechanism described above. The target must not send
40761 asynchronous stop reply notifications until the sequence is complete.
40762 If all threads are running when the target receives the @samp{?} packet,
40763 or if the target is not attached to any process, it shall respond
40766 @node Packet Acknowledgment
40767 @section Packet Acknowledgment
40769 @cindex acknowledgment, for @value{GDBN} remote
40770 @cindex packet acknowledgment, for @value{GDBN} remote
40771 By default, when either the host or the target machine receives a packet,
40772 the first response expected is an acknowledgment: either @samp{+} (to indicate
40773 the package was received correctly) or @samp{-} (to request retransmission).
40774 This mechanism allows the @value{GDBN} remote protocol to operate over
40775 unreliable transport mechanisms, such as a serial line.
40777 In cases where the transport mechanism is itself reliable (such as a pipe or
40778 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
40779 It may be desirable to disable them in that case to reduce communication
40780 overhead, or for other reasons. This can be accomplished by means of the
40781 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
40783 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
40784 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
40785 and response format still includes the normal checksum, as described in
40786 @ref{Overview}, but the checksum may be ignored by the receiver.
40788 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
40789 no-acknowledgment mode, it should report that to @value{GDBN}
40790 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
40791 @pxref{qSupported}.
40792 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
40793 disabled via the @code{set remote noack-packet off} command
40794 (@pxref{Remote Configuration}),
40795 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
40796 Only then may the stub actually turn off packet acknowledgments.
40797 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
40798 response, which can be safely ignored by the stub.
40800 Note that @code{set remote noack-packet} command only affects negotiation
40801 between @value{GDBN} and the stub when subsequent connections are made;
40802 it does not affect the protocol acknowledgment state for any current
40804 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
40805 new connection is established,
40806 there is also no protocol request to re-enable the acknowledgments
40807 for the current connection, once disabled.
40812 Example sequence of a target being re-started. Notice how the restart
40813 does not get any direct output:
40818 @emph{target restarts}
40821 <- @code{T001:1234123412341234}
40825 Example sequence of a target being stepped by a single instruction:
40828 -> @code{G1445@dots{}}
40833 <- @code{T001:1234123412341234}
40837 <- @code{1455@dots{}}
40841 @node File-I/O Remote Protocol Extension
40842 @section File-I/O Remote Protocol Extension
40843 @cindex File-I/O remote protocol extension
40846 * File-I/O Overview::
40847 * Protocol Basics::
40848 * The F Request Packet::
40849 * The F Reply Packet::
40850 * The Ctrl-C Message::
40852 * List of Supported Calls::
40853 * Protocol-specific Representation of Datatypes::
40855 * File-I/O Examples::
40858 @node File-I/O Overview
40859 @subsection File-I/O Overview
40860 @cindex file-i/o overview
40862 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
40863 target to use the host's file system and console I/O to perform various
40864 system calls. System calls on the target system are translated into a
40865 remote protocol packet to the host system, which then performs the needed
40866 actions and returns a response packet to the target system.
40867 This simulates file system operations even on targets that lack file systems.
40869 The protocol is defined to be independent of both the host and target systems.
40870 It uses its own internal representation of datatypes and values. Both
40871 @value{GDBN} and the target's @value{GDBN} stub are responsible for
40872 translating the system-dependent value representations into the internal
40873 protocol representations when data is transmitted.
40875 The communication is synchronous. A system call is possible only when
40876 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
40877 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
40878 the target is stopped to allow deterministic access to the target's
40879 memory. Therefore File-I/O is not interruptible by target signals. On
40880 the other hand, it is possible to interrupt File-I/O by a user interrupt
40881 (@samp{Ctrl-C}) within @value{GDBN}.
40883 The target's request to perform a host system call does not finish
40884 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
40885 after finishing the system call, the target returns to continuing the
40886 previous activity (continue, step). No additional continue or step
40887 request from @value{GDBN} is required.
40890 (@value{GDBP}) continue
40891 <- target requests 'system call X'
40892 target is stopped, @value{GDBN} executes system call
40893 -> @value{GDBN} returns result
40894 ... target continues, @value{GDBN} returns to wait for the target
40895 <- target hits breakpoint and sends a Txx packet
40898 The protocol only supports I/O on the console and to regular files on
40899 the host file system. Character or block special devices, pipes,
40900 named pipes, sockets or any other communication method on the host
40901 system are not supported by this protocol.
40903 File I/O is not supported in non-stop mode.
40905 @node Protocol Basics
40906 @subsection Protocol Basics
40907 @cindex protocol basics, file-i/o
40909 The File-I/O protocol uses the @code{F} packet as the request as well
40910 as reply packet. Since a File-I/O system call can only occur when
40911 @value{GDBN} is waiting for a response from the continuing or stepping target,
40912 the File-I/O request is a reply that @value{GDBN} has to expect as a result
40913 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
40914 This @code{F} packet contains all information needed to allow @value{GDBN}
40915 to call the appropriate host system call:
40919 A unique identifier for the requested system call.
40922 All parameters to the system call. Pointers are given as addresses
40923 in the target memory address space. Pointers to strings are given as
40924 pointer/length pair. Numerical values are given as they are.
40925 Numerical control flags are given in a protocol-specific representation.
40929 At this point, @value{GDBN} has to perform the following actions.
40933 If the parameters include pointer values to data needed as input to a
40934 system call, @value{GDBN} requests this data from the target with a
40935 standard @code{m} packet request. This additional communication has to be
40936 expected by the target implementation and is handled as any other @code{m}
40940 @value{GDBN} translates all value from protocol representation to host
40941 representation as needed. Datatypes are coerced into the host types.
40944 @value{GDBN} calls the system call.
40947 It then coerces datatypes back to protocol representation.
40950 If the system call is expected to return data in buffer space specified
40951 by pointer parameters to the call, the data is transmitted to the
40952 target using a @code{M} or @code{X} packet. This packet has to be expected
40953 by the target implementation and is handled as any other @code{M} or @code{X}
40958 Eventually @value{GDBN} replies with another @code{F} packet which contains all
40959 necessary information for the target to continue. This at least contains
40966 @code{errno}, if has been changed by the system call.
40973 After having done the needed type and value coercion, the target continues
40974 the latest continue or step action.
40976 @node The F Request Packet
40977 @subsection The @code{F} Request Packet
40978 @cindex file-i/o request packet
40979 @cindex @code{F} request packet
40981 The @code{F} request packet has the following format:
40984 @item F@var{call-id},@var{parameter@dots{}}
40986 @var{call-id} is the identifier to indicate the host system call to be called.
40987 This is just the name of the function.
40989 @var{parameter@dots{}} are the parameters to the system call.
40990 Parameters are hexadecimal integer values, either the actual values in case
40991 of scalar datatypes, pointers to target buffer space in case of compound
40992 datatypes and unspecified memory areas, or pointer/length pairs in case
40993 of string parameters. These are appended to the @var{call-id} as a
40994 comma-delimited list. All values are transmitted in ASCII
40995 string representation, pointer/length pairs separated by a slash.
41001 @node The F Reply Packet
41002 @subsection The @code{F} Reply Packet
41003 @cindex file-i/o reply packet
41004 @cindex @code{F} reply packet
41006 The @code{F} reply packet has the following format:
41010 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
41012 @var{retcode} is the return code of the system call as hexadecimal value.
41014 @var{errno} is the @code{errno} set by the call, in protocol-specific
41016 This parameter can be omitted if the call was successful.
41018 @var{Ctrl-C flag} is only sent if the user requested a break. In this
41019 case, @var{errno} must be sent as well, even if the call was successful.
41020 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
41027 or, if the call was interrupted before the host call has been performed:
41034 assuming 4 is the protocol-specific representation of @code{EINTR}.
41039 @node The Ctrl-C Message
41040 @subsection The @samp{Ctrl-C} Message
41041 @cindex ctrl-c message, in file-i/o protocol
41043 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
41044 reply packet (@pxref{The F Reply Packet}),
41045 the target should behave as if it had
41046 gotten a break message. The meaning for the target is ``system call
41047 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
41048 (as with a break message) and return to @value{GDBN} with a @code{T02}
41051 It's important for the target to know in which
41052 state the system call was interrupted. There are two possible cases:
41056 The system call hasn't been performed on the host yet.
41059 The system call on the host has been finished.
41063 These two states can be distinguished by the target by the value of the
41064 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
41065 call hasn't been performed. This is equivalent to the @code{EINTR} handling
41066 on POSIX systems. In any other case, the target may presume that the
41067 system call has been finished --- successfully or not --- and should behave
41068 as if the break message arrived right after the system call.
41070 @value{GDBN} must behave reliably. If the system call has not been called
41071 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
41072 @code{errno} in the packet. If the system call on the host has been finished
41073 before the user requests a break, the full action must be finished by
41074 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
41075 The @code{F} packet may only be sent when either nothing has happened
41076 or the full action has been completed.
41079 @subsection Console I/O
41080 @cindex console i/o as part of file-i/o
41082 By default and if not explicitly closed by the target system, the file
41083 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
41084 on the @value{GDBN} console is handled as any other file output operation
41085 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
41086 by @value{GDBN} so that after the target read request from file descriptor
41087 0 all following typing is buffered until either one of the following
41092 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
41094 system call is treated as finished.
41097 The user presses @key{RET}. This is treated as end of input with a trailing
41101 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
41102 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
41106 If the user has typed more characters than fit in the buffer given to
41107 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
41108 either another @code{read(0, @dots{})} is requested by the target, or debugging
41109 is stopped at the user's request.
41112 @node List of Supported Calls
41113 @subsection List of Supported Calls
41114 @cindex list of supported file-i/o calls
41131 @unnumberedsubsubsec open
41132 @cindex open, file-i/o system call
41137 int open(const char *pathname, int flags);
41138 int open(const char *pathname, int flags, mode_t mode);
41142 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
41145 @var{flags} is the bitwise @code{OR} of the following values:
41149 If the file does not exist it will be created. The host
41150 rules apply as far as file ownership and time stamps
41154 When used with @code{O_CREAT}, if the file already exists it is
41155 an error and open() fails.
41158 If the file already exists and the open mode allows
41159 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
41160 truncated to zero length.
41163 The file is opened in append mode.
41166 The file is opened for reading only.
41169 The file is opened for writing only.
41172 The file is opened for reading and writing.
41176 Other bits are silently ignored.
41180 @var{mode} is the bitwise @code{OR} of the following values:
41184 User has read permission.
41187 User has write permission.
41190 Group has read permission.
41193 Group has write permission.
41196 Others have read permission.
41199 Others have write permission.
41203 Other bits are silently ignored.
41206 @item Return value:
41207 @code{open} returns the new file descriptor or -1 if an error
41214 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
41217 @var{pathname} refers to a directory.
41220 The requested access is not allowed.
41223 @var{pathname} was too long.
41226 A directory component in @var{pathname} does not exist.
41229 @var{pathname} refers to a device, pipe, named pipe or socket.
41232 @var{pathname} refers to a file on a read-only filesystem and
41233 write access was requested.
41236 @var{pathname} is an invalid pointer value.
41239 No space on device to create the file.
41242 The process already has the maximum number of files open.
41245 The limit on the total number of files open on the system
41249 The call was interrupted by the user.
41255 @unnumberedsubsubsec close
41256 @cindex close, file-i/o system call
41265 @samp{Fclose,@var{fd}}
41267 @item Return value:
41268 @code{close} returns zero on success, or -1 if an error occurred.
41274 @var{fd} isn't a valid open file descriptor.
41277 The call was interrupted by the user.
41283 @unnumberedsubsubsec read
41284 @cindex read, file-i/o system call
41289 int read(int fd, void *buf, unsigned int count);
41293 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
41295 @item Return value:
41296 On success, the number of bytes read is returned.
41297 Zero indicates end of file. If count is zero, read
41298 returns zero as well. On error, -1 is returned.
41304 @var{fd} is not a valid file descriptor or is not open for
41308 @var{bufptr} is an invalid pointer value.
41311 The call was interrupted by the user.
41317 @unnumberedsubsubsec write
41318 @cindex write, file-i/o system call
41323 int write(int fd, const void *buf, unsigned int count);
41327 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
41329 @item Return value:
41330 On success, the number of bytes written are returned.
41331 Zero indicates nothing was written. On error, -1
41338 @var{fd} is not a valid file descriptor or is not open for
41342 @var{bufptr} is an invalid pointer value.
41345 An attempt was made to write a file that exceeds the
41346 host-specific maximum file size allowed.
41349 No space on device to write the data.
41352 The call was interrupted by the user.
41358 @unnumberedsubsubsec lseek
41359 @cindex lseek, file-i/o system call
41364 long lseek (int fd, long offset, int flag);
41368 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
41370 @var{flag} is one of:
41374 The offset is set to @var{offset} bytes.
41377 The offset is set to its current location plus @var{offset}
41381 The offset is set to the size of the file plus @var{offset}
41385 @item Return value:
41386 On success, the resulting unsigned offset in bytes from
41387 the beginning of the file is returned. Otherwise, a
41388 value of -1 is returned.
41394 @var{fd} is not a valid open file descriptor.
41397 @var{fd} is associated with the @value{GDBN} console.
41400 @var{flag} is not a proper value.
41403 The call was interrupted by the user.
41409 @unnumberedsubsubsec rename
41410 @cindex rename, file-i/o system call
41415 int rename(const char *oldpath, const char *newpath);
41419 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
41421 @item Return value:
41422 On success, zero is returned. On error, -1 is returned.
41428 @var{newpath} is an existing directory, but @var{oldpath} is not a
41432 @var{newpath} is a non-empty directory.
41435 @var{oldpath} or @var{newpath} is a directory that is in use by some
41439 An attempt was made to make a directory a subdirectory
41443 A component used as a directory in @var{oldpath} or new
41444 path is not a directory. Or @var{oldpath} is a directory
41445 and @var{newpath} exists but is not a directory.
41448 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
41451 No access to the file or the path of the file.
41455 @var{oldpath} or @var{newpath} was too long.
41458 A directory component in @var{oldpath} or @var{newpath} does not exist.
41461 The file is on a read-only filesystem.
41464 The device containing the file has no room for the new
41468 The call was interrupted by the user.
41474 @unnumberedsubsubsec unlink
41475 @cindex unlink, file-i/o system call
41480 int unlink(const char *pathname);
41484 @samp{Funlink,@var{pathnameptr}/@var{len}}
41486 @item Return value:
41487 On success, zero is returned. On error, -1 is returned.
41493 No access to the file or the path of the file.
41496 The system does not allow unlinking of directories.
41499 The file @var{pathname} cannot be unlinked because it's
41500 being used by another process.
41503 @var{pathnameptr} is an invalid pointer value.
41506 @var{pathname} was too long.
41509 A directory component in @var{pathname} does not exist.
41512 A component of the path is not a directory.
41515 The file is on a read-only filesystem.
41518 The call was interrupted by the user.
41524 @unnumberedsubsubsec stat/fstat
41525 @cindex fstat, file-i/o system call
41526 @cindex stat, file-i/o system call
41531 int stat(const char *pathname, struct stat *buf);
41532 int fstat(int fd, struct stat *buf);
41536 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
41537 @samp{Ffstat,@var{fd},@var{bufptr}}
41539 @item Return value:
41540 On success, zero is returned. On error, -1 is returned.
41546 @var{fd} is not a valid open file.
41549 A directory component in @var{pathname} does not exist or the
41550 path is an empty string.
41553 A component of the path is not a directory.
41556 @var{pathnameptr} is an invalid pointer value.
41559 No access to the file or the path of the file.
41562 @var{pathname} was too long.
41565 The call was interrupted by the user.
41571 @unnumberedsubsubsec gettimeofday
41572 @cindex gettimeofday, file-i/o system call
41577 int gettimeofday(struct timeval *tv, void *tz);
41581 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
41583 @item Return value:
41584 On success, 0 is returned, -1 otherwise.
41590 @var{tz} is a non-NULL pointer.
41593 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
41599 @unnumberedsubsubsec isatty
41600 @cindex isatty, file-i/o system call
41605 int isatty(int fd);
41609 @samp{Fisatty,@var{fd}}
41611 @item Return value:
41612 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
41618 The call was interrupted by the user.
41623 Note that the @code{isatty} call is treated as a special case: it returns
41624 1 to the target if the file descriptor is attached
41625 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
41626 would require implementing @code{ioctl} and would be more complex than
41631 @unnumberedsubsubsec system
41632 @cindex system, file-i/o system call
41637 int system(const char *command);
41641 @samp{Fsystem,@var{commandptr}/@var{len}}
41643 @item Return value:
41644 If @var{len} is zero, the return value indicates whether a shell is
41645 available. A zero return value indicates a shell is not available.
41646 For non-zero @var{len}, the value returned is -1 on error and the
41647 return status of the command otherwise. Only the exit status of the
41648 command is returned, which is extracted from the host's @code{system}
41649 return value by calling @code{WEXITSTATUS(retval)}. In case
41650 @file{/bin/sh} could not be executed, 127 is returned.
41656 The call was interrupted by the user.
41661 @value{GDBN} takes over the full task of calling the necessary host calls
41662 to perform the @code{system} call. The return value of @code{system} on
41663 the host is simplified before it's returned
41664 to the target. Any termination signal information from the child process
41665 is discarded, and the return value consists
41666 entirely of the exit status of the called command.
41668 Due to security concerns, the @code{system} call is by default refused
41669 by @value{GDBN}. The user has to allow this call explicitly with the
41670 @code{set remote system-call-allowed 1} command.
41673 @item set remote system-call-allowed
41674 @kindex set remote system-call-allowed
41675 Control whether to allow the @code{system} calls in the File I/O
41676 protocol for the remote target. The default is zero (disabled).
41678 @item show remote system-call-allowed
41679 @kindex show remote system-call-allowed
41680 Show whether the @code{system} calls are allowed in the File I/O
41684 @node Protocol-specific Representation of Datatypes
41685 @subsection Protocol-specific Representation of Datatypes
41686 @cindex protocol-specific representation of datatypes, in file-i/o protocol
41689 * Integral Datatypes::
41691 * Memory Transfer::
41696 @node Integral Datatypes
41697 @unnumberedsubsubsec Integral Datatypes
41698 @cindex integral datatypes, in file-i/o protocol
41700 The integral datatypes used in the system calls are @code{int},
41701 @code{unsigned int}, @code{long}, @code{unsigned long},
41702 @code{mode_t}, and @code{time_t}.
41704 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
41705 implemented as 32 bit values in this protocol.
41707 @code{long} and @code{unsigned long} are implemented as 64 bit types.
41709 @xref{Limits}, for corresponding MIN and MAX values (similar to those
41710 in @file{limits.h}) to allow range checking on host and target.
41712 @code{time_t} datatypes are defined as seconds since the Epoch.
41714 All integral datatypes transferred as part of a memory read or write of a
41715 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
41718 @node Pointer Values
41719 @unnumberedsubsubsec Pointer Values
41720 @cindex pointer values, in file-i/o protocol
41722 Pointers to target data are transmitted as they are. An exception
41723 is made for pointers to buffers for which the length isn't
41724 transmitted as part of the function call, namely strings. Strings
41725 are transmitted as a pointer/length pair, both as hex values, e.g.@:
41732 which is a pointer to data of length 18 bytes at position 0x1aaf.
41733 The length is defined as the full string length in bytes, including
41734 the trailing null byte. For example, the string @code{"hello world"}
41735 at address 0x123456 is transmitted as
41741 @node Memory Transfer
41742 @unnumberedsubsubsec Memory Transfer
41743 @cindex memory transfer, in file-i/o protocol
41745 Structured data which is transferred using a memory read or write (for
41746 example, a @code{struct stat}) is expected to be in a protocol-specific format
41747 with all scalar multibyte datatypes being big endian. Translation to
41748 this representation needs to be done both by the target before the @code{F}
41749 packet is sent, and by @value{GDBN} before
41750 it transfers memory to the target. Transferred pointers to structured
41751 data should point to the already-coerced data at any time.
41755 @unnumberedsubsubsec struct stat
41756 @cindex struct stat, in file-i/o protocol
41758 The buffer of type @code{struct stat} used by the target and @value{GDBN}
41759 is defined as follows:
41763 unsigned int st_dev; /* device */
41764 unsigned int st_ino; /* inode */
41765 mode_t st_mode; /* protection */
41766 unsigned int st_nlink; /* number of hard links */
41767 unsigned int st_uid; /* user ID of owner */
41768 unsigned int st_gid; /* group ID of owner */
41769 unsigned int st_rdev; /* device type (if inode device) */
41770 unsigned long st_size; /* total size, in bytes */
41771 unsigned long st_blksize; /* blocksize for filesystem I/O */
41772 unsigned long st_blocks; /* number of blocks allocated */
41773 time_t st_atime; /* time of last access */
41774 time_t st_mtime; /* time of last modification */
41775 time_t st_ctime; /* time of last change */
41779 The integral datatypes conform to the definitions given in the
41780 appropriate section (see @ref{Integral Datatypes}, for details) so this
41781 structure is of size 64 bytes.
41783 The values of several fields have a restricted meaning and/or
41789 A value of 0 represents a file, 1 the console.
41792 No valid meaning for the target. Transmitted unchanged.
41795 Valid mode bits are described in @ref{Constants}. Any other
41796 bits have currently no meaning for the target.
41801 No valid meaning for the target. Transmitted unchanged.
41806 These values have a host and file system dependent
41807 accuracy. Especially on Windows hosts, the file system may not
41808 support exact timing values.
41811 The target gets a @code{struct stat} of the above representation and is
41812 responsible for coercing it to the target representation before
41815 Note that due to size differences between the host, target, and protocol
41816 representations of @code{struct stat} members, these members could eventually
41817 get truncated on the target.
41819 @node struct timeval
41820 @unnumberedsubsubsec struct timeval
41821 @cindex struct timeval, in file-i/o protocol
41823 The buffer of type @code{struct timeval} used by the File-I/O protocol
41824 is defined as follows:
41828 time_t tv_sec; /* second */
41829 long tv_usec; /* microsecond */
41833 The integral datatypes conform to the definitions given in the
41834 appropriate section (see @ref{Integral Datatypes}, for details) so this
41835 structure is of size 8 bytes.
41838 @subsection Constants
41839 @cindex constants, in file-i/o protocol
41841 The following values are used for the constants inside of the
41842 protocol. @value{GDBN} and target are responsible for translating these
41843 values before and after the call as needed.
41854 @unnumberedsubsubsec Open Flags
41855 @cindex open flags, in file-i/o protocol
41857 All values are given in hexadecimal representation.
41869 @node mode_t Values
41870 @unnumberedsubsubsec mode_t Values
41871 @cindex mode_t values, in file-i/o protocol
41873 All values are given in octal representation.
41890 @unnumberedsubsubsec Errno Values
41891 @cindex errno values, in file-i/o protocol
41893 All values are given in decimal representation.
41918 @code{EUNKNOWN} is used as a fallback error value if a host system returns
41919 any error value not in the list of supported error numbers.
41922 @unnumberedsubsubsec Lseek Flags
41923 @cindex lseek flags, in file-i/o protocol
41932 @unnumberedsubsubsec Limits
41933 @cindex limits, in file-i/o protocol
41935 All values are given in decimal representation.
41938 INT_MIN -2147483648
41940 UINT_MAX 4294967295
41941 LONG_MIN -9223372036854775808
41942 LONG_MAX 9223372036854775807
41943 ULONG_MAX 18446744073709551615
41946 @node File-I/O Examples
41947 @subsection File-I/O Examples
41948 @cindex file-i/o examples
41950 Example sequence of a write call, file descriptor 3, buffer is at target
41951 address 0x1234, 6 bytes should be written:
41954 <- @code{Fwrite,3,1234,6}
41955 @emph{request memory read from target}
41958 @emph{return "6 bytes written"}
41962 Example sequence of a read call, file descriptor 3, buffer is at target
41963 address 0x1234, 6 bytes should be read:
41966 <- @code{Fread,3,1234,6}
41967 @emph{request memory write to target}
41968 -> @code{X1234,6:XXXXXX}
41969 @emph{return "6 bytes read"}
41973 Example sequence of a read call, call fails on the host due to invalid
41974 file descriptor (@code{EBADF}):
41977 <- @code{Fread,3,1234,6}
41981 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
41985 <- @code{Fread,3,1234,6}
41990 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
41994 <- @code{Fread,3,1234,6}
41995 -> @code{X1234,6:XXXXXX}
41999 @node Library List Format
42000 @section Library List Format
42001 @cindex library list format, remote protocol
42003 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
42004 same process as your application to manage libraries. In this case,
42005 @value{GDBN} can use the loader's symbol table and normal memory
42006 operations to maintain a list of shared libraries. On other
42007 platforms, the operating system manages loaded libraries.
42008 @value{GDBN} can not retrieve the list of currently loaded libraries
42009 through memory operations, so it uses the @samp{qXfer:libraries:read}
42010 packet (@pxref{qXfer library list read}) instead. The remote stub
42011 queries the target's operating system and reports which libraries
42014 The @samp{qXfer:libraries:read} packet returns an XML document which
42015 lists loaded libraries and their offsets. Each library has an
42016 associated name and one or more segment or section base addresses,
42017 which report where the library was loaded in memory.
42019 For the common case of libraries that are fully linked binaries, the
42020 library should have a list of segments. If the target supports
42021 dynamic linking of a relocatable object file, its library XML element
42022 should instead include a list of allocated sections. The segment or
42023 section bases are start addresses, not relocation offsets; they do not
42024 depend on the library's link-time base addresses.
42026 @value{GDBN} must be linked with the Expat library to support XML
42027 library lists. @xref{Expat}.
42029 A simple memory map, with one loaded library relocated by a single
42030 offset, looks like this:
42034 <library name="/lib/libc.so.6">
42035 <segment address="0x10000000"/>
42040 Another simple memory map, with one loaded library with three
42041 allocated sections (.text, .data, .bss), looks like this:
42045 <library name="sharedlib.o">
42046 <section address="0x10000000"/>
42047 <section address="0x20000000"/>
42048 <section address="0x30000000"/>
42053 The format of a library list is described by this DTD:
42056 <!-- library-list: Root element with versioning -->
42057 <!ELEMENT library-list (library)*>
42058 <!ATTLIST library-list version CDATA #FIXED "1.0">
42059 <!ELEMENT library (segment*, section*)>
42060 <!ATTLIST library name CDATA #REQUIRED>
42061 <!ELEMENT segment EMPTY>
42062 <!ATTLIST segment address CDATA #REQUIRED>
42063 <!ELEMENT section EMPTY>
42064 <!ATTLIST section address CDATA #REQUIRED>
42067 In addition, segments and section descriptors cannot be mixed within a
42068 single library element, and you must supply at least one segment or
42069 section for each library.
42071 @node Library List Format for SVR4 Targets
42072 @section Library List Format for SVR4 Targets
42073 @cindex library list format, remote protocol
42075 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
42076 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
42077 shared libraries. Still a special library list provided by this packet is
42078 more efficient for the @value{GDBN} remote protocol.
42080 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
42081 loaded libraries and their SVR4 linker parameters. For each library on SVR4
42082 target, the following parameters are reported:
42086 @code{name}, the absolute file name from the @code{l_name} field of
42087 @code{struct link_map}.
42089 @code{lm} with address of @code{struct link_map} used for TLS
42090 (Thread Local Storage) access.
42092 @code{l_addr}, the displacement as read from the field @code{l_addr} of
42093 @code{struct link_map}. For prelinked libraries this is not an absolute
42094 memory address. It is a displacement of absolute memory address against
42095 address the file was prelinked to during the library load.
42097 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
42100 Additionally the single @code{main-lm} attribute specifies address of
42101 @code{struct link_map} used for the main executable. This parameter is used
42102 for TLS access and its presence is optional.
42104 @value{GDBN} must be linked with the Expat library to support XML
42105 SVR4 library lists. @xref{Expat}.
42107 A simple memory map, with two loaded libraries (which do not use prelink),
42111 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
42112 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
42114 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
42116 </library-list-svr>
42119 The format of an SVR4 library list is described by this DTD:
42122 <!-- library-list-svr4: Root element with versioning -->
42123 <!ELEMENT library-list-svr4 (library)*>
42124 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
42125 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
42126 <!ELEMENT library EMPTY>
42127 <!ATTLIST library name CDATA #REQUIRED>
42128 <!ATTLIST library lm CDATA #REQUIRED>
42129 <!ATTLIST library l_addr CDATA #REQUIRED>
42130 <!ATTLIST library l_ld CDATA #REQUIRED>
42133 @node Memory Map Format
42134 @section Memory Map Format
42135 @cindex memory map format
42137 To be able to write into flash memory, @value{GDBN} needs to obtain a
42138 memory map from the target. This section describes the format of the
42141 The memory map is obtained using the @samp{qXfer:memory-map:read}
42142 (@pxref{qXfer memory map read}) packet and is an XML document that
42143 lists memory regions.
42145 @value{GDBN} must be linked with the Expat library to support XML
42146 memory maps. @xref{Expat}.
42148 The top-level structure of the document is shown below:
42151 <?xml version="1.0"?>
42152 <!DOCTYPE memory-map
42153 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
42154 "http://sourceware.org/gdb/gdb-memory-map.dtd">
42160 Each region can be either:
42165 A region of RAM starting at @var{addr} and extending for @var{length}
42169 <memory type="ram" start="@var{addr}" length="@var{length}"/>
42174 A region of read-only memory:
42177 <memory type="rom" start="@var{addr}" length="@var{length}"/>
42182 A region of flash memory, with erasure blocks @var{blocksize}
42186 <memory type="flash" start="@var{addr}" length="@var{length}">
42187 <property name="blocksize">@var{blocksize}</property>
42193 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
42194 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
42195 packets to write to addresses in such ranges.
42197 The formal DTD for memory map format is given below:
42200 <!-- ................................................... -->
42201 <!-- Memory Map XML DTD ................................ -->
42202 <!-- File: memory-map.dtd .............................. -->
42203 <!-- .................................... .............. -->
42204 <!-- memory-map.dtd -->
42205 <!-- memory-map: Root element with versioning -->
42206 <!ELEMENT memory-map (memory | property)>
42207 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
42208 <!ELEMENT memory (property)>
42209 <!-- memory: Specifies a memory region,
42210 and its type, or device. -->
42211 <!ATTLIST memory type CDATA #REQUIRED
42212 start CDATA #REQUIRED
42213 length CDATA #REQUIRED
42214 device CDATA #IMPLIED>
42215 <!-- property: Generic attribute tag -->
42216 <!ELEMENT property (#PCDATA | property)*>
42217 <!ATTLIST property name CDATA #REQUIRED>
42220 @node Thread List Format
42221 @section Thread List Format
42222 @cindex thread list format
42224 To efficiently update the list of threads and their attributes,
42225 @value{GDBN} issues the @samp{qXfer:threads:read} packet
42226 (@pxref{qXfer threads read}) and obtains the XML document with
42227 the following structure:
42230 <?xml version="1.0"?>
42232 <thread id="id" core="0">
42233 ... description ...
42238 Each @samp{thread} element must have the @samp{id} attribute that
42239 identifies the thread (@pxref{thread-id syntax}). The
42240 @samp{core} attribute, if present, specifies which processor core
42241 the thread was last executing on. The content of the of @samp{thread}
42242 element is interpreted as human-readable auxilliary information.
42244 @node Traceframe Info Format
42245 @section Traceframe Info Format
42246 @cindex traceframe info format
42248 To be able to know which objects in the inferior can be examined when
42249 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
42250 memory ranges, registers and trace state variables that have been
42251 collected in a traceframe.
42253 This list is obtained using the @samp{qXfer:traceframe-info:read}
42254 (@pxref{qXfer traceframe info read}) packet and is an XML document.
42256 @value{GDBN} must be linked with the Expat library to support XML
42257 traceframe info discovery. @xref{Expat}.
42259 The top-level structure of the document is shown below:
42262 <?xml version="1.0"?>
42263 <!DOCTYPE traceframe-info
42264 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
42265 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
42271 Each traceframe block can be either:
42276 A region of collected memory starting at @var{addr} and extending for
42277 @var{length} bytes from there:
42280 <memory start="@var{addr}" length="@var{length}"/>
42284 A block indicating trace state variable numbered @var{number} has been
42288 <tvar id="@var{number}"/>
42293 The formal DTD for the traceframe info format is given below:
42296 <!ELEMENT traceframe-info (memory | tvar)* >
42297 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
42299 <!ELEMENT memory EMPTY>
42300 <!ATTLIST memory start CDATA #REQUIRED
42301 length CDATA #REQUIRED>
42303 <!ATTLIST tvar id CDATA #REQUIRED>
42306 @node Branch Trace Format
42307 @section Branch Trace Format
42308 @cindex branch trace format
42310 In order to display the branch trace of an inferior thread,
42311 @value{GDBN} needs to obtain the list of branches. This list is
42312 represented as list of sequential code blocks that are connected via
42313 branches. The code in each block has been executed sequentially.
42315 This list is obtained using the @samp{qXfer:btrace:read}
42316 (@pxref{qXfer btrace read}) packet and is an XML document.
42318 @value{GDBN} must be linked with the Expat library to support XML
42319 traceframe info discovery. @xref{Expat}.
42321 The top-level structure of the document is shown below:
42324 <?xml version="1.0"?>
42326 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
42327 "http://sourceware.org/gdb/gdb-btrace.dtd">
42336 A block of sequentially executed instructions starting at @var{begin}
42337 and ending at @var{end}:
42340 <block begin="@var{begin}" end="@var{end}"/>
42345 The formal DTD for the branch trace format is given below:
42348 <!ELEMENT btrace (block)* >
42349 <!ATTLIST btrace version CDATA #FIXED "1.0">
42351 <!ELEMENT block EMPTY>
42352 <!ATTLIST block begin CDATA #REQUIRED
42353 end CDATA #REQUIRED>
42356 @include agentexpr.texi
42358 @node Target Descriptions
42359 @appendix Target Descriptions
42360 @cindex target descriptions
42362 One of the challenges of using @value{GDBN} to debug embedded systems
42363 is that there are so many minor variants of each processor
42364 architecture in use. It is common practice for vendors to start with
42365 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
42366 and then make changes to adapt it to a particular market niche. Some
42367 architectures have hundreds of variants, available from dozens of
42368 vendors. This leads to a number of problems:
42372 With so many different customized processors, it is difficult for
42373 the @value{GDBN} maintainers to keep up with the changes.
42375 Since individual variants may have short lifetimes or limited
42376 audiences, it may not be worthwhile to carry information about every
42377 variant in the @value{GDBN} source tree.
42379 When @value{GDBN} does support the architecture of the embedded system
42380 at hand, the task of finding the correct architecture name to give the
42381 @command{set architecture} command can be error-prone.
42384 To address these problems, the @value{GDBN} remote protocol allows a
42385 target system to not only identify itself to @value{GDBN}, but to
42386 actually describe its own features. This lets @value{GDBN} support
42387 processor variants it has never seen before --- to the extent that the
42388 descriptions are accurate, and that @value{GDBN} understands them.
42390 @value{GDBN} must be linked with the Expat library to support XML
42391 target descriptions. @xref{Expat}.
42394 * Retrieving Descriptions:: How descriptions are fetched from a target.
42395 * Target Description Format:: The contents of a target description.
42396 * Predefined Target Types:: Standard types available for target
42398 * Standard Target Features:: Features @value{GDBN} knows about.
42401 @node Retrieving Descriptions
42402 @section Retrieving Descriptions
42404 Target descriptions can be read from the target automatically, or
42405 specified by the user manually. The default behavior is to read the
42406 description from the target. @value{GDBN} retrieves it via the remote
42407 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
42408 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
42409 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
42410 XML document, of the form described in @ref{Target Description
42413 Alternatively, you can specify a file to read for the target description.
42414 If a file is set, the target will not be queried. The commands to
42415 specify a file are:
42418 @cindex set tdesc filename
42419 @item set tdesc filename @var{path}
42420 Read the target description from @var{path}.
42422 @cindex unset tdesc filename
42423 @item unset tdesc filename
42424 Do not read the XML target description from a file. @value{GDBN}
42425 will use the description supplied by the current target.
42427 @cindex show tdesc filename
42428 @item show tdesc filename
42429 Show the filename to read for a target description, if any.
42433 @node Target Description Format
42434 @section Target Description Format
42435 @cindex target descriptions, XML format
42437 A target description annex is an @uref{http://www.w3.org/XML/, XML}
42438 document which complies with the Document Type Definition provided in
42439 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
42440 means you can use generally available tools like @command{xmllint} to
42441 check that your feature descriptions are well-formed and valid.
42442 However, to help people unfamiliar with XML write descriptions for
42443 their targets, we also describe the grammar here.
42445 Target descriptions can identify the architecture of the remote target
42446 and (for some architectures) provide information about custom register
42447 sets. They can also identify the OS ABI of the remote target.
42448 @value{GDBN} can use this information to autoconfigure for your
42449 target, or to warn you if you connect to an unsupported target.
42451 Here is a simple target description:
42454 <target version="1.0">
42455 <architecture>i386:x86-64</architecture>
42460 This minimal description only says that the target uses
42461 the x86-64 architecture.
42463 A target description has the following overall form, with [ ] marking
42464 optional elements and @dots{} marking repeatable elements. The elements
42465 are explained further below.
42468 <?xml version="1.0"?>
42469 <!DOCTYPE target SYSTEM "gdb-target.dtd">
42470 <target version="1.0">
42471 @r{[}@var{architecture}@r{]}
42472 @r{[}@var{osabi}@r{]}
42473 @r{[}@var{compatible}@r{]}
42474 @r{[}@var{feature}@dots{}@r{]}
42479 The description is generally insensitive to whitespace and line
42480 breaks, under the usual common-sense rules. The XML version
42481 declaration and document type declaration can generally be omitted
42482 (@value{GDBN} does not require them), but specifying them may be
42483 useful for XML validation tools. The @samp{version} attribute for
42484 @samp{<target>} may also be omitted, but we recommend
42485 including it; if future versions of @value{GDBN} use an incompatible
42486 revision of @file{gdb-target.dtd}, they will detect and report
42487 the version mismatch.
42489 @subsection Inclusion
42490 @cindex target descriptions, inclusion
42493 @cindex <xi:include>
42496 It can sometimes be valuable to split a target description up into
42497 several different annexes, either for organizational purposes, or to
42498 share files between different possible target descriptions. You can
42499 divide a description into multiple files by replacing any element of
42500 the target description with an inclusion directive of the form:
42503 <xi:include href="@var{document}"/>
42507 When @value{GDBN} encounters an element of this form, it will retrieve
42508 the named XML @var{document}, and replace the inclusion directive with
42509 the contents of that document. If the current description was read
42510 using @samp{qXfer}, then so will be the included document;
42511 @var{document} will be interpreted as the name of an annex. If the
42512 current description was read from a file, @value{GDBN} will look for
42513 @var{document} as a file in the same directory where it found the
42514 original description.
42516 @subsection Architecture
42517 @cindex <architecture>
42519 An @samp{<architecture>} element has this form:
42522 <architecture>@var{arch}</architecture>
42525 @var{arch} is one of the architectures from the set accepted by
42526 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
42529 @cindex @code{<osabi>}
42531 This optional field was introduced in @value{GDBN} version 7.0.
42532 Previous versions of @value{GDBN} ignore it.
42534 An @samp{<osabi>} element has this form:
42537 <osabi>@var{abi-name}</osabi>
42540 @var{abi-name} is an OS ABI name from the same selection accepted by
42541 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
42543 @subsection Compatible Architecture
42544 @cindex @code{<compatible>}
42546 This optional field was introduced in @value{GDBN} version 7.0.
42547 Previous versions of @value{GDBN} ignore it.
42549 A @samp{<compatible>} element has this form:
42552 <compatible>@var{arch}</compatible>
42555 @var{arch} is one of the architectures from the set accepted by
42556 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
42558 A @samp{<compatible>} element is used to specify that the target
42559 is able to run binaries in some other than the main target architecture
42560 given by the @samp{<architecture>} element. For example, on the
42561 Cell Broadband Engine, the main architecture is @code{powerpc:common}
42562 or @code{powerpc:common64}, but the system is able to run binaries
42563 in the @code{spu} architecture as well. The way to describe this
42564 capability with @samp{<compatible>} is as follows:
42567 <architecture>powerpc:common</architecture>
42568 <compatible>spu</compatible>
42571 @subsection Features
42574 Each @samp{<feature>} describes some logical portion of the target
42575 system. Features are currently used to describe available CPU
42576 registers and the types of their contents. A @samp{<feature>} element
42580 <feature name="@var{name}">
42581 @r{[}@var{type}@dots{}@r{]}
42587 Each feature's name should be unique within the description. The name
42588 of a feature does not matter unless @value{GDBN} has some special
42589 knowledge of the contents of that feature; if it does, the feature
42590 should have its standard name. @xref{Standard Target Features}.
42594 Any register's value is a collection of bits which @value{GDBN} must
42595 interpret. The default interpretation is a two's complement integer,
42596 but other types can be requested by name in the register description.
42597 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
42598 Target Types}), and the description can define additional composite types.
42600 Each type element must have an @samp{id} attribute, which gives
42601 a unique (within the containing @samp{<feature>}) name to the type.
42602 Types must be defined before they are used.
42605 Some targets offer vector registers, which can be treated as arrays
42606 of scalar elements. These types are written as @samp{<vector>} elements,
42607 specifying the array element type, @var{type}, and the number of elements,
42611 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
42615 If a register's value is usefully viewed in multiple ways, define it
42616 with a union type containing the useful representations. The
42617 @samp{<union>} element contains one or more @samp{<field>} elements,
42618 each of which has a @var{name} and a @var{type}:
42621 <union id="@var{id}">
42622 <field name="@var{name}" type="@var{type}"/>
42628 If a register's value is composed from several separate values, define
42629 it with a structure type. There are two forms of the @samp{<struct>}
42630 element; a @samp{<struct>} element must either contain only bitfields
42631 or contain no bitfields. If the structure contains only bitfields,
42632 its total size in bytes must be specified, each bitfield must have an
42633 explicit start and end, and bitfields are automatically assigned an
42634 integer type. The field's @var{start} should be less than or
42635 equal to its @var{end}, and zero represents the least significant bit.
42638 <struct id="@var{id}" size="@var{size}">
42639 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
42644 If the structure contains no bitfields, then each field has an
42645 explicit type, and no implicit padding is added.
42648 <struct id="@var{id}">
42649 <field name="@var{name}" type="@var{type}"/>
42655 If a register's value is a series of single-bit flags, define it with
42656 a flags type. The @samp{<flags>} element has an explicit @var{size}
42657 and contains one or more @samp{<field>} elements. Each field has a
42658 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
42662 <flags id="@var{id}" size="@var{size}">
42663 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
42668 @subsection Registers
42671 Each register is represented as an element with this form:
42674 <reg name="@var{name}"
42675 bitsize="@var{size}"
42676 @r{[}regnum="@var{num}"@r{]}
42677 @r{[}save-restore="@var{save-restore}"@r{]}
42678 @r{[}type="@var{type}"@r{]}
42679 @r{[}group="@var{group}"@r{]}/>
42683 The components are as follows:
42688 The register's name; it must be unique within the target description.
42691 The register's size, in bits.
42694 The register's number. If omitted, a register's number is one greater
42695 than that of the previous register (either in the current feature or in
42696 a preceding feature); the first register in the target description
42697 defaults to zero. This register number is used to read or write
42698 the register; e.g.@: it is used in the remote @code{p} and @code{P}
42699 packets, and registers appear in the @code{g} and @code{G} packets
42700 in order of increasing register number.
42703 Whether the register should be preserved across inferior function
42704 calls; this must be either @code{yes} or @code{no}. The default is
42705 @code{yes}, which is appropriate for most registers except for
42706 some system control registers; this is not related to the target's
42710 The type of the register. @var{type} may be a predefined type, a type
42711 defined in the current feature, or one of the special types @code{int}
42712 and @code{float}. @code{int} is an integer type of the correct size
42713 for @var{bitsize}, and @code{float} is a floating point type (in the
42714 architecture's normal floating point format) of the correct size for
42715 @var{bitsize}. The default is @code{int}.
42718 The register group to which this register belongs. @var{group} must
42719 be either @code{general}, @code{float}, or @code{vector}. If no
42720 @var{group} is specified, @value{GDBN} will not display the register
42721 in @code{info registers}.
42725 @node Predefined Target Types
42726 @section Predefined Target Types
42727 @cindex target descriptions, predefined types
42729 Type definitions in the self-description can build up composite types
42730 from basic building blocks, but can not define fundamental types. Instead,
42731 standard identifiers are provided by @value{GDBN} for the fundamental
42732 types. The currently supported types are:
42741 Signed integer types holding the specified number of bits.
42748 Unsigned integer types holding the specified number of bits.
42752 Pointers to unspecified code and data. The program counter and
42753 any dedicated return address register may be marked as code
42754 pointers; printing a code pointer converts it into a symbolic
42755 address. The stack pointer and any dedicated address registers
42756 may be marked as data pointers.
42759 Single precision IEEE floating point.
42762 Double precision IEEE floating point.
42765 The 12-byte extended precision format used by ARM FPA registers.
42768 The 10-byte extended precision format used by x87 registers.
42771 32bit @sc{eflags} register used by x86.
42774 32bit @sc{mxcsr} register used by x86.
42778 @node Standard Target Features
42779 @section Standard Target Features
42780 @cindex target descriptions, standard features
42782 A target description must contain either no registers or all the
42783 target's registers. If the description contains no registers, then
42784 @value{GDBN} will assume a default register layout, selected based on
42785 the architecture. If the description contains any registers, the
42786 default layout will not be used; the standard registers must be
42787 described in the target description, in such a way that @value{GDBN}
42788 can recognize them.
42790 This is accomplished by giving specific names to feature elements
42791 which contain standard registers. @value{GDBN} will look for features
42792 with those names and verify that they contain the expected registers;
42793 if any known feature is missing required registers, or if any required
42794 feature is missing, @value{GDBN} will reject the target
42795 description. You can add additional registers to any of the
42796 standard features --- @value{GDBN} will display them just as if
42797 they were added to an unrecognized feature.
42799 This section lists the known features and their expected contents.
42800 Sample XML documents for these features are included in the
42801 @value{GDBN} source tree, in the directory @file{gdb/features}.
42803 Names recognized by @value{GDBN} should include the name of the
42804 company or organization which selected the name, and the overall
42805 architecture to which the feature applies; so e.g.@: the feature
42806 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
42808 The names of registers are not case sensitive for the purpose
42809 of recognizing standard features, but @value{GDBN} will only display
42810 registers using the capitalization used in the description.
42813 * AArch64 Features::
42818 * Nios II Features::
42819 * PowerPC Features::
42820 * S/390 and System z Features::
42825 @node AArch64 Features
42826 @subsection AArch64 Features
42827 @cindex target descriptions, AArch64 features
42829 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
42830 targets. It should contain registers @samp{x0} through @samp{x30},
42831 @samp{sp}, @samp{pc}, and @samp{cpsr}.
42833 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
42834 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
42838 @subsection ARM Features
42839 @cindex target descriptions, ARM features
42841 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
42843 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
42844 @samp{lr}, @samp{pc}, and @samp{cpsr}.
42846 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
42847 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
42848 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
42851 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
42852 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
42854 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
42855 it should contain at least registers @samp{wR0} through @samp{wR15} and
42856 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
42857 @samp{wCSSF}, and @samp{wCASF} registers are optional.
42859 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
42860 should contain at least registers @samp{d0} through @samp{d15}. If
42861 they are present, @samp{d16} through @samp{d31} should also be included.
42862 @value{GDBN} will synthesize the single-precision registers from
42863 halves of the double-precision registers.
42865 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
42866 need to contain registers; it instructs @value{GDBN} to display the
42867 VFP double-precision registers as vectors and to synthesize the
42868 quad-precision registers from pairs of double-precision registers.
42869 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
42870 be present and include 32 double-precision registers.
42872 @node i386 Features
42873 @subsection i386 Features
42874 @cindex target descriptions, i386 features
42876 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
42877 targets. It should describe the following registers:
42881 @samp{eax} through @samp{edi} plus @samp{eip} for i386
42883 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
42885 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
42886 @samp{fs}, @samp{gs}
42888 @samp{st0} through @samp{st7}
42890 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
42891 @samp{foseg}, @samp{fooff} and @samp{fop}
42894 The register sets may be different, depending on the target.
42896 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
42897 describe registers:
42901 @samp{xmm0} through @samp{xmm7} for i386
42903 @samp{xmm0} through @samp{xmm15} for amd64
42908 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
42909 @samp{org.gnu.gdb.i386.sse} feature. It should
42910 describe the upper 128 bits of @sc{ymm} registers:
42914 @samp{ymm0h} through @samp{ymm7h} for i386
42916 @samp{ymm0h} through @samp{ymm15h} for amd64
42919 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
42920 describe a single register, @samp{orig_eax}.
42922 @node MIPS Features
42923 @subsection @acronym{MIPS} Features
42924 @cindex target descriptions, @acronym{MIPS} features
42926 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
42927 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
42928 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
42931 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
42932 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
42933 registers. They may be 32-bit or 64-bit depending on the target.
42935 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
42936 it may be optional in a future version of @value{GDBN}. It should
42937 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
42938 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
42940 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
42941 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
42942 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
42943 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
42945 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
42946 contain a single register, @samp{restart}, which is used by the
42947 Linux kernel to control restartable syscalls.
42949 @node M68K Features
42950 @subsection M68K Features
42951 @cindex target descriptions, M68K features
42954 @item @samp{org.gnu.gdb.m68k.core}
42955 @itemx @samp{org.gnu.gdb.coldfire.core}
42956 @itemx @samp{org.gnu.gdb.fido.core}
42957 One of those features must be always present.
42958 The feature that is present determines which flavor of m68k is
42959 used. The feature that is present should contain registers
42960 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
42961 @samp{sp}, @samp{ps} and @samp{pc}.
42963 @item @samp{org.gnu.gdb.coldfire.fp}
42964 This feature is optional. If present, it should contain registers
42965 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
42969 @node Nios II Features
42970 @subsection Nios II Features
42971 @cindex target descriptions, Nios II features
42973 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
42974 targets. It should contain the 32 core registers (@samp{zero},
42975 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
42976 @samp{pc}, and the 16 control registers (@samp{status} through
42979 @node PowerPC Features
42980 @subsection PowerPC Features
42981 @cindex target descriptions, PowerPC features
42983 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
42984 targets. It should contain registers @samp{r0} through @samp{r31},
42985 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
42986 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
42988 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
42989 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
42991 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
42992 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
42995 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
42996 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
42997 will combine these registers with the floating point registers
42998 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
42999 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
43000 through @samp{vs63}, the set of vector registers for POWER7.
43002 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
43003 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
43004 @samp{spefscr}. SPE targets should provide 32-bit registers in
43005 @samp{org.gnu.gdb.power.core} and provide the upper halves in
43006 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
43007 these to present registers @samp{ev0} through @samp{ev31} to the
43010 @node S/390 and System z Features
43011 @subsection S/390 and System z Features
43012 @cindex target descriptions, S/390 features
43013 @cindex target descriptions, System z features
43015 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
43016 System z targets. It should contain the PSW and the 16 general
43017 registers. In particular, System z targets should provide the 64-bit
43018 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
43019 S/390 targets should provide the 32-bit versions of these registers.
43020 A System z target that runs in 31-bit addressing mode should provide
43021 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
43022 register's upper halves @samp{r0h} through @samp{r15h}, and their
43023 lower halves @samp{r0l} through @samp{r15l}.
43025 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
43026 contain the 64-bit registers @samp{f0} through @samp{f15}, and
43029 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
43030 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
43032 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
43033 contain the register @samp{orig_r2}, which is 64-bit wide on System z
43034 targets and 32-bit otherwise. In addition, the feature may contain
43035 the @samp{last_break} register, whose width depends on the addressing
43036 mode, as well as the @samp{system_call} register, which is always
43039 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
43040 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
43041 @samp{atia}, and @samp{tr0} through @samp{tr15}.
43043 @node TIC6x Features
43044 @subsection TMS320C6x Features
43045 @cindex target descriptions, TIC6x features
43046 @cindex target descriptions, TMS320C6x features
43047 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
43048 targets. It should contain registers @samp{A0} through @samp{A15},
43049 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
43051 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
43052 contain registers @samp{A16} through @samp{A31} and @samp{B16}
43053 through @samp{B31}.
43055 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
43056 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
43058 @node Operating System Information
43059 @appendix Operating System Information
43060 @cindex operating system information
43066 Users of @value{GDBN} often wish to obtain information about the state of
43067 the operating system running on the target---for example the list of
43068 processes, or the list of open files. This section describes the
43069 mechanism that makes it possible. This mechanism is similar to the
43070 target features mechanism (@pxref{Target Descriptions}), but focuses
43071 on a different aspect of target.
43073 Operating system information is retrived from the target via the
43074 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
43075 read}). The object name in the request should be @samp{osdata}, and
43076 the @var{annex} identifies the data to be fetched.
43079 @appendixsection Process list
43080 @cindex operating system information, process list
43082 When requesting the process list, the @var{annex} field in the
43083 @samp{qXfer} request should be @samp{processes}. The returned data is
43084 an XML document. The formal syntax of this document is defined in
43085 @file{gdb/features/osdata.dtd}.
43087 An example document is:
43090 <?xml version="1.0"?>
43091 <!DOCTYPE target SYSTEM "osdata.dtd">
43092 <osdata type="processes">
43094 <column name="pid">1</column>
43095 <column name="user">root</column>
43096 <column name="command">/sbin/init</column>
43097 <column name="cores">1,2,3</column>
43102 Each item should include a column whose name is @samp{pid}. The value
43103 of that column should identify the process on the target. The
43104 @samp{user} and @samp{command} columns are optional, and will be
43105 displayed by @value{GDBN}. The @samp{cores} column, if present,
43106 should contain a comma-separated list of cores that this process
43107 is running on. Target may provide additional columns,
43108 which @value{GDBN} currently ignores.
43110 @node Trace File Format
43111 @appendix Trace File Format
43112 @cindex trace file format
43114 The trace file comes in three parts: a header, a textual description
43115 section, and a trace frame section with binary data.
43117 The header has the form @code{\x7fTRACE0\n}. The first byte is
43118 @code{0x7f} so as to indicate that the file contains binary data,
43119 while the @code{0} is a version number that may have different values
43122 The description section consists of multiple lines of @sc{ascii} text
43123 separated by newline characters (@code{0xa}). The lines may include a
43124 variety of optional descriptive or context-setting information, such
43125 as tracepoint definitions or register set size. @value{GDBN} will
43126 ignore any line that it does not recognize. An empty line marks the end
43129 @c FIXME add some specific types of data
43131 The trace frame section consists of a number of consecutive frames.
43132 Each frame begins with a two-byte tracepoint number, followed by a
43133 four-byte size giving the amount of data in the frame. The data in
43134 the frame consists of a number of blocks, each introduced by a
43135 character indicating its type (at least register, memory, and trace
43136 state variable). The data in this section is raw binary, not a
43137 hexadecimal or other encoding; its endianness matches the target's
43140 @c FIXME bi-arch may require endianness/arch info in description section
43143 @item R @var{bytes}
43144 Register block. The number and ordering of bytes matches that of a
43145 @code{g} packet in the remote protocol. Note that these are the
43146 actual bytes, in target order and @value{GDBN} register order, not a
43147 hexadecimal encoding.
43149 @item M @var{address} @var{length} @var{bytes}...
43150 Memory block. This is a contiguous block of memory, at the 8-byte
43151 address @var{address}, with a 2-byte length @var{length}, followed by
43152 @var{length} bytes.
43154 @item V @var{number} @var{value}
43155 Trace state variable block. This records the 8-byte signed value
43156 @var{value} of trace state variable numbered @var{number}.
43160 Future enhancements of the trace file format may include additional types
43163 @node Index Section Format
43164 @appendix @code{.gdb_index} section format
43165 @cindex .gdb_index section format
43166 @cindex index section format
43168 This section documents the index section that is created by @code{save
43169 gdb-index} (@pxref{Index Files}). The index section is
43170 DWARF-specific; some knowledge of DWARF is assumed in this
43173 The mapped index file format is designed to be directly
43174 @code{mmap}able on any architecture. In most cases, a datum is
43175 represented using a little-endian 32-bit integer value, called an
43176 @code{offset_type}. Big endian machines must byte-swap the values
43177 before using them. Exceptions to this rule are noted. The data is
43178 laid out such that alignment is always respected.
43180 A mapped index consists of several areas, laid out in order.
43184 The file header. This is a sequence of values, of @code{offset_type}
43185 unless otherwise noted:
43189 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
43190 Version 4 uses a different hashing function from versions 5 and 6.
43191 Version 6 includes symbols for inlined functions, whereas versions 4
43192 and 5 do not. Version 7 adds attributes to the CU indices in the
43193 symbol table. Version 8 specifies that symbols from DWARF type units
43194 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
43195 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
43197 @value{GDBN} will only read version 4, 5, or 6 indices
43198 by specifying @code{set use-deprecated-index-sections on}.
43199 GDB has a workaround for potentially broken version 7 indices so it is
43200 currently not flagged as deprecated.
43203 The offset, from the start of the file, of the CU list.
43206 The offset, from the start of the file, of the types CU list. Note
43207 that this area can be empty, in which case this offset will be equal
43208 to the next offset.
43211 The offset, from the start of the file, of the address area.
43214 The offset, from the start of the file, of the symbol table.
43217 The offset, from the start of the file, of the constant pool.
43221 The CU list. This is a sequence of pairs of 64-bit little-endian
43222 values, sorted by the CU offset. The first element in each pair is
43223 the offset of a CU in the @code{.debug_info} section. The second
43224 element in each pair is the length of that CU. References to a CU
43225 elsewhere in the map are done using a CU index, which is just the
43226 0-based index into this table. Note that if there are type CUs, then
43227 conceptually CUs and type CUs form a single list for the purposes of
43231 The types CU list. This is a sequence of triplets of 64-bit
43232 little-endian values. In a triplet, the first value is the CU offset,
43233 the second value is the type offset in the CU, and the third value is
43234 the type signature. The types CU list is not sorted.
43237 The address area. The address area consists of a sequence of address
43238 entries. Each address entry has three elements:
43242 The low address. This is a 64-bit little-endian value.
43245 The high address. This is a 64-bit little-endian value. Like
43246 @code{DW_AT_high_pc}, the value is one byte beyond the end.
43249 The CU index. This is an @code{offset_type} value.
43253 The symbol table. This is an open-addressed hash table. The size of
43254 the hash table is always a power of 2.
43256 Each slot in the hash table consists of a pair of @code{offset_type}
43257 values. The first value is the offset of the symbol's name in the
43258 constant pool. The second value is the offset of the CU vector in the
43261 If both values are 0, then this slot in the hash table is empty. This
43262 is ok because while 0 is a valid constant pool index, it cannot be a
43263 valid index for both a string and a CU vector.
43265 The hash value for a table entry is computed by applying an
43266 iterative hash function to the symbol's name. Starting with an
43267 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
43268 the string is incorporated into the hash using the formula depending on the
43273 The formula is @code{r = r * 67 + c - 113}.
43275 @item Versions 5 to 7
43276 The formula is @code{r = r * 67 + tolower (c) - 113}.
43279 The terminating @samp{\0} is not incorporated into the hash.
43281 The step size used in the hash table is computed via
43282 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
43283 value, and @samp{size} is the size of the hash table. The step size
43284 is used to find the next candidate slot when handling a hash
43287 The names of C@t{++} symbols in the hash table are canonicalized. We
43288 don't currently have a simple description of the canonicalization
43289 algorithm; if you intend to create new index sections, you must read
43293 The constant pool. This is simply a bunch of bytes. It is organized
43294 so that alignment is correct: CU vectors are stored first, followed by
43297 A CU vector in the constant pool is a sequence of @code{offset_type}
43298 values. The first value is the number of CU indices in the vector.
43299 Each subsequent value is the index and symbol attributes of a CU in
43300 the CU list. This element in the hash table is used to indicate which
43301 CUs define the symbol and how the symbol is used.
43302 See below for the format of each CU index+attributes entry.
43304 A string in the constant pool is zero-terminated.
43307 Attributes were added to CU index values in @code{.gdb_index} version 7.
43308 If a symbol has multiple uses within a CU then there is one
43309 CU index+attributes value for each use.
43311 The format of each CU index+attributes entry is as follows
43317 This is the index of the CU in the CU list.
43319 These bits are reserved for future purposes and must be zero.
43321 The kind of the symbol in the CU.
43325 This value is reserved and should not be used.
43326 By reserving zero the full @code{offset_type} value is backwards compatible
43327 with previous versions of the index.
43329 The symbol is a type.
43331 The symbol is a variable or an enum value.
43333 The symbol is a function.
43335 Any other kind of symbol.
43337 These values are reserved.
43341 This bit is zero if the value is global and one if it is static.
43343 The determination of whether a symbol is global or static is complicated.
43344 The authorative reference is the file @file{dwarf2read.c} in
43345 @value{GDBN} sources.
43349 This pseudo-code describes the computation of a symbol's kind and
43350 global/static attributes in the index.
43353 is_external = get_attribute (die, DW_AT_external);
43354 language = get_attribute (cu_die, DW_AT_language);
43357 case DW_TAG_typedef:
43358 case DW_TAG_base_type:
43359 case DW_TAG_subrange_type:
43363 case DW_TAG_enumerator:
43365 is_static = (language != CPLUS && language != JAVA);
43367 case DW_TAG_subprogram:
43369 is_static = ! (is_external || language == ADA);
43371 case DW_TAG_constant:
43373 is_static = ! is_external;
43375 case DW_TAG_variable:
43377 is_static = ! is_external;
43379 case DW_TAG_namespace:
43383 case DW_TAG_class_type:
43384 case DW_TAG_interface_type:
43385 case DW_TAG_structure_type:
43386 case DW_TAG_union_type:
43387 case DW_TAG_enumeration_type:
43389 is_static = (language != CPLUS && language != JAVA);
43397 @appendix Manual pages
43401 * gdb man:: The GNU Debugger man page
43402 * gdbserver man:: Remote Server for the GNU Debugger man page
43403 * gcore man:: Generate a core file of a running program
43404 * gdbinit man:: gdbinit scripts
43410 @c man title gdb The GNU Debugger
43412 @c man begin SYNOPSIS gdb
43413 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
43414 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
43415 [@option{-b}@w{ }@var{bps}]
43416 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
43417 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
43418 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
43419 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
43420 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
43423 @c man begin DESCRIPTION gdb
43424 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
43425 going on ``inside'' another program while it executes -- or what another
43426 program was doing at the moment it crashed.
43428 @value{GDBN} can do four main kinds of things (plus other things in support of
43429 these) to help you catch bugs in the act:
43433 Start your program, specifying anything that might affect its behavior.
43436 Make your program stop on specified conditions.
43439 Examine what has happened, when your program has stopped.
43442 Change things in your program, so you can experiment with correcting the
43443 effects of one bug and go on to learn about another.
43446 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
43449 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
43450 commands from the terminal until you tell it to exit with the @value{GDBN}
43451 command @code{quit}. You can get online help from @value{GDBN} itself
43452 by using the command @code{help}.
43454 You can run @code{gdb} with no arguments or options; but the most
43455 usual way to start @value{GDBN} is with one argument or two, specifying an
43456 executable program as the argument:
43462 You can also start with both an executable program and a core file specified:
43468 You can, instead, specify a process ID as a second argument, if you want
43469 to debug a running process:
43477 would attach @value{GDBN} to process @code{1234} (unless you also have a file
43478 named @file{1234}; @value{GDBN} does check for a core file first).
43479 With option @option{-p} you can omit the @var{program} filename.
43481 Here are some of the most frequently needed @value{GDBN} commands:
43483 @c pod2man highlights the right hand side of the @item lines.
43485 @item break [@var{file}:]@var{functiop}
43486 Set a breakpoint at @var{function} (in @var{file}).
43488 @item run [@var{arglist}]
43489 Start your program (with @var{arglist}, if specified).
43492 Backtrace: display the program stack.
43494 @item print @var{expr}
43495 Display the value of an expression.
43498 Continue running your program (after stopping, e.g. at a breakpoint).
43501 Execute next program line (after stopping); step @emph{over} any
43502 function calls in the line.
43504 @item edit [@var{file}:]@var{function}
43505 look at the program line where it is presently stopped.
43507 @item list [@var{file}:]@var{function}
43508 type the text of the program in the vicinity of where it is presently stopped.
43511 Execute next program line (after stopping); step @emph{into} any
43512 function calls in the line.
43514 @item help [@var{name}]
43515 Show information about @value{GDBN} command @var{name}, or general information
43516 about using @value{GDBN}.
43519 Exit from @value{GDBN}.
43523 For full details on @value{GDBN},
43524 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
43525 by Richard M. Stallman and Roland H. Pesch. The same text is available online
43526 as the @code{gdb} entry in the @code{info} program.
43530 @c man begin OPTIONS gdb
43531 Any arguments other than options specify an executable
43532 file and core file (or process ID); that is, the first argument
43533 encountered with no
43534 associated option flag is equivalent to a @option{-se} option, and the second,
43535 if any, is equivalent to a @option{-c} option if it's the name of a file.
43537 both long and short forms; both are shown here. The long forms are also
43538 recognized if you truncate them, so long as enough of the option is
43539 present to be unambiguous. (If you prefer, you can flag option
43540 arguments with @option{+} rather than @option{-}, though we illustrate the
43541 more usual convention.)
43543 All the options and command line arguments you give are processed
43544 in sequential order. The order makes a difference when the @option{-x}
43550 List all options, with brief explanations.
43552 @item -symbols=@var{file}
43553 @itemx -s @var{file}
43554 Read symbol table from file @var{file}.
43557 Enable writing into executable and core files.
43559 @item -exec=@var{file}
43560 @itemx -e @var{file}
43561 Use file @var{file} as the executable file to execute when
43562 appropriate, and for examining pure data in conjunction with a core
43565 @item -se=@var{file}
43566 Read symbol table from file @var{file} and use it as the executable
43569 @item -core=@var{file}
43570 @itemx -c @var{file}
43571 Use file @var{file} as a core dump to examine.
43573 @item -command=@var{file}
43574 @itemx -x @var{file}
43575 Execute @value{GDBN} commands from file @var{file}.
43577 @item -ex @var{command}
43578 Execute given @value{GDBN} @var{command}.
43580 @item -directory=@var{directory}
43581 @itemx -d @var{directory}
43582 Add @var{directory} to the path to search for source files.
43585 Do not execute commands from @file{~/.gdbinit}.
43589 Do not execute commands from any @file{.gdbinit} initialization files.
43593 ``Quiet''. Do not print the introductory and copyright messages. These
43594 messages are also suppressed in batch mode.
43597 Run in batch mode. Exit with status @code{0} after processing all the command
43598 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
43599 Exit with nonzero status if an error occurs in executing the @value{GDBN}
43600 commands in the command files.
43602 Batch mode may be useful for running @value{GDBN} as a filter, for example to
43603 download and run a program on another computer; in order to make this
43604 more useful, the message
43607 Program exited normally.
43611 (which is ordinarily issued whenever a program running under @value{GDBN} control
43612 terminates) is not issued when running in batch mode.
43614 @item -cd=@var{directory}
43615 Run @value{GDBN} using @var{directory} as its working directory,
43616 instead of the current directory.
43620 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
43621 @value{GDBN} to output the full file name and line number in a standard,
43622 recognizable fashion each time a stack frame is displayed (which
43623 includes each time the program stops). This recognizable format looks
43624 like two @samp{\032} characters, followed by the file name, line number
43625 and character position separated by colons, and a newline. The
43626 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
43627 characters as a signal to display the source code for the frame.
43630 Set the line speed (baud rate or bits per second) of any serial
43631 interface used by @value{GDBN} for remote debugging.
43633 @item -tty=@var{device}
43634 Run using @var{device} for your program's standard input and output.
43638 @c man begin SEEALSO gdb
43640 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
43641 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
43642 documentation are properly installed at your site, the command
43649 should give you access to the complete manual.
43651 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
43652 Richard M. Stallman and Roland H. Pesch, July 1991.
43656 @node gdbserver man
43657 @heading gdbserver man
43659 @c man title gdbserver Remote Server for the GNU Debugger
43661 @c man begin SYNOPSIS gdbserver
43662 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
43664 gdbserver --attach @var{comm} @var{pid}
43666 gdbserver --multi @var{comm}
43670 @c man begin DESCRIPTION gdbserver
43671 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
43672 than the one which is running the program being debugged.
43675 @subheading Usage (server (target) side)
43678 Usage (server (target) side):
43681 First, you need to have a copy of the program you want to debug put onto
43682 the target system. The program can be stripped to save space if needed, as
43683 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
43684 the @value{GDBN} running on the host system.
43686 To use the server, you log on to the target system, and run the @command{gdbserver}
43687 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
43688 your program, and (c) its arguments. The general syntax is:
43691 target> gdbserver @var{comm} @var{program} [@var{args} ...]
43694 For example, using a serial port, you might say:
43698 @c @file would wrap it as F</dev/com1>.
43699 target> gdbserver /dev/com1 emacs foo.txt
43702 target> gdbserver @file{/dev/com1} emacs foo.txt
43706 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
43707 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
43708 waits patiently for the host @value{GDBN} to communicate with it.
43710 To use a TCP connection, you could say:
43713 target> gdbserver host:2345 emacs foo.txt
43716 This says pretty much the same thing as the last example, except that we are
43717 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
43718 that we are expecting to see a TCP connection from @code{host} to local TCP port
43719 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
43720 want for the port number as long as it does not conflict with any existing TCP
43721 ports on the target system. This same port number must be used in the host
43722 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
43723 you chose a port number that conflicts with another service, @command{gdbserver} will
43724 print an error message and exit.
43726 @command{gdbserver} can also attach to running programs.
43727 This is accomplished via the @option{--attach} argument. The syntax is:
43730 target> gdbserver --attach @var{comm} @var{pid}
43733 @var{pid} is the process ID of a currently running process. It isn't
43734 necessary to point @command{gdbserver} at a binary for the running process.
43736 To start @code{gdbserver} without supplying an initial command to run
43737 or process ID to attach, use the @option{--multi} command line option.
43738 In such case you should connect using @kbd{target extended-remote} to start
43739 the program you want to debug.
43742 target> gdbserver --multi @var{comm}
43746 @subheading Usage (host side)
43752 You need an unstripped copy of the target program on your host system, since
43753 @value{GDBN} needs to examine it's symbol tables and such. Start up @value{GDBN} as you normally
43754 would, with the target program as the first argument. (You may need to use the
43755 @option{--baud} option if the serial line is running at anything except 9600 baud.)
43756 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
43757 new command you need to know about is @code{target remote}
43758 (or @code{target extended-remote}). Its argument is either
43759 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
43760 descriptor. For example:
43764 @c @file would wrap it as F</dev/ttyb>.
43765 (gdb) target remote /dev/ttyb
43768 (gdb) target remote @file{/dev/ttyb}
43773 communicates with the server via serial line @file{/dev/ttyb}, and:
43776 (gdb) target remote the-target:2345
43780 communicates via a TCP connection to port 2345 on host `the-target', where
43781 you previously started up @command{gdbserver} with the same port number. Note that for
43782 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
43783 command, otherwise you may get an error that looks something like
43784 `Connection refused'.
43786 @command{gdbserver} can also debug multiple inferiors at once,
43789 the @value{GDBN} manual in node @code{Inferiors and Programs}
43790 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
43793 @ref{Inferiors and Programs}.
43795 In such case use the @code{extended-remote} @value{GDBN} command variant:
43798 (gdb) target extended-remote the-target:2345
43801 The @command{gdbserver} option @option{--multi} may or may not be used in such
43805 @c man begin OPTIONS gdbserver
43806 There are three different modes for invoking @command{gdbserver}:
43811 Debug a specific program specified by its program name:
43814 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
43817 The @var{comm} parameter specifies how should the server communicate
43818 with @value{GDBN}; it is either a device name (to use a serial line),
43819 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
43820 stdin/stdout of @code{gdbserver}. Specify the name of the program to
43821 debug in @var{prog}. Any remaining arguments will be passed to the
43822 program verbatim. When the program exits, @value{GDBN} will close the
43823 connection, and @code{gdbserver} will exit.
43826 Debug a specific program by specifying the process ID of a running
43830 gdbserver --attach @var{comm} @var{pid}
43833 The @var{comm} parameter is as described above. Supply the process ID
43834 of a running program in @var{pid}; @value{GDBN} will do everything
43835 else. Like with the previous mode, when the process @var{pid} exits,
43836 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
43839 Multi-process mode -- debug more than one program/process:
43842 gdbserver --multi @var{comm}
43845 In this mode, @value{GDBN} can instruct @command{gdbserver} which
43846 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
43847 close the connection when a process being debugged exits, so you can
43848 debug several processes in the same session.
43851 In each of the modes you may specify these options:
43856 List all options, with brief explanations.
43859 This option causes @command{gdbserver} to print its version number and exit.
43862 @command{gdbserver} will attach to a running program. The syntax is:
43865 target> gdbserver --attach @var{comm} @var{pid}
43868 @var{pid} is the process ID of a currently running process. It isn't
43869 necessary to point @command{gdbserver} at a binary for the running process.
43872 To start @code{gdbserver} without supplying an initial command to run
43873 or process ID to attach, use this command line option.
43874 Then you can connect using @kbd{target extended-remote} and start
43875 the program you want to debug. The syntax is:
43878 target> gdbserver --multi @var{comm}
43882 Instruct @code{gdbserver} to display extra status information about the debugging
43884 This option is intended for @code{gdbserver} development and for bug reports to
43887 @item --remote-debug
43888 Instruct @code{gdbserver} to display remote protocol debug output.
43889 This option is intended for @code{gdbserver} development and for bug reports to
43893 Specify a wrapper to launch programs
43894 for debugging. The option should be followed by the name of the
43895 wrapper, then any command-line arguments to pass to the wrapper, then
43896 @kbd{--} indicating the end of the wrapper arguments.
43899 By default, @command{gdbserver} keeps the listening TCP port open, so that
43900 additional connections are possible. However, if you start @code{gdbserver}
43901 with the @option{--once} option, it will stop listening for any further
43902 connection attempts after connecting to the first @value{GDBN} session.
43904 @c --disable-packet is not documented for users.
43906 @c --disable-randomization and --no-disable-randomization are superseded by
43907 @c QDisableRandomization.
43912 @c man begin SEEALSO gdbserver
43914 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
43915 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
43916 documentation are properly installed at your site, the command
43922 should give you access to the complete manual.
43924 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
43925 Richard M. Stallman and Roland H. Pesch, July 1991.
43932 @c man title gcore Generate a core file of a running program
43935 @c man begin SYNOPSIS gcore
43936 gcore [-o @var{filename}] @var{pid}
43940 @c man begin DESCRIPTION gcore
43941 Generate a core dump of a running program with process ID @var{pid}.
43942 Produced file is equivalent to a kernel produced core file as if the process
43943 crashed (and if @kbd{ulimit -c} were used to set up an appropriate core dump
43944 limit). Unlike after a crash, after @command{gcore} the program remains
43945 running without any change.
43948 @c man begin OPTIONS gcore
43950 @item -o @var{filename}
43951 The optional argument
43952 @var{filename} specifies the file name where to put the core dump.
43953 If not specified, the file name defaults to @file{core.@var{pid}},
43954 where @var{pid} is the running program process ID.
43958 @c man begin SEEALSO gcore
43960 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
43961 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
43962 documentation are properly installed at your site, the command
43969 should give you access to the complete manual.
43971 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
43972 Richard M. Stallman and Roland H. Pesch, July 1991.
43979 @c man title gdbinit GDB initialization scripts
43982 @c man begin SYNOPSIS gdbinit
43983 @ifset SYSTEM_GDBINIT
43984 @value{SYSTEM_GDBINIT}
43993 @c man begin DESCRIPTION gdbinit
43994 These files contain @value{GDBN} commands to automatically execute during
43995 @value{GDBN} startup. The lines of contents are canned sequences of commands,
43998 the @value{GDBN} manual in node @code{Sequences}
43999 -- shell command @code{info -f gdb -n Sequences}.
44005 Please read more in
44007 the @value{GDBN} manual in node @code{Startup}
44008 -- shell command @code{info -f gdb -n Startup}.
44015 @ifset SYSTEM_GDBINIT
44016 @item @value{SYSTEM_GDBINIT}
44018 @ifclear SYSTEM_GDBINIT
44019 @item (not enabled with @code{--with-system-gdbinit} during compilation)
44021 System-wide initialization file. It is executed unless user specified
44022 @value{GDBN} option @code{-nx} or @code{-n}.
44025 the @value{GDBN} manual in node @code{System-wide configuration}
44026 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
44029 @ref{System-wide configuration}.
44033 User initialization file. It is executed unless user specified
44034 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
44037 Initialization file for current directory. It may need to be enabled with
44038 @value{GDBN} security command @code{set auto-load local-gdbinit}.
44041 the @value{GDBN} manual in node @code{Init File in the Current Directory}
44042 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
44045 @ref{Init File in the Current Directory}.
44050 @c man begin SEEALSO gdbinit
44052 gdb(1), @code{info -f gdb -n Startup}
44054 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
44055 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
44056 documentation are properly installed at your site, the command
44062 should give you access to the complete manual.
44064 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44065 Richard M. Stallman and Roland H. Pesch, July 1991.
44071 @node GNU Free Documentation License
44072 @appendix GNU Free Documentation License
44075 @node Concept Index
44076 @unnumbered Concept Index
44080 @node Command and Variable Index
44081 @unnumbered Command, Variable, and Function Index
44086 % I think something like @@colophon should be in texinfo. In the
44088 \long\def\colophon{\hbox to0pt{}\vfill
44089 \centerline{The body of this manual is set in}
44090 \centerline{\fontname\tenrm,}
44091 \centerline{with headings in {\bf\fontname\tenbf}}
44092 \centerline{and examples in {\tt\fontname\tentt}.}
44093 \centerline{{\it\fontname\tenit\/},}
44094 \centerline{{\bf\fontname\tenbf}, and}
44095 \centerline{{\sl\fontname\tensl\/}}
44096 \centerline{are used for emphasis.}\vfill}
44098 % Blame: doc@@cygnus.com, 1991.