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.
2015 @xref{Arguments, ,Your Program's Arguments}.
2017 @item The @emph{environment.}
2018 Your program normally inherits its environment from @value{GDBN}, but you can
2019 use the @value{GDBN} commands @code{set environment} and @code{unset
2020 environment} to change parts of the environment that affect
2021 your program. @xref{Environment, ,Your Program's Environment}.
2023 @item The @emph{working directory.}
2024 Your program inherits its working directory from @value{GDBN}. You can set
2025 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
2026 @xref{Working Directory, ,Your Program's Working Directory}.
2028 @item The @emph{standard input and output.}
2029 Your program normally uses the same device for standard input and
2030 standard output as @value{GDBN} is using. You can redirect input and output
2031 in the @code{run} command line, or you can use the @code{tty} command to
2032 set a different device for your program.
2033 @xref{Input/Output, ,Your Program's Input and Output}.
2036 @emph{Warning:} While input and output redirection work, you cannot use
2037 pipes to pass the output of the program you are debugging to another
2038 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2042 When you issue the @code{run} command, your program begins to execute
2043 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2044 of how to arrange for your program to stop. Once your program has
2045 stopped, you may call functions in your program, using the @code{print}
2046 or @code{call} commands. @xref{Data, ,Examining Data}.
2048 If the modification time of your symbol file has changed since the last
2049 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2050 table, and reads it again. When it does this, @value{GDBN} tries to retain
2051 your current breakpoints.
2056 @cindex run to main procedure
2057 The name of the main procedure can vary from language to language.
2058 With C or C@t{++}, the main procedure name is always @code{main}, but
2059 other languages such as Ada do not require a specific name for their
2060 main procedure. The debugger provides a convenient way to start the
2061 execution of the program and to stop at the beginning of the main
2062 procedure, depending on the language used.
2064 The @samp{start} command does the equivalent of setting a temporary
2065 breakpoint at the beginning of the main procedure and then invoking
2066 the @samp{run} command.
2068 @cindex elaboration phase
2069 Some programs contain an @dfn{elaboration} phase where some startup code is
2070 executed before the main procedure is called. This depends on the
2071 languages used to write your program. In C@t{++}, for instance,
2072 constructors for static and global objects are executed before
2073 @code{main} is called. It is therefore possible that the debugger stops
2074 before reaching the main procedure. However, the temporary breakpoint
2075 will remain to halt execution.
2077 Specify the arguments to give to your program as arguments to the
2078 @samp{start} command. These arguments will be given verbatim to the
2079 underlying @samp{run} command. Note that the same arguments will be
2080 reused if no argument is provided during subsequent calls to
2081 @samp{start} or @samp{run}.
2083 It is sometimes necessary to debug the program during elaboration. In
2084 these cases, using the @code{start} command would stop the execution of
2085 your program too late, as the program would have already completed the
2086 elaboration phase. Under these circumstances, insert breakpoints in your
2087 elaboration code before running your program.
2089 @kindex set exec-wrapper
2090 @item set exec-wrapper @var{wrapper}
2091 @itemx show exec-wrapper
2092 @itemx unset exec-wrapper
2093 When @samp{exec-wrapper} is set, the specified wrapper is used to
2094 launch programs for debugging. @value{GDBN} starts your program
2095 with a shell command of the form @kbd{exec @var{wrapper}
2096 @var{program}}. Quoting is added to @var{program} and its
2097 arguments, but not to @var{wrapper}, so you should add quotes if
2098 appropriate for your shell. The wrapper runs until it executes
2099 your program, and then @value{GDBN} takes control.
2101 You can use any program that eventually calls @code{execve} with
2102 its arguments as a wrapper. Several standard Unix utilities do
2103 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2104 with @code{exec "$@@"} will also work.
2106 For example, you can use @code{env} to pass an environment variable to
2107 the debugged program, without setting the variable in your shell's
2111 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2115 This command is available when debugging locally on most targets, excluding
2116 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2118 @kindex set disable-randomization
2119 @item set disable-randomization
2120 @itemx set disable-randomization on
2121 This option (enabled by default in @value{GDBN}) will turn off the native
2122 randomization of the virtual address space of the started program. This option
2123 is useful for multiple debugging sessions to make the execution better
2124 reproducible and memory addresses reusable across debugging sessions.
2126 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2127 On @sc{gnu}/Linux you can get the same behavior using
2130 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2133 @item set disable-randomization off
2134 Leave the behavior of the started executable unchanged. Some bugs rear their
2135 ugly heads only when the program is loaded at certain addresses. If your bug
2136 disappears when you run the program under @value{GDBN}, that might be because
2137 @value{GDBN} by default disables the address randomization on platforms, such
2138 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2139 disable-randomization off} to try to reproduce such elusive bugs.
2141 On targets where it is available, virtual address space randomization
2142 protects the programs against certain kinds of security attacks. In these
2143 cases the attacker needs to know the exact location of a concrete executable
2144 code. Randomizing its location makes it impossible to inject jumps misusing
2145 a code at its expected addresses.
2147 Prelinking shared libraries provides a startup performance advantage but it
2148 makes addresses in these libraries predictable for privileged processes by
2149 having just unprivileged access at the target system. Reading the shared
2150 library binary gives enough information for assembling the malicious code
2151 misusing it. Still even a prelinked shared library can get loaded at a new
2152 random address just requiring the regular relocation process during the
2153 startup. Shared libraries not already prelinked are always loaded at
2154 a randomly chosen address.
2156 Position independent executables (PIE) contain position independent code
2157 similar to the shared libraries and therefore such executables get loaded at
2158 a randomly chosen address upon startup. PIE executables always load even
2159 already prelinked shared libraries at a random address. You can build such
2160 executable using @command{gcc -fPIE -pie}.
2162 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2163 (as long as the randomization is enabled).
2165 @item show disable-randomization
2166 Show the current setting of the explicit disable of the native randomization of
2167 the virtual address space of the started program.
2172 @section Your Program's Arguments
2174 @cindex arguments (to your program)
2175 The arguments to your program can be specified by the arguments of the
2177 They are passed to a shell, which expands wildcard characters and
2178 performs redirection of I/O, and thence to your program. Your
2179 @code{SHELL} environment variable (if it exists) specifies what shell
2180 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2181 the default shell (@file{/bin/sh} on Unix).
2183 On non-Unix systems, the program is usually invoked directly by
2184 @value{GDBN}, which emulates I/O redirection via the appropriate system
2185 calls, and the wildcard characters are expanded by the startup code of
2186 the program, not by the shell.
2188 @code{run} with no arguments uses the same arguments used by the previous
2189 @code{run}, or those set by the @code{set args} command.
2194 Specify the arguments to be used the next time your program is run. If
2195 @code{set args} has no arguments, @code{run} executes your program
2196 with no arguments. Once you have run your program with arguments,
2197 using @code{set args} before the next @code{run} is the only way to run
2198 it again without arguments.
2202 Show the arguments to give your program when it is started.
2206 @section Your Program's Environment
2208 @cindex environment (of your program)
2209 The @dfn{environment} consists of a set of environment variables and
2210 their values. Environment variables conventionally record such things as
2211 your user name, your home directory, your terminal type, and your search
2212 path for programs to run. Usually you set up environment variables with
2213 the shell and they are inherited by all the other programs you run. When
2214 debugging, it can be useful to try running your program with a modified
2215 environment without having to start @value{GDBN} over again.
2219 @item path @var{directory}
2220 Add @var{directory} to the front of the @code{PATH} environment variable
2221 (the search path for executables) that will be passed to your program.
2222 The value of @code{PATH} used by @value{GDBN} does not change.
2223 You may specify several directory names, separated by whitespace or by a
2224 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2225 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2226 is moved to the front, so it is searched sooner.
2228 You can use the string @samp{$cwd} to refer to whatever is the current
2229 working directory at the time @value{GDBN} searches the path. If you
2230 use @samp{.} instead, it refers to the directory where you executed the
2231 @code{path} command. @value{GDBN} replaces @samp{.} in the
2232 @var{directory} argument (with the current path) before adding
2233 @var{directory} to the search path.
2234 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2235 @c document that, since repeating it would be a no-op.
2239 Display the list of search paths for executables (the @code{PATH}
2240 environment variable).
2242 @kindex show environment
2243 @item show environment @r{[}@var{varname}@r{]}
2244 Print the value of environment variable @var{varname} to be given to
2245 your program when it starts. If you do not supply @var{varname},
2246 print the names and values of all environment variables to be given to
2247 your program. You can abbreviate @code{environment} as @code{env}.
2249 @kindex set environment
2250 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2251 Set environment variable @var{varname} to @var{value}. The value
2252 changes for your program only, not for @value{GDBN} itself. @var{value} may
2253 be any string; the values of environment variables are just strings, and
2254 any interpretation is supplied by your program itself. The @var{value}
2255 parameter is optional; if it is eliminated, the variable is set to a
2257 @c "any string" here does not include leading, trailing
2258 @c blanks. Gnu asks: does anyone care?
2260 For example, this command:
2267 tells the debugged program, when subsequently run, that its user is named
2268 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2269 are not actually required.)
2271 @kindex unset environment
2272 @item unset environment @var{varname}
2273 Remove variable @var{varname} from the environment to be passed to your
2274 program. This is different from @samp{set env @var{varname} =};
2275 @code{unset environment} removes the variable from the environment,
2276 rather than assigning it an empty value.
2279 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2281 by your @code{SHELL} environment variable if it exists (or
2282 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2283 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2284 @file{.bashrc} for BASH---any variables you set in that file affect
2285 your program. You may wish to move setting of environment variables to
2286 files that are only run when you sign on, such as @file{.login} or
2289 @node Working Directory
2290 @section Your Program's Working Directory
2292 @cindex working directory (of your program)
2293 Each time you start your program with @code{run}, it inherits its
2294 working directory from the current working directory of @value{GDBN}.
2295 The @value{GDBN} working directory is initially whatever it inherited
2296 from its parent process (typically the shell), but you can specify a new
2297 working directory in @value{GDBN} with the @code{cd} command.
2299 The @value{GDBN} working directory also serves as a default for the commands
2300 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2305 @cindex change working directory
2306 @item cd @r{[}@var{directory}@r{]}
2307 Set the @value{GDBN} working directory to @var{directory}. If not
2308 given, @var{directory} uses @file{'~'}.
2312 Print the @value{GDBN} working directory.
2315 It is generally impossible to find the current working directory of
2316 the process being debugged (since a program can change its directory
2317 during its run). If you work on a system where @value{GDBN} is
2318 configured with the @file{/proc} support, you can use the @code{info
2319 proc} command (@pxref{SVR4 Process Information}) to find out the
2320 current working directory of the debuggee.
2323 @section Your Program's Input and Output
2328 By default, the program you run under @value{GDBN} does input and output to
2329 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2330 to its own terminal modes to interact with you, but it records the terminal
2331 modes your program was using and switches back to them when you continue
2332 running your program.
2335 @kindex info terminal
2337 Displays information recorded by @value{GDBN} about the terminal modes your
2341 You can redirect your program's input and/or output using shell
2342 redirection with the @code{run} command. For example,
2349 starts your program, diverting its output to the file @file{outfile}.
2352 @cindex controlling terminal
2353 Another way to specify where your program should do input and output is
2354 with the @code{tty} command. This command accepts a file name as
2355 argument, and causes this file to be the default for future @code{run}
2356 commands. It also resets the controlling terminal for the child
2357 process, for future @code{run} commands. For example,
2364 directs that processes started with subsequent @code{run} commands
2365 default to do input and output on the terminal @file{/dev/ttyb} and have
2366 that as their controlling terminal.
2368 An explicit redirection in @code{run} overrides the @code{tty} command's
2369 effect on the input/output device, but not its effect on the controlling
2372 When you use the @code{tty} command or redirect input in the @code{run}
2373 command, only the input @emph{for your program} is affected. The input
2374 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2375 for @code{set inferior-tty}.
2377 @cindex inferior tty
2378 @cindex set inferior controlling terminal
2379 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2380 display the name of the terminal that will be used for future runs of your
2384 @item set inferior-tty /dev/ttyb
2385 @kindex set inferior-tty
2386 Set the tty for the program being debugged to /dev/ttyb.
2388 @item show inferior-tty
2389 @kindex show inferior-tty
2390 Show the current tty for the program being debugged.
2394 @section Debugging an Already-running Process
2399 @item attach @var{process-id}
2400 This command attaches to a running process---one that was started
2401 outside @value{GDBN}. (@code{info files} shows your active
2402 targets.) The command takes as argument a process ID. The usual way to
2403 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2404 or with the @samp{jobs -l} shell command.
2406 @code{attach} does not repeat if you press @key{RET} a second time after
2407 executing the command.
2410 To use @code{attach}, your program must be running in an environment
2411 which supports processes; for example, @code{attach} does not work for
2412 programs on bare-board targets that lack an operating system. You must
2413 also have permission to send the process a signal.
2415 When you use @code{attach}, the debugger finds the program running in
2416 the process first by looking in the current working directory, then (if
2417 the program is not found) by using the source file search path
2418 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2419 the @code{file} command to load the program. @xref{Files, ,Commands to
2422 The first thing @value{GDBN} does after arranging to debug the specified
2423 process is to stop it. You can examine and modify an attached process
2424 with all the @value{GDBN} commands that are ordinarily available when
2425 you start processes with @code{run}. You can insert breakpoints; you
2426 can step and continue; you can modify storage. If you would rather the
2427 process continue running, you may use the @code{continue} command after
2428 attaching @value{GDBN} to the process.
2433 When you have finished debugging the attached process, you can use the
2434 @code{detach} command to release it from @value{GDBN} control. Detaching
2435 the process continues its execution. After the @code{detach} command,
2436 that process and @value{GDBN} become completely independent once more, and you
2437 are ready to @code{attach} another process or start one with @code{run}.
2438 @code{detach} does not repeat if you press @key{RET} again after
2439 executing the command.
2442 If you exit @value{GDBN} while you have an attached process, you detach
2443 that process. If you use the @code{run} command, you kill that process.
2444 By default, @value{GDBN} asks for confirmation if you try to do either of these
2445 things; you can control whether or not you need to confirm by using the
2446 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2450 @section Killing the Child Process
2455 Kill the child process in which your program is running under @value{GDBN}.
2458 This command is useful if you wish to debug a core dump instead of a
2459 running process. @value{GDBN} ignores any core dump file while your program
2462 On some operating systems, a program cannot be executed outside @value{GDBN}
2463 while you have breakpoints set on it inside @value{GDBN}. You can use the
2464 @code{kill} command in this situation to permit running your program
2465 outside the debugger.
2467 The @code{kill} command is also useful if you wish to recompile and
2468 relink your program, since on many systems it is impossible to modify an
2469 executable file while it is running in a process. In this case, when you
2470 next type @code{run}, @value{GDBN} notices that the file has changed, and
2471 reads the symbol table again (while trying to preserve your current
2472 breakpoint settings).
2474 @node Inferiors and Programs
2475 @section Debugging Multiple Inferiors and Programs
2477 @value{GDBN} lets you run and debug multiple programs in a single
2478 session. In addition, @value{GDBN} on some systems may let you run
2479 several programs simultaneously (otherwise you have to exit from one
2480 before starting another). In the most general case, you can have
2481 multiple threads of execution in each of multiple processes, launched
2482 from multiple executables.
2485 @value{GDBN} represents the state of each program execution with an
2486 object called an @dfn{inferior}. An inferior typically corresponds to
2487 a process, but is more general and applies also to targets that do not
2488 have processes. Inferiors may be created before a process runs, and
2489 may be retained after a process exits. Inferiors have unique
2490 identifiers that are different from process ids. Usually each
2491 inferior will also have its own distinct address space, although some
2492 embedded targets may have several inferiors running in different parts
2493 of a single address space. Each inferior may in turn have multiple
2494 threads running in it.
2496 To find out what inferiors exist at any moment, use @w{@code{info
2500 @kindex info inferiors
2501 @item info inferiors
2502 Print a list of all inferiors currently being managed by @value{GDBN}.
2504 @value{GDBN} displays for each inferior (in this order):
2508 the inferior number assigned by @value{GDBN}
2511 the target system's inferior identifier
2514 the name of the executable the inferior is running.
2519 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2520 indicates the current inferior.
2524 @c end table here to get a little more width for example
2527 (@value{GDBP}) info inferiors
2528 Num Description Executable
2529 2 process 2307 hello
2530 * 1 process 3401 goodbye
2533 To switch focus between inferiors, use the @code{inferior} command:
2536 @kindex inferior @var{infno}
2537 @item inferior @var{infno}
2538 Make inferior number @var{infno} the current inferior. The argument
2539 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2540 in the first field of the @samp{info inferiors} display.
2544 You can get multiple executables into a debugging session via the
2545 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2546 systems @value{GDBN} can add inferiors to the debug session
2547 automatically by following calls to @code{fork} and @code{exec}. To
2548 remove inferiors from the debugging session use the
2549 @w{@code{remove-inferiors}} command.
2552 @kindex add-inferior
2553 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2554 Adds @var{n} inferiors to be run using @var{executable} as the
2555 executable. @var{n} defaults to 1. If no executable is specified,
2556 the inferiors begins empty, with no program. You can still assign or
2557 change the program assigned to the inferior at any time by using the
2558 @code{file} command with the executable name as its argument.
2560 @kindex clone-inferior
2561 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2562 Adds @var{n} inferiors ready to execute the same program as inferior
2563 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2564 number of the current inferior. This is a convenient command when you
2565 want to run another instance of the inferior you are debugging.
2568 (@value{GDBP}) info inferiors
2569 Num Description Executable
2570 * 1 process 29964 helloworld
2571 (@value{GDBP}) clone-inferior
2574 (@value{GDBP}) info inferiors
2575 Num Description Executable
2577 * 1 process 29964 helloworld
2580 You can now simply switch focus to inferior 2 and run it.
2582 @kindex remove-inferiors
2583 @item remove-inferiors @var{infno}@dots{}
2584 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2585 possible to remove an inferior that is running with this command. For
2586 those, use the @code{kill} or @code{detach} command first.
2590 To quit debugging one of the running inferiors that is not the current
2591 inferior, you can either detach from it by using the @w{@code{detach
2592 inferior}} command (allowing it to run independently), or kill it
2593 using the @w{@code{kill inferiors}} command:
2596 @kindex detach inferiors @var{infno}@dots{}
2597 @item detach inferior @var{infno}@dots{}
2598 Detach from the inferior or inferiors identified by @value{GDBN}
2599 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2600 still stays on the list of inferiors shown by @code{info inferiors},
2601 but its Description will show @samp{<null>}.
2603 @kindex kill inferiors @var{infno}@dots{}
2604 @item kill inferiors @var{infno}@dots{}
2605 Kill the inferior or inferiors identified by @value{GDBN} inferior
2606 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2607 stays on the list of inferiors shown by @code{info inferiors}, but its
2608 Description will show @samp{<null>}.
2611 After the successful completion of a command such as @code{detach},
2612 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2613 a normal process exit, the inferior is still valid and listed with
2614 @code{info inferiors}, ready to be restarted.
2617 To be notified when inferiors are started or exit under @value{GDBN}'s
2618 control use @w{@code{set print inferior-events}}:
2621 @kindex set print inferior-events
2622 @cindex print messages on inferior start and exit
2623 @item set print inferior-events
2624 @itemx set print inferior-events on
2625 @itemx set print inferior-events off
2626 The @code{set print inferior-events} command allows you to enable or
2627 disable printing of messages when @value{GDBN} notices that new
2628 inferiors have started or that inferiors have exited or have been
2629 detached. By default, these messages will not be printed.
2631 @kindex show print inferior-events
2632 @item show print inferior-events
2633 Show whether messages will be printed when @value{GDBN} detects that
2634 inferiors have started, exited or have been detached.
2637 Many commands will work the same with multiple programs as with a
2638 single program: e.g., @code{print myglobal} will simply display the
2639 value of @code{myglobal} in the current inferior.
2642 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2643 get more info about the relationship of inferiors, programs, address
2644 spaces in a debug session. You can do that with the @w{@code{maint
2645 info program-spaces}} command.
2648 @kindex maint info program-spaces
2649 @item maint info program-spaces
2650 Print a list of all program spaces currently being managed by
2653 @value{GDBN} displays for each program space (in this order):
2657 the program space number assigned by @value{GDBN}
2660 the name of the executable loaded into the program space, with e.g.,
2661 the @code{file} command.
2666 An asterisk @samp{*} preceding the @value{GDBN} program space number
2667 indicates the current program space.
2669 In addition, below each program space line, @value{GDBN} prints extra
2670 information that isn't suitable to display in tabular form. For
2671 example, the list of inferiors bound to the program space.
2674 (@value{GDBP}) maint info program-spaces
2677 Bound inferiors: ID 1 (process 21561)
2681 Here we can see that no inferior is running the program @code{hello},
2682 while @code{process 21561} is running the program @code{goodbye}. On
2683 some targets, it is possible that multiple inferiors are bound to the
2684 same program space. The most common example is that of debugging both
2685 the parent and child processes of a @code{vfork} call. For example,
2688 (@value{GDBP}) maint info program-spaces
2691 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2694 Here, both inferior 2 and inferior 1 are running in the same program
2695 space as a result of inferior 1 having executed a @code{vfork} call.
2699 @section Debugging Programs with Multiple Threads
2701 @cindex threads of execution
2702 @cindex multiple threads
2703 @cindex switching threads
2704 In some operating systems, such as HP-UX and Solaris, a single program
2705 may have more than one @dfn{thread} of execution. The precise semantics
2706 of threads differ from one operating system to another, but in general
2707 the threads of a single program are akin to multiple processes---except
2708 that they share one address space (that is, they can all examine and
2709 modify the same variables). On the other hand, each thread has its own
2710 registers and execution stack, and perhaps private memory.
2712 @value{GDBN} provides these facilities for debugging multi-thread
2716 @item automatic notification of new threads
2717 @item @samp{thread @var{threadno}}, a command to switch among threads
2718 @item @samp{info threads}, a command to inquire about existing threads
2719 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2720 a command to apply a command to a list of threads
2721 @item thread-specific breakpoints
2722 @item @samp{set print thread-events}, which controls printing of
2723 messages on thread start and exit.
2724 @item @samp{set libthread-db-search-path @var{path}}, which lets
2725 the user specify which @code{libthread_db} to use if the default choice
2726 isn't compatible with the program.
2730 @emph{Warning:} These facilities are not yet available on every
2731 @value{GDBN} configuration where the operating system supports threads.
2732 If your @value{GDBN} does not support threads, these commands have no
2733 effect. For example, a system without thread support shows no output
2734 from @samp{info threads}, and always rejects the @code{thread} command,
2738 (@value{GDBP}) info threads
2739 (@value{GDBP}) thread 1
2740 Thread ID 1 not known. Use the "info threads" command to
2741 see the IDs of currently known threads.
2743 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2744 @c doesn't support threads"?
2747 @cindex focus of debugging
2748 @cindex current thread
2749 The @value{GDBN} thread debugging facility allows you to observe all
2750 threads while your program runs---but whenever @value{GDBN} takes
2751 control, one thread in particular is always the focus of debugging.
2752 This thread is called the @dfn{current thread}. Debugging commands show
2753 program information from the perspective of the current thread.
2755 @cindex @code{New} @var{systag} message
2756 @cindex thread identifier (system)
2757 @c FIXME-implementors!! It would be more helpful if the [New...] message
2758 @c included GDB's numeric thread handle, so you could just go to that
2759 @c thread without first checking `info threads'.
2760 Whenever @value{GDBN} detects a new thread in your program, it displays
2761 the target system's identification for the thread with a message in the
2762 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2763 whose form varies depending on the particular system. For example, on
2764 @sc{gnu}/Linux, you might see
2767 [New Thread 0x41e02940 (LWP 25582)]
2771 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2772 the @var{systag} is simply something like @samp{process 368}, with no
2775 @c FIXME!! (1) Does the [New...] message appear even for the very first
2776 @c thread of a program, or does it only appear for the
2777 @c second---i.e.@: when it becomes obvious we have a multithread
2779 @c (2) *Is* there necessarily a first thread always? Or do some
2780 @c multithread systems permit starting a program with multiple
2781 @c threads ab initio?
2783 @cindex thread number
2784 @cindex thread identifier (GDB)
2785 For debugging purposes, @value{GDBN} associates its own thread
2786 number---always a single integer---with each thread in your program.
2789 @kindex info threads
2790 @item info threads @r{[}@var{id}@dots{}@r{]}
2791 Display a summary of all threads currently in your program. Optional
2792 argument @var{id}@dots{} is one or more thread ids separated by spaces, and
2793 means to print information only about the specified thread or threads.
2794 @value{GDBN} displays for each thread (in this order):
2798 the thread number assigned by @value{GDBN}
2801 the target system's thread identifier (@var{systag})
2804 the thread's name, if one is known. A thread can either be named by
2805 the user (see @code{thread name}, below), or, in some cases, by the
2809 the current stack frame summary for that thread
2813 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2814 indicates the current thread.
2818 @c end table here to get a little more width for example
2821 (@value{GDBP}) info threads
2823 3 process 35 thread 27 0x34e5 in sigpause ()
2824 2 process 35 thread 23 0x34e5 in sigpause ()
2825 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2829 On Solaris, you can display more information about user threads with a
2830 Solaris-specific command:
2833 @item maint info sol-threads
2834 @kindex maint info sol-threads
2835 @cindex thread info (Solaris)
2836 Display info on Solaris user threads.
2840 @kindex thread @var{threadno}
2841 @item thread @var{threadno}
2842 Make thread number @var{threadno} the current thread. The command
2843 argument @var{threadno} is the internal @value{GDBN} thread number, as
2844 shown in the first field of the @samp{info threads} display.
2845 @value{GDBN} responds by displaying the system identifier of the thread
2846 you selected, and its current stack frame summary:
2849 (@value{GDBP}) thread 2
2850 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
2851 #0 some_function (ignore=0x0) at example.c:8
2852 8 printf ("hello\n");
2856 As with the @samp{[New @dots{}]} message, the form of the text after
2857 @samp{Switching to} depends on your system's conventions for identifying
2860 @vindex $_thread@r{, convenience variable}
2861 The debugger convenience variable @samp{$_thread} contains the number
2862 of the current thread. You may find this useful in writing breakpoint
2863 conditional expressions, command scripts, and so forth. See
2864 @xref{Convenience Vars,, Convenience Variables}, for general
2865 information on convenience variables.
2867 @kindex thread apply
2868 @cindex apply command to several threads
2869 @item thread apply [@var{threadno} | all] @var{command}
2870 The @code{thread apply} command allows you to apply the named
2871 @var{command} to one or more threads. Specify the numbers of the
2872 threads that you want affected with the command argument
2873 @var{threadno}. It can be a single thread number, one of the numbers
2874 shown in the first field of the @samp{info threads} display; or it
2875 could be a range of thread numbers, as in @code{2-4}. To apply a
2876 command to all threads, type @kbd{thread apply all @var{command}}.
2879 @cindex name a thread
2880 @item thread name [@var{name}]
2881 This command assigns a name to the current thread. If no argument is
2882 given, any existing user-specified name is removed. The thread name
2883 appears in the @samp{info threads} display.
2885 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
2886 determine the name of the thread as given by the OS. On these
2887 systems, a name specified with @samp{thread name} will override the
2888 system-give name, and removing the user-specified name will cause
2889 @value{GDBN} to once again display the system-specified name.
2892 @cindex search for a thread
2893 @item thread find [@var{regexp}]
2894 Search for and display thread ids whose name or @var{systag}
2895 matches the supplied regular expression.
2897 As well as being the complement to the @samp{thread name} command,
2898 this command also allows you to identify a thread by its target
2899 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
2903 (@value{GDBN}) thread find 26688
2904 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
2905 (@value{GDBN}) info thread 4
2907 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
2910 @kindex set print thread-events
2911 @cindex print messages on thread start and exit
2912 @item set print thread-events
2913 @itemx set print thread-events on
2914 @itemx set print thread-events off
2915 The @code{set print thread-events} command allows you to enable or
2916 disable printing of messages when @value{GDBN} notices that new threads have
2917 started or that threads have exited. By default, these messages will
2918 be printed if detection of these events is supported by the target.
2919 Note that these messages cannot be disabled on all targets.
2921 @kindex show print thread-events
2922 @item show print thread-events
2923 Show whether messages will be printed when @value{GDBN} detects that threads
2924 have started and exited.
2927 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2928 more information about how @value{GDBN} behaves when you stop and start
2929 programs with multiple threads.
2931 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2932 watchpoints in programs with multiple threads.
2934 @anchor{set libthread-db-search-path}
2936 @kindex set libthread-db-search-path
2937 @cindex search path for @code{libthread_db}
2938 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2939 If this variable is set, @var{path} is a colon-separated list of
2940 directories @value{GDBN} will use to search for @code{libthread_db}.
2941 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2942 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
2943 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
2946 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2947 @code{libthread_db} library to obtain information about threads in the
2948 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2949 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
2950 specific thread debugging library loading is enabled
2951 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
2953 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
2954 refers to the default system directories that are
2955 normally searched for loading shared libraries. The @samp{$sdir} entry
2956 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
2957 (@pxref{libthread_db.so.1 file}).
2959 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
2960 refers to the directory from which @code{libpthread}
2961 was loaded in the inferior process.
2963 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2964 @value{GDBN} attempts to initialize it with the current inferior process.
2965 If this initialization fails (which could happen because of a version
2966 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2967 will unload @code{libthread_db}, and continue with the next directory.
2968 If none of @code{libthread_db} libraries initialize successfully,
2969 @value{GDBN} will issue a warning and thread debugging will be disabled.
2971 Setting @code{libthread-db-search-path} is currently implemented
2972 only on some platforms.
2974 @kindex show libthread-db-search-path
2975 @item show libthread-db-search-path
2976 Display current libthread_db search path.
2978 @kindex set debug libthread-db
2979 @kindex show debug libthread-db
2980 @cindex debugging @code{libthread_db}
2981 @item set debug libthread-db
2982 @itemx show debug libthread-db
2983 Turns on or off display of @code{libthread_db}-related events.
2984 Use @code{1} to enable, @code{0} to disable.
2988 @section Debugging Forks
2990 @cindex fork, debugging programs which call
2991 @cindex multiple processes
2992 @cindex processes, multiple
2993 On most systems, @value{GDBN} has no special support for debugging
2994 programs which create additional processes using the @code{fork}
2995 function. When a program forks, @value{GDBN} will continue to debug the
2996 parent process and the child process will run unimpeded. If you have
2997 set a breakpoint in any code which the child then executes, the child
2998 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2999 will cause it to terminate.
3001 However, if you want to debug the child process there is a workaround
3002 which isn't too painful. Put a call to @code{sleep} in the code which
3003 the child process executes after the fork. It may be useful to sleep
3004 only if a certain environment variable is set, or a certain file exists,
3005 so that the delay need not occur when you don't want to run @value{GDBN}
3006 on the child. While the child is sleeping, use the @code{ps} program to
3007 get its process ID. Then tell @value{GDBN} (a new invocation of
3008 @value{GDBN} if you are also debugging the parent process) to attach to
3009 the child process (@pxref{Attach}). From that point on you can debug
3010 the child process just like any other process which you attached to.
3012 On some systems, @value{GDBN} provides support for debugging programs that
3013 create additional processes using the @code{fork} or @code{vfork} functions.
3014 Currently, the only platforms with this feature are HP-UX (11.x and later
3015 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
3017 By default, when a program forks, @value{GDBN} will continue to debug
3018 the parent process and the child process will run unimpeded.
3020 If you want to follow the child process instead of the parent process,
3021 use the command @w{@code{set follow-fork-mode}}.
3024 @kindex set follow-fork-mode
3025 @item set follow-fork-mode @var{mode}
3026 Set the debugger response to a program call of @code{fork} or
3027 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3028 process. The @var{mode} argument can be:
3032 The original process is debugged after a fork. The child process runs
3033 unimpeded. This is the default.
3036 The new process is debugged after a fork. The parent process runs
3041 @kindex show follow-fork-mode
3042 @item show follow-fork-mode
3043 Display the current debugger response to a @code{fork} or @code{vfork} call.
3046 @cindex debugging multiple processes
3047 On Linux, if you want to debug both the parent and child processes, use the
3048 command @w{@code{set detach-on-fork}}.
3051 @kindex set detach-on-fork
3052 @item set detach-on-fork @var{mode}
3053 Tells gdb whether to detach one of the processes after a fork, or
3054 retain debugger control over them both.
3058 The child process (or parent process, depending on the value of
3059 @code{follow-fork-mode}) will be detached and allowed to run
3060 independently. This is the default.
3063 Both processes will be held under the control of @value{GDBN}.
3064 One process (child or parent, depending on the value of
3065 @code{follow-fork-mode}) is debugged as usual, while the other
3070 @kindex show detach-on-fork
3071 @item show detach-on-fork
3072 Show whether detach-on-fork mode is on/off.
3075 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3076 will retain control of all forked processes (including nested forks).
3077 You can list the forked processes under the control of @value{GDBN} by
3078 using the @w{@code{info inferiors}} command, and switch from one fork
3079 to another by using the @code{inferior} command (@pxref{Inferiors and
3080 Programs, ,Debugging Multiple Inferiors and Programs}).
3082 To quit debugging one of the forked processes, you can either detach
3083 from it by using the @w{@code{detach inferiors}} command (allowing it
3084 to run independently), or kill it using the @w{@code{kill inferiors}}
3085 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3088 If you ask to debug a child process and a @code{vfork} is followed by an
3089 @code{exec}, @value{GDBN} executes the new target up to the first
3090 breakpoint in the new target. If you have a breakpoint set on
3091 @code{main} in your original program, the breakpoint will also be set on
3092 the child process's @code{main}.
3094 On some systems, when a child process is spawned by @code{vfork}, you
3095 cannot debug the child or parent until an @code{exec} call completes.
3097 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3098 call executes, the new target restarts. To restart the parent
3099 process, use the @code{file} command with the parent executable name
3100 as its argument. By default, after an @code{exec} call executes,
3101 @value{GDBN} discards the symbols of the previous executable image.
3102 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3106 @kindex set follow-exec-mode
3107 @item set follow-exec-mode @var{mode}
3109 Set debugger response to a program call of @code{exec}. An
3110 @code{exec} call replaces the program image of a process.
3112 @code{follow-exec-mode} can be:
3116 @value{GDBN} creates a new inferior and rebinds the process to this
3117 new inferior. The program the process was running before the
3118 @code{exec} call can be restarted afterwards by restarting the
3124 (@value{GDBP}) info inferiors
3126 Id Description Executable
3129 process 12020 is executing new program: prog2
3130 Program exited normally.
3131 (@value{GDBP}) info inferiors
3132 Id Description Executable
3138 @value{GDBN} keeps the process bound to the same inferior. The new
3139 executable image replaces the previous executable loaded in the
3140 inferior. Restarting the inferior after the @code{exec} call, with
3141 e.g., the @code{run} command, restarts the executable the process was
3142 running after the @code{exec} call. This is the default mode.
3147 (@value{GDBP}) info inferiors
3148 Id Description Executable
3151 process 12020 is executing new program: prog2
3152 Program exited normally.
3153 (@value{GDBP}) info inferiors
3154 Id Description Executable
3161 You can use the @code{catch} command to make @value{GDBN} stop whenever
3162 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3163 Catchpoints, ,Setting Catchpoints}.
3165 @node Checkpoint/Restart
3166 @section Setting a @emph{Bookmark} to Return to Later
3171 @cindex snapshot of a process
3172 @cindex rewind program state
3174 On certain operating systems@footnote{Currently, only
3175 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3176 program's state, called a @dfn{checkpoint}, and come back to it
3179 Returning to a checkpoint effectively undoes everything that has
3180 happened in the program since the @code{checkpoint} was saved. This
3181 includes changes in memory, registers, and even (within some limits)
3182 system state. Effectively, it is like going back in time to the
3183 moment when the checkpoint was saved.
3185 Thus, if you're stepping thru a program and you think you're
3186 getting close to the point where things go wrong, you can save
3187 a checkpoint. Then, if you accidentally go too far and miss
3188 the critical statement, instead of having to restart your program
3189 from the beginning, you can just go back to the checkpoint and
3190 start again from there.
3192 This can be especially useful if it takes a lot of time or
3193 steps to reach the point where you think the bug occurs.
3195 To use the @code{checkpoint}/@code{restart} method of debugging:
3200 Save a snapshot of the debugged program's current execution state.
3201 The @code{checkpoint} command takes no arguments, but each checkpoint
3202 is assigned a small integer id, similar to a breakpoint id.
3204 @kindex info checkpoints
3205 @item info checkpoints
3206 List the checkpoints that have been saved in the current debugging
3207 session. For each checkpoint, the following information will be
3214 @item Source line, or label
3217 @kindex restart @var{checkpoint-id}
3218 @item restart @var{checkpoint-id}
3219 Restore the program state that was saved as checkpoint number
3220 @var{checkpoint-id}. All program variables, registers, stack frames
3221 etc.@: will be returned to the values that they had when the checkpoint
3222 was saved. In essence, gdb will ``wind back the clock'' to the point
3223 in time when the checkpoint was saved.
3225 Note that breakpoints, @value{GDBN} variables, command history etc.
3226 are not affected by restoring a checkpoint. In general, a checkpoint
3227 only restores things that reside in the program being debugged, not in
3230 @kindex delete checkpoint @var{checkpoint-id}
3231 @item delete checkpoint @var{checkpoint-id}
3232 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3236 Returning to a previously saved checkpoint will restore the user state
3237 of the program being debugged, plus a significant subset of the system
3238 (OS) state, including file pointers. It won't ``un-write'' data from
3239 a file, but it will rewind the file pointer to the previous location,
3240 so that the previously written data can be overwritten. For files
3241 opened in read mode, the pointer will also be restored so that the
3242 previously read data can be read again.
3244 Of course, characters that have been sent to a printer (or other
3245 external device) cannot be ``snatched back'', and characters received
3246 from eg.@: a serial device can be removed from internal program buffers,
3247 but they cannot be ``pushed back'' into the serial pipeline, ready to
3248 be received again. Similarly, the actual contents of files that have
3249 been changed cannot be restored (at this time).
3251 However, within those constraints, you actually can ``rewind'' your
3252 program to a previously saved point in time, and begin debugging it
3253 again --- and you can change the course of events so as to debug a
3254 different execution path this time.
3256 @cindex checkpoints and process id
3257 Finally, there is one bit of internal program state that will be
3258 different when you return to a checkpoint --- the program's process
3259 id. Each checkpoint will have a unique process id (or @var{pid}),
3260 and each will be different from the program's original @var{pid}.
3261 If your program has saved a local copy of its process id, this could
3262 potentially pose a problem.
3264 @subsection A Non-obvious Benefit of Using Checkpoints
3266 On some systems such as @sc{gnu}/Linux, address space randomization
3267 is performed on new processes for security reasons. This makes it
3268 difficult or impossible to set a breakpoint, or watchpoint, on an
3269 absolute address if you have to restart the program, since the
3270 absolute location of a symbol will change from one execution to the
3273 A checkpoint, however, is an @emph{identical} copy of a process.
3274 Therefore if you create a checkpoint at (eg.@:) the start of main,
3275 and simply return to that checkpoint instead of restarting the
3276 process, you can avoid the effects of address randomization and
3277 your symbols will all stay in the same place.
3280 @chapter Stopping and Continuing
3282 The principal purposes of using a debugger are so that you can stop your
3283 program before it terminates; or so that, if your program runs into
3284 trouble, you can investigate and find out why.
3286 Inside @value{GDBN}, your program may stop for any of several reasons,
3287 such as a signal, a breakpoint, or reaching a new line after a
3288 @value{GDBN} command such as @code{step}. You may then examine and
3289 change variables, set new breakpoints or remove old ones, and then
3290 continue execution. Usually, the messages shown by @value{GDBN} provide
3291 ample explanation of the status of your program---but you can also
3292 explicitly request this information at any time.
3295 @kindex info program
3297 Display information about the status of your program: whether it is
3298 running or not, what process it is, and why it stopped.
3302 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3303 * Continuing and Stepping:: Resuming execution
3304 * Skipping Over Functions and Files::
3305 Skipping over functions and files
3307 * Thread Stops:: Stopping and starting multi-thread programs
3311 @section Breakpoints, Watchpoints, and Catchpoints
3314 A @dfn{breakpoint} makes your program stop whenever a certain point in
3315 the program is reached. For each breakpoint, you can add conditions to
3316 control in finer detail whether your program stops. You can set
3317 breakpoints with the @code{break} command and its variants (@pxref{Set
3318 Breaks, ,Setting Breakpoints}), to specify the place where your program
3319 should stop by line number, function name or exact address in the
3322 On some systems, you can set breakpoints in shared libraries before
3323 the executable is run. There is a minor limitation on HP-UX systems:
3324 you must wait until the executable is run in order to set breakpoints
3325 in shared library routines that are not called directly by the program
3326 (for example, routines that are arguments in a @code{pthread_create}
3330 @cindex data breakpoints
3331 @cindex memory tracing
3332 @cindex breakpoint on memory address
3333 @cindex breakpoint on variable modification
3334 A @dfn{watchpoint} is a special breakpoint that stops your program
3335 when the value of an expression changes. The expression may be a value
3336 of a variable, or it could involve values of one or more variables
3337 combined by operators, such as @samp{a + b}. This is sometimes called
3338 @dfn{data breakpoints}. You must use a different command to set
3339 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3340 from that, you can manage a watchpoint like any other breakpoint: you
3341 enable, disable, and delete both breakpoints and watchpoints using the
3344 You can arrange to have values from your program displayed automatically
3345 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3349 @cindex breakpoint on events
3350 A @dfn{catchpoint} is another special breakpoint that stops your program
3351 when a certain kind of event occurs, such as the throwing of a C@t{++}
3352 exception or the loading of a library. As with watchpoints, you use a
3353 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3354 Catchpoints}), but aside from that, you can manage a catchpoint like any
3355 other breakpoint. (To stop when your program receives a signal, use the
3356 @code{handle} command; see @ref{Signals, ,Signals}.)
3358 @cindex breakpoint numbers
3359 @cindex numbers for breakpoints
3360 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3361 catchpoint when you create it; these numbers are successive integers
3362 starting with one. In many of the commands for controlling various
3363 features of breakpoints you use the breakpoint number to say which
3364 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3365 @dfn{disabled}; if disabled, it has no effect on your program until you
3368 @cindex breakpoint ranges
3369 @cindex ranges of breakpoints
3370 Some @value{GDBN} commands accept a range of breakpoints on which to
3371 operate. A breakpoint range is either a single breakpoint number, like
3372 @samp{5}, or two such numbers, in increasing order, separated by a
3373 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3374 all breakpoints in that range are operated on.
3377 * Set Breaks:: Setting breakpoints
3378 * Set Watchpoints:: Setting watchpoints
3379 * Set Catchpoints:: Setting catchpoints
3380 * Delete Breaks:: Deleting breakpoints
3381 * Disabling:: Disabling breakpoints
3382 * Conditions:: Break conditions
3383 * Break Commands:: Breakpoint command lists
3384 * Dynamic Printf:: Dynamic printf
3385 * Save Breakpoints:: How to save breakpoints in a file
3386 * Static Probe Points:: Listing static probe points
3387 * Error in Breakpoints:: ``Cannot insert breakpoints''
3388 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3392 @subsection Setting Breakpoints
3394 @c FIXME LMB what does GDB do if no code on line of breakpt?
3395 @c consider in particular declaration with/without initialization.
3397 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3400 @kindex b @r{(@code{break})}
3401 @vindex $bpnum@r{, convenience variable}
3402 @cindex latest breakpoint
3403 Breakpoints are set with the @code{break} command (abbreviated
3404 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3405 number of the breakpoint you've set most recently; see @ref{Convenience
3406 Vars,, Convenience Variables}, for a discussion of what you can do with
3407 convenience variables.
3410 @item break @var{location}
3411 Set a breakpoint at the given @var{location}, which can specify a
3412 function name, a line number, or an address of an instruction.
3413 (@xref{Specify Location}, for a list of all the possible ways to
3414 specify a @var{location}.) The breakpoint will stop your program just
3415 before it executes any of the code in the specified @var{location}.
3417 When using source languages that permit overloading of symbols, such as
3418 C@t{++}, a function name may refer to more than one possible place to break.
3419 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3422 It is also possible to insert a breakpoint that will stop the program
3423 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3424 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3427 When called without any arguments, @code{break} sets a breakpoint at
3428 the next instruction to be executed in the selected stack frame
3429 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3430 innermost, this makes your program stop as soon as control
3431 returns to that frame. This is similar to the effect of a
3432 @code{finish} command in the frame inside the selected frame---except
3433 that @code{finish} does not leave an active breakpoint. If you use
3434 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3435 the next time it reaches the current location; this may be useful
3438 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3439 least one instruction has been executed. If it did not do this, you
3440 would be unable to proceed past a breakpoint without first disabling the
3441 breakpoint. This rule applies whether or not the breakpoint already
3442 existed when your program stopped.
3444 @item break @dots{} if @var{cond}
3445 Set a breakpoint with condition @var{cond}; evaluate the expression
3446 @var{cond} each time the breakpoint is reached, and stop only if the
3447 value is nonzero---that is, if @var{cond} evaluates as true.
3448 @samp{@dots{}} stands for one of the possible arguments described
3449 above (or no argument) specifying where to break. @xref{Conditions,
3450 ,Break Conditions}, for more information on breakpoint conditions.
3453 @item tbreak @var{args}
3454 Set a breakpoint enabled only for one stop. @var{args} are the
3455 same as for the @code{break} command, and the breakpoint is set in the same
3456 way, but the breakpoint is automatically deleted after the first time your
3457 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3460 @cindex hardware breakpoints
3461 @item hbreak @var{args}
3462 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3463 @code{break} command and the breakpoint is set in the same way, but the
3464 breakpoint requires hardware support and some target hardware may not
3465 have this support. The main purpose of this is EPROM/ROM code
3466 debugging, so you can set a breakpoint at an instruction without
3467 changing the instruction. This can be used with the new trap-generation
3468 provided by SPARClite DSU and most x86-based targets. These targets
3469 will generate traps when a program accesses some data or instruction
3470 address that is assigned to the debug registers. However the hardware
3471 breakpoint registers can take a limited number of breakpoints. For
3472 example, on the DSU, only two data breakpoints can be set at a time, and
3473 @value{GDBN} will reject this command if more than two are used. Delete
3474 or disable unused hardware breakpoints before setting new ones
3475 (@pxref{Disabling, ,Disabling Breakpoints}).
3476 @xref{Conditions, ,Break Conditions}.
3477 For remote targets, you can restrict the number of hardware
3478 breakpoints @value{GDBN} will use, see @ref{set remote
3479 hardware-breakpoint-limit}.
3482 @item thbreak @var{args}
3483 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3484 are the same as for the @code{hbreak} command and the breakpoint is set in
3485 the same way. However, like the @code{tbreak} command,
3486 the breakpoint is automatically deleted after the
3487 first time your program stops there. Also, like the @code{hbreak}
3488 command, the breakpoint requires hardware support and some target hardware
3489 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3490 See also @ref{Conditions, ,Break Conditions}.
3493 @cindex regular expression
3494 @cindex breakpoints at functions matching a regexp
3495 @cindex set breakpoints in many functions
3496 @item rbreak @var{regex}
3497 Set breakpoints on all functions matching the regular expression
3498 @var{regex}. This command sets an unconditional breakpoint on all
3499 matches, printing a list of all breakpoints it set. Once these
3500 breakpoints are set, they are treated just like the breakpoints set with
3501 the @code{break} command. You can delete them, disable them, or make
3502 them conditional the same way as any other breakpoint.
3504 The syntax of the regular expression is the standard one used with tools
3505 like @file{grep}. Note that this is different from the syntax used by
3506 shells, so for instance @code{foo*} matches all functions that include
3507 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3508 @code{.*} leading and trailing the regular expression you supply, so to
3509 match only functions that begin with @code{foo}, use @code{^foo}.
3511 @cindex non-member C@t{++} functions, set breakpoint in
3512 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3513 breakpoints on overloaded functions that are not members of any special
3516 @cindex set breakpoints on all functions
3517 The @code{rbreak} command can be used to set breakpoints in
3518 @strong{all} the functions in a program, like this:
3521 (@value{GDBP}) rbreak .
3524 @item rbreak @var{file}:@var{regex}
3525 If @code{rbreak} is called with a filename qualification, it limits
3526 the search for functions matching the given regular expression to the
3527 specified @var{file}. This can be used, for example, to set breakpoints on
3528 every function in a given file:
3531 (@value{GDBP}) rbreak file.c:.
3534 The colon separating the filename qualifier from the regex may
3535 optionally be surrounded by spaces.
3537 @kindex info breakpoints
3538 @cindex @code{$_} and @code{info breakpoints}
3539 @item info breakpoints @r{[}@var{n}@dots{}@r{]}
3540 @itemx info break @r{[}@var{n}@dots{}@r{]}
3541 Print a table of all breakpoints, watchpoints, and catchpoints set and
3542 not deleted. Optional argument @var{n} means print information only
3543 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3544 For each breakpoint, following columns are printed:
3547 @item Breakpoint Numbers
3549 Breakpoint, watchpoint, or catchpoint.
3551 Whether the breakpoint is marked to be disabled or deleted when hit.
3552 @item Enabled or Disabled
3553 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3554 that are not enabled.
3556 Where the breakpoint is in your program, as a memory address. For a
3557 pending breakpoint whose address is not yet known, this field will
3558 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3559 library that has the symbol or line referred by breakpoint is loaded.
3560 See below for details. A breakpoint with several locations will
3561 have @samp{<MULTIPLE>} in this field---see below for details.
3563 Where the breakpoint is in the source for your program, as a file and
3564 line number. For a pending breakpoint, the original string passed to
3565 the breakpoint command will be listed as it cannot be resolved until
3566 the appropriate shared library is loaded in the future.
3570 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
3571 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
3572 @value{GDBN} on the host's side. If it is ``target'', then the condition
3573 is evaluated by the target. The @code{info break} command shows
3574 the condition on the line following the affected breakpoint, together with
3575 its condition evaluation mode in between parentheses.
3577 Breakpoint commands, if any, are listed after that. A pending breakpoint is
3578 allowed to have a condition specified for it. The condition is not parsed for
3579 validity until a shared library is loaded that allows the pending
3580 breakpoint to resolve to a valid location.
3583 @code{info break} with a breakpoint
3584 number @var{n} as argument lists only that breakpoint. The
3585 convenience variable @code{$_} and the default examining-address for
3586 the @code{x} command are set to the address of the last breakpoint
3587 listed (@pxref{Memory, ,Examining Memory}).
3590 @code{info break} displays a count of the number of times the breakpoint
3591 has been hit. This is especially useful in conjunction with the
3592 @code{ignore} command. You can ignore a large number of breakpoint
3593 hits, look at the breakpoint info to see how many times the breakpoint
3594 was hit, and then run again, ignoring one less than that number. This
3595 will get you quickly to the last hit of that breakpoint.
3598 For a breakpoints with an enable count (xref) greater than 1,
3599 @code{info break} also displays that count.
3603 @value{GDBN} allows you to set any number of breakpoints at the same place in
3604 your program. There is nothing silly or meaningless about this. When
3605 the breakpoints are conditional, this is even useful
3606 (@pxref{Conditions, ,Break Conditions}).
3608 @cindex multiple locations, breakpoints
3609 @cindex breakpoints, multiple locations
3610 It is possible that a breakpoint corresponds to several locations
3611 in your program. Examples of this situation are:
3615 Multiple functions in the program may have the same name.
3618 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3619 instances of the function body, used in different cases.
3622 For a C@t{++} template function, a given line in the function can
3623 correspond to any number of instantiations.
3626 For an inlined function, a given source line can correspond to
3627 several places where that function is inlined.
3630 In all those cases, @value{GDBN} will insert a breakpoint at all
3631 the relevant locations.
3633 A breakpoint with multiple locations is displayed in the breakpoint
3634 table using several rows---one header row, followed by one row for
3635 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3636 address column. The rows for individual locations contain the actual
3637 addresses for locations, and show the functions to which those
3638 locations belong. The number column for a location is of the form
3639 @var{breakpoint-number}.@var{location-number}.
3644 Num Type Disp Enb Address What
3645 1 breakpoint keep y <MULTIPLE>
3647 breakpoint already hit 1 time
3648 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3649 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3652 Each location can be individually enabled or disabled by passing
3653 @var{breakpoint-number}.@var{location-number} as argument to the
3654 @code{enable} and @code{disable} commands. Note that you cannot
3655 delete the individual locations from the list, you can only delete the
3656 entire list of locations that belong to their parent breakpoint (with
3657 the @kbd{delete @var{num}} command, where @var{num} is the number of
3658 the parent breakpoint, 1 in the above example). Disabling or enabling
3659 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3660 that belong to that breakpoint.
3662 @cindex pending breakpoints
3663 It's quite common to have a breakpoint inside a shared library.
3664 Shared libraries can be loaded and unloaded explicitly,
3665 and possibly repeatedly, as the program is executed. To support
3666 this use case, @value{GDBN} updates breakpoint locations whenever
3667 any shared library is loaded or unloaded. Typically, you would
3668 set a breakpoint in a shared library at the beginning of your
3669 debugging session, when the library is not loaded, and when the
3670 symbols from the library are not available. When you try to set
3671 breakpoint, @value{GDBN} will ask you if you want to set
3672 a so called @dfn{pending breakpoint}---breakpoint whose address
3673 is not yet resolved.
3675 After the program is run, whenever a new shared library is loaded,
3676 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3677 shared library contains the symbol or line referred to by some
3678 pending breakpoint, that breakpoint is resolved and becomes an
3679 ordinary breakpoint. When a library is unloaded, all breakpoints
3680 that refer to its symbols or source lines become pending again.
3682 This logic works for breakpoints with multiple locations, too. For
3683 example, if you have a breakpoint in a C@t{++} template function, and
3684 a newly loaded shared library has an instantiation of that template,
3685 a new location is added to the list of locations for the breakpoint.
3687 Except for having unresolved address, pending breakpoints do not
3688 differ from regular breakpoints. You can set conditions or commands,
3689 enable and disable them and perform other breakpoint operations.
3691 @value{GDBN} provides some additional commands for controlling what
3692 happens when the @samp{break} command cannot resolve breakpoint
3693 address specification to an address:
3695 @kindex set breakpoint pending
3696 @kindex show breakpoint pending
3698 @item set breakpoint pending auto
3699 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3700 location, it queries you whether a pending breakpoint should be created.
3702 @item set breakpoint pending on
3703 This indicates that an unrecognized breakpoint location should automatically
3704 result in a pending breakpoint being created.
3706 @item set breakpoint pending off
3707 This indicates that pending breakpoints are not to be created. Any
3708 unrecognized breakpoint location results in an error. This setting does
3709 not affect any pending breakpoints previously created.
3711 @item show breakpoint pending
3712 Show the current behavior setting for creating pending breakpoints.
3715 The settings above only affect the @code{break} command and its
3716 variants. Once breakpoint is set, it will be automatically updated
3717 as shared libraries are loaded and unloaded.
3719 @cindex automatic hardware breakpoints
3720 For some targets, @value{GDBN} can automatically decide if hardware or
3721 software breakpoints should be used, depending on whether the
3722 breakpoint address is read-only or read-write. This applies to
3723 breakpoints set with the @code{break} command as well as to internal
3724 breakpoints set by commands like @code{next} and @code{finish}. For
3725 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3728 You can control this automatic behaviour with the following commands::
3730 @kindex set breakpoint auto-hw
3731 @kindex show breakpoint auto-hw
3733 @item set breakpoint auto-hw on
3734 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3735 will try to use the target memory map to decide if software or hardware
3736 breakpoint must be used.
3738 @item set breakpoint auto-hw off
3739 This indicates @value{GDBN} should not automatically select breakpoint
3740 type. If the target provides a memory map, @value{GDBN} will warn when
3741 trying to set software breakpoint at a read-only address.
3744 @value{GDBN} normally implements breakpoints by replacing the program code
3745 at the breakpoint address with a special instruction, which, when
3746 executed, given control to the debugger. By default, the program
3747 code is so modified only when the program is resumed. As soon as
3748 the program stops, @value{GDBN} restores the original instructions. This
3749 behaviour guards against leaving breakpoints inserted in the
3750 target should gdb abrubptly disconnect. However, with slow remote
3751 targets, inserting and removing breakpoint can reduce the performance.
3752 This behavior can be controlled with the following commands::
3754 @kindex set breakpoint always-inserted
3755 @kindex show breakpoint always-inserted
3757 @item set breakpoint always-inserted off
3758 All breakpoints, including newly added by the user, are inserted in
3759 the target only when the target is resumed. All breakpoints are
3760 removed from the target when it stops.
3762 @item set breakpoint always-inserted on
3763 Causes all breakpoints to be inserted in the target at all times. If
3764 the user adds a new breakpoint, or changes an existing breakpoint, the
3765 breakpoints in the target are updated immediately. A breakpoint is
3766 removed from the target only when breakpoint itself is removed.
3768 @cindex non-stop mode, and @code{breakpoint always-inserted}
3769 @item set breakpoint always-inserted auto
3770 This is the default mode. If @value{GDBN} is controlling the inferior
3771 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3772 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3773 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3774 @code{breakpoint always-inserted} mode is off.
3777 @value{GDBN} handles conditional breakpoints by evaluating these conditions
3778 when a breakpoint breaks. If the condition is true, then the process being
3779 debugged stops, otherwise the process is resumed.
3781 If the target supports evaluating conditions on its end, @value{GDBN} may
3782 download the breakpoint, together with its conditions, to it.
3784 This feature can be controlled via the following commands:
3786 @kindex set breakpoint condition-evaluation
3787 @kindex show breakpoint condition-evaluation
3789 @item set breakpoint condition-evaluation host
3790 This option commands @value{GDBN} to evaluate the breakpoint
3791 conditions on the host's side. Unconditional breakpoints are sent to
3792 the target which in turn receives the triggers and reports them back to GDB
3793 for condition evaluation. This is the standard evaluation mode.
3795 @item set breakpoint condition-evaluation target
3796 This option commands @value{GDBN} to download breakpoint conditions
3797 to the target at the moment of their insertion. The target
3798 is responsible for evaluating the conditional expression and reporting
3799 breakpoint stop events back to @value{GDBN} whenever the condition
3800 is true. Due to limitations of target-side evaluation, some conditions
3801 cannot be evaluated there, e.g., conditions that depend on local data
3802 that is only known to the host. Examples include
3803 conditional expressions involving convenience variables, complex types
3804 that cannot be handled by the agent expression parser and expressions
3805 that are too long to be sent over to the target, specially when the
3806 target is a remote system. In these cases, the conditions will be
3807 evaluated by @value{GDBN}.
3809 @item set breakpoint condition-evaluation auto
3810 This is the default mode. If the target supports evaluating breakpoint
3811 conditions on its end, @value{GDBN} will download breakpoint conditions to
3812 the target (limitations mentioned previously apply). If the target does
3813 not support breakpoint condition evaluation, then @value{GDBN} will fallback
3814 to evaluating all these conditions on the host's side.
3818 @cindex negative breakpoint numbers
3819 @cindex internal @value{GDBN} breakpoints
3820 @value{GDBN} itself sometimes sets breakpoints in your program for
3821 special purposes, such as proper handling of @code{longjmp} (in C
3822 programs). These internal breakpoints are assigned negative numbers,
3823 starting with @code{-1}; @samp{info breakpoints} does not display them.
3824 You can see these breakpoints with the @value{GDBN} maintenance command
3825 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3828 @node Set Watchpoints
3829 @subsection Setting Watchpoints
3831 @cindex setting watchpoints
3832 You can use a watchpoint to stop execution whenever the value of an
3833 expression changes, without having to predict a particular place where
3834 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3835 The expression may be as simple as the value of a single variable, or
3836 as complex as many variables combined by operators. Examples include:
3840 A reference to the value of a single variable.
3843 An address cast to an appropriate data type. For example,
3844 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3845 address (assuming an @code{int} occupies 4 bytes).
3848 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3849 expression can use any operators valid in the program's native
3850 language (@pxref{Languages}).
3853 You can set a watchpoint on an expression even if the expression can
3854 not be evaluated yet. For instance, you can set a watchpoint on
3855 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3856 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3857 the expression produces a valid value. If the expression becomes
3858 valid in some other way than changing a variable (e.g.@: if the memory
3859 pointed to by @samp{*global_ptr} becomes readable as the result of a
3860 @code{malloc} call), @value{GDBN} may not stop until the next time
3861 the expression changes.
3863 @cindex software watchpoints
3864 @cindex hardware watchpoints
3865 Depending on your system, watchpoints may be implemented in software or
3866 hardware. @value{GDBN} does software watchpointing by single-stepping your
3867 program and testing the variable's value each time, which is hundreds of
3868 times slower than normal execution. (But this may still be worth it, to
3869 catch errors where you have no clue what part of your program is the
3872 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3873 x86-based targets, @value{GDBN} includes support for hardware
3874 watchpoints, which do not slow down the running of your program.
3878 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3879 Set a watchpoint for an expression. @value{GDBN} will break when the
3880 expression @var{expr} is written into by the program and its value
3881 changes. The simplest (and the most popular) use of this command is
3882 to watch the value of a single variable:
3885 (@value{GDBP}) watch foo
3888 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3889 argument, @value{GDBN} breaks only when the thread identified by
3890 @var{threadnum} changes the value of @var{expr}. If any other threads
3891 change the value of @var{expr}, @value{GDBN} will not break. Note
3892 that watchpoints restricted to a single thread in this way only work
3893 with Hardware Watchpoints.
3895 Ordinarily a watchpoint respects the scope of variables in @var{expr}
3896 (see below). The @code{-location} argument tells @value{GDBN} to
3897 instead watch the memory referred to by @var{expr}. In this case,
3898 @value{GDBN} will evaluate @var{expr}, take the address of the result,
3899 and watch the memory at that address. The type of the result is used
3900 to determine the size of the watched memory. If the expression's
3901 result does not have an address, then @value{GDBN} will print an
3904 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
3905 of masked watchpoints, if the current architecture supports this
3906 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
3907 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
3908 to an address to watch. The mask specifies that some bits of an address
3909 (the bits which are reset in the mask) should be ignored when matching
3910 the address accessed by the inferior against the watchpoint address.
3911 Thus, a masked watchpoint watches many addresses simultaneously---those
3912 addresses whose unmasked bits are identical to the unmasked bits in the
3913 watchpoint address. The @code{mask} argument implies @code{-location}.
3917 (@value{GDBP}) watch foo mask 0xffff00ff
3918 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
3922 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3923 Set a watchpoint that will break when the value of @var{expr} is read
3927 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3928 Set a watchpoint that will break when @var{expr} is either read from
3929 or written into by the program.
3931 @kindex info watchpoints @r{[}@var{n}@dots{}@r{]}
3932 @item info watchpoints @r{[}@var{n}@dots{}@r{]}
3933 This command prints a list of watchpoints, using the same format as
3934 @code{info break} (@pxref{Set Breaks}).
3937 If you watch for a change in a numerically entered address you need to
3938 dereference it, as the address itself is just a constant number which will
3939 never change. @value{GDBN} refuses to create a watchpoint that watches
3940 a never-changing value:
3943 (@value{GDBP}) watch 0x600850
3944 Cannot watch constant value 0x600850.
3945 (@value{GDBP}) watch *(int *) 0x600850
3946 Watchpoint 1: *(int *) 6293584
3949 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3950 watchpoints execute very quickly, and the debugger reports a change in
3951 value at the exact instruction where the change occurs. If @value{GDBN}
3952 cannot set a hardware watchpoint, it sets a software watchpoint, which
3953 executes more slowly and reports the change in value at the next
3954 @emph{statement}, not the instruction, after the change occurs.
3956 @cindex use only software watchpoints
3957 You can force @value{GDBN} to use only software watchpoints with the
3958 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3959 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3960 the underlying system supports them. (Note that hardware-assisted
3961 watchpoints that were set @emph{before} setting
3962 @code{can-use-hw-watchpoints} to zero will still use the hardware
3963 mechanism of watching expression values.)
3966 @item set can-use-hw-watchpoints
3967 @kindex set can-use-hw-watchpoints
3968 Set whether or not to use hardware watchpoints.
3970 @item show can-use-hw-watchpoints
3971 @kindex show can-use-hw-watchpoints
3972 Show the current mode of using hardware watchpoints.
3975 For remote targets, you can restrict the number of hardware
3976 watchpoints @value{GDBN} will use, see @ref{set remote
3977 hardware-breakpoint-limit}.
3979 When you issue the @code{watch} command, @value{GDBN} reports
3982 Hardware watchpoint @var{num}: @var{expr}
3986 if it was able to set a hardware watchpoint.
3988 Currently, the @code{awatch} and @code{rwatch} commands can only set
3989 hardware watchpoints, because accesses to data that don't change the
3990 value of the watched expression cannot be detected without examining
3991 every instruction as it is being executed, and @value{GDBN} does not do
3992 that currently. If @value{GDBN} finds that it is unable to set a
3993 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3994 will print a message like this:
3997 Expression cannot be implemented with read/access watchpoint.
4000 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4001 data type of the watched expression is wider than what a hardware
4002 watchpoint on the target machine can handle. For example, some systems
4003 can only watch regions that are up to 4 bytes wide; on such systems you
4004 cannot set hardware watchpoints for an expression that yields a
4005 double-precision floating-point number (which is typically 8 bytes
4006 wide). As a work-around, it might be possible to break the large region
4007 into a series of smaller ones and watch them with separate watchpoints.
4009 If you set too many hardware watchpoints, @value{GDBN} might be unable
4010 to insert all of them when you resume the execution of your program.
4011 Since the precise number of active watchpoints is unknown until such
4012 time as the program is about to be resumed, @value{GDBN} might not be
4013 able to warn you about this when you set the watchpoints, and the
4014 warning will be printed only when the program is resumed:
4017 Hardware watchpoint @var{num}: Could not insert watchpoint
4021 If this happens, delete or disable some of the watchpoints.
4023 Watching complex expressions that reference many variables can also
4024 exhaust the resources available for hardware-assisted watchpoints.
4025 That's because @value{GDBN} needs to watch every variable in the
4026 expression with separately allocated resources.
4028 If you call a function interactively using @code{print} or @code{call},
4029 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4030 kind of breakpoint or the call completes.
4032 @value{GDBN} automatically deletes watchpoints that watch local
4033 (automatic) variables, or expressions that involve such variables, when
4034 they go out of scope, that is, when the execution leaves the block in
4035 which these variables were defined. In particular, when the program
4036 being debugged terminates, @emph{all} local variables go out of scope,
4037 and so only watchpoints that watch global variables remain set. If you
4038 rerun the program, you will need to set all such watchpoints again. One
4039 way of doing that would be to set a code breakpoint at the entry to the
4040 @code{main} function and when it breaks, set all the watchpoints.
4042 @cindex watchpoints and threads
4043 @cindex threads and watchpoints
4044 In multi-threaded programs, watchpoints will detect changes to the
4045 watched expression from every thread.
4048 @emph{Warning:} In multi-threaded programs, software watchpoints
4049 have only limited usefulness. If @value{GDBN} creates a software
4050 watchpoint, it can only watch the value of an expression @emph{in a
4051 single thread}. If you are confident that the expression can only
4052 change due to the current thread's activity (and if you are also
4053 confident that no other thread can become current), then you can use
4054 software watchpoints as usual. However, @value{GDBN} may not notice
4055 when a non-current thread's activity changes the expression. (Hardware
4056 watchpoints, in contrast, watch an expression in all threads.)
4059 @xref{set remote hardware-watchpoint-limit}.
4061 @node Set Catchpoints
4062 @subsection Setting Catchpoints
4063 @cindex catchpoints, setting
4064 @cindex exception handlers
4065 @cindex event handling
4067 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4068 kinds of program events, such as C@t{++} exceptions or the loading of a
4069 shared library. Use the @code{catch} command to set a catchpoint.
4073 @item catch @var{event}
4074 Stop when @var{event} occurs. @var{event} can be any of the following:
4077 @item throw @r{[}@var{regexp}@r{]}
4078 @itemx rethrow @r{[}@var{regexp}@r{]}
4079 @itemx catch @r{[}@var{regexp}@r{]}
4080 @cindex stop on C@t{++} exceptions
4081 The throwing, re-throwing, or catching of a C@t{++} exception.
4083 If @var{regexp} is given, then only exceptions whose type matches the
4084 regular expression will be caught.
4086 @vindex $_exception@r{, convenience variable}
4087 The convenience variable @code{$_exception} is available at an
4088 exception-related catchpoint, on some systems. This holds the
4089 exception being thrown.
4091 There are currently some limitations to C@t{++} exception handling in
4096 The support for these commands is system-dependent. Currently, only
4097 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4101 The regular expression feature and the @code{$_exception} convenience
4102 variable rely on the presence of some SDT probes in @code{libstdc++}.
4103 If these probes are not present, then these features cannot be used.
4104 These probes were first available in the GCC 4.8 release, but whether
4105 or not they are available in your GCC also depends on how it was
4109 The @code{$_exception} convenience variable is only valid at the
4110 instruction at which an exception-related catchpoint is set.
4113 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4114 location in the system library which implements runtime exception
4115 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4116 (@pxref{Selection}) to get to your code.
4119 If you call a function interactively, @value{GDBN} normally returns
4120 control to you when the function has finished executing. If the call
4121 raises an exception, however, the call may bypass the mechanism that
4122 returns control to you and cause your program either to abort or to
4123 simply continue running until it hits a breakpoint, catches a signal
4124 that @value{GDBN} is listening for, or exits. This is the case even if
4125 you set a catchpoint for the exception; catchpoints on exceptions are
4126 disabled within interactive calls. @xref{Calling}, for information on
4127 controlling this with @code{set unwind-on-terminating-exception}.
4130 You cannot raise an exception interactively.
4133 You cannot install an exception handler interactively.
4137 @cindex Ada exception catching
4138 @cindex catch Ada exceptions
4139 An Ada exception being raised. If an exception name is specified
4140 at the end of the command (eg @code{catch exception Program_Error}),
4141 the debugger will stop only when this specific exception is raised.
4142 Otherwise, the debugger stops execution when any Ada exception is raised.
4144 When inserting an exception catchpoint on a user-defined exception whose
4145 name is identical to one of the exceptions defined by the language, the
4146 fully qualified name must be used as the exception name. Otherwise,
4147 @value{GDBN} will assume that it should stop on the pre-defined exception
4148 rather than the user-defined one. For instance, assuming an exception
4149 called @code{Constraint_Error} is defined in package @code{Pck}, then
4150 the command to use to catch such exceptions is @kbd{catch exception
4151 Pck.Constraint_Error}.
4153 @item exception unhandled
4154 An exception that was raised but is not handled by the program.
4157 A failed Ada assertion.
4160 @cindex break on fork/exec
4161 A call to @code{exec}. This is currently only available for HP-UX
4165 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
4166 @cindex break on a system call.
4167 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4168 syscall is a mechanism for application programs to request a service
4169 from the operating system (OS) or one of the OS system services.
4170 @value{GDBN} can catch some or all of the syscalls issued by the
4171 debuggee, and show the related information for each syscall. If no
4172 argument is specified, calls to and returns from all system calls
4175 @var{name} can be any system call name that is valid for the
4176 underlying OS. Just what syscalls are valid depends on the OS. On
4177 GNU and Unix systems, you can find the full list of valid syscall
4178 names on @file{/usr/include/asm/unistd.h}.
4180 @c For MS-Windows, the syscall names and the corresponding numbers
4181 @c can be found, e.g., on this URL:
4182 @c http://www.metasploit.com/users/opcode/syscalls.html
4183 @c but we don't support Windows syscalls yet.
4185 Normally, @value{GDBN} knows in advance which syscalls are valid for
4186 each OS, so you can use the @value{GDBN} command-line completion
4187 facilities (@pxref{Completion,, command completion}) to list the
4190 You may also specify the system call numerically. A syscall's
4191 number is the value passed to the OS's syscall dispatcher to
4192 identify the requested service. When you specify the syscall by its
4193 name, @value{GDBN} uses its database of syscalls to convert the name
4194 into the corresponding numeric code, but using the number directly
4195 may be useful if @value{GDBN}'s database does not have the complete
4196 list of syscalls on your system (e.g., because @value{GDBN} lags
4197 behind the OS upgrades).
4199 The example below illustrates how this command works if you don't provide
4203 (@value{GDBP}) catch syscall
4204 Catchpoint 1 (syscall)
4206 Starting program: /tmp/catch-syscall
4208 Catchpoint 1 (call to syscall 'close'), \
4209 0xffffe424 in __kernel_vsyscall ()
4213 Catchpoint 1 (returned from syscall 'close'), \
4214 0xffffe424 in __kernel_vsyscall ()
4218 Here is an example of catching a system call by name:
4221 (@value{GDBP}) catch syscall chroot
4222 Catchpoint 1 (syscall 'chroot' [61])
4224 Starting program: /tmp/catch-syscall
4226 Catchpoint 1 (call to syscall 'chroot'), \
4227 0xffffe424 in __kernel_vsyscall ()
4231 Catchpoint 1 (returned from syscall 'chroot'), \
4232 0xffffe424 in __kernel_vsyscall ()
4236 An example of specifying a system call numerically. In the case
4237 below, the syscall number has a corresponding entry in the XML
4238 file, so @value{GDBN} finds its name and prints it:
4241 (@value{GDBP}) catch syscall 252
4242 Catchpoint 1 (syscall(s) 'exit_group')
4244 Starting program: /tmp/catch-syscall
4246 Catchpoint 1 (call to syscall 'exit_group'), \
4247 0xffffe424 in __kernel_vsyscall ()
4251 Program exited normally.
4255 However, there can be situations when there is no corresponding name
4256 in XML file for that syscall number. In this case, @value{GDBN} prints
4257 a warning message saying that it was not able to find the syscall name,
4258 but the catchpoint will be set anyway. See the example below:
4261 (@value{GDBP}) catch syscall 764
4262 warning: The number '764' does not represent a known syscall.
4263 Catchpoint 2 (syscall 764)
4267 If you configure @value{GDBN} using the @samp{--without-expat} option,
4268 it will not be able to display syscall names. Also, if your
4269 architecture does not have an XML file describing its system calls,
4270 you will not be able to see the syscall names. It is important to
4271 notice that these two features are used for accessing the syscall
4272 name database. In either case, you will see a warning like this:
4275 (@value{GDBP}) catch syscall
4276 warning: Could not open "syscalls/i386-linux.xml"
4277 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4278 GDB will not be able to display syscall names.
4279 Catchpoint 1 (syscall)
4283 Of course, the file name will change depending on your architecture and system.
4285 Still using the example above, you can also try to catch a syscall by its
4286 number. In this case, you would see something like:
4289 (@value{GDBP}) catch syscall 252
4290 Catchpoint 1 (syscall(s) 252)
4293 Again, in this case @value{GDBN} would not be able to display syscall's names.
4296 A call to @code{fork}. This is currently only available for HP-UX
4300 A call to @code{vfork}. This is currently only available for HP-UX
4303 @item load @r{[}regexp@r{]}
4304 @itemx unload @r{[}regexp@r{]}
4305 The loading or unloading of a shared library. If @var{regexp} is
4306 given, then the catchpoint will stop only if the regular expression
4307 matches one of the affected libraries.
4309 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4310 The delivery of a signal.
4312 With no arguments, this catchpoint will catch any signal that is not
4313 used internally by @value{GDBN}, specifically, all signals except
4314 @samp{SIGTRAP} and @samp{SIGINT}.
4316 With the argument @samp{all}, all signals, including those used by
4317 @value{GDBN}, will be caught. This argument cannot be used with other
4320 Otherwise, the arguments are a list of signal names as given to
4321 @code{handle} (@pxref{Signals}). Only signals specified in this list
4324 One reason that @code{catch signal} can be more useful than
4325 @code{handle} is that you can attach commands and conditions to the
4328 When a signal is caught by a catchpoint, the signal's @code{stop} and
4329 @code{print} settings, as specified by @code{handle}, are ignored.
4330 However, whether the signal is still delivered to the inferior depends
4331 on the @code{pass} setting; this can be changed in the catchpoint's
4336 @item tcatch @var{event}
4337 Set a catchpoint that is enabled only for one stop. The catchpoint is
4338 automatically deleted after the first time the event is caught.
4342 Use the @code{info break} command to list the current catchpoints.
4346 @subsection Deleting Breakpoints
4348 @cindex clearing breakpoints, watchpoints, catchpoints
4349 @cindex deleting breakpoints, watchpoints, catchpoints
4350 It is often necessary to eliminate a breakpoint, watchpoint, or
4351 catchpoint once it has done its job and you no longer want your program
4352 to stop there. This is called @dfn{deleting} the breakpoint. A
4353 breakpoint that has been deleted no longer exists; it is forgotten.
4355 With the @code{clear} command you can delete breakpoints according to
4356 where they are in your program. With the @code{delete} command you can
4357 delete individual breakpoints, watchpoints, or catchpoints by specifying
4358 their breakpoint numbers.
4360 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4361 automatically ignores breakpoints on the first instruction to be executed
4362 when you continue execution without changing the execution address.
4367 Delete any breakpoints at the next instruction to be executed in the
4368 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4369 the innermost frame is selected, this is a good way to delete a
4370 breakpoint where your program just stopped.
4372 @item clear @var{location}
4373 Delete any breakpoints set at the specified @var{location}.
4374 @xref{Specify Location}, for the various forms of @var{location}; the
4375 most useful ones are listed below:
4378 @item clear @var{function}
4379 @itemx clear @var{filename}:@var{function}
4380 Delete any breakpoints set at entry to the named @var{function}.
4382 @item clear @var{linenum}
4383 @itemx clear @var{filename}:@var{linenum}
4384 Delete any breakpoints set at or within the code of the specified
4385 @var{linenum} of the specified @var{filename}.
4388 @cindex delete breakpoints
4390 @kindex d @r{(@code{delete})}
4391 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4392 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4393 ranges specified as arguments. If no argument is specified, delete all
4394 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4395 confirm off}). You can abbreviate this command as @code{d}.
4399 @subsection Disabling Breakpoints
4401 @cindex enable/disable a breakpoint
4402 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4403 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4404 it had been deleted, but remembers the information on the breakpoint so
4405 that you can @dfn{enable} it again later.
4407 You disable and enable breakpoints, watchpoints, and catchpoints with
4408 the @code{enable} and @code{disable} commands, optionally specifying
4409 one or more breakpoint numbers as arguments. Use @code{info break} to
4410 print a list of all breakpoints, watchpoints, and catchpoints if you
4411 do not know which numbers to use.
4413 Disabling and enabling a breakpoint that has multiple locations
4414 affects all of its locations.
4416 A breakpoint, watchpoint, or catchpoint can have any of several
4417 different states of enablement:
4421 Enabled. The breakpoint stops your program. A breakpoint set
4422 with the @code{break} command starts out in this state.
4424 Disabled. The breakpoint has no effect on your program.
4426 Enabled once. The breakpoint stops your program, but then becomes
4429 Enabled for a count. The breakpoint stops your program for the next
4430 N times, then becomes disabled.
4432 Enabled for deletion. The breakpoint stops your program, but
4433 immediately after it does so it is deleted permanently. A breakpoint
4434 set with the @code{tbreak} command starts out in this state.
4437 You can use the following commands to enable or disable breakpoints,
4438 watchpoints, and catchpoints:
4442 @kindex dis @r{(@code{disable})}
4443 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4444 Disable the specified breakpoints---or all breakpoints, if none are
4445 listed. A disabled breakpoint has no effect but is not forgotten. All
4446 options such as ignore-counts, conditions and commands are remembered in
4447 case the breakpoint is enabled again later. You may abbreviate
4448 @code{disable} as @code{dis}.
4451 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4452 Enable the specified breakpoints (or all defined breakpoints). They
4453 become effective once again in stopping your program.
4455 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4456 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4457 of these breakpoints immediately after stopping your program.
4459 @item enable @r{[}breakpoints@r{]} count @var{count} @var{range}@dots{}
4460 Enable the specified breakpoints temporarily. @value{GDBN} records
4461 @var{count} with each of the specified breakpoints, and decrements a
4462 breakpoint's count when it is hit. When any count reaches 0,
4463 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4464 count (@pxref{Conditions, ,Break Conditions}), that will be
4465 decremented to 0 before @var{count} is affected.
4467 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4468 Enable the specified breakpoints to work once, then die. @value{GDBN}
4469 deletes any of these breakpoints as soon as your program stops there.
4470 Breakpoints set by the @code{tbreak} command start out in this state.
4473 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4474 @c confusing: tbreak is also initially enabled.
4475 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4476 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4477 subsequently, they become disabled or enabled only when you use one of
4478 the commands above. (The command @code{until} can set and delete a
4479 breakpoint of its own, but it does not change the state of your other
4480 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4484 @subsection Break Conditions
4485 @cindex conditional breakpoints
4486 @cindex breakpoint conditions
4488 @c FIXME what is scope of break condition expr? Context where wanted?
4489 @c in particular for a watchpoint?
4490 The simplest sort of breakpoint breaks every time your program reaches a
4491 specified place. You can also specify a @dfn{condition} for a
4492 breakpoint. A condition is just a Boolean expression in your
4493 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4494 a condition evaluates the expression each time your program reaches it,
4495 and your program stops only if the condition is @emph{true}.
4497 This is the converse of using assertions for program validation; in that
4498 situation, you want to stop when the assertion is violated---that is,
4499 when the condition is false. In C, if you want to test an assertion expressed
4500 by the condition @var{assert}, you should set the condition
4501 @samp{! @var{assert}} on the appropriate breakpoint.
4503 Conditions are also accepted for watchpoints; you may not need them,
4504 since a watchpoint is inspecting the value of an expression anyhow---but
4505 it might be simpler, say, to just set a watchpoint on a variable name,
4506 and specify a condition that tests whether the new value is an interesting
4509 Break conditions can have side effects, and may even call functions in
4510 your program. This can be useful, for example, to activate functions
4511 that log program progress, or to use your own print functions to
4512 format special data structures. The effects are completely predictable
4513 unless there is another enabled breakpoint at the same address. (In
4514 that case, @value{GDBN} might see the other breakpoint first and stop your
4515 program without checking the condition of this one.) Note that
4516 breakpoint commands are usually more convenient and flexible than break
4518 purpose of performing side effects when a breakpoint is reached
4519 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4521 Breakpoint conditions can also be evaluated on the target's side if
4522 the target supports it. Instead of evaluating the conditions locally,
4523 @value{GDBN} encodes the expression into an agent expression
4524 (@pxref{Agent Expressions}) suitable for execution on the target,
4525 independently of @value{GDBN}. Global variables become raw memory
4526 locations, locals become stack accesses, and so forth.
4528 In this case, @value{GDBN} will only be notified of a breakpoint trigger
4529 when its condition evaluates to true. This mechanism may provide faster
4530 response times depending on the performance characteristics of the target
4531 since it does not need to keep @value{GDBN} informed about
4532 every breakpoint trigger, even those with false conditions.
4534 Break conditions can be specified when a breakpoint is set, by using
4535 @samp{if} in the arguments to the @code{break} command. @xref{Set
4536 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4537 with the @code{condition} command.
4539 You can also use the @code{if} keyword with the @code{watch} command.
4540 The @code{catch} command does not recognize the @code{if} keyword;
4541 @code{condition} is the only way to impose a further condition on a
4546 @item condition @var{bnum} @var{expression}
4547 Specify @var{expression} as the break condition for breakpoint,
4548 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4549 breakpoint @var{bnum} stops your program only if the value of
4550 @var{expression} is true (nonzero, in C). When you use
4551 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4552 syntactic correctness, and to determine whether symbols in it have
4553 referents in the context of your breakpoint. If @var{expression} uses
4554 symbols not referenced in the context of the breakpoint, @value{GDBN}
4555 prints an error message:
4558 No symbol "foo" in current context.
4563 not actually evaluate @var{expression} at the time the @code{condition}
4564 command (or a command that sets a breakpoint with a condition, like
4565 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4567 @item condition @var{bnum}
4568 Remove the condition from breakpoint number @var{bnum}. It becomes
4569 an ordinary unconditional breakpoint.
4572 @cindex ignore count (of breakpoint)
4573 A special case of a breakpoint condition is to stop only when the
4574 breakpoint has been reached a certain number of times. This is so
4575 useful that there is a special way to do it, using the @dfn{ignore
4576 count} of the breakpoint. Every breakpoint has an ignore count, which
4577 is an integer. Most of the time, the ignore count is zero, and
4578 therefore has no effect. But if your program reaches a breakpoint whose
4579 ignore count is positive, then instead of stopping, it just decrements
4580 the ignore count by one and continues. As a result, if the ignore count
4581 value is @var{n}, the breakpoint does not stop the next @var{n} times
4582 your program reaches it.
4586 @item ignore @var{bnum} @var{count}
4587 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4588 The next @var{count} times the breakpoint is reached, your program's
4589 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4592 To make the breakpoint stop the next time it is reached, specify
4595 When you use @code{continue} to resume execution of your program from a
4596 breakpoint, you can specify an ignore count directly as an argument to
4597 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4598 Stepping,,Continuing and Stepping}.
4600 If a breakpoint has a positive ignore count and a condition, the
4601 condition is not checked. Once the ignore count reaches zero,
4602 @value{GDBN} resumes checking the condition.
4604 You could achieve the effect of the ignore count with a condition such
4605 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4606 is decremented each time. @xref{Convenience Vars, ,Convenience
4610 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4613 @node Break Commands
4614 @subsection Breakpoint Command Lists
4616 @cindex breakpoint commands
4617 You can give any breakpoint (or watchpoint or catchpoint) a series of
4618 commands to execute when your program stops due to that breakpoint. For
4619 example, you might want to print the values of certain expressions, or
4620 enable other breakpoints.
4624 @kindex end@r{ (breakpoint commands)}
4625 @item commands @r{[}@var{range}@dots{}@r{]}
4626 @itemx @dots{} @var{command-list} @dots{}
4628 Specify a list of commands for the given breakpoints. The commands
4629 themselves appear on the following lines. Type a line containing just
4630 @code{end} to terminate the commands.
4632 To remove all commands from a breakpoint, type @code{commands} and
4633 follow it immediately with @code{end}; that is, give no commands.
4635 With no argument, @code{commands} refers to the last breakpoint,
4636 watchpoint, or catchpoint set (not to the breakpoint most recently
4637 encountered). If the most recent breakpoints were set with a single
4638 command, then the @code{commands} will apply to all the breakpoints
4639 set by that command. This applies to breakpoints set by
4640 @code{rbreak}, and also applies when a single @code{break} command
4641 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4645 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4646 disabled within a @var{command-list}.
4648 You can use breakpoint commands to start your program up again. Simply
4649 use the @code{continue} command, or @code{step}, or any other command
4650 that resumes execution.
4652 Any other commands in the command list, after a command that resumes
4653 execution, are ignored. This is because any time you resume execution
4654 (even with a simple @code{next} or @code{step}), you may encounter
4655 another breakpoint---which could have its own command list, leading to
4656 ambiguities about which list to execute.
4659 If the first command you specify in a command list is @code{silent}, the
4660 usual message about stopping at a breakpoint is not printed. This may
4661 be desirable for breakpoints that are to print a specific message and
4662 then continue. If none of the remaining commands print anything, you
4663 see no sign that the breakpoint was reached. @code{silent} is
4664 meaningful only at the beginning of a breakpoint command list.
4666 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4667 print precisely controlled output, and are often useful in silent
4668 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4670 For example, here is how you could use breakpoint commands to print the
4671 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4677 printf "x is %d\n",x
4682 One application for breakpoint commands is to compensate for one bug so
4683 you can test for another. Put a breakpoint just after the erroneous line
4684 of code, give it a condition to detect the case in which something
4685 erroneous has been done, and give it commands to assign correct values
4686 to any variables that need them. End with the @code{continue} command
4687 so that your program does not stop, and start with the @code{silent}
4688 command so that no output is produced. Here is an example:
4699 @node Dynamic Printf
4700 @subsection Dynamic Printf
4702 @cindex dynamic printf
4704 The dynamic printf command @code{dprintf} combines a breakpoint with
4705 formatted printing of your program's data to give you the effect of
4706 inserting @code{printf} calls into your program on-the-fly, without
4707 having to recompile it.
4709 In its most basic form, the output goes to the GDB console. However,
4710 you can set the variable @code{dprintf-style} for alternate handling.
4711 For instance, you can ask to format the output by calling your
4712 program's @code{printf} function. This has the advantage that the
4713 characters go to the program's output device, so they can recorded in
4714 redirects to files and so forth.
4716 If you are doing remote debugging with a stub or agent, you can also
4717 ask to have the printf handled by the remote agent. In addition to
4718 ensuring that the output goes to the remote program's device along
4719 with any other output the program might produce, you can also ask that
4720 the dprintf remain active even after disconnecting from the remote
4721 target. Using the stub/agent is also more efficient, as it can do
4722 everything without needing to communicate with @value{GDBN}.
4726 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
4727 Whenever execution reaches @var{location}, print the values of one or
4728 more @var{expressions} under the control of the string @var{template}.
4729 To print several values, separate them with commas.
4731 @item set dprintf-style @var{style}
4732 Set the dprintf output to be handled in one of several different
4733 styles enumerated below. A change of style affects all existing
4734 dynamic printfs immediately. (If you need individual control over the
4735 print commands, simply define normal breakpoints with
4736 explicitly-supplied command lists.)
4739 @kindex dprintf-style gdb
4740 Handle the output using the @value{GDBN} @code{printf} command.
4743 @kindex dprintf-style call
4744 Handle the output by calling a function in your program (normally
4748 @kindex dprintf-style agent
4749 Have the remote debugging agent (such as @code{gdbserver}) handle
4750 the output itself. This style is only available for agents that
4751 support running commands on the target.
4753 @item set dprintf-function @var{function}
4754 Set the function to call if the dprintf style is @code{call}. By
4755 default its value is @code{printf}. You may set it to any expression.
4756 that @value{GDBN} can evaluate to a function, as per the @code{call}
4759 @item set dprintf-channel @var{channel}
4760 Set a ``channel'' for dprintf. If set to a non-empty value,
4761 @value{GDBN} will evaluate it as an expression and pass the result as
4762 a first argument to the @code{dprintf-function}, in the manner of
4763 @code{fprintf} and similar functions. Otherwise, the dprintf format
4764 string will be the first argument, in the manner of @code{printf}.
4766 As an example, if you wanted @code{dprintf} output to go to a logfile
4767 that is a standard I/O stream assigned to the variable @code{mylog},
4768 you could do the following:
4771 (gdb) set dprintf-style call
4772 (gdb) set dprintf-function fprintf
4773 (gdb) set dprintf-channel mylog
4774 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
4775 Dprintf 1 at 0x123456: file main.c, line 25.
4777 1 dprintf keep y 0x00123456 in main at main.c:25
4778 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
4783 Note that the @code{info break} displays the dynamic printf commands
4784 as normal breakpoint commands; you can thus easily see the effect of
4785 the variable settings.
4787 @item set disconnected-dprintf on
4788 @itemx set disconnected-dprintf off
4789 @kindex set disconnected-dprintf
4790 Choose whether @code{dprintf} commands should continue to run if
4791 @value{GDBN} has disconnected from the target. This only applies
4792 if the @code{dprintf-style} is @code{agent}.
4794 @item show disconnected-dprintf off
4795 @kindex show disconnected-dprintf
4796 Show the current choice for disconnected @code{dprintf}.
4800 @value{GDBN} does not check the validity of function and channel,
4801 relying on you to supply values that are meaningful for the contexts
4802 in which they are being used. For instance, the function and channel
4803 may be the values of local variables, but if that is the case, then
4804 all enabled dynamic prints must be at locations within the scope of
4805 those locals. If evaluation fails, @value{GDBN} will report an error.
4807 @node Save Breakpoints
4808 @subsection How to save breakpoints to a file
4810 To save breakpoint definitions to a file use the @w{@code{save
4811 breakpoints}} command.
4814 @kindex save breakpoints
4815 @cindex save breakpoints to a file for future sessions
4816 @item save breakpoints [@var{filename}]
4817 This command saves all current breakpoint definitions together with
4818 their commands and ignore counts, into a file @file{@var{filename}}
4819 suitable for use in a later debugging session. This includes all
4820 types of breakpoints (breakpoints, watchpoints, catchpoints,
4821 tracepoints). To read the saved breakpoint definitions, use the
4822 @code{source} command (@pxref{Command Files}). Note that watchpoints
4823 with expressions involving local variables may fail to be recreated
4824 because it may not be possible to access the context where the
4825 watchpoint is valid anymore. Because the saved breakpoint definitions
4826 are simply a sequence of @value{GDBN} commands that recreate the
4827 breakpoints, you can edit the file in your favorite editing program,
4828 and remove the breakpoint definitions you're not interested in, or
4829 that can no longer be recreated.
4832 @node Static Probe Points
4833 @subsection Static Probe Points
4835 @cindex static probe point, SystemTap
4836 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
4837 for Statically Defined Tracing, and the probes are designed to have a tiny
4838 runtime code and data footprint, and no dynamic relocations. They are
4839 usable from assembly, C and C@t{++} languages. See
4840 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
4841 for a good reference on how the @acronym{SDT} probes are implemented.
4843 Currently, @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
4844 @acronym{SDT} probes are supported on ELF-compatible systems. See
4845 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
4846 for more information on how to add @code{SystemTap} @acronym{SDT} probes
4847 in your applications.
4849 @cindex semaphores on static probe points
4850 Some probes have an associated semaphore variable; for instance, this
4851 happens automatically if you defined your probe using a DTrace-style
4852 @file{.d} file. If your probe has a semaphore, @value{GDBN} will
4853 automatically enable it when you specify a breakpoint using the
4854 @samp{-probe-stap} notation. But, if you put a breakpoint at a probe's
4855 location by some other method (e.g., @code{break file:line}), then
4856 @value{GDBN} will not automatically set the semaphore.
4858 You can examine the available static static probes using @code{info
4859 probes}, with optional arguments:
4863 @item info probes stap @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
4864 If given, @var{provider} is a regular expression used to match against provider
4865 names when selecting which probes to list. If omitted, probes by all
4866 probes from all providers are listed.
4868 If given, @var{name} is a regular expression to match against probe names
4869 when selecting which probes to list. If omitted, probe names are not
4870 considered when deciding whether to display them.
4872 If given, @var{objfile} is a regular expression used to select which
4873 object files (executable or shared libraries) to examine. If not
4874 given, all object files are considered.
4876 @item info probes all
4877 List the available static probes, from all types.
4880 @vindex $_probe_arg@r{, convenience variable}
4881 A probe may specify up to twelve arguments. These are available at the
4882 point at which the probe is defined---that is, when the current PC is
4883 at the probe's location. The arguments are available using the
4884 convenience variables (@pxref{Convenience Vars})
4885 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. Each probe argument is
4886 an integer of the appropriate size; types are not preserved. The
4887 convenience variable @code{$_probe_argc} holds the number of arguments
4888 at the current probe point.
4890 These variables are always available, but attempts to access them at
4891 any location other than a probe point will cause @value{GDBN} to give
4895 @c @ifclear BARETARGET
4896 @node Error in Breakpoints
4897 @subsection ``Cannot insert breakpoints''
4899 If you request too many active hardware-assisted breakpoints and
4900 watchpoints, you will see this error message:
4902 @c FIXME: the precise wording of this message may change; the relevant
4903 @c source change is not committed yet (Sep 3, 1999).
4905 Stopped; cannot insert breakpoints.
4906 You may have requested too many hardware breakpoints and watchpoints.
4910 This message is printed when you attempt to resume the program, since
4911 only then @value{GDBN} knows exactly how many hardware breakpoints and
4912 watchpoints it needs to insert.
4914 When this message is printed, you need to disable or remove some of the
4915 hardware-assisted breakpoints and watchpoints, and then continue.
4917 @node Breakpoint-related Warnings
4918 @subsection ``Breakpoint address adjusted...''
4919 @cindex breakpoint address adjusted
4921 Some processor architectures place constraints on the addresses at
4922 which breakpoints may be placed. For architectures thus constrained,
4923 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4924 with the constraints dictated by the architecture.
4926 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4927 a VLIW architecture in which a number of RISC-like instructions may be
4928 bundled together for parallel execution. The FR-V architecture
4929 constrains the location of a breakpoint instruction within such a
4930 bundle to the instruction with the lowest address. @value{GDBN}
4931 honors this constraint by adjusting a breakpoint's address to the
4932 first in the bundle.
4934 It is not uncommon for optimized code to have bundles which contain
4935 instructions from different source statements, thus it may happen that
4936 a breakpoint's address will be adjusted from one source statement to
4937 another. Since this adjustment may significantly alter @value{GDBN}'s
4938 breakpoint related behavior from what the user expects, a warning is
4939 printed when the breakpoint is first set and also when the breakpoint
4942 A warning like the one below is printed when setting a breakpoint
4943 that's been subject to address adjustment:
4946 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4949 Such warnings are printed both for user settable and @value{GDBN}'s
4950 internal breakpoints. If you see one of these warnings, you should
4951 verify that a breakpoint set at the adjusted address will have the
4952 desired affect. If not, the breakpoint in question may be removed and
4953 other breakpoints may be set which will have the desired behavior.
4954 E.g., it may be sufficient to place the breakpoint at a later
4955 instruction. A conditional breakpoint may also be useful in some
4956 cases to prevent the breakpoint from triggering too often.
4958 @value{GDBN} will also issue a warning when stopping at one of these
4959 adjusted breakpoints:
4962 warning: Breakpoint 1 address previously adjusted from 0x00010414
4966 When this warning is encountered, it may be too late to take remedial
4967 action except in cases where the breakpoint is hit earlier or more
4968 frequently than expected.
4970 @node Continuing and Stepping
4971 @section Continuing and Stepping
4975 @cindex resuming execution
4976 @dfn{Continuing} means resuming program execution until your program
4977 completes normally. In contrast, @dfn{stepping} means executing just
4978 one more ``step'' of your program, where ``step'' may mean either one
4979 line of source code, or one machine instruction (depending on what
4980 particular command you use). Either when continuing or when stepping,
4981 your program may stop even sooner, due to a breakpoint or a signal. (If
4982 it stops due to a signal, you may want to use @code{handle}, or use
4983 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4987 @kindex c @r{(@code{continue})}
4988 @kindex fg @r{(resume foreground execution)}
4989 @item continue @r{[}@var{ignore-count}@r{]}
4990 @itemx c @r{[}@var{ignore-count}@r{]}
4991 @itemx fg @r{[}@var{ignore-count}@r{]}
4992 Resume program execution, at the address where your program last stopped;
4993 any breakpoints set at that address are bypassed. The optional argument
4994 @var{ignore-count} allows you to specify a further number of times to
4995 ignore a breakpoint at this location; its effect is like that of
4996 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4998 The argument @var{ignore-count} is meaningful only when your program
4999 stopped due to a breakpoint. At other times, the argument to
5000 @code{continue} is ignored.
5002 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5003 debugged program is deemed to be the foreground program) are provided
5004 purely for convenience, and have exactly the same behavior as
5008 To resume execution at a different place, you can use @code{return}
5009 (@pxref{Returning, ,Returning from a Function}) to go back to the
5010 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5011 Different Address}) to go to an arbitrary location in your program.
5013 A typical technique for using stepping is to set a breakpoint
5014 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5015 beginning of the function or the section of your program where a problem
5016 is believed to lie, run your program until it stops at that breakpoint,
5017 and then step through the suspect area, examining the variables that are
5018 interesting, until you see the problem happen.
5022 @kindex s @r{(@code{step})}
5024 Continue running your program until control reaches a different source
5025 line, then stop it and return control to @value{GDBN}. This command is
5026 abbreviated @code{s}.
5029 @c "without debugging information" is imprecise; actually "without line
5030 @c numbers in the debugging information". (gcc -g1 has debugging info but
5031 @c not line numbers). But it seems complex to try to make that
5032 @c distinction here.
5033 @emph{Warning:} If you use the @code{step} command while control is
5034 within a function that was compiled without debugging information,
5035 execution proceeds until control reaches a function that does have
5036 debugging information. Likewise, it will not step into a function which
5037 is compiled without debugging information. To step through functions
5038 without debugging information, use the @code{stepi} command, described
5042 The @code{step} command only stops at the first instruction of a source
5043 line. This prevents the multiple stops that could otherwise occur in
5044 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5045 to stop if a function that has debugging information is called within
5046 the line. In other words, @code{step} @emph{steps inside} any functions
5047 called within the line.
5049 Also, the @code{step} command only enters a function if there is line
5050 number information for the function. Otherwise it acts like the
5051 @code{next} command. This avoids problems when using @code{cc -gl}
5052 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5053 was any debugging information about the routine.
5055 @item step @var{count}
5056 Continue running as in @code{step}, but do so @var{count} times. If a
5057 breakpoint is reached, or a signal not related to stepping occurs before
5058 @var{count} steps, stepping stops right away.
5061 @kindex n @r{(@code{next})}
5062 @item next @r{[}@var{count}@r{]}
5063 Continue to the next source line in the current (innermost) stack frame.
5064 This is similar to @code{step}, but function calls that appear within
5065 the line of code are executed without stopping. Execution stops when
5066 control reaches a different line of code at the original stack level
5067 that was executing when you gave the @code{next} command. This command
5068 is abbreviated @code{n}.
5070 An argument @var{count} is a repeat count, as for @code{step}.
5073 @c FIX ME!! Do we delete this, or is there a way it fits in with
5074 @c the following paragraph? --- Vctoria
5076 @c @code{next} within a function that lacks debugging information acts like
5077 @c @code{step}, but any function calls appearing within the code of the
5078 @c function are executed without stopping.
5080 The @code{next} command only stops at the first instruction of a
5081 source line. This prevents multiple stops that could otherwise occur in
5082 @code{switch} statements, @code{for} loops, etc.
5084 @kindex set step-mode
5086 @cindex functions without line info, and stepping
5087 @cindex stepping into functions with no line info
5088 @itemx set step-mode on
5089 The @code{set step-mode on} command causes the @code{step} command to
5090 stop at the first instruction of a function which contains no debug line
5091 information rather than stepping over it.
5093 This is useful in cases where you may be interested in inspecting the
5094 machine instructions of a function which has no symbolic info and do not
5095 want @value{GDBN} to automatically skip over this function.
5097 @item set step-mode off
5098 Causes the @code{step} command to step over any functions which contains no
5099 debug information. This is the default.
5101 @item show step-mode
5102 Show whether @value{GDBN} will stop in or step over functions without
5103 source line debug information.
5106 @kindex fin @r{(@code{finish})}
5108 Continue running until just after function in the selected stack frame
5109 returns. Print the returned value (if any). This command can be
5110 abbreviated as @code{fin}.
5112 Contrast this with the @code{return} command (@pxref{Returning,
5113 ,Returning from a Function}).
5116 @kindex u @r{(@code{until})}
5117 @cindex run until specified location
5120 Continue running until a source line past the current line, in the
5121 current stack frame, is reached. This command is used to avoid single
5122 stepping through a loop more than once. It is like the @code{next}
5123 command, except that when @code{until} encounters a jump, it
5124 automatically continues execution until the program counter is greater
5125 than the address of the jump.
5127 This means that when you reach the end of a loop after single stepping
5128 though it, @code{until} makes your program continue execution until it
5129 exits the loop. In contrast, a @code{next} command at the end of a loop
5130 simply steps back to the beginning of the loop, which forces you to step
5131 through the next iteration.
5133 @code{until} always stops your program if it attempts to exit the current
5136 @code{until} may produce somewhat counterintuitive results if the order
5137 of machine code does not match the order of the source lines. For
5138 example, in the following excerpt from a debugging session, the @code{f}
5139 (@code{frame}) command shows that execution is stopped at line
5140 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5144 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5146 (@value{GDBP}) until
5147 195 for ( ; argc > 0; NEXTARG) @{
5150 This happened because, for execution efficiency, the compiler had
5151 generated code for the loop closure test at the end, rather than the
5152 start, of the loop---even though the test in a C @code{for}-loop is
5153 written before the body of the loop. The @code{until} command appeared
5154 to step back to the beginning of the loop when it advanced to this
5155 expression; however, it has not really gone to an earlier
5156 statement---not in terms of the actual machine code.
5158 @code{until} with no argument works by means of single
5159 instruction stepping, and hence is slower than @code{until} with an
5162 @item until @var{location}
5163 @itemx u @var{location}
5164 Continue running your program until either the specified location is
5165 reached, or the current stack frame returns. @var{location} is any of
5166 the forms described in @ref{Specify Location}.
5167 This form of the command uses temporary breakpoints, and
5168 hence is quicker than @code{until} without an argument. The specified
5169 location is actually reached only if it is in the current frame. This
5170 implies that @code{until} can be used to skip over recursive function
5171 invocations. For instance in the code below, if the current location is
5172 line @code{96}, issuing @code{until 99} will execute the program up to
5173 line @code{99} in the same invocation of factorial, i.e., after the inner
5174 invocations have returned.
5177 94 int factorial (int value)
5179 96 if (value > 1) @{
5180 97 value *= factorial (value - 1);
5187 @kindex advance @var{location}
5188 @item advance @var{location}
5189 Continue running the program up to the given @var{location}. An argument is
5190 required, which should be of one of the forms described in
5191 @ref{Specify Location}.
5192 Execution will also stop upon exit from the current stack
5193 frame. This command is similar to @code{until}, but @code{advance} will
5194 not skip over recursive function calls, and the target location doesn't
5195 have to be in the same frame as the current one.
5199 @kindex si @r{(@code{stepi})}
5201 @itemx stepi @var{arg}
5203 Execute one machine instruction, then stop and return to the debugger.
5205 It is often useful to do @samp{display/i $pc} when stepping by machine
5206 instructions. This makes @value{GDBN} automatically display the next
5207 instruction to be executed, each time your program stops. @xref{Auto
5208 Display,, Automatic Display}.
5210 An argument is a repeat count, as in @code{step}.
5214 @kindex ni @r{(@code{nexti})}
5216 @itemx nexti @var{arg}
5218 Execute one machine instruction, but if it is a function call,
5219 proceed until the function returns.
5221 An argument is a repeat count, as in @code{next}.
5224 @node Skipping Over Functions and Files
5225 @section Skipping Over Functions and Files
5226 @cindex skipping over functions and files
5228 The program you are debugging may contain some functions which are
5229 uninteresting to debug. The @code{skip} comand lets you tell @value{GDBN} to
5230 skip a function or all functions in a file when stepping.
5232 For example, consider the following C function:
5243 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5244 are not interested in stepping through @code{boring}. If you run @code{step}
5245 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5246 step over both @code{foo} and @code{boring}!
5248 One solution is to @code{step} into @code{boring} and use the @code{finish}
5249 command to immediately exit it. But this can become tedious if @code{boring}
5250 is called from many places.
5252 A more flexible solution is to execute @kbd{skip boring}. This instructs
5253 @value{GDBN} never to step into @code{boring}. Now when you execute
5254 @code{step} at line 103, you'll step over @code{boring} and directly into
5257 You can also instruct @value{GDBN} to skip all functions in a file, with, for
5258 example, @code{skip file boring.c}.
5261 @kindex skip function
5262 @item skip @r{[}@var{linespec}@r{]}
5263 @itemx skip function @r{[}@var{linespec}@r{]}
5264 After running this command, the function named by @var{linespec} or the
5265 function containing the line named by @var{linespec} will be skipped over when
5266 stepping. @xref{Specify Location}.
5268 If you do not specify @var{linespec}, the function you're currently debugging
5271 (If you have a function called @code{file} that you want to skip, use
5272 @kbd{skip function file}.)
5275 @item skip file @r{[}@var{filename}@r{]}
5276 After running this command, any function whose source lives in @var{filename}
5277 will be skipped over when stepping.
5279 If you do not specify @var{filename}, functions whose source lives in the file
5280 you're currently debugging will be skipped.
5283 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5284 These are the commands for managing your list of skips:
5288 @item info skip @r{[}@var{range}@r{]}
5289 Print details about the specified skip(s). If @var{range} is not specified,
5290 print a table with details about all functions and files marked for skipping.
5291 @code{info skip} prints the following information about each skip:
5295 A number identifying this skip.
5297 The type of this skip, either @samp{function} or @samp{file}.
5298 @item Enabled or Disabled
5299 Enabled skips are marked with @samp{y}. Disabled skips are marked with @samp{n}.
5301 For function skips, this column indicates the address in memory of the function
5302 being skipped. If you've set a function skip on a function which has not yet
5303 been loaded, this field will contain @samp{<PENDING>}. Once a shared library
5304 which has the function is loaded, @code{info skip} will show the function's
5307 For file skips, this field contains the filename being skipped. For functions
5308 skips, this field contains the function name and its line number in the file
5309 where it is defined.
5313 @item skip delete @r{[}@var{range}@r{]}
5314 Delete the specified skip(s). If @var{range} is not specified, delete all
5318 @item skip enable @r{[}@var{range}@r{]}
5319 Enable the specified skip(s). If @var{range} is not specified, enable all
5322 @kindex skip disable
5323 @item skip disable @r{[}@var{range}@r{]}
5324 Disable the specified skip(s). If @var{range} is not specified, disable all
5333 A signal is an asynchronous event that can happen in a program. The
5334 operating system defines the possible kinds of signals, and gives each
5335 kind a name and a number. For example, in Unix @code{SIGINT} is the
5336 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
5337 @code{SIGSEGV} is the signal a program gets from referencing a place in
5338 memory far away from all the areas in use; @code{SIGALRM} occurs when
5339 the alarm clock timer goes off (which happens only if your program has
5340 requested an alarm).
5342 @cindex fatal signals
5343 Some signals, including @code{SIGALRM}, are a normal part of the
5344 functioning of your program. Others, such as @code{SIGSEGV}, indicate
5345 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
5346 program has not specified in advance some other way to handle the signal.
5347 @code{SIGINT} does not indicate an error in your program, but it is normally
5348 fatal so it can carry out the purpose of the interrupt: to kill the program.
5350 @value{GDBN} has the ability to detect any occurrence of a signal in your
5351 program. You can tell @value{GDBN} in advance what to do for each kind of
5354 @cindex handling signals
5355 Normally, @value{GDBN} is set up to let the non-erroneous signals like
5356 @code{SIGALRM} be silently passed to your program
5357 (so as not to interfere with their role in the program's functioning)
5358 but to stop your program immediately whenever an error signal happens.
5359 You can change these settings with the @code{handle} command.
5362 @kindex info signals
5366 Print a table of all the kinds of signals and how @value{GDBN} has been told to
5367 handle each one. You can use this to see the signal numbers of all
5368 the defined types of signals.
5370 @item info signals @var{sig}
5371 Similar, but print information only about the specified signal number.
5373 @code{info handle} is an alias for @code{info signals}.
5375 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5376 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
5377 for details about this command.
5380 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
5381 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
5382 can be the number of a signal or its name (with or without the
5383 @samp{SIG} at the beginning); a list of signal numbers of the form
5384 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
5385 known signals. Optional arguments @var{keywords}, described below,
5386 say what change to make.
5390 The keywords allowed by the @code{handle} command can be abbreviated.
5391 Their full names are:
5395 @value{GDBN} should not stop your program when this signal happens. It may
5396 still print a message telling you that the signal has come in.
5399 @value{GDBN} should stop your program when this signal happens. This implies
5400 the @code{print} keyword as well.
5403 @value{GDBN} should print a message when this signal happens.
5406 @value{GDBN} should not mention the occurrence of the signal at all. This
5407 implies the @code{nostop} keyword as well.
5411 @value{GDBN} should allow your program to see this signal; your program
5412 can handle the signal, or else it may terminate if the signal is fatal
5413 and not handled. @code{pass} and @code{noignore} are synonyms.
5417 @value{GDBN} should not allow your program to see this signal.
5418 @code{nopass} and @code{ignore} are synonyms.
5422 When a signal stops your program, the signal is not visible to the
5424 continue. Your program sees the signal then, if @code{pass} is in
5425 effect for the signal in question @emph{at that time}. In other words,
5426 after @value{GDBN} reports a signal, you can use the @code{handle}
5427 command with @code{pass} or @code{nopass} to control whether your
5428 program sees that signal when you continue.
5430 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
5431 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
5432 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
5435 You can also use the @code{signal} command to prevent your program from
5436 seeing a signal, or cause it to see a signal it normally would not see,
5437 or to give it any signal at any time. For example, if your program stopped
5438 due to some sort of memory reference error, you might store correct
5439 values into the erroneous variables and continue, hoping to see more
5440 execution; but your program would probably terminate immediately as
5441 a result of the fatal signal once it saw the signal. To prevent this,
5442 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
5445 @cindex extra signal information
5446 @anchor{extra signal information}
5448 On some targets, @value{GDBN} can inspect extra signal information
5449 associated with the intercepted signal, before it is actually
5450 delivered to the program being debugged. This information is exported
5451 by the convenience variable @code{$_siginfo}, and consists of data
5452 that is passed by the kernel to the signal handler at the time of the
5453 receipt of a signal. The data type of the information itself is
5454 target dependent. You can see the data type using the @code{ptype
5455 $_siginfo} command. On Unix systems, it typically corresponds to the
5456 standard @code{siginfo_t} type, as defined in the @file{signal.h}
5459 Here's an example, on a @sc{gnu}/Linux system, printing the stray
5460 referenced address that raised a segmentation fault.
5464 (@value{GDBP}) continue
5465 Program received signal SIGSEGV, Segmentation fault.
5466 0x0000000000400766 in main ()
5468 (@value{GDBP}) ptype $_siginfo
5475 struct @{...@} _kill;
5476 struct @{...@} _timer;
5478 struct @{...@} _sigchld;
5479 struct @{...@} _sigfault;
5480 struct @{...@} _sigpoll;
5483 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5487 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5488 $1 = (void *) 0x7ffff7ff7000
5492 Depending on target support, @code{$_siginfo} may also be writable.
5495 @section Stopping and Starting Multi-thread Programs
5497 @cindex stopped threads
5498 @cindex threads, stopped
5500 @cindex continuing threads
5501 @cindex threads, continuing
5503 @value{GDBN} supports debugging programs with multiple threads
5504 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
5505 are two modes of controlling execution of your program within the
5506 debugger. In the default mode, referred to as @dfn{all-stop mode},
5507 when any thread in your program stops (for example, at a breakpoint
5508 or while being stepped), all other threads in the program are also stopped by
5509 @value{GDBN}. On some targets, @value{GDBN} also supports
5510 @dfn{non-stop mode}, in which other threads can continue to run freely while
5511 you examine the stopped thread in the debugger.
5514 * All-Stop Mode:: All threads stop when GDB takes control
5515 * Non-Stop Mode:: Other threads continue to execute
5516 * Background Execution:: Running your program asynchronously
5517 * Thread-Specific Breakpoints:: Controlling breakpoints
5518 * Interrupted System Calls:: GDB may interfere with system calls
5519 * Observer Mode:: GDB does not alter program behavior
5523 @subsection All-Stop Mode
5525 @cindex all-stop mode
5527 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
5528 @emph{all} threads of execution stop, not just the current thread. This
5529 allows you to examine the overall state of the program, including
5530 switching between threads, without worrying that things may change
5533 Conversely, whenever you restart the program, @emph{all} threads start
5534 executing. @emph{This is true even when single-stepping} with commands
5535 like @code{step} or @code{next}.
5537 In particular, @value{GDBN} cannot single-step all threads in lockstep.
5538 Since thread scheduling is up to your debugging target's operating
5539 system (not controlled by @value{GDBN}), other threads may
5540 execute more than one statement while the current thread completes a
5541 single step. Moreover, in general other threads stop in the middle of a
5542 statement, rather than at a clean statement boundary, when the program
5545 You might even find your program stopped in another thread after
5546 continuing or even single-stepping. This happens whenever some other
5547 thread runs into a breakpoint, a signal, or an exception before the
5548 first thread completes whatever you requested.
5550 @cindex automatic thread selection
5551 @cindex switching threads automatically
5552 @cindex threads, automatic switching
5553 Whenever @value{GDBN} stops your program, due to a breakpoint or a
5554 signal, it automatically selects the thread where that breakpoint or
5555 signal happened. @value{GDBN} alerts you to the context switch with a
5556 message such as @samp{[Switching to Thread @var{n}]} to identify the
5559 On some OSes, you can modify @value{GDBN}'s default behavior by
5560 locking the OS scheduler to allow only a single thread to run.
5563 @item set scheduler-locking @var{mode}
5564 @cindex scheduler locking mode
5565 @cindex lock scheduler
5566 Set the scheduler locking mode. If it is @code{off}, then there is no
5567 locking and any thread may run at any time. If @code{on}, then only the
5568 current thread may run when the inferior is resumed. The @code{step}
5569 mode optimizes for single-stepping; it prevents other threads
5570 from preempting the current thread while you are stepping, so that
5571 the focus of debugging does not change unexpectedly.
5572 Other threads only rarely (or never) get a chance to run
5573 when you step. They are more likely to run when you @samp{next} over a
5574 function call, and they are completely free to run when you use commands
5575 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
5576 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
5577 the current thread away from the thread that you are debugging.
5579 @item show scheduler-locking
5580 Display the current scheduler locking mode.
5583 @cindex resume threads of multiple processes simultaneously
5584 By default, when you issue one of the execution commands such as
5585 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5586 threads of the current inferior to run. For example, if @value{GDBN}
5587 is attached to two inferiors, each with two threads, the
5588 @code{continue} command resumes only the two threads of the current
5589 inferior. This is useful, for example, when you debug a program that
5590 forks and you want to hold the parent stopped (so that, for instance,
5591 it doesn't run to exit), while you debug the child. In other
5592 situations, you may not be interested in inspecting the current state
5593 of any of the processes @value{GDBN} is attached to, and you may want
5594 to resume them all until some breakpoint is hit. In the latter case,
5595 you can instruct @value{GDBN} to allow all threads of all the
5596 inferiors to run with the @w{@code{set schedule-multiple}} command.
5599 @kindex set schedule-multiple
5600 @item set schedule-multiple
5601 Set the mode for allowing threads of multiple processes to be resumed
5602 when an execution command is issued. When @code{on}, all threads of
5603 all processes are allowed to run. When @code{off}, only the threads
5604 of the current process are resumed. The default is @code{off}. The
5605 @code{scheduler-locking} mode takes precedence when set to @code{on},
5606 or while you are stepping and set to @code{step}.
5608 @item show schedule-multiple
5609 Display the current mode for resuming the execution of threads of
5614 @subsection Non-Stop Mode
5616 @cindex non-stop mode
5618 @c This section is really only a place-holder, and needs to be expanded
5619 @c with more details.
5621 For some multi-threaded targets, @value{GDBN} supports an optional
5622 mode of operation in which you can examine stopped program threads in
5623 the debugger while other threads continue to execute freely. This
5624 minimizes intrusion when debugging live systems, such as programs
5625 where some threads have real-time constraints or must continue to
5626 respond to external events. This is referred to as @dfn{non-stop} mode.
5628 In non-stop mode, when a thread stops to report a debugging event,
5629 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5630 threads as well, in contrast to the all-stop mode behavior. Additionally,
5631 execution commands such as @code{continue} and @code{step} apply by default
5632 only to the current thread in non-stop mode, rather than all threads as
5633 in all-stop mode. This allows you to control threads explicitly in
5634 ways that are not possible in all-stop mode --- for example, stepping
5635 one thread while allowing others to run freely, stepping
5636 one thread while holding all others stopped, or stepping several threads
5637 independently and simultaneously.
5639 To enter non-stop mode, use this sequence of commands before you run
5640 or attach to your program:
5643 # Enable the async interface.
5646 # If using the CLI, pagination breaks non-stop.
5649 # Finally, turn it on!
5653 You can use these commands to manipulate the non-stop mode setting:
5656 @kindex set non-stop
5657 @item set non-stop on
5658 Enable selection of non-stop mode.
5659 @item set non-stop off
5660 Disable selection of non-stop mode.
5661 @kindex show non-stop
5663 Show the current non-stop enablement setting.
5666 Note these commands only reflect whether non-stop mode is enabled,
5667 not whether the currently-executing program is being run in non-stop mode.
5668 In particular, the @code{set non-stop} preference is only consulted when
5669 @value{GDBN} starts or connects to the target program, and it is generally
5670 not possible to switch modes once debugging has started. Furthermore,
5671 since not all targets support non-stop mode, even when you have enabled
5672 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5675 In non-stop mode, all execution commands apply only to the current thread
5676 by default. That is, @code{continue} only continues one thread.
5677 To continue all threads, issue @code{continue -a} or @code{c -a}.
5679 You can use @value{GDBN}'s background execution commands
5680 (@pxref{Background Execution}) to run some threads in the background
5681 while you continue to examine or step others from @value{GDBN}.
5682 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5683 always executed asynchronously in non-stop mode.
5685 Suspending execution is done with the @code{interrupt} command when
5686 running in the background, or @kbd{Ctrl-c} during foreground execution.
5687 In all-stop mode, this stops the whole process;
5688 but in non-stop mode the interrupt applies only to the current thread.
5689 To stop the whole program, use @code{interrupt -a}.
5691 Other execution commands do not currently support the @code{-a} option.
5693 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5694 that thread current, as it does in all-stop mode. This is because the
5695 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5696 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5697 changed to a different thread just as you entered a command to operate on the
5698 previously current thread.
5700 @node Background Execution
5701 @subsection Background Execution
5703 @cindex foreground execution
5704 @cindex background execution
5705 @cindex asynchronous execution
5706 @cindex execution, foreground, background and asynchronous
5708 @value{GDBN}'s execution commands have two variants: the normal
5709 foreground (synchronous) behavior, and a background
5710 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5711 the program to report that some thread has stopped before prompting for
5712 another command. In background execution, @value{GDBN} immediately gives
5713 a command prompt so that you can issue other commands while your program runs.
5715 You need to explicitly enable asynchronous mode before you can use
5716 background execution commands. You can use these commands to
5717 manipulate the asynchronous mode setting:
5720 @kindex set target-async
5721 @item set target-async on
5722 Enable asynchronous mode.
5723 @item set target-async off
5724 Disable asynchronous mode.
5725 @kindex show target-async
5726 @item show target-async
5727 Show the current target-async setting.
5730 If the target doesn't support async mode, @value{GDBN} issues an error
5731 message if you attempt to use the background execution commands.
5733 To specify background execution, add a @code{&} to the command. For example,
5734 the background form of the @code{continue} command is @code{continue&}, or
5735 just @code{c&}. The execution commands that accept background execution
5741 @xref{Starting, , Starting your Program}.
5745 @xref{Attach, , Debugging an Already-running Process}.
5749 @xref{Continuing and Stepping, step}.
5753 @xref{Continuing and Stepping, stepi}.
5757 @xref{Continuing and Stepping, next}.
5761 @xref{Continuing and Stepping, nexti}.
5765 @xref{Continuing and Stepping, continue}.
5769 @xref{Continuing and Stepping, finish}.
5773 @xref{Continuing and Stepping, until}.
5777 Background execution is especially useful in conjunction with non-stop
5778 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5779 However, you can also use these commands in the normal all-stop mode with
5780 the restriction that you cannot issue another execution command until the
5781 previous one finishes. Examples of commands that are valid in all-stop
5782 mode while the program is running include @code{help} and @code{info break}.
5784 You can interrupt your program while it is running in the background by
5785 using the @code{interrupt} command.
5792 Suspend execution of the running program. In all-stop mode,
5793 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5794 only the current thread. To stop the whole program in non-stop mode,
5795 use @code{interrupt -a}.
5798 @node Thread-Specific Breakpoints
5799 @subsection Thread-Specific Breakpoints
5801 When your program has multiple threads (@pxref{Threads,, Debugging
5802 Programs with Multiple Threads}), you can choose whether to set
5803 breakpoints on all threads, or on a particular thread.
5806 @cindex breakpoints and threads
5807 @cindex thread breakpoints
5808 @kindex break @dots{} thread @var{threadno}
5809 @item break @var{linespec} thread @var{threadno}
5810 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5811 @var{linespec} specifies source lines; there are several ways of
5812 writing them (@pxref{Specify Location}), but the effect is always to
5813 specify some source line.
5815 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5816 to specify that you only want @value{GDBN} to stop the program when a
5817 particular thread reaches this breakpoint. @var{threadno} is one of the
5818 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5819 column of the @samp{info threads} display.
5821 If you do not specify @samp{thread @var{threadno}} when you set a
5822 breakpoint, the breakpoint applies to @emph{all} threads of your
5825 You can use the @code{thread} qualifier on conditional breakpoints as
5826 well; in this case, place @samp{thread @var{threadno}} before or
5827 after the breakpoint condition, like this:
5830 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5835 @node Interrupted System Calls
5836 @subsection Interrupted System Calls
5838 @cindex thread breakpoints and system calls
5839 @cindex system calls and thread breakpoints
5840 @cindex premature return from system calls
5841 There is an unfortunate side effect when using @value{GDBN} to debug
5842 multi-threaded programs. If one thread stops for a
5843 breakpoint, or for some other reason, and another thread is blocked in a
5844 system call, then the system call may return prematurely. This is a
5845 consequence of the interaction between multiple threads and the signals
5846 that @value{GDBN} uses to implement breakpoints and other events that
5849 To handle this problem, your program should check the return value of
5850 each system call and react appropriately. This is good programming
5853 For example, do not write code like this:
5859 The call to @code{sleep} will return early if a different thread stops
5860 at a breakpoint or for some other reason.
5862 Instead, write this:
5867 unslept = sleep (unslept);
5870 A system call is allowed to return early, so the system is still
5871 conforming to its specification. But @value{GDBN} does cause your
5872 multi-threaded program to behave differently than it would without
5875 Also, @value{GDBN} uses internal breakpoints in the thread library to
5876 monitor certain events such as thread creation and thread destruction.
5877 When such an event happens, a system call in another thread may return
5878 prematurely, even though your program does not appear to stop.
5881 @subsection Observer Mode
5883 If you want to build on non-stop mode and observe program behavior
5884 without any chance of disruption by @value{GDBN}, you can set
5885 variables to disable all of the debugger's attempts to modify state,
5886 whether by writing memory, inserting breakpoints, etc. These operate
5887 at a low level, intercepting operations from all commands.
5889 When all of these are set to @code{off}, then @value{GDBN} is said to
5890 be @dfn{observer mode}. As a convenience, the variable
5891 @code{observer} can be set to disable these, plus enable non-stop
5894 Note that @value{GDBN} will not prevent you from making nonsensical
5895 combinations of these settings. For instance, if you have enabled
5896 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
5897 then breakpoints that work by writing trap instructions into the code
5898 stream will still not be able to be placed.
5903 @item set observer on
5904 @itemx set observer off
5905 When set to @code{on}, this disables all the permission variables
5906 below (except for @code{insert-fast-tracepoints}), plus enables
5907 non-stop debugging. Setting this to @code{off} switches back to
5908 normal debugging, though remaining in non-stop mode.
5911 Show whether observer mode is on or off.
5913 @kindex may-write-registers
5914 @item set may-write-registers on
5915 @itemx set may-write-registers off
5916 This controls whether @value{GDBN} will attempt to alter the values of
5917 registers, such as with assignment expressions in @code{print}, or the
5918 @code{jump} command. It defaults to @code{on}.
5920 @item show may-write-registers
5921 Show the current permission to write registers.
5923 @kindex may-write-memory
5924 @item set may-write-memory on
5925 @itemx set may-write-memory off
5926 This controls whether @value{GDBN} will attempt to alter the contents
5927 of memory, such as with assignment expressions in @code{print}. It
5928 defaults to @code{on}.
5930 @item show may-write-memory
5931 Show the current permission to write memory.
5933 @kindex may-insert-breakpoints
5934 @item set may-insert-breakpoints on
5935 @itemx set may-insert-breakpoints off
5936 This controls whether @value{GDBN} will attempt to insert breakpoints.
5937 This affects all breakpoints, including internal breakpoints defined
5938 by @value{GDBN}. It defaults to @code{on}.
5940 @item show may-insert-breakpoints
5941 Show the current permission to insert breakpoints.
5943 @kindex may-insert-tracepoints
5944 @item set may-insert-tracepoints on
5945 @itemx set may-insert-tracepoints off
5946 This controls whether @value{GDBN} will attempt to insert (regular)
5947 tracepoints at the beginning of a tracing experiment. It affects only
5948 non-fast tracepoints, fast tracepoints being under the control of
5949 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
5951 @item show may-insert-tracepoints
5952 Show the current permission to insert tracepoints.
5954 @kindex may-insert-fast-tracepoints
5955 @item set may-insert-fast-tracepoints on
5956 @itemx set may-insert-fast-tracepoints off
5957 This controls whether @value{GDBN} will attempt to insert fast
5958 tracepoints at the beginning of a tracing experiment. It affects only
5959 fast tracepoints, regular (non-fast) tracepoints being under the
5960 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
5962 @item show may-insert-fast-tracepoints
5963 Show the current permission to insert fast tracepoints.
5965 @kindex may-interrupt
5966 @item set may-interrupt on
5967 @itemx set may-interrupt off
5968 This controls whether @value{GDBN} will attempt to interrupt or stop
5969 program execution. When this variable is @code{off}, the
5970 @code{interrupt} command will have no effect, nor will
5971 @kbd{Ctrl-c}. It defaults to @code{on}.
5973 @item show may-interrupt
5974 Show the current permission to interrupt or stop the program.
5978 @node Reverse Execution
5979 @chapter Running programs backward
5980 @cindex reverse execution
5981 @cindex running programs backward
5983 When you are debugging a program, it is not unusual to realize that
5984 you have gone too far, and some event of interest has already happened.
5985 If the target environment supports it, @value{GDBN} can allow you to
5986 ``rewind'' the program by running it backward.
5988 A target environment that supports reverse execution should be able
5989 to ``undo'' the changes in machine state that have taken place as the
5990 program was executing normally. Variables, registers etc.@: should
5991 revert to their previous values. Obviously this requires a great
5992 deal of sophistication on the part of the target environment; not
5993 all target environments can support reverse execution.
5995 When a program is executed in reverse, the instructions that
5996 have most recently been executed are ``un-executed'', in reverse
5997 order. The program counter runs backward, following the previous
5998 thread of execution in reverse. As each instruction is ``un-executed'',
5999 the values of memory and/or registers that were changed by that
6000 instruction are reverted to their previous states. After executing
6001 a piece of source code in reverse, all side effects of that code
6002 should be ``undone'', and all variables should be returned to their
6003 prior values@footnote{
6004 Note that some side effects are easier to undo than others. For instance,
6005 memory and registers are relatively easy, but device I/O is hard. Some
6006 targets may be able undo things like device I/O, and some may not.
6008 The contract between @value{GDBN} and the reverse executing target
6009 requires only that the target do something reasonable when
6010 @value{GDBN} tells it to execute backwards, and then report the
6011 results back to @value{GDBN}. Whatever the target reports back to
6012 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6013 assumes that the memory and registers that the target reports are in a
6014 consistant state, but @value{GDBN} accepts whatever it is given.
6017 If you are debugging in a target environment that supports
6018 reverse execution, @value{GDBN} provides the following commands.
6021 @kindex reverse-continue
6022 @kindex rc @r{(@code{reverse-continue})}
6023 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6024 @itemx rc @r{[}@var{ignore-count}@r{]}
6025 Beginning at the point where your program last stopped, start executing
6026 in reverse. Reverse execution will stop for breakpoints and synchronous
6027 exceptions (signals), just like normal execution. Behavior of
6028 asynchronous signals depends on the target environment.
6030 @kindex reverse-step
6031 @kindex rs @r{(@code{step})}
6032 @item reverse-step @r{[}@var{count}@r{]}
6033 Run the program backward until control reaches the start of a
6034 different source line; then stop it, and return control to @value{GDBN}.
6036 Like the @code{step} command, @code{reverse-step} will only stop
6037 at the beginning of a source line. It ``un-executes'' the previously
6038 executed source line. If the previous source line included calls to
6039 debuggable functions, @code{reverse-step} will step (backward) into
6040 the called function, stopping at the beginning of the @emph{last}
6041 statement in the called function (typically a return statement).
6043 Also, as with the @code{step} command, if non-debuggable functions are
6044 called, @code{reverse-step} will run thru them backward without stopping.
6046 @kindex reverse-stepi
6047 @kindex rsi @r{(@code{reverse-stepi})}
6048 @item reverse-stepi @r{[}@var{count}@r{]}
6049 Reverse-execute one machine instruction. Note that the instruction
6050 to be reverse-executed is @emph{not} the one pointed to by the program
6051 counter, but the instruction executed prior to that one. For instance,
6052 if the last instruction was a jump, @code{reverse-stepi} will take you
6053 back from the destination of the jump to the jump instruction itself.
6055 @kindex reverse-next
6056 @kindex rn @r{(@code{reverse-next})}
6057 @item reverse-next @r{[}@var{count}@r{]}
6058 Run backward to the beginning of the previous line executed in
6059 the current (innermost) stack frame. If the line contains function
6060 calls, they will be ``un-executed'' without stopping. Starting from
6061 the first line of a function, @code{reverse-next} will take you back
6062 to the caller of that function, @emph{before} the function was called,
6063 just as the normal @code{next} command would take you from the last
6064 line of a function back to its return to its caller
6065 @footnote{Unless the code is too heavily optimized.}.
6067 @kindex reverse-nexti
6068 @kindex rni @r{(@code{reverse-nexti})}
6069 @item reverse-nexti @r{[}@var{count}@r{]}
6070 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6071 in reverse, except that called functions are ``un-executed'' atomically.
6072 That is, if the previously executed instruction was a return from
6073 another function, @code{reverse-nexti} will continue to execute
6074 in reverse until the call to that function (from the current stack
6077 @kindex reverse-finish
6078 @item reverse-finish
6079 Just as the @code{finish} command takes you to the point where the
6080 current function returns, @code{reverse-finish} takes you to the point
6081 where it was called. Instead of ending up at the end of the current
6082 function invocation, you end up at the beginning.
6084 @kindex set exec-direction
6085 @item set exec-direction
6086 Set the direction of target execution.
6087 @item set exec-direction reverse
6088 @cindex execute forward or backward in time
6089 @value{GDBN} will perform all execution commands in reverse, until the
6090 exec-direction mode is changed to ``forward''. Affected commands include
6091 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6092 command cannot be used in reverse mode.
6093 @item set exec-direction forward
6094 @value{GDBN} will perform all execution commands in the normal fashion.
6095 This is the default.
6099 @node Process Record and Replay
6100 @chapter Recording Inferior's Execution and Replaying It
6101 @cindex process record and replay
6102 @cindex recording inferior's execution and replaying it
6104 On some platforms, @value{GDBN} provides a special @dfn{process record
6105 and replay} target that can record a log of the process execution, and
6106 replay it later with both forward and reverse execution commands.
6109 When this target is in use, if the execution log includes the record
6110 for the next instruction, @value{GDBN} will debug in @dfn{replay
6111 mode}. In the replay mode, the inferior does not really execute code
6112 instructions. Instead, all the events that normally happen during
6113 code execution are taken from the execution log. While code is not
6114 really executed in replay mode, the values of registers (including the
6115 program counter register) and the memory of the inferior are still
6116 changed as they normally would. Their contents are taken from the
6120 If the record for the next instruction is not in the execution log,
6121 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6122 inferior executes normally, and @value{GDBN} records the execution log
6125 The process record and replay target supports reverse execution
6126 (@pxref{Reverse Execution}), even if the platform on which the
6127 inferior runs does not. However, the reverse execution is limited in
6128 this case by the range of the instructions recorded in the execution
6129 log. In other words, reverse execution on platforms that don't
6130 support it directly can only be done in the replay mode.
6132 When debugging in the reverse direction, @value{GDBN} will work in
6133 replay mode as long as the execution log includes the record for the
6134 previous instruction; otherwise, it will work in record mode, if the
6135 platform supports reverse execution, or stop if not.
6137 For architecture environments that support process record and replay,
6138 @value{GDBN} provides the following commands:
6141 @kindex target record
6142 @kindex target record-full
6143 @kindex target record-btrace
6146 @kindex record btrace
6150 @item record @var{method}
6151 This command starts the process record and replay target. The
6152 recording method can be specified as parameter. Without a parameter
6153 the command uses the @code{full} recording method. The following
6154 recording methods are available:
6158 Full record/replay recording using @value{GDBN}'s software record and
6159 replay implementation. This method allows replaying and reverse
6163 Hardware-supported instruction recording. This method does not allow
6164 replaying and reverse execution.
6166 This recording method may not be available on all processors.
6169 The process record and replay target can only debug a process that is
6170 already running. Therefore, you need first to start the process with
6171 the @kbd{run} or @kbd{start} commands, and then start the recording
6172 with the @kbd{record @var{method}} command.
6174 Both @code{record @var{method}} and @code{rec @var{method}} are
6175 aliases of @code{target record-@var{method}}.
6177 @cindex displaced stepping, and process record and replay
6178 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6179 will be automatically disabled when process record and replay target
6180 is started. That's because the process record and replay target
6181 doesn't support displaced stepping.
6183 @cindex non-stop mode, and process record and replay
6184 @cindex asynchronous execution, and process record and replay
6185 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6186 the asynchronous execution mode (@pxref{Background Execution}), not
6187 all recording methods are available. The @code{full} recording method
6188 does not support these two modes.
6193 Stop the process record and replay target. When process record and
6194 replay target stops, the entire execution log will be deleted and the
6195 inferior will either be terminated, or will remain in its final state.
6197 When you stop the process record and replay target in record mode (at
6198 the end of the execution log), the inferior will be stopped at the
6199 next instruction that would have been recorded. In other words, if
6200 you record for a while and then stop recording, the inferior process
6201 will be left in the same state as if the recording never happened.
6203 On the other hand, if the process record and replay target is stopped
6204 while in replay mode (that is, not at the end of the execution log,
6205 but at some earlier point), the inferior process will become ``live''
6206 at that earlier state, and it will then be possible to continue the
6207 usual ``live'' debugging of the process from that state.
6209 When the inferior process exits, or @value{GDBN} detaches from it,
6210 process record and replay target will automatically stop itself.
6214 Go to a specific location in the execution log. There are several
6215 ways to specify the location to go to:
6218 @item record goto begin
6219 @itemx record goto start
6220 Go to the beginning of the execution log.
6222 @item record goto end
6223 Go to the end of the execution log.
6225 @item record goto @var{n}
6226 Go to instruction number @var{n} in the execution log.
6230 @item record save @var{filename}
6231 Save the execution log to a file @file{@var{filename}}.
6232 Default filename is @file{gdb_record.@var{process_id}}, where
6233 @var{process_id} is the process ID of the inferior.
6235 This command may not be available for all recording methods.
6237 @kindex record restore
6238 @item record restore @var{filename}
6239 Restore the execution log from a file @file{@var{filename}}.
6240 File must have been created with @code{record save}.
6242 @kindex set record full
6243 @item set record full insn-number-max @var{limit}
6244 @itemx set record full insn-number-max unlimited
6245 Set the limit of instructions to be recorded for the @code{full}
6246 recording method. Default value is 200000.
6248 If @var{limit} is a positive number, then @value{GDBN} will start
6249 deleting instructions from the log once the number of the record
6250 instructions becomes greater than @var{limit}. For every new recorded
6251 instruction, @value{GDBN} will delete the earliest recorded
6252 instruction to keep the number of recorded instructions at the limit.
6253 (Since deleting recorded instructions loses information, @value{GDBN}
6254 lets you control what happens when the limit is reached, by means of
6255 the @code{stop-at-limit} option, described below.)
6257 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
6258 delete recorded instructions from the execution log. The number of
6259 recorded instructions is limited only by the available memory.
6261 @kindex show record full
6262 @item show record full insn-number-max
6263 Show the limit of instructions to be recorded with the @code{full}
6266 @item set record full stop-at-limit
6267 Control the behavior of the @code{full} recording method when the
6268 number of recorded instructions reaches the limit. If ON (the
6269 default), @value{GDBN} will stop when the limit is reached for the
6270 first time and ask you whether you want to stop the inferior or
6271 continue running it and recording the execution log. If you decide
6272 to continue recording, each new recorded instruction will cause the
6273 oldest one to be deleted.
6275 If this option is OFF, @value{GDBN} will automatically delete the
6276 oldest record to make room for each new one, without asking.
6278 @item show record full stop-at-limit
6279 Show the current setting of @code{stop-at-limit}.
6281 @item set record full memory-query
6282 Control the behavior when @value{GDBN} is unable to record memory
6283 changes caused by an instruction for the @code{full} recording method.
6284 If ON, @value{GDBN} will query whether to stop the inferior in that
6287 If this option is OFF (the default), @value{GDBN} will automatically
6288 ignore the effect of such instructions on memory. Later, when
6289 @value{GDBN} replays this execution log, it will mark the log of this
6290 instruction as not accessible, and it will not affect the replay
6293 @item show record full memory-query
6294 Show the current setting of @code{memory-query}.
6298 Show various statistics about the recording depending on the recording
6303 For the @code{full} recording method, it shows the state of process
6304 record and its in-memory execution log buffer, including:
6308 Whether in record mode or replay mode.
6310 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
6312 Highest recorded instruction number.
6314 Current instruction about to be replayed (if in replay mode).
6316 Number of instructions contained in the execution log.
6318 Maximum number of instructions that may be contained in the execution log.
6322 For the @code{btrace} recording method, it shows the number of
6323 instructions that have been recorded and the number of blocks of
6324 sequential control-flow that is formed by the recorded instructions.
6327 @kindex record delete
6330 When record target runs in replay mode (``in the past''), delete the
6331 subsequent execution log and begin to record a new execution log starting
6332 from the current address. This means you will abandon the previously
6333 recorded ``future'' and begin recording a new ``future''.
6335 @kindex record instruction-history
6336 @kindex rec instruction-history
6337 @item record instruction-history
6338 Disassembles instructions from the recorded execution log. By
6339 default, ten instructions are disassembled. This can be changed using
6340 the @code{set record instruction-history-size} command. Instructions
6341 are printed in execution order. There are several ways to specify
6342 what part of the execution log to disassemble:
6345 @item record instruction-history @var{insn}
6346 Disassembles ten instructions starting from instruction number
6349 @item record instruction-history @var{insn}, +/-@var{n}
6350 Disassembles @var{n} instructions around instruction number
6351 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
6352 @var{n} instructions after instruction number @var{insn}. If
6353 @var{n} is preceded with @code{-}, disassembles @var{n}
6354 instructions before instruction number @var{insn}.
6356 @item record instruction-history
6357 Disassembles ten more instructions after the last disassembly.
6359 @item record instruction-history -
6360 Disassembles ten more instructions before the last disassembly.
6362 @item record instruction-history @var{begin} @var{end}
6363 Disassembles instructions beginning with instruction number
6364 @var{begin} until instruction number @var{end}. The instruction
6365 number @var{end} is not included.
6368 This command may not be available for all recording methods.
6371 @item set record instruction-history-size @var{size}
6372 @itemx set record instruction-history-size unlimited
6373 Define how many instructions to disassemble in the @code{record
6374 instruction-history} command. The default value is 10.
6375 A @var{size} of @code{unlimited} means unlimited instructions.
6378 @item show record instruction-history-size
6379 Show how many instructions to disassemble in the @code{record
6380 instruction-history} command.
6382 @kindex record function-call-history
6383 @kindex rec function-call-history
6384 @item record function-call-history
6385 Prints the execution history at function granularity. It prints one
6386 line for each sequence of instructions that belong to the same
6387 function giving the name of that function, the source lines
6388 for this instruction sequence (if the @code{/l} modifier is
6389 specified), and the instructions numbers that form the sequence (if
6390 the @code{/i} modifier is specified).
6393 (@value{GDBP}) @b{list 1, 10}
6404 (@value{GDBP}) @b{record function-call-history /l}
6410 By default, ten lines are printed. This can be changed using the
6411 @code{set record function-call-history-size} command. Functions are
6412 printed in execution order. There are several ways to specify what
6416 @item record function-call-history @var{func}
6417 Prints ten functions starting from function number @var{func}.
6419 @item record function-call-history @var{func}, +/-@var{n}
6420 Prints @var{n} functions around function number @var{func}. If
6421 @var{n} is preceded with @code{+}, prints @var{n} functions after
6422 function number @var{func}. If @var{n} is preceded with @code{-},
6423 prints @var{n} functions before function number @var{func}.
6425 @item record function-call-history
6426 Prints ten more functions after the last ten-line print.
6428 @item record function-call-history -
6429 Prints ten more functions before the last ten-line print.
6431 @item record function-call-history @var{begin} @var{end}
6432 Prints functions beginning with function number @var{begin} until
6433 function number @var{end}. The function number @var{end} is not
6437 This command may not be available for all recording methods.
6439 @item set record function-call-history-size @var{size}
6440 @itemx set record function-call-history-size unlimited
6441 Define how many lines to print in the
6442 @code{record function-call-history} command. The default value is 10.
6443 A size of @code{unlimited} means unlimited lines.
6445 @item show record function-call-history-size
6446 Show how many lines to print in the
6447 @code{record function-call-history} command.
6452 @chapter Examining the Stack
6454 When your program has stopped, the first thing you need to know is where it
6455 stopped and how it got there.
6458 Each time your program performs a function call, information about the call
6460 That information includes the location of the call in your program,
6461 the arguments of the call,
6462 and the local variables of the function being called.
6463 The information is saved in a block of data called a @dfn{stack frame}.
6464 The stack frames are allocated in a region of memory called the @dfn{call
6467 When your program stops, the @value{GDBN} commands for examining the
6468 stack allow you to see all of this information.
6470 @cindex selected frame
6471 One of the stack frames is @dfn{selected} by @value{GDBN} and many
6472 @value{GDBN} commands refer implicitly to the selected frame. In
6473 particular, whenever you ask @value{GDBN} for the value of a variable in
6474 your program, the value is found in the selected frame. There are
6475 special @value{GDBN} commands to select whichever frame you are
6476 interested in. @xref{Selection, ,Selecting a Frame}.
6478 When your program stops, @value{GDBN} automatically selects the
6479 currently executing frame and describes it briefly, similar to the
6480 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
6483 * Frames:: Stack frames
6484 * Backtrace:: Backtraces
6485 * Frame Filter Management:: Managing frame filters
6486 * Selection:: Selecting a frame
6487 * Frame Info:: Information on a frame
6492 @section Stack Frames
6494 @cindex frame, definition
6496 The call stack is divided up into contiguous pieces called @dfn{stack
6497 frames}, or @dfn{frames} for short; each frame is the data associated
6498 with one call to one function. The frame contains the arguments given
6499 to the function, the function's local variables, and the address at
6500 which the function is executing.
6502 @cindex initial frame
6503 @cindex outermost frame
6504 @cindex innermost frame
6505 When your program is started, the stack has only one frame, that of the
6506 function @code{main}. This is called the @dfn{initial} frame or the
6507 @dfn{outermost} frame. Each time a function is called, a new frame is
6508 made. Each time a function returns, the frame for that function invocation
6509 is eliminated. If a function is recursive, there can be many frames for
6510 the same function. The frame for the function in which execution is
6511 actually occurring is called the @dfn{innermost} frame. This is the most
6512 recently created of all the stack frames that still exist.
6514 @cindex frame pointer
6515 Inside your program, stack frames are identified by their addresses. A
6516 stack frame consists of many bytes, each of which has its own address; each
6517 kind of computer has a convention for choosing one byte whose
6518 address serves as the address of the frame. Usually this address is kept
6519 in a register called the @dfn{frame pointer register}
6520 (@pxref{Registers, $fp}) while execution is going on in that frame.
6522 @cindex frame number
6523 @value{GDBN} assigns numbers to all existing stack frames, starting with
6524 zero for the innermost frame, one for the frame that called it,
6525 and so on upward. These numbers do not really exist in your program;
6526 they are assigned by @value{GDBN} to give you a way of designating stack
6527 frames in @value{GDBN} commands.
6529 @c The -fomit-frame-pointer below perennially causes hbox overflow
6530 @c underflow problems.
6531 @cindex frameless execution
6532 Some compilers provide a way to compile functions so that they operate
6533 without stack frames. (For example, the @value{NGCC} option
6535 @samp{-fomit-frame-pointer}
6537 generates functions without a frame.)
6538 This is occasionally done with heavily used library functions to save
6539 the frame setup time. @value{GDBN} has limited facilities for dealing
6540 with these function invocations. If the innermost function invocation
6541 has no stack frame, @value{GDBN} nevertheless regards it as though
6542 it had a separate frame, which is numbered zero as usual, allowing
6543 correct tracing of the function call chain. However, @value{GDBN} has
6544 no provision for frameless functions elsewhere in the stack.
6547 @kindex frame@r{, command}
6548 @cindex current stack frame
6549 @item frame @var{args}
6550 The @code{frame} command allows you to move from one stack frame to another,
6551 and to print the stack frame you select. @var{args} may be either the
6552 address of the frame or the stack frame number. Without an argument,
6553 @code{frame} prints the current stack frame.
6555 @kindex select-frame
6556 @cindex selecting frame silently
6558 The @code{select-frame} command allows you to move from one stack frame
6559 to another without printing the frame. This is the silent version of
6567 @cindex call stack traces
6568 A backtrace is a summary of how your program got where it is. It shows one
6569 line per frame, for many frames, starting with the currently executing
6570 frame (frame zero), followed by its caller (frame one), and on up the
6573 @anchor{backtrace-command}
6576 @kindex bt @r{(@code{backtrace})}
6579 Print a backtrace of the entire stack: one line per frame for all
6580 frames in the stack.
6582 You can stop the backtrace at any time by typing the system interrupt
6583 character, normally @kbd{Ctrl-c}.
6585 @item backtrace @var{n}
6587 Similar, but print only the innermost @var{n} frames.
6589 @item backtrace -@var{n}
6591 Similar, but print only the outermost @var{n} frames.
6593 @item backtrace full
6595 @itemx bt full @var{n}
6596 @itemx bt full -@var{n}
6597 Print the values of the local variables also. @var{n} specifies the
6598 number of frames to print, as described above.
6600 @item backtrace no-filters
6601 @itemx bt no-filters
6602 @itemx bt no-filters @var{n}
6603 @itemx bt no-filters -@var{n}
6604 @itemx bt no-filters full
6605 @itemx bt no-filters full @var{n}
6606 @itemx bt no-filters full -@var{n}
6607 Do not run Python frame filters on this backtrace. @xref{Frame
6608 Filter API}, for more information. Additionally use @ref{disable
6609 frame-filter all} to turn off all frame filters. This is only
6610 relevant when @value{GDBN} has been configured with @code{Python}
6616 The names @code{where} and @code{info stack} (abbreviated @code{info s})
6617 are additional aliases for @code{backtrace}.
6619 @cindex multiple threads, backtrace
6620 In a multi-threaded program, @value{GDBN} by default shows the
6621 backtrace only for the current thread. To display the backtrace for
6622 several or all of the threads, use the command @code{thread apply}
6623 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
6624 apply all backtrace}, @value{GDBN} will display the backtrace for all
6625 the threads; this is handy when you debug a core dump of a
6626 multi-threaded program.
6628 Each line in the backtrace shows the frame number and the function name.
6629 The program counter value is also shown---unless you use @code{set
6630 print address off}. The backtrace also shows the source file name and
6631 line number, as well as the arguments to the function. The program
6632 counter value is omitted if it is at the beginning of the code for that
6635 Here is an example of a backtrace. It was made with the command
6636 @samp{bt 3}, so it shows the innermost three frames.
6640 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6642 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
6643 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
6645 (More stack frames follow...)
6650 The display for frame zero does not begin with a program counter
6651 value, indicating that your program has stopped at the beginning of the
6652 code for line @code{993} of @code{builtin.c}.
6655 The value of parameter @code{data} in frame 1 has been replaced by
6656 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
6657 only if it is a scalar (integer, pointer, enumeration, etc). See command
6658 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
6659 on how to configure the way function parameter values are printed.
6661 @cindex optimized out, in backtrace
6662 @cindex function call arguments, optimized out
6663 If your program was compiled with optimizations, some compilers will
6664 optimize away arguments passed to functions if those arguments are
6665 never used after the call. Such optimizations generate code that
6666 passes arguments through registers, but doesn't store those arguments
6667 in the stack frame. @value{GDBN} has no way of displaying such
6668 arguments in stack frames other than the innermost one. Here's what
6669 such a backtrace might look like:
6673 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6675 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
6676 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
6678 (More stack frames follow...)
6683 The values of arguments that were not saved in their stack frames are
6684 shown as @samp{<optimized out>}.
6686 If you need to display the values of such optimized-out arguments,
6687 either deduce that from other variables whose values depend on the one
6688 you are interested in, or recompile without optimizations.
6690 @cindex backtrace beyond @code{main} function
6691 @cindex program entry point
6692 @cindex startup code, and backtrace
6693 Most programs have a standard user entry point---a place where system
6694 libraries and startup code transition into user code. For C this is
6695 @code{main}@footnote{
6696 Note that embedded programs (the so-called ``free-standing''
6697 environment) are not required to have a @code{main} function as the
6698 entry point. They could even have multiple entry points.}.
6699 When @value{GDBN} finds the entry function in a backtrace
6700 it will terminate the backtrace, to avoid tracing into highly
6701 system-specific (and generally uninteresting) code.
6703 If you need to examine the startup code, or limit the number of levels
6704 in a backtrace, you can change this behavior:
6707 @item set backtrace past-main
6708 @itemx set backtrace past-main on
6709 @kindex set backtrace
6710 Backtraces will continue past the user entry point.
6712 @item set backtrace past-main off
6713 Backtraces will stop when they encounter the user entry point. This is the
6716 @item show backtrace past-main
6717 @kindex show backtrace
6718 Display the current user entry point backtrace policy.
6720 @item set backtrace past-entry
6721 @itemx set backtrace past-entry on
6722 Backtraces will continue past the internal entry point of an application.
6723 This entry point is encoded by the linker when the application is built,
6724 and is likely before the user entry point @code{main} (or equivalent) is called.
6726 @item set backtrace past-entry off
6727 Backtraces will stop when they encounter the internal entry point of an
6728 application. This is the default.
6730 @item show backtrace past-entry
6731 Display the current internal entry point backtrace policy.
6733 @item set backtrace limit @var{n}
6734 @itemx set backtrace limit 0
6735 @itemx set backtrace limit unlimited
6736 @cindex backtrace limit
6737 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
6738 or zero means unlimited levels.
6740 @item show backtrace limit
6741 Display the current limit on backtrace levels.
6744 You can control how file names are displayed.
6747 @item set filename-display
6748 @itemx set filename-display relative
6749 @cindex filename-display
6750 Display file names relative to the compilation directory. This is the default.
6752 @item set filename-display basename
6753 Display only basename of a filename.
6755 @item set filename-display absolute
6756 Display an absolute filename.
6758 @item show filename-display
6759 Show the current way to display filenames.
6762 @node Frame Filter Management
6763 @section Management of Frame Filters.
6764 @cindex managing frame filters
6766 Frame filters are Python based utilities to manage and decorate the
6767 output of frames. @xref{Frame Filter API}, for further information.
6769 Managing frame filters is performed by several commands available
6770 within @value{GDBN}, detailed here.
6773 @kindex info frame-filter
6774 @item info frame-filter
6775 Print a list of installed frame filters from all dictionaries, showing
6776 their name, priority and enabled status.
6778 @kindex disable frame-filter
6779 @anchor{disable frame-filter all}
6780 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
6781 Disable a frame filter in the dictionary matching
6782 @var{filter-dictionary}, or @code{all}, and @var{filter-name}.
6783 @var{filter-dictionary} may be @code{all}, @code{global},
6784 @code{progspace} or the name of the object file where the frame filter
6785 dictionary resides. When @code{all} is specified, all frame filters
6786 across all dictionaries are disabled. @var{filter-name} is the name
6787 of the frame filter and is used when @code{all} is not the option for
6788 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
6789 may be enabled again later.
6791 @kindex enable frame-filter
6792 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
6793 Enable a frame filter in the dictionary matching
6794 @var{filter-dictionary}, or @code{all}, and @var{filter-name}.
6795 @var{filter-dictionary} may be @code{all}, @code{global},
6796 @code{progspace} or the name of the object file where the frame filter
6797 dictionary resides. When @code{all} is specified, all frame filters across
6798 all dictionaries are enabled. @var{filter-name} is the name of the frame
6799 filter and is used when @code{all} is not the option for
6800 @var{filter-dictionary}.
6805 (gdb) info frame-filter
6807 global frame-filters:
6808 Priority Enabled Name
6809 1000 No PrimaryFunctionFilter
6812 progspace /build/test frame-filters:
6813 Priority Enabled Name
6814 100 Yes ProgspaceFilter
6816 objfile /build/test frame-filters:
6817 Priority Enabled Name
6818 999 Yes BuildProgra Filter
6820 (gdb) disable frame-filter /build/test BuildProgramFilter
6821 (gdb) info frame-filter
6823 global frame-filters:
6824 Priority Enabled Name
6825 1000 No PrimaryFunctionFilter
6828 progspace /build/test frame-filters:
6829 Priority Enabled Name
6830 100 Yes ProgspaceFilter
6832 objfile /build/test frame-filters:
6833 Priority Enabled Name
6834 999 No BuildProgramFilter
6836 (gdb) enable frame-filter global PrimaryFunctionFilter
6837 (gdb) info frame-filter
6839 global frame-filters:
6840 Priority Enabled Name
6841 1000 Yes PrimaryFunctionFilter
6844 progspace /build/test frame-filters:
6845 Priority Enabled Name
6846 100 Yes ProgspaceFilter
6848 objfile /build/test frame-filters:
6849 Priority Enabled Name
6850 999 No BuildProgramFilter
6853 @kindex set frame-filter priority
6854 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
6855 Set the @var{priority} of a frame filter in the dictionary matching
6856 @var{filter-dictionary}, and the frame filter name matching
6857 @var{filter-name}. @var{filter-dictionary} may be @code{global},
6858 @code{progspace} or the name of the object file where the frame filter
6859 dictionary resides. @var{priority} is an integer.
6861 @kindex show frame-filter priority
6862 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
6863 Show the @var{priority} of a frame filter in the dictionary matching
6864 @var{filter-dictionary}, and the frame filter name matching
6865 @var{filter-name}. @var{filter-dictionary} may be @code{global},
6866 @code{progspace} or the name of the object file where the frame filter
6872 (gdb) info frame-filter
6874 global frame-filters:
6875 Priority Enabled Name
6876 1000 Yes PrimaryFunctionFilter
6879 progspace /build/test frame-filters:
6880 Priority Enabled Name
6881 100 Yes ProgspaceFilter
6883 objfile /build/test frame-filters:
6884 Priority Enabled Name
6885 999 No BuildProgramFilter
6887 (gdb) set frame-filter priority global Reverse 50
6888 (gdb) info frame-filter
6890 global frame-filters:
6891 Priority Enabled Name
6892 1000 Yes PrimaryFunctionFilter
6895 progspace /build/test frame-filters:
6896 Priority Enabled Name
6897 100 Yes ProgspaceFilter
6899 objfile /build/test frame-filters:
6900 Priority Enabled Name
6901 999 No BuildProgramFilter
6906 @section Selecting a Frame
6908 Most commands for examining the stack and other data in your program work on
6909 whichever stack frame is selected at the moment. Here are the commands for
6910 selecting a stack frame; all of them finish by printing a brief description
6911 of the stack frame just selected.
6914 @kindex frame@r{, selecting}
6915 @kindex f @r{(@code{frame})}
6918 Select frame number @var{n}. Recall that frame zero is the innermost
6919 (currently executing) frame, frame one is the frame that called the
6920 innermost one, and so on. The highest-numbered frame is the one for
6923 @item frame @var{addr}
6925 Select the frame at address @var{addr}. This is useful mainly if the
6926 chaining of stack frames has been damaged by a bug, making it
6927 impossible for @value{GDBN} to assign numbers properly to all frames. In
6928 addition, this can be useful when your program has multiple stacks and
6929 switches between them.
6931 On the SPARC architecture, @code{frame} needs two addresses to
6932 select an arbitrary frame: a frame pointer and a stack pointer.
6934 On the @acronym{MIPS} and Alpha architecture, it needs two addresses: a stack
6935 pointer and a program counter.
6937 On the 29k architecture, it needs three addresses: a register stack
6938 pointer, a program counter, and a memory stack pointer.
6942 Move @var{n} frames up the stack. For positive numbers @var{n}, this
6943 advances toward the outermost frame, to higher frame numbers, to frames
6944 that have existed longer. @var{n} defaults to one.
6947 @kindex do @r{(@code{down})}
6949 Move @var{n} frames down the stack. For positive numbers @var{n}, this
6950 advances toward the innermost frame, to lower frame numbers, to frames
6951 that were created more recently. @var{n} defaults to one. You may
6952 abbreviate @code{down} as @code{do}.
6955 All of these commands end by printing two lines of output describing the
6956 frame. The first line shows the frame number, the function name, the
6957 arguments, and the source file and line number of execution in that
6958 frame. The second line shows the text of that source line.
6966 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
6968 10 read_input_file (argv[i]);
6972 After such a printout, the @code{list} command with no arguments
6973 prints ten lines centered on the point of execution in the frame.
6974 You can also edit the program at the point of execution with your favorite
6975 editing program by typing @code{edit}.
6976 @xref{List, ,Printing Source Lines},
6980 @kindex down-silently
6982 @item up-silently @var{n}
6983 @itemx down-silently @var{n}
6984 These two commands are variants of @code{up} and @code{down},
6985 respectively; they differ in that they do their work silently, without
6986 causing display of the new frame. They are intended primarily for use
6987 in @value{GDBN} command scripts, where the output might be unnecessary and
6992 @section Information About a Frame
6994 There are several other commands to print information about the selected
7000 When used without any argument, this command does not change which
7001 frame is selected, but prints a brief description of the currently
7002 selected stack frame. It can be abbreviated @code{f}. With an
7003 argument, this command is used to select a stack frame.
7004 @xref{Selection, ,Selecting a Frame}.
7007 @kindex info f @r{(@code{info frame})}
7010 This command prints a verbose description of the selected stack frame,
7015 the address of the frame
7017 the address of the next frame down (called by this frame)
7019 the address of the next frame up (caller of this frame)
7021 the language in which the source code corresponding to this frame is written
7023 the address of the frame's arguments
7025 the address of the frame's local variables
7027 the program counter saved in it (the address of execution in the caller frame)
7029 which registers were saved in the frame
7032 @noindent The verbose description is useful when
7033 something has gone wrong that has made the stack format fail to fit
7034 the usual conventions.
7036 @item info frame @var{addr}
7037 @itemx info f @var{addr}
7038 Print a verbose description of the frame at address @var{addr}, without
7039 selecting that frame. The selected frame remains unchanged by this
7040 command. This requires the same kind of address (more than one for some
7041 architectures) that you specify in the @code{frame} command.
7042 @xref{Selection, ,Selecting a Frame}.
7046 Print the arguments of the selected frame, each on a separate line.
7050 Print the local variables of the selected frame, each on a separate
7051 line. These are all variables (declared either static or automatic)
7052 accessible at the point of execution of the selected frame.
7058 @chapter Examining Source Files
7060 @value{GDBN} can print parts of your program's source, since the debugging
7061 information recorded in the program tells @value{GDBN} what source files were
7062 used to build it. When your program stops, @value{GDBN} spontaneously prints
7063 the line where it stopped. Likewise, when you select a stack frame
7064 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
7065 execution in that frame has stopped. You can print other portions of
7066 source files by explicit command.
7068 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
7069 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
7070 @value{GDBN} under @sc{gnu} Emacs}.
7073 * List:: Printing source lines
7074 * Specify Location:: How to specify code locations
7075 * Edit:: Editing source files
7076 * Search:: Searching source files
7077 * Source Path:: Specifying source directories
7078 * Machine Code:: Source and machine code
7082 @section Printing Source Lines
7085 @kindex l @r{(@code{list})}
7086 To print lines from a source file, use the @code{list} command
7087 (abbreviated @code{l}). By default, ten lines are printed.
7088 There are several ways to specify what part of the file you want to
7089 print; see @ref{Specify Location}, for the full list.
7091 Here are the forms of the @code{list} command most commonly used:
7094 @item list @var{linenum}
7095 Print lines centered around line number @var{linenum} in the
7096 current source file.
7098 @item list @var{function}
7099 Print lines centered around the beginning of function
7103 Print more lines. If the last lines printed were printed with a
7104 @code{list} command, this prints lines following the last lines
7105 printed; however, if the last line printed was a solitary line printed
7106 as part of displaying a stack frame (@pxref{Stack, ,Examining the
7107 Stack}), this prints lines centered around that line.
7110 Print lines just before the lines last printed.
7113 @cindex @code{list}, how many lines to display
7114 By default, @value{GDBN} prints ten source lines with any of these forms of
7115 the @code{list} command. You can change this using @code{set listsize}:
7118 @kindex set listsize
7119 @item set listsize @var{count}
7120 @itemx set listsize unlimited
7121 Make the @code{list} command display @var{count} source lines (unless
7122 the @code{list} argument explicitly specifies some other number).
7123 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
7125 @kindex show listsize
7127 Display the number of lines that @code{list} prints.
7130 Repeating a @code{list} command with @key{RET} discards the argument,
7131 so it is equivalent to typing just @code{list}. This is more useful
7132 than listing the same lines again. An exception is made for an
7133 argument of @samp{-}; that argument is preserved in repetition so that
7134 each repetition moves up in the source file.
7136 In general, the @code{list} command expects you to supply zero, one or two
7137 @dfn{linespecs}. Linespecs specify source lines; there are several ways
7138 of writing them (@pxref{Specify Location}), but the effect is always
7139 to specify some source line.
7141 Here is a complete description of the possible arguments for @code{list}:
7144 @item list @var{linespec}
7145 Print lines centered around the line specified by @var{linespec}.
7147 @item list @var{first},@var{last}
7148 Print lines from @var{first} to @var{last}. Both arguments are
7149 linespecs. When a @code{list} command has two linespecs, and the
7150 source file of the second linespec is omitted, this refers to
7151 the same source file as the first linespec.
7153 @item list ,@var{last}
7154 Print lines ending with @var{last}.
7156 @item list @var{first},
7157 Print lines starting with @var{first}.
7160 Print lines just after the lines last printed.
7163 Print lines just before the lines last printed.
7166 As described in the preceding table.
7169 @node Specify Location
7170 @section Specifying a Location
7171 @cindex specifying location
7174 Several @value{GDBN} commands accept arguments that specify a location
7175 of your program's code. Since @value{GDBN} is a source-level
7176 debugger, a location usually specifies some line in the source code;
7177 for that reason, locations are also known as @dfn{linespecs}.
7179 Here are all the different ways of specifying a code location that
7180 @value{GDBN} understands:
7184 Specifies the line number @var{linenum} of the current source file.
7187 @itemx +@var{offset}
7188 Specifies the line @var{offset} lines before or after the @dfn{current
7189 line}. For the @code{list} command, the current line is the last one
7190 printed; for the breakpoint commands, this is the line at which
7191 execution stopped in the currently selected @dfn{stack frame}
7192 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
7193 used as the second of the two linespecs in a @code{list} command,
7194 this specifies the line @var{offset} lines up or down from the first
7197 @item @var{filename}:@var{linenum}
7198 Specifies the line @var{linenum} in the source file @var{filename}.
7199 If @var{filename} is a relative file name, then it will match any
7200 source file name with the same trailing components. For example, if
7201 @var{filename} is @samp{gcc/expr.c}, then it will match source file
7202 name of @file{/build/trunk/gcc/expr.c}, but not
7203 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
7205 @item @var{function}
7206 Specifies the line that begins the body of the function @var{function}.
7207 For example, in C, this is the line with the open brace.
7209 @item @var{function}:@var{label}
7210 Specifies the line where @var{label} appears in @var{function}.
7212 @item @var{filename}:@var{function}
7213 Specifies the line that begins the body of the function @var{function}
7214 in the file @var{filename}. You only need the file name with a
7215 function name to avoid ambiguity when there are identically named
7216 functions in different source files.
7219 Specifies the line at which the label named @var{label} appears.
7220 @value{GDBN} searches for the label in the function corresponding to
7221 the currently selected stack frame. If there is no current selected
7222 stack frame (for instance, if the inferior is not running), then
7223 @value{GDBN} will not search for a label.
7225 @item *@var{address}
7226 Specifies the program address @var{address}. For line-oriented
7227 commands, such as @code{list} and @code{edit}, this specifies a source
7228 line that contains @var{address}. For @code{break} and other
7229 breakpoint oriented commands, this can be used to set breakpoints in
7230 parts of your program which do not have debugging information or
7233 Here @var{address} may be any expression valid in the current working
7234 language (@pxref{Languages, working language}) that specifies a code
7235 address. In addition, as a convenience, @value{GDBN} extends the
7236 semantics of expressions used in locations to cover the situations
7237 that frequently happen during debugging. Here are the various forms
7241 @item @var{expression}
7242 Any expression valid in the current working language.
7244 @item @var{funcaddr}
7245 An address of a function or procedure derived from its name. In C,
7246 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
7247 simply the function's name @var{function} (and actually a special case
7248 of a valid expression). In Pascal and Modula-2, this is
7249 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
7250 (although the Pascal form also works).
7252 This form specifies the address of the function's first instruction,
7253 before the stack frame and arguments have been set up.
7255 @item '@var{filename}'::@var{funcaddr}
7256 Like @var{funcaddr} above, but also specifies the name of the source
7257 file explicitly. This is useful if the name of the function does not
7258 specify the function unambiguously, e.g., if there are several
7259 functions with identical names in different source files.
7262 @cindex breakpoint at static probe point
7263 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
7264 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
7265 applications to embed static probes. @xref{Static Probe Points}, for more
7266 information on finding and using static probes. This form of linespec
7267 specifies the location of such a static probe.
7269 If @var{objfile} is given, only probes coming from that shared library
7270 or executable matching @var{objfile} as a regular expression are considered.
7271 If @var{provider} is given, then only probes from that provider are considered.
7272 If several probes match the spec, @value{GDBN} will insert a breakpoint at
7273 each one of those probes.
7279 @section Editing Source Files
7280 @cindex editing source files
7283 @kindex e @r{(@code{edit})}
7284 To edit the lines in a source file, use the @code{edit} command.
7285 The editing program of your choice
7286 is invoked with the current line set to
7287 the active line in the program.
7288 Alternatively, there are several ways to specify what part of the file you
7289 want to print if you want to see other parts of the program:
7292 @item edit @var{location}
7293 Edit the source file specified by @code{location}. Editing starts at
7294 that @var{location}, e.g., at the specified source line of the
7295 specified file. @xref{Specify Location}, for all the possible forms
7296 of the @var{location} argument; here are the forms of the @code{edit}
7297 command most commonly used:
7300 @item edit @var{number}
7301 Edit the current source file with @var{number} as the active line number.
7303 @item edit @var{function}
7304 Edit the file containing @var{function} at the beginning of its definition.
7309 @subsection Choosing your Editor
7310 You can customize @value{GDBN} to use any editor you want
7312 The only restriction is that your editor (say @code{ex}), recognizes the
7313 following command-line syntax:
7315 ex +@var{number} file
7317 The optional numeric value +@var{number} specifies the number of the line in
7318 the file where to start editing.}.
7319 By default, it is @file{@value{EDITOR}}, but you can change this
7320 by setting the environment variable @code{EDITOR} before using
7321 @value{GDBN}. For example, to configure @value{GDBN} to use the
7322 @code{vi} editor, you could use these commands with the @code{sh} shell:
7328 or in the @code{csh} shell,
7330 setenv EDITOR /usr/bin/vi
7335 @section Searching Source Files
7336 @cindex searching source files
7338 There are two commands for searching through the current source file for a
7343 @kindex forward-search
7344 @kindex fo @r{(@code{forward-search})}
7345 @item forward-search @var{regexp}
7346 @itemx search @var{regexp}
7347 The command @samp{forward-search @var{regexp}} checks each line,
7348 starting with the one following the last line listed, for a match for
7349 @var{regexp}. It lists the line that is found. You can use the
7350 synonym @samp{search @var{regexp}} or abbreviate the command name as
7353 @kindex reverse-search
7354 @item reverse-search @var{regexp}
7355 The command @samp{reverse-search @var{regexp}} checks each line, starting
7356 with the one before the last line listed and going backward, for a match
7357 for @var{regexp}. It lists the line that is found. You can abbreviate
7358 this command as @code{rev}.
7362 @section Specifying Source Directories
7365 @cindex directories for source files
7366 Executable programs sometimes do not record the directories of the source
7367 files from which they were compiled, just the names. Even when they do,
7368 the directories could be moved between the compilation and your debugging
7369 session. @value{GDBN} has a list of directories to search for source files;
7370 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
7371 it tries all the directories in the list, in the order they are present
7372 in the list, until it finds a file with the desired name.
7374 For example, suppose an executable references the file
7375 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
7376 @file{/mnt/cross}. The file is first looked up literally; if this
7377 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
7378 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
7379 message is printed. @value{GDBN} does not look up the parts of the
7380 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
7381 Likewise, the subdirectories of the source path are not searched: if
7382 the source path is @file{/mnt/cross}, and the binary refers to
7383 @file{foo.c}, @value{GDBN} would not find it under
7384 @file{/mnt/cross/usr/src/foo-1.0/lib}.
7386 Plain file names, relative file names with leading directories, file
7387 names containing dots, etc.@: are all treated as described above; for
7388 instance, if the source path is @file{/mnt/cross}, and the source file
7389 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
7390 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
7391 that---@file{/mnt/cross/foo.c}.
7393 Note that the executable search path is @emph{not} used to locate the
7396 Whenever you reset or rearrange the source path, @value{GDBN} clears out
7397 any information it has cached about where source files are found and where
7398 each line is in the file.
7402 When you start @value{GDBN}, its source path includes only @samp{cdir}
7403 and @samp{cwd}, in that order.
7404 To add other directories, use the @code{directory} command.
7406 The search path is used to find both program source files and @value{GDBN}
7407 script files (read using the @samp{-command} option and @samp{source} command).
7409 In addition to the source path, @value{GDBN} provides a set of commands
7410 that manage a list of source path substitution rules. A @dfn{substitution
7411 rule} specifies how to rewrite source directories stored in the program's
7412 debug information in case the sources were moved to a different
7413 directory between compilation and debugging. A rule is made of
7414 two strings, the first specifying what needs to be rewritten in
7415 the path, and the second specifying how it should be rewritten.
7416 In @ref{set substitute-path}, we name these two parts @var{from} and
7417 @var{to} respectively. @value{GDBN} does a simple string replacement
7418 of @var{from} with @var{to} at the start of the directory part of the
7419 source file name, and uses that result instead of the original file
7420 name to look up the sources.
7422 Using the previous example, suppose the @file{foo-1.0} tree has been
7423 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
7424 @value{GDBN} to replace @file{/usr/src} in all source path names with
7425 @file{/mnt/cross}. The first lookup will then be
7426 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
7427 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
7428 substitution rule, use the @code{set substitute-path} command
7429 (@pxref{set substitute-path}).
7431 To avoid unexpected substitution results, a rule is applied only if the
7432 @var{from} part of the directory name ends at a directory separator.
7433 For instance, a rule substituting @file{/usr/source} into
7434 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
7435 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
7436 is applied only at the beginning of the directory name, this rule will
7437 not be applied to @file{/root/usr/source/baz.c} either.
7439 In many cases, you can achieve the same result using the @code{directory}
7440 command. However, @code{set substitute-path} can be more efficient in
7441 the case where the sources are organized in a complex tree with multiple
7442 subdirectories. With the @code{directory} command, you need to add each
7443 subdirectory of your project. If you moved the entire tree while
7444 preserving its internal organization, then @code{set substitute-path}
7445 allows you to direct the debugger to all the sources with one single
7448 @code{set substitute-path} is also more than just a shortcut command.
7449 The source path is only used if the file at the original location no
7450 longer exists. On the other hand, @code{set substitute-path} modifies
7451 the debugger behavior to look at the rewritten location instead. So, if
7452 for any reason a source file that is not relevant to your executable is
7453 located at the original location, a substitution rule is the only
7454 method available to point @value{GDBN} at the new location.
7456 @cindex @samp{--with-relocated-sources}
7457 @cindex default source path substitution
7458 You can configure a default source path substitution rule by
7459 configuring @value{GDBN} with the
7460 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
7461 should be the name of a directory under @value{GDBN}'s configured
7462 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
7463 directory names in debug information under @var{dir} will be adjusted
7464 automatically if the installed @value{GDBN} is moved to a new
7465 location. This is useful if @value{GDBN}, libraries or executables
7466 with debug information and corresponding source code are being moved
7470 @item directory @var{dirname} @dots{}
7471 @item dir @var{dirname} @dots{}
7472 Add directory @var{dirname} to the front of the source path. Several
7473 directory names may be given to this command, separated by @samp{:}
7474 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
7475 part of absolute file names) or
7476 whitespace. You may specify a directory that is already in the source
7477 path; this moves it forward, so @value{GDBN} searches it sooner.
7481 @vindex $cdir@r{, convenience variable}
7482 @vindex $cwd@r{, convenience variable}
7483 @cindex compilation directory
7484 @cindex current directory
7485 @cindex working directory
7486 @cindex directory, current
7487 @cindex directory, compilation
7488 You can use the string @samp{$cdir} to refer to the compilation
7489 directory (if one is recorded), and @samp{$cwd} to refer to the current
7490 working directory. @samp{$cwd} is not the same as @samp{.}---the former
7491 tracks the current working directory as it changes during your @value{GDBN}
7492 session, while the latter is immediately expanded to the current
7493 directory at the time you add an entry to the source path.
7496 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
7498 @c RET-repeat for @code{directory} is explicitly disabled, but since
7499 @c repeating it would be a no-op we do not say that. (thanks to RMS)
7501 @item set directories @var{path-list}
7502 @kindex set directories
7503 Set the source path to @var{path-list}.
7504 @samp{$cdir:$cwd} are added if missing.
7506 @item show directories
7507 @kindex show directories
7508 Print the source path: show which directories it contains.
7510 @anchor{set substitute-path}
7511 @item set substitute-path @var{from} @var{to}
7512 @kindex set substitute-path
7513 Define a source path substitution rule, and add it at the end of the
7514 current list of existing substitution rules. If a rule with the same
7515 @var{from} was already defined, then the old rule is also deleted.
7517 For example, if the file @file{/foo/bar/baz.c} was moved to
7518 @file{/mnt/cross/baz.c}, then the command
7521 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
7525 will tell @value{GDBN} to replace @samp{/usr/src} with
7526 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
7527 @file{baz.c} even though it was moved.
7529 In the case when more than one substitution rule have been defined,
7530 the rules are evaluated one by one in the order where they have been
7531 defined. The first one matching, if any, is selected to perform
7534 For instance, if we had entered the following commands:
7537 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
7538 (@value{GDBP}) set substitute-path /usr/src /mnt/src
7542 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
7543 @file{/mnt/include/defs.h} by using the first rule. However, it would
7544 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
7545 @file{/mnt/src/lib/foo.c}.
7548 @item unset substitute-path [path]
7549 @kindex unset substitute-path
7550 If a path is specified, search the current list of substitution rules
7551 for a rule that would rewrite that path. Delete that rule if found.
7552 A warning is emitted by the debugger if no rule could be found.
7554 If no path is specified, then all substitution rules are deleted.
7556 @item show substitute-path [path]
7557 @kindex show substitute-path
7558 If a path is specified, then print the source path substitution rule
7559 which would rewrite that path, if any.
7561 If no path is specified, then print all existing source path substitution
7566 If your source path is cluttered with directories that are no longer of
7567 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
7568 versions of source. You can correct the situation as follows:
7572 Use @code{directory} with no argument to reset the source path to its default value.
7575 Use @code{directory} with suitable arguments to reinstall the
7576 directories you want in the source path. You can add all the
7577 directories in one command.
7581 @section Source and Machine Code
7582 @cindex source line and its code address
7584 You can use the command @code{info line} to map source lines to program
7585 addresses (and vice versa), and the command @code{disassemble} to display
7586 a range of addresses as machine instructions. You can use the command
7587 @code{set disassemble-next-line} to set whether to disassemble next
7588 source line when execution stops. When run under @sc{gnu} Emacs
7589 mode, the @code{info line} command causes the arrow to point to the
7590 line specified. Also, @code{info line} prints addresses in symbolic form as
7595 @item info line @var{linespec}
7596 Print the starting and ending addresses of the compiled code for
7597 source line @var{linespec}. You can specify source lines in any of
7598 the ways documented in @ref{Specify Location}.
7601 For example, we can use @code{info line} to discover the location of
7602 the object code for the first line of function
7603 @code{m4_changequote}:
7605 @c FIXME: I think this example should also show the addresses in
7606 @c symbolic form, as they usually would be displayed.
7608 (@value{GDBP}) info line m4_changequote
7609 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
7613 @cindex code address and its source line
7614 We can also inquire (using @code{*@var{addr}} as the form for
7615 @var{linespec}) what source line covers a particular address:
7617 (@value{GDBP}) info line *0x63ff
7618 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
7621 @cindex @code{$_} and @code{info line}
7622 @cindex @code{x} command, default address
7623 @kindex x@r{(examine), and} info line
7624 After @code{info line}, the default address for the @code{x} command
7625 is changed to the starting address of the line, so that @samp{x/i} is
7626 sufficient to begin examining the machine code (@pxref{Memory,
7627 ,Examining Memory}). Also, this address is saved as the value of the
7628 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
7633 @cindex assembly instructions
7634 @cindex instructions, assembly
7635 @cindex machine instructions
7636 @cindex listing machine instructions
7638 @itemx disassemble /m
7639 @itemx disassemble /r
7640 This specialized command dumps a range of memory as machine
7641 instructions. It can also print mixed source+disassembly by specifying
7642 the @code{/m} modifier and print the raw instructions in hex as well as
7643 in symbolic form by specifying the @code{/r}.
7644 The default memory range is the function surrounding the
7645 program counter of the selected frame. A single argument to this
7646 command is a program counter value; @value{GDBN} dumps the function
7647 surrounding this value. When two arguments are given, they should
7648 be separated by a comma, possibly surrounded by whitespace. The
7649 arguments specify a range of addresses to dump, in one of two forms:
7652 @item @var{start},@var{end}
7653 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
7654 @item @var{start},+@var{length}
7655 the addresses from @var{start} (inclusive) to
7656 @code{@var{start}+@var{length}} (exclusive).
7660 When 2 arguments are specified, the name of the function is also
7661 printed (since there could be several functions in the given range).
7663 The argument(s) can be any expression yielding a numeric value, such as
7664 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
7666 If the range of memory being disassembled contains current program counter,
7667 the instruction at that location is shown with a @code{=>} marker.
7670 The following example shows the disassembly of a range of addresses of
7671 HP PA-RISC 2.0 code:
7674 (@value{GDBP}) disas 0x32c4, 0x32e4
7675 Dump of assembler code from 0x32c4 to 0x32e4:
7676 0x32c4 <main+204>: addil 0,dp
7677 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
7678 0x32cc <main+212>: ldil 0x3000,r31
7679 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
7680 0x32d4 <main+220>: ldo 0(r31),rp
7681 0x32d8 <main+224>: addil -0x800,dp
7682 0x32dc <main+228>: ldo 0x588(r1),r26
7683 0x32e0 <main+232>: ldil 0x3000,r31
7684 End of assembler dump.
7687 Here is an example showing mixed source+assembly for Intel x86, when the
7688 program is stopped just after function prologue:
7691 (@value{GDBP}) disas /m main
7692 Dump of assembler code for function main:
7694 0x08048330 <+0>: push %ebp
7695 0x08048331 <+1>: mov %esp,%ebp
7696 0x08048333 <+3>: sub $0x8,%esp
7697 0x08048336 <+6>: and $0xfffffff0,%esp
7698 0x08048339 <+9>: sub $0x10,%esp
7700 6 printf ("Hello.\n");
7701 => 0x0804833c <+12>: movl $0x8048440,(%esp)
7702 0x08048343 <+19>: call 0x8048284 <puts@@plt>
7706 0x08048348 <+24>: mov $0x0,%eax
7707 0x0804834d <+29>: leave
7708 0x0804834e <+30>: ret
7710 End of assembler dump.
7713 Here is another example showing raw instructions in hex for AMD x86-64,
7716 (gdb) disas /r 0x400281,+10
7717 Dump of assembler code from 0x400281 to 0x40028b:
7718 0x0000000000400281: 38 36 cmp %dh,(%rsi)
7719 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
7720 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
7721 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
7722 End of assembler dump.
7725 Addresses cannot be specified as a linespec (@pxref{Specify Location}).
7726 So, for example, if you want to disassemble function @code{bar}
7727 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
7728 and not @samp{disassemble foo.c:bar}.
7730 Some architectures have more than one commonly-used set of instruction
7731 mnemonics or other syntax.
7733 For programs that were dynamically linked and use shared libraries,
7734 instructions that call functions or branch to locations in the shared
7735 libraries might show a seemingly bogus location---it's actually a
7736 location of the relocation table. On some architectures, @value{GDBN}
7737 might be able to resolve these to actual function names.
7740 @kindex set disassembly-flavor
7741 @cindex Intel disassembly flavor
7742 @cindex AT&T disassembly flavor
7743 @item set disassembly-flavor @var{instruction-set}
7744 Select the instruction set to use when disassembling the
7745 program via the @code{disassemble} or @code{x/i} commands.
7747 Currently this command is only defined for the Intel x86 family. You
7748 can set @var{instruction-set} to either @code{intel} or @code{att}.
7749 The default is @code{att}, the AT&T flavor used by default by Unix
7750 assemblers for x86-based targets.
7752 @kindex show disassembly-flavor
7753 @item show disassembly-flavor
7754 Show the current setting of the disassembly flavor.
7758 @kindex set disassemble-next-line
7759 @kindex show disassemble-next-line
7760 @item set disassemble-next-line
7761 @itemx show disassemble-next-line
7762 Control whether or not @value{GDBN} will disassemble the next source
7763 line or instruction when execution stops. If ON, @value{GDBN} will
7764 display disassembly of the next source line when execution of the
7765 program being debugged stops. This is @emph{in addition} to
7766 displaying the source line itself, which @value{GDBN} always does if
7767 possible. If the next source line cannot be displayed for some reason
7768 (e.g., if @value{GDBN} cannot find the source file, or there's no line
7769 info in the debug info), @value{GDBN} will display disassembly of the
7770 next @emph{instruction} instead of showing the next source line. If
7771 AUTO, @value{GDBN} will display disassembly of next instruction only
7772 if the source line cannot be displayed. This setting causes
7773 @value{GDBN} to display some feedback when you step through a function
7774 with no line info or whose source file is unavailable. The default is
7775 OFF, which means never display the disassembly of the next line or
7781 @chapter Examining Data
7783 @cindex printing data
7784 @cindex examining data
7787 The usual way to examine data in your program is with the @code{print}
7788 command (abbreviated @code{p}), or its synonym @code{inspect}. It
7789 evaluates and prints the value of an expression of the language your
7790 program is written in (@pxref{Languages, ,Using @value{GDBN} with
7791 Different Languages}). It may also print the expression using a
7792 Python-based pretty-printer (@pxref{Pretty Printing}).
7795 @item print @var{expr}
7796 @itemx print /@var{f} @var{expr}
7797 @var{expr} is an expression (in the source language). By default the
7798 value of @var{expr} is printed in a format appropriate to its data type;
7799 you can choose a different format by specifying @samp{/@var{f}}, where
7800 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
7804 @itemx print /@var{f}
7805 @cindex reprint the last value
7806 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
7807 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
7808 conveniently inspect the same value in an alternative format.
7811 A more low-level way of examining data is with the @code{x} command.
7812 It examines data in memory at a specified address and prints it in a
7813 specified format. @xref{Memory, ,Examining Memory}.
7815 If you are interested in information about types, or about how the
7816 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
7817 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
7820 @cindex exploring hierarchical data structures
7822 Another way of examining values of expressions and type information is
7823 through the Python extension command @code{explore} (available only if
7824 the @value{GDBN} build is configured with @code{--with-python}). It
7825 offers an interactive way to start at the highest level (or, the most
7826 abstract level) of the data type of an expression (or, the data type
7827 itself) and explore all the way down to leaf scalar values/fields
7828 embedded in the higher level data types.
7831 @item explore @var{arg}
7832 @var{arg} is either an expression (in the source language), or a type
7833 visible in the current context of the program being debugged.
7836 The working of the @code{explore} command can be illustrated with an
7837 example. If a data type @code{struct ComplexStruct} is defined in your
7847 struct ComplexStruct
7849 struct SimpleStruct *ss_p;
7855 followed by variable declarations as
7858 struct SimpleStruct ss = @{ 10, 1.11 @};
7859 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
7863 then, the value of the variable @code{cs} can be explored using the
7864 @code{explore} command as follows.
7868 The value of `cs' is a struct/class of type `struct ComplexStruct' with
7869 the following fields:
7871 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
7872 arr = <Enter 1 to explore this field of type `int [10]'>
7874 Enter the field number of choice:
7878 Since the fields of @code{cs} are not scalar values, you are being
7879 prompted to chose the field you want to explore. Let's say you choose
7880 the field @code{ss_p} by entering @code{0}. Then, since this field is a
7881 pointer, you will be asked if it is pointing to a single value. From
7882 the declaration of @code{cs} above, it is indeed pointing to a single
7883 value, hence you enter @code{y}. If you enter @code{n}, then you will
7884 be asked if it were pointing to an array of values, in which case this
7885 field will be explored as if it were an array.
7888 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
7889 Continue exploring it as a pointer to a single value [y/n]: y
7890 The value of `*(cs.ss_p)' is a struct/class of type `struct
7891 SimpleStruct' with the following fields:
7893 i = 10 .. (Value of type `int')
7894 d = 1.1100000000000001 .. (Value of type `double')
7896 Press enter to return to parent value:
7900 If the field @code{arr} of @code{cs} was chosen for exploration by
7901 entering @code{1} earlier, then since it is as array, you will be
7902 prompted to enter the index of the element in the array that you want
7906 `cs.arr' is an array of `int'.
7907 Enter the index of the element you want to explore in `cs.arr': 5
7909 `(cs.arr)[5]' is a scalar value of type `int'.
7913 Press enter to return to parent value:
7916 In general, at any stage of exploration, you can go deeper towards the
7917 leaf values by responding to the prompts appropriately, or hit the
7918 return key to return to the enclosing data structure (the @i{higher}
7919 level data structure).
7921 Similar to exploring values, you can use the @code{explore} command to
7922 explore types. Instead of specifying a value (which is typically a
7923 variable name or an expression valid in the current context of the
7924 program being debugged), you specify a type name. If you consider the
7925 same example as above, your can explore the type
7926 @code{struct ComplexStruct} by passing the argument
7927 @code{struct ComplexStruct} to the @code{explore} command.
7930 (gdb) explore struct ComplexStruct
7934 By responding to the prompts appropriately in the subsequent interactive
7935 session, you can explore the type @code{struct ComplexStruct} in a
7936 manner similar to how the value @code{cs} was explored in the above
7939 The @code{explore} command also has two sub-commands,
7940 @code{explore value} and @code{explore type}. The former sub-command is
7941 a way to explicitly specify that value exploration of the argument is
7942 being invoked, while the latter is a way to explicitly specify that type
7943 exploration of the argument is being invoked.
7946 @item explore value @var{expr}
7947 @cindex explore value
7948 This sub-command of @code{explore} explores the value of the
7949 expression @var{expr} (if @var{expr} is an expression valid in the
7950 current context of the program being debugged). The behavior of this
7951 command is identical to that of the behavior of the @code{explore}
7952 command being passed the argument @var{expr}.
7954 @item explore type @var{arg}
7955 @cindex explore type
7956 This sub-command of @code{explore} explores the type of @var{arg} (if
7957 @var{arg} is a type visible in the current context of program being
7958 debugged), or the type of the value/expression @var{arg} (if @var{arg}
7959 is an expression valid in the current context of the program being
7960 debugged). If @var{arg} is a type, then the behavior of this command is
7961 identical to that of the @code{explore} command being passed the
7962 argument @var{arg}. If @var{arg} is an expression, then the behavior of
7963 this command will be identical to that of the @code{explore} command
7964 being passed the type of @var{arg} as the argument.
7968 * Expressions:: Expressions
7969 * Ambiguous Expressions:: Ambiguous Expressions
7970 * Variables:: Program variables
7971 * Arrays:: Artificial arrays
7972 * Output Formats:: Output formats
7973 * Memory:: Examining memory
7974 * Auto Display:: Automatic display
7975 * Print Settings:: Print settings
7976 * Pretty Printing:: Python pretty printing
7977 * Value History:: Value history
7978 * Convenience Vars:: Convenience variables
7979 * Convenience Funs:: Convenience functions
7980 * Registers:: Registers
7981 * Floating Point Hardware:: Floating point hardware
7982 * Vector Unit:: Vector Unit
7983 * OS Information:: Auxiliary data provided by operating system
7984 * Memory Region Attributes:: Memory region attributes
7985 * Dump/Restore Files:: Copy between memory and a file
7986 * Core File Generation:: Cause a program dump its core
7987 * Character Sets:: Debugging programs that use a different
7988 character set than GDB does
7989 * Caching Remote Data:: Data caching for remote targets
7990 * Searching Memory:: Searching memory for a sequence of bytes
7994 @section Expressions
7997 @code{print} and many other @value{GDBN} commands accept an expression and
7998 compute its value. Any kind of constant, variable or operator defined
7999 by the programming language you are using is valid in an expression in
8000 @value{GDBN}. This includes conditional expressions, function calls,
8001 casts, and string constants. It also includes preprocessor macros, if
8002 you compiled your program to include this information; see
8005 @cindex arrays in expressions
8006 @value{GDBN} supports array constants in expressions input by
8007 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
8008 you can use the command @code{print @{1, 2, 3@}} to create an array
8009 of three integers. If you pass an array to a function or assign it
8010 to a program variable, @value{GDBN} copies the array to memory that
8011 is @code{malloc}ed in the target program.
8013 Because C is so widespread, most of the expressions shown in examples in
8014 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
8015 Languages}, for information on how to use expressions in other
8018 In this section, we discuss operators that you can use in @value{GDBN}
8019 expressions regardless of your programming language.
8021 @cindex casts, in expressions
8022 Casts are supported in all languages, not just in C, because it is so
8023 useful to cast a number into a pointer in order to examine a structure
8024 at that address in memory.
8025 @c FIXME: casts supported---Mod2 true?
8027 @value{GDBN} supports these operators, in addition to those common
8028 to programming languages:
8032 @samp{@@} is a binary operator for treating parts of memory as arrays.
8033 @xref{Arrays, ,Artificial Arrays}, for more information.
8036 @samp{::} allows you to specify a variable in terms of the file or
8037 function where it is defined. @xref{Variables, ,Program Variables}.
8039 @cindex @{@var{type}@}
8040 @cindex type casting memory
8041 @cindex memory, viewing as typed object
8042 @cindex casts, to view memory
8043 @item @{@var{type}@} @var{addr}
8044 Refers to an object of type @var{type} stored at address @var{addr} in
8045 memory. @var{addr} may be any expression whose value is an integer or
8046 pointer (but parentheses are required around binary operators, just as in
8047 a cast). This construct is allowed regardless of what kind of data is
8048 normally supposed to reside at @var{addr}.
8051 @node Ambiguous Expressions
8052 @section Ambiguous Expressions
8053 @cindex ambiguous expressions
8055 Expressions can sometimes contain some ambiguous elements. For instance,
8056 some programming languages (notably Ada, C@t{++} and Objective-C) permit
8057 a single function name to be defined several times, for application in
8058 different contexts. This is called @dfn{overloading}. Another example
8059 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
8060 templates and is typically instantiated several times, resulting in
8061 the same function name being defined in different contexts.
8063 In some cases and depending on the language, it is possible to adjust
8064 the expression to remove the ambiguity. For instance in C@t{++}, you
8065 can specify the signature of the function you want to break on, as in
8066 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
8067 qualified name of your function often makes the expression unambiguous
8070 When an ambiguity that needs to be resolved is detected, the debugger
8071 has the capability to display a menu of numbered choices for each
8072 possibility, and then waits for the selection with the prompt @samp{>}.
8073 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
8074 aborts the current command. If the command in which the expression was
8075 used allows more than one choice to be selected, the next option in the
8076 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
8079 For example, the following session excerpt shows an attempt to set a
8080 breakpoint at the overloaded symbol @code{String::after}.
8081 We choose three particular definitions of that function name:
8083 @c FIXME! This is likely to change to show arg type lists, at least
8086 (@value{GDBP}) b String::after
8089 [2] file:String.cc; line number:867
8090 [3] file:String.cc; line number:860
8091 [4] file:String.cc; line number:875
8092 [5] file:String.cc; line number:853
8093 [6] file:String.cc; line number:846
8094 [7] file:String.cc; line number:735
8096 Breakpoint 1 at 0xb26c: file String.cc, line 867.
8097 Breakpoint 2 at 0xb344: file String.cc, line 875.
8098 Breakpoint 3 at 0xafcc: file String.cc, line 846.
8099 Multiple breakpoints were set.
8100 Use the "delete" command to delete unwanted
8107 @kindex set multiple-symbols
8108 @item set multiple-symbols @var{mode}
8109 @cindex multiple-symbols menu
8111 This option allows you to adjust the debugger behavior when an expression
8114 By default, @var{mode} is set to @code{all}. If the command with which
8115 the expression is used allows more than one choice, then @value{GDBN}
8116 automatically selects all possible choices. For instance, inserting
8117 a breakpoint on a function using an ambiguous name results in a breakpoint
8118 inserted on each possible match. However, if a unique choice must be made,
8119 then @value{GDBN} uses the menu to help you disambiguate the expression.
8120 For instance, printing the address of an overloaded function will result
8121 in the use of the menu.
8123 When @var{mode} is set to @code{ask}, the debugger always uses the menu
8124 when an ambiguity is detected.
8126 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
8127 an error due to the ambiguity and the command is aborted.
8129 @kindex show multiple-symbols
8130 @item show multiple-symbols
8131 Show the current value of the @code{multiple-symbols} setting.
8135 @section Program Variables
8137 The most common kind of expression to use is the name of a variable
8140 Variables in expressions are understood in the selected stack frame
8141 (@pxref{Selection, ,Selecting a Frame}); they must be either:
8145 global (or file-static)
8152 visible according to the scope rules of the
8153 programming language from the point of execution in that frame
8156 @noindent This means that in the function
8171 you can examine and use the variable @code{a} whenever your program is
8172 executing within the function @code{foo}, but you can only use or
8173 examine the variable @code{b} while your program is executing inside
8174 the block where @code{b} is declared.
8176 @cindex variable name conflict
8177 There is an exception: you can refer to a variable or function whose
8178 scope is a single source file even if the current execution point is not
8179 in this file. But it is possible to have more than one such variable or
8180 function with the same name (in different source files). If that
8181 happens, referring to that name has unpredictable effects. If you wish,
8182 you can specify a static variable in a particular function or file by
8183 using the colon-colon (@code{::}) notation:
8185 @cindex colon-colon, context for variables/functions
8187 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
8188 @cindex @code{::}, context for variables/functions
8191 @var{file}::@var{variable}
8192 @var{function}::@var{variable}
8196 Here @var{file} or @var{function} is the name of the context for the
8197 static @var{variable}. In the case of file names, you can use quotes to
8198 make sure @value{GDBN} parses the file name as a single word---for example,
8199 to print a global value of @code{x} defined in @file{f2.c}:
8202 (@value{GDBP}) p 'f2.c'::x
8205 The @code{::} notation is normally used for referring to
8206 static variables, since you typically disambiguate uses of local variables
8207 in functions by selecting the appropriate frame and using the
8208 simple name of the variable. However, you may also use this notation
8209 to refer to local variables in frames enclosing the selected frame:
8218 process (a); /* Stop here */
8229 For example, if there is a breakpoint at the commented line,
8230 here is what you might see
8231 when the program stops after executing the call @code{bar(0)}:
8236 (@value{GDBP}) p bar::a
8239 #2 0x080483d0 in foo (a=5) at foobar.c:12
8242 (@value{GDBP}) p bar::a
8246 @cindex C@t{++} scope resolution
8247 These uses of @samp{::} are very rarely in conflict with the very similar
8248 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
8249 scope resolution operator in @value{GDBN} expressions.
8250 @c FIXME: Um, so what happens in one of those rare cases where it's in
8253 @cindex wrong values
8254 @cindex variable values, wrong
8255 @cindex function entry/exit, wrong values of variables
8256 @cindex optimized code, wrong values of variables
8258 @emph{Warning:} Occasionally, a local variable may appear to have the
8259 wrong value at certain points in a function---just after entry to a new
8260 scope, and just before exit.
8262 You may see this problem when you are stepping by machine instructions.
8263 This is because, on most machines, it takes more than one instruction to
8264 set up a stack frame (including local variable definitions); if you are
8265 stepping by machine instructions, variables may appear to have the wrong
8266 values until the stack frame is completely built. On exit, it usually
8267 also takes more than one machine instruction to destroy a stack frame;
8268 after you begin stepping through that group of instructions, local
8269 variable definitions may be gone.
8271 This may also happen when the compiler does significant optimizations.
8272 To be sure of always seeing accurate values, turn off all optimization
8275 @cindex ``No symbol "foo" in current context''
8276 Another possible effect of compiler optimizations is to optimize
8277 unused variables out of existence, or assign variables to registers (as
8278 opposed to memory addresses). Depending on the support for such cases
8279 offered by the debug info format used by the compiler, @value{GDBN}
8280 might not be able to display values for such local variables. If that
8281 happens, @value{GDBN} will print a message like this:
8284 No symbol "foo" in current context.
8287 To solve such problems, either recompile without optimizations, or use a
8288 different debug info format, if the compiler supports several such
8289 formats. @xref{Compilation}, for more information on choosing compiler
8290 options. @xref{C, ,C and C@t{++}}, for more information about debug
8291 info formats that are best suited to C@t{++} programs.
8293 If you ask to print an object whose contents are unknown to
8294 @value{GDBN}, e.g., because its data type is not completely specified
8295 by the debug information, @value{GDBN} will say @samp{<incomplete
8296 type>}. @xref{Symbols, incomplete type}, for more about this.
8298 If you append @kbd{@@entry} string to a function parameter name you get its
8299 value at the time the function got called. If the value is not available an
8300 error message is printed. Entry values are available only with some compilers.
8301 Entry values are normally also printed at the function parameter list according
8302 to @ref{set print entry-values}.
8305 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
8311 (gdb) print i@@entry
8315 Strings are identified as arrays of @code{char} values without specified
8316 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
8317 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
8318 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
8319 defines literal string type @code{"char"} as @code{char} without a sign.
8324 signed char var1[] = "A";
8327 You get during debugging
8332 $2 = @{65 'A', 0 '\0'@}
8336 @section Artificial Arrays
8338 @cindex artificial array
8340 @kindex @@@r{, referencing memory as an array}
8341 It is often useful to print out several successive objects of the
8342 same type in memory; a section of an array, or an array of
8343 dynamically determined size for which only a pointer exists in the
8346 You can do this by referring to a contiguous span of memory as an
8347 @dfn{artificial array}, using the binary operator @samp{@@}. The left
8348 operand of @samp{@@} should be the first element of the desired array
8349 and be an individual object. The right operand should be the desired length
8350 of the array. The result is an array value whose elements are all of
8351 the type of the left argument. The first element is actually the left
8352 argument; the second element comes from bytes of memory immediately
8353 following those that hold the first element, and so on. Here is an
8354 example. If a program says
8357 int *array = (int *) malloc (len * sizeof (int));
8361 you can print the contents of @code{array} with
8367 The left operand of @samp{@@} must reside in memory. Array values made
8368 with @samp{@@} in this way behave just like other arrays in terms of
8369 subscripting, and are coerced to pointers when used in expressions.
8370 Artificial arrays most often appear in expressions via the value history
8371 (@pxref{Value History, ,Value History}), after printing one out.
8373 Another way to create an artificial array is to use a cast.
8374 This re-interprets a value as if it were an array.
8375 The value need not be in memory:
8377 (@value{GDBP}) p/x (short[2])0x12345678
8378 $1 = @{0x1234, 0x5678@}
8381 As a convenience, if you leave the array length out (as in
8382 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
8383 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
8385 (@value{GDBP}) p/x (short[])0x12345678
8386 $2 = @{0x1234, 0x5678@}
8389 Sometimes the artificial array mechanism is not quite enough; in
8390 moderately complex data structures, the elements of interest may not
8391 actually be adjacent---for example, if you are interested in the values
8392 of pointers in an array. One useful work-around in this situation is
8393 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
8394 Variables}) as a counter in an expression that prints the first
8395 interesting value, and then repeat that expression via @key{RET}. For
8396 instance, suppose you have an array @code{dtab} of pointers to
8397 structures, and you are interested in the values of a field @code{fv}
8398 in each structure. Here is an example of what you might type:
8408 @node Output Formats
8409 @section Output Formats
8411 @cindex formatted output
8412 @cindex output formats
8413 By default, @value{GDBN} prints a value according to its data type. Sometimes
8414 this is not what you want. For example, you might want to print a number
8415 in hex, or a pointer in decimal. Or you might want to view data in memory
8416 at a certain address as a character string or as an instruction. To do
8417 these things, specify an @dfn{output format} when you print a value.
8419 The simplest use of output formats is to say how to print a value
8420 already computed. This is done by starting the arguments of the
8421 @code{print} command with a slash and a format letter. The format
8422 letters supported are:
8426 Regard the bits of the value as an integer, and print the integer in
8430 Print as integer in signed decimal.
8433 Print as integer in unsigned decimal.
8436 Print as integer in octal.
8439 Print as integer in binary. The letter @samp{t} stands for ``two''.
8440 @footnote{@samp{b} cannot be used because these format letters are also
8441 used with the @code{x} command, where @samp{b} stands for ``byte'';
8442 see @ref{Memory,,Examining Memory}.}
8445 @cindex unknown address, locating
8446 @cindex locate address
8447 Print as an address, both absolute in hexadecimal and as an offset from
8448 the nearest preceding symbol. You can use this format used to discover
8449 where (in what function) an unknown address is located:
8452 (@value{GDBP}) p/a 0x54320
8453 $3 = 0x54320 <_initialize_vx+396>
8457 The command @code{info symbol 0x54320} yields similar results.
8458 @xref{Symbols, info symbol}.
8461 Regard as an integer and print it as a character constant. This
8462 prints both the numerical value and its character representation. The
8463 character representation is replaced with the octal escape @samp{\nnn}
8464 for characters outside the 7-bit @sc{ascii} range.
8466 Without this format, @value{GDBN} displays @code{char},
8467 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
8468 constants. Single-byte members of vectors are displayed as integer
8472 Regard the bits of the value as a floating point number and print
8473 using typical floating point syntax.
8476 @cindex printing strings
8477 @cindex printing byte arrays
8478 Regard as a string, if possible. With this format, pointers to single-byte
8479 data are displayed as null-terminated strings and arrays of single-byte data
8480 are displayed as fixed-length strings. Other values are displayed in their
8483 Without this format, @value{GDBN} displays pointers to and arrays of
8484 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
8485 strings. Single-byte members of a vector are displayed as an integer
8489 @cindex raw printing
8490 Print using the @samp{raw} formatting. By default, @value{GDBN} will
8491 use a Python-based pretty-printer, if one is available (@pxref{Pretty
8492 Printing}). This typically results in a higher-level display of the
8493 value's contents. The @samp{r} format bypasses any Python
8494 pretty-printer which might exist.
8497 For example, to print the program counter in hex (@pxref{Registers}), type
8504 Note that no space is required before the slash; this is because command
8505 names in @value{GDBN} cannot contain a slash.
8507 To reprint the last value in the value history with a different format,
8508 you can use the @code{print} command with just a format and no
8509 expression. For example, @samp{p/x} reprints the last value in hex.
8512 @section Examining Memory
8514 You can use the command @code{x} (for ``examine'') to examine memory in
8515 any of several formats, independently of your program's data types.
8517 @cindex examining memory
8519 @kindex x @r{(examine memory)}
8520 @item x/@var{nfu} @var{addr}
8523 Use the @code{x} command to examine memory.
8526 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
8527 much memory to display and how to format it; @var{addr} is an
8528 expression giving the address where you want to start displaying memory.
8529 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
8530 Several commands set convenient defaults for @var{addr}.
8533 @item @var{n}, the repeat count
8534 The repeat count is a decimal integer; the default is 1. It specifies
8535 how much memory (counting by units @var{u}) to display.
8536 @c This really is **decimal**; unaffected by 'set radix' as of GDB
8539 @item @var{f}, the display format
8540 The display format is one of the formats used by @code{print}
8541 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
8542 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
8543 The default is @samp{x} (hexadecimal) initially. The default changes
8544 each time you use either @code{x} or @code{print}.
8546 @item @var{u}, the unit size
8547 The unit size is any of
8553 Halfwords (two bytes).
8555 Words (four bytes). This is the initial default.
8557 Giant words (eight bytes).
8560 Each time you specify a unit size with @code{x}, that size becomes the
8561 default unit the next time you use @code{x}. For the @samp{i} format,
8562 the unit size is ignored and is normally not written. For the @samp{s} format,
8563 the unit size defaults to @samp{b}, unless it is explicitly given.
8564 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
8565 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
8566 Note that the results depend on the programming language of the
8567 current compilation unit. If the language is C, the @samp{s}
8568 modifier will use the UTF-16 encoding while @samp{w} will use
8569 UTF-32. The encoding is set by the programming language and cannot
8572 @item @var{addr}, starting display address
8573 @var{addr} is the address where you want @value{GDBN} to begin displaying
8574 memory. The expression need not have a pointer value (though it may);
8575 it is always interpreted as an integer address of a byte of memory.
8576 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
8577 @var{addr} is usually just after the last address examined---but several
8578 other commands also set the default address: @code{info breakpoints} (to
8579 the address of the last breakpoint listed), @code{info line} (to the
8580 starting address of a line), and @code{print} (if you use it to display
8581 a value from memory).
8584 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
8585 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
8586 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
8587 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
8588 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
8590 Since the letters indicating unit sizes are all distinct from the
8591 letters specifying output formats, you do not have to remember whether
8592 unit size or format comes first; either order works. The output
8593 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
8594 (However, the count @var{n} must come first; @samp{wx4} does not work.)
8596 Even though the unit size @var{u} is ignored for the formats @samp{s}
8597 and @samp{i}, you might still want to use a count @var{n}; for example,
8598 @samp{3i} specifies that you want to see three machine instructions,
8599 including any operands. For convenience, especially when used with
8600 the @code{display} command, the @samp{i} format also prints branch delay
8601 slot instructions, if any, beyond the count specified, which immediately
8602 follow the last instruction that is within the count. The command
8603 @code{disassemble} gives an alternative way of inspecting machine
8604 instructions; see @ref{Machine Code,,Source and Machine Code}.
8606 All the defaults for the arguments to @code{x} are designed to make it
8607 easy to continue scanning memory with minimal specifications each time
8608 you use @code{x}. For example, after you have inspected three machine
8609 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
8610 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
8611 the repeat count @var{n} is used again; the other arguments default as
8612 for successive uses of @code{x}.
8614 When examining machine instructions, the instruction at current program
8615 counter is shown with a @code{=>} marker. For example:
8618 (@value{GDBP}) x/5i $pc-6
8619 0x804837f <main+11>: mov %esp,%ebp
8620 0x8048381 <main+13>: push %ecx
8621 0x8048382 <main+14>: sub $0x4,%esp
8622 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
8623 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
8626 @cindex @code{$_}, @code{$__}, and value history
8627 The addresses and contents printed by the @code{x} command are not saved
8628 in the value history because there is often too much of them and they
8629 would get in the way. Instead, @value{GDBN} makes these values available for
8630 subsequent use in expressions as values of the convenience variables
8631 @code{$_} and @code{$__}. After an @code{x} command, the last address
8632 examined is available for use in expressions in the convenience variable
8633 @code{$_}. The contents of that address, as examined, are available in
8634 the convenience variable @code{$__}.
8636 If the @code{x} command has a repeat count, the address and contents saved
8637 are from the last memory unit printed; this is not the same as the last
8638 address printed if several units were printed on the last line of output.
8640 @cindex remote memory comparison
8641 @cindex verify remote memory image
8642 When you are debugging a program running on a remote target machine
8643 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
8644 remote machine's memory against the executable file you downloaded to
8645 the target. The @code{compare-sections} command is provided for such
8649 @kindex compare-sections
8650 @item compare-sections @r{[}@var{section-name}@r{]}
8651 Compare the data of a loadable section @var{section-name} in the
8652 executable file of the program being debugged with the same section in
8653 the remote machine's memory, and report any mismatches. With no
8654 arguments, compares all loadable sections. This command's
8655 availability depends on the target's support for the @code{"qCRC"}
8660 @section Automatic Display
8661 @cindex automatic display
8662 @cindex display of expressions
8664 If you find that you want to print the value of an expression frequently
8665 (to see how it changes), you might want to add it to the @dfn{automatic
8666 display list} so that @value{GDBN} prints its value each time your program stops.
8667 Each expression added to the list is given a number to identify it;
8668 to remove an expression from the list, you specify that number.
8669 The automatic display looks like this:
8673 3: bar[5] = (struct hack *) 0x3804
8677 This display shows item numbers, expressions and their current values. As with
8678 displays you request manually using @code{x} or @code{print}, you can
8679 specify the output format you prefer; in fact, @code{display} decides
8680 whether to use @code{print} or @code{x} depending your format
8681 specification---it uses @code{x} if you specify either the @samp{i}
8682 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
8686 @item display @var{expr}
8687 Add the expression @var{expr} to the list of expressions to display
8688 each time your program stops. @xref{Expressions, ,Expressions}.
8690 @code{display} does not repeat if you press @key{RET} again after using it.
8692 @item display/@var{fmt} @var{expr}
8693 For @var{fmt} specifying only a display format and not a size or
8694 count, add the expression @var{expr} to the auto-display list but
8695 arrange to display it each time in the specified format @var{fmt}.
8696 @xref{Output Formats,,Output Formats}.
8698 @item display/@var{fmt} @var{addr}
8699 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
8700 number of units, add the expression @var{addr} as a memory address to
8701 be examined each time your program stops. Examining means in effect
8702 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
8705 For example, @samp{display/i $pc} can be helpful, to see the machine
8706 instruction about to be executed each time execution stops (@samp{$pc}
8707 is a common name for the program counter; @pxref{Registers, ,Registers}).
8710 @kindex delete display
8712 @item undisplay @var{dnums}@dots{}
8713 @itemx delete display @var{dnums}@dots{}
8714 Remove items from the list of expressions to display. Specify the
8715 numbers of the displays that you want affected with the command
8716 argument @var{dnums}. It can be a single display number, one of the
8717 numbers shown in the first field of the @samp{info display} display;
8718 or it could be a range of display numbers, as in @code{2-4}.
8720 @code{undisplay} does not repeat if you press @key{RET} after using it.
8721 (Otherwise you would just get the error @samp{No display number @dots{}}.)
8723 @kindex disable display
8724 @item disable display @var{dnums}@dots{}
8725 Disable the display of item numbers @var{dnums}. A disabled display
8726 item is not printed automatically, but is not forgotten. It may be
8727 enabled again later. Specify the numbers of the displays that you
8728 want affected with the command argument @var{dnums}. It can be a
8729 single display number, one of the numbers shown in the first field of
8730 the @samp{info display} display; or it could be a range of display
8731 numbers, as in @code{2-4}.
8733 @kindex enable display
8734 @item enable display @var{dnums}@dots{}
8735 Enable display of item numbers @var{dnums}. It becomes effective once
8736 again in auto display of its expression, until you specify otherwise.
8737 Specify the numbers of the displays that you want affected with the
8738 command argument @var{dnums}. It can be a single display number, one
8739 of the numbers shown in the first field of the @samp{info display}
8740 display; or it could be a range of display numbers, as in @code{2-4}.
8743 Display the current values of the expressions on the list, just as is
8744 done when your program stops.
8746 @kindex info display
8748 Print the list of expressions previously set up to display
8749 automatically, each one with its item number, but without showing the
8750 values. This includes disabled expressions, which are marked as such.
8751 It also includes expressions which would not be displayed right now
8752 because they refer to automatic variables not currently available.
8755 @cindex display disabled out of scope
8756 If a display expression refers to local variables, then it does not make
8757 sense outside the lexical context for which it was set up. Such an
8758 expression is disabled when execution enters a context where one of its
8759 variables is not defined. For example, if you give the command
8760 @code{display last_char} while inside a function with an argument
8761 @code{last_char}, @value{GDBN} displays this argument while your program
8762 continues to stop inside that function. When it stops elsewhere---where
8763 there is no variable @code{last_char}---the display is disabled
8764 automatically. The next time your program stops where @code{last_char}
8765 is meaningful, you can enable the display expression once again.
8767 @node Print Settings
8768 @section Print Settings
8770 @cindex format options
8771 @cindex print settings
8772 @value{GDBN} provides the following ways to control how arrays, structures,
8773 and symbols are printed.
8776 These settings are useful for debugging programs in any language:
8780 @item set print address
8781 @itemx set print address on
8782 @cindex print/don't print memory addresses
8783 @value{GDBN} prints memory addresses showing the location of stack
8784 traces, structure values, pointer values, breakpoints, and so forth,
8785 even when it also displays the contents of those addresses. The default
8786 is @code{on}. For example, this is what a stack frame display looks like with
8787 @code{set print address on}:
8792 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
8794 530 if (lquote != def_lquote)
8798 @item set print address off
8799 Do not print addresses when displaying their contents. For example,
8800 this is the same stack frame displayed with @code{set print address off}:
8804 (@value{GDBP}) set print addr off
8806 #0 set_quotes (lq="<<", rq=">>") at input.c:530
8807 530 if (lquote != def_lquote)
8811 You can use @samp{set print address off} to eliminate all machine
8812 dependent displays from the @value{GDBN} interface. For example, with
8813 @code{print address off}, you should get the same text for backtraces on
8814 all machines---whether or not they involve pointer arguments.
8817 @item show print address
8818 Show whether or not addresses are to be printed.
8821 When @value{GDBN} prints a symbolic address, it normally prints the
8822 closest earlier symbol plus an offset. If that symbol does not uniquely
8823 identify the address (for example, it is a name whose scope is a single
8824 source file), you may need to clarify. One way to do this is with
8825 @code{info line}, for example @samp{info line *0x4537}. Alternately,
8826 you can set @value{GDBN} to print the source file and line number when
8827 it prints a symbolic address:
8830 @item set print symbol-filename on
8831 @cindex source file and line of a symbol
8832 @cindex symbol, source file and line
8833 Tell @value{GDBN} to print the source file name and line number of a
8834 symbol in the symbolic form of an address.
8836 @item set print symbol-filename off
8837 Do not print source file name and line number of a symbol. This is the
8840 @item show print symbol-filename
8841 Show whether or not @value{GDBN} will print the source file name and
8842 line number of a symbol in the symbolic form of an address.
8845 Another situation where it is helpful to show symbol filenames and line
8846 numbers is when disassembling code; @value{GDBN} shows you the line
8847 number and source file that corresponds to each instruction.
8849 Also, you may wish to see the symbolic form only if the address being
8850 printed is reasonably close to the closest earlier symbol:
8853 @item set print max-symbolic-offset @var{max-offset}
8854 @itemx set print max-symbolic-offset unlimited
8855 @cindex maximum value for offset of closest symbol
8856 Tell @value{GDBN} to only display the symbolic form of an address if the
8857 offset between the closest earlier symbol and the address is less than
8858 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
8859 to always print the symbolic form of an address if any symbol precedes
8860 it. Zero is equivalent to @code{unlimited}.
8862 @item show print max-symbolic-offset
8863 Ask how large the maximum offset is that @value{GDBN} prints in a
8867 @cindex wild pointer, interpreting
8868 @cindex pointer, finding referent
8869 If you have a pointer and you are not sure where it points, try
8870 @samp{set print symbol-filename on}. Then you can determine the name
8871 and source file location of the variable where it points, using
8872 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
8873 For example, here @value{GDBN} shows that a variable @code{ptt} points
8874 at another variable @code{t}, defined in @file{hi2.c}:
8877 (@value{GDBP}) set print symbol-filename on
8878 (@value{GDBP}) p/a ptt
8879 $4 = 0xe008 <t in hi2.c>
8883 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
8884 does not show the symbol name and filename of the referent, even with
8885 the appropriate @code{set print} options turned on.
8888 You can also enable @samp{/a}-like formatting all the time using
8889 @samp{set print symbol on}:
8892 @item set print symbol on
8893 Tell @value{GDBN} to print the symbol corresponding to an address, if
8896 @item set print symbol off
8897 Tell @value{GDBN} not to print the symbol corresponding to an
8898 address. In this mode, @value{GDBN} will still print the symbol
8899 corresponding to pointers to functions. This is the default.
8901 @item show print symbol
8902 Show whether @value{GDBN} will display the symbol corresponding to an
8906 Other settings control how different kinds of objects are printed:
8909 @item set print array
8910 @itemx set print array on
8911 @cindex pretty print arrays
8912 Pretty print arrays. This format is more convenient to read,
8913 but uses more space. The default is off.
8915 @item set print array off
8916 Return to compressed format for arrays.
8918 @item show print array
8919 Show whether compressed or pretty format is selected for displaying
8922 @cindex print array indexes
8923 @item set print array-indexes
8924 @itemx set print array-indexes on
8925 Print the index of each element when displaying arrays. May be more
8926 convenient to locate a given element in the array or quickly find the
8927 index of a given element in that printed array. The default is off.
8929 @item set print array-indexes off
8930 Stop printing element indexes when displaying arrays.
8932 @item show print array-indexes
8933 Show whether the index of each element is printed when displaying
8936 @item set print elements @var{number-of-elements}
8937 @itemx set print elements unlimited
8938 @cindex number of array elements to print
8939 @cindex limit on number of printed array elements
8940 Set a limit on how many elements of an array @value{GDBN} will print.
8941 If @value{GDBN} is printing a large array, it stops printing after it has
8942 printed the number of elements set by the @code{set print elements} command.
8943 This limit also applies to the display of strings.
8944 When @value{GDBN} starts, this limit is set to 200.
8945 Setting @var{number-of-elements} to @code{unlimited} or zero means
8946 that the number of elements to print is unlimited.
8948 @item show print elements
8949 Display the number of elements of a large array that @value{GDBN} will print.
8950 If the number is 0, then the printing is unlimited.
8952 @item set print frame-arguments @var{value}
8953 @kindex set print frame-arguments
8954 @cindex printing frame argument values
8955 @cindex print all frame argument values
8956 @cindex print frame argument values for scalars only
8957 @cindex do not print frame argument values
8958 This command allows to control how the values of arguments are printed
8959 when the debugger prints a frame (@pxref{Frames}). The possible
8964 The values of all arguments are printed.
8967 Print the value of an argument only if it is a scalar. The value of more
8968 complex arguments such as arrays, structures, unions, etc, is replaced
8969 by @code{@dots{}}. This is the default. Here is an example where
8970 only scalar arguments are shown:
8973 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
8978 None of the argument values are printed. Instead, the value of each argument
8979 is replaced by @code{@dots{}}. In this case, the example above now becomes:
8982 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
8987 By default, only scalar arguments are printed. This command can be used
8988 to configure the debugger to print the value of all arguments, regardless
8989 of their type. However, it is often advantageous to not print the value
8990 of more complex parameters. For instance, it reduces the amount of
8991 information printed in each frame, making the backtrace more readable.
8992 Also, it improves performance when displaying Ada frames, because
8993 the computation of large arguments can sometimes be CPU-intensive,
8994 especially in large applications. Setting @code{print frame-arguments}
8995 to @code{scalars} (the default) or @code{none} avoids this computation,
8996 thus speeding up the display of each Ada frame.
8998 @item show print frame-arguments
8999 Show how the value of arguments should be displayed when printing a frame.
9001 @anchor{set print entry-values}
9002 @item set print entry-values @var{value}
9003 @kindex set print entry-values
9004 Set printing of frame argument values at function entry. In some cases
9005 @value{GDBN} can determine the value of function argument which was passed by
9006 the function caller, even if the value was modified inside the called function
9007 and therefore is different. With optimized code, the current value could be
9008 unavailable, but the entry value may still be known.
9010 The default value is @code{default} (see below for its description). Older
9011 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
9012 this feature will behave in the @code{default} setting the same way as with the
9015 This functionality is currently supported only by DWARF 2 debugging format and
9016 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
9017 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
9020 The @var{value} parameter can be one of the following:
9024 Print only actual parameter values, never print values from function entry
9028 #0 different (val=6)
9029 #0 lost (val=<optimized out>)
9031 #0 invalid (val=<optimized out>)
9035 Print only parameter values from function entry point. The actual parameter
9036 values are never printed.
9038 #0 equal (val@@entry=5)
9039 #0 different (val@@entry=5)
9040 #0 lost (val@@entry=5)
9041 #0 born (val@@entry=<optimized out>)
9042 #0 invalid (val@@entry=<optimized out>)
9046 Print only parameter values from function entry point. If value from function
9047 entry point is not known while the actual value is known, print the actual
9048 value for such parameter.
9050 #0 equal (val@@entry=5)
9051 #0 different (val@@entry=5)
9052 #0 lost (val@@entry=5)
9054 #0 invalid (val@@entry=<optimized out>)
9058 Print actual parameter values. If actual parameter value is not known while
9059 value from function entry point is known, print the entry point value for such
9063 #0 different (val=6)
9064 #0 lost (val@@entry=5)
9066 #0 invalid (val=<optimized out>)
9070 Always print both the actual parameter value and its value from function entry
9071 point, even if values of one or both are not available due to compiler
9074 #0 equal (val=5, val@@entry=5)
9075 #0 different (val=6, val@@entry=5)
9076 #0 lost (val=<optimized out>, val@@entry=5)
9077 #0 born (val=10, val@@entry=<optimized out>)
9078 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
9082 Print the actual parameter value if it is known and also its value from
9083 function entry point if it is known. If neither is known, print for the actual
9084 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
9085 values are known and identical, print the shortened
9086 @code{param=param@@entry=VALUE} notation.
9088 #0 equal (val=val@@entry=5)
9089 #0 different (val=6, val@@entry=5)
9090 #0 lost (val@@entry=5)
9092 #0 invalid (val=<optimized out>)
9096 Always print the actual parameter value. Print also its value from function
9097 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
9098 if both values are known and identical, print the shortened
9099 @code{param=param@@entry=VALUE} notation.
9101 #0 equal (val=val@@entry=5)
9102 #0 different (val=6, val@@entry=5)
9103 #0 lost (val=<optimized out>, val@@entry=5)
9105 #0 invalid (val=<optimized out>)
9109 For analysis messages on possible failures of frame argument values at function
9110 entry resolution see @ref{set debug entry-values}.
9112 @item show print entry-values
9113 Show the method being used for printing of frame argument values at function
9116 @item set print repeats @var{number-of-repeats}
9117 @itemx set print repeats unlimited
9118 @cindex repeated array elements
9119 Set the threshold for suppressing display of repeated array
9120 elements. When the number of consecutive identical elements of an
9121 array exceeds the threshold, @value{GDBN} prints the string
9122 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
9123 identical repetitions, instead of displaying the identical elements
9124 themselves. Setting the threshold to @code{unlimited} or zero will
9125 cause all elements to be individually printed. The default threshold
9128 @item show print repeats
9129 Display the current threshold for printing repeated identical
9132 @item set print null-stop
9133 @cindex @sc{null} elements in arrays
9134 Cause @value{GDBN} to stop printing the characters of an array when the first
9135 @sc{null} is encountered. This is useful when large arrays actually
9136 contain only short strings.
9139 @item show print null-stop
9140 Show whether @value{GDBN} stops printing an array on the first
9141 @sc{null} character.
9143 @item set print pretty on
9144 @cindex print structures in indented form
9145 @cindex indentation in structure display
9146 Cause @value{GDBN} to print structures in an indented format with one member
9147 per line, like this:
9162 @item set print pretty off
9163 Cause @value{GDBN} to print structures in a compact format, like this:
9167 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
9168 meat = 0x54 "Pork"@}
9173 This is the default format.
9175 @item show print pretty
9176 Show which format @value{GDBN} is using to print structures.
9178 @item set print sevenbit-strings on
9179 @cindex eight-bit characters in strings
9180 @cindex octal escapes in strings
9181 Print using only seven-bit characters; if this option is set,
9182 @value{GDBN} displays any eight-bit characters (in strings or
9183 character values) using the notation @code{\}@var{nnn}. This setting is
9184 best if you are working in English (@sc{ascii}) and you use the
9185 high-order bit of characters as a marker or ``meta'' bit.
9187 @item set print sevenbit-strings off
9188 Print full eight-bit characters. This allows the use of more
9189 international character sets, and is the default.
9191 @item show print sevenbit-strings
9192 Show whether or not @value{GDBN} is printing only seven-bit characters.
9194 @item set print union on
9195 @cindex unions in structures, printing
9196 Tell @value{GDBN} to print unions which are contained in structures
9197 and other unions. This is the default setting.
9199 @item set print union off
9200 Tell @value{GDBN} not to print unions which are contained in
9201 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
9204 @item show print union
9205 Ask @value{GDBN} whether or not it will print unions which are contained in
9206 structures and other unions.
9208 For example, given the declarations
9211 typedef enum @{Tree, Bug@} Species;
9212 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
9213 typedef enum @{Caterpillar, Cocoon, Butterfly@}
9224 struct thing foo = @{Tree, @{Acorn@}@};
9228 with @code{set print union on} in effect @samp{p foo} would print
9231 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
9235 and with @code{set print union off} in effect it would print
9238 $1 = @{it = Tree, form = @{...@}@}
9242 @code{set print union} affects programs written in C-like languages
9248 These settings are of interest when debugging C@t{++} programs:
9251 @cindex demangling C@t{++} names
9252 @item set print demangle
9253 @itemx set print demangle on
9254 Print C@t{++} names in their source form rather than in the encoded
9255 (``mangled'') form passed to the assembler and linker for type-safe
9256 linkage. The default is on.
9258 @item show print demangle
9259 Show whether C@t{++} names are printed in mangled or demangled form.
9261 @item set print asm-demangle
9262 @itemx set print asm-demangle on
9263 Print C@t{++} names in their source form rather than their mangled form, even
9264 in assembler code printouts such as instruction disassemblies.
9267 @item show print asm-demangle
9268 Show whether C@t{++} names in assembly listings are printed in mangled
9271 @cindex C@t{++} symbol decoding style
9272 @cindex symbol decoding style, C@t{++}
9273 @kindex set demangle-style
9274 @item set demangle-style @var{style}
9275 Choose among several encoding schemes used by different compilers to
9276 represent C@t{++} names. The choices for @var{style} are currently:
9280 Allow @value{GDBN} to choose a decoding style by inspecting your program.
9281 This is the default.
9284 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
9287 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
9290 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
9293 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
9294 @strong{Warning:} this setting alone is not sufficient to allow
9295 debugging @code{cfront}-generated executables. @value{GDBN} would
9296 require further enhancement to permit that.
9299 If you omit @var{style}, you will see a list of possible formats.
9301 @item show demangle-style
9302 Display the encoding style currently in use for decoding C@t{++} symbols.
9304 @item set print object
9305 @itemx set print object on
9306 @cindex derived type of an object, printing
9307 @cindex display derived types
9308 When displaying a pointer to an object, identify the @emph{actual}
9309 (derived) type of the object rather than the @emph{declared} type, using
9310 the virtual function table. Note that the virtual function table is
9311 required---this feature can only work for objects that have run-time
9312 type identification; a single virtual method in the object's declared
9313 type is sufficient. Note that this setting is also taken into account when
9314 working with variable objects via MI (@pxref{GDB/MI}).
9316 @item set print object off
9317 Display only the declared type of objects, without reference to the
9318 virtual function table. This is the default setting.
9320 @item show print object
9321 Show whether actual, or declared, object types are displayed.
9323 @item set print static-members
9324 @itemx set print static-members on
9325 @cindex static members of C@t{++} objects
9326 Print static members when displaying a C@t{++} object. The default is on.
9328 @item set print static-members off
9329 Do not print static members when displaying a C@t{++} object.
9331 @item show print static-members
9332 Show whether C@t{++} static members are printed or not.
9334 @item set print pascal_static-members
9335 @itemx set print pascal_static-members on
9336 @cindex static members of Pascal objects
9337 @cindex Pascal objects, static members display
9338 Print static members when displaying a Pascal object. The default is on.
9340 @item set print pascal_static-members off
9341 Do not print static members when displaying a Pascal object.
9343 @item show print pascal_static-members
9344 Show whether Pascal static members are printed or not.
9346 @c These don't work with HP ANSI C++ yet.
9347 @item set print vtbl
9348 @itemx set print vtbl on
9349 @cindex pretty print C@t{++} virtual function tables
9350 @cindex virtual functions (C@t{++}) display
9351 @cindex VTBL display
9352 Pretty print C@t{++} virtual function tables. The default is off.
9353 (The @code{vtbl} commands do not work on programs compiled with the HP
9354 ANSI C@t{++} compiler (@code{aCC}).)
9356 @item set print vtbl off
9357 Do not pretty print C@t{++} virtual function tables.
9359 @item show print vtbl
9360 Show whether C@t{++} virtual function tables are pretty printed, or not.
9363 @node Pretty Printing
9364 @section Pretty Printing
9366 @value{GDBN} provides a mechanism to allow pretty-printing of values using
9367 Python code. It greatly simplifies the display of complex objects. This
9368 mechanism works for both MI and the CLI.
9371 * Pretty-Printer Introduction:: Introduction to pretty-printers
9372 * Pretty-Printer Example:: An example pretty-printer
9373 * Pretty-Printer Commands:: Pretty-printer commands
9376 @node Pretty-Printer Introduction
9377 @subsection Pretty-Printer Introduction
9379 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
9380 registered for the value. If there is then @value{GDBN} invokes the
9381 pretty-printer to print the value. Otherwise the value is printed normally.
9383 Pretty-printers are normally named. This makes them easy to manage.
9384 The @samp{info pretty-printer} command will list all the installed
9385 pretty-printers with their names.
9386 If a pretty-printer can handle multiple data types, then its
9387 @dfn{subprinters} are the printers for the individual data types.
9388 Each such subprinter has its own name.
9389 The format of the name is @var{printer-name};@var{subprinter-name}.
9391 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
9392 Typically they are automatically loaded and registered when the corresponding
9393 debug information is loaded, thus making them available without having to
9394 do anything special.
9396 There are three places where a pretty-printer can be registered.
9400 Pretty-printers registered globally are available when debugging
9404 Pretty-printers registered with a program space are available only
9405 when debugging that program.
9406 @xref{Progspaces In Python}, for more details on program spaces in Python.
9409 Pretty-printers registered with an objfile are loaded and unloaded
9410 with the corresponding objfile (e.g., shared library).
9411 @xref{Objfiles In Python}, for more details on objfiles in Python.
9414 @xref{Selecting Pretty-Printers}, for further information on how
9415 pretty-printers are selected,
9417 @xref{Writing a Pretty-Printer}, for implementing pretty printers
9420 @node Pretty-Printer Example
9421 @subsection Pretty-Printer Example
9423 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
9426 (@value{GDBP}) print s
9428 static npos = 4294967295,
9430 <std::allocator<char>> = @{
9431 <__gnu_cxx::new_allocator<char>> = @{
9432 <No data fields>@}, <No data fields>
9434 members of std::basic_string<char, std::char_traits<char>,
9435 std::allocator<char> >::_Alloc_hider:
9436 _M_p = 0x804a014 "abcd"
9441 With a pretty-printer for @code{std::string} only the contents are printed:
9444 (@value{GDBP}) print s
9448 @node Pretty-Printer Commands
9449 @subsection Pretty-Printer Commands
9450 @cindex pretty-printer commands
9453 @kindex info pretty-printer
9454 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9455 Print the list of installed pretty-printers.
9456 This includes disabled pretty-printers, which are marked as such.
9458 @var{object-regexp} is a regular expression matching the objects
9459 whose pretty-printers to list.
9460 Objects can be @code{global}, the program space's file
9461 (@pxref{Progspaces In Python}),
9462 and the object files within that program space (@pxref{Objfiles In Python}).
9463 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
9464 looks up a printer from these three objects.
9466 @var{name-regexp} is a regular expression matching the name of the printers
9469 @kindex disable pretty-printer
9470 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9471 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9472 A disabled pretty-printer is not forgotten, it may be enabled again later.
9474 @kindex enable pretty-printer
9475 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9476 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9481 Suppose we have three pretty-printers installed: one from library1.so
9482 named @code{foo} that prints objects of type @code{foo}, and
9483 another from library2.so named @code{bar} that prints two types of objects,
9484 @code{bar1} and @code{bar2}.
9487 (gdb) info pretty-printer
9494 (gdb) info pretty-printer library2
9499 (gdb) disable pretty-printer library1
9501 2 of 3 printers enabled
9502 (gdb) info pretty-printer
9509 (gdb) disable pretty-printer library2 bar:bar1
9511 1 of 3 printers enabled
9512 (gdb) info pretty-printer library2
9519 (gdb) disable pretty-printer library2 bar
9521 0 of 3 printers enabled
9522 (gdb) info pretty-printer library2
9531 Note that for @code{bar} the entire printer can be disabled,
9532 as can each individual subprinter.
9535 @section Value History
9537 @cindex value history
9538 @cindex history of values printed by @value{GDBN}
9539 Values printed by the @code{print} command are saved in the @value{GDBN}
9540 @dfn{value history}. This allows you to refer to them in other expressions.
9541 Values are kept until the symbol table is re-read or discarded
9542 (for example with the @code{file} or @code{symbol-file} commands).
9543 When the symbol table changes, the value history is discarded,
9544 since the values may contain pointers back to the types defined in the
9549 @cindex history number
9550 The values printed are given @dfn{history numbers} by which you can
9551 refer to them. These are successive integers starting with one.
9552 @code{print} shows you the history number assigned to a value by
9553 printing @samp{$@var{num} = } before the value; here @var{num} is the
9556 To refer to any previous value, use @samp{$} followed by the value's
9557 history number. The way @code{print} labels its output is designed to
9558 remind you of this. Just @code{$} refers to the most recent value in
9559 the history, and @code{$$} refers to the value before that.
9560 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
9561 is the value just prior to @code{$$}, @code{$$1} is equivalent to
9562 @code{$$}, and @code{$$0} is equivalent to @code{$}.
9564 For example, suppose you have just printed a pointer to a structure and
9565 want to see the contents of the structure. It suffices to type
9571 If you have a chain of structures where the component @code{next} points
9572 to the next one, you can print the contents of the next one with this:
9579 You can print successive links in the chain by repeating this
9580 command---which you can do by just typing @key{RET}.
9582 Note that the history records values, not expressions. If the value of
9583 @code{x} is 4 and you type these commands:
9591 then the value recorded in the value history by the @code{print} command
9592 remains 4 even though the value of @code{x} has changed.
9597 Print the last ten values in the value history, with their item numbers.
9598 This is like @samp{p@ $$9} repeated ten times, except that @code{show
9599 values} does not change the history.
9601 @item show values @var{n}
9602 Print ten history values centered on history item number @var{n}.
9605 Print ten history values just after the values last printed. If no more
9606 values are available, @code{show values +} produces no display.
9609 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
9610 same effect as @samp{show values +}.
9612 @node Convenience Vars
9613 @section Convenience Variables
9615 @cindex convenience variables
9616 @cindex user-defined variables
9617 @value{GDBN} provides @dfn{convenience variables} that you can use within
9618 @value{GDBN} to hold on to a value and refer to it later. These variables
9619 exist entirely within @value{GDBN}; they are not part of your program, and
9620 setting a convenience variable has no direct effect on further execution
9621 of your program. That is why you can use them freely.
9623 Convenience variables are prefixed with @samp{$}. Any name preceded by
9624 @samp{$} can be used for a convenience variable, unless it is one of
9625 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
9626 (Value history references, in contrast, are @emph{numbers} preceded
9627 by @samp{$}. @xref{Value History, ,Value History}.)
9629 You can save a value in a convenience variable with an assignment
9630 expression, just as you would set a variable in your program.
9634 set $foo = *object_ptr
9638 would save in @code{$foo} the value contained in the object pointed to by
9641 Using a convenience variable for the first time creates it, but its
9642 value is @code{void} until you assign a new value. You can alter the
9643 value with another assignment at any time.
9645 Convenience variables have no fixed types. You can assign a convenience
9646 variable any type of value, including structures and arrays, even if
9647 that variable already has a value of a different type. The convenience
9648 variable, when used as an expression, has the type of its current value.
9651 @kindex show convenience
9652 @cindex show all user variables and functions
9653 @item show convenience
9654 Print a list of convenience variables used so far, and their values,
9655 as well as a list of the convenience functions.
9656 Abbreviated @code{show conv}.
9658 @kindex init-if-undefined
9659 @cindex convenience variables, initializing
9660 @item init-if-undefined $@var{variable} = @var{expression}
9661 Set a convenience variable if it has not already been set. This is useful
9662 for user-defined commands that keep some state. It is similar, in concept,
9663 to using local static variables with initializers in C (except that
9664 convenience variables are global). It can also be used to allow users to
9665 override default values used in a command script.
9667 If the variable is already defined then the expression is not evaluated so
9668 any side-effects do not occur.
9671 One of the ways to use a convenience variable is as a counter to be
9672 incremented or a pointer to be advanced. For example, to print
9673 a field from successive elements of an array of structures:
9677 print bar[$i++]->contents
9681 Repeat that command by typing @key{RET}.
9683 Some convenience variables are created automatically by @value{GDBN} and given
9684 values likely to be useful.
9687 @vindex $_@r{, convenience variable}
9689 The variable @code{$_} is automatically set by the @code{x} command to
9690 the last address examined (@pxref{Memory, ,Examining Memory}). Other
9691 commands which provide a default address for @code{x} to examine also
9692 set @code{$_} to that address; these commands include @code{info line}
9693 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
9694 except when set by the @code{x} command, in which case it is a pointer
9695 to the type of @code{$__}.
9697 @vindex $__@r{, convenience variable}
9699 The variable @code{$__} is automatically set by the @code{x} command
9700 to the value found in the last address examined. Its type is chosen
9701 to match the format in which the data was printed.
9704 @vindex $_exitcode@r{, convenience variable}
9705 The variable @code{$_exitcode} is automatically set to the exit code when
9706 the program being debugged terminates.
9709 The variable @code{$_exception} is set to the exception object being
9710 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
9713 @itemx $_probe_arg0@dots{}$_probe_arg11
9714 Arguments to a static probe. @xref{Static Probe Points}.
9717 @vindex $_sdata@r{, inspect, convenience variable}
9718 The variable @code{$_sdata} contains extra collected static tracepoint
9719 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
9720 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
9721 if extra static tracepoint data has not been collected.
9724 @vindex $_siginfo@r{, convenience variable}
9725 The variable @code{$_siginfo} contains extra signal information
9726 (@pxref{extra signal information}). Note that @code{$_siginfo}
9727 could be empty, if the application has not yet received any signals.
9728 For example, it will be empty before you execute the @code{run} command.
9731 @vindex $_tlb@r{, convenience variable}
9732 The variable @code{$_tlb} is automatically set when debugging
9733 applications running on MS-Windows in native mode or connected to
9734 gdbserver that supports the @code{qGetTIBAddr} request.
9735 @xref{General Query Packets}.
9736 This variable contains the address of the thread information block.
9740 On HP-UX systems, if you refer to a function or variable name that
9741 begins with a dollar sign, @value{GDBN} searches for a user or system
9742 name first, before it searches for a convenience variable.
9744 @node Convenience Funs
9745 @section Convenience Functions
9747 @cindex convenience functions
9748 @value{GDBN} also supplies some @dfn{convenience functions}. These
9749 have a syntax similar to convenience variables. A convenience
9750 function can be used in an expression just like an ordinary function;
9751 however, a convenience function is implemented internally to
9754 These functions require @value{GDBN} to be configured with
9755 @code{Python} support.
9759 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
9760 @findex $_memeq@r{, convenience function}
9761 Returns one if the @var{length} bytes at the addresses given by
9762 @var{buf1} and @var{buf2} are equal.
9763 Otherwise it returns zero.
9765 @item $_regex(@var{str}, @var{regex})
9766 @findex $_regex@r{, convenience function}
9767 Returns one if the string @var{str} matches the regular expression
9768 @var{regex}. Otherwise it returns zero.
9769 The syntax of the regular expression is that specified by @code{Python}'s
9770 regular expression support.
9772 @item $_streq(@var{str1}, @var{str2})
9773 @findex $_streq@r{, convenience function}
9774 Returns one if the strings @var{str1} and @var{str2} are equal.
9775 Otherwise it returns zero.
9777 @item $_strlen(@var{str})
9778 @findex $_strlen@r{, convenience function}
9779 Returns the length of string @var{str}.
9783 @value{GDBN} provides the ability to list and get help on
9784 convenience functions.
9788 @kindex help function
9789 @cindex show all convenience functions
9790 Print a list of all convenience functions.
9797 You can refer to machine register contents, in expressions, as variables
9798 with names starting with @samp{$}. The names of registers are different
9799 for each machine; use @code{info registers} to see the names used on
9803 @kindex info registers
9804 @item info registers
9805 Print the names and values of all registers except floating-point
9806 and vector registers (in the selected stack frame).
9808 @kindex info all-registers
9809 @cindex floating point registers
9810 @item info all-registers
9811 Print the names and values of all registers, including floating-point
9812 and vector registers (in the selected stack frame).
9814 @item info registers @var{regname} @dots{}
9815 Print the @dfn{relativized} value of each specified register @var{regname}.
9816 As discussed in detail below, register values are normally relative to
9817 the selected stack frame. @var{regname} may be any register name valid on
9818 the machine you are using, with or without the initial @samp{$}.
9821 @cindex stack pointer register
9822 @cindex program counter register
9823 @cindex process status register
9824 @cindex frame pointer register
9825 @cindex standard registers
9826 @value{GDBN} has four ``standard'' register names that are available (in
9827 expressions) on most machines---whenever they do not conflict with an
9828 architecture's canonical mnemonics for registers. The register names
9829 @code{$pc} and @code{$sp} are used for the program counter register and
9830 the stack pointer. @code{$fp} is used for a register that contains a
9831 pointer to the current stack frame, and @code{$ps} is used for a
9832 register that contains the processor status. For example,
9833 you could print the program counter in hex with
9840 or print the instruction to be executed next with
9847 or add four to the stack pointer@footnote{This is a way of removing
9848 one word from the stack, on machines where stacks grow downward in
9849 memory (most machines, nowadays). This assumes that the innermost
9850 stack frame is selected; setting @code{$sp} is not allowed when other
9851 stack frames are selected. To pop entire frames off the stack,
9852 regardless of machine architecture, use @code{return};
9853 see @ref{Returning, ,Returning from a Function}.} with
9859 Whenever possible, these four standard register names are available on
9860 your machine even though the machine has different canonical mnemonics,
9861 so long as there is no conflict. The @code{info registers} command
9862 shows the canonical names. For example, on the SPARC, @code{info
9863 registers} displays the processor status register as @code{$psr} but you
9864 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
9865 is an alias for the @sc{eflags} register.
9867 @value{GDBN} always considers the contents of an ordinary register as an
9868 integer when the register is examined in this way. Some machines have
9869 special registers which can hold nothing but floating point; these
9870 registers are considered to have floating point values. There is no way
9871 to refer to the contents of an ordinary register as floating point value
9872 (although you can @emph{print} it as a floating point value with
9873 @samp{print/f $@var{regname}}).
9875 Some registers have distinct ``raw'' and ``virtual'' data formats. This
9876 means that the data format in which the register contents are saved by
9877 the operating system is not the same one that your program normally
9878 sees. For example, the registers of the 68881 floating point
9879 coprocessor are always saved in ``extended'' (raw) format, but all C
9880 programs expect to work with ``double'' (virtual) format. In such
9881 cases, @value{GDBN} normally works with the virtual format only (the format
9882 that makes sense for your program), but the @code{info registers} command
9883 prints the data in both formats.
9885 @cindex SSE registers (x86)
9886 @cindex MMX registers (x86)
9887 Some machines have special registers whose contents can be interpreted
9888 in several different ways. For example, modern x86-based machines
9889 have SSE and MMX registers that can hold several values packed
9890 together in several different formats. @value{GDBN} refers to such
9891 registers in @code{struct} notation:
9894 (@value{GDBP}) print $xmm1
9896 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
9897 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
9898 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
9899 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
9900 v4_int32 = @{0, 20657912, 11, 13@},
9901 v2_int64 = @{88725056443645952, 55834574859@},
9902 uint128 = 0x0000000d0000000b013b36f800000000
9907 To set values of such registers, you need to tell @value{GDBN} which
9908 view of the register you wish to change, as if you were assigning
9909 value to a @code{struct} member:
9912 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
9915 Normally, register values are relative to the selected stack frame
9916 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
9917 value that the register would contain if all stack frames farther in
9918 were exited and their saved registers restored. In order to see the
9919 true contents of hardware registers, you must select the innermost
9920 frame (with @samp{frame 0}).
9922 However, @value{GDBN} must deduce where registers are saved, from the machine
9923 code generated by your compiler. If some registers are not saved, or if
9924 @value{GDBN} is unable to locate the saved registers, the selected stack
9925 frame makes no difference.
9927 @node Floating Point Hardware
9928 @section Floating Point Hardware
9929 @cindex floating point
9931 Depending on the configuration, @value{GDBN} may be able to give
9932 you more information about the status of the floating point hardware.
9937 Display hardware-dependent information about the floating
9938 point unit. The exact contents and layout vary depending on the
9939 floating point chip. Currently, @samp{info float} is supported on
9940 the ARM and x86 machines.
9944 @section Vector Unit
9947 Depending on the configuration, @value{GDBN} may be able to give you
9948 more information about the status of the vector unit.
9953 Display information about the vector unit. The exact contents and
9954 layout vary depending on the hardware.
9957 @node OS Information
9958 @section Operating System Auxiliary Information
9959 @cindex OS information
9961 @value{GDBN} provides interfaces to useful OS facilities that can help
9962 you debug your program.
9964 @cindex auxiliary vector
9965 @cindex vector, auxiliary
9966 Some operating systems supply an @dfn{auxiliary vector} to programs at
9967 startup. This is akin to the arguments and environment that you
9968 specify for a program, but contains a system-dependent variety of
9969 binary values that tell system libraries important details about the
9970 hardware, operating system, and process. Each value's purpose is
9971 identified by an integer tag; the meanings are well-known but system-specific.
9972 Depending on the configuration and operating system facilities,
9973 @value{GDBN} may be able to show you this information. For remote
9974 targets, this functionality may further depend on the remote stub's
9975 support of the @samp{qXfer:auxv:read} packet, see
9976 @ref{qXfer auxiliary vector read}.
9981 Display the auxiliary vector of the inferior, which can be either a
9982 live process or a core dump file. @value{GDBN} prints each tag value
9983 numerically, and also shows names and text descriptions for recognized
9984 tags. Some values in the vector are numbers, some bit masks, and some
9985 pointers to strings or other data. @value{GDBN} displays each value in the
9986 most appropriate form for a recognized tag, and in hexadecimal for
9987 an unrecognized tag.
9990 On some targets, @value{GDBN} can access operating system-specific
9991 information and show it to you. The types of information available
9992 will differ depending on the type of operating system running on the
9993 target. The mechanism used to fetch the data is described in
9994 @ref{Operating System Information}. For remote targets, this
9995 functionality depends on the remote stub's support of the
9996 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
10000 @item info os @var{infotype}
10002 Display OS information of the requested type.
10004 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
10006 @anchor{linux info os infotypes}
10008 @kindex info os processes
10010 Display the list of processes on the target. For each process,
10011 @value{GDBN} prints the process identifier, the name of the user, the
10012 command corresponding to the process, and the list of processor cores
10013 that the process is currently running on. (To understand what these
10014 properties mean, for this and the following info types, please consult
10015 the general @sc{gnu}/Linux documentation.)
10017 @kindex info os procgroups
10019 Display the list of process groups on the target. For each process,
10020 @value{GDBN} prints the identifier of the process group that it belongs
10021 to, the command corresponding to the process group leader, the process
10022 identifier, and the command line of the process. The list is sorted
10023 first by the process group identifier, then by the process identifier,
10024 so that processes belonging to the same process group are grouped together
10025 and the process group leader is listed first.
10027 @kindex info os threads
10029 Display the list of threads running on the target. For each thread,
10030 @value{GDBN} prints the identifier of the process that the thread
10031 belongs to, the command of the process, the thread identifier, and the
10032 processor core that it is currently running on. The main thread of a
10033 process is not listed.
10035 @kindex info os files
10037 Display the list of open file descriptors on the target. For each
10038 file descriptor, @value{GDBN} prints the identifier of the process
10039 owning the descriptor, the command of the owning process, the value
10040 of the descriptor, and the target of the descriptor.
10042 @kindex info os sockets
10044 Display the list of Internet-domain sockets on the target. For each
10045 socket, @value{GDBN} prints the address and port of the local and
10046 remote endpoints, the current state of the connection, the creator of
10047 the socket, the IP address family of the socket, and the type of the
10050 @kindex info os shm
10052 Display the list of all System V shared-memory regions on the target.
10053 For each shared-memory region, @value{GDBN} prints the region key,
10054 the shared-memory identifier, the access permissions, the size of the
10055 region, the process that created the region, the process that last
10056 attached to or detached from the region, the current number of live
10057 attaches to the region, and the times at which the region was last
10058 attached to, detach from, and changed.
10060 @kindex info os semaphores
10062 Display the list of all System V semaphore sets on the target. For each
10063 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
10064 set identifier, the access permissions, the number of semaphores in the
10065 set, the user and group of the owner and creator of the semaphore set,
10066 and the times at which the semaphore set was operated upon and changed.
10068 @kindex info os msg
10070 Display the list of all System V message queues on the target. For each
10071 message queue, @value{GDBN} prints the message queue key, the message
10072 queue identifier, the access permissions, the current number of bytes
10073 on the queue, the current number of messages on the queue, the processes
10074 that last sent and received a message on the queue, the user and group
10075 of the owner and creator of the message queue, the times at which a
10076 message was last sent and received on the queue, and the time at which
10077 the message queue was last changed.
10079 @kindex info os modules
10081 Display the list of all loaded kernel modules on the target. For each
10082 module, @value{GDBN} prints the module name, the size of the module in
10083 bytes, the number of times the module is used, the dependencies of the
10084 module, the status of the module, and the address of the loaded module
10089 If @var{infotype} is omitted, then list the possible values for
10090 @var{infotype} and the kind of OS information available for each
10091 @var{infotype}. If the target does not return a list of possible
10092 types, this command will report an error.
10095 @node Memory Region Attributes
10096 @section Memory Region Attributes
10097 @cindex memory region attributes
10099 @dfn{Memory region attributes} allow you to describe special handling
10100 required by regions of your target's memory. @value{GDBN} uses
10101 attributes to determine whether to allow certain types of memory
10102 accesses; whether to use specific width accesses; and whether to cache
10103 target memory. By default the description of memory regions is
10104 fetched from the target (if the current target supports this), but the
10105 user can override the fetched regions.
10107 Defined memory regions can be individually enabled and disabled. When a
10108 memory region is disabled, @value{GDBN} uses the default attributes when
10109 accessing memory in that region. Similarly, if no memory regions have
10110 been defined, @value{GDBN} uses the default attributes when accessing
10113 When a memory region is defined, it is given a number to identify it;
10114 to enable, disable, or remove a memory region, you specify that number.
10118 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
10119 Define a memory region bounded by @var{lower} and @var{upper} with
10120 attributes @var{attributes}@dots{}, and add it to the list of regions
10121 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
10122 case: it is treated as the target's maximum memory address.
10123 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
10126 Discard any user changes to the memory regions and use target-supplied
10127 regions, if available, or no regions if the target does not support.
10130 @item delete mem @var{nums}@dots{}
10131 Remove memory regions @var{nums}@dots{} from the list of regions
10132 monitored by @value{GDBN}.
10134 @kindex disable mem
10135 @item disable mem @var{nums}@dots{}
10136 Disable monitoring of memory regions @var{nums}@dots{}.
10137 A disabled memory region is not forgotten.
10138 It may be enabled again later.
10141 @item enable mem @var{nums}@dots{}
10142 Enable monitoring of memory regions @var{nums}@dots{}.
10146 Print a table of all defined memory regions, with the following columns
10150 @item Memory Region Number
10151 @item Enabled or Disabled.
10152 Enabled memory regions are marked with @samp{y}.
10153 Disabled memory regions are marked with @samp{n}.
10156 The address defining the inclusive lower bound of the memory region.
10159 The address defining the exclusive upper bound of the memory region.
10162 The list of attributes set for this memory region.
10167 @subsection Attributes
10169 @subsubsection Memory Access Mode
10170 The access mode attributes set whether @value{GDBN} may make read or
10171 write accesses to a memory region.
10173 While these attributes prevent @value{GDBN} from performing invalid
10174 memory accesses, they do nothing to prevent the target system, I/O DMA,
10175 etc.@: from accessing memory.
10179 Memory is read only.
10181 Memory is write only.
10183 Memory is read/write. This is the default.
10186 @subsubsection Memory Access Size
10187 The access size attribute tells @value{GDBN} to use specific sized
10188 accesses in the memory region. Often memory mapped device registers
10189 require specific sized accesses. If no access size attribute is
10190 specified, @value{GDBN} may use accesses of any size.
10194 Use 8 bit memory accesses.
10196 Use 16 bit memory accesses.
10198 Use 32 bit memory accesses.
10200 Use 64 bit memory accesses.
10203 @c @subsubsection Hardware/Software Breakpoints
10204 @c The hardware/software breakpoint attributes set whether @value{GDBN}
10205 @c will use hardware or software breakpoints for the internal breakpoints
10206 @c used by the step, next, finish, until, etc. commands.
10210 @c Always use hardware breakpoints
10211 @c @item swbreak (default)
10214 @subsubsection Data Cache
10215 The data cache attributes set whether @value{GDBN} will cache target
10216 memory. While this generally improves performance by reducing debug
10217 protocol overhead, it can lead to incorrect results because @value{GDBN}
10218 does not know about volatile variables or memory mapped device
10223 Enable @value{GDBN} to cache target memory.
10225 Disable @value{GDBN} from caching target memory. This is the default.
10228 @subsection Memory Access Checking
10229 @value{GDBN} can be instructed to refuse accesses to memory that is
10230 not explicitly described. This can be useful if accessing such
10231 regions has undesired effects for a specific target, or to provide
10232 better error checking. The following commands control this behaviour.
10235 @kindex set mem inaccessible-by-default
10236 @item set mem inaccessible-by-default [on|off]
10237 If @code{on} is specified, make @value{GDBN} treat memory not
10238 explicitly described by the memory ranges as non-existent and refuse accesses
10239 to such memory. The checks are only performed if there's at least one
10240 memory range defined. If @code{off} is specified, make @value{GDBN}
10241 treat the memory not explicitly described by the memory ranges as RAM.
10242 The default value is @code{on}.
10243 @kindex show mem inaccessible-by-default
10244 @item show mem inaccessible-by-default
10245 Show the current handling of accesses to unknown memory.
10249 @c @subsubsection Memory Write Verification
10250 @c The memory write verification attributes set whether @value{GDBN}
10251 @c will re-reads data after each write to verify the write was successful.
10255 @c @item noverify (default)
10258 @node Dump/Restore Files
10259 @section Copy Between Memory and a File
10260 @cindex dump/restore files
10261 @cindex append data to a file
10262 @cindex dump data to a file
10263 @cindex restore data from a file
10265 You can use the commands @code{dump}, @code{append}, and
10266 @code{restore} to copy data between target memory and a file. The
10267 @code{dump} and @code{append} commands write data to a file, and the
10268 @code{restore} command reads data from a file back into the inferior's
10269 memory. Files may be in binary, Motorola S-record, Intel hex, or
10270 Tektronix Hex format; however, @value{GDBN} can only append to binary
10276 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
10277 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
10278 Dump the contents of memory from @var{start_addr} to @var{end_addr},
10279 or the value of @var{expr}, to @var{filename} in the given format.
10281 The @var{format} parameter may be any one of:
10288 Motorola S-record format.
10290 Tektronix Hex format.
10293 @value{GDBN} uses the same definitions of these formats as the
10294 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
10295 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
10299 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
10300 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
10301 Append the contents of memory from @var{start_addr} to @var{end_addr},
10302 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
10303 (@value{GDBN} can only append data to files in raw binary form.)
10306 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
10307 Restore the contents of file @var{filename} into memory. The
10308 @code{restore} command can automatically recognize any known @sc{bfd}
10309 file format, except for raw binary. To restore a raw binary file you
10310 must specify the optional keyword @code{binary} after the filename.
10312 If @var{bias} is non-zero, its value will be added to the addresses
10313 contained in the file. Binary files always start at address zero, so
10314 they will be restored at address @var{bias}. Other bfd files have
10315 a built-in location; they will be restored at offset @var{bias}
10316 from that location.
10318 If @var{start} and/or @var{end} are non-zero, then only data between
10319 file offset @var{start} and file offset @var{end} will be restored.
10320 These offsets are relative to the addresses in the file, before
10321 the @var{bias} argument is applied.
10325 @node Core File Generation
10326 @section How to Produce a Core File from Your Program
10327 @cindex dump core from inferior
10329 A @dfn{core file} or @dfn{core dump} is a file that records the memory
10330 image of a running process and its process status (register values
10331 etc.). Its primary use is post-mortem debugging of a program that
10332 crashed while it ran outside a debugger. A program that crashes
10333 automatically produces a core file, unless this feature is disabled by
10334 the user. @xref{Files}, for information on invoking @value{GDBN} in
10335 the post-mortem debugging mode.
10337 Occasionally, you may wish to produce a core file of the program you
10338 are debugging in order to preserve a snapshot of its state.
10339 @value{GDBN} has a special command for that.
10343 @kindex generate-core-file
10344 @item generate-core-file [@var{file}]
10345 @itemx gcore [@var{file}]
10346 Produce a core dump of the inferior process. The optional argument
10347 @var{file} specifies the file name where to put the core dump. If not
10348 specified, the file name defaults to @file{core.@var{pid}}, where
10349 @var{pid} is the inferior process ID.
10351 Note that this command is implemented only for some systems (as of
10352 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
10355 @node Character Sets
10356 @section Character Sets
10357 @cindex character sets
10359 @cindex translating between character sets
10360 @cindex host character set
10361 @cindex target character set
10363 If the program you are debugging uses a different character set to
10364 represent characters and strings than the one @value{GDBN} uses itself,
10365 @value{GDBN} can automatically translate between the character sets for
10366 you. The character set @value{GDBN} uses we call the @dfn{host
10367 character set}; the one the inferior program uses we call the
10368 @dfn{target character set}.
10370 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
10371 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
10372 remote protocol (@pxref{Remote Debugging}) to debug a program
10373 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
10374 then the host character set is Latin-1, and the target character set is
10375 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
10376 target-charset EBCDIC-US}, then @value{GDBN} translates between
10377 @sc{ebcdic} and Latin 1 as you print character or string values, or use
10378 character and string literals in expressions.
10380 @value{GDBN} has no way to automatically recognize which character set
10381 the inferior program uses; you must tell it, using the @code{set
10382 target-charset} command, described below.
10384 Here are the commands for controlling @value{GDBN}'s character set
10388 @item set target-charset @var{charset}
10389 @kindex set target-charset
10390 Set the current target character set to @var{charset}. To display the
10391 list of supported target character sets, type
10392 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
10394 @item set host-charset @var{charset}
10395 @kindex set host-charset
10396 Set the current host character set to @var{charset}.
10398 By default, @value{GDBN} uses a host character set appropriate to the
10399 system it is running on; you can override that default using the
10400 @code{set host-charset} command. On some systems, @value{GDBN} cannot
10401 automatically determine the appropriate host character set. In this
10402 case, @value{GDBN} uses @samp{UTF-8}.
10404 @value{GDBN} can only use certain character sets as its host character
10405 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
10406 @value{GDBN} will list the host character sets it supports.
10408 @item set charset @var{charset}
10409 @kindex set charset
10410 Set the current host and target character sets to @var{charset}. As
10411 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
10412 @value{GDBN} will list the names of the character sets that can be used
10413 for both host and target.
10416 @kindex show charset
10417 Show the names of the current host and target character sets.
10419 @item show host-charset
10420 @kindex show host-charset
10421 Show the name of the current host character set.
10423 @item show target-charset
10424 @kindex show target-charset
10425 Show the name of the current target character set.
10427 @item set target-wide-charset @var{charset}
10428 @kindex set target-wide-charset
10429 Set the current target's wide character set to @var{charset}. This is
10430 the character set used by the target's @code{wchar_t} type. To
10431 display the list of supported wide character sets, type
10432 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
10434 @item show target-wide-charset
10435 @kindex show target-wide-charset
10436 Show the name of the current target's wide character set.
10439 Here is an example of @value{GDBN}'s character set support in action.
10440 Assume that the following source code has been placed in the file
10441 @file{charset-test.c}:
10447 = @{72, 101, 108, 108, 111, 44, 32, 119,
10448 111, 114, 108, 100, 33, 10, 0@};
10449 char ibm1047_hello[]
10450 = @{200, 133, 147, 147, 150, 107, 64, 166,
10451 150, 153, 147, 132, 90, 37, 0@};
10455 printf ("Hello, world!\n");
10459 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
10460 containing the string @samp{Hello, world!} followed by a newline,
10461 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
10463 We compile the program, and invoke the debugger on it:
10466 $ gcc -g charset-test.c -o charset-test
10467 $ gdb -nw charset-test
10468 GNU gdb 2001-12-19-cvs
10469 Copyright 2001 Free Software Foundation, Inc.
10474 We can use the @code{show charset} command to see what character sets
10475 @value{GDBN} is currently using to interpret and display characters and
10479 (@value{GDBP}) show charset
10480 The current host and target character set is `ISO-8859-1'.
10484 For the sake of printing this manual, let's use @sc{ascii} as our
10485 initial character set:
10487 (@value{GDBP}) set charset ASCII
10488 (@value{GDBP}) show charset
10489 The current host and target character set is `ASCII'.
10493 Let's assume that @sc{ascii} is indeed the correct character set for our
10494 host system --- in other words, let's assume that if @value{GDBN} prints
10495 characters using the @sc{ascii} character set, our terminal will display
10496 them properly. Since our current target character set is also
10497 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
10500 (@value{GDBP}) print ascii_hello
10501 $1 = 0x401698 "Hello, world!\n"
10502 (@value{GDBP}) print ascii_hello[0]
10507 @value{GDBN} uses the target character set for character and string
10508 literals you use in expressions:
10511 (@value{GDBP}) print '+'
10516 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
10519 @value{GDBN} relies on the user to tell it which character set the
10520 target program uses. If we print @code{ibm1047_hello} while our target
10521 character set is still @sc{ascii}, we get jibberish:
10524 (@value{GDBP}) print ibm1047_hello
10525 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
10526 (@value{GDBP}) print ibm1047_hello[0]
10531 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
10532 @value{GDBN} tells us the character sets it supports:
10535 (@value{GDBP}) set target-charset
10536 ASCII EBCDIC-US IBM1047 ISO-8859-1
10537 (@value{GDBP}) set target-charset
10540 We can select @sc{ibm1047} as our target character set, and examine the
10541 program's strings again. Now the @sc{ascii} string is wrong, but
10542 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
10543 target character set, @sc{ibm1047}, to the host character set,
10544 @sc{ascii}, and they display correctly:
10547 (@value{GDBP}) set target-charset IBM1047
10548 (@value{GDBP}) show charset
10549 The current host character set is `ASCII'.
10550 The current target character set is `IBM1047'.
10551 (@value{GDBP}) print ascii_hello
10552 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
10553 (@value{GDBP}) print ascii_hello[0]
10555 (@value{GDBP}) print ibm1047_hello
10556 $8 = 0x4016a8 "Hello, world!\n"
10557 (@value{GDBP}) print ibm1047_hello[0]
10562 As above, @value{GDBN} uses the target character set for character and
10563 string literals you use in expressions:
10566 (@value{GDBP}) print '+'
10571 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
10574 @node Caching Remote Data
10575 @section Caching Data of Remote Targets
10576 @cindex caching data of remote targets
10578 @value{GDBN} caches data exchanged between the debugger and a
10579 remote target (@pxref{Remote Debugging}). Such caching generally improves
10580 performance, because it reduces the overhead of the remote protocol by
10581 bundling memory reads and writes into large chunks. Unfortunately, simply
10582 caching everything would lead to incorrect results, since @value{GDBN}
10583 does not necessarily know anything about volatile values, memory-mapped I/O
10584 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
10585 memory can be changed @emph{while} a gdb command is executing.
10586 Therefore, by default, @value{GDBN} only caches data
10587 known to be on the stack@footnote{In non-stop mode, it is moderately
10588 rare for a running thread to modify the stack of a stopped thread
10589 in a way that would interfere with a backtrace, and caching of
10590 stack reads provides a significant speed up of remote backtraces.}.
10591 Other regions of memory can be explicitly marked as
10592 cacheable; see @pxref{Memory Region Attributes}.
10595 @kindex set remotecache
10596 @item set remotecache on
10597 @itemx set remotecache off
10598 This option no longer does anything; it exists for compatibility
10601 @kindex show remotecache
10602 @item show remotecache
10603 Show the current state of the obsolete remotecache flag.
10605 @kindex set stack-cache
10606 @item set stack-cache on
10607 @itemx set stack-cache off
10608 Enable or disable caching of stack accesses. When @code{ON}, use
10609 caching. By default, this option is @code{ON}.
10611 @kindex show stack-cache
10612 @item show stack-cache
10613 Show the current state of data caching for memory accesses.
10615 @kindex info dcache
10616 @item info dcache @r{[}line@r{]}
10617 Print the information about the data cache performance. The
10618 information displayed includes the dcache width and depth, and for
10619 each cache line, its number, address, and how many times it was
10620 referenced. This command is useful for debugging the data cache
10623 If a line number is specified, the contents of that line will be
10626 @item set dcache size @var{size}
10627 @cindex dcache size
10628 @kindex set dcache size
10629 Set maximum number of entries in dcache (dcache depth above).
10631 @item set dcache line-size @var{line-size}
10632 @cindex dcache line-size
10633 @kindex set dcache line-size
10634 Set number of bytes each dcache entry caches (dcache width above).
10635 Must be a power of 2.
10637 @item show dcache size
10638 @kindex show dcache size
10639 Show maximum number of dcache entries. See also @ref{Caching Remote Data, info dcache}.
10641 @item show dcache line-size
10642 @kindex show dcache line-size
10643 Show default size of dcache lines. See also @ref{Caching Remote Data, info dcache}.
10647 @node Searching Memory
10648 @section Search Memory
10649 @cindex searching memory
10651 Memory can be searched for a particular sequence of bytes with the
10652 @code{find} command.
10656 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
10657 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
10658 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
10659 etc. The search begins at address @var{start_addr} and continues for either
10660 @var{len} bytes or through to @var{end_addr} inclusive.
10663 @var{s} and @var{n} are optional parameters.
10664 They may be specified in either order, apart or together.
10667 @item @var{s}, search query size
10668 The size of each search query value.
10674 halfwords (two bytes)
10678 giant words (eight bytes)
10681 All values are interpreted in the current language.
10682 This means, for example, that if the current source language is C/C@t{++}
10683 then searching for the string ``hello'' includes the trailing '\0'.
10685 If the value size is not specified, it is taken from the
10686 value's type in the current language.
10687 This is useful when one wants to specify the search
10688 pattern as a mixture of types.
10689 Note that this means, for example, that in the case of C-like languages
10690 a search for an untyped 0x42 will search for @samp{(int) 0x42}
10691 which is typically four bytes.
10693 @item @var{n}, maximum number of finds
10694 The maximum number of matches to print. The default is to print all finds.
10697 You can use strings as search values. Quote them with double-quotes
10699 The string value is copied into the search pattern byte by byte,
10700 regardless of the endianness of the target and the size specification.
10702 The address of each match found is printed as well as a count of the
10703 number of matches found.
10705 The address of the last value found is stored in convenience variable
10707 A count of the number of matches is stored in @samp{$numfound}.
10709 For example, if stopped at the @code{printf} in this function:
10715 static char hello[] = "hello-hello";
10716 static struct @{ char c; short s; int i; @}
10717 __attribute__ ((packed)) mixed
10718 = @{ 'c', 0x1234, 0x87654321 @};
10719 printf ("%s\n", hello);
10724 you get during debugging:
10727 (gdb) find &hello[0], +sizeof(hello), "hello"
10728 0x804956d <hello.1620+6>
10730 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
10731 0x8049567 <hello.1620>
10732 0x804956d <hello.1620+6>
10734 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
10735 0x8049567 <hello.1620>
10737 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
10738 0x8049560 <mixed.1625>
10740 (gdb) print $numfound
10743 $2 = (void *) 0x8049560
10746 @node Optimized Code
10747 @chapter Debugging Optimized Code
10748 @cindex optimized code, debugging
10749 @cindex debugging optimized code
10751 Almost all compilers support optimization. With optimization
10752 disabled, the compiler generates assembly code that corresponds
10753 directly to your source code, in a simplistic way. As the compiler
10754 applies more powerful optimizations, the generated assembly code
10755 diverges from your original source code. With help from debugging
10756 information generated by the compiler, @value{GDBN} can map from
10757 the running program back to constructs from your original source.
10759 @value{GDBN} is more accurate with optimization disabled. If you
10760 can recompile without optimization, it is easier to follow the
10761 progress of your program during debugging. But, there are many cases
10762 where you may need to debug an optimized version.
10764 When you debug a program compiled with @samp{-g -O}, remember that the
10765 optimizer has rearranged your code; the debugger shows you what is
10766 really there. Do not be too surprised when the execution path does not
10767 exactly match your source file! An extreme example: if you define a
10768 variable, but never use it, @value{GDBN} never sees that
10769 variable---because the compiler optimizes it out of existence.
10771 Some things do not work as well with @samp{-g -O} as with just
10772 @samp{-g}, particularly on machines with instruction scheduling. If in
10773 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
10774 please report it to us as a bug (including a test case!).
10775 @xref{Variables}, for more information about debugging optimized code.
10778 * Inline Functions:: How @value{GDBN} presents inlining
10779 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
10782 @node Inline Functions
10783 @section Inline Functions
10784 @cindex inline functions, debugging
10786 @dfn{Inlining} is an optimization that inserts a copy of the function
10787 body directly at each call site, instead of jumping to a shared
10788 routine. @value{GDBN} displays inlined functions just like
10789 non-inlined functions. They appear in backtraces. You can view their
10790 arguments and local variables, step into them with @code{step}, skip
10791 them with @code{next}, and escape from them with @code{finish}.
10792 You can check whether a function was inlined by using the
10793 @code{info frame} command.
10795 For @value{GDBN} to support inlined functions, the compiler must
10796 record information about inlining in the debug information ---
10797 @value{NGCC} using the @sc{dwarf 2} format does this, and several
10798 other compilers do also. @value{GDBN} only supports inlined functions
10799 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
10800 do not emit two required attributes (@samp{DW_AT_call_file} and
10801 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
10802 function calls with earlier versions of @value{NGCC}. It instead
10803 displays the arguments and local variables of inlined functions as
10804 local variables in the caller.
10806 The body of an inlined function is directly included at its call site;
10807 unlike a non-inlined function, there are no instructions devoted to
10808 the call. @value{GDBN} still pretends that the call site and the
10809 start of the inlined function are different instructions. Stepping to
10810 the call site shows the call site, and then stepping again shows
10811 the first line of the inlined function, even though no additional
10812 instructions are executed.
10814 This makes source-level debugging much clearer; you can see both the
10815 context of the call and then the effect of the call. Only stepping by
10816 a single instruction using @code{stepi} or @code{nexti} does not do
10817 this; single instruction steps always show the inlined body.
10819 There are some ways that @value{GDBN} does not pretend that inlined
10820 function calls are the same as normal calls:
10824 Setting breakpoints at the call site of an inlined function may not
10825 work, because the call site does not contain any code. @value{GDBN}
10826 may incorrectly move the breakpoint to the next line of the enclosing
10827 function, after the call. This limitation will be removed in a future
10828 version of @value{GDBN}; until then, set a breakpoint on an earlier line
10829 or inside the inlined function instead.
10832 @value{GDBN} cannot locate the return value of inlined calls after
10833 using the @code{finish} command. This is a limitation of compiler-generated
10834 debugging information; after @code{finish}, you can step to the next line
10835 and print a variable where your program stored the return value.
10839 @node Tail Call Frames
10840 @section Tail Call Frames
10841 @cindex tail call frames, debugging
10843 Function @code{B} can call function @code{C} in its very last statement. In
10844 unoptimized compilation the call of @code{C} is immediately followed by return
10845 instruction at the end of @code{B} code. Optimizing compiler may replace the
10846 call and return in function @code{B} into one jump to function @code{C}
10847 instead. Such use of a jump instruction is called @dfn{tail call}.
10849 During execution of function @code{C}, there will be no indication in the
10850 function call stack frames that it was tail-called from @code{B}. If function
10851 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
10852 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
10853 some cases @value{GDBN} can determine that @code{C} was tail-called from
10854 @code{B}, and it will then create fictitious call frame for that, with the
10855 return address set up as if @code{B} called @code{C} normally.
10857 This functionality is currently supported only by DWARF 2 debugging format and
10858 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
10859 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
10862 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
10863 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
10867 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
10869 Stack level 1, frame at 0x7fffffffda30:
10870 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
10871 tail call frame, caller of frame at 0x7fffffffda30
10872 source language c++.
10873 Arglist at unknown address.
10874 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
10877 The detection of all the possible code path executions can find them ambiguous.
10878 There is no execution history stored (possible @ref{Reverse Execution} is never
10879 used for this purpose) and the last known caller could have reached the known
10880 callee by multiple different jump sequences. In such case @value{GDBN} still
10881 tries to show at least all the unambiguous top tail callers and all the
10882 unambiguous bottom tail calees, if any.
10885 @anchor{set debug entry-values}
10886 @item set debug entry-values
10887 @kindex set debug entry-values
10888 When set to on, enables printing of analysis messages for both frame argument
10889 values at function entry and tail calls. It will show all the possible valid
10890 tail calls code paths it has considered. It will also print the intersection
10891 of them with the final unambiguous (possibly partial or even empty) code path
10894 @item show debug entry-values
10895 @kindex show debug entry-values
10896 Show the current state of analysis messages printing for both frame argument
10897 values at function entry and tail calls.
10900 The analysis messages for tail calls can for example show why the virtual tail
10901 call frame for function @code{c} has not been recognized (due to the indirect
10902 reference by variable @code{x}):
10905 static void __attribute__((noinline, noclone)) c (void);
10906 void (*x) (void) = c;
10907 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
10908 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
10909 int main (void) @{ x (); return 0; @}
10911 Breakpoint 1, DW_OP_GNU_entry_value resolving cannot find
10912 DW_TAG_GNU_call_site 0x40039a in main
10914 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
10917 #1 0x000000000040039a in main () at t.c:5
10920 Another possibility is an ambiguous virtual tail call frames resolution:
10924 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
10925 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
10926 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
10927 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
10928 static void __attribute__((noinline, noclone)) b (void)
10929 @{ if (i) c (); else e (); @}
10930 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
10931 int main (void) @{ a (); return 0; @}
10933 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
10934 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
10935 tailcall: reduced: 0x4004d2(a) |
10938 #1 0x00000000004004d2 in a () at t.c:8
10939 #2 0x0000000000400395 in main () at t.c:9
10942 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
10943 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
10945 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
10946 @ifset HAVE_MAKEINFO_CLICK
10947 @set ARROW @click{}
10948 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
10949 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
10951 @ifclear HAVE_MAKEINFO_CLICK
10953 @set CALLSEQ1B @value{CALLSEQ1A}
10954 @set CALLSEQ2B @value{CALLSEQ2A}
10957 Frames #0 and #2 are real, #1 is a virtual tail call frame.
10958 The code can have possible execution paths @value{CALLSEQ1B} or
10959 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
10961 @code{initial:} state shows some random possible calling sequence @value{GDBN}
10962 has found. It then finds another possible calling sequcen - that one is
10963 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
10964 printed as the @code{reduced:} calling sequence. That one could have many
10965 futher @code{compare:} and @code{reduced:} statements as long as there remain
10966 any non-ambiguous sequence entries.
10968 For the frame of function @code{b} in both cases there are different possible
10969 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
10970 also ambigous. The only non-ambiguous frame is the one for function @code{a},
10971 therefore this one is displayed to the user while the ambiguous frames are
10974 There can be also reasons why printing of frame argument values at function
10979 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
10980 static void __attribute__((noinline, noclone)) a (int i);
10981 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
10982 static void __attribute__((noinline, noclone)) a (int i)
10983 @{ if (i) b (i - 1); else c (0); @}
10984 int main (void) @{ a (5); return 0; @}
10987 #0 c (i=i@@entry=0) at t.c:2
10988 #1 0x0000000000400428 in a (DW_OP_GNU_entry_value resolving has found
10989 function "a" at 0x400420 can call itself via tail calls
10990 i=<optimized out>) at t.c:6
10991 #2 0x000000000040036e in main () at t.c:7
10994 @value{GDBN} cannot find out from the inferior state if and how many times did
10995 function @code{a} call itself (via function @code{b}) as these calls would be
10996 tail calls. Such tail calls would modify thue @code{i} variable, therefore
10997 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
10998 prints @code{<optimized out>} instead.
11001 @chapter C Preprocessor Macros
11003 Some languages, such as C and C@t{++}, provide a way to define and invoke
11004 ``preprocessor macros'' which expand into strings of tokens.
11005 @value{GDBN} can evaluate expressions containing macro invocations, show
11006 the result of macro expansion, and show a macro's definition, including
11007 where it was defined.
11009 You may need to compile your program specially to provide @value{GDBN}
11010 with information about preprocessor macros. Most compilers do not
11011 include macros in their debugging information, even when you compile
11012 with the @option{-g} flag. @xref{Compilation}.
11014 A program may define a macro at one point, remove that definition later,
11015 and then provide a different definition after that. Thus, at different
11016 points in the program, a macro may have different definitions, or have
11017 no definition at all. If there is a current stack frame, @value{GDBN}
11018 uses the macros in scope at that frame's source code line. Otherwise,
11019 @value{GDBN} uses the macros in scope at the current listing location;
11022 Whenever @value{GDBN} evaluates an expression, it always expands any
11023 macro invocations present in the expression. @value{GDBN} also provides
11024 the following commands for working with macros explicitly.
11028 @kindex macro expand
11029 @cindex macro expansion, showing the results of preprocessor
11030 @cindex preprocessor macro expansion, showing the results of
11031 @cindex expanding preprocessor macros
11032 @item macro expand @var{expression}
11033 @itemx macro exp @var{expression}
11034 Show the results of expanding all preprocessor macro invocations in
11035 @var{expression}. Since @value{GDBN} simply expands macros, but does
11036 not parse the result, @var{expression} need not be a valid expression;
11037 it can be any string of tokens.
11040 @item macro expand-once @var{expression}
11041 @itemx macro exp1 @var{expression}
11042 @cindex expand macro once
11043 @i{(This command is not yet implemented.)} Show the results of
11044 expanding those preprocessor macro invocations that appear explicitly in
11045 @var{expression}. Macro invocations appearing in that expansion are
11046 left unchanged. This command allows you to see the effect of a
11047 particular macro more clearly, without being confused by further
11048 expansions. Since @value{GDBN} simply expands macros, but does not
11049 parse the result, @var{expression} need not be a valid expression; it
11050 can be any string of tokens.
11053 @cindex macro definition, showing
11054 @cindex definition of a macro, showing
11055 @cindex macros, from debug info
11056 @item info macro [-a|-all] [--] @var{macro}
11057 Show the current definition or all definitions of the named @var{macro},
11058 and describe the source location or compiler command-line where that
11059 definition was established. The optional double dash is to signify the end of
11060 argument processing and the beginning of @var{macro} for non C-like macros where
11061 the macro may begin with a hyphen.
11063 @kindex info macros
11064 @item info macros @var{linespec}
11065 Show all macro definitions that are in effect at the location specified
11066 by @var{linespec}, and describe the source location or compiler
11067 command-line where those definitions were established.
11069 @kindex macro define
11070 @cindex user-defined macros
11071 @cindex defining macros interactively
11072 @cindex macros, user-defined
11073 @item macro define @var{macro} @var{replacement-list}
11074 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
11075 Introduce a definition for a preprocessor macro named @var{macro},
11076 invocations of which are replaced by the tokens given in
11077 @var{replacement-list}. The first form of this command defines an
11078 ``object-like'' macro, which takes no arguments; the second form
11079 defines a ``function-like'' macro, which takes the arguments given in
11082 A definition introduced by this command is in scope in every
11083 expression evaluated in @value{GDBN}, until it is removed with the
11084 @code{macro undef} command, described below. The definition overrides
11085 all definitions for @var{macro} present in the program being debugged,
11086 as well as any previous user-supplied definition.
11088 @kindex macro undef
11089 @item macro undef @var{macro}
11090 Remove any user-supplied definition for the macro named @var{macro}.
11091 This command only affects definitions provided with the @code{macro
11092 define} command, described above; it cannot remove definitions present
11093 in the program being debugged.
11097 List all the macros defined using the @code{macro define} command.
11100 @cindex macros, example of debugging with
11101 Here is a transcript showing the above commands in action. First, we
11102 show our source files:
11107 #include "sample.h"
11110 #define ADD(x) (M + x)
11115 printf ("Hello, world!\n");
11117 printf ("We're so creative.\n");
11119 printf ("Goodbye, world!\n");
11126 Now, we compile the program using the @sc{gnu} C compiler,
11127 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
11128 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
11129 and @option{-gdwarf-4}; we recommend always choosing the most recent
11130 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
11131 includes information about preprocessor macros in the debugging
11135 $ gcc -gdwarf-2 -g3 sample.c -o sample
11139 Now, we start @value{GDBN} on our sample program:
11143 GNU gdb 2002-05-06-cvs
11144 Copyright 2002 Free Software Foundation, Inc.
11145 GDB is free software, @dots{}
11149 We can expand macros and examine their definitions, even when the
11150 program is not running. @value{GDBN} uses the current listing position
11151 to decide which macro definitions are in scope:
11154 (@value{GDBP}) list main
11157 5 #define ADD(x) (M + x)
11162 10 printf ("Hello, world!\n");
11164 12 printf ("We're so creative.\n");
11165 (@value{GDBP}) info macro ADD
11166 Defined at /home/jimb/gdb/macros/play/sample.c:5
11167 #define ADD(x) (M + x)
11168 (@value{GDBP}) info macro Q
11169 Defined at /home/jimb/gdb/macros/play/sample.h:1
11170 included at /home/jimb/gdb/macros/play/sample.c:2
11172 (@value{GDBP}) macro expand ADD(1)
11173 expands to: (42 + 1)
11174 (@value{GDBP}) macro expand-once ADD(1)
11175 expands to: once (M + 1)
11179 In the example above, note that @code{macro expand-once} expands only
11180 the macro invocation explicit in the original text --- the invocation of
11181 @code{ADD} --- but does not expand the invocation of the macro @code{M},
11182 which was introduced by @code{ADD}.
11184 Once the program is running, @value{GDBN} uses the macro definitions in
11185 force at the source line of the current stack frame:
11188 (@value{GDBP}) break main
11189 Breakpoint 1 at 0x8048370: file sample.c, line 10.
11191 Starting program: /home/jimb/gdb/macros/play/sample
11193 Breakpoint 1, main () at sample.c:10
11194 10 printf ("Hello, world!\n");
11198 At line 10, the definition of the macro @code{N} at line 9 is in force:
11201 (@value{GDBP}) info macro N
11202 Defined at /home/jimb/gdb/macros/play/sample.c:9
11204 (@value{GDBP}) macro expand N Q M
11205 expands to: 28 < 42
11206 (@value{GDBP}) print N Q M
11211 As we step over directives that remove @code{N}'s definition, and then
11212 give it a new definition, @value{GDBN} finds the definition (or lack
11213 thereof) in force at each point:
11216 (@value{GDBP}) next
11218 12 printf ("We're so creative.\n");
11219 (@value{GDBP}) info macro N
11220 The symbol `N' has no definition as a C/C++ preprocessor macro
11221 at /home/jimb/gdb/macros/play/sample.c:12
11222 (@value{GDBP}) next
11224 14 printf ("Goodbye, world!\n");
11225 (@value{GDBP}) info macro N
11226 Defined at /home/jimb/gdb/macros/play/sample.c:13
11228 (@value{GDBP}) macro expand N Q M
11229 expands to: 1729 < 42
11230 (@value{GDBP}) print N Q M
11235 In addition to source files, macros can be defined on the compilation command
11236 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
11237 such a way, @value{GDBN} displays the location of their definition as line zero
11238 of the source file submitted to the compiler.
11241 (@value{GDBP}) info macro __STDC__
11242 Defined at /home/jimb/gdb/macros/play/sample.c:0
11249 @chapter Tracepoints
11250 @c This chapter is based on the documentation written by Michael
11251 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
11253 @cindex tracepoints
11254 In some applications, it is not feasible for the debugger to interrupt
11255 the program's execution long enough for the developer to learn
11256 anything helpful about its behavior. If the program's correctness
11257 depends on its real-time behavior, delays introduced by a debugger
11258 might cause the program to change its behavior drastically, or perhaps
11259 fail, even when the code itself is correct. It is useful to be able
11260 to observe the program's behavior without interrupting it.
11262 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
11263 specify locations in the program, called @dfn{tracepoints}, and
11264 arbitrary expressions to evaluate when those tracepoints are reached.
11265 Later, using the @code{tfind} command, you can examine the values
11266 those expressions had when the program hit the tracepoints. The
11267 expressions may also denote objects in memory---structures or arrays,
11268 for example---whose values @value{GDBN} should record; while visiting
11269 a particular tracepoint, you may inspect those objects as if they were
11270 in memory at that moment. However, because @value{GDBN} records these
11271 values without interacting with you, it can do so quickly and
11272 unobtrusively, hopefully not disturbing the program's behavior.
11274 The tracepoint facility is currently available only for remote
11275 targets. @xref{Targets}. In addition, your remote target must know
11276 how to collect trace data. This functionality is implemented in the
11277 remote stub; however, none of the stubs distributed with @value{GDBN}
11278 support tracepoints as of this writing. The format of the remote
11279 packets used to implement tracepoints are described in @ref{Tracepoint
11282 It is also possible to get trace data from a file, in a manner reminiscent
11283 of corefiles; you specify the filename, and use @code{tfind} to search
11284 through the file. @xref{Trace Files}, for more details.
11286 This chapter describes the tracepoint commands and features.
11289 * Set Tracepoints::
11290 * Analyze Collected Data::
11291 * Tracepoint Variables::
11295 @node Set Tracepoints
11296 @section Commands to Set Tracepoints
11298 Before running such a @dfn{trace experiment}, an arbitrary number of
11299 tracepoints can be set. A tracepoint is actually a special type of
11300 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
11301 standard breakpoint commands. For instance, as with breakpoints,
11302 tracepoint numbers are successive integers starting from one, and many
11303 of the commands associated with tracepoints take the tracepoint number
11304 as their argument, to identify which tracepoint to work on.
11306 For each tracepoint, you can specify, in advance, some arbitrary set
11307 of data that you want the target to collect in the trace buffer when
11308 it hits that tracepoint. The collected data can include registers,
11309 local variables, or global data. Later, you can use @value{GDBN}
11310 commands to examine the values these data had at the time the
11311 tracepoint was hit.
11313 Tracepoints do not support every breakpoint feature. Ignore counts on
11314 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
11315 commands when they are hit. Tracepoints may not be thread-specific
11318 @cindex fast tracepoints
11319 Some targets may support @dfn{fast tracepoints}, which are inserted in
11320 a different way (such as with a jump instead of a trap), that is
11321 faster but possibly restricted in where they may be installed.
11323 @cindex static tracepoints
11324 @cindex markers, static tracepoints
11325 @cindex probing markers, static tracepoints
11326 Regular and fast tracepoints are dynamic tracing facilities, meaning
11327 that they can be used to insert tracepoints at (almost) any location
11328 in the target. Some targets may also support controlling @dfn{static
11329 tracepoints} from @value{GDBN}. With static tracing, a set of
11330 instrumentation points, also known as @dfn{markers}, are embedded in
11331 the target program, and can be activated or deactivated by name or
11332 address. These are usually placed at locations which facilitate
11333 investigating what the target is actually doing. @value{GDBN}'s
11334 support for static tracing includes being able to list instrumentation
11335 points, and attach them with @value{GDBN} defined high level
11336 tracepoints that expose the whole range of convenience of
11337 @value{GDBN}'s tracepoints support. Namely, support for collecting
11338 registers values and values of global or local (to the instrumentation
11339 point) variables; tracepoint conditions and trace state variables.
11340 The act of installing a @value{GDBN} static tracepoint on an
11341 instrumentation point, or marker, is referred to as @dfn{probing} a
11342 static tracepoint marker.
11344 @code{gdbserver} supports tracepoints on some target systems.
11345 @xref{Server,,Tracepoints support in @code{gdbserver}}.
11347 This section describes commands to set tracepoints and associated
11348 conditions and actions.
11351 * Create and Delete Tracepoints::
11352 * Enable and Disable Tracepoints::
11353 * Tracepoint Passcounts::
11354 * Tracepoint Conditions::
11355 * Trace State Variables::
11356 * Tracepoint Actions::
11357 * Listing Tracepoints::
11358 * Listing Static Tracepoint Markers::
11359 * Starting and Stopping Trace Experiments::
11360 * Tracepoint Restrictions::
11363 @node Create and Delete Tracepoints
11364 @subsection Create and Delete Tracepoints
11367 @cindex set tracepoint
11369 @item trace @var{location}
11370 The @code{trace} command is very similar to the @code{break} command.
11371 Its argument @var{location} can be a source line, a function name, or
11372 an address in the target program. @xref{Specify Location}. The
11373 @code{trace} command defines a tracepoint, which is a point in the
11374 target program where the debugger will briefly stop, collect some
11375 data, and then allow the program to continue. Setting a tracepoint or
11376 changing its actions takes effect immediately if the remote stub
11377 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
11379 If remote stub doesn't support the @samp{InstallInTrace} feature, all
11380 these changes don't take effect until the next @code{tstart}
11381 command, and once a trace experiment is running, further changes will
11382 not have any effect until the next trace experiment starts. In addition,
11383 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
11384 address is not yet resolved. (This is similar to pending breakpoints.)
11385 Pending tracepoints are not downloaded to the target and not installed
11386 until they are resolved. The resolution of pending tracepoints requires
11387 @value{GDBN} support---when debugging with the remote target, and
11388 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
11389 tracing}), pending tracepoints can not be resolved (and downloaded to
11390 the remote stub) while @value{GDBN} is disconnected.
11392 Here are some examples of using the @code{trace} command:
11395 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
11397 (@value{GDBP}) @b{trace +2} // 2 lines forward
11399 (@value{GDBP}) @b{trace my_function} // first source line of function
11401 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
11403 (@value{GDBP}) @b{trace *0x2117c4} // an address
11407 You can abbreviate @code{trace} as @code{tr}.
11409 @item trace @var{location} if @var{cond}
11410 Set a tracepoint with condition @var{cond}; evaluate the expression
11411 @var{cond} each time the tracepoint is reached, and collect data only
11412 if the value is nonzero---that is, if @var{cond} evaluates as true.
11413 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
11414 information on tracepoint conditions.
11416 @item ftrace @var{location} [ if @var{cond} ]
11417 @cindex set fast tracepoint
11418 @cindex fast tracepoints, setting
11420 The @code{ftrace} command sets a fast tracepoint. For targets that
11421 support them, fast tracepoints will use a more efficient but possibly
11422 less general technique to trigger data collection, such as a jump
11423 instruction instead of a trap, or some sort of hardware support. It
11424 may not be possible to create a fast tracepoint at the desired
11425 location, in which case the command will exit with an explanatory
11428 @value{GDBN} handles arguments to @code{ftrace} exactly as for
11431 On 32-bit x86-architecture systems, fast tracepoints normally need to
11432 be placed at an instruction that is 5 bytes or longer, but can be
11433 placed at 4-byte instructions if the low 64K of memory of the target
11434 program is available to install trampolines. Some Unix-type systems,
11435 such as @sc{gnu}/Linux, exclude low addresses from the program's
11436 address space; but for instance with the Linux kernel it is possible
11437 to let @value{GDBN} use this area by doing a @command{sysctl} command
11438 to set the @code{mmap_min_addr} kernel parameter, as in
11441 sudo sysctl -w vm.mmap_min_addr=32768
11445 which sets the low address to 32K, which leaves plenty of room for
11446 trampolines. The minimum address should be set to a page boundary.
11448 @item strace @var{location} [ if @var{cond} ]
11449 @cindex set static tracepoint
11450 @cindex static tracepoints, setting
11451 @cindex probe static tracepoint marker
11453 The @code{strace} command sets a static tracepoint. For targets that
11454 support it, setting a static tracepoint probes a static
11455 instrumentation point, or marker, found at @var{location}. It may not
11456 be possible to set a static tracepoint at the desired location, in
11457 which case the command will exit with an explanatory message.
11459 @value{GDBN} handles arguments to @code{strace} exactly as for
11460 @code{trace}, with the addition that the user can also specify
11461 @code{-m @var{marker}} as @var{location}. This probes the marker
11462 identified by the @var{marker} string identifier. This identifier
11463 depends on the static tracepoint backend library your program is
11464 using. You can find all the marker identifiers in the @samp{ID} field
11465 of the @code{info static-tracepoint-markers} command output.
11466 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
11467 Markers}. For example, in the following small program using the UST
11473 trace_mark(ust, bar33, "str %s", "FOOBAZ");
11478 the marker id is composed of joining the first two arguments to the
11479 @code{trace_mark} call with a slash, which translates to:
11482 (@value{GDBP}) info static-tracepoint-markers
11483 Cnt Enb ID Address What
11484 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
11490 so you may probe the marker above with:
11493 (@value{GDBP}) strace -m ust/bar33
11496 Static tracepoints accept an extra collect action --- @code{collect
11497 $_sdata}. This collects arbitrary user data passed in the probe point
11498 call to the tracing library. In the UST example above, you'll see
11499 that the third argument to @code{trace_mark} is a printf-like format
11500 string. The user data is then the result of running that formating
11501 string against the following arguments. Note that @code{info
11502 static-tracepoint-markers} command output lists that format string in
11503 the @samp{Data:} field.
11505 You can inspect this data when analyzing the trace buffer, by printing
11506 the $_sdata variable like any other variable available to
11507 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
11510 @cindex last tracepoint number
11511 @cindex recent tracepoint number
11512 @cindex tracepoint number
11513 The convenience variable @code{$tpnum} records the tracepoint number
11514 of the most recently set tracepoint.
11516 @kindex delete tracepoint
11517 @cindex tracepoint deletion
11518 @item delete tracepoint @r{[}@var{num}@r{]}
11519 Permanently delete one or more tracepoints. With no argument, the
11520 default is to delete all tracepoints. Note that the regular
11521 @code{delete} command can remove tracepoints also.
11526 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
11528 (@value{GDBP}) @b{delete trace} // remove all tracepoints
11532 You can abbreviate this command as @code{del tr}.
11535 @node Enable and Disable Tracepoints
11536 @subsection Enable and Disable Tracepoints
11538 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
11541 @kindex disable tracepoint
11542 @item disable tracepoint @r{[}@var{num}@r{]}
11543 Disable tracepoint @var{num}, or all tracepoints if no argument
11544 @var{num} is given. A disabled tracepoint will have no effect during
11545 a trace experiment, but it is not forgotten. You can re-enable
11546 a disabled tracepoint using the @code{enable tracepoint} command.
11547 If the command is issued during a trace experiment and the debug target
11548 has support for disabling tracepoints during a trace experiment, then the
11549 change will be effective immediately. Otherwise, it will be applied to the
11550 next trace experiment.
11552 @kindex enable tracepoint
11553 @item enable tracepoint @r{[}@var{num}@r{]}
11554 Enable tracepoint @var{num}, or all tracepoints. If this command is
11555 issued during a trace experiment and the debug target supports enabling
11556 tracepoints during a trace experiment, then the enabled tracepoints will
11557 become effective immediately. Otherwise, they will become effective the
11558 next time a trace experiment is run.
11561 @node Tracepoint Passcounts
11562 @subsection Tracepoint Passcounts
11566 @cindex tracepoint pass count
11567 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
11568 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
11569 automatically stop a trace experiment. If a tracepoint's passcount is
11570 @var{n}, then the trace experiment will be automatically stopped on
11571 the @var{n}'th time that tracepoint is hit. If the tracepoint number
11572 @var{num} is not specified, the @code{passcount} command sets the
11573 passcount of the most recently defined tracepoint. If no passcount is
11574 given, the trace experiment will run until stopped explicitly by the
11580 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
11581 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
11583 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
11584 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
11585 (@value{GDBP}) @b{trace foo}
11586 (@value{GDBP}) @b{pass 3}
11587 (@value{GDBP}) @b{trace bar}
11588 (@value{GDBP}) @b{pass 2}
11589 (@value{GDBP}) @b{trace baz}
11590 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
11591 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
11592 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
11593 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
11597 @node Tracepoint Conditions
11598 @subsection Tracepoint Conditions
11599 @cindex conditional tracepoints
11600 @cindex tracepoint conditions
11602 The simplest sort of tracepoint collects data every time your program
11603 reaches a specified place. You can also specify a @dfn{condition} for
11604 a tracepoint. A condition is just a Boolean expression in your
11605 programming language (@pxref{Expressions, ,Expressions}). A
11606 tracepoint with a condition evaluates the expression each time your
11607 program reaches it, and data collection happens only if the condition
11610 Tracepoint conditions can be specified when a tracepoint is set, by
11611 using @samp{if} in the arguments to the @code{trace} command.
11612 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
11613 also be set or changed at any time with the @code{condition} command,
11614 just as with breakpoints.
11616 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
11617 the conditional expression itself. Instead, @value{GDBN} encodes the
11618 expression into an agent expression (@pxref{Agent Expressions})
11619 suitable for execution on the target, independently of @value{GDBN}.
11620 Global variables become raw memory locations, locals become stack
11621 accesses, and so forth.
11623 For instance, suppose you have a function that is usually called
11624 frequently, but should not be called after an error has occurred. You
11625 could use the following tracepoint command to collect data about calls
11626 of that function that happen while the error code is propagating
11627 through the program; an unconditional tracepoint could end up
11628 collecting thousands of useless trace frames that you would have to
11632 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
11635 @node Trace State Variables
11636 @subsection Trace State Variables
11637 @cindex trace state variables
11639 A @dfn{trace state variable} is a special type of variable that is
11640 created and managed by target-side code. The syntax is the same as
11641 that for GDB's convenience variables (a string prefixed with ``$''),
11642 but they are stored on the target. They must be created explicitly,
11643 using a @code{tvariable} command. They are always 64-bit signed
11646 Trace state variables are remembered by @value{GDBN}, and downloaded
11647 to the target along with tracepoint information when the trace
11648 experiment starts. There are no intrinsic limits on the number of
11649 trace state variables, beyond memory limitations of the target.
11651 @cindex convenience variables, and trace state variables
11652 Although trace state variables are managed by the target, you can use
11653 them in print commands and expressions as if they were convenience
11654 variables; @value{GDBN} will get the current value from the target
11655 while the trace experiment is running. Trace state variables share
11656 the same namespace as other ``$'' variables, which means that you
11657 cannot have trace state variables with names like @code{$23} or
11658 @code{$pc}, nor can you have a trace state variable and a convenience
11659 variable with the same name.
11663 @item tvariable $@var{name} [ = @var{expression} ]
11665 The @code{tvariable} command creates a new trace state variable named
11666 @code{$@var{name}}, and optionally gives it an initial value of
11667 @var{expression}. @var{expression} is evaluated when this command is
11668 entered; the result will be converted to an integer if possible,
11669 otherwise @value{GDBN} will report an error. A subsequent
11670 @code{tvariable} command specifying the same name does not create a
11671 variable, but instead assigns the supplied initial value to the
11672 existing variable of that name, overwriting any previous initial
11673 value. The default initial value is 0.
11675 @item info tvariables
11676 @kindex info tvariables
11677 List all the trace state variables along with their initial values.
11678 Their current values may also be displayed, if the trace experiment is
11681 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
11682 @kindex delete tvariable
11683 Delete the given trace state variables, or all of them if no arguments
11688 @node Tracepoint Actions
11689 @subsection Tracepoint Action Lists
11693 @cindex tracepoint actions
11694 @item actions @r{[}@var{num}@r{]}
11695 This command will prompt for a list of actions to be taken when the
11696 tracepoint is hit. If the tracepoint number @var{num} is not
11697 specified, this command sets the actions for the one that was most
11698 recently defined (so that you can define a tracepoint and then say
11699 @code{actions} without bothering about its number). You specify the
11700 actions themselves on the following lines, one action at a time, and
11701 terminate the actions list with a line containing just @code{end}. So
11702 far, the only defined actions are @code{collect}, @code{teval}, and
11703 @code{while-stepping}.
11705 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
11706 Commands, ,Breakpoint Command Lists}), except that only the defined
11707 actions are allowed; any other @value{GDBN} command is rejected.
11709 @cindex remove actions from a tracepoint
11710 To remove all actions from a tracepoint, type @samp{actions @var{num}}
11711 and follow it immediately with @samp{end}.
11714 (@value{GDBP}) @b{collect @var{data}} // collect some data
11716 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
11718 (@value{GDBP}) @b{end} // signals the end of actions.
11721 In the following example, the action list begins with @code{collect}
11722 commands indicating the things to be collected when the tracepoint is
11723 hit. Then, in order to single-step and collect additional data
11724 following the tracepoint, a @code{while-stepping} command is used,
11725 followed by the list of things to be collected after each step in a
11726 sequence of single steps. The @code{while-stepping} command is
11727 terminated by its own separate @code{end} command. Lastly, the action
11728 list is terminated by an @code{end} command.
11731 (@value{GDBP}) @b{trace foo}
11732 (@value{GDBP}) @b{actions}
11733 Enter actions for tracepoint 1, one per line:
11736 > while-stepping 12
11737 > collect $pc, arr[i]
11742 @kindex collect @r{(tracepoints)}
11743 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
11744 Collect values of the given expressions when the tracepoint is hit.
11745 This command accepts a comma-separated list of any valid expressions.
11746 In addition to global, static, or local variables, the following
11747 special arguments are supported:
11751 Collect all registers.
11754 Collect all function arguments.
11757 Collect all local variables.
11760 Collect the return address. This is helpful if you want to see more
11764 Collects the number of arguments from the static probe at which the
11765 tracepoint is located.
11766 @xref{Static Probe Points}.
11768 @item $_probe_arg@var{n}
11769 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
11770 from the static probe at which the tracepoint is located.
11771 @xref{Static Probe Points}.
11774 @vindex $_sdata@r{, collect}
11775 Collect static tracepoint marker specific data. Only available for
11776 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
11777 Lists}. On the UST static tracepoints library backend, an
11778 instrumentation point resembles a @code{printf} function call. The
11779 tracing library is able to collect user specified data formatted to a
11780 character string using the format provided by the programmer that
11781 instrumented the program. Other backends have similar mechanisms.
11782 Here's an example of a UST marker call:
11785 const char master_name[] = "$your_name";
11786 trace_mark(channel1, marker1, "hello %s", master_name)
11789 In this case, collecting @code{$_sdata} collects the string
11790 @samp{hello $yourname}. When analyzing the trace buffer, you can
11791 inspect @samp{$_sdata} like any other variable available to
11795 You can give several consecutive @code{collect} commands, each one
11796 with a single argument, or one @code{collect} command with several
11797 arguments separated by commas; the effect is the same.
11799 The optional @var{mods} changes the usual handling of the arguments.
11800 @code{s} requests that pointers to chars be handled as strings, in
11801 particular collecting the contents of the memory being pointed at, up
11802 to the first zero. The upper bound is by default the value of the
11803 @code{print elements} variable; if @code{s} is followed by a decimal
11804 number, that is the upper bound instead. So for instance
11805 @samp{collect/s25 mystr} collects as many as 25 characters at
11808 The command @code{info scope} (@pxref{Symbols, info scope}) is
11809 particularly useful for figuring out what data to collect.
11811 @kindex teval @r{(tracepoints)}
11812 @item teval @var{expr1}, @var{expr2}, @dots{}
11813 Evaluate the given expressions when the tracepoint is hit. This
11814 command accepts a comma-separated list of expressions. The results
11815 are discarded, so this is mainly useful for assigning values to trace
11816 state variables (@pxref{Trace State Variables}) without adding those
11817 values to the trace buffer, as would be the case if the @code{collect}
11820 @kindex while-stepping @r{(tracepoints)}
11821 @item while-stepping @var{n}
11822 Perform @var{n} single-step instruction traces after the tracepoint,
11823 collecting new data after each step. The @code{while-stepping}
11824 command is followed by the list of what to collect while stepping
11825 (followed by its own @code{end} command):
11828 > while-stepping 12
11829 > collect $regs, myglobal
11835 Note that @code{$pc} is not automatically collected by
11836 @code{while-stepping}; you need to explicitly collect that register if
11837 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
11840 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
11841 @kindex set default-collect
11842 @cindex default collection action
11843 This variable is a list of expressions to collect at each tracepoint
11844 hit. It is effectively an additional @code{collect} action prepended
11845 to every tracepoint action list. The expressions are parsed
11846 individually for each tracepoint, so for instance a variable named
11847 @code{xyz} may be interpreted as a global for one tracepoint, and a
11848 local for another, as appropriate to the tracepoint's location.
11850 @item show default-collect
11851 @kindex show default-collect
11852 Show the list of expressions that are collected by default at each
11857 @node Listing Tracepoints
11858 @subsection Listing Tracepoints
11861 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
11862 @kindex info tp @r{[}@var{n}@dots{}@r{]}
11863 @cindex information about tracepoints
11864 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
11865 Display information about the tracepoint @var{num}. If you don't
11866 specify a tracepoint number, displays information about all the
11867 tracepoints defined so far. The format is similar to that used for
11868 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
11869 command, simply restricting itself to tracepoints.
11871 A tracepoint's listing may include additional information specific to
11876 its passcount as given by the @code{passcount @var{n}} command
11879 the state about installed on target of each location
11883 (@value{GDBP}) @b{info trace}
11884 Num Type Disp Enb Address What
11885 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
11887 collect globfoo, $regs
11892 2 tracepoint keep y <MULTIPLE>
11894 2.1 y 0x0804859c in func4 at change-loc.h:35
11895 installed on target
11896 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
11897 installed on target
11898 2.3 y <PENDING> set_tracepoint
11899 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
11900 not installed on target
11905 This command can be abbreviated @code{info tp}.
11908 @node Listing Static Tracepoint Markers
11909 @subsection Listing Static Tracepoint Markers
11912 @kindex info static-tracepoint-markers
11913 @cindex information about static tracepoint markers
11914 @item info static-tracepoint-markers
11915 Display information about all static tracepoint markers defined in the
11918 For each marker, the following columns are printed:
11922 An incrementing counter, output to help readability. This is not a
11925 The marker ID, as reported by the target.
11926 @item Enabled or Disabled
11927 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
11928 that are not enabled.
11930 Where the marker is in your program, as a memory address.
11932 Where the marker is in the source for your program, as a file and line
11933 number. If the debug information included in the program does not
11934 allow @value{GDBN} to locate the source of the marker, this column
11935 will be left blank.
11939 In addition, the following information may be printed for each marker:
11943 User data passed to the tracing library by the marker call. In the
11944 UST backend, this is the format string passed as argument to the
11946 @item Static tracepoints probing the marker
11947 The list of static tracepoints attached to the marker.
11951 (@value{GDBP}) info static-tracepoint-markers
11952 Cnt ID Enb Address What
11953 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
11954 Data: number1 %d number2 %d
11955 Probed by static tracepoints: #2
11956 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
11962 @node Starting and Stopping Trace Experiments
11963 @subsection Starting and Stopping Trace Experiments
11966 @kindex tstart [ @var{notes} ]
11967 @cindex start a new trace experiment
11968 @cindex collected data discarded
11970 This command starts the trace experiment, and begins collecting data.
11971 It has the side effect of discarding all the data collected in the
11972 trace buffer during the previous trace experiment. If any arguments
11973 are supplied, they are taken as a note and stored with the trace
11974 experiment's state. The notes may be arbitrary text, and are
11975 especially useful with disconnected tracing in a multi-user context;
11976 the notes can explain what the trace is doing, supply user contact
11977 information, and so forth.
11979 @kindex tstop [ @var{notes} ]
11980 @cindex stop a running trace experiment
11982 This command stops the trace experiment. If any arguments are
11983 supplied, they are recorded with the experiment as a note. This is
11984 useful if you are stopping a trace started by someone else, for
11985 instance if the trace is interfering with the system's behavior and
11986 needs to be stopped quickly.
11988 @strong{Note}: a trace experiment and data collection may stop
11989 automatically if any tracepoint's passcount is reached
11990 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
11993 @cindex status of trace data collection
11994 @cindex trace experiment, status of
11996 This command displays the status of the current trace data
12000 Here is an example of the commands we described so far:
12003 (@value{GDBP}) @b{trace gdb_c_test}
12004 (@value{GDBP}) @b{actions}
12005 Enter actions for tracepoint #1, one per line.
12006 > collect $regs,$locals,$args
12007 > while-stepping 11
12011 (@value{GDBP}) @b{tstart}
12012 [time passes @dots{}]
12013 (@value{GDBP}) @b{tstop}
12016 @anchor{disconnected tracing}
12017 @cindex disconnected tracing
12018 You can choose to continue running the trace experiment even if
12019 @value{GDBN} disconnects from the target, voluntarily or
12020 involuntarily. For commands such as @code{detach}, the debugger will
12021 ask what you want to do with the trace. But for unexpected
12022 terminations (@value{GDBN} crash, network outage), it would be
12023 unfortunate to lose hard-won trace data, so the variable
12024 @code{disconnected-tracing} lets you decide whether the trace should
12025 continue running without @value{GDBN}.
12028 @item set disconnected-tracing on
12029 @itemx set disconnected-tracing off
12030 @kindex set disconnected-tracing
12031 Choose whether a tracing run should continue to run if @value{GDBN}
12032 has disconnected from the target. Note that @code{detach} or
12033 @code{quit} will ask you directly what to do about a running trace no
12034 matter what this variable's setting, so the variable is mainly useful
12035 for handling unexpected situations, such as loss of the network.
12037 @item show disconnected-tracing
12038 @kindex show disconnected-tracing
12039 Show the current choice for disconnected tracing.
12043 When you reconnect to the target, the trace experiment may or may not
12044 still be running; it might have filled the trace buffer in the
12045 meantime, or stopped for one of the other reasons. If it is running,
12046 it will continue after reconnection.
12048 Upon reconnection, the target will upload information about the
12049 tracepoints in effect. @value{GDBN} will then compare that
12050 information to the set of tracepoints currently defined, and attempt
12051 to match them up, allowing for the possibility that the numbers may
12052 have changed due to creation and deletion in the meantime. If one of
12053 the target's tracepoints does not match any in @value{GDBN}, the
12054 debugger will create a new tracepoint, so that you have a number with
12055 which to specify that tracepoint. This matching-up process is
12056 necessarily heuristic, and it may result in useless tracepoints being
12057 created; you may simply delete them if they are of no use.
12059 @cindex circular trace buffer
12060 If your target agent supports a @dfn{circular trace buffer}, then you
12061 can run a trace experiment indefinitely without filling the trace
12062 buffer; when space runs out, the agent deletes already-collected trace
12063 frames, oldest first, until there is enough room to continue
12064 collecting. This is especially useful if your tracepoints are being
12065 hit too often, and your trace gets terminated prematurely because the
12066 buffer is full. To ask for a circular trace buffer, simply set
12067 @samp{circular-trace-buffer} to on. You can set this at any time,
12068 including during tracing; if the agent can do it, it will change
12069 buffer handling on the fly, otherwise it will not take effect until
12073 @item set circular-trace-buffer on
12074 @itemx set circular-trace-buffer off
12075 @kindex set circular-trace-buffer
12076 Choose whether a tracing run should use a linear or circular buffer
12077 for trace data. A linear buffer will not lose any trace data, but may
12078 fill up prematurely, while a circular buffer will discard old trace
12079 data, but it will have always room for the latest tracepoint hits.
12081 @item show circular-trace-buffer
12082 @kindex show circular-trace-buffer
12083 Show the current choice for the trace buffer. Note that this may not
12084 match the agent's current buffer handling, nor is it guaranteed to
12085 match the setting that might have been in effect during a past run,
12086 for instance if you are looking at frames from a trace file.
12091 @item set trace-buffer-size @var{n}
12092 @itemx set trace-buffer-size unlimited
12093 @kindex set trace-buffer-size
12094 Request that the target use a trace buffer of @var{n} bytes. Not all
12095 targets will honor the request; they may have a compiled-in size for
12096 the trace buffer, or some other limitation. Set to a value of
12097 @code{unlimited} or @code{-1} to let the target use whatever size it
12098 likes. This is also the default.
12100 @item show trace-buffer-size
12101 @kindex show trace-buffer-size
12102 Show the current requested size for the trace buffer. Note that this
12103 will only match the actual size if the target supports size-setting,
12104 and was able to handle the requested size. For instance, if the
12105 target can only change buffer size between runs, this variable will
12106 not reflect the change until the next run starts. Use @code{tstatus}
12107 to get a report of the actual buffer size.
12111 @item set trace-user @var{text}
12112 @kindex set trace-user
12114 @item show trace-user
12115 @kindex show trace-user
12117 @item set trace-notes @var{text}
12118 @kindex set trace-notes
12119 Set the trace run's notes.
12121 @item show trace-notes
12122 @kindex show trace-notes
12123 Show the trace run's notes.
12125 @item set trace-stop-notes @var{text}
12126 @kindex set trace-stop-notes
12127 Set the trace run's stop notes. The handling of the note is as for
12128 @code{tstop} arguments; the set command is convenient way to fix a
12129 stop note that is mistaken or incomplete.
12131 @item show trace-stop-notes
12132 @kindex show trace-stop-notes
12133 Show the trace run's stop notes.
12137 @node Tracepoint Restrictions
12138 @subsection Tracepoint Restrictions
12140 @cindex tracepoint restrictions
12141 There are a number of restrictions on the use of tracepoints. As
12142 described above, tracepoint data gathering occurs on the target
12143 without interaction from @value{GDBN}. Thus the full capabilities of
12144 the debugger are not available during data gathering, and then at data
12145 examination time, you will be limited by only having what was
12146 collected. The following items describe some common problems, but it
12147 is not exhaustive, and you may run into additional difficulties not
12153 Tracepoint expressions are intended to gather objects (lvalues). Thus
12154 the full flexibility of GDB's expression evaluator is not available.
12155 You cannot call functions, cast objects to aggregate types, access
12156 convenience variables or modify values (except by assignment to trace
12157 state variables). Some language features may implicitly call
12158 functions (for instance Objective-C fields with accessors), and therefore
12159 cannot be collected either.
12162 Collection of local variables, either individually or in bulk with
12163 @code{$locals} or @code{$args}, during @code{while-stepping} may
12164 behave erratically. The stepping action may enter a new scope (for
12165 instance by stepping into a function), or the location of the variable
12166 may change (for instance it is loaded into a register). The
12167 tracepoint data recorded uses the location information for the
12168 variables that is correct for the tracepoint location. When the
12169 tracepoint is created, it is not possible, in general, to determine
12170 where the steps of a @code{while-stepping} sequence will advance the
12171 program---particularly if a conditional branch is stepped.
12174 Collection of an incompletely-initialized or partially-destroyed object
12175 may result in something that @value{GDBN} cannot display, or displays
12176 in a misleading way.
12179 When @value{GDBN} displays a pointer to character it automatically
12180 dereferences the pointer to also display characters of the string
12181 being pointed to. However, collecting the pointer during tracing does
12182 not automatically collect the string. You need to explicitly
12183 dereference the pointer and provide size information if you want to
12184 collect not only the pointer, but the memory pointed to. For example,
12185 @code{*ptr@@50} can be used to collect the 50 element array pointed to
12189 It is not possible to collect a complete stack backtrace at a
12190 tracepoint. Instead, you may collect the registers and a few hundred
12191 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
12192 (adjust to use the name of the actual stack pointer register on your
12193 target architecture, and the amount of stack you wish to capture).
12194 Then the @code{backtrace} command will show a partial backtrace when
12195 using a trace frame. The number of stack frames that can be examined
12196 depends on the sizes of the frames in the collected stack. Note that
12197 if you ask for a block so large that it goes past the bottom of the
12198 stack, the target agent may report an error trying to read from an
12202 If you do not collect registers at a tracepoint, @value{GDBN} can
12203 infer that the value of @code{$pc} must be the same as the address of
12204 the tracepoint and use that when you are looking at a trace frame
12205 for that tracepoint. However, this cannot work if the tracepoint has
12206 multiple locations (for instance if it was set in a function that was
12207 inlined), or if it has a @code{while-stepping} loop. In those cases
12208 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
12213 @node Analyze Collected Data
12214 @section Using the Collected Data
12216 After the tracepoint experiment ends, you use @value{GDBN} commands
12217 for examining the trace data. The basic idea is that each tracepoint
12218 collects a trace @dfn{snapshot} every time it is hit and another
12219 snapshot every time it single-steps. All these snapshots are
12220 consecutively numbered from zero and go into a buffer, and you can
12221 examine them later. The way you examine them is to @dfn{focus} on a
12222 specific trace snapshot. When the remote stub is focused on a trace
12223 snapshot, it will respond to all @value{GDBN} requests for memory and
12224 registers by reading from the buffer which belongs to that snapshot,
12225 rather than from @emph{real} memory or registers of the program being
12226 debugged. This means that @strong{all} @value{GDBN} commands
12227 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
12228 behave as if we were currently debugging the program state as it was
12229 when the tracepoint occurred. Any requests for data that are not in
12230 the buffer will fail.
12233 * tfind:: How to select a trace snapshot
12234 * tdump:: How to display all data for a snapshot
12235 * save tracepoints:: How to save tracepoints for a future run
12239 @subsection @code{tfind @var{n}}
12242 @cindex select trace snapshot
12243 @cindex find trace snapshot
12244 The basic command for selecting a trace snapshot from the buffer is
12245 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
12246 counting from zero. If no argument @var{n} is given, the next
12247 snapshot is selected.
12249 Here are the various forms of using the @code{tfind} command.
12253 Find the first snapshot in the buffer. This is a synonym for
12254 @code{tfind 0} (since 0 is the number of the first snapshot).
12257 Stop debugging trace snapshots, resume @emph{live} debugging.
12260 Same as @samp{tfind none}.
12263 No argument means find the next trace snapshot.
12266 Find the previous trace snapshot before the current one. This permits
12267 retracing earlier steps.
12269 @item tfind tracepoint @var{num}
12270 Find the next snapshot associated with tracepoint @var{num}. Search
12271 proceeds forward from the last examined trace snapshot. If no
12272 argument @var{num} is given, it means find the next snapshot collected
12273 for the same tracepoint as the current snapshot.
12275 @item tfind pc @var{addr}
12276 Find the next snapshot associated with the value @var{addr} of the
12277 program counter. Search proceeds forward from the last examined trace
12278 snapshot. If no argument @var{addr} is given, it means find the next
12279 snapshot with the same value of PC as the current snapshot.
12281 @item tfind outside @var{addr1}, @var{addr2}
12282 Find the next snapshot whose PC is outside the given range of
12283 addresses (exclusive).
12285 @item tfind range @var{addr1}, @var{addr2}
12286 Find the next snapshot whose PC is between @var{addr1} and
12287 @var{addr2} (inclusive).
12289 @item tfind line @r{[}@var{file}:@r{]}@var{n}
12290 Find the next snapshot associated with the source line @var{n}. If
12291 the optional argument @var{file} is given, refer to line @var{n} in
12292 that source file. Search proceeds forward from the last examined
12293 trace snapshot. If no argument @var{n} is given, it means find the
12294 next line other than the one currently being examined; thus saying
12295 @code{tfind line} repeatedly can appear to have the same effect as
12296 stepping from line to line in a @emph{live} debugging session.
12299 The default arguments for the @code{tfind} commands are specifically
12300 designed to make it easy to scan through the trace buffer. For
12301 instance, @code{tfind} with no argument selects the next trace
12302 snapshot, and @code{tfind -} with no argument selects the previous
12303 trace snapshot. So, by giving one @code{tfind} command, and then
12304 simply hitting @key{RET} repeatedly you can examine all the trace
12305 snapshots in order. Or, by saying @code{tfind -} and then hitting
12306 @key{RET} repeatedly you can examine the snapshots in reverse order.
12307 The @code{tfind line} command with no argument selects the snapshot
12308 for the next source line executed. The @code{tfind pc} command with
12309 no argument selects the next snapshot with the same program counter
12310 (PC) as the current frame. The @code{tfind tracepoint} command with
12311 no argument selects the next trace snapshot collected by the same
12312 tracepoint as the current one.
12314 In addition to letting you scan through the trace buffer manually,
12315 these commands make it easy to construct @value{GDBN} scripts that
12316 scan through the trace buffer and print out whatever collected data
12317 you are interested in. Thus, if we want to examine the PC, FP, and SP
12318 registers from each trace frame in the buffer, we can say this:
12321 (@value{GDBP}) @b{tfind start}
12322 (@value{GDBP}) @b{while ($trace_frame != -1)}
12323 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
12324 $trace_frame, $pc, $sp, $fp
12328 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
12329 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
12330 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
12331 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
12332 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
12333 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
12334 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
12335 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
12336 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
12337 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
12338 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
12341 Or, if we want to examine the variable @code{X} at each source line in
12345 (@value{GDBP}) @b{tfind start}
12346 (@value{GDBP}) @b{while ($trace_frame != -1)}
12347 > printf "Frame %d, X == %d\n", $trace_frame, X
12357 @subsection @code{tdump}
12359 @cindex dump all data collected at tracepoint
12360 @cindex tracepoint data, display
12362 This command takes no arguments. It prints all the data collected at
12363 the current trace snapshot.
12366 (@value{GDBP}) @b{trace 444}
12367 (@value{GDBP}) @b{actions}
12368 Enter actions for tracepoint #2, one per line:
12369 > collect $regs, $locals, $args, gdb_long_test
12372 (@value{GDBP}) @b{tstart}
12374 (@value{GDBP}) @b{tfind line 444}
12375 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
12377 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
12379 (@value{GDBP}) @b{tdump}
12380 Data collected at tracepoint 2, trace frame 1:
12381 d0 0xc4aa0085 -995491707
12385 d4 0x71aea3d 119204413
12388 d7 0x380035 3670069
12389 a0 0x19e24a 1696330
12390 a1 0x3000668 50333288
12392 a3 0x322000 3284992
12393 a4 0x3000698 50333336
12394 a5 0x1ad3cc 1758156
12395 fp 0x30bf3c 0x30bf3c
12396 sp 0x30bf34 0x30bf34
12398 pc 0x20b2c8 0x20b2c8
12402 p = 0x20e5b4 "gdb-test"
12409 gdb_long_test = 17 '\021'
12414 @code{tdump} works by scanning the tracepoint's current collection
12415 actions and printing the value of each expression listed. So
12416 @code{tdump} can fail, if after a run, you change the tracepoint's
12417 actions to mention variables that were not collected during the run.
12419 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
12420 uses the collected value of @code{$pc} to distinguish between trace
12421 frames that were collected at the tracepoint hit, and frames that were
12422 collected while stepping. This allows it to correctly choose whether
12423 to display the basic list of collections, or the collections from the
12424 body of the while-stepping loop. However, if @code{$pc} was not collected,
12425 then @code{tdump} will always attempt to dump using the basic collection
12426 list, and may fail if a while-stepping frame does not include all the
12427 same data that is collected at the tracepoint hit.
12428 @c This is getting pretty arcane, example would be good.
12430 @node save tracepoints
12431 @subsection @code{save tracepoints @var{filename}}
12432 @kindex save tracepoints
12433 @kindex save-tracepoints
12434 @cindex save tracepoints for future sessions
12436 This command saves all current tracepoint definitions together with
12437 their actions and passcounts, into a file @file{@var{filename}}
12438 suitable for use in a later debugging session. To read the saved
12439 tracepoint definitions, use the @code{source} command (@pxref{Command
12440 Files}). The @w{@code{save-tracepoints}} command is a deprecated
12441 alias for @w{@code{save tracepoints}}
12443 @node Tracepoint Variables
12444 @section Convenience Variables for Tracepoints
12445 @cindex tracepoint variables
12446 @cindex convenience variables for tracepoints
12449 @vindex $trace_frame
12450 @item (int) $trace_frame
12451 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
12452 snapshot is selected.
12454 @vindex $tracepoint
12455 @item (int) $tracepoint
12456 The tracepoint for the current trace snapshot.
12458 @vindex $trace_line
12459 @item (int) $trace_line
12460 The line number for the current trace snapshot.
12462 @vindex $trace_file
12463 @item (char []) $trace_file
12464 The source file for the current trace snapshot.
12466 @vindex $trace_func
12467 @item (char []) $trace_func
12468 The name of the function containing @code{$tracepoint}.
12471 Note: @code{$trace_file} is not suitable for use in @code{printf},
12472 use @code{output} instead.
12474 Here's a simple example of using these convenience variables for
12475 stepping through all the trace snapshots and printing some of their
12476 data. Note that these are not the same as trace state variables,
12477 which are managed by the target.
12480 (@value{GDBP}) @b{tfind start}
12482 (@value{GDBP}) @b{while $trace_frame != -1}
12483 > output $trace_file
12484 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
12490 @section Using Trace Files
12491 @cindex trace files
12493 In some situations, the target running a trace experiment may no
12494 longer be available; perhaps it crashed, or the hardware was needed
12495 for a different activity. To handle these cases, you can arrange to
12496 dump the trace data into a file, and later use that file as a source
12497 of trace data, via the @code{target tfile} command.
12502 @item tsave [ -r ] @var{filename}
12503 @itemx tsave [-ctf] @var{dirname}
12504 Save the trace data to @var{filename}. By default, this command
12505 assumes that @var{filename} refers to the host filesystem, so if
12506 necessary @value{GDBN} will copy raw trace data up from the target and
12507 then save it. If the target supports it, you can also supply the
12508 optional argument @code{-r} (``remote'') to direct the target to save
12509 the data directly into @var{filename} in its own filesystem, which may be
12510 more efficient if the trace buffer is very large. (Note, however, that
12511 @code{target tfile} can only read from files accessible to the host.)
12512 By default, this command will save trace frame in tfile format.
12513 You can supply the optional argument @code{-ctf} to save date in CTF
12514 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
12515 that can be shared by multiple debugging and tracing tools. Please go to
12516 @indicateurl{http://www.efficios.com/ctf} to get more information.
12518 @kindex target tfile
12522 @item target tfile @var{filename}
12523 @itemx target ctf @var{dirname}
12524 Use the file named @var{filename} or directory named @var{dirname} as
12525 a source of trace data. Commands that examine data work as they do with
12526 a live target, but it is not possible to run any new trace experiments.
12527 @code{tstatus} will report the state of the trace run at the moment
12528 the data was saved, as well as the current trace frame you are examining.
12529 @var{filename} or @var{dirname} must be on a filesystem accessible to
12533 (@value{GDBP}) target ctf ctf.ctf
12534 (@value{GDBP}) tfind
12535 Found trace frame 0, tracepoint 2
12536 39 ++a; /* set tracepoint 1 here */
12537 (@value{GDBP}) tdump
12538 Data collected at tracepoint 2, trace frame 0:
12542 c = @{"123", "456", "789", "123", "456", "789"@}
12543 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
12551 @chapter Debugging Programs That Use Overlays
12554 If your program is too large to fit completely in your target system's
12555 memory, you can sometimes use @dfn{overlays} to work around this
12556 problem. @value{GDBN} provides some support for debugging programs that
12560 * How Overlays Work:: A general explanation of overlays.
12561 * Overlay Commands:: Managing overlays in @value{GDBN}.
12562 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
12563 mapped by asking the inferior.
12564 * Overlay Sample Program:: A sample program using overlays.
12567 @node How Overlays Work
12568 @section How Overlays Work
12569 @cindex mapped overlays
12570 @cindex unmapped overlays
12571 @cindex load address, overlay's
12572 @cindex mapped address
12573 @cindex overlay area
12575 Suppose you have a computer whose instruction address space is only 64
12576 kilobytes long, but which has much more memory which can be accessed by
12577 other means: special instructions, segment registers, or memory
12578 management hardware, for example. Suppose further that you want to
12579 adapt a program which is larger than 64 kilobytes to run on this system.
12581 One solution is to identify modules of your program which are relatively
12582 independent, and need not call each other directly; call these modules
12583 @dfn{overlays}. Separate the overlays from the main program, and place
12584 their machine code in the larger memory. Place your main program in
12585 instruction memory, but leave at least enough space there to hold the
12586 largest overlay as well.
12588 Now, to call a function located in an overlay, you must first copy that
12589 overlay's machine code from the large memory into the space set aside
12590 for it in the instruction memory, and then jump to its entry point
12593 @c NB: In the below the mapped area's size is greater or equal to the
12594 @c size of all overlays. This is intentional to remind the developer
12595 @c that overlays don't necessarily need to be the same size.
12599 Data Instruction Larger
12600 Address Space Address Space Address Space
12601 +-----------+ +-----------+ +-----------+
12603 +-----------+ +-----------+ +-----------+<-- overlay 1
12604 | program | | main | .----| overlay 1 | load address
12605 | variables | | program | | +-----------+
12606 | and heap | | | | | |
12607 +-----------+ | | | +-----------+<-- overlay 2
12608 | | +-----------+ | | | load address
12609 +-----------+ | | | .-| overlay 2 |
12611 mapped --->+-----------+ | | +-----------+
12612 address | | | | | |
12613 | overlay | <-' | | |
12614 | area | <---' +-----------+<-- overlay 3
12615 | | <---. | | load address
12616 +-----------+ `--| overlay 3 |
12623 @anchor{A code overlay}A code overlay
12627 The diagram (@pxref{A code overlay}) shows a system with separate data
12628 and instruction address spaces. To map an overlay, the program copies
12629 its code from the larger address space to the instruction address space.
12630 Since the overlays shown here all use the same mapped address, only one
12631 may be mapped at a time. For a system with a single address space for
12632 data and instructions, the diagram would be similar, except that the
12633 program variables and heap would share an address space with the main
12634 program and the overlay area.
12636 An overlay loaded into instruction memory and ready for use is called a
12637 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
12638 instruction memory. An overlay not present (or only partially present)
12639 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
12640 is its address in the larger memory. The mapped address is also called
12641 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
12642 called the @dfn{load memory address}, or @dfn{LMA}.
12644 Unfortunately, overlays are not a completely transparent way to adapt a
12645 program to limited instruction memory. They introduce a new set of
12646 global constraints you must keep in mind as you design your program:
12651 Before calling or returning to a function in an overlay, your program
12652 must make sure that overlay is actually mapped. Otherwise, the call or
12653 return will transfer control to the right address, but in the wrong
12654 overlay, and your program will probably crash.
12657 If the process of mapping an overlay is expensive on your system, you
12658 will need to choose your overlays carefully to minimize their effect on
12659 your program's performance.
12662 The executable file you load onto your system must contain each
12663 overlay's instructions, appearing at the overlay's load address, not its
12664 mapped address. However, each overlay's instructions must be relocated
12665 and its symbols defined as if the overlay were at its mapped address.
12666 You can use GNU linker scripts to specify different load and relocation
12667 addresses for pieces of your program; see @ref{Overlay Description,,,
12668 ld.info, Using ld: the GNU linker}.
12671 The procedure for loading executable files onto your system must be able
12672 to load their contents into the larger address space as well as the
12673 instruction and data spaces.
12677 The overlay system described above is rather simple, and could be
12678 improved in many ways:
12683 If your system has suitable bank switch registers or memory management
12684 hardware, you could use those facilities to make an overlay's load area
12685 contents simply appear at their mapped address in instruction space.
12686 This would probably be faster than copying the overlay to its mapped
12687 area in the usual way.
12690 If your overlays are small enough, you could set aside more than one
12691 overlay area, and have more than one overlay mapped at a time.
12694 You can use overlays to manage data, as well as instructions. In
12695 general, data overlays are even less transparent to your design than
12696 code overlays: whereas code overlays only require care when you call or
12697 return to functions, data overlays require care every time you access
12698 the data. Also, if you change the contents of a data overlay, you
12699 must copy its contents back out to its load address before you can copy a
12700 different data overlay into the same mapped area.
12705 @node Overlay Commands
12706 @section Overlay Commands
12708 To use @value{GDBN}'s overlay support, each overlay in your program must
12709 correspond to a separate section of the executable file. The section's
12710 virtual memory address and load memory address must be the overlay's
12711 mapped and load addresses. Identifying overlays with sections allows
12712 @value{GDBN} to determine the appropriate address of a function or
12713 variable, depending on whether the overlay is mapped or not.
12715 @value{GDBN}'s overlay commands all start with the word @code{overlay};
12716 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
12721 Disable @value{GDBN}'s overlay support. When overlay support is
12722 disabled, @value{GDBN} assumes that all functions and variables are
12723 always present at their mapped addresses. By default, @value{GDBN}'s
12724 overlay support is disabled.
12726 @item overlay manual
12727 @cindex manual overlay debugging
12728 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
12729 relies on you to tell it which overlays are mapped, and which are not,
12730 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
12731 commands described below.
12733 @item overlay map-overlay @var{overlay}
12734 @itemx overlay map @var{overlay}
12735 @cindex map an overlay
12736 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
12737 be the name of the object file section containing the overlay. When an
12738 overlay is mapped, @value{GDBN} assumes it can find the overlay's
12739 functions and variables at their mapped addresses. @value{GDBN} assumes
12740 that any other overlays whose mapped ranges overlap that of
12741 @var{overlay} are now unmapped.
12743 @item overlay unmap-overlay @var{overlay}
12744 @itemx overlay unmap @var{overlay}
12745 @cindex unmap an overlay
12746 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
12747 must be the name of the object file section containing the overlay.
12748 When an overlay is unmapped, @value{GDBN} assumes it can find the
12749 overlay's functions and variables at their load addresses.
12752 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
12753 consults a data structure the overlay manager maintains in the inferior
12754 to see which overlays are mapped. For details, see @ref{Automatic
12755 Overlay Debugging}.
12757 @item overlay load-target
12758 @itemx overlay load
12759 @cindex reloading the overlay table
12760 Re-read the overlay table from the inferior. Normally, @value{GDBN}
12761 re-reads the table @value{GDBN} automatically each time the inferior
12762 stops, so this command should only be necessary if you have changed the
12763 overlay mapping yourself using @value{GDBN}. This command is only
12764 useful when using automatic overlay debugging.
12766 @item overlay list-overlays
12767 @itemx overlay list
12768 @cindex listing mapped overlays
12769 Display a list of the overlays currently mapped, along with their mapped
12770 addresses, load addresses, and sizes.
12774 Normally, when @value{GDBN} prints a code address, it includes the name
12775 of the function the address falls in:
12778 (@value{GDBP}) print main
12779 $3 = @{int ()@} 0x11a0 <main>
12782 When overlay debugging is enabled, @value{GDBN} recognizes code in
12783 unmapped overlays, and prints the names of unmapped functions with
12784 asterisks around them. For example, if @code{foo} is a function in an
12785 unmapped overlay, @value{GDBN} prints it this way:
12788 (@value{GDBP}) overlay list
12789 No sections are mapped.
12790 (@value{GDBP}) print foo
12791 $5 = @{int (int)@} 0x100000 <*foo*>
12794 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
12798 (@value{GDBP}) overlay list
12799 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
12800 mapped at 0x1016 - 0x104a
12801 (@value{GDBP}) print foo
12802 $6 = @{int (int)@} 0x1016 <foo>
12805 When overlay debugging is enabled, @value{GDBN} can find the correct
12806 address for functions and variables in an overlay, whether or not the
12807 overlay is mapped. This allows most @value{GDBN} commands, like
12808 @code{break} and @code{disassemble}, to work normally, even on unmapped
12809 code. However, @value{GDBN}'s breakpoint support has some limitations:
12813 @cindex breakpoints in overlays
12814 @cindex overlays, setting breakpoints in
12815 You can set breakpoints in functions in unmapped overlays, as long as
12816 @value{GDBN} can write to the overlay at its load address.
12818 @value{GDBN} can not set hardware or simulator-based breakpoints in
12819 unmapped overlays. However, if you set a breakpoint at the end of your
12820 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
12821 you are using manual overlay management), @value{GDBN} will re-set its
12822 breakpoints properly.
12826 @node Automatic Overlay Debugging
12827 @section Automatic Overlay Debugging
12828 @cindex automatic overlay debugging
12830 @value{GDBN} can automatically track which overlays are mapped and which
12831 are not, given some simple co-operation from the overlay manager in the
12832 inferior. If you enable automatic overlay debugging with the
12833 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
12834 looks in the inferior's memory for certain variables describing the
12835 current state of the overlays.
12837 Here are the variables your overlay manager must define to support
12838 @value{GDBN}'s automatic overlay debugging:
12842 @item @code{_ovly_table}:
12843 This variable must be an array of the following structures:
12848 /* The overlay's mapped address. */
12851 /* The size of the overlay, in bytes. */
12852 unsigned long size;
12854 /* The overlay's load address. */
12857 /* Non-zero if the overlay is currently mapped;
12859 unsigned long mapped;
12863 @item @code{_novlys}:
12864 This variable must be a four-byte signed integer, holding the total
12865 number of elements in @code{_ovly_table}.
12869 To decide whether a particular overlay is mapped or not, @value{GDBN}
12870 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
12871 @code{lma} members equal the VMA and LMA of the overlay's section in the
12872 executable file. When @value{GDBN} finds a matching entry, it consults
12873 the entry's @code{mapped} member to determine whether the overlay is
12876 In addition, your overlay manager may define a function called
12877 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
12878 will silently set a breakpoint there. If the overlay manager then
12879 calls this function whenever it has changed the overlay table, this
12880 will enable @value{GDBN} to accurately keep track of which overlays
12881 are in program memory, and update any breakpoints that may be set
12882 in overlays. This will allow breakpoints to work even if the
12883 overlays are kept in ROM or other non-writable memory while they
12884 are not being executed.
12886 @node Overlay Sample Program
12887 @section Overlay Sample Program
12888 @cindex overlay example program
12890 When linking a program which uses overlays, you must place the overlays
12891 at their load addresses, while relocating them to run at their mapped
12892 addresses. To do this, you must write a linker script (@pxref{Overlay
12893 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
12894 since linker scripts are specific to a particular host system, target
12895 architecture, and target memory layout, this manual cannot provide
12896 portable sample code demonstrating @value{GDBN}'s overlay support.
12898 However, the @value{GDBN} source distribution does contain an overlaid
12899 program, with linker scripts for a few systems, as part of its test
12900 suite. The program consists of the following files from
12901 @file{gdb/testsuite/gdb.base}:
12905 The main program file.
12907 A simple overlay manager, used by @file{overlays.c}.
12912 Overlay modules, loaded and used by @file{overlays.c}.
12915 Linker scripts for linking the test program on the @code{d10v-elf}
12916 and @code{m32r-elf} targets.
12919 You can build the test program using the @code{d10v-elf} GCC
12920 cross-compiler like this:
12923 $ d10v-elf-gcc -g -c overlays.c
12924 $ d10v-elf-gcc -g -c ovlymgr.c
12925 $ d10v-elf-gcc -g -c foo.c
12926 $ d10v-elf-gcc -g -c bar.c
12927 $ d10v-elf-gcc -g -c baz.c
12928 $ d10v-elf-gcc -g -c grbx.c
12929 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
12930 baz.o grbx.o -Wl,-Td10v.ld -o overlays
12933 The build process is identical for any other architecture, except that
12934 you must substitute the appropriate compiler and linker script for the
12935 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
12939 @chapter Using @value{GDBN} with Different Languages
12942 Although programming languages generally have common aspects, they are
12943 rarely expressed in the same manner. For instance, in ANSI C,
12944 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
12945 Modula-2, it is accomplished by @code{p^}. Values can also be
12946 represented (and displayed) differently. Hex numbers in C appear as
12947 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
12949 @cindex working language
12950 Language-specific information is built into @value{GDBN} for some languages,
12951 allowing you to express operations like the above in your program's
12952 native language, and allowing @value{GDBN} to output values in a manner
12953 consistent with the syntax of your program's native language. The
12954 language you use to build expressions is called the @dfn{working
12958 * Setting:: Switching between source languages
12959 * Show:: Displaying the language
12960 * Checks:: Type and range checks
12961 * Supported Languages:: Supported languages
12962 * Unsupported Languages:: Unsupported languages
12966 @section Switching Between Source Languages
12968 There are two ways to control the working language---either have @value{GDBN}
12969 set it automatically, or select it manually yourself. You can use the
12970 @code{set language} command for either purpose. On startup, @value{GDBN}
12971 defaults to setting the language automatically. The working language is
12972 used to determine how expressions you type are interpreted, how values
12975 In addition to the working language, every source file that
12976 @value{GDBN} knows about has its own working language. For some object
12977 file formats, the compiler might indicate which language a particular
12978 source file is in. However, most of the time @value{GDBN} infers the
12979 language from the name of the file. The language of a source file
12980 controls whether C@t{++} names are demangled---this way @code{backtrace} can
12981 show each frame appropriately for its own language. There is no way to
12982 set the language of a source file from within @value{GDBN}, but you can
12983 set the language associated with a filename extension. @xref{Show, ,
12984 Displaying the Language}.
12986 This is most commonly a problem when you use a program, such
12987 as @code{cfront} or @code{f2c}, that generates C but is written in
12988 another language. In that case, make the
12989 program use @code{#line} directives in its C output; that way
12990 @value{GDBN} will know the correct language of the source code of the original
12991 program, and will display that source code, not the generated C code.
12994 * Filenames:: Filename extensions and languages.
12995 * Manually:: Setting the working language manually
12996 * Automatically:: Having @value{GDBN} infer the source language
13000 @subsection List of Filename Extensions and Languages
13002 If a source file name ends in one of the following extensions, then
13003 @value{GDBN} infers that its language is the one indicated.
13021 C@t{++} source file
13027 Objective-C source file
13031 Fortran source file
13034 Modula-2 source file
13038 Assembler source file. This actually behaves almost like C, but
13039 @value{GDBN} does not skip over function prologues when stepping.
13042 In addition, you may set the language associated with a filename
13043 extension. @xref{Show, , Displaying the Language}.
13046 @subsection Setting the Working Language
13048 If you allow @value{GDBN} to set the language automatically,
13049 expressions are interpreted the same way in your debugging session and
13052 @kindex set language
13053 If you wish, you may set the language manually. To do this, issue the
13054 command @samp{set language @var{lang}}, where @var{lang} is the name of
13055 a language, such as
13056 @code{c} or @code{modula-2}.
13057 For a list of the supported languages, type @samp{set language}.
13059 Setting the language manually prevents @value{GDBN} from updating the working
13060 language automatically. This can lead to confusion if you try
13061 to debug a program when the working language is not the same as the
13062 source language, when an expression is acceptable to both
13063 languages---but means different things. For instance, if the current
13064 source file were written in C, and @value{GDBN} was parsing Modula-2, a
13072 might not have the effect you intended. In C, this means to add
13073 @code{b} and @code{c} and place the result in @code{a}. The result
13074 printed would be the value of @code{a}. In Modula-2, this means to compare
13075 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
13077 @node Automatically
13078 @subsection Having @value{GDBN} Infer the Source Language
13080 To have @value{GDBN} set the working language automatically, use
13081 @samp{set language local} or @samp{set language auto}. @value{GDBN}
13082 then infers the working language. That is, when your program stops in a
13083 frame (usually by encountering a breakpoint), @value{GDBN} sets the
13084 working language to the language recorded for the function in that
13085 frame. If the language for a frame is unknown (that is, if the function
13086 or block corresponding to the frame was defined in a source file that
13087 does not have a recognized extension), the current working language is
13088 not changed, and @value{GDBN} issues a warning.
13090 This may not seem necessary for most programs, which are written
13091 entirely in one source language. However, program modules and libraries
13092 written in one source language can be used by a main program written in
13093 a different source language. Using @samp{set language auto} in this
13094 case frees you from having to set the working language manually.
13097 @section Displaying the Language
13099 The following commands help you find out which language is the
13100 working language, and also what language source files were written in.
13103 @item show language
13104 @kindex show language
13105 Display the current working language. This is the
13106 language you can use with commands such as @code{print} to
13107 build and compute expressions that may involve variables in your program.
13110 @kindex info frame@r{, show the source language}
13111 Display the source language for this frame. This language becomes the
13112 working language if you use an identifier from this frame.
13113 @xref{Frame Info, ,Information about a Frame}, to identify the other
13114 information listed here.
13117 @kindex info source@r{, show the source language}
13118 Display the source language of this source file.
13119 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
13120 information listed here.
13123 In unusual circumstances, you may have source files with extensions
13124 not in the standard list. You can then set the extension associated
13125 with a language explicitly:
13128 @item set extension-language @var{ext} @var{language}
13129 @kindex set extension-language
13130 Tell @value{GDBN} that source files with extension @var{ext} are to be
13131 assumed as written in the source language @var{language}.
13133 @item info extensions
13134 @kindex info extensions
13135 List all the filename extensions and the associated languages.
13139 @section Type and Range Checking
13141 Some languages are designed to guard you against making seemingly common
13142 errors through a series of compile- and run-time checks. These include
13143 checking the type of arguments to functions and operators and making
13144 sure mathematical overflows are caught at run time. Checks such as
13145 these help to ensure a program's correctness once it has been compiled
13146 by eliminating type mismatches and providing active checks for range
13147 errors when your program is running.
13149 By default @value{GDBN} checks for these errors according to the
13150 rules of the current source language. Although @value{GDBN} does not check
13151 the statements in your program, it can check expressions entered directly
13152 into @value{GDBN} for evaluation via the @code{print} command, for example.
13155 * Type Checking:: An overview of type checking
13156 * Range Checking:: An overview of range checking
13159 @cindex type checking
13160 @cindex checks, type
13161 @node Type Checking
13162 @subsection An Overview of Type Checking
13164 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
13165 arguments to operators and functions have to be of the correct type,
13166 otherwise an error occurs. These checks prevent type mismatch
13167 errors from ever causing any run-time problems. For example,
13170 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
13172 (@value{GDBP}) print obj.my_method (0)
13175 (@value{GDBP}) print obj.my_method (0x1234)
13176 Cannot resolve method klass::my_method to any overloaded instance
13179 The second example fails because in C@t{++} the integer constant
13180 @samp{0x1234} is not type-compatible with the pointer parameter type.
13182 For the expressions you use in @value{GDBN} commands, you can tell
13183 @value{GDBN} to not enforce strict type checking or
13184 to treat any mismatches as errors and abandon the expression;
13185 When type checking is disabled, @value{GDBN} successfully evaluates
13186 expressions like the second example above.
13188 Even if type checking is off, there may be other reasons
13189 related to type that prevent @value{GDBN} from evaluating an expression.
13190 For instance, @value{GDBN} does not know how to add an @code{int} and
13191 a @code{struct foo}. These particular type errors have nothing to do
13192 with the language in use and usually arise from expressions which make
13193 little sense to evaluate anyway.
13195 @value{GDBN} provides some additional commands for controlling type checking:
13197 @kindex set check type
13198 @kindex show check type
13200 @item set check type on
13201 @itemx set check type off
13202 Set strict type checking on or off. If any type mismatches occur in
13203 evaluating an expression while type checking is on, @value{GDBN} prints a
13204 message and aborts evaluation of the expression.
13206 @item show check type
13207 Show the current setting of type checking and whether @value{GDBN}
13208 is enforcing strict type checking rules.
13211 @cindex range checking
13212 @cindex checks, range
13213 @node Range Checking
13214 @subsection An Overview of Range Checking
13216 In some languages (such as Modula-2), it is an error to exceed the
13217 bounds of a type; this is enforced with run-time checks. Such range
13218 checking is meant to ensure program correctness by making sure
13219 computations do not overflow, or indices on an array element access do
13220 not exceed the bounds of the array.
13222 For expressions you use in @value{GDBN} commands, you can tell
13223 @value{GDBN} to treat range errors in one of three ways: ignore them,
13224 always treat them as errors and abandon the expression, or issue
13225 warnings but evaluate the expression anyway.
13227 A range error can result from numerical overflow, from exceeding an
13228 array index bound, or when you type a constant that is not a member
13229 of any type. Some languages, however, do not treat overflows as an
13230 error. In many implementations of C, mathematical overflow causes the
13231 result to ``wrap around'' to lower values---for example, if @var{m} is
13232 the largest integer value, and @var{s} is the smallest, then
13235 @var{m} + 1 @result{} @var{s}
13238 This, too, is specific to individual languages, and in some cases
13239 specific to individual compilers or machines. @xref{Supported Languages, ,
13240 Supported Languages}, for further details on specific languages.
13242 @value{GDBN} provides some additional commands for controlling the range checker:
13244 @kindex set check range
13245 @kindex show check range
13247 @item set check range auto
13248 Set range checking on or off based on the current working language.
13249 @xref{Supported Languages, ,Supported Languages}, for the default settings for
13252 @item set check range on
13253 @itemx set check range off
13254 Set range checking on or off, overriding the default setting for the
13255 current working language. A warning is issued if the setting does not
13256 match the language default. If a range error occurs and range checking is on,
13257 then a message is printed and evaluation of the expression is aborted.
13259 @item set check range warn
13260 Output messages when the @value{GDBN} range checker detects a range error,
13261 but attempt to evaluate the expression anyway. Evaluating the
13262 expression may still be impossible for other reasons, such as accessing
13263 memory that the process does not own (a typical example from many Unix
13267 Show the current setting of the range checker, and whether or not it is
13268 being set automatically by @value{GDBN}.
13271 @node Supported Languages
13272 @section Supported Languages
13274 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran, Java,
13275 OpenCL C, Pascal, assembly, Modula-2, and Ada.
13276 @c This is false ...
13277 Some @value{GDBN} features may be used in expressions regardless of the
13278 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
13279 and the @samp{@{type@}addr} construct (@pxref{Expressions,
13280 ,Expressions}) can be used with the constructs of any supported
13283 The following sections detail to what degree each source language is
13284 supported by @value{GDBN}. These sections are not meant to be language
13285 tutorials or references, but serve only as a reference guide to what the
13286 @value{GDBN} expression parser accepts, and what input and output
13287 formats should look like for different languages. There are many good
13288 books written on each of these languages; please look to these for a
13289 language reference or tutorial.
13292 * C:: C and C@t{++}
13295 * Objective-C:: Objective-C
13296 * OpenCL C:: OpenCL C
13297 * Fortran:: Fortran
13299 * Modula-2:: Modula-2
13304 @subsection C and C@t{++}
13306 @cindex C and C@t{++}
13307 @cindex expressions in C or C@t{++}
13309 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
13310 to both languages. Whenever this is the case, we discuss those languages
13314 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
13315 @cindex @sc{gnu} C@t{++}
13316 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
13317 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
13318 effectively, you must compile your C@t{++} programs with a supported
13319 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
13320 compiler (@code{aCC}).
13323 * C Operators:: C and C@t{++} operators
13324 * C Constants:: C and C@t{++} constants
13325 * C Plus Plus Expressions:: C@t{++} expressions
13326 * C Defaults:: Default settings for C and C@t{++}
13327 * C Checks:: C and C@t{++} type and range checks
13328 * Debugging C:: @value{GDBN} and C
13329 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
13330 * Decimal Floating Point:: Numbers in Decimal Floating Point format
13334 @subsubsection C and C@t{++} Operators
13336 @cindex C and C@t{++} operators
13338 Operators must be defined on values of specific types. For instance,
13339 @code{+} is defined on numbers, but not on structures. Operators are
13340 often defined on groups of types.
13342 For the purposes of C and C@t{++}, the following definitions hold:
13347 @emph{Integral types} include @code{int} with any of its storage-class
13348 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
13351 @emph{Floating-point types} include @code{float}, @code{double}, and
13352 @code{long double} (if supported by the target platform).
13355 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
13358 @emph{Scalar types} include all of the above.
13363 The following operators are supported. They are listed here
13364 in order of increasing precedence:
13368 The comma or sequencing operator. Expressions in a comma-separated list
13369 are evaluated from left to right, with the result of the entire
13370 expression being the last expression evaluated.
13373 Assignment. The value of an assignment expression is the value
13374 assigned. Defined on scalar types.
13377 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
13378 and translated to @w{@code{@var{a} = @var{a op b}}}.
13379 @w{@code{@var{op}=}} and @code{=} have the same precedence.
13380 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
13381 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
13384 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
13385 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
13389 Logical @sc{or}. Defined on integral types.
13392 Logical @sc{and}. Defined on integral types.
13395 Bitwise @sc{or}. Defined on integral types.
13398 Bitwise exclusive-@sc{or}. Defined on integral types.
13401 Bitwise @sc{and}. Defined on integral types.
13404 Equality and inequality. Defined on scalar types. The value of these
13405 expressions is 0 for false and non-zero for true.
13407 @item <@r{, }>@r{, }<=@r{, }>=
13408 Less than, greater than, less than or equal, greater than or equal.
13409 Defined on scalar types. The value of these expressions is 0 for false
13410 and non-zero for true.
13413 left shift, and right shift. Defined on integral types.
13416 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
13419 Addition and subtraction. Defined on integral types, floating-point types and
13422 @item *@r{, }/@r{, }%
13423 Multiplication, division, and modulus. Multiplication and division are
13424 defined on integral and floating-point types. Modulus is defined on
13428 Increment and decrement. When appearing before a variable, the
13429 operation is performed before the variable is used in an expression;
13430 when appearing after it, the variable's value is used before the
13431 operation takes place.
13434 Pointer dereferencing. Defined on pointer types. Same precedence as
13438 Address operator. Defined on variables. Same precedence as @code{++}.
13440 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
13441 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
13442 to examine the address
13443 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
13447 Negative. Defined on integral and floating-point types. Same
13448 precedence as @code{++}.
13451 Logical negation. Defined on integral types. Same precedence as
13455 Bitwise complement operator. Defined on integral types. Same precedence as
13460 Structure member, and pointer-to-structure member. For convenience,
13461 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
13462 pointer based on the stored type information.
13463 Defined on @code{struct} and @code{union} data.
13466 Dereferences of pointers to members.
13469 Array indexing. @code{@var{a}[@var{i}]} is defined as
13470 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
13473 Function parameter list. Same precedence as @code{->}.
13476 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
13477 and @code{class} types.
13480 Doubled colons also represent the @value{GDBN} scope operator
13481 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
13485 If an operator is redefined in the user code, @value{GDBN} usually
13486 attempts to invoke the redefined version instead of using the operator's
13487 predefined meaning.
13490 @subsubsection C and C@t{++} Constants
13492 @cindex C and C@t{++} constants
13494 @value{GDBN} allows you to express the constants of C and C@t{++} in the
13499 Integer constants are a sequence of digits. Octal constants are
13500 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
13501 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
13502 @samp{l}, specifying that the constant should be treated as a
13506 Floating point constants are a sequence of digits, followed by a decimal
13507 point, followed by a sequence of digits, and optionally followed by an
13508 exponent. An exponent is of the form:
13509 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
13510 sequence of digits. The @samp{+} is optional for positive exponents.
13511 A floating-point constant may also end with a letter @samp{f} or
13512 @samp{F}, specifying that the constant should be treated as being of
13513 the @code{float} (as opposed to the default @code{double}) type; or with
13514 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
13518 Enumerated constants consist of enumerated identifiers, or their
13519 integral equivalents.
13522 Character constants are a single character surrounded by single quotes
13523 (@code{'}), or a number---the ordinal value of the corresponding character
13524 (usually its @sc{ascii} value). Within quotes, the single character may
13525 be represented by a letter or by @dfn{escape sequences}, which are of
13526 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
13527 of the character's ordinal value; or of the form @samp{\@var{x}}, where
13528 @samp{@var{x}} is a predefined special character---for example,
13529 @samp{\n} for newline.
13531 Wide character constants can be written by prefixing a character
13532 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
13533 form of @samp{x}. The target wide character set is used when
13534 computing the value of this constant (@pxref{Character Sets}).
13537 String constants are a sequence of character constants surrounded by
13538 double quotes (@code{"}). Any valid character constant (as described
13539 above) may appear. Double quotes within the string must be preceded by
13540 a backslash, so for instance @samp{"a\"b'c"} is a string of five
13543 Wide string constants can be written by prefixing a string constant
13544 with @samp{L}, as in C. The target wide character set is used when
13545 computing the value of this constant (@pxref{Character Sets}).
13548 Pointer constants are an integral value. You can also write pointers
13549 to constants using the C operator @samp{&}.
13552 Array constants are comma-separated lists surrounded by braces @samp{@{}
13553 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
13554 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
13555 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
13558 @node C Plus Plus Expressions
13559 @subsubsection C@t{++} Expressions
13561 @cindex expressions in C@t{++}
13562 @value{GDBN} expression handling can interpret most C@t{++} expressions.
13564 @cindex debugging C@t{++} programs
13565 @cindex C@t{++} compilers
13566 @cindex debug formats and C@t{++}
13567 @cindex @value{NGCC} and C@t{++}
13569 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
13570 the proper compiler and the proper debug format. Currently,
13571 @value{GDBN} works best when debugging C@t{++} code that is compiled
13572 with the most recent version of @value{NGCC} possible. The DWARF
13573 debugging format is preferred; @value{NGCC} defaults to this on most
13574 popular platforms. Other compilers and/or debug formats are likely to
13575 work badly or not at all when using @value{GDBN} to debug C@t{++}
13576 code. @xref{Compilation}.
13581 @cindex member functions
13583 Member function calls are allowed; you can use expressions like
13586 count = aml->GetOriginal(x, y)
13589 @vindex this@r{, inside C@t{++} member functions}
13590 @cindex namespace in C@t{++}
13592 While a member function is active (in the selected stack frame), your
13593 expressions have the same namespace available as the member function;
13594 that is, @value{GDBN} allows implicit references to the class instance
13595 pointer @code{this} following the same rules as C@t{++}. @code{using}
13596 declarations in the current scope are also respected by @value{GDBN}.
13598 @cindex call overloaded functions
13599 @cindex overloaded functions, calling
13600 @cindex type conversions in C@t{++}
13602 You can call overloaded functions; @value{GDBN} resolves the function
13603 call to the right definition, with some restrictions. @value{GDBN} does not
13604 perform overload resolution involving user-defined type conversions,
13605 calls to constructors, or instantiations of templates that do not exist
13606 in the program. It also cannot handle ellipsis argument lists or
13609 It does perform integral conversions and promotions, floating-point
13610 promotions, arithmetic conversions, pointer conversions, conversions of
13611 class objects to base classes, and standard conversions such as those of
13612 functions or arrays to pointers; it requires an exact match on the
13613 number of function arguments.
13615 Overload resolution is always performed, unless you have specified
13616 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
13617 ,@value{GDBN} Features for C@t{++}}.
13619 You must specify @code{set overload-resolution off} in order to use an
13620 explicit function signature to call an overloaded function, as in
13622 p 'foo(char,int)'('x', 13)
13625 The @value{GDBN} command-completion facility can simplify this;
13626 see @ref{Completion, ,Command Completion}.
13628 @cindex reference declarations
13630 @value{GDBN} understands variables declared as C@t{++} references; you can use
13631 them in expressions just as you do in C@t{++} source---they are automatically
13634 In the parameter list shown when @value{GDBN} displays a frame, the values of
13635 reference variables are not displayed (unlike other variables); this
13636 avoids clutter, since references are often used for large structures.
13637 The @emph{address} of a reference variable is always shown, unless
13638 you have specified @samp{set print address off}.
13641 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
13642 expressions can use it just as expressions in your program do. Since
13643 one scope may be defined in another, you can use @code{::} repeatedly if
13644 necessary, for example in an expression like
13645 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
13646 resolving name scope by reference to source files, in both C and C@t{++}
13647 debugging (@pxref{Variables, ,Program Variables}).
13650 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
13655 @subsubsection C and C@t{++} Defaults
13657 @cindex C and C@t{++} defaults
13659 If you allow @value{GDBN} to set range checking automatically, it
13660 defaults to @code{off} whenever the working language changes to
13661 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
13662 selects the working language.
13664 If you allow @value{GDBN} to set the language automatically, it
13665 recognizes source files whose names end with @file{.c}, @file{.C}, or
13666 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
13667 these files, it sets the working language to C or C@t{++}.
13668 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
13669 for further details.
13672 @subsubsection C and C@t{++} Type and Range Checks
13674 @cindex C and C@t{++} checks
13676 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
13677 checking is used. However, if you turn type checking off, @value{GDBN}
13678 will allow certain non-standard conversions, such as promoting integer
13679 constants to pointers.
13681 Range checking, if turned on, is done on mathematical operations. Array
13682 indices are not checked, since they are often used to index a pointer
13683 that is not itself an array.
13686 @subsubsection @value{GDBN} and C
13688 The @code{set print union} and @code{show print union} commands apply to
13689 the @code{union} type. When set to @samp{on}, any @code{union} that is
13690 inside a @code{struct} or @code{class} is also printed. Otherwise, it
13691 appears as @samp{@{...@}}.
13693 The @code{@@} operator aids in the debugging of dynamic arrays, formed
13694 with pointers and a memory allocation function. @xref{Expressions,
13697 @node Debugging C Plus Plus
13698 @subsubsection @value{GDBN} Features for C@t{++}
13700 @cindex commands for C@t{++}
13702 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
13703 designed specifically for use with C@t{++}. Here is a summary:
13706 @cindex break in overloaded functions
13707 @item @r{breakpoint menus}
13708 When you want a breakpoint in a function whose name is overloaded,
13709 @value{GDBN} has the capability to display a menu of possible breakpoint
13710 locations to help you specify which function definition you want.
13711 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
13713 @cindex overloading in C@t{++}
13714 @item rbreak @var{regex}
13715 Setting breakpoints using regular expressions is helpful for setting
13716 breakpoints on overloaded functions that are not members of any special
13718 @xref{Set Breaks, ,Setting Breakpoints}.
13720 @cindex C@t{++} exception handling
13722 @itemx catch rethrow
13724 Debug C@t{++} exception handling using these commands. @xref{Set
13725 Catchpoints, , Setting Catchpoints}.
13727 @cindex inheritance
13728 @item ptype @var{typename}
13729 Print inheritance relationships as well as other information for type
13731 @xref{Symbols, ,Examining the Symbol Table}.
13733 @item info vtbl @var{expression}.
13734 The @code{info vtbl} command can be used to display the virtual
13735 method tables of the object computed by @var{expression}. This shows
13736 one entry per virtual table; there may be multiple virtual tables when
13737 multiple inheritance is in use.
13739 @cindex C@t{++} symbol display
13740 @item set print demangle
13741 @itemx show print demangle
13742 @itemx set print asm-demangle
13743 @itemx show print asm-demangle
13744 Control whether C@t{++} symbols display in their source form, both when
13745 displaying code as C@t{++} source and when displaying disassemblies.
13746 @xref{Print Settings, ,Print Settings}.
13748 @item set print object
13749 @itemx show print object
13750 Choose whether to print derived (actual) or declared types of objects.
13751 @xref{Print Settings, ,Print Settings}.
13753 @item set print vtbl
13754 @itemx show print vtbl
13755 Control the format for printing virtual function tables.
13756 @xref{Print Settings, ,Print Settings}.
13757 (The @code{vtbl} commands do not work on programs compiled with the HP
13758 ANSI C@t{++} compiler (@code{aCC}).)
13760 @kindex set overload-resolution
13761 @cindex overloaded functions, overload resolution
13762 @item set overload-resolution on
13763 Enable overload resolution for C@t{++} expression evaluation. The default
13764 is on. For overloaded functions, @value{GDBN} evaluates the arguments
13765 and searches for a function whose signature matches the argument types,
13766 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
13767 Expressions, ,C@t{++} Expressions}, for details).
13768 If it cannot find a match, it emits a message.
13770 @item set overload-resolution off
13771 Disable overload resolution for C@t{++} expression evaluation. For
13772 overloaded functions that are not class member functions, @value{GDBN}
13773 chooses the first function of the specified name that it finds in the
13774 symbol table, whether or not its arguments are of the correct type. For
13775 overloaded functions that are class member functions, @value{GDBN}
13776 searches for a function whose signature @emph{exactly} matches the
13779 @kindex show overload-resolution
13780 @item show overload-resolution
13781 Show the current setting of overload resolution.
13783 @item @r{Overloaded symbol names}
13784 You can specify a particular definition of an overloaded symbol, using
13785 the same notation that is used to declare such symbols in C@t{++}: type
13786 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
13787 also use the @value{GDBN} command-line word completion facilities to list the
13788 available choices, or to finish the type list for you.
13789 @xref{Completion,, Command Completion}, for details on how to do this.
13792 @node Decimal Floating Point
13793 @subsubsection Decimal Floating Point format
13794 @cindex decimal floating point format
13796 @value{GDBN} can examine, set and perform computations with numbers in
13797 decimal floating point format, which in the C language correspond to the
13798 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
13799 specified by the extension to support decimal floating-point arithmetic.
13801 There are two encodings in use, depending on the architecture: BID (Binary
13802 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
13803 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
13806 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
13807 to manipulate decimal floating point numbers, it is not possible to convert
13808 (using a cast, for example) integers wider than 32-bit to decimal float.
13810 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
13811 point computations, error checking in decimal float operations ignores
13812 underflow, overflow and divide by zero exceptions.
13814 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
13815 to inspect @code{_Decimal128} values stored in floating point registers.
13816 See @ref{PowerPC,,PowerPC} for more details.
13822 @value{GDBN} can be used to debug programs written in D and compiled with
13823 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
13824 specific feature --- dynamic arrays.
13829 @cindex Go (programming language)
13830 @value{GDBN} can be used to debug programs written in Go and compiled with
13831 @file{gccgo} or @file{6g} compilers.
13833 Here is a summary of the Go-specific features and restrictions:
13836 @cindex current Go package
13837 @item The current Go package
13838 The name of the current package does not need to be specified when
13839 specifying global variables and functions.
13841 For example, given the program:
13845 var myglob = "Shall we?"
13851 When stopped inside @code{main} either of these work:
13855 (gdb) p main.myglob
13858 @cindex builtin Go types
13859 @item Builtin Go types
13860 The @code{string} type is recognized by @value{GDBN} and is printed
13863 @cindex builtin Go functions
13864 @item Builtin Go functions
13865 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
13866 function and handles it internally.
13868 @cindex restrictions on Go expressions
13869 @item Restrictions on Go expressions
13870 All Go operators are supported except @code{&^}.
13871 The Go @code{_} ``blank identifier'' is not supported.
13872 Automatic dereferencing of pointers is not supported.
13876 @subsection Objective-C
13878 @cindex Objective-C
13879 This section provides information about some commands and command
13880 options that are useful for debugging Objective-C code. See also
13881 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
13882 few more commands specific to Objective-C support.
13885 * Method Names in Commands::
13886 * The Print Command with Objective-C::
13889 @node Method Names in Commands
13890 @subsubsection Method Names in Commands
13892 The following commands have been extended to accept Objective-C method
13893 names as line specifications:
13895 @kindex clear@r{, and Objective-C}
13896 @kindex break@r{, and Objective-C}
13897 @kindex info line@r{, and Objective-C}
13898 @kindex jump@r{, and Objective-C}
13899 @kindex list@r{, and Objective-C}
13903 @item @code{info line}
13908 A fully qualified Objective-C method name is specified as
13911 -[@var{Class} @var{methodName}]
13914 where the minus sign is used to indicate an instance method and a
13915 plus sign (not shown) is used to indicate a class method. The class
13916 name @var{Class} and method name @var{methodName} are enclosed in
13917 brackets, similar to the way messages are specified in Objective-C
13918 source code. For example, to set a breakpoint at the @code{create}
13919 instance method of class @code{Fruit} in the program currently being
13923 break -[Fruit create]
13926 To list ten program lines around the @code{initialize} class method,
13930 list +[NSText initialize]
13933 In the current version of @value{GDBN}, the plus or minus sign is
13934 required. In future versions of @value{GDBN}, the plus or minus
13935 sign will be optional, but you can use it to narrow the search. It
13936 is also possible to specify just a method name:
13942 You must specify the complete method name, including any colons. If
13943 your program's source files contain more than one @code{create} method,
13944 you'll be presented with a numbered list of classes that implement that
13945 method. Indicate your choice by number, or type @samp{0} to exit if
13948 As another example, to clear a breakpoint established at the
13949 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
13952 clear -[NSWindow makeKeyAndOrderFront:]
13955 @node The Print Command with Objective-C
13956 @subsubsection The Print Command With Objective-C
13957 @cindex Objective-C, print objects
13958 @kindex print-object
13959 @kindex po @r{(@code{print-object})}
13961 The print command has also been extended to accept methods. For example:
13964 print -[@var{object} hash]
13967 @cindex print an Objective-C object description
13968 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
13970 will tell @value{GDBN} to send the @code{hash} message to @var{object}
13971 and print the result. Also, an additional command has been added,
13972 @code{print-object} or @code{po} for short, which is meant to print
13973 the description of an object. However, this command may only work
13974 with certain Objective-C libraries that have a particular hook
13975 function, @code{_NSPrintForDebugger}, defined.
13978 @subsection OpenCL C
13981 This section provides information about @value{GDBN}s OpenCL C support.
13984 * OpenCL C Datatypes::
13985 * OpenCL C Expressions::
13986 * OpenCL C Operators::
13989 @node OpenCL C Datatypes
13990 @subsubsection OpenCL C Datatypes
13992 @cindex OpenCL C Datatypes
13993 @value{GDBN} supports the builtin scalar and vector datatypes specified
13994 by OpenCL 1.1. In addition the half- and double-precision floating point
13995 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
13996 extensions are also known to @value{GDBN}.
13998 @node OpenCL C Expressions
13999 @subsubsection OpenCL C Expressions
14001 @cindex OpenCL C Expressions
14002 @value{GDBN} supports accesses to vector components including the access as
14003 lvalue where possible. Since OpenCL C is based on C99 most C expressions
14004 supported by @value{GDBN} can be used as well.
14006 @node OpenCL C Operators
14007 @subsubsection OpenCL C Operators
14009 @cindex OpenCL C Operators
14010 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
14014 @subsection Fortran
14015 @cindex Fortran-specific support in @value{GDBN}
14017 @value{GDBN} can be used to debug programs written in Fortran, but it
14018 currently supports only the features of Fortran 77 language.
14020 @cindex trailing underscore, in Fortran symbols
14021 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
14022 among them) append an underscore to the names of variables and
14023 functions. When you debug programs compiled by those compilers, you
14024 will need to refer to variables and functions with a trailing
14028 * Fortran Operators:: Fortran operators and expressions
14029 * Fortran Defaults:: Default settings for Fortran
14030 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
14033 @node Fortran Operators
14034 @subsubsection Fortran Operators and Expressions
14036 @cindex Fortran operators and expressions
14038 Operators must be defined on values of specific types. For instance,
14039 @code{+} is defined on numbers, but not on characters or other non-
14040 arithmetic types. Operators are often defined on groups of types.
14044 The exponentiation operator. It raises the first operand to the power
14048 The range operator. Normally used in the form of array(low:high) to
14049 represent a section of array.
14052 The access component operator. Normally used to access elements in derived
14053 types. Also suitable for unions. As unions aren't part of regular Fortran,
14054 this can only happen when accessing a register that uses a gdbarch-defined
14058 @node Fortran Defaults
14059 @subsubsection Fortran Defaults
14061 @cindex Fortran Defaults
14063 Fortran symbols are usually case-insensitive, so @value{GDBN} by
14064 default uses case-insensitive matches for Fortran symbols. You can
14065 change that with the @samp{set case-insensitive} command, see
14066 @ref{Symbols}, for the details.
14068 @node Special Fortran Commands
14069 @subsubsection Special Fortran Commands
14071 @cindex Special Fortran commands
14073 @value{GDBN} has some commands to support Fortran-specific features,
14074 such as displaying common blocks.
14077 @cindex @code{COMMON} blocks, Fortran
14078 @kindex info common
14079 @item info common @r{[}@var{common-name}@r{]}
14080 This command prints the values contained in the Fortran @code{COMMON}
14081 block whose name is @var{common-name}. With no argument, the names of
14082 all @code{COMMON} blocks visible at the current program location are
14089 @cindex Pascal support in @value{GDBN}, limitations
14090 Debugging Pascal programs which use sets, subranges, file variables, or
14091 nested functions does not currently work. @value{GDBN} does not support
14092 entering expressions, printing values, or similar features using Pascal
14095 The Pascal-specific command @code{set print pascal_static-members}
14096 controls whether static members of Pascal objects are displayed.
14097 @xref{Print Settings, pascal_static-members}.
14100 @subsection Modula-2
14102 @cindex Modula-2, @value{GDBN} support
14104 The extensions made to @value{GDBN} to support Modula-2 only support
14105 output from the @sc{gnu} Modula-2 compiler (which is currently being
14106 developed). Other Modula-2 compilers are not currently supported, and
14107 attempting to debug executables produced by them is most likely
14108 to give an error as @value{GDBN} reads in the executable's symbol
14111 @cindex expressions in Modula-2
14113 * M2 Operators:: Built-in operators
14114 * Built-In Func/Proc:: Built-in functions and procedures
14115 * M2 Constants:: Modula-2 constants
14116 * M2 Types:: Modula-2 types
14117 * M2 Defaults:: Default settings for Modula-2
14118 * Deviations:: Deviations from standard Modula-2
14119 * M2 Checks:: Modula-2 type and range checks
14120 * M2 Scope:: The scope operators @code{::} and @code{.}
14121 * GDB/M2:: @value{GDBN} and Modula-2
14125 @subsubsection Operators
14126 @cindex Modula-2 operators
14128 Operators must be defined on values of specific types. For instance,
14129 @code{+} is defined on numbers, but not on structures. Operators are
14130 often defined on groups of types. For the purposes of Modula-2, the
14131 following definitions hold:
14136 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
14140 @emph{Character types} consist of @code{CHAR} and its subranges.
14143 @emph{Floating-point types} consist of @code{REAL}.
14146 @emph{Pointer types} consist of anything declared as @code{POINTER TO
14150 @emph{Scalar types} consist of all of the above.
14153 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
14156 @emph{Boolean types} consist of @code{BOOLEAN}.
14160 The following operators are supported, and appear in order of
14161 increasing precedence:
14165 Function argument or array index separator.
14168 Assignment. The value of @var{var} @code{:=} @var{value} is
14172 Less than, greater than on integral, floating-point, or enumerated
14176 Less than or equal to, greater than or equal to
14177 on integral, floating-point and enumerated types, or set inclusion on
14178 set types. Same precedence as @code{<}.
14180 @item =@r{, }<>@r{, }#
14181 Equality and two ways of expressing inequality, valid on scalar types.
14182 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
14183 available for inequality, since @code{#} conflicts with the script
14187 Set membership. Defined on set types and the types of their members.
14188 Same precedence as @code{<}.
14191 Boolean disjunction. Defined on boolean types.
14194 Boolean conjunction. Defined on boolean types.
14197 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
14200 Addition and subtraction on integral and floating-point types, or union
14201 and difference on set types.
14204 Multiplication on integral and floating-point types, or set intersection
14208 Division on floating-point types, or symmetric set difference on set
14209 types. Same precedence as @code{*}.
14212 Integer division and remainder. Defined on integral types. Same
14213 precedence as @code{*}.
14216 Negative. Defined on @code{INTEGER} and @code{REAL} data.
14219 Pointer dereferencing. Defined on pointer types.
14222 Boolean negation. Defined on boolean types. Same precedence as
14226 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
14227 precedence as @code{^}.
14230 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
14233 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
14237 @value{GDBN} and Modula-2 scope operators.
14241 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
14242 treats the use of the operator @code{IN}, or the use of operators
14243 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
14244 @code{<=}, and @code{>=} on sets as an error.
14248 @node Built-In Func/Proc
14249 @subsubsection Built-in Functions and Procedures
14250 @cindex Modula-2 built-ins
14252 Modula-2 also makes available several built-in procedures and functions.
14253 In describing these, the following metavariables are used:
14258 represents an @code{ARRAY} variable.
14261 represents a @code{CHAR} constant or variable.
14264 represents a variable or constant of integral type.
14267 represents an identifier that belongs to a set. Generally used in the
14268 same function with the metavariable @var{s}. The type of @var{s} should
14269 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
14272 represents a variable or constant of integral or floating-point type.
14275 represents a variable or constant of floating-point type.
14281 represents a variable.
14284 represents a variable or constant of one of many types. See the
14285 explanation of the function for details.
14288 All Modula-2 built-in procedures also return a result, described below.
14292 Returns the absolute value of @var{n}.
14295 If @var{c} is a lower case letter, it returns its upper case
14296 equivalent, otherwise it returns its argument.
14299 Returns the character whose ordinal value is @var{i}.
14302 Decrements the value in the variable @var{v} by one. Returns the new value.
14304 @item DEC(@var{v},@var{i})
14305 Decrements the value in the variable @var{v} by @var{i}. Returns the
14308 @item EXCL(@var{m},@var{s})
14309 Removes the element @var{m} from the set @var{s}. Returns the new
14312 @item FLOAT(@var{i})
14313 Returns the floating point equivalent of the integer @var{i}.
14315 @item HIGH(@var{a})
14316 Returns the index of the last member of @var{a}.
14319 Increments the value in the variable @var{v} by one. Returns the new value.
14321 @item INC(@var{v},@var{i})
14322 Increments the value in the variable @var{v} by @var{i}. Returns the
14325 @item INCL(@var{m},@var{s})
14326 Adds the element @var{m} to the set @var{s} if it is not already
14327 there. Returns the new set.
14330 Returns the maximum value of the type @var{t}.
14333 Returns the minimum value of the type @var{t}.
14336 Returns boolean TRUE if @var{i} is an odd number.
14339 Returns the ordinal value of its argument. For example, the ordinal
14340 value of a character is its @sc{ascii} value (on machines supporting the
14341 @sc{ascii} character set). @var{x} must be of an ordered type, which include
14342 integral, character and enumerated types.
14344 @item SIZE(@var{x})
14345 Returns the size of its argument. @var{x} can be a variable or a type.
14347 @item TRUNC(@var{r})
14348 Returns the integral part of @var{r}.
14350 @item TSIZE(@var{x})
14351 Returns the size of its argument. @var{x} can be a variable or a type.
14353 @item VAL(@var{t},@var{i})
14354 Returns the member of the type @var{t} whose ordinal value is @var{i}.
14358 @emph{Warning:} Sets and their operations are not yet supported, so
14359 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
14363 @cindex Modula-2 constants
14365 @subsubsection Constants
14367 @value{GDBN} allows you to express the constants of Modula-2 in the following
14373 Integer constants are simply a sequence of digits. When used in an
14374 expression, a constant is interpreted to be type-compatible with the
14375 rest of the expression. Hexadecimal integers are specified by a
14376 trailing @samp{H}, and octal integers by a trailing @samp{B}.
14379 Floating point constants appear as a sequence of digits, followed by a
14380 decimal point and another sequence of digits. An optional exponent can
14381 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
14382 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
14383 digits of the floating point constant must be valid decimal (base 10)
14387 Character constants consist of a single character enclosed by a pair of
14388 like quotes, either single (@code{'}) or double (@code{"}). They may
14389 also be expressed by their ordinal value (their @sc{ascii} value, usually)
14390 followed by a @samp{C}.
14393 String constants consist of a sequence of characters enclosed by a
14394 pair of like quotes, either single (@code{'}) or double (@code{"}).
14395 Escape sequences in the style of C are also allowed. @xref{C
14396 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
14400 Enumerated constants consist of an enumerated identifier.
14403 Boolean constants consist of the identifiers @code{TRUE} and
14407 Pointer constants consist of integral values only.
14410 Set constants are not yet supported.
14414 @subsubsection Modula-2 Types
14415 @cindex Modula-2 types
14417 Currently @value{GDBN} can print the following data types in Modula-2
14418 syntax: array types, record types, set types, pointer types, procedure
14419 types, enumerated types, subrange types and base types. You can also
14420 print the contents of variables declared using these type.
14421 This section gives a number of simple source code examples together with
14422 sample @value{GDBN} sessions.
14424 The first example contains the following section of code:
14433 and you can request @value{GDBN} to interrogate the type and value of
14434 @code{r} and @code{s}.
14437 (@value{GDBP}) print s
14439 (@value{GDBP}) ptype s
14441 (@value{GDBP}) print r
14443 (@value{GDBP}) ptype r
14448 Likewise if your source code declares @code{s} as:
14452 s: SET ['A'..'Z'] ;
14456 then you may query the type of @code{s} by:
14459 (@value{GDBP}) ptype s
14460 type = SET ['A'..'Z']
14464 Note that at present you cannot interactively manipulate set
14465 expressions using the debugger.
14467 The following example shows how you might declare an array in Modula-2
14468 and how you can interact with @value{GDBN} to print its type and contents:
14472 s: ARRAY [-10..10] OF CHAR ;
14476 (@value{GDBP}) ptype s
14477 ARRAY [-10..10] OF CHAR
14480 Note that the array handling is not yet complete and although the type
14481 is printed correctly, expression handling still assumes that all
14482 arrays have a lower bound of zero and not @code{-10} as in the example
14485 Here are some more type related Modula-2 examples:
14489 colour = (blue, red, yellow, green) ;
14490 t = [blue..yellow] ;
14498 The @value{GDBN} interaction shows how you can query the data type
14499 and value of a variable.
14502 (@value{GDBP}) print s
14504 (@value{GDBP}) ptype t
14505 type = [blue..yellow]
14509 In this example a Modula-2 array is declared and its contents
14510 displayed. Observe that the contents are written in the same way as
14511 their @code{C} counterparts.
14515 s: ARRAY [1..5] OF CARDINAL ;
14521 (@value{GDBP}) print s
14522 $1 = @{1, 0, 0, 0, 0@}
14523 (@value{GDBP}) ptype s
14524 type = ARRAY [1..5] OF CARDINAL
14527 The Modula-2 language interface to @value{GDBN} also understands
14528 pointer types as shown in this example:
14532 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
14539 and you can request that @value{GDBN} describes the type of @code{s}.
14542 (@value{GDBP}) ptype s
14543 type = POINTER TO ARRAY [1..5] OF CARDINAL
14546 @value{GDBN} handles compound types as we can see in this example.
14547 Here we combine array types, record types, pointer types and subrange
14558 myarray = ARRAY myrange OF CARDINAL ;
14559 myrange = [-2..2] ;
14561 s: POINTER TO ARRAY myrange OF foo ;
14565 and you can ask @value{GDBN} to describe the type of @code{s} as shown
14569 (@value{GDBP}) ptype s
14570 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
14573 f3 : ARRAY [-2..2] OF CARDINAL;
14578 @subsubsection Modula-2 Defaults
14579 @cindex Modula-2 defaults
14581 If type and range checking are set automatically by @value{GDBN}, they
14582 both default to @code{on} whenever the working language changes to
14583 Modula-2. This happens regardless of whether you or @value{GDBN}
14584 selected the working language.
14586 If you allow @value{GDBN} to set the language automatically, then entering
14587 code compiled from a file whose name ends with @file{.mod} sets the
14588 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
14589 Infer the Source Language}, for further details.
14592 @subsubsection Deviations from Standard Modula-2
14593 @cindex Modula-2, deviations from
14595 A few changes have been made to make Modula-2 programs easier to debug.
14596 This is done primarily via loosening its type strictness:
14600 Unlike in standard Modula-2, pointer constants can be formed by
14601 integers. This allows you to modify pointer variables during
14602 debugging. (In standard Modula-2, the actual address contained in a
14603 pointer variable is hidden from you; it can only be modified
14604 through direct assignment to another pointer variable or expression that
14605 returned a pointer.)
14608 C escape sequences can be used in strings and characters to represent
14609 non-printable characters. @value{GDBN} prints out strings with these
14610 escape sequences embedded. Single non-printable characters are
14611 printed using the @samp{CHR(@var{nnn})} format.
14614 The assignment operator (@code{:=}) returns the value of its right-hand
14618 All built-in procedures both modify @emph{and} return their argument.
14622 @subsubsection Modula-2 Type and Range Checks
14623 @cindex Modula-2 checks
14626 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
14629 @c FIXME remove warning when type/range checks added
14631 @value{GDBN} considers two Modula-2 variables type equivalent if:
14635 They are of types that have been declared equivalent via a @code{TYPE
14636 @var{t1} = @var{t2}} statement
14639 They have been declared on the same line. (Note: This is true of the
14640 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
14643 As long as type checking is enabled, any attempt to combine variables
14644 whose types are not equivalent is an error.
14646 Range checking is done on all mathematical operations, assignment, array
14647 index bounds, and all built-in functions and procedures.
14650 @subsubsection The Scope Operators @code{::} and @code{.}
14652 @cindex @code{.}, Modula-2 scope operator
14653 @cindex colon, doubled as scope operator
14655 @vindex colon-colon@r{, in Modula-2}
14656 @c Info cannot handle :: but TeX can.
14659 @vindex ::@r{, in Modula-2}
14662 There are a few subtle differences between the Modula-2 scope operator
14663 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
14668 @var{module} . @var{id}
14669 @var{scope} :: @var{id}
14673 where @var{scope} is the name of a module or a procedure,
14674 @var{module} the name of a module, and @var{id} is any declared
14675 identifier within your program, except another module.
14677 Using the @code{::} operator makes @value{GDBN} search the scope
14678 specified by @var{scope} for the identifier @var{id}. If it is not
14679 found in the specified scope, then @value{GDBN} searches all scopes
14680 enclosing the one specified by @var{scope}.
14682 Using the @code{.} operator makes @value{GDBN} search the current scope for
14683 the identifier specified by @var{id} that was imported from the
14684 definition module specified by @var{module}. With this operator, it is
14685 an error if the identifier @var{id} was not imported from definition
14686 module @var{module}, or if @var{id} is not an identifier in
14690 @subsubsection @value{GDBN} and Modula-2
14692 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
14693 Five subcommands of @code{set print} and @code{show print} apply
14694 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
14695 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
14696 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
14697 analogue in Modula-2.
14699 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
14700 with any language, is not useful with Modula-2. Its
14701 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
14702 created in Modula-2 as they can in C or C@t{++}. However, because an
14703 address can be specified by an integral constant, the construct
14704 @samp{@{@var{type}@}@var{adrexp}} is still useful.
14706 @cindex @code{#} in Modula-2
14707 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
14708 interpreted as the beginning of a comment. Use @code{<>} instead.
14714 The extensions made to @value{GDBN} for Ada only support
14715 output from the @sc{gnu} Ada (GNAT) compiler.
14716 Other Ada compilers are not currently supported, and
14717 attempting to debug executables produced by them is most likely
14721 @cindex expressions in Ada
14723 * Ada Mode Intro:: General remarks on the Ada syntax
14724 and semantics supported by Ada mode
14726 * Omissions from Ada:: Restrictions on the Ada expression syntax.
14727 * Additions to Ada:: Extensions of the Ada expression syntax.
14728 * Stopping Before Main Program:: Debugging the program during elaboration.
14729 * Ada Tasks:: Listing and setting breakpoints in tasks.
14730 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
14731 * Ravenscar Profile:: Tasking Support when using the Ravenscar
14733 * Ada Glitches:: Known peculiarities of Ada mode.
14736 @node Ada Mode Intro
14737 @subsubsection Introduction
14738 @cindex Ada mode, general
14740 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
14741 syntax, with some extensions.
14742 The philosophy behind the design of this subset is
14746 That @value{GDBN} should provide basic literals and access to operations for
14747 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
14748 leaving more sophisticated computations to subprograms written into the
14749 program (which therefore may be called from @value{GDBN}).
14752 That type safety and strict adherence to Ada language restrictions
14753 are not particularly important to the @value{GDBN} user.
14756 That brevity is important to the @value{GDBN} user.
14759 Thus, for brevity, the debugger acts as if all names declared in
14760 user-written packages are directly visible, even if they are not visible
14761 according to Ada rules, thus making it unnecessary to fully qualify most
14762 names with their packages, regardless of context. Where this causes
14763 ambiguity, @value{GDBN} asks the user's intent.
14765 The debugger will start in Ada mode if it detects an Ada main program.
14766 As for other languages, it will enter Ada mode when stopped in a program that
14767 was translated from an Ada source file.
14769 While in Ada mode, you may use `@t{--}' for comments. This is useful
14770 mostly for documenting command files. The standard @value{GDBN} comment
14771 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
14772 middle (to allow based literals).
14774 The debugger supports limited overloading. Given a subprogram call in which
14775 the function symbol has multiple definitions, it will use the number of
14776 actual parameters and some information about their types to attempt to narrow
14777 the set of definitions. It also makes very limited use of context, preferring
14778 procedures to functions in the context of the @code{call} command, and
14779 functions to procedures elsewhere.
14781 @node Omissions from Ada
14782 @subsubsection Omissions from Ada
14783 @cindex Ada, omissions from
14785 Here are the notable omissions from the subset:
14789 Only a subset of the attributes are supported:
14793 @t{'First}, @t{'Last}, and @t{'Length}
14794 on array objects (not on types and subtypes).
14797 @t{'Min} and @t{'Max}.
14800 @t{'Pos} and @t{'Val}.
14806 @t{'Range} on array objects (not subtypes), but only as the right
14807 operand of the membership (@code{in}) operator.
14810 @t{'Access}, @t{'Unchecked_Access}, and
14811 @t{'Unrestricted_Access} (a GNAT extension).
14819 @code{Characters.Latin_1} are not available and
14820 concatenation is not implemented. Thus, escape characters in strings are
14821 not currently available.
14824 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
14825 equality of representations. They will generally work correctly
14826 for strings and arrays whose elements have integer or enumeration types.
14827 They may not work correctly for arrays whose element
14828 types have user-defined equality, for arrays of real values
14829 (in particular, IEEE-conformant floating point, because of negative
14830 zeroes and NaNs), and for arrays whose elements contain unused bits with
14831 indeterminate values.
14834 The other component-by-component array operations (@code{and}, @code{or},
14835 @code{xor}, @code{not}, and relational tests other than equality)
14836 are not implemented.
14839 @cindex array aggregates (Ada)
14840 @cindex record aggregates (Ada)
14841 @cindex aggregates (Ada)
14842 There is limited support for array and record aggregates. They are
14843 permitted only on the right sides of assignments, as in these examples:
14846 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
14847 (@value{GDBP}) set An_Array := (1, others => 0)
14848 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
14849 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
14850 (@value{GDBP}) set A_Record := (1, "Peter", True);
14851 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
14855 discriminant's value by assigning an aggregate has an
14856 undefined effect if that discriminant is used within the record.
14857 However, you can first modify discriminants by directly assigning to
14858 them (which normally would not be allowed in Ada), and then performing an
14859 aggregate assignment. For example, given a variable @code{A_Rec}
14860 declared to have a type such as:
14863 type Rec (Len : Small_Integer := 0) is record
14865 Vals : IntArray (1 .. Len);
14869 you can assign a value with a different size of @code{Vals} with two
14873 (@value{GDBP}) set A_Rec.Len := 4
14874 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
14877 As this example also illustrates, @value{GDBN} is very loose about the usual
14878 rules concerning aggregates. You may leave out some of the
14879 components of an array or record aggregate (such as the @code{Len}
14880 component in the assignment to @code{A_Rec} above); they will retain their
14881 original values upon assignment. You may freely use dynamic values as
14882 indices in component associations. You may even use overlapping or
14883 redundant component associations, although which component values are
14884 assigned in such cases is not defined.
14887 Calls to dispatching subprograms are not implemented.
14890 The overloading algorithm is much more limited (i.e., less selective)
14891 than that of real Ada. It makes only limited use of the context in
14892 which a subexpression appears to resolve its meaning, and it is much
14893 looser in its rules for allowing type matches. As a result, some
14894 function calls will be ambiguous, and the user will be asked to choose
14895 the proper resolution.
14898 The @code{new} operator is not implemented.
14901 Entry calls are not implemented.
14904 Aside from printing, arithmetic operations on the native VAX floating-point
14905 formats are not supported.
14908 It is not possible to slice a packed array.
14911 The names @code{True} and @code{False}, when not part of a qualified name,
14912 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
14914 Should your program
14915 redefine these names in a package or procedure (at best a dubious practice),
14916 you will have to use fully qualified names to access their new definitions.
14919 @node Additions to Ada
14920 @subsubsection Additions to Ada
14921 @cindex Ada, deviations from
14923 As it does for other languages, @value{GDBN} makes certain generic
14924 extensions to Ada (@pxref{Expressions}):
14928 If the expression @var{E} is a variable residing in memory (typically
14929 a local variable or array element) and @var{N} is a positive integer,
14930 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
14931 @var{N}-1 adjacent variables following it in memory as an array. In
14932 Ada, this operator is generally not necessary, since its prime use is
14933 in displaying parts of an array, and slicing will usually do this in
14934 Ada. However, there are occasional uses when debugging programs in
14935 which certain debugging information has been optimized away.
14938 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
14939 appears in function or file @var{B}.'' When @var{B} is a file name,
14940 you must typically surround it in single quotes.
14943 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
14944 @var{type} that appears at address @var{addr}.''
14947 A name starting with @samp{$} is a convenience variable
14948 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
14951 In addition, @value{GDBN} provides a few other shortcuts and outright
14952 additions specific to Ada:
14956 The assignment statement is allowed as an expression, returning
14957 its right-hand operand as its value. Thus, you may enter
14960 (@value{GDBP}) set x := y + 3
14961 (@value{GDBP}) print A(tmp := y + 1)
14965 The semicolon is allowed as an ``operator,'' returning as its value
14966 the value of its right-hand operand.
14967 This allows, for example,
14968 complex conditional breaks:
14971 (@value{GDBP}) break f
14972 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
14976 Rather than use catenation and symbolic character names to introduce special
14977 characters into strings, one may instead use a special bracket notation,
14978 which is also used to print strings. A sequence of characters of the form
14979 @samp{["@var{XX}"]} within a string or character literal denotes the
14980 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
14981 sequence of characters @samp{["""]} also denotes a single quotation mark
14982 in strings. For example,
14984 "One line.["0a"]Next line.["0a"]"
14987 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
14991 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
14992 @t{'Max} is optional (and is ignored in any case). For example, it is valid
14996 (@value{GDBP}) print 'max(x, y)
15000 When printing arrays, @value{GDBN} uses positional notation when the
15001 array has a lower bound of 1, and uses a modified named notation otherwise.
15002 For example, a one-dimensional array of three integers with a lower bound
15003 of 3 might print as
15010 That is, in contrast to valid Ada, only the first component has a @code{=>}
15014 You may abbreviate attributes in expressions with any unique,
15015 multi-character subsequence of
15016 their names (an exact match gets preference).
15017 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
15018 in place of @t{a'length}.
15021 @cindex quoting Ada internal identifiers
15022 Since Ada is case-insensitive, the debugger normally maps identifiers you type
15023 to lower case. The GNAT compiler uses upper-case characters for
15024 some of its internal identifiers, which are normally of no interest to users.
15025 For the rare occasions when you actually have to look at them,
15026 enclose them in angle brackets to avoid the lower-case mapping.
15029 (@value{GDBP}) print <JMPBUF_SAVE>[0]
15033 Printing an object of class-wide type or dereferencing an
15034 access-to-class-wide value will display all the components of the object's
15035 specific type (as indicated by its run-time tag). Likewise, component
15036 selection on such a value will operate on the specific type of the
15041 @node Stopping Before Main Program
15042 @subsubsection Stopping at the Very Beginning
15044 @cindex breakpointing Ada elaboration code
15045 It is sometimes necessary to debug the program during elaboration, and
15046 before reaching the main procedure.
15047 As defined in the Ada Reference
15048 Manual, the elaboration code is invoked from a procedure called
15049 @code{adainit}. To run your program up to the beginning of
15050 elaboration, simply use the following two commands:
15051 @code{tbreak adainit} and @code{run}.
15054 @subsubsection Extensions for Ada Tasks
15055 @cindex Ada, tasking
15057 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
15058 @value{GDBN} provides the following task-related commands:
15063 This command shows a list of current Ada tasks, as in the following example:
15070 (@value{GDBP}) info tasks
15071 ID TID P-ID Pri State Name
15072 1 8088000 0 15 Child Activation Wait main_task
15073 2 80a4000 1 15 Accept Statement b
15074 3 809a800 1 15 Child Activation Wait a
15075 * 4 80ae800 3 15 Runnable c
15080 In this listing, the asterisk before the last task indicates it to be the
15081 task currently being inspected.
15085 Represents @value{GDBN}'s internal task number.
15091 The parent's task ID (@value{GDBN}'s internal task number).
15094 The base priority of the task.
15097 Current state of the task.
15101 The task has been created but has not been activated. It cannot be
15105 The task is not blocked for any reason known to Ada. (It may be waiting
15106 for a mutex, though.) It is conceptually "executing" in normal mode.
15109 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
15110 that were waiting on terminate alternatives have been awakened and have
15111 terminated themselves.
15113 @item Child Activation Wait
15114 The task is waiting for created tasks to complete activation.
15116 @item Accept Statement
15117 The task is waiting on an accept or selective wait statement.
15119 @item Waiting on entry call
15120 The task is waiting on an entry call.
15122 @item Async Select Wait
15123 The task is waiting to start the abortable part of an asynchronous
15127 The task is waiting on a select statement with only a delay
15130 @item Child Termination Wait
15131 The task is sleeping having completed a master within itself, and is
15132 waiting for the tasks dependent on that master to become terminated or
15133 waiting on a terminate Phase.
15135 @item Wait Child in Term Alt
15136 The task is sleeping waiting for tasks on terminate alternatives to
15137 finish terminating.
15139 @item Accepting RV with @var{taskno}
15140 The task is accepting a rendez-vous with the task @var{taskno}.
15144 Name of the task in the program.
15148 @kindex info task @var{taskno}
15149 @item info task @var{taskno}
15150 This command shows detailled informations on the specified task, as in
15151 the following example:
15156 (@value{GDBP}) info tasks
15157 ID TID P-ID Pri State Name
15158 1 8077880 0 15 Child Activation Wait main_task
15159 * 2 807c468 1 15 Runnable task_1
15160 (@value{GDBP}) info task 2
15161 Ada Task: 0x807c468
15164 Parent: 1 (main_task)
15170 @kindex task@r{ (Ada)}
15171 @cindex current Ada task ID
15172 This command prints the ID of the current task.
15178 (@value{GDBP}) info tasks
15179 ID TID P-ID Pri State Name
15180 1 8077870 0 15 Child Activation Wait main_task
15181 * 2 807c458 1 15 Runnable t
15182 (@value{GDBP}) task
15183 [Current task is 2]
15186 @item task @var{taskno}
15187 @cindex Ada task switching
15188 This command is like the @code{thread @var{threadno}}
15189 command (@pxref{Threads}). It switches the context of debugging
15190 from the current task to the given task.
15196 (@value{GDBP}) info tasks
15197 ID TID P-ID Pri State Name
15198 1 8077870 0 15 Child Activation Wait main_task
15199 * 2 807c458 1 15 Runnable t
15200 (@value{GDBP}) task 1
15201 [Switching to task 1]
15202 #0 0x8067726 in pthread_cond_wait ()
15204 #0 0x8067726 in pthread_cond_wait ()
15205 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
15206 #2 0x805cb63 in system.task_primitives.operations.sleep ()
15207 #3 0x806153e in system.tasking.stages.activate_tasks ()
15208 #4 0x804aacc in un () at un.adb:5
15211 @item break @var{linespec} task @var{taskno}
15212 @itemx break @var{linespec} task @var{taskno} if @dots{}
15213 @cindex breakpoints and tasks, in Ada
15214 @cindex task breakpoints, in Ada
15215 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
15216 These commands are like the @code{break @dots{} thread @dots{}}
15217 command (@pxref{Thread Stops}).
15218 @var{linespec} specifies source lines, as described
15219 in @ref{Specify Location}.
15221 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
15222 to specify that you only want @value{GDBN} to stop the program when a
15223 particular Ada task reaches this breakpoint. @var{taskno} is one of the
15224 numeric task identifiers assigned by @value{GDBN}, shown in the first
15225 column of the @samp{info tasks} display.
15227 If you do not specify @samp{task @var{taskno}} when you set a
15228 breakpoint, the breakpoint applies to @emph{all} tasks of your
15231 You can use the @code{task} qualifier on conditional breakpoints as
15232 well; in this case, place @samp{task @var{taskno}} before the
15233 breakpoint condition (before the @code{if}).
15241 (@value{GDBP}) info tasks
15242 ID TID P-ID Pri State Name
15243 1 140022020 0 15 Child Activation Wait main_task
15244 2 140045060 1 15 Accept/Select Wait t2
15245 3 140044840 1 15 Runnable t1
15246 * 4 140056040 1 15 Runnable t3
15247 (@value{GDBP}) b 15 task 2
15248 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
15249 (@value{GDBP}) cont
15254 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
15256 (@value{GDBP}) info tasks
15257 ID TID P-ID Pri State Name
15258 1 140022020 0 15 Child Activation Wait main_task
15259 * 2 140045060 1 15 Runnable t2
15260 3 140044840 1 15 Runnable t1
15261 4 140056040 1 15 Delay Sleep t3
15265 @node Ada Tasks and Core Files
15266 @subsubsection Tasking Support when Debugging Core Files
15267 @cindex Ada tasking and core file debugging
15269 When inspecting a core file, as opposed to debugging a live program,
15270 tasking support may be limited or even unavailable, depending on
15271 the platform being used.
15272 For instance, on x86-linux, the list of tasks is available, but task
15273 switching is not supported. On Tru64, however, task switching will work
15276 On certain platforms, including Tru64, the debugger needs to perform some
15277 memory writes in order to provide Ada tasking support. When inspecting
15278 a core file, this means that the core file must be opened with read-write
15279 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
15280 Under these circumstances, you should make a backup copy of the core
15281 file before inspecting it with @value{GDBN}.
15283 @node Ravenscar Profile
15284 @subsubsection Tasking Support when using the Ravenscar Profile
15285 @cindex Ravenscar Profile
15287 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
15288 specifically designed for systems with safety-critical real-time
15292 @kindex set ravenscar task-switching on
15293 @cindex task switching with program using Ravenscar Profile
15294 @item set ravenscar task-switching on
15295 Allows task switching when debugging a program that uses the Ravenscar
15296 Profile. This is the default.
15298 @kindex set ravenscar task-switching off
15299 @item set ravenscar task-switching off
15300 Turn off task switching when debugging a program that uses the Ravenscar
15301 Profile. This is mostly intended to disable the code that adds support
15302 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
15303 the Ravenscar runtime is preventing @value{GDBN} from working properly.
15304 To be effective, this command should be run before the program is started.
15306 @kindex show ravenscar task-switching
15307 @item show ravenscar task-switching
15308 Show whether it is possible to switch from task to task in a program
15309 using the Ravenscar Profile.
15314 @subsubsection Known Peculiarities of Ada Mode
15315 @cindex Ada, problems
15317 Besides the omissions listed previously (@pxref{Omissions from Ada}),
15318 we know of several problems with and limitations of Ada mode in
15320 some of which will be fixed with planned future releases of the debugger
15321 and the GNU Ada compiler.
15325 Static constants that the compiler chooses not to materialize as objects in
15326 storage are invisible to the debugger.
15329 Named parameter associations in function argument lists are ignored (the
15330 argument lists are treated as positional).
15333 Many useful library packages are currently invisible to the debugger.
15336 Fixed-point arithmetic, conversions, input, and output is carried out using
15337 floating-point arithmetic, and may give results that only approximate those on
15341 The GNAT compiler never generates the prefix @code{Standard} for any of
15342 the standard symbols defined by the Ada language. @value{GDBN} knows about
15343 this: it will strip the prefix from names when you use it, and will never
15344 look for a name you have so qualified among local symbols, nor match against
15345 symbols in other packages or subprograms. If you have
15346 defined entities anywhere in your program other than parameters and
15347 local variables whose simple names match names in @code{Standard},
15348 GNAT's lack of qualification here can cause confusion. When this happens,
15349 you can usually resolve the confusion
15350 by qualifying the problematic names with package
15351 @code{Standard} explicitly.
15354 Older versions of the compiler sometimes generate erroneous debugging
15355 information, resulting in the debugger incorrectly printing the value
15356 of affected entities. In some cases, the debugger is able to work
15357 around an issue automatically. In other cases, the debugger is able
15358 to work around the issue, but the work-around has to be specifically
15361 @kindex set ada trust-PAD-over-XVS
15362 @kindex show ada trust-PAD-over-XVS
15365 @item set ada trust-PAD-over-XVS on
15366 Configure GDB to strictly follow the GNAT encoding when computing the
15367 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
15368 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
15369 a complete description of the encoding used by the GNAT compiler).
15370 This is the default.
15372 @item set ada trust-PAD-over-XVS off
15373 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
15374 sometimes prints the wrong value for certain entities, changing @code{ada
15375 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
15376 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
15377 @code{off}, but this incurs a slight performance penalty, so it is
15378 recommended to leave this setting to @code{on} unless necessary.
15382 @node Unsupported Languages
15383 @section Unsupported Languages
15385 @cindex unsupported languages
15386 @cindex minimal language
15387 In addition to the other fully-supported programming languages,
15388 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
15389 It does not represent a real programming language, but provides a set
15390 of capabilities close to what the C or assembly languages provide.
15391 This should allow most simple operations to be performed while debugging
15392 an application that uses a language currently not supported by @value{GDBN}.
15394 If the language is set to @code{auto}, @value{GDBN} will automatically
15395 select this language if the current frame corresponds to an unsupported
15399 @chapter Examining the Symbol Table
15401 The commands described in this chapter allow you to inquire about the
15402 symbols (names of variables, functions and types) defined in your
15403 program. This information is inherent in the text of your program and
15404 does not change as your program executes. @value{GDBN} finds it in your
15405 program's symbol table, in the file indicated when you started @value{GDBN}
15406 (@pxref{File Options, ,Choosing Files}), or by one of the
15407 file-management commands (@pxref{Files, ,Commands to Specify Files}).
15409 @cindex symbol names
15410 @cindex names of symbols
15411 @cindex quoting names
15412 Occasionally, you may need to refer to symbols that contain unusual
15413 characters, which @value{GDBN} ordinarily treats as word delimiters. The
15414 most frequent case is in referring to static variables in other
15415 source files (@pxref{Variables,,Program Variables}). File names
15416 are recorded in object files as debugging symbols, but @value{GDBN} would
15417 ordinarily parse a typical file name, like @file{foo.c}, as the three words
15418 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
15419 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
15426 looks up the value of @code{x} in the scope of the file @file{foo.c}.
15429 @cindex case-insensitive symbol names
15430 @cindex case sensitivity in symbol names
15431 @kindex set case-sensitive
15432 @item set case-sensitive on
15433 @itemx set case-sensitive off
15434 @itemx set case-sensitive auto
15435 Normally, when @value{GDBN} looks up symbols, it matches their names
15436 with case sensitivity determined by the current source language.
15437 Occasionally, you may wish to control that. The command @code{set
15438 case-sensitive} lets you do that by specifying @code{on} for
15439 case-sensitive matches or @code{off} for case-insensitive ones. If
15440 you specify @code{auto}, case sensitivity is reset to the default
15441 suitable for the source language. The default is case-sensitive
15442 matches for all languages except for Fortran, for which the default is
15443 case-insensitive matches.
15445 @kindex show case-sensitive
15446 @item show case-sensitive
15447 This command shows the current setting of case sensitivity for symbols
15450 @kindex set print type methods
15451 @item set print type methods
15452 @itemx set print type methods on
15453 @itemx set print type methods off
15454 Normally, when @value{GDBN} prints a class, it displays any methods
15455 declared in that class. You can control this behavior either by
15456 passing the appropriate flag to @code{ptype}, or using @command{set
15457 print type methods}. Specifying @code{on} will cause @value{GDBN} to
15458 display the methods; this is the default. Specifying @code{off} will
15459 cause @value{GDBN} to omit the methods.
15461 @kindex show print type methods
15462 @item show print type methods
15463 This command shows the current setting of method display when printing
15466 @kindex set print type typedefs
15467 @item set print type typedefs
15468 @itemx set print type typedefs on
15469 @itemx set print type typedefs off
15471 Normally, when @value{GDBN} prints a class, it displays any typedefs
15472 defined in that class. You can control this behavior either by
15473 passing the appropriate flag to @code{ptype}, or using @command{set
15474 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
15475 display the typedef definitions; this is the default. Specifying
15476 @code{off} will cause @value{GDBN} to omit the typedef definitions.
15477 Note that this controls whether the typedef definition itself is
15478 printed, not whether typedef names are substituted when printing other
15481 @kindex show print type typedefs
15482 @item show print type typedefs
15483 This command shows the current setting of typedef display when
15486 @kindex info address
15487 @cindex address of a symbol
15488 @item info address @var{symbol}
15489 Describe where the data for @var{symbol} is stored. For a register
15490 variable, this says which register it is kept in. For a non-register
15491 local variable, this prints the stack-frame offset at which the variable
15494 Note the contrast with @samp{print &@var{symbol}}, which does not work
15495 at all for a register variable, and for a stack local variable prints
15496 the exact address of the current instantiation of the variable.
15498 @kindex info symbol
15499 @cindex symbol from address
15500 @cindex closest symbol and offset for an address
15501 @item info symbol @var{addr}
15502 Print the name of a symbol which is stored at the address @var{addr}.
15503 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
15504 nearest symbol and an offset from it:
15507 (@value{GDBP}) info symbol 0x54320
15508 _initialize_vx + 396 in section .text
15512 This is the opposite of the @code{info address} command. You can use
15513 it to find out the name of a variable or a function given its address.
15515 For dynamically linked executables, the name of executable or shared
15516 library containing the symbol is also printed:
15519 (@value{GDBP}) info symbol 0x400225
15520 _start + 5 in section .text of /tmp/a.out
15521 (@value{GDBP}) info symbol 0x2aaaac2811cf
15522 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
15526 @item whatis[/@var{flags}] [@var{arg}]
15527 Print the data type of @var{arg}, which can be either an expression
15528 or a name of a data type. With no argument, print the data type of
15529 @code{$}, the last value in the value history.
15531 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
15532 is not actually evaluated, and any side-effecting operations (such as
15533 assignments or function calls) inside it do not take place.
15535 If @var{arg} is a variable or an expression, @code{whatis} prints its
15536 literal type as it is used in the source code. If the type was
15537 defined using a @code{typedef}, @code{whatis} will @emph{not} print
15538 the data type underlying the @code{typedef}. If the type of the
15539 variable or the expression is a compound data type, such as
15540 @code{struct} or @code{class}, @code{whatis} never prints their
15541 fields or methods. It just prints the @code{struct}/@code{class}
15542 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
15543 such a compound data type, use @code{ptype}.
15545 If @var{arg} is a type name that was defined using @code{typedef},
15546 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
15547 Unrolling means that @code{whatis} will show the underlying type used
15548 in the @code{typedef} declaration of @var{arg}. However, if that
15549 underlying type is also a @code{typedef}, @code{whatis} will not
15552 For C code, the type names may also have the form @samp{class
15553 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
15554 @var{union-tag}} or @samp{enum @var{enum-tag}}.
15556 @var{flags} can be used to modify how the type is displayed.
15557 Available flags are:
15561 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
15562 parameters and typedefs defined in a class when printing the class'
15563 members. The @code{/r} flag disables this.
15566 Do not print methods defined in the class.
15569 Print methods defined in the class. This is the default, but the flag
15570 exists in case you change the default with @command{set print type methods}.
15573 Do not print typedefs defined in the class. Note that this controls
15574 whether the typedef definition itself is printed, not whether typedef
15575 names are substituted when printing other types.
15578 Print typedefs defined in the class. This is the default, but the flag
15579 exists in case you change the default with @command{set print type typedefs}.
15583 @item ptype[/@var{flags}] [@var{arg}]
15584 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
15585 detailed description of the type, instead of just the name of the type.
15586 @xref{Expressions, ,Expressions}.
15588 Contrary to @code{whatis}, @code{ptype} always unrolls any
15589 @code{typedef}s in its argument declaration, whether the argument is
15590 a variable, expression, or a data type. This means that @code{ptype}
15591 of a variable or an expression will not print literally its type as
15592 present in the source code---use @code{whatis} for that. @code{typedef}s at
15593 the pointer or reference targets are also unrolled. Only @code{typedef}s of
15594 fields, methods and inner @code{class typedef}s of @code{struct}s,
15595 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
15597 For example, for this variable declaration:
15600 typedef double real_t;
15601 struct complex @{ real_t real; double imag; @};
15602 typedef struct complex complex_t;
15604 real_t *real_pointer_var;
15608 the two commands give this output:
15612 (@value{GDBP}) whatis var
15614 (@value{GDBP}) ptype var
15615 type = struct complex @{
15619 (@value{GDBP}) whatis complex_t
15620 type = struct complex
15621 (@value{GDBP}) whatis struct complex
15622 type = struct complex
15623 (@value{GDBP}) ptype struct complex
15624 type = struct complex @{
15628 (@value{GDBP}) whatis real_pointer_var
15630 (@value{GDBP}) ptype real_pointer_var
15636 As with @code{whatis}, using @code{ptype} without an argument refers to
15637 the type of @code{$}, the last value in the value history.
15639 @cindex incomplete type
15640 Sometimes, programs use opaque data types or incomplete specifications
15641 of complex data structure. If the debug information included in the
15642 program does not allow @value{GDBN} to display a full declaration of
15643 the data type, it will say @samp{<incomplete type>}. For example,
15644 given these declarations:
15648 struct foo *fooptr;
15652 but no definition for @code{struct foo} itself, @value{GDBN} will say:
15655 (@value{GDBP}) ptype foo
15656 $1 = <incomplete type>
15660 ``Incomplete type'' is C terminology for data types that are not
15661 completely specified.
15664 @item info types @var{regexp}
15666 Print a brief description of all types whose names match the regular
15667 expression @var{regexp} (or all types in your program, if you supply
15668 no argument). Each complete typename is matched as though it were a
15669 complete line; thus, @samp{i type value} gives information on all
15670 types in your program whose names include the string @code{value}, but
15671 @samp{i type ^value$} gives information only on types whose complete
15672 name is @code{value}.
15674 This command differs from @code{ptype} in two ways: first, like
15675 @code{whatis}, it does not print a detailed description; second, it
15676 lists all source files where a type is defined.
15678 @kindex info type-printers
15679 @item info type-printers
15680 Versions of @value{GDBN} that ship with Python scripting enabled may
15681 have ``type printers'' available. When using @command{ptype} or
15682 @command{whatis}, these printers are consulted when the name of a type
15683 is needed. @xref{Type Printing API}, for more information on writing
15686 @code{info type-printers} displays all the available type printers.
15688 @kindex enable type-printer
15689 @kindex disable type-printer
15690 @item enable type-printer @var{name}@dots{}
15691 @item disable type-printer @var{name}@dots{}
15692 These commands can be used to enable or disable type printers.
15695 @cindex local variables
15696 @item info scope @var{location}
15697 List all the variables local to a particular scope. This command
15698 accepts a @var{location} argument---a function name, a source line, or
15699 an address preceded by a @samp{*}, and prints all the variables local
15700 to the scope defined by that location. (@xref{Specify Location}, for
15701 details about supported forms of @var{location}.) For example:
15704 (@value{GDBP}) @b{info scope command_line_handler}
15705 Scope for command_line_handler:
15706 Symbol rl is an argument at stack/frame offset 8, length 4.
15707 Symbol linebuffer is in static storage at address 0x150a18, length 4.
15708 Symbol linelength is in static storage at address 0x150a1c, length 4.
15709 Symbol p is a local variable in register $esi, length 4.
15710 Symbol p1 is a local variable in register $ebx, length 4.
15711 Symbol nline is a local variable in register $edx, length 4.
15712 Symbol repeat is a local variable at frame offset -8, length 4.
15716 This command is especially useful for determining what data to collect
15717 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
15720 @kindex info source
15722 Show information about the current source file---that is, the source file for
15723 the function containing the current point of execution:
15726 the name of the source file, and the directory containing it,
15728 the directory it was compiled in,
15730 its length, in lines,
15732 which programming language it is written in,
15734 whether the executable includes debugging information for that file, and
15735 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
15737 whether the debugging information includes information about
15738 preprocessor macros.
15742 @kindex info sources
15744 Print the names of all source files in your program for which there is
15745 debugging information, organized into two lists: files whose symbols
15746 have already been read, and files whose symbols will be read when needed.
15748 @kindex info functions
15749 @item info functions
15750 Print the names and data types of all defined functions.
15752 @item info functions @var{regexp}
15753 Print the names and data types of all defined functions
15754 whose names contain a match for regular expression @var{regexp}.
15755 Thus, @samp{info fun step} finds all functions whose names
15756 include @code{step}; @samp{info fun ^step} finds those whose names
15757 start with @code{step}. If a function name contains characters
15758 that conflict with the regular expression language (e.g.@:
15759 @samp{operator*()}), they may be quoted with a backslash.
15761 @kindex info variables
15762 @item info variables
15763 Print the names and data types of all variables that are defined
15764 outside of functions (i.e.@: excluding local variables).
15766 @item info variables @var{regexp}
15767 Print the names and data types of all variables (except for local
15768 variables) whose names contain a match for regular expression
15771 @kindex info classes
15772 @cindex Objective-C, classes and selectors
15774 @itemx info classes @var{regexp}
15775 Display all Objective-C classes in your program, or
15776 (with the @var{regexp} argument) all those matching a particular regular
15779 @kindex info selectors
15780 @item info selectors
15781 @itemx info selectors @var{regexp}
15782 Display all Objective-C selectors in your program, or
15783 (with the @var{regexp} argument) all those matching a particular regular
15787 This was never implemented.
15788 @kindex info methods
15790 @itemx info methods @var{regexp}
15791 The @code{info methods} command permits the user to examine all defined
15792 methods within C@t{++} program, or (with the @var{regexp} argument) a
15793 specific set of methods found in the various C@t{++} classes. Many
15794 C@t{++} classes provide a large number of methods. Thus, the output
15795 from the @code{ptype} command can be overwhelming and hard to use. The
15796 @code{info-methods} command filters the methods, printing only those
15797 which match the regular-expression @var{regexp}.
15800 @cindex opaque data types
15801 @kindex set opaque-type-resolution
15802 @item set opaque-type-resolution on
15803 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
15804 declared as a pointer to a @code{struct}, @code{class}, or
15805 @code{union}---for example, @code{struct MyType *}---that is used in one
15806 source file although the full declaration of @code{struct MyType} is in
15807 another source file. The default is on.
15809 A change in the setting of this subcommand will not take effect until
15810 the next time symbols for a file are loaded.
15812 @item set opaque-type-resolution off
15813 Tell @value{GDBN} not to resolve opaque types. In this case, the type
15814 is printed as follows:
15816 @{<no data fields>@}
15819 @kindex show opaque-type-resolution
15820 @item show opaque-type-resolution
15821 Show whether opaque types are resolved or not.
15823 @kindex maint print symbols
15824 @cindex symbol dump
15825 @kindex maint print psymbols
15826 @cindex partial symbol dump
15827 @item maint print symbols @var{filename}
15828 @itemx maint print psymbols @var{filename}
15829 @itemx maint print msymbols @var{filename}
15830 Write a dump of debugging symbol data into the file @var{filename}.
15831 These commands are used to debug the @value{GDBN} symbol-reading code. Only
15832 symbols with debugging data are included. If you use @samp{maint print
15833 symbols}, @value{GDBN} includes all the symbols for which it has already
15834 collected full details: that is, @var{filename} reflects symbols for
15835 only those files whose symbols @value{GDBN} has read. You can use the
15836 command @code{info sources} to find out which files these are. If you
15837 use @samp{maint print psymbols} instead, the dump shows information about
15838 symbols that @value{GDBN} only knows partially---that is, symbols defined in
15839 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
15840 @samp{maint print msymbols} dumps just the minimal symbol information
15841 required for each object file from which @value{GDBN} has read some symbols.
15842 @xref{Files, ,Commands to Specify Files}, for a discussion of how
15843 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
15845 @kindex maint info symtabs
15846 @kindex maint info psymtabs
15847 @cindex listing @value{GDBN}'s internal symbol tables
15848 @cindex symbol tables, listing @value{GDBN}'s internal
15849 @cindex full symbol tables, listing @value{GDBN}'s internal
15850 @cindex partial symbol tables, listing @value{GDBN}'s internal
15851 @item maint info symtabs @r{[} @var{regexp} @r{]}
15852 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
15854 List the @code{struct symtab} or @code{struct partial_symtab}
15855 structures whose names match @var{regexp}. If @var{regexp} is not
15856 given, list them all. The output includes expressions which you can
15857 copy into a @value{GDBN} debugging this one to examine a particular
15858 structure in more detail. For example:
15861 (@value{GDBP}) maint info psymtabs dwarf2read
15862 @{ objfile /home/gnu/build/gdb/gdb
15863 ((struct objfile *) 0x82e69d0)
15864 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
15865 ((struct partial_symtab *) 0x8474b10)
15868 text addresses 0x814d3c8 -- 0x8158074
15869 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
15870 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
15871 dependencies (none)
15874 (@value{GDBP}) maint info symtabs
15878 We see that there is one partial symbol table whose filename contains
15879 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
15880 and we see that @value{GDBN} has not read in any symtabs yet at all.
15881 If we set a breakpoint on a function, that will cause @value{GDBN} to
15882 read the symtab for the compilation unit containing that function:
15885 (@value{GDBP}) break dwarf2_psymtab_to_symtab
15886 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
15888 (@value{GDBP}) maint info symtabs
15889 @{ objfile /home/gnu/build/gdb/gdb
15890 ((struct objfile *) 0x82e69d0)
15891 @{ symtab /home/gnu/src/gdb/dwarf2read.c
15892 ((struct symtab *) 0x86c1f38)
15895 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
15896 linetable ((struct linetable *) 0x8370fa0)
15897 debugformat DWARF 2
15906 @chapter Altering Execution
15908 Once you think you have found an error in your program, you might want to
15909 find out for certain whether correcting the apparent error would lead to
15910 correct results in the rest of the run. You can find the answer by
15911 experiment, using the @value{GDBN} features for altering execution of the
15914 For example, you can store new values into variables or memory
15915 locations, give your program a signal, restart it at a different
15916 address, or even return prematurely from a function.
15919 * Assignment:: Assignment to variables
15920 * Jumping:: Continuing at a different address
15921 * Signaling:: Giving your program a signal
15922 * Returning:: Returning from a function
15923 * Calling:: Calling your program's functions
15924 * Patching:: Patching your program
15928 @section Assignment to Variables
15931 @cindex setting variables
15932 To alter the value of a variable, evaluate an assignment expression.
15933 @xref{Expressions, ,Expressions}. For example,
15940 stores the value 4 into the variable @code{x}, and then prints the
15941 value of the assignment expression (which is 4).
15942 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
15943 information on operators in supported languages.
15945 @kindex set variable
15946 @cindex variables, setting
15947 If you are not interested in seeing the value of the assignment, use the
15948 @code{set} command instead of the @code{print} command. @code{set} is
15949 really the same as @code{print} except that the expression's value is
15950 not printed and is not put in the value history (@pxref{Value History,
15951 ,Value History}). The expression is evaluated only for its effects.
15953 If the beginning of the argument string of the @code{set} command
15954 appears identical to a @code{set} subcommand, use the @code{set
15955 variable} command instead of just @code{set}. This command is identical
15956 to @code{set} except for its lack of subcommands. For example, if your
15957 program has a variable @code{width}, you get an error if you try to set
15958 a new value with just @samp{set width=13}, because @value{GDBN} has the
15959 command @code{set width}:
15962 (@value{GDBP}) whatis width
15964 (@value{GDBP}) p width
15966 (@value{GDBP}) set width=47
15967 Invalid syntax in expression.
15971 The invalid expression, of course, is @samp{=47}. In
15972 order to actually set the program's variable @code{width}, use
15975 (@value{GDBP}) set var width=47
15978 Because the @code{set} command has many subcommands that can conflict
15979 with the names of program variables, it is a good idea to use the
15980 @code{set variable} command instead of just @code{set}. For example, if
15981 your program has a variable @code{g}, you run into problems if you try
15982 to set a new value with just @samp{set g=4}, because @value{GDBN} has
15983 the command @code{set gnutarget}, abbreviated @code{set g}:
15987 (@value{GDBP}) whatis g
15991 (@value{GDBP}) set g=4
15995 The program being debugged has been started already.
15996 Start it from the beginning? (y or n) y
15997 Starting program: /home/smith/cc_progs/a.out
15998 "/home/smith/cc_progs/a.out": can't open to read symbols:
15999 Invalid bfd target.
16000 (@value{GDBP}) show g
16001 The current BFD target is "=4".
16006 The program variable @code{g} did not change, and you silently set the
16007 @code{gnutarget} to an invalid value. In order to set the variable
16011 (@value{GDBP}) set var g=4
16014 @value{GDBN} allows more implicit conversions in assignments than C; you can
16015 freely store an integer value into a pointer variable or vice versa,
16016 and you can convert any structure to any other structure that is the
16017 same length or shorter.
16018 @comment FIXME: how do structs align/pad in these conversions?
16019 @comment /doc@cygnus.com 18dec1990
16021 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
16022 construct to generate a value of specified type at a specified address
16023 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
16024 to memory location @code{0x83040} as an integer (which implies a certain size
16025 and representation in memory), and
16028 set @{int@}0x83040 = 4
16032 stores the value 4 into that memory location.
16035 @section Continuing at a Different Address
16037 Ordinarily, when you continue your program, you do so at the place where
16038 it stopped, with the @code{continue} command. You can instead continue at
16039 an address of your own choosing, with the following commands:
16043 @kindex j @r{(@code{jump})}
16044 @item jump @var{linespec}
16045 @itemx j @var{linespec}
16046 @itemx jump @var{location}
16047 @itemx j @var{location}
16048 Resume execution at line @var{linespec} or at address given by
16049 @var{location}. Execution stops again immediately if there is a
16050 breakpoint there. @xref{Specify Location}, for a description of the
16051 different forms of @var{linespec} and @var{location}. It is common
16052 practice to use the @code{tbreak} command in conjunction with
16053 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
16055 The @code{jump} command does not change the current stack frame, or
16056 the stack pointer, or the contents of any memory location or any
16057 register other than the program counter. If line @var{linespec} is in
16058 a different function from the one currently executing, the results may
16059 be bizarre if the two functions expect different patterns of arguments or
16060 of local variables. For this reason, the @code{jump} command requests
16061 confirmation if the specified line is not in the function currently
16062 executing. However, even bizarre results are predictable if you are
16063 well acquainted with the machine-language code of your program.
16066 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
16067 On many systems, you can get much the same effect as the @code{jump}
16068 command by storing a new value into the register @code{$pc}. The
16069 difference is that this does not start your program running; it only
16070 changes the address of where it @emph{will} run when you continue. For
16078 makes the next @code{continue} command or stepping command execute at
16079 address @code{0x485}, rather than at the address where your program stopped.
16080 @xref{Continuing and Stepping, ,Continuing and Stepping}.
16082 The most common occasion to use the @code{jump} command is to back
16083 up---perhaps with more breakpoints set---over a portion of a program
16084 that has already executed, in order to examine its execution in more
16089 @section Giving your Program a Signal
16090 @cindex deliver a signal to a program
16094 @item signal @var{signal}
16095 Resume execution where your program stopped, but immediately give it the
16096 signal @var{signal}. @var{signal} can be the name or the number of a
16097 signal. For example, on many systems @code{signal 2} and @code{signal
16098 SIGINT} are both ways of sending an interrupt signal.
16100 Alternatively, if @var{signal} is zero, continue execution without
16101 giving a signal. This is useful when your program stopped on account of
16102 a signal and would ordinarily see the signal when resumed with the
16103 @code{continue} command; @samp{signal 0} causes it to resume without a
16106 @code{signal} does not repeat when you press @key{RET} a second time
16107 after executing the command.
16111 Invoking the @code{signal} command is not the same as invoking the
16112 @code{kill} utility from the shell. Sending a signal with @code{kill}
16113 causes @value{GDBN} to decide what to do with the signal depending on
16114 the signal handling tables (@pxref{Signals}). The @code{signal} command
16115 passes the signal directly to your program.
16119 @section Returning from a Function
16122 @cindex returning from a function
16125 @itemx return @var{expression}
16126 You can cancel execution of a function call with the @code{return}
16127 command. If you give an
16128 @var{expression} argument, its value is used as the function's return
16132 When you use @code{return}, @value{GDBN} discards the selected stack frame
16133 (and all frames within it). You can think of this as making the
16134 discarded frame return prematurely. If you wish to specify a value to
16135 be returned, give that value as the argument to @code{return}.
16137 This pops the selected stack frame (@pxref{Selection, ,Selecting a
16138 Frame}), and any other frames inside of it, leaving its caller as the
16139 innermost remaining frame. That frame becomes selected. The
16140 specified value is stored in the registers used for returning values
16143 The @code{return} command does not resume execution; it leaves the
16144 program stopped in the state that would exist if the function had just
16145 returned. In contrast, the @code{finish} command (@pxref{Continuing
16146 and Stepping, ,Continuing and Stepping}) resumes execution until the
16147 selected stack frame returns naturally.
16149 @value{GDBN} needs to know how the @var{expression} argument should be set for
16150 the inferior. The concrete registers assignment depends on the OS ABI and the
16151 type being returned by the selected stack frame. For example it is common for
16152 OS ABI to return floating point values in FPU registers while integer values in
16153 CPU registers. Still some ABIs return even floating point values in CPU
16154 registers. Larger integer widths (such as @code{long long int}) also have
16155 specific placement rules. @value{GDBN} already knows the OS ABI from its
16156 current target so it needs to find out also the type being returned to make the
16157 assignment into the right register(s).
16159 Normally, the selected stack frame has debug info. @value{GDBN} will always
16160 use the debug info instead of the implicit type of @var{expression} when the
16161 debug info is available. For example, if you type @kbd{return -1}, and the
16162 function in the current stack frame is declared to return a @code{long long
16163 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
16164 into a @code{long long int}:
16167 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
16169 (@value{GDBP}) return -1
16170 Make func return now? (y or n) y
16171 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
16172 43 printf ("result=%lld\n", func ());
16176 However, if the selected stack frame does not have a debug info, e.g., if the
16177 function was compiled without debug info, @value{GDBN} has to find out the type
16178 to return from user. Specifying a different type by mistake may set the value
16179 in different inferior registers than the caller code expects. For example,
16180 typing @kbd{return -1} with its implicit type @code{int} would set only a part
16181 of a @code{long long int} result for a debug info less function (on 32-bit
16182 architectures). Therefore the user is required to specify the return type by
16183 an appropriate cast explicitly:
16186 Breakpoint 2, 0x0040050b in func ()
16187 (@value{GDBP}) return -1
16188 Return value type not available for selected stack frame.
16189 Please use an explicit cast of the value to return.
16190 (@value{GDBP}) return (long long int) -1
16191 Make selected stack frame return now? (y or n) y
16192 #0 0x00400526 in main ()
16197 @section Calling Program Functions
16200 @cindex calling functions
16201 @cindex inferior functions, calling
16202 @item print @var{expr}
16203 Evaluate the expression @var{expr} and display the resulting value.
16204 @var{expr} may include calls to functions in the program being
16208 @item call @var{expr}
16209 Evaluate the expression @var{expr} without displaying @code{void}
16212 You can use this variant of the @code{print} command if you want to
16213 execute a function from your program that does not return anything
16214 (a.k.a.@: @dfn{a void function}), but without cluttering the output
16215 with @code{void} returned values that @value{GDBN} will otherwise
16216 print. If the result is not void, it is printed and saved in the
16220 It is possible for the function you call via the @code{print} or
16221 @code{call} command to generate a signal (e.g., if there's a bug in
16222 the function, or if you passed it incorrect arguments). What happens
16223 in that case is controlled by the @code{set unwindonsignal} command.
16225 Similarly, with a C@t{++} program it is possible for the function you
16226 call via the @code{print} or @code{call} command to generate an
16227 exception that is not handled due to the constraints of the dummy
16228 frame. In this case, any exception that is raised in the frame, but has
16229 an out-of-frame exception handler will not be found. GDB builds a
16230 dummy-frame for the inferior function call, and the unwinder cannot
16231 seek for exception handlers outside of this dummy-frame. What happens
16232 in that case is controlled by the
16233 @code{set unwind-on-terminating-exception} command.
16236 @item set unwindonsignal
16237 @kindex set unwindonsignal
16238 @cindex unwind stack in called functions
16239 @cindex call dummy stack unwinding
16240 Set unwinding of the stack if a signal is received while in a function
16241 that @value{GDBN} called in the program being debugged. If set to on,
16242 @value{GDBN} unwinds the stack it created for the call and restores
16243 the context to what it was before the call. If set to off (the
16244 default), @value{GDBN} stops in the frame where the signal was
16247 @item show unwindonsignal
16248 @kindex show unwindonsignal
16249 Show the current setting of stack unwinding in the functions called by
16252 @item set unwind-on-terminating-exception
16253 @kindex set unwind-on-terminating-exception
16254 @cindex unwind stack in called functions with unhandled exceptions
16255 @cindex call dummy stack unwinding on unhandled exception.
16256 Set unwinding of the stack if a C@t{++} exception is raised, but left
16257 unhandled while in a function that @value{GDBN} called in the program being
16258 debugged. If set to on (the default), @value{GDBN} unwinds the stack
16259 it created for the call and restores the context to what it was before
16260 the call. If set to off, @value{GDBN} the exception is delivered to
16261 the default C@t{++} exception handler and the inferior terminated.
16263 @item show unwind-on-terminating-exception
16264 @kindex show unwind-on-terminating-exception
16265 Show the current setting of stack unwinding in the functions called by
16270 @cindex weak alias functions
16271 Sometimes, a function you wish to call is actually a @dfn{weak alias}
16272 for another function. In such case, @value{GDBN} might not pick up
16273 the type information, including the types of the function arguments,
16274 which causes @value{GDBN} to call the inferior function incorrectly.
16275 As a result, the called function will function erroneously and may
16276 even crash. A solution to that is to use the name of the aliased
16280 @section Patching Programs
16282 @cindex patching binaries
16283 @cindex writing into executables
16284 @cindex writing into corefiles
16286 By default, @value{GDBN} opens the file containing your program's
16287 executable code (or the corefile) read-only. This prevents accidental
16288 alterations to machine code; but it also prevents you from intentionally
16289 patching your program's binary.
16291 If you'd like to be able to patch the binary, you can specify that
16292 explicitly with the @code{set write} command. For example, you might
16293 want to turn on internal debugging flags, or even to make emergency
16299 @itemx set write off
16300 If you specify @samp{set write on}, @value{GDBN} opens executable and
16301 core files for both reading and writing; if you specify @kbd{set write
16302 off} (the default), @value{GDBN} opens them read-only.
16304 If you have already loaded a file, you must load it again (using the
16305 @code{exec-file} or @code{core-file} command) after changing @code{set
16306 write}, for your new setting to take effect.
16310 Display whether executable files and core files are opened for writing
16311 as well as reading.
16315 @chapter @value{GDBN} Files
16317 @value{GDBN} needs to know the file name of the program to be debugged,
16318 both in order to read its symbol table and in order to start your
16319 program. To debug a core dump of a previous run, you must also tell
16320 @value{GDBN} the name of the core dump file.
16323 * Files:: Commands to specify files
16324 * Separate Debug Files:: Debugging information in separate files
16325 * MiniDebugInfo:: Debugging information in a special section
16326 * Index Files:: Index files speed up GDB
16327 * Symbol Errors:: Errors reading symbol files
16328 * Data Files:: GDB data files
16332 @section Commands to Specify Files
16334 @cindex symbol table
16335 @cindex core dump file
16337 You may want to specify executable and core dump file names. The usual
16338 way to do this is at start-up time, using the arguments to
16339 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
16340 Out of @value{GDBN}}).
16342 Occasionally it is necessary to change to a different file during a
16343 @value{GDBN} session. Or you may run @value{GDBN} and forget to
16344 specify a file you want to use. Or you are debugging a remote target
16345 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
16346 Program}). In these situations the @value{GDBN} commands to specify
16347 new files are useful.
16350 @cindex executable file
16352 @item file @var{filename}
16353 Use @var{filename} as the program to be debugged. It is read for its
16354 symbols and for the contents of pure memory. It is also the program
16355 executed when you use the @code{run} command. If you do not specify a
16356 directory and the file is not found in the @value{GDBN} working directory,
16357 @value{GDBN} uses the environment variable @code{PATH} as a list of
16358 directories to search, just as the shell does when looking for a program
16359 to run. You can change the value of this variable, for both @value{GDBN}
16360 and your program, using the @code{path} command.
16362 @cindex unlinked object files
16363 @cindex patching object files
16364 You can load unlinked object @file{.o} files into @value{GDBN} using
16365 the @code{file} command. You will not be able to ``run'' an object
16366 file, but you can disassemble functions and inspect variables. Also,
16367 if the underlying BFD functionality supports it, you could use
16368 @kbd{gdb -write} to patch object files using this technique. Note
16369 that @value{GDBN} can neither interpret nor modify relocations in this
16370 case, so branches and some initialized variables will appear to go to
16371 the wrong place. But this feature is still handy from time to time.
16374 @code{file} with no argument makes @value{GDBN} discard any information it
16375 has on both executable file and the symbol table.
16378 @item exec-file @r{[} @var{filename} @r{]}
16379 Specify that the program to be run (but not the symbol table) is found
16380 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
16381 if necessary to locate your program. Omitting @var{filename} means to
16382 discard information on the executable file.
16384 @kindex symbol-file
16385 @item symbol-file @r{[} @var{filename} @r{]}
16386 Read symbol table information from file @var{filename}. @code{PATH} is
16387 searched when necessary. Use the @code{file} command to get both symbol
16388 table and program to run from the same file.
16390 @code{symbol-file} with no argument clears out @value{GDBN} information on your
16391 program's symbol table.
16393 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
16394 some breakpoints and auto-display expressions. This is because they may
16395 contain pointers to the internal data recording symbols and data types,
16396 which are part of the old symbol table data being discarded inside
16399 @code{symbol-file} does not repeat if you press @key{RET} again after
16402 When @value{GDBN} is configured for a particular environment, it
16403 understands debugging information in whatever format is the standard
16404 generated for that environment; you may use either a @sc{gnu} compiler, or
16405 other compilers that adhere to the local conventions.
16406 Best results are usually obtained from @sc{gnu} compilers; for example,
16407 using @code{@value{NGCC}} you can generate debugging information for
16410 For most kinds of object files, with the exception of old SVR3 systems
16411 using COFF, the @code{symbol-file} command does not normally read the
16412 symbol table in full right away. Instead, it scans the symbol table
16413 quickly to find which source files and which symbols are present. The
16414 details are read later, one source file at a time, as they are needed.
16416 The purpose of this two-stage reading strategy is to make @value{GDBN}
16417 start up faster. For the most part, it is invisible except for
16418 occasional pauses while the symbol table details for a particular source
16419 file are being read. (The @code{set verbose} command can turn these
16420 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
16421 Warnings and Messages}.)
16423 We have not implemented the two-stage strategy for COFF yet. When the
16424 symbol table is stored in COFF format, @code{symbol-file} reads the
16425 symbol table data in full right away. Note that ``stabs-in-COFF''
16426 still does the two-stage strategy, since the debug info is actually
16430 @cindex reading symbols immediately
16431 @cindex symbols, reading immediately
16432 @item symbol-file @r{[} -readnow @r{]} @var{filename}
16433 @itemx file @r{[} -readnow @r{]} @var{filename}
16434 You can override the @value{GDBN} two-stage strategy for reading symbol
16435 tables by using the @samp{-readnow} option with any of the commands that
16436 load symbol table information, if you want to be sure @value{GDBN} has the
16437 entire symbol table available.
16439 @c FIXME: for now no mention of directories, since this seems to be in
16440 @c flux. 13mar1992 status is that in theory GDB would look either in
16441 @c current dir or in same dir as myprog; but issues like competing
16442 @c GDB's, or clutter in system dirs, mean that in practice right now
16443 @c only current dir is used. FFish says maybe a special GDB hierarchy
16444 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
16448 @item core-file @r{[}@var{filename}@r{]}
16450 Specify the whereabouts of a core dump file to be used as the ``contents
16451 of memory''. Traditionally, core files contain only some parts of the
16452 address space of the process that generated them; @value{GDBN} can access the
16453 executable file itself for other parts.
16455 @code{core-file} with no argument specifies that no core file is
16458 Note that the core file is ignored when your program is actually running
16459 under @value{GDBN}. So, if you have been running your program and you
16460 wish to debug a core file instead, you must kill the subprocess in which
16461 the program is running. To do this, use the @code{kill} command
16462 (@pxref{Kill Process, ,Killing the Child Process}).
16464 @kindex add-symbol-file
16465 @cindex dynamic linking
16466 @item add-symbol-file @var{filename} @var{address}
16467 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
16468 @itemx add-symbol-file @var{filename} @var{address} -s @var{section} @var{address} @dots{}
16469 The @code{add-symbol-file} command reads additional symbol table
16470 information from the file @var{filename}. You would use this command
16471 when @var{filename} has been dynamically loaded (by some other means)
16472 into the program that is running. @var{address} should be the memory
16473 address at which the file has been loaded; @value{GDBN} cannot figure
16474 this out for itself. You can additionally specify an arbitrary number
16475 of @samp{-s @var{section} @var{address}} pairs, to give an explicit
16476 section name and base address for that section. You can specify any
16477 @var{address} as an expression.
16479 The symbol table of the file @var{filename} is added to the symbol table
16480 originally read with the @code{symbol-file} command. You can use the
16481 @code{add-symbol-file} command any number of times; the new symbol data
16482 thus read keeps adding to the old. To discard all old symbol data
16483 instead, use the @code{symbol-file} command without any arguments.
16485 @cindex relocatable object files, reading symbols from
16486 @cindex object files, relocatable, reading symbols from
16487 @cindex reading symbols from relocatable object files
16488 @cindex symbols, reading from relocatable object files
16489 @cindex @file{.o} files, reading symbols from
16490 Although @var{filename} is typically a shared library file, an
16491 executable file, or some other object file which has been fully
16492 relocated for loading into a process, you can also load symbolic
16493 information from relocatable @file{.o} files, as long as:
16497 the file's symbolic information refers only to linker symbols defined in
16498 that file, not to symbols defined by other object files,
16500 every section the file's symbolic information refers to has actually
16501 been loaded into the inferior, as it appears in the file, and
16503 you can determine the address at which every section was loaded, and
16504 provide these to the @code{add-symbol-file} command.
16508 Some embedded operating systems, like Sun Chorus and VxWorks, can load
16509 relocatable files into an already running program; such systems
16510 typically make the requirements above easy to meet. However, it's
16511 important to recognize that many native systems use complex link
16512 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
16513 assembly, for example) that make the requirements difficult to meet. In
16514 general, one cannot assume that using @code{add-symbol-file} to read a
16515 relocatable object file's symbolic information will have the same effect
16516 as linking the relocatable object file into the program in the normal
16519 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
16521 @kindex add-symbol-file-from-memory
16522 @cindex @code{syscall DSO}
16523 @cindex load symbols from memory
16524 @item add-symbol-file-from-memory @var{address}
16525 Load symbols from the given @var{address} in a dynamically loaded
16526 object file whose image is mapped directly into the inferior's memory.
16527 For example, the Linux kernel maps a @code{syscall DSO} into each
16528 process's address space; this DSO provides kernel-specific code for
16529 some system calls. The argument can be any expression whose
16530 evaluation yields the address of the file's shared object file header.
16531 For this command to work, you must have used @code{symbol-file} or
16532 @code{exec-file} commands in advance.
16534 @kindex add-shared-symbol-files
16536 @item add-shared-symbol-files @var{library-file}
16537 @itemx assf @var{library-file}
16538 The @code{add-shared-symbol-files} command can currently be used only
16539 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
16540 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
16541 @value{GDBN} automatically looks for shared libraries, however if
16542 @value{GDBN} does not find yours, you can invoke
16543 @code{add-shared-symbol-files}. It takes one argument: the shared
16544 library's file name. @code{assf} is a shorthand alias for
16545 @code{add-shared-symbol-files}.
16548 @item section @var{section} @var{addr}
16549 The @code{section} command changes the base address of the named
16550 @var{section} of the exec file to @var{addr}. This can be used if the
16551 exec file does not contain section addresses, (such as in the
16552 @code{a.out} format), or when the addresses specified in the file
16553 itself are wrong. Each section must be changed separately. The
16554 @code{info files} command, described below, lists all the sections and
16558 @kindex info target
16561 @code{info files} and @code{info target} are synonymous; both print the
16562 current target (@pxref{Targets, ,Specifying a Debugging Target}),
16563 including the names of the executable and core dump files currently in
16564 use by @value{GDBN}, and the files from which symbols were loaded. The
16565 command @code{help target} lists all possible targets rather than
16568 @kindex maint info sections
16569 @item maint info sections
16570 Another command that can give you extra information about program sections
16571 is @code{maint info sections}. In addition to the section information
16572 displayed by @code{info files}, this command displays the flags and file
16573 offset of each section in the executable and core dump files. In addition,
16574 @code{maint info sections} provides the following command options (which
16575 may be arbitrarily combined):
16579 Display sections for all loaded object files, including shared libraries.
16580 @item @var{sections}
16581 Display info only for named @var{sections}.
16582 @item @var{section-flags}
16583 Display info only for sections for which @var{section-flags} are true.
16584 The section flags that @value{GDBN} currently knows about are:
16587 Section will have space allocated in the process when loaded.
16588 Set for all sections except those containing debug information.
16590 Section will be loaded from the file into the child process memory.
16591 Set for pre-initialized code and data, clear for @code{.bss} sections.
16593 Section needs to be relocated before loading.
16595 Section cannot be modified by the child process.
16597 Section contains executable code only.
16599 Section contains data only (no executable code).
16601 Section will reside in ROM.
16603 Section contains data for constructor/destructor lists.
16605 Section is not empty.
16607 An instruction to the linker to not output the section.
16608 @item COFF_SHARED_LIBRARY
16609 A notification to the linker that the section contains
16610 COFF shared library information.
16612 Section contains common symbols.
16615 @kindex set trust-readonly-sections
16616 @cindex read-only sections
16617 @item set trust-readonly-sections on
16618 Tell @value{GDBN} that readonly sections in your object file
16619 really are read-only (i.e.@: that their contents will not change).
16620 In that case, @value{GDBN} can fetch values from these sections
16621 out of the object file, rather than from the target program.
16622 For some targets (notably embedded ones), this can be a significant
16623 enhancement to debugging performance.
16625 The default is off.
16627 @item set trust-readonly-sections off
16628 Tell @value{GDBN} not to trust readonly sections. This means that
16629 the contents of the section might change while the program is running,
16630 and must therefore be fetched from the target when needed.
16632 @item show trust-readonly-sections
16633 Show the current setting of trusting readonly sections.
16636 All file-specifying commands allow both absolute and relative file names
16637 as arguments. @value{GDBN} always converts the file name to an absolute file
16638 name and remembers it that way.
16640 @cindex shared libraries
16641 @anchor{Shared Libraries}
16642 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
16643 and IBM RS/6000 AIX shared libraries.
16645 On MS-Windows @value{GDBN} must be linked with the Expat library to support
16646 shared libraries. @xref{Expat}.
16648 @value{GDBN} automatically loads symbol definitions from shared libraries
16649 when you use the @code{run} command, or when you examine a core file.
16650 (Before you issue the @code{run} command, @value{GDBN} does not understand
16651 references to a function in a shared library, however---unless you are
16652 debugging a core file).
16654 On HP-UX, if the program loads a library explicitly, @value{GDBN}
16655 automatically loads the symbols at the time of the @code{shl_load} call.
16657 @c FIXME: some @value{GDBN} release may permit some refs to undef
16658 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
16659 @c FIXME...lib; check this from time to time when updating manual
16661 There are times, however, when you may wish to not automatically load
16662 symbol definitions from shared libraries, such as when they are
16663 particularly large or there are many of them.
16665 To control the automatic loading of shared library symbols, use the
16669 @kindex set auto-solib-add
16670 @item set auto-solib-add @var{mode}
16671 If @var{mode} is @code{on}, symbols from all shared object libraries
16672 will be loaded automatically when the inferior begins execution, you
16673 attach to an independently started inferior, or when the dynamic linker
16674 informs @value{GDBN} that a new library has been loaded. If @var{mode}
16675 is @code{off}, symbols must be loaded manually, using the
16676 @code{sharedlibrary} command. The default value is @code{on}.
16678 @cindex memory used for symbol tables
16679 If your program uses lots of shared libraries with debug info that
16680 takes large amounts of memory, you can decrease the @value{GDBN}
16681 memory footprint by preventing it from automatically loading the
16682 symbols from shared libraries. To that end, type @kbd{set
16683 auto-solib-add off} before running the inferior, then load each
16684 library whose debug symbols you do need with @kbd{sharedlibrary
16685 @var{regexp}}, where @var{regexp} is a regular expression that matches
16686 the libraries whose symbols you want to be loaded.
16688 @kindex show auto-solib-add
16689 @item show auto-solib-add
16690 Display the current autoloading mode.
16693 @cindex load shared library
16694 To explicitly load shared library symbols, use the @code{sharedlibrary}
16698 @kindex info sharedlibrary
16700 @item info share @var{regex}
16701 @itemx info sharedlibrary @var{regex}
16702 Print the names of the shared libraries which are currently loaded
16703 that match @var{regex}. If @var{regex} is omitted then print
16704 all shared libraries that are loaded.
16706 @kindex sharedlibrary
16708 @item sharedlibrary @var{regex}
16709 @itemx share @var{regex}
16710 Load shared object library symbols for files matching a
16711 Unix regular expression.
16712 As with files loaded automatically, it only loads shared libraries
16713 required by your program for a core file or after typing @code{run}. If
16714 @var{regex} is omitted all shared libraries required by your program are
16717 @item nosharedlibrary
16718 @kindex nosharedlibrary
16719 @cindex unload symbols from shared libraries
16720 Unload all shared object library symbols. This discards all symbols
16721 that have been loaded from all shared libraries. Symbols from shared
16722 libraries that were loaded by explicit user requests are not
16726 Sometimes you may wish that @value{GDBN} stops and gives you control
16727 when any of shared library events happen. The best way to do this is
16728 to use @code{catch load} and @code{catch unload} (@pxref{Set
16731 @value{GDBN} also supports the the @code{set stop-on-solib-events}
16732 command for this. This command exists for historical reasons. It is
16733 less useful than setting a catchpoint, because it does not allow for
16734 conditions or commands as a catchpoint does.
16737 @item set stop-on-solib-events
16738 @kindex set stop-on-solib-events
16739 This command controls whether @value{GDBN} should give you control
16740 when the dynamic linker notifies it about some shared library event.
16741 The most common event of interest is loading or unloading of a new
16744 @item show stop-on-solib-events
16745 @kindex show stop-on-solib-events
16746 Show whether @value{GDBN} stops and gives you control when shared
16747 library events happen.
16750 Shared libraries are also supported in many cross or remote debugging
16751 configurations. @value{GDBN} needs to have access to the target's libraries;
16752 this can be accomplished either by providing copies of the libraries
16753 on the host system, or by asking @value{GDBN} to automatically retrieve the
16754 libraries from the target. If copies of the target libraries are
16755 provided, they need to be the same as the target libraries, although the
16756 copies on the target can be stripped as long as the copies on the host are
16759 @cindex where to look for shared libraries
16760 For remote debugging, you need to tell @value{GDBN} where the target
16761 libraries are, so that it can load the correct copies---otherwise, it
16762 may try to load the host's libraries. @value{GDBN} has two variables
16763 to specify the search directories for target libraries.
16766 @cindex prefix for shared library file names
16767 @cindex system root, alternate
16768 @kindex set solib-absolute-prefix
16769 @kindex set sysroot
16770 @item set sysroot @var{path}
16771 Use @var{path} as the system root for the program being debugged. Any
16772 absolute shared library paths will be prefixed with @var{path}; many
16773 runtime loaders store the absolute paths to the shared library in the
16774 target program's memory. If you use @code{set sysroot} to find shared
16775 libraries, they need to be laid out in the same way that they are on
16776 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
16779 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
16780 retrieve the target libraries from the remote system. This is only
16781 supported when using a remote target that supports the @code{remote get}
16782 command (@pxref{File Transfer,,Sending files to a remote system}).
16783 The part of @var{path} following the initial @file{remote:}
16784 (if present) is used as system root prefix on the remote file system.
16785 @footnote{If you want to specify a local system root using a directory
16786 that happens to be named @file{remote:}, you need to use some equivalent
16787 variant of the name like @file{./remote:}.}
16789 For targets with an MS-DOS based filesystem, such as MS-Windows and
16790 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
16791 absolute file name with @var{path}. But first, on Unix hosts,
16792 @value{GDBN} converts all backslash directory separators into forward
16793 slashes, because the backslash is not a directory separator on Unix:
16796 c:\foo\bar.dll @result{} c:/foo/bar.dll
16799 Then, @value{GDBN} attempts prefixing the target file name with
16800 @var{path}, and looks for the resulting file name in the host file
16804 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
16807 If that does not find the shared library, @value{GDBN} tries removing
16808 the @samp{:} character from the drive spec, both for convenience, and,
16809 for the case of the host file system not supporting file names with
16813 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
16816 This makes it possible to have a system root that mirrors a target
16817 with more than one drive. E.g., you may want to setup your local
16818 copies of the target system shared libraries like so (note @samp{c} vs
16822 @file{/path/to/sysroot/c/sys/bin/foo.dll}
16823 @file{/path/to/sysroot/c/sys/bin/bar.dll}
16824 @file{/path/to/sysroot/z/sys/bin/bar.dll}
16828 and point the system root at @file{/path/to/sysroot}, so that
16829 @value{GDBN} can find the correct copies of both
16830 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
16832 If that still does not find the shared library, @value{GDBN} tries
16833 removing the whole drive spec from the target file name:
16836 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
16839 This last lookup makes it possible to not care about the drive name,
16840 if you don't want or need to.
16842 The @code{set solib-absolute-prefix} command is an alias for @code{set
16845 @cindex default system root
16846 @cindex @samp{--with-sysroot}
16847 You can set the default system root by using the configure-time
16848 @samp{--with-sysroot} option. If the system root is inside
16849 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
16850 @samp{--exec-prefix}), then the default system root will be updated
16851 automatically if the installed @value{GDBN} is moved to a new
16854 @kindex show sysroot
16856 Display the current shared library prefix.
16858 @kindex set solib-search-path
16859 @item set solib-search-path @var{path}
16860 If this variable is set, @var{path} is a colon-separated list of
16861 directories to search for shared libraries. @samp{solib-search-path}
16862 is used after @samp{sysroot} fails to locate the library, or if the
16863 path to the library is relative instead of absolute. If you want to
16864 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
16865 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
16866 finding your host's libraries. @samp{sysroot} is preferred; setting
16867 it to a nonexistent directory may interfere with automatic loading
16868 of shared library symbols.
16870 @kindex show solib-search-path
16871 @item show solib-search-path
16872 Display the current shared library search path.
16874 @cindex DOS file-name semantics of file names.
16875 @kindex set target-file-system-kind (unix|dos-based|auto)
16876 @kindex show target-file-system-kind
16877 @item set target-file-system-kind @var{kind}
16878 Set assumed file system kind for target reported file names.
16880 Shared library file names as reported by the target system may not
16881 make sense as is on the system @value{GDBN} is running on. For
16882 example, when remote debugging a target that has MS-DOS based file
16883 system semantics, from a Unix host, the target may be reporting to
16884 @value{GDBN} a list of loaded shared libraries with file names such as
16885 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
16886 drive letters, so the @samp{c:\} prefix is not normally understood as
16887 indicating an absolute file name, and neither is the backslash
16888 normally considered a directory separator character. In that case,
16889 the native file system would interpret this whole absolute file name
16890 as a relative file name with no directory components. This would make
16891 it impossible to point @value{GDBN} at a copy of the remote target's
16892 shared libraries on the host using @code{set sysroot}, and impractical
16893 with @code{set solib-search-path}. Setting
16894 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
16895 to interpret such file names similarly to how the target would, and to
16896 map them to file names valid on @value{GDBN}'s native file system
16897 semantics. The value of @var{kind} can be @code{"auto"}, in addition
16898 to one of the supported file system kinds. In that case, @value{GDBN}
16899 tries to determine the appropriate file system variant based on the
16900 current target's operating system (@pxref{ABI, ,Configuring the
16901 Current ABI}). The supported file system settings are:
16905 Instruct @value{GDBN} to assume the target file system is of Unix
16906 kind. Only file names starting the forward slash (@samp{/}) character
16907 are considered absolute, and the directory separator character is also
16911 Instruct @value{GDBN} to assume the target file system is DOS based.
16912 File names starting with either a forward slash, or a drive letter
16913 followed by a colon (e.g., @samp{c:}), are considered absolute, and
16914 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
16915 considered directory separators.
16918 Instruct @value{GDBN} to use the file system kind associated with the
16919 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
16920 This is the default.
16924 @cindex file name canonicalization
16925 @cindex base name differences
16926 When processing file names provided by the user, @value{GDBN}
16927 frequently needs to compare them to the file names recorded in the
16928 program's debug info. Normally, @value{GDBN} compares just the
16929 @dfn{base names} of the files as strings, which is reasonably fast
16930 even for very large programs. (The base name of a file is the last
16931 portion of its name, after stripping all the leading directories.)
16932 This shortcut in comparison is based upon the assumption that files
16933 cannot have more than one base name. This is usually true, but
16934 references to files that use symlinks or similar filesystem
16935 facilities violate that assumption. If your program records files
16936 using such facilities, or if you provide file names to @value{GDBN}
16937 using symlinks etc., you can set @code{basenames-may-differ} to
16938 @code{true} to instruct @value{GDBN} to completely canonicalize each
16939 pair of file names it needs to compare. This will make file-name
16940 comparisons accurate, but at a price of a significant slowdown.
16943 @item set basenames-may-differ
16944 @kindex set basenames-may-differ
16945 Set whether a source file may have multiple base names.
16947 @item show basenames-may-differ
16948 @kindex show basenames-may-differ
16949 Show whether a source file may have multiple base names.
16952 @node Separate Debug Files
16953 @section Debugging Information in Separate Files
16954 @cindex separate debugging information files
16955 @cindex debugging information in separate files
16956 @cindex @file{.debug} subdirectories
16957 @cindex debugging information directory, global
16958 @cindex global debugging information directories
16959 @cindex build ID, and separate debugging files
16960 @cindex @file{.build-id} directory
16962 @value{GDBN} allows you to put a program's debugging information in a
16963 file separate from the executable itself, in a way that allows
16964 @value{GDBN} to find and load the debugging information automatically.
16965 Since debugging information can be very large---sometimes larger
16966 than the executable code itself---some systems distribute debugging
16967 information for their executables in separate files, which users can
16968 install only when they need to debug a problem.
16970 @value{GDBN} supports two ways of specifying the separate debug info
16975 The executable contains a @dfn{debug link} that specifies the name of
16976 the separate debug info file. The separate debug file's name is
16977 usually @file{@var{executable}.debug}, where @var{executable} is the
16978 name of the corresponding executable file without leading directories
16979 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
16980 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
16981 checksum for the debug file, which @value{GDBN} uses to validate that
16982 the executable and the debug file came from the same build.
16985 The executable contains a @dfn{build ID}, a unique bit string that is
16986 also present in the corresponding debug info file. (This is supported
16987 only on some operating systems, notably those which use the ELF format
16988 for binary files and the @sc{gnu} Binutils.) For more details about
16989 this feature, see the description of the @option{--build-id}
16990 command-line option in @ref{Options, , Command Line Options, ld.info,
16991 The GNU Linker}. The debug info file's name is not specified
16992 explicitly by the build ID, but can be computed from the build ID, see
16996 Depending on the way the debug info file is specified, @value{GDBN}
16997 uses two different methods of looking for the debug file:
17001 For the ``debug link'' method, @value{GDBN} looks up the named file in
17002 the directory of the executable file, then in a subdirectory of that
17003 directory named @file{.debug}, and finally under each one of the global debug
17004 directories, in a subdirectory whose name is identical to the leading
17005 directories of the executable's absolute file name.
17008 For the ``build ID'' method, @value{GDBN} looks in the
17009 @file{.build-id} subdirectory of each one of the global debug directories for
17010 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
17011 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
17012 are the rest of the bit string. (Real build ID strings are 32 or more
17013 hex characters, not 10.)
17016 So, for example, suppose you ask @value{GDBN} to debug
17017 @file{/usr/bin/ls}, which has a debug link that specifies the
17018 file @file{ls.debug}, and a build ID whose value in hex is
17019 @code{abcdef1234}. If the list of the global debug directories includes
17020 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
17021 debug information files, in the indicated order:
17025 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
17027 @file{/usr/bin/ls.debug}
17029 @file{/usr/bin/.debug/ls.debug}
17031 @file{/usr/lib/debug/usr/bin/ls.debug}.
17034 @anchor{debug-file-directory}
17035 Global debugging info directories default to what is set by @value{GDBN}
17036 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
17037 you can also set the global debugging info directories, and view the list
17038 @value{GDBN} is currently using.
17042 @kindex set debug-file-directory
17043 @item set debug-file-directory @var{directories}
17044 Set the directories which @value{GDBN} searches for separate debugging
17045 information files to @var{directory}. Multiple path components can be set
17046 concatenating them by a path separator.
17048 @kindex show debug-file-directory
17049 @item show debug-file-directory
17050 Show the directories @value{GDBN} searches for separate debugging
17055 @cindex @code{.gnu_debuglink} sections
17056 @cindex debug link sections
17057 A debug link is a special section of the executable file named
17058 @code{.gnu_debuglink}. The section must contain:
17062 A filename, with any leading directory components removed, followed by
17065 zero to three bytes of padding, as needed to reach the next four-byte
17066 boundary within the section, and
17068 a four-byte CRC checksum, stored in the same endianness used for the
17069 executable file itself. The checksum is computed on the debugging
17070 information file's full contents by the function given below, passing
17071 zero as the @var{crc} argument.
17074 Any executable file format can carry a debug link, as long as it can
17075 contain a section named @code{.gnu_debuglink} with the contents
17078 @cindex @code{.note.gnu.build-id} sections
17079 @cindex build ID sections
17080 The build ID is a special section in the executable file (and in other
17081 ELF binary files that @value{GDBN} may consider). This section is
17082 often named @code{.note.gnu.build-id}, but that name is not mandatory.
17083 It contains unique identification for the built files---the ID remains
17084 the same across multiple builds of the same build tree. The default
17085 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
17086 content for the build ID string. The same section with an identical
17087 value is present in the original built binary with symbols, in its
17088 stripped variant, and in the separate debugging information file.
17090 The debugging information file itself should be an ordinary
17091 executable, containing a full set of linker symbols, sections, and
17092 debugging information. The sections of the debugging information file
17093 should have the same names, addresses, and sizes as the original file,
17094 but they need not contain any data---much like a @code{.bss} section
17095 in an ordinary executable.
17097 The @sc{gnu} binary utilities (Binutils) package includes the
17098 @samp{objcopy} utility that can produce
17099 the separated executable / debugging information file pairs using the
17100 following commands:
17103 @kbd{objcopy --only-keep-debug foo foo.debug}
17108 These commands remove the debugging
17109 information from the executable file @file{foo} and place it in the file
17110 @file{foo.debug}. You can use the first, second or both methods to link the
17115 The debug link method needs the following additional command to also leave
17116 behind a debug link in @file{foo}:
17119 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
17122 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
17123 a version of the @code{strip} command such that the command @kbd{strip foo -f
17124 foo.debug} has the same functionality as the two @code{objcopy} commands and
17125 the @code{ln -s} command above, together.
17128 Build ID gets embedded into the main executable using @code{ld --build-id} or
17129 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
17130 compatibility fixes for debug files separation are present in @sc{gnu} binary
17131 utilities (Binutils) package since version 2.18.
17136 @cindex CRC algorithm definition
17137 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
17138 IEEE 802.3 using the polynomial:
17140 @c TexInfo requires naked braces for multi-digit exponents for Tex
17141 @c output, but this causes HTML output to barf. HTML has to be set using
17142 @c raw commands. So we end up having to specify this equation in 2
17147 <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>
17148 + <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
17154 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
17155 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
17159 The function is computed byte at a time, taking the least
17160 significant bit of each byte first. The initial pattern
17161 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
17162 the final result is inverted to ensure trailing zeros also affect the
17165 @emph{Note:} This is the same CRC polynomial as used in handling the
17166 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
17167 , @value{GDBN} Remote Serial Protocol}). However in the
17168 case of the Remote Serial Protocol, the CRC is computed @emph{most}
17169 significant bit first, and the result is not inverted, so trailing
17170 zeros have no effect on the CRC value.
17172 To complete the description, we show below the code of the function
17173 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
17174 initially supplied @code{crc} argument means that an initial call to
17175 this function passing in zero will start computing the CRC using
17178 @kindex gnu_debuglink_crc32
17181 gnu_debuglink_crc32 (unsigned long crc,
17182 unsigned char *buf, size_t len)
17184 static const unsigned long crc32_table[256] =
17186 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
17187 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
17188 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
17189 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
17190 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
17191 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
17192 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
17193 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
17194 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
17195 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
17196 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
17197 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
17198 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
17199 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
17200 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
17201 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
17202 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
17203 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
17204 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
17205 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
17206 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
17207 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
17208 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
17209 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
17210 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
17211 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
17212 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
17213 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
17214 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
17215 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
17216 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
17217 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
17218 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
17219 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
17220 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
17221 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
17222 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
17223 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
17224 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
17225 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
17226 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
17227 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
17228 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
17229 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
17230 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
17231 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
17232 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
17233 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
17234 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
17235 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
17236 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
17239 unsigned char *end;
17241 crc = ~crc & 0xffffffff;
17242 for (end = buf + len; buf < end; ++buf)
17243 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
17244 return ~crc & 0xffffffff;
17249 This computation does not apply to the ``build ID'' method.
17251 @node MiniDebugInfo
17252 @section Debugging information in a special section
17253 @cindex separate debug sections
17254 @cindex @samp{.gnu_debugdata} section
17256 Some systems ship pre-built executables and libraries that have a
17257 special @samp{.gnu_debugdata} section. This feature is called
17258 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
17259 is used to supply extra symbols for backtraces.
17261 The intent of this section is to provide extra minimal debugging
17262 information for use in simple backtraces. It is not intended to be a
17263 replacement for full separate debugging information (@pxref{Separate
17264 Debug Files}). The example below shows the intended use; however,
17265 @value{GDBN} does not currently put restrictions on what sort of
17266 debugging information might be included in the section.
17268 @value{GDBN} has support for this extension. If the section exists,
17269 then it is used provided that no other source of debugging information
17270 can be found, and that @value{GDBN} was configured with LZMA support.
17272 This section can be easily created using @command{objcopy} and other
17273 standard utilities:
17276 # Extract the dynamic symbols from the main binary, there is no need
17277 # to also have these in the normal symbol table
17278 nm -D @var{binary} --format=posix --defined-only \
17279 | awk '@{ print $1 @}' | sort > dynsyms
17281 # Extract all the text (i.e. function) symbols from the debuginfo .
17282 nm @var{binary} --format=posix --defined-only \
17283 | awk '@{ if ($2 == "T" || $2 == "t") print $1 @}' \
17286 # Keep all the function symbols not already in the dynamic symbol
17288 comm -13 dynsyms funcsyms > keep_symbols
17290 # Copy the full debuginfo, keeping only a minimal set of symbols and
17291 # removing some unnecessary sections.
17292 objcopy -S --remove-section .gdb_index --remove-section .comment \
17293 --keep-symbols=keep_symbols @var{binary} mini_debuginfo
17295 # Inject the compressed data into the .gnu_debugdata section of the
17298 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
17302 @section Index Files Speed Up @value{GDBN}
17303 @cindex index files
17304 @cindex @samp{.gdb_index} section
17306 When @value{GDBN} finds a symbol file, it scans the symbols in the
17307 file in order to construct an internal symbol table. This lets most
17308 @value{GDBN} operations work quickly---at the cost of a delay early
17309 on. For large programs, this delay can be quite lengthy, so
17310 @value{GDBN} provides a way to build an index, which speeds up
17313 The index is stored as a section in the symbol file. @value{GDBN} can
17314 write the index to a file, then you can put it into the symbol file
17315 using @command{objcopy}.
17317 To create an index file, use the @code{save gdb-index} command:
17320 @item save gdb-index @var{directory}
17321 @kindex save gdb-index
17322 Create an index file for each symbol file currently known by
17323 @value{GDBN}. Each file is named after its corresponding symbol file,
17324 with @samp{.gdb-index} appended, and is written into the given
17328 Once you have created an index file you can merge it into your symbol
17329 file, here named @file{symfile}, using @command{objcopy}:
17332 $ objcopy --add-section .gdb_index=symfile.gdb-index \
17333 --set-section-flags .gdb_index=readonly symfile symfile
17336 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
17337 sections that have been deprecated. Usually they are deprecated because
17338 they are missing a new feature or have performance issues.
17339 To tell @value{GDBN} to use a deprecated index section anyway
17340 specify @code{set use-deprecated-index-sections on}.
17341 The default is @code{off}.
17342 This can speed up startup, but may result in some functionality being lost.
17343 @xref{Index Section Format}.
17345 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
17346 must be done before gdb reads the file. The following will not work:
17349 $ gdb -ex "set use-deprecated-index-sections on" <program>
17352 Instead you must do, for example,
17355 $ gdb -iex "set use-deprecated-index-sections on" <program>
17358 There are currently some limitation on indices. They only work when
17359 for DWARF debugging information, not stabs. And, they do not
17360 currently work for programs using Ada.
17362 @node Symbol Errors
17363 @section Errors Reading Symbol Files
17365 While reading a symbol file, @value{GDBN} occasionally encounters problems,
17366 such as symbol types it does not recognize, or known bugs in compiler
17367 output. By default, @value{GDBN} does not notify you of such problems, since
17368 they are relatively common and primarily of interest to people
17369 debugging compilers. If you are interested in seeing information
17370 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
17371 only one message about each such type of problem, no matter how many
17372 times the problem occurs; or you can ask @value{GDBN} to print more messages,
17373 to see how many times the problems occur, with the @code{set
17374 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
17377 The messages currently printed, and their meanings, include:
17380 @item inner block not inside outer block in @var{symbol}
17382 The symbol information shows where symbol scopes begin and end
17383 (such as at the start of a function or a block of statements). This
17384 error indicates that an inner scope block is not fully contained
17385 in its outer scope blocks.
17387 @value{GDBN} circumvents the problem by treating the inner block as if it had
17388 the same scope as the outer block. In the error message, @var{symbol}
17389 may be shown as ``@code{(don't know)}'' if the outer block is not a
17392 @item block at @var{address} out of order
17394 The symbol information for symbol scope blocks should occur in
17395 order of increasing addresses. This error indicates that it does not
17398 @value{GDBN} does not circumvent this problem, and has trouble
17399 locating symbols in the source file whose symbols it is reading. (You
17400 can often determine what source file is affected by specifying
17401 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
17404 @item bad block start address patched
17406 The symbol information for a symbol scope block has a start address
17407 smaller than the address of the preceding source line. This is known
17408 to occur in the SunOS 4.1.1 (and earlier) C compiler.
17410 @value{GDBN} circumvents the problem by treating the symbol scope block as
17411 starting on the previous source line.
17413 @item bad string table offset in symbol @var{n}
17416 Symbol number @var{n} contains a pointer into the string table which is
17417 larger than the size of the string table.
17419 @value{GDBN} circumvents the problem by considering the symbol to have the
17420 name @code{foo}, which may cause other problems if many symbols end up
17423 @item unknown symbol type @code{0x@var{nn}}
17425 The symbol information contains new data types that @value{GDBN} does
17426 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
17427 uncomprehended information, in hexadecimal.
17429 @value{GDBN} circumvents the error by ignoring this symbol information.
17430 This usually allows you to debug your program, though certain symbols
17431 are not accessible. If you encounter such a problem and feel like
17432 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
17433 on @code{complain}, then go up to the function @code{read_dbx_symtab}
17434 and examine @code{*bufp} to see the symbol.
17436 @item stub type has NULL name
17438 @value{GDBN} could not find the full definition for a struct or class.
17440 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
17441 The symbol information for a C@t{++} member function is missing some
17442 information that recent versions of the compiler should have output for
17445 @item info mismatch between compiler and debugger
17447 @value{GDBN} could not parse a type specification output by the compiler.
17452 @section GDB Data Files
17454 @cindex prefix for data files
17455 @value{GDBN} will sometimes read an auxiliary data file. These files
17456 are kept in a directory known as the @dfn{data directory}.
17458 You can set the data directory's name, and view the name @value{GDBN}
17459 is currently using.
17462 @kindex set data-directory
17463 @item set data-directory @var{directory}
17464 Set the directory which @value{GDBN} searches for auxiliary data files
17465 to @var{directory}.
17467 @kindex show data-directory
17468 @item show data-directory
17469 Show the directory @value{GDBN} searches for auxiliary data files.
17472 @cindex default data directory
17473 @cindex @samp{--with-gdb-datadir}
17474 You can set the default data directory by using the configure-time
17475 @samp{--with-gdb-datadir} option. If the data directory is inside
17476 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
17477 @samp{--exec-prefix}), then the default data directory will be updated
17478 automatically if the installed @value{GDBN} is moved to a new
17481 The data directory may also be specified with the
17482 @code{--data-directory} command line option.
17483 @xref{Mode Options}.
17486 @chapter Specifying a Debugging Target
17488 @cindex debugging target
17489 A @dfn{target} is the execution environment occupied by your program.
17491 Often, @value{GDBN} runs in the same host environment as your program;
17492 in that case, the debugging target is specified as a side effect when
17493 you use the @code{file} or @code{core} commands. When you need more
17494 flexibility---for example, running @value{GDBN} on a physically separate
17495 host, or controlling a standalone system over a serial port or a
17496 realtime system over a TCP/IP connection---you can use the @code{target}
17497 command to specify one of the target types configured for @value{GDBN}
17498 (@pxref{Target Commands, ,Commands for Managing Targets}).
17500 @cindex target architecture
17501 It is possible to build @value{GDBN} for several different @dfn{target
17502 architectures}. When @value{GDBN} is built like that, you can choose
17503 one of the available architectures with the @kbd{set architecture}
17507 @kindex set architecture
17508 @kindex show architecture
17509 @item set architecture @var{arch}
17510 This command sets the current target architecture to @var{arch}. The
17511 value of @var{arch} can be @code{"auto"}, in addition to one of the
17512 supported architectures.
17514 @item show architecture
17515 Show the current target architecture.
17517 @item set processor
17519 @kindex set processor
17520 @kindex show processor
17521 These are alias commands for, respectively, @code{set architecture}
17522 and @code{show architecture}.
17526 * Active Targets:: Active targets
17527 * Target Commands:: Commands for managing targets
17528 * Byte Order:: Choosing target byte order
17531 @node Active Targets
17532 @section Active Targets
17534 @cindex stacking targets
17535 @cindex active targets
17536 @cindex multiple targets
17538 There are multiple classes of targets such as: processes, executable files or
17539 recording sessions. Core files belong to the process class, making core file
17540 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
17541 on multiple active targets, one in each class. This allows you to (for
17542 example) start a process and inspect its activity, while still having access to
17543 the executable file after the process finishes. Or if you start process
17544 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
17545 presented a virtual layer of the recording target, while the process target
17546 remains stopped at the chronologically last point of the process execution.
17548 Use the @code{core-file} and @code{exec-file} commands to select a new core
17549 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
17550 specify as a target a process that is already running, use the @code{attach}
17551 command (@pxref{Attach, ,Debugging an Already-running Process}).
17553 @node Target Commands
17554 @section Commands for Managing Targets
17557 @item target @var{type} @var{parameters}
17558 Connects the @value{GDBN} host environment to a target machine or
17559 process. A target is typically a protocol for talking to debugging
17560 facilities. You use the argument @var{type} to specify the type or
17561 protocol of the target machine.
17563 Further @var{parameters} are interpreted by the target protocol, but
17564 typically include things like device names or host names to connect
17565 with, process numbers, and baud rates.
17567 The @code{target} command does not repeat if you press @key{RET} again
17568 after executing the command.
17570 @kindex help target
17572 Displays the names of all targets available. To display targets
17573 currently selected, use either @code{info target} or @code{info files}
17574 (@pxref{Files, ,Commands to Specify Files}).
17576 @item help target @var{name}
17577 Describe a particular target, including any parameters necessary to
17580 @kindex set gnutarget
17581 @item set gnutarget @var{args}
17582 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
17583 knows whether it is reading an @dfn{executable},
17584 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
17585 with the @code{set gnutarget} command. Unlike most @code{target} commands,
17586 with @code{gnutarget} the @code{target} refers to a program, not a machine.
17589 @emph{Warning:} To specify a file format with @code{set gnutarget},
17590 you must know the actual BFD name.
17594 @xref{Files, , Commands to Specify Files}.
17596 @kindex show gnutarget
17597 @item show gnutarget
17598 Use the @code{show gnutarget} command to display what file format
17599 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
17600 @value{GDBN} will determine the file format for each file automatically,
17601 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
17604 @cindex common targets
17605 Here are some common targets (available, or not, depending on the GDB
17610 @item target exec @var{program}
17611 @cindex executable file target
17612 An executable file. @samp{target exec @var{program}} is the same as
17613 @samp{exec-file @var{program}}.
17615 @item target core @var{filename}
17616 @cindex core dump file target
17617 A core dump file. @samp{target core @var{filename}} is the same as
17618 @samp{core-file @var{filename}}.
17620 @item target remote @var{medium}
17621 @cindex remote target
17622 A remote system connected to @value{GDBN} via a serial line or network
17623 connection. This command tells @value{GDBN} to use its own remote
17624 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
17626 For example, if you have a board connected to @file{/dev/ttya} on the
17627 machine running @value{GDBN}, you could say:
17630 target remote /dev/ttya
17633 @code{target remote} supports the @code{load} command. This is only
17634 useful if you have some other way of getting the stub to the target
17635 system, and you can put it somewhere in memory where it won't get
17636 clobbered by the download.
17638 @item target sim @r{[}@var{simargs}@r{]} @dots{}
17639 @cindex built-in simulator target
17640 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
17648 works; however, you cannot assume that a specific memory map, device
17649 drivers, or even basic I/O is available, although some simulators do
17650 provide these. For info about any processor-specific simulator details,
17651 see the appropriate section in @ref{Embedded Processors, ,Embedded
17656 Some configurations may include these targets as well:
17660 @item target nrom @var{dev}
17661 @cindex NetROM ROM emulator target
17662 NetROM ROM emulator. This target only supports downloading.
17666 Different targets are available on different configurations of @value{GDBN};
17667 your configuration may have more or fewer targets.
17669 Many remote targets require you to download the executable's code once
17670 you've successfully established a connection. You may wish to control
17671 various aspects of this process.
17676 @kindex set hash@r{, for remote monitors}
17677 @cindex hash mark while downloading
17678 This command controls whether a hash mark @samp{#} is displayed while
17679 downloading a file to the remote monitor. If on, a hash mark is
17680 displayed after each S-record is successfully downloaded to the
17684 @kindex show hash@r{, for remote monitors}
17685 Show the current status of displaying the hash mark.
17687 @item set debug monitor
17688 @kindex set debug monitor
17689 @cindex display remote monitor communications
17690 Enable or disable display of communications messages between
17691 @value{GDBN} and the remote monitor.
17693 @item show debug monitor
17694 @kindex show debug monitor
17695 Show the current status of displaying communications between
17696 @value{GDBN} and the remote monitor.
17701 @kindex load @var{filename}
17702 @item load @var{filename}
17704 Depending on what remote debugging facilities are configured into
17705 @value{GDBN}, the @code{load} command may be available. Where it exists, it
17706 is meant to make @var{filename} (an executable) available for debugging
17707 on the remote system---by downloading, or dynamic linking, for example.
17708 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
17709 the @code{add-symbol-file} command.
17711 If your @value{GDBN} does not have a @code{load} command, attempting to
17712 execute it gets the error message ``@code{You can't do that when your
17713 target is @dots{}}''
17715 The file is loaded at whatever address is specified in the executable.
17716 For some object file formats, you can specify the load address when you
17717 link the program; for other formats, like a.out, the object file format
17718 specifies a fixed address.
17719 @c FIXME! This would be a good place for an xref to the GNU linker doc.
17721 Depending on the remote side capabilities, @value{GDBN} may be able to
17722 load programs into flash memory.
17724 @code{load} does not repeat if you press @key{RET} again after using it.
17728 @section Choosing Target Byte Order
17730 @cindex choosing target byte order
17731 @cindex target byte order
17733 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
17734 offer the ability to run either big-endian or little-endian byte
17735 orders. Usually the executable or symbol will include a bit to
17736 designate the endian-ness, and you will not need to worry about
17737 which to use. However, you may still find it useful to adjust
17738 @value{GDBN}'s idea of processor endian-ness manually.
17742 @item set endian big
17743 Instruct @value{GDBN} to assume the target is big-endian.
17745 @item set endian little
17746 Instruct @value{GDBN} to assume the target is little-endian.
17748 @item set endian auto
17749 Instruct @value{GDBN} to use the byte order associated with the
17753 Display @value{GDBN}'s current idea of the target byte order.
17757 Note that these commands merely adjust interpretation of symbolic
17758 data on the host, and that they have absolutely no effect on the
17762 @node Remote Debugging
17763 @chapter Debugging Remote Programs
17764 @cindex remote debugging
17766 If you are trying to debug a program running on a machine that cannot run
17767 @value{GDBN} in the usual way, it is often useful to use remote debugging.
17768 For example, you might use remote debugging on an operating system kernel,
17769 or on a small system which does not have a general purpose operating system
17770 powerful enough to run a full-featured debugger.
17772 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
17773 to make this work with particular debugging targets. In addition,
17774 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
17775 but not specific to any particular target system) which you can use if you
17776 write the remote stubs---the code that runs on the remote system to
17777 communicate with @value{GDBN}.
17779 Other remote targets may be available in your
17780 configuration of @value{GDBN}; use @code{help target} to list them.
17783 * Connecting:: Connecting to a remote target
17784 * File Transfer:: Sending files to a remote system
17785 * Server:: Using the gdbserver program
17786 * Remote Configuration:: Remote configuration
17787 * Remote Stub:: Implementing a remote stub
17791 @section Connecting to a Remote Target
17793 On the @value{GDBN} host machine, you will need an unstripped copy of
17794 your program, since @value{GDBN} needs symbol and debugging information.
17795 Start up @value{GDBN} as usual, using the name of the local copy of your
17796 program as the first argument.
17798 @cindex @code{target remote}
17799 @value{GDBN} can communicate with the target over a serial line, or
17800 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
17801 each case, @value{GDBN} uses the same protocol for debugging your
17802 program; only the medium carrying the debugging packets varies. The
17803 @code{target remote} command establishes a connection to the target.
17804 Its arguments indicate which medium to use:
17808 @item target remote @var{serial-device}
17809 @cindex serial line, @code{target remote}
17810 Use @var{serial-device} to communicate with the target. For example,
17811 to use a serial line connected to the device named @file{/dev/ttyb}:
17814 target remote /dev/ttyb
17817 If you're using a serial line, you may want to give @value{GDBN} the
17818 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
17819 (@pxref{Remote Configuration, set remotebaud}) before the
17820 @code{target} command.
17822 @item target remote @code{@var{host}:@var{port}}
17823 @itemx target remote @code{tcp:@var{host}:@var{port}}
17824 @cindex @acronym{TCP} port, @code{target remote}
17825 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
17826 The @var{host} may be either a host name or a numeric @acronym{IP}
17827 address; @var{port} must be a decimal number. The @var{host} could be
17828 the target machine itself, if it is directly connected to the net, or
17829 it might be a terminal server which in turn has a serial line to the
17832 For example, to connect to port 2828 on a terminal server named
17836 target remote manyfarms:2828
17839 If your remote target is actually running on the same machine as your
17840 debugger session (e.g.@: a simulator for your target running on the
17841 same host), you can omit the hostname. For example, to connect to
17842 port 1234 on your local machine:
17845 target remote :1234
17849 Note that the colon is still required here.
17851 @item target remote @code{udp:@var{host}:@var{port}}
17852 @cindex @acronym{UDP} port, @code{target remote}
17853 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
17854 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
17857 target remote udp:manyfarms:2828
17860 When using a @acronym{UDP} connection for remote debugging, you should
17861 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
17862 can silently drop packets on busy or unreliable networks, which will
17863 cause havoc with your debugging session.
17865 @item target remote | @var{command}
17866 @cindex pipe, @code{target remote} to
17867 Run @var{command} in the background and communicate with it using a
17868 pipe. The @var{command} is a shell command, to be parsed and expanded
17869 by the system's command shell, @code{/bin/sh}; it should expect remote
17870 protocol packets on its standard input, and send replies on its
17871 standard output. You could use this to run a stand-alone simulator
17872 that speaks the remote debugging protocol, to make net connections
17873 using programs like @code{ssh}, or for other similar tricks.
17875 If @var{command} closes its standard output (perhaps by exiting),
17876 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
17877 program has already exited, this will have no effect.)
17881 Once the connection has been established, you can use all the usual
17882 commands to examine and change data. The remote program is already
17883 running; you can use @kbd{step} and @kbd{continue}, and you do not
17884 need to use @kbd{run}.
17886 @cindex interrupting remote programs
17887 @cindex remote programs, interrupting
17888 Whenever @value{GDBN} is waiting for the remote program, if you type the
17889 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
17890 program. This may or may not succeed, depending in part on the hardware
17891 and the serial drivers the remote system uses. If you type the
17892 interrupt character once again, @value{GDBN} displays this prompt:
17895 Interrupted while waiting for the program.
17896 Give up (and stop debugging it)? (y or n)
17899 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
17900 (If you decide you want to try again later, you can use @samp{target
17901 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
17902 goes back to waiting.
17905 @kindex detach (remote)
17907 When you have finished debugging the remote program, you can use the
17908 @code{detach} command to release it from @value{GDBN} control.
17909 Detaching from the target normally resumes its execution, but the results
17910 will depend on your particular remote stub. After the @code{detach}
17911 command, @value{GDBN} is free to connect to another target.
17915 The @code{disconnect} command behaves like @code{detach}, except that
17916 the target is generally not resumed. It will wait for @value{GDBN}
17917 (this instance or another one) to connect and continue debugging. After
17918 the @code{disconnect} command, @value{GDBN} is again free to connect to
17921 @cindex send command to remote monitor
17922 @cindex extend @value{GDBN} for remote targets
17923 @cindex add new commands for external monitor
17925 @item monitor @var{cmd}
17926 This command allows you to send arbitrary commands directly to the
17927 remote monitor. Since @value{GDBN} doesn't care about the commands it
17928 sends like this, this command is the way to extend @value{GDBN}---you
17929 can add new commands that only the external monitor will understand
17933 @node File Transfer
17934 @section Sending files to a remote system
17935 @cindex remote target, file transfer
17936 @cindex file transfer
17937 @cindex sending files to remote systems
17939 Some remote targets offer the ability to transfer files over the same
17940 connection used to communicate with @value{GDBN}. This is convenient
17941 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
17942 running @code{gdbserver} over a network interface. For other targets,
17943 e.g.@: embedded devices with only a single serial port, this may be
17944 the only way to upload or download files.
17946 Not all remote targets support these commands.
17950 @item remote put @var{hostfile} @var{targetfile}
17951 Copy file @var{hostfile} from the host system (the machine running
17952 @value{GDBN}) to @var{targetfile} on the target system.
17955 @item remote get @var{targetfile} @var{hostfile}
17956 Copy file @var{targetfile} from the target system to @var{hostfile}
17957 on the host system.
17959 @kindex remote delete
17960 @item remote delete @var{targetfile}
17961 Delete @var{targetfile} from the target system.
17966 @section Using the @code{gdbserver} Program
17969 @cindex remote connection without stubs
17970 @code{gdbserver} is a control program for Unix-like systems, which
17971 allows you to connect your program with a remote @value{GDBN} via
17972 @code{target remote}---but without linking in the usual debugging stub.
17974 @code{gdbserver} is not a complete replacement for the debugging stubs,
17975 because it requires essentially the same operating-system facilities
17976 that @value{GDBN} itself does. In fact, a system that can run
17977 @code{gdbserver} to connect to a remote @value{GDBN} could also run
17978 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
17979 because it is a much smaller program than @value{GDBN} itself. It is
17980 also easier to port than all of @value{GDBN}, so you may be able to get
17981 started more quickly on a new system by using @code{gdbserver}.
17982 Finally, if you develop code for real-time systems, you may find that
17983 the tradeoffs involved in real-time operation make it more convenient to
17984 do as much development work as possible on another system, for example
17985 by cross-compiling. You can use @code{gdbserver} to make a similar
17986 choice for debugging.
17988 @value{GDBN} and @code{gdbserver} communicate via either a serial line
17989 or a TCP connection, using the standard @value{GDBN} remote serial
17993 @emph{Warning:} @code{gdbserver} does not have any built-in security.
17994 Do not run @code{gdbserver} connected to any public network; a
17995 @value{GDBN} connection to @code{gdbserver} provides access to the
17996 target system with the same privileges as the user running
18000 @subsection Running @code{gdbserver}
18001 @cindex arguments, to @code{gdbserver}
18002 @cindex @code{gdbserver}, command-line arguments
18004 Run @code{gdbserver} on the target system. You need a copy of the
18005 program you want to debug, including any libraries it requires.
18006 @code{gdbserver} does not need your program's symbol table, so you can
18007 strip the program if necessary to save space. @value{GDBN} on the host
18008 system does all the symbol handling.
18010 To use the server, you must tell it how to communicate with @value{GDBN};
18011 the name of your program; and the arguments for your program. The usual
18015 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
18018 @var{comm} is either a device name (to use a serial line), or a TCP
18019 hostname and portnumber, or @code{-} or @code{stdio} to use
18020 stdin/stdout of @code{gdbserver}.
18021 For example, to debug Emacs with the argument
18022 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
18026 target> gdbserver /dev/com1 emacs foo.txt
18029 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
18032 To use a TCP connection instead of a serial line:
18035 target> gdbserver host:2345 emacs foo.txt
18038 The only difference from the previous example is the first argument,
18039 specifying that you are communicating with the host @value{GDBN} via
18040 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
18041 expect a TCP connection from machine @samp{host} to local TCP port 2345.
18042 (Currently, the @samp{host} part is ignored.) You can choose any number
18043 you want for the port number as long as it does not conflict with any
18044 TCP ports already in use on the target system (for example, @code{23} is
18045 reserved for @code{telnet}).@footnote{If you choose a port number that
18046 conflicts with another service, @code{gdbserver} prints an error message
18047 and exits.} You must use the same port number with the host @value{GDBN}
18048 @code{target remote} command.
18050 The @code{stdio} connection is useful when starting @code{gdbserver}
18054 (gdb) target remote | ssh -T hostname gdbserver - hello
18057 The @samp{-T} option to ssh is provided because we don't need a remote pty,
18058 and we don't want escape-character handling. Ssh does this by default when
18059 a command is provided, the flag is provided to make it explicit.
18060 You could elide it if you want to.
18062 Programs started with stdio-connected gdbserver have @file{/dev/null} for
18063 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
18064 display through a pipe connected to gdbserver.
18065 Both @code{stdout} and @code{stderr} use the same pipe.
18067 @subsubsection Attaching to a Running Program
18068 @cindex attach to a program, @code{gdbserver}
18069 @cindex @option{--attach}, @code{gdbserver} option
18071 On some targets, @code{gdbserver} can also attach to running programs.
18072 This is accomplished via the @code{--attach} argument. The syntax is:
18075 target> gdbserver --attach @var{comm} @var{pid}
18078 @var{pid} is the process ID of a currently running process. It isn't necessary
18079 to point @code{gdbserver} at a binary for the running process.
18082 You can debug processes by name instead of process ID if your target has the
18083 @code{pidof} utility:
18086 target> gdbserver --attach @var{comm} `pidof @var{program}`
18089 In case more than one copy of @var{program} is running, or @var{program}
18090 has multiple threads, most versions of @code{pidof} support the
18091 @code{-s} option to only return the first process ID.
18093 @subsubsection Multi-Process Mode for @code{gdbserver}
18094 @cindex @code{gdbserver}, multiple processes
18095 @cindex multiple processes with @code{gdbserver}
18097 When you connect to @code{gdbserver} using @code{target remote},
18098 @code{gdbserver} debugs the specified program only once. When the
18099 program exits, or you detach from it, @value{GDBN} closes the connection
18100 and @code{gdbserver} exits.
18102 If you connect using @kbd{target extended-remote}, @code{gdbserver}
18103 enters multi-process mode. When the debugged program exits, or you
18104 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
18105 though no program is running. The @code{run} and @code{attach}
18106 commands instruct @code{gdbserver} to run or attach to a new program.
18107 The @code{run} command uses @code{set remote exec-file} (@pxref{set
18108 remote exec-file}) to select the program to run. Command line
18109 arguments are supported, except for wildcard expansion and I/O
18110 redirection (@pxref{Arguments}).
18112 @cindex @option{--multi}, @code{gdbserver} option
18113 To start @code{gdbserver} without supplying an initial command to run
18114 or process ID to attach, use the @option{--multi} command line option.
18115 Then you can connect using @kbd{target extended-remote} and start
18116 the program you want to debug.
18118 In multi-process mode @code{gdbserver} does not automatically exit unless you
18119 use the option @option{--once}. You can terminate it by using
18120 @code{monitor exit} (@pxref{Monitor Commands for gdbserver}). Note that the
18121 conditions under which @code{gdbserver} terminates depend on how @value{GDBN}
18122 connects to it (@kbd{target remote} or @kbd{target extended-remote}). The
18123 @option{--multi} option to @code{gdbserver} has no influence on that.
18125 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
18127 This section applies only when @code{gdbserver} is run to listen on a TCP port.
18129 @code{gdbserver} normally terminates after all of its debugged processes have
18130 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
18131 extended-remote}, @code{gdbserver} stays running even with no processes left.
18132 @value{GDBN} normally terminates the spawned debugged process on its exit,
18133 which normally also terminates @code{gdbserver} in the @kbd{target remote}
18134 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
18135 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
18136 stays running even in the @kbd{target remote} mode.
18138 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
18139 Such reconnecting is useful for features like @ref{disconnected tracing}. For
18140 completeness, at most one @value{GDBN} can be connected at a time.
18142 @cindex @option{--once}, @code{gdbserver} option
18143 By default, @code{gdbserver} keeps the listening TCP port open, so that
18144 additional connections are possible. However, if you start @code{gdbserver}
18145 with the @option{--once} option, it will stop listening for any further
18146 connection attempts after connecting to the first @value{GDBN} session. This
18147 means no further connections to @code{gdbserver} will be possible after the
18148 first one. It also means @code{gdbserver} will terminate after the first
18149 connection with remote @value{GDBN} has closed, even for unexpectedly closed
18150 connections and even in the @kbd{target extended-remote} mode. The
18151 @option{--once} option allows reusing the same port number for connecting to
18152 multiple instances of @code{gdbserver} running on the same host, since each
18153 instance closes its port after the first connection.
18155 @subsubsection Other Command-Line Arguments for @code{gdbserver}
18157 @cindex @option{--debug}, @code{gdbserver} option
18158 The @option{--debug} option tells @code{gdbserver} to display extra
18159 status information about the debugging process.
18160 @cindex @option{--remote-debug}, @code{gdbserver} option
18161 The @option{--remote-debug} option tells @code{gdbserver} to display
18162 remote protocol debug output. These options are intended for
18163 @code{gdbserver} development and for bug reports to the developers.
18165 @cindex @option{--wrapper}, @code{gdbserver} option
18166 The @option{--wrapper} option specifies a wrapper to launch programs
18167 for debugging. The option should be followed by the name of the
18168 wrapper, then any command-line arguments to pass to the wrapper, then
18169 @kbd{--} indicating the end of the wrapper arguments.
18171 @code{gdbserver} runs the specified wrapper program with a combined
18172 command line including the wrapper arguments, then the name of the
18173 program to debug, then any arguments to the program. The wrapper
18174 runs until it executes your program, and then @value{GDBN} gains control.
18176 You can use any program that eventually calls @code{execve} with
18177 its arguments as a wrapper. Several standard Unix utilities do
18178 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
18179 with @code{exec "$@@"} will also work.
18181 For example, you can use @code{env} to pass an environment variable to
18182 the debugged program, without setting the variable in @code{gdbserver}'s
18186 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
18189 @subsection Connecting to @code{gdbserver}
18191 Run @value{GDBN} on the host system.
18193 First make sure you have the necessary symbol files. Load symbols for
18194 your application using the @code{file} command before you connect. Use
18195 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
18196 was compiled with the correct sysroot using @code{--with-sysroot}).
18198 The symbol file and target libraries must exactly match the executable
18199 and libraries on the target, with one exception: the files on the host
18200 system should not be stripped, even if the files on the target system
18201 are. Mismatched or missing files will lead to confusing results
18202 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
18203 files may also prevent @code{gdbserver} from debugging multi-threaded
18206 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
18207 For TCP connections, you must start up @code{gdbserver} prior to using
18208 the @code{target remote} command. Otherwise you may get an error whose
18209 text depends on the host system, but which usually looks something like
18210 @samp{Connection refused}. Don't use the @code{load}
18211 command in @value{GDBN} when using @code{gdbserver}, since the program is
18212 already on the target.
18214 @subsection Monitor Commands for @code{gdbserver}
18215 @cindex monitor commands, for @code{gdbserver}
18216 @anchor{Monitor Commands for gdbserver}
18218 During a @value{GDBN} session using @code{gdbserver}, you can use the
18219 @code{monitor} command to send special requests to @code{gdbserver}.
18220 Here are the available commands.
18224 List the available monitor commands.
18226 @item monitor set debug 0
18227 @itemx monitor set debug 1
18228 Disable or enable general debugging messages.
18230 @item monitor set remote-debug 0
18231 @itemx monitor set remote-debug 1
18232 Disable or enable specific debugging messages associated with the remote
18233 protocol (@pxref{Remote Protocol}).
18235 @item monitor set libthread-db-search-path [PATH]
18236 @cindex gdbserver, search path for @code{libthread_db}
18237 When this command is issued, @var{path} is a colon-separated list of
18238 directories to search for @code{libthread_db} (@pxref{Threads,,set
18239 libthread-db-search-path}). If you omit @var{path},
18240 @samp{libthread-db-search-path} will be reset to its default value.
18242 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
18243 not supported in @code{gdbserver}.
18246 Tell gdbserver to exit immediately. This command should be followed by
18247 @code{disconnect} to close the debugging session. @code{gdbserver} will
18248 detach from any attached processes and kill any processes it created.
18249 Use @code{monitor exit} to terminate @code{gdbserver} at the end
18250 of a multi-process mode debug session.
18254 @subsection Tracepoints support in @code{gdbserver}
18255 @cindex tracepoints support in @code{gdbserver}
18257 On some targets, @code{gdbserver} supports tracepoints, fast
18258 tracepoints and static tracepoints.
18260 For fast or static tracepoints to work, a special library called the
18261 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
18262 This library is built and distributed as an integral part of
18263 @code{gdbserver}. In addition, support for static tracepoints
18264 requires building the in-process agent library with static tracepoints
18265 support. At present, the UST (LTTng Userspace Tracer,
18266 @url{http://lttng.org/ust}) tracing engine is supported. This support
18267 is automatically available if UST development headers are found in the
18268 standard include path when @code{gdbserver} is built, or if
18269 @code{gdbserver} was explicitly configured using @option{--with-ust}
18270 to point at such headers. You can explicitly disable the support
18271 using @option{--with-ust=no}.
18273 There are several ways to load the in-process agent in your program:
18276 @item Specifying it as dependency at link time
18278 You can link your program dynamically with the in-process agent
18279 library. On most systems, this is accomplished by adding
18280 @code{-linproctrace} to the link command.
18282 @item Using the system's preloading mechanisms
18284 You can force loading the in-process agent at startup time by using
18285 your system's support for preloading shared libraries. Many Unixes
18286 support the concept of preloading user defined libraries. In most
18287 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
18288 in the environment. See also the description of @code{gdbserver}'s
18289 @option{--wrapper} command line option.
18291 @item Using @value{GDBN} to force loading the agent at run time
18293 On some systems, you can force the inferior to load a shared library,
18294 by calling a dynamic loader function in the inferior that takes care
18295 of dynamically looking up and loading a shared library. On most Unix
18296 systems, the function is @code{dlopen}. You'll use the @code{call}
18297 command for that. For example:
18300 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
18303 Note that on most Unix systems, for the @code{dlopen} function to be
18304 available, the program needs to be linked with @code{-ldl}.
18307 On systems that have a userspace dynamic loader, like most Unix
18308 systems, when you connect to @code{gdbserver} using @code{target
18309 remote}, you'll find that the program is stopped at the dynamic
18310 loader's entry point, and no shared library has been loaded in the
18311 program's address space yet, including the in-process agent. In that
18312 case, before being able to use any of the fast or static tracepoints
18313 features, you need to let the loader run and load the shared
18314 libraries. The simplest way to do that is to run the program to the
18315 main procedure. E.g., if debugging a C or C@t{++} program, start
18316 @code{gdbserver} like so:
18319 $ gdbserver :9999 myprogram
18322 Start GDB and connect to @code{gdbserver} like so, and run to main:
18326 (@value{GDBP}) target remote myhost:9999
18327 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
18328 (@value{GDBP}) b main
18329 (@value{GDBP}) continue
18332 The in-process tracing agent library should now be loaded into the
18333 process; you can confirm it with the @code{info sharedlibrary}
18334 command, which will list @file{libinproctrace.so} as loaded in the
18335 process. You are now ready to install fast tracepoints, list static
18336 tracepoint markers, probe static tracepoints markers, and start
18339 @node Remote Configuration
18340 @section Remote Configuration
18343 @kindex show remote
18344 This section documents the configuration options available when
18345 debugging remote programs. For the options related to the File I/O
18346 extensions of the remote protocol, see @ref{system,
18347 system-call-allowed}.
18350 @item set remoteaddresssize @var{bits}
18351 @cindex address size for remote targets
18352 @cindex bits in remote address
18353 Set the maximum size of address in a memory packet to the specified
18354 number of bits. @value{GDBN} will mask off the address bits above
18355 that number, when it passes addresses to the remote target. The
18356 default value is the number of bits in the target's address.
18358 @item show remoteaddresssize
18359 Show the current value of remote address size in bits.
18361 @item set remotebaud @var{n}
18362 @cindex baud rate for remote targets
18363 Set the baud rate for the remote serial I/O to @var{n} baud. The
18364 value is used to set the speed of the serial port used for debugging
18367 @item show remotebaud
18368 Show the current speed of the remote connection.
18370 @item set remotebreak
18371 @cindex interrupt remote programs
18372 @cindex BREAK signal instead of Ctrl-C
18373 @anchor{set remotebreak}
18374 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
18375 when you type @kbd{Ctrl-c} to interrupt the program running
18376 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
18377 character instead. The default is off, since most remote systems
18378 expect to see @samp{Ctrl-C} as the interrupt signal.
18380 @item show remotebreak
18381 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
18382 interrupt the remote program.
18384 @item set remoteflow on
18385 @itemx set remoteflow off
18386 @kindex set remoteflow
18387 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
18388 on the serial port used to communicate to the remote target.
18390 @item show remoteflow
18391 @kindex show remoteflow
18392 Show the current setting of hardware flow control.
18394 @item set remotelogbase @var{base}
18395 Set the base (a.k.a.@: radix) of logging serial protocol
18396 communications to @var{base}. Supported values of @var{base} are:
18397 @code{ascii}, @code{octal}, and @code{hex}. The default is
18400 @item show remotelogbase
18401 Show the current setting of the radix for logging remote serial
18404 @item set remotelogfile @var{file}
18405 @cindex record serial communications on file
18406 Record remote serial communications on the named @var{file}. The
18407 default is not to record at all.
18409 @item show remotelogfile.
18410 Show the current setting of the file name on which to record the
18411 serial communications.
18413 @item set remotetimeout @var{num}
18414 @cindex timeout for serial communications
18415 @cindex remote timeout
18416 Set the timeout limit to wait for the remote target to respond to
18417 @var{num} seconds. The default is 2 seconds.
18419 @item show remotetimeout
18420 Show the current number of seconds to wait for the remote target
18423 @cindex limit hardware breakpoints and watchpoints
18424 @cindex remote target, limit break- and watchpoints
18425 @anchor{set remote hardware-watchpoint-limit}
18426 @anchor{set remote hardware-breakpoint-limit}
18427 @item set remote hardware-watchpoint-limit @var{limit}
18428 @itemx set remote hardware-breakpoint-limit @var{limit}
18429 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
18430 watchpoints. A limit of -1, the default, is treated as unlimited.
18432 @cindex limit hardware watchpoints length
18433 @cindex remote target, limit watchpoints length
18434 @anchor{set remote hardware-watchpoint-length-limit}
18435 @item set remote hardware-watchpoint-length-limit @var{limit}
18436 Restrict @value{GDBN} to using @var{limit} bytes for the maximum length of
18437 a remote hardware watchpoint. A limit of -1, the default, is treated
18440 @item show remote hardware-watchpoint-length-limit
18441 Show the current limit (in bytes) of the maximum length of
18442 a remote hardware watchpoint.
18444 @item set remote exec-file @var{filename}
18445 @itemx show remote exec-file
18446 @anchor{set remote exec-file}
18447 @cindex executable file, for remote target
18448 Select the file used for @code{run} with @code{target
18449 extended-remote}. This should be set to a filename valid on the
18450 target system. If it is not set, the target will use a default
18451 filename (e.g.@: the last program run).
18453 @item set remote interrupt-sequence
18454 @cindex interrupt remote programs
18455 @cindex select Ctrl-C, BREAK or BREAK-g
18456 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
18457 @samp{BREAK-g} as the
18458 sequence to the remote target in order to interrupt the execution.
18459 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
18460 is high level of serial line for some certain time.
18461 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
18462 It is @code{BREAK} signal followed by character @code{g}.
18464 @item show interrupt-sequence
18465 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
18466 is sent by @value{GDBN} to interrupt the remote program.
18467 @code{BREAK-g} is BREAK signal followed by @code{g} and
18468 also known as Magic SysRq g.
18470 @item set remote interrupt-on-connect
18471 @cindex send interrupt-sequence on start
18472 Specify whether interrupt-sequence is sent to remote target when
18473 @value{GDBN} connects to it. This is mostly needed when you debug
18474 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
18475 which is known as Magic SysRq g in order to connect @value{GDBN}.
18477 @item show interrupt-on-connect
18478 Show whether interrupt-sequence is sent
18479 to remote target when @value{GDBN} connects to it.
18483 @item set tcp auto-retry on
18484 @cindex auto-retry, for remote TCP target
18485 Enable auto-retry for remote TCP connections. This is useful if the remote
18486 debugging agent is launched in parallel with @value{GDBN}; there is a race
18487 condition because the agent may not become ready to accept the connection
18488 before @value{GDBN} attempts to connect. When auto-retry is
18489 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
18490 to establish the connection using the timeout specified by
18491 @code{set tcp connect-timeout}.
18493 @item set tcp auto-retry off
18494 Do not auto-retry failed TCP connections.
18496 @item show tcp auto-retry
18497 Show the current auto-retry setting.
18499 @item set tcp connect-timeout @var{seconds}
18500 @itemx set tcp connect-timeout unlimited
18501 @cindex connection timeout, for remote TCP target
18502 @cindex timeout, for remote target connection
18503 Set the timeout for establishing a TCP connection to the remote target to
18504 @var{seconds}. The timeout affects both polling to retry failed connections
18505 (enabled by @code{set tcp auto-retry on}) and waiting for connections
18506 that are merely slow to complete, and represents an approximate cumulative
18507 value. If @var{seconds} is @code{unlimited}, there is no timeout and
18508 @value{GDBN} will keep attempting to establish a connection forever,
18509 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
18511 @item show tcp connect-timeout
18512 Show the current connection timeout setting.
18515 @cindex remote packets, enabling and disabling
18516 The @value{GDBN} remote protocol autodetects the packets supported by
18517 your debugging stub. If you need to override the autodetection, you
18518 can use these commands to enable or disable individual packets. Each
18519 packet can be set to @samp{on} (the remote target supports this
18520 packet), @samp{off} (the remote target does not support this packet),
18521 or @samp{auto} (detect remote target support for this packet). They
18522 all default to @samp{auto}. For more information about each packet,
18523 see @ref{Remote Protocol}.
18525 During normal use, you should not have to use any of these commands.
18526 If you do, that may be a bug in your remote debugging stub, or a bug
18527 in @value{GDBN}. You may want to report the problem to the
18528 @value{GDBN} developers.
18530 For each packet @var{name}, the command to enable or disable the
18531 packet is @code{set remote @var{name}-packet}. The available settings
18534 @multitable @columnfractions 0.28 0.32 0.25
18537 @tab Related Features
18539 @item @code{fetch-register}
18541 @tab @code{info registers}
18543 @item @code{set-register}
18547 @item @code{binary-download}
18549 @tab @code{load}, @code{set}
18551 @item @code{read-aux-vector}
18552 @tab @code{qXfer:auxv:read}
18553 @tab @code{info auxv}
18555 @item @code{symbol-lookup}
18556 @tab @code{qSymbol}
18557 @tab Detecting multiple threads
18559 @item @code{attach}
18560 @tab @code{vAttach}
18563 @item @code{verbose-resume}
18565 @tab Stepping or resuming multiple threads
18571 @item @code{software-breakpoint}
18575 @item @code{hardware-breakpoint}
18579 @item @code{write-watchpoint}
18583 @item @code{read-watchpoint}
18587 @item @code{access-watchpoint}
18591 @item @code{target-features}
18592 @tab @code{qXfer:features:read}
18593 @tab @code{set architecture}
18595 @item @code{library-info}
18596 @tab @code{qXfer:libraries:read}
18597 @tab @code{info sharedlibrary}
18599 @item @code{memory-map}
18600 @tab @code{qXfer:memory-map:read}
18601 @tab @code{info mem}
18603 @item @code{read-sdata-object}
18604 @tab @code{qXfer:sdata:read}
18605 @tab @code{print $_sdata}
18607 @item @code{read-spu-object}
18608 @tab @code{qXfer:spu:read}
18609 @tab @code{info spu}
18611 @item @code{write-spu-object}
18612 @tab @code{qXfer:spu:write}
18613 @tab @code{info spu}
18615 @item @code{read-siginfo-object}
18616 @tab @code{qXfer:siginfo:read}
18617 @tab @code{print $_siginfo}
18619 @item @code{write-siginfo-object}
18620 @tab @code{qXfer:siginfo:write}
18621 @tab @code{set $_siginfo}
18623 @item @code{threads}
18624 @tab @code{qXfer:threads:read}
18625 @tab @code{info threads}
18627 @item @code{get-thread-local-@*storage-address}
18628 @tab @code{qGetTLSAddr}
18629 @tab Displaying @code{__thread} variables
18631 @item @code{get-thread-information-block-address}
18632 @tab @code{qGetTIBAddr}
18633 @tab Display MS-Windows Thread Information Block.
18635 @item @code{search-memory}
18636 @tab @code{qSearch:memory}
18639 @item @code{supported-packets}
18640 @tab @code{qSupported}
18641 @tab Remote communications parameters
18643 @item @code{pass-signals}
18644 @tab @code{QPassSignals}
18645 @tab @code{handle @var{signal}}
18647 @item @code{program-signals}
18648 @tab @code{QProgramSignals}
18649 @tab @code{handle @var{signal}}
18651 @item @code{hostio-close-packet}
18652 @tab @code{vFile:close}
18653 @tab @code{remote get}, @code{remote put}
18655 @item @code{hostio-open-packet}
18656 @tab @code{vFile:open}
18657 @tab @code{remote get}, @code{remote put}
18659 @item @code{hostio-pread-packet}
18660 @tab @code{vFile:pread}
18661 @tab @code{remote get}, @code{remote put}
18663 @item @code{hostio-pwrite-packet}
18664 @tab @code{vFile:pwrite}
18665 @tab @code{remote get}, @code{remote put}
18667 @item @code{hostio-unlink-packet}
18668 @tab @code{vFile:unlink}
18669 @tab @code{remote delete}
18671 @item @code{hostio-readlink-packet}
18672 @tab @code{vFile:readlink}
18675 @item @code{noack-packet}
18676 @tab @code{QStartNoAckMode}
18677 @tab Packet acknowledgment
18679 @item @code{osdata}
18680 @tab @code{qXfer:osdata:read}
18681 @tab @code{info os}
18683 @item @code{query-attached}
18684 @tab @code{qAttached}
18685 @tab Querying remote process attach state.
18687 @item @code{trace-buffer-size}
18688 @tab @code{QTBuffer:size}
18689 @tab @code{set trace-buffer-size}
18691 @item @code{trace-status}
18692 @tab @code{qTStatus}
18693 @tab @code{tstatus}
18695 @item @code{traceframe-info}
18696 @tab @code{qXfer:traceframe-info:read}
18697 @tab Traceframe info
18699 @item @code{install-in-trace}
18700 @tab @code{InstallInTrace}
18701 @tab Install tracepoint in tracing
18703 @item @code{disable-randomization}
18704 @tab @code{QDisableRandomization}
18705 @tab @code{set disable-randomization}
18707 @item @code{conditional-breakpoints-packet}
18708 @tab @code{Z0 and Z1}
18709 @tab @code{Support for target-side breakpoint condition evaluation}
18713 @section Implementing a Remote Stub
18715 @cindex debugging stub, example
18716 @cindex remote stub, example
18717 @cindex stub example, remote debugging
18718 The stub files provided with @value{GDBN} implement the target side of the
18719 communication protocol, and the @value{GDBN} side is implemented in the
18720 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
18721 these subroutines to communicate, and ignore the details. (If you're
18722 implementing your own stub file, you can still ignore the details: start
18723 with one of the existing stub files. @file{sparc-stub.c} is the best
18724 organized, and therefore the easiest to read.)
18726 @cindex remote serial debugging, overview
18727 To debug a program running on another machine (the debugging
18728 @dfn{target} machine), you must first arrange for all the usual
18729 prerequisites for the program to run by itself. For example, for a C
18734 A startup routine to set up the C runtime environment; these usually
18735 have a name like @file{crt0}. The startup routine may be supplied by
18736 your hardware supplier, or you may have to write your own.
18739 A C subroutine library to support your program's
18740 subroutine calls, notably managing input and output.
18743 A way of getting your program to the other machine---for example, a
18744 download program. These are often supplied by the hardware
18745 manufacturer, but you may have to write your own from hardware
18749 The next step is to arrange for your program to use a serial port to
18750 communicate with the machine where @value{GDBN} is running (the @dfn{host}
18751 machine). In general terms, the scheme looks like this:
18755 @value{GDBN} already understands how to use this protocol; when everything
18756 else is set up, you can simply use the @samp{target remote} command
18757 (@pxref{Targets,,Specifying a Debugging Target}).
18759 @item On the target,
18760 you must link with your program a few special-purpose subroutines that
18761 implement the @value{GDBN} remote serial protocol. The file containing these
18762 subroutines is called a @dfn{debugging stub}.
18764 On certain remote targets, you can use an auxiliary program
18765 @code{gdbserver} instead of linking a stub into your program.
18766 @xref{Server,,Using the @code{gdbserver} Program}, for details.
18769 The debugging stub is specific to the architecture of the remote
18770 machine; for example, use @file{sparc-stub.c} to debug programs on
18773 @cindex remote serial stub list
18774 These working remote stubs are distributed with @value{GDBN}:
18779 @cindex @file{i386-stub.c}
18782 For Intel 386 and compatible architectures.
18785 @cindex @file{m68k-stub.c}
18786 @cindex Motorola 680x0
18788 For Motorola 680x0 architectures.
18791 @cindex @file{sh-stub.c}
18794 For Renesas SH architectures.
18797 @cindex @file{sparc-stub.c}
18799 For @sc{sparc} architectures.
18801 @item sparcl-stub.c
18802 @cindex @file{sparcl-stub.c}
18805 For Fujitsu @sc{sparclite} architectures.
18809 The @file{README} file in the @value{GDBN} distribution may list other
18810 recently added stubs.
18813 * Stub Contents:: What the stub can do for you
18814 * Bootstrapping:: What you must do for the stub
18815 * Debug Session:: Putting it all together
18818 @node Stub Contents
18819 @subsection What the Stub Can Do for You
18821 @cindex remote serial stub
18822 The debugging stub for your architecture supplies these three
18826 @item set_debug_traps
18827 @findex set_debug_traps
18828 @cindex remote serial stub, initialization
18829 This routine arranges for @code{handle_exception} to run when your
18830 program stops. You must call this subroutine explicitly in your
18831 program's startup code.
18833 @item handle_exception
18834 @findex handle_exception
18835 @cindex remote serial stub, main routine
18836 This is the central workhorse, but your program never calls it
18837 explicitly---the setup code arranges for @code{handle_exception} to
18838 run when a trap is triggered.
18840 @code{handle_exception} takes control when your program stops during
18841 execution (for example, on a breakpoint), and mediates communications
18842 with @value{GDBN} on the host machine. This is where the communications
18843 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
18844 representative on the target machine. It begins by sending summary
18845 information on the state of your program, then continues to execute,
18846 retrieving and transmitting any information @value{GDBN} needs, until you
18847 execute a @value{GDBN} command that makes your program resume; at that point,
18848 @code{handle_exception} returns control to your own code on the target
18852 @cindex @code{breakpoint} subroutine, remote
18853 Use this auxiliary subroutine to make your program contain a
18854 breakpoint. Depending on the particular situation, this may be the only
18855 way for @value{GDBN} to get control. For instance, if your target
18856 machine has some sort of interrupt button, you won't need to call this;
18857 pressing the interrupt button transfers control to
18858 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
18859 simply receiving characters on the serial port may also trigger a trap;
18860 again, in that situation, you don't need to call @code{breakpoint} from
18861 your own program---simply running @samp{target remote} from the host
18862 @value{GDBN} session gets control.
18864 Call @code{breakpoint} if none of these is true, or if you simply want
18865 to make certain your program stops at a predetermined point for the
18866 start of your debugging session.
18869 @node Bootstrapping
18870 @subsection What You Must Do for the Stub
18872 @cindex remote stub, support routines
18873 The debugging stubs that come with @value{GDBN} are set up for a particular
18874 chip architecture, but they have no information about the rest of your
18875 debugging target machine.
18877 First of all you need to tell the stub how to communicate with the
18881 @item int getDebugChar()
18882 @findex getDebugChar
18883 Write this subroutine to read a single character from the serial port.
18884 It may be identical to @code{getchar} for your target system; a
18885 different name is used to allow you to distinguish the two if you wish.
18887 @item void putDebugChar(int)
18888 @findex putDebugChar
18889 Write this subroutine to write a single character to the serial port.
18890 It may be identical to @code{putchar} for your target system; a
18891 different name is used to allow you to distinguish the two if you wish.
18894 @cindex control C, and remote debugging
18895 @cindex interrupting remote targets
18896 If you want @value{GDBN} to be able to stop your program while it is
18897 running, you need to use an interrupt-driven serial driver, and arrange
18898 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
18899 character). That is the character which @value{GDBN} uses to tell the
18900 remote system to stop.
18902 Getting the debugging target to return the proper status to @value{GDBN}
18903 probably requires changes to the standard stub; one quick and dirty way
18904 is to just execute a breakpoint instruction (the ``dirty'' part is that
18905 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
18907 Other routines you need to supply are:
18910 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
18911 @findex exceptionHandler
18912 Write this function to install @var{exception_address} in the exception
18913 handling tables. You need to do this because the stub does not have any
18914 way of knowing what the exception handling tables on your target system
18915 are like (for example, the processor's table might be in @sc{rom},
18916 containing entries which point to a table in @sc{ram}).
18917 @var{exception_number} is the exception number which should be changed;
18918 its meaning is architecture-dependent (for example, different numbers
18919 might represent divide by zero, misaligned access, etc). When this
18920 exception occurs, control should be transferred directly to
18921 @var{exception_address}, and the processor state (stack, registers,
18922 and so on) should be just as it is when a processor exception occurs. So if
18923 you want to use a jump instruction to reach @var{exception_address}, it
18924 should be a simple jump, not a jump to subroutine.
18926 For the 386, @var{exception_address} should be installed as an interrupt
18927 gate so that interrupts are masked while the handler runs. The gate
18928 should be at privilege level 0 (the most privileged level). The
18929 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
18930 help from @code{exceptionHandler}.
18932 @item void flush_i_cache()
18933 @findex flush_i_cache
18934 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
18935 instruction cache, if any, on your target machine. If there is no
18936 instruction cache, this subroutine may be a no-op.
18938 On target machines that have instruction caches, @value{GDBN} requires this
18939 function to make certain that the state of your program is stable.
18943 You must also make sure this library routine is available:
18946 @item void *memset(void *, int, int)
18948 This is the standard library function @code{memset} that sets an area of
18949 memory to a known value. If you have one of the free versions of
18950 @code{libc.a}, @code{memset} can be found there; otherwise, you must
18951 either obtain it from your hardware manufacturer, or write your own.
18954 If you do not use the GNU C compiler, you may need other standard
18955 library subroutines as well; this varies from one stub to another,
18956 but in general the stubs are likely to use any of the common library
18957 subroutines which @code{@value{NGCC}} generates as inline code.
18960 @node Debug Session
18961 @subsection Putting it All Together
18963 @cindex remote serial debugging summary
18964 In summary, when your program is ready to debug, you must follow these
18969 Make sure you have defined the supporting low-level routines
18970 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
18972 @code{getDebugChar}, @code{putDebugChar},
18973 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
18977 Insert these lines in your program's startup code, before the main
18978 procedure is called:
18985 On some machines, when a breakpoint trap is raised, the hardware
18986 automatically makes the PC point to the instruction after the
18987 breakpoint. If your machine doesn't do that, you may need to adjust
18988 @code{handle_exception} to arrange for it to return to the instruction
18989 after the breakpoint on this first invocation, so that your program
18990 doesn't keep hitting the initial breakpoint instead of making
18994 For the 680x0 stub only, you need to provide a variable called
18995 @code{exceptionHook}. Normally you just use:
18998 void (*exceptionHook)() = 0;
19002 but if before calling @code{set_debug_traps}, you set it to point to a
19003 function in your program, that function is called when
19004 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
19005 error). The function indicated by @code{exceptionHook} is called with
19006 one parameter: an @code{int} which is the exception number.
19009 Compile and link together: your program, the @value{GDBN} debugging stub for
19010 your target architecture, and the supporting subroutines.
19013 Make sure you have a serial connection between your target machine and
19014 the @value{GDBN} host, and identify the serial port on the host.
19017 @c The "remote" target now provides a `load' command, so we should
19018 @c document that. FIXME.
19019 Download your program to your target machine (or get it there by
19020 whatever means the manufacturer provides), and start it.
19023 Start @value{GDBN} on the host, and connect to the target
19024 (@pxref{Connecting,,Connecting to a Remote Target}).
19028 @node Configurations
19029 @chapter Configuration-Specific Information
19031 While nearly all @value{GDBN} commands are available for all native and
19032 cross versions of the debugger, there are some exceptions. This chapter
19033 describes things that are only available in certain configurations.
19035 There are three major categories of configurations: native
19036 configurations, where the host and target are the same, embedded
19037 operating system configurations, which are usually the same for several
19038 different processor architectures, and bare embedded processors, which
19039 are quite different from each other.
19044 * Embedded Processors::
19051 This section describes details specific to particular native
19056 * BSD libkvm Interface:: Debugging BSD kernel memory images
19057 * SVR4 Process Information:: SVR4 process information
19058 * DJGPP Native:: Features specific to the DJGPP port
19059 * Cygwin Native:: Features specific to the Cygwin port
19060 * Hurd Native:: Features specific to @sc{gnu} Hurd
19061 * Darwin:: Features specific to Darwin
19067 On HP-UX systems, if you refer to a function or variable name that
19068 begins with a dollar sign, @value{GDBN} searches for a user or system
19069 name first, before it searches for a convenience variable.
19072 @node BSD libkvm Interface
19073 @subsection BSD libkvm Interface
19076 @cindex kernel memory image
19077 @cindex kernel crash dump
19079 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
19080 interface that provides a uniform interface for accessing kernel virtual
19081 memory images, including live systems and crash dumps. @value{GDBN}
19082 uses this interface to allow you to debug live kernels and kernel crash
19083 dumps on many native BSD configurations. This is implemented as a
19084 special @code{kvm} debugging target. For debugging a live system, load
19085 the currently running kernel into @value{GDBN} and connect to the
19089 (@value{GDBP}) @b{target kvm}
19092 For debugging crash dumps, provide the file name of the crash dump as an
19096 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
19099 Once connected to the @code{kvm} target, the following commands are
19105 Set current context from the @dfn{Process Control Block} (PCB) address.
19108 Set current context from proc address. This command isn't available on
19109 modern FreeBSD systems.
19112 @node SVR4 Process Information
19113 @subsection SVR4 Process Information
19115 @cindex examine process image
19116 @cindex process info via @file{/proc}
19118 Many versions of SVR4 and compatible systems provide a facility called
19119 @samp{/proc} that can be used to examine the image of a running
19120 process using file-system subroutines.
19122 If @value{GDBN} is configured for an operating system with this
19123 facility, the command @code{info proc} is available to report
19124 information about the process running your program, or about any
19125 process running on your system. This includes, as of this writing,
19126 @sc{gnu}/Linux, OSF/1 (Digital Unix), Solaris, and Irix, but
19127 not HP-UX, for example.
19129 This command may also work on core files that were created on a system
19130 that has the @samp{/proc} facility.
19136 @itemx info proc @var{process-id}
19137 Summarize available information about any running process. If a
19138 process ID is specified by @var{process-id}, display information about
19139 that process; otherwise display information about the program being
19140 debugged. The summary includes the debugged process ID, the command
19141 line used to invoke it, its current working directory, and its
19142 executable file's absolute file name.
19144 On some systems, @var{process-id} can be of the form
19145 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
19146 within a process. If the optional @var{pid} part is missing, it means
19147 a thread from the process being debugged (the leading @samp{/} still
19148 needs to be present, or else @value{GDBN} will interpret the number as
19149 a process ID rather than a thread ID).
19151 @item info proc cmdline
19152 @cindex info proc cmdline
19153 Show the original command line of the process. This command is
19154 specific to @sc{gnu}/Linux.
19156 @item info proc cwd
19157 @cindex info proc cwd
19158 Show the current working directory of the process. This command is
19159 specific to @sc{gnu}/Linux.
19161 @item info proc exe
19162 @cindex info proc exe
19163 Show the name of executable of the process. This command is specific
19166 @item info proc mappings
19167 @cindex memory address space mappings
19168 Report the memory address space ranges accessible in the program, with
19169 information on whether the process has read, write, or execute access
19170 rights to each range. On @sc{gnu}/Linux systems, each memory range
19171 includes the object file which is mapped to that range, instead of the
19172 memory access rights to that range.
19174 @item info proc stat
19175 @itemx info proc status
19176 @cindex process detailed status information
19177 These subcommands are specific to @sc{gnu}/Linux systems. They show
19178 the process-related information, including the user ID and group ID;
19179 how many threads are there in the process; its virtual memory usage;
19180 the signals that are pending, blocked, and ignored; its TTY; its
19181 consumption of system and user time; its stack size; its @samp{nice}
19182 value; etc. For more information, see the @samp{proc} man page
19183 (type @kbd{man 5 proc} from your shell prompt).
19185 @item info proc all
19186 Show all the information about the process described under all of the
19187 above @code{info proc} subcommands.
19190 @comment These sub-options of 'info proc' were not included when
19191 @comment procfs.c was re-written. Keep their descriptions around
19192 @comment against the day when someone finds the time to put them back in.
19193 @kindex info proc times
19194 @item info proc times
19195 Starting time, user CPU time, and system CPU time for your program and
19198 @kindex info proc id
19200 Report on the process IDs related to your program: its own process ID,
19201 the ID of its parent, the process group ID, and the session ID.
19204 @item set procfs-trace
19205 @kindex set procfs-trace
19206 @cindex @code{procfs} API calls
19207 This command enables and disables tracing of @code{procfs} API calls.
19209 @item show procfs-trace
19210 @kindex show procfs-trace
19211 Show the current state of @code{procfs} API call tracing.
19213 @item set procfs-file @var{file}
19214 @kindex set procfs-file
19215 Tell @value{GDBN} to write @code{procfs} API trace to the named
19216 @var{file}. @value{GDBN} appends the trace info to the previous
19217 contents of the file. The default is to display the trace on the
19220 @item show procfs-file
19221 @kindex show procfs-file
19222 Show the file to which @code{procfs} API trace is written.
19224 @item proc-trace-entry
19225 @itemx proc-trace-exit
19226 @itemx proc-untrace-entry
19227 @itemx proc-untrace-exit
19228 @kindex proc-trace-entry
19229 @kindex proc-trace-exit
19230 @kindex proc-untrace-entry
19231 @kindex proc-untrace-exit
19232 These commands enable and disable tracing of entries into and exits
19233 from the @code{syscall} interface.
19236 @kindex info pidlist
19237 @cindex process list, QNX Neutrino
19238 For QNX Neutrino only, this command displays the list of all the
19239 processes and all the threads within each process.
19242 @kindex info meminfo
19243 @cindex mapinfo list, QNX Neutrino
19244 For QNX Neutrino only, this command displays the list of all mapinfos.
19248 @subsection Features for Debugging @sc{djgpp} Programs
19249 @cindex @sc{djgpp} debugging
19250 @cindex native @sc{djgpp} debugging
19251 @cindex MS-DOS-specific commands
19254 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
19255 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
19256 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
19257 top of real-mode DOS systems and their emulations.
19259 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
19260 defines a few commands specific to the @sc{djgpp} port. This
19261 subsection describes those commands.
19266 This is a prefix of @sc{djgpp}-specific commands which print
19267 information about the target system and important OS structures.
19270 @cindex MS-DOS system info
19271 @cindex free memory information (MS-DOS)
19272 @item info dos sysinfo
19273 This command displays assorted information about the underlying
19274 platform: the CPU type and features, the OS version and flavor, the
19275 DPMI version, and the available conventional and DPMI memory.
19280 @cindex segment descriptor tables
19281 @cindex descriptor tables display
19283 @itemx info dos ldt
19284 @itemx info dos idt
19285 These 3 commands display entries from, respectively, Global, Local,
19286 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
19287 tables are data structures which store a descriptor for each segment
19288 that is currently in use. The segment's selector is an index into a
19289 descriptor table; the table entry for that index holds the
19290 descriptor's base address and limit, and its attributes and access
19293 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
19294 segment (used for both data and the stack), and a DOS segment (which
19295 allows access to DOS/BIOS data structures and absolute addresses in
19296 conventional memory). However, the DPMI host will usually define
19297 additional segments in order to support the DPMI environment.
19299 @cindex garbled pointers
19300 These commands allow to display entries from the descriptor tables.
19301 Without an argument, all entries from the specified table are
19302 displayed. An argument, which should be an integer expression, means
19303 display a single entry whose index is given by the argument. For
19304 example, here's a convenient way to display information about the
19305 debugged program's data segment:
19308 @exdent @code{(@value{GDBP}) info dos ldt $ds}
19309 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
19313 This comes in handy when you want to see whether a pointer is outside
19314 the data segment's limit (i.e.@: @dfn{garbled}).
19316 @cindex page tables display (MS-DOS)
19318 @itemx info dos pte
19319 These two commands display entries from, respectively, the Page
19320 Directory and the Page Tables. Page Directories and Page Tables are
19321 data structures which control how virtual memory addresses are mapped
19322 into physical addresses. A Page Table includes an entry for every
19323 page of memory that is mapped into the program's address space; there
19324 may be several Page Tables, each one holding up to 4096 entries. A
19325 Page Directory has up to 4096 entries, one each for every Page Table
19326 that is currently in use.
19328 Without an argument, @kbd{info dos pde} displays the entire Page
19329 Directory, and @kbd{info dos pte} displays all the entries in all of
19330 the Page Tables. An argument, an integer expression, given to the
19331 @kbd{info dos pde} command means display only that entry from the Page
19332 Directory table. An argument given to the @kbd{info dos pte} command
19333 means display entries from a single Page Table, the one pointed to by
19334 the specified entry in the Page Directory.
19336 @cindex direct memory access (DMA) on MS-DOS
19337 These commands are useful when your program uses @dfn{DMA} (Direct
19338 Memory Access), which needs physical addresses to program the DMA
19341 These commands are supported only with some DPMI servers.
19343 @cindex physical address from linear address
19344 @item info dos address-pte @var{addr}
19345 This command displays the Page Table entry for a specified linear
19346 address. The argument @var{addr} is a linear address which should
19347 already have the appropriate segment's base address added to it,
19348 because this command accepts addresses which may belong to @emph{any}
19349 segment. For example, here's how to display the Page Table entry for
19350 the page where a variable @code{i} is stored:
19353 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
19354 @exdent @code{Page Table entry for address 0x11a00d30:}
19355 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
19359 This says that @code{i} is stored at offset @code{0xd30} from the page
19360 whose physical base address is @code{0x02698000}, and shows all the
19361 attributes of that page.
19363 Note that you must cast the addresses of variables to a @code{char *},
19364 since otherwise the value of @code{__djgpp_base_address}, the base
19365 address of all variables and functions in a @sc{djgpp} program, will
19366 be added using the rules of C pointer arithmetics: if @code{i} is
19367 declared an @code{int}, @value{GDBN} will add 4 times the value of
19368 @code{__djgpp_base_address} to the address of @code{i}.
19370 Here's another example, it displays the Page Table entry for the
19374 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
19375 @exdent @code{Page Table entry for address 0x29110:}
19376 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
19380 (The @code{+ 3} offset is because the transfer buffer's address is the
19381 3rd member of the @code{_go32_info_block} structure.) The output
19382 clearly shows that this DPMI server maps the addresses in conventional
19383 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
19384 linear (@code{0x29110}) addresses are identical.
19386 This command is supported only with some DPMI servers.
19389 @cindex DOS serial data link, remote debugging
19390 In addition to native debugging, the DJGPP port supports remote
19391 debugging via a serial data link. The following commands are specific
19392 to remote serial debugging in the DJGPP port of @value{GDBN}.
19395 @kindex set com1base
19396 @kindex set com1irq
19397 @kindex set com2base
19398 @kindex set com2irq
19399 @kindex set com3base
19400 @kindex set com3irq
19401 @kindex set com4base
19402 @kindex set com4irq
19403 @item set com1base @var{addr}
19404 This command sets the base I/O port address of the @file{COM1} serial
19407 @item set com1irq @var{irq}
19408 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
19409 for the @file{COM1} serial port.
19411 There are similar commands @samp{set com2base}, @samp{set com3irq},
19412 etc.@: for setting the port address and the @code{IRQ} lines for the
19415 @kindex show com1base
19416 @kindex show com1irq
19417 @kindex show com2base
19418 @kindex show com2irq
19419 @kindex show com3base
19420 @kindex show com3irq
19421 @kindex show com4base
19422 @kindex show com4irq
19423 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
19424 display the current settings of the base address and the @code{IRQ}
19425 lines used by the COM ports.
19428 @kindex info serial
19429 @cindex DOS serial port status
19430 This command prints the status of the 4 DOS serial ports. For each
19431 port, it prints whether it's active or not, its I/O base address and
19432 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
19433 counts of various errors encountered so far.
19437 @node Cygwin Native
19438 @subsection Features for Debugging MS Windows PE Executables
19439 @cindex MS Windows debugging
19440 @cindex native Cygwin debugging
19441 @cindex Cygwin-specific commands
19443 @value{GDBN} supports native debugging of MS Windows programs, including
19444 DLLs with and without symbolic debugging information.
19446 @cindex Ctrl-BREAK, MS-Windows
19447 @cindex interrupt debuggee on MS-Windows
19448 MS-Windows programs that call @code{SetConsoleMode} to switch off the
19449 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
19450 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
19451 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
19452 sequence, which can be used to interrupt the debuggee even if it
19455 There are various additional Cygwin-specific commands, described in
19456 this section. Working with DLLs that have no debugging symbols is
19457 described in @ref{Non-debug DLL Symbols}.
19462 This is a prefix of MS Windows-specific commands which print
19463 information about the target system and important OS structures.
19465 @item info w32 selector
19466 This command displays information returned by
19467 the Win32 API @code{GetThreadSelectorEntry} function.
19468 It takes an optional argument that is evaluated to
19469 a long value to give the information about this given selector.
19470 Without argument, this command displays information
19471 about the six segment registers.
19473 @item info w32 thread-information-block
19474 This command displays thread specific information stored in the
19475 Thread Information Block (readable on the X86 CPU family using @code{$fs}
19476 selector for 32-bit programs and @code{$gs} for 64-bit programs).
19480 This is a Cygwin-specific alias of @code{info shared}.
19482 @kindex dll-symbols
19484 This command loads symbols from a dll similarly to
19485 add-sym command but without the need to specify a base address.
19487 @kindex set cygwin-exceptions
19488 @cindex debugging the Cygwin DLL
19489 @cindex Cygwin DLL, debugging
19490 @item set cygwin-exceptions @var{mode}
19491 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
19492 happen inside the Cygwin DLL. If @var{mode} is @code{off},
19493 @value{GDBN} will delay recognition of exceptions, and may ignore some
19494 exceptions which seem to be caused by internal Cygwin DLL
19495 ``bookkeeping''. This option is meant primarily for debugging the
19496 Cygwin DLL itself; the default value is @code{off} to avoid annoying
19497 @value{GDBN} users with false @code{SIGSEGV} signals.
19499 @kindex show cygwin-exceptions
19500 @item show cygwin-exceptions
19501 Displays whether @value{GDBN} will break on exceptions that happen
19502 inside the Cygwin DLL itself.
19504 @kindex set new-console
19505 @item set new-console @var{mode}
19506 If @var{mode} is @code{on} the debuggee will
19507 be started in a new console on next start.
19508 If @var{mode} is @code{off}, the debuggee will
19509 be started in the same console as the debugger.
19511 @kindex show new-console
19512 @item show new-console
19513 Displays whether a new console is used
19514 when the debuggee is started.
19516 @kindex set new-group
19517 @item set new-group @var{mode}
19518 This boolean value controls whether the debuggee should
19519 start a new group or stay in the same group as the debugger.
19520 This affects the way the Windows OS handles
19523 @kindex show new-group
19524 @item show new-group
19525 Displays current value of new-group boolean.
19527 @kindex set debugevents
19528 @item set debugevents
19529 This boolean value adds debug output concerning kernel events related
19530 to the debuggee seen by the debugger. This includes events that
19531 signal thread and process creation and exit, DLL loading and
19532 unloading, console interrupts, and debugging messages produced by the
19533 Windows @code{OutputDebugString} API call.
19535 @kindex set debugexec
19536 @item set debugexec
19537 This boolean value adds debug output concerning execute events
19538 (such as resume thread) seen by the debugger.
19540 @kindex set debugexceptions
19541 @item set debugexceptions
19542 This boolean value adds debug output concerning exceptions in the
19543 debuggee seen by the debugger.
19545 @kindex set debugmemory
19546 @item set debugmemory
19547 This boolean value adds debug output concerning debuggee memory reads
19548 and writes by the debugger.
19552 This boolean values specifies whether the debuggee is called
19553 via a shell or directly (default value is on).
19557 Displays if the debuggee will be started with a shell.
19562 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
19565 @node Non-debug DLL Symbols
19566 @subsubsection Support for DLLs without Debugging Symbols
19567 @cindex DLLs with no debugging symbols
19568 @cindex Minimal symbols and DLLs
19570 Very often on windows, some of the DLLs that your program relies on do
19571 not include symbolic debugging information (for example,
19572 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
19573 symbols in a DLL, it relies on the minimal amount of symbolic
19574 information contained in the DLL's export table. This section
19575 describes working with such symbols, known internally to @value{GDBN} as
19576 ``minimal symbols''.
19578 Note that before the debugged program has started execution, no DLLs
19579 will have been loaded. The easiest way around this problem is simply to
19580 start the program --- either by setting a breakpoint or letting the
19581 program run once to completion. It is also possible to force
19582 @value{GDBN} to load a particular DLL before starting the executable ---
19583 see the shared library information in @ref{Files}, or the
19584 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
19585 explicitly loading symbols from a DLL with no debugging information will
19586 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
19587 which may adversely affect symbol lookup performance.
19589 @subsubsection DLL Name Prefixes
19591 In keeping with the naming conventions used by the Microsoft debugging
19592 tools, DLL export symbols are made available with a prefix based on the
19593 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
19594 also entered into the symbol table, so @code{CreateFileA} is often
19595 sufficient. In some cases there will be name clashes within a program
19596 (particularly if the executable itself includes full debugging symbols)
19597 necessitating the use of the fully qualified name when referring to the
19598 contents of the DLL. Use single-quotes around the name to avoid the
19599 exclamation mark (``!'') being interpreted as a language operator.
19601 Note that the internal name of the DLL may be all upper-case, even
19602 though the file name of the DLL is lower-case, or vice-versa. Since
19603 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
19604 some confusion. If in doubt, try the @code{info functions} and
19605 @code{info variables} commands or even @code{maint print msymbols}
19606 (@pxref{Symbols}). Here's an example:
19609 (@value{GDBP}) info function CreateFileA
19610 All functions matching regular expression "CreateFileA":
19612 Non-debugging symbols:
19613 0x77e885f4 CreateFileA
19614 0x77e885f4 KERNEL32!CreateFileA
19618 (@value{GDBP}) info function !
19619 All functions matching regular expression "!":
19621 Non-debugging symbols:
19622 0x6100114c cygwin1!__assert
19623 0x61004034 cygwin1!_dll_crt0@@0
19624 0x61004240 cygwin1!dll_crt0(per_process *)
19628 @subsubsection Working with Minimal Symbols
19630 Symbols extracted from a DLL's export table do not contain very much
19631 type information. All that @value{GDBN} can do is guess whether a symbol
19632 refers to a function or variable depending on the linker section that
19633 contains the symbol. Also note that the actual contents of the memory
19634 contained in a DLL are not available unless the program is running. This
19635 means that you cannot examine the contents of a variable or disassemble
19636 a function within a DLL without a running program.
19638 Variables are generally treated as pointers and dereferenced
19639 automatically. For this reason, it is often necessary to prefix a
19640 variable name with the address-of operator (``&'') and provide explicit
19641 type information in the command. Here's an example of the type of
19645 (@value{GDBP}) print 'cygwin1!__argv'
19650 (@value{GDBP}) x 'cygwin1!__argv'
19651 0x10021610: "\230y\""
19654 And two possible solutions:
19657 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
19658 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
19662 (@value{GDBP}) x/2x &'cygwin1!__argv'
19663 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
19664 (@value{GDBP}) x/x 0x10021608
19665 0x10021608: 0x0022fd98
19666 (@value{GDBP}) x/s 0x0022fd98
19667 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
19670 Setting a break point within a DLL is possible even before the program
19671 starts execution. However, under these circumstances, @value{GDBN} can't
19672 examine the initial instructions of the function in order to skip the
19673 function's frame set-up code. You can work around this by using ``*&''
19674 to set the breakpoint at a raw memory address:
19677 (@value{GDBP}) break *&'python22!PyOS_Readline'
19678 Breakpoint 1 at 0x1e04eff0
19681 The author of these extensions is not entirely convinced that setting a
19682 break point within a shared DLL like @file{kernel32.dll} is completely
19686 @subsection Commands Specific to @sc{gnu} Hurd Systems
19687 @cindex @sc{gnu} Hurd debugging
19689 This subsection describes @value{GDBN} commands specific to the
19690 @sc{gnu} Hurd native debugging.
19695 @kindex set signals@r{, Hurd command}
19696 @kindex set sigs@r{, Hurd command}
19697 This command toggles the state of inferior signal interception by
19698 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
19699 affected by this command. @code{sigs} is a shorthand alias for
19704 @kindex show signals@r{, Hurd command}
19705 @kindex show sigs@r{, Hurd command}
19706 Show the current state of intercepting inferior's signals.
19708 @item set signal-thread
19709 @itemx set sigthread
19710 @kindex set signal-thread
19711 @kindex set sigthread
19712 This command tells @value{GDBN} which thread is the @code{libc} signal
19713 thread. That thread is run when a signal is delivered to a running
19714 process. @code{set sigthread} is the shorthand alias of @code{set
19717 @item show signal-thread
19718 @itemx show sigthread
19719 @kindex show signal-thread
19720 @kindex show sigthread
19721 These two commands show which thread will run when the inferior is
19722 delivered a signal.
19725 @kindex set stopped@r{, Hurd command}
19726 This commands tells @value{GDBN} that the inferior process is stopped,
19727 as with the @code{SIGSTOP} signal. The stopped process can be
19728 continued by delivering a signal to it.
19731 @kindex show stopped@r{, Hurd command}
19732 This command shows whether @value{GDBN} thinks the debuggee is
19735 @item set exceptions
19736 @kindex set exceptions@r{, Hurd command}
19737 Use this command to turn off trapping of exceptions in the inferior.
19738 When exception trapping is off, neither breakpoints nor
19739 single-stepping will work. To restore the default, set exception
19742 @item show exceptions
19743 @kindex show exceptions@r{, Hurd command}
19744 Show the current state of trapping exceptions in the inferior.
19746 @item set task pause
19747 @kindex set task@r{, Hurd commands}
19748 @cindex task attributes (@sc{gnu} Hurd)
19749 @cindex pause current task (@sc{gnu} Hurd)
19750 This command toggles task suspension when @value{GDBN} has control.
19751 Setting it to on takes effect immediately, and the task is suspended
19752 whenever @value{GDBN} gets control. Setting it to off will take
19753 effect the next time the inferior is continued. If this option is set
19754 to off, you can use @code{set thread default pause on} or @code{set
19755 thread pause on} (see below) to pause individual threads.
19757 @item show task pause
19758 @kindex show task@r{, Hurd commands}
19759 Show the current state of task suspension.
19761 @item set task detach-suspend-count
19762 @cindex task suspend count
19763 @cindex detach from task, @sc{gnu} Hurd
19764 This command sets the suspend count the task will be left with when
19765 @value{GDBN} detaches from it.
19767 @item show task detach-suspend-count
19768 Show the suspend count the task will be left with when detaching.
19770 @item set task exception-port
19771 @itemx set task excp
19772 @cindex task exception port, @sc{gnu} Hurd
19773 This command sets the task exception port to which @value{GDBN} will
19774 forward exceptions. The argument should be the value of the @dfn{send
19775 rights} of the task. @code{set task excp} is a shorthand alias.
19777 @item set noninvasive
19778 @cindex noninvasive task options
19779 This command switches @value{GDBN} to a mode that is the least
19780 invasive as far as interfering with the inferior is concerned. This
19781 is the same as using @code{set task pause}, @code{set exceptions}, and
19782 @code{set signals} to values opposite to the defaults.
19784 @item info send-rights
19785 @itemx info receive-rights
19786 @itemx info port-rights
19787 @itemx info port-sets
19788 @itemx info dead-names
19791 @cindex send rights, @sc{gnu} Hurd
19792 @cindex receive rights, @sc{gnu} Hurd
19793 @cindex port rights, @sc{gnu} Hurd
19794 @cindex port sets, @sc{gnu} Hurd
19795 @cindex dead names, @sc{gnu} Hurd
19796 These commands display information about, respectively, send rights,
19797 receive rights, port rights, port sets, and dead names of a task.
19798 There are also shorthand aliases: @code{info ports} for @code{info
19799 port-rights} and @code{info psets} for @code{info port-sets}.
19801 @item set thread pause
19802 @kindex set thread@r{, Hurd command}
19803 @cindex thread properties, @sc{gnu} Hurd
19804 @cindex pause current thread (@sc{gnu} Hurd)
19805 This command toggles current thread suspension when @value{GDBN} has
19806 control. Setting it to on takes effect immediately, and the current
19807 thread is suspended whenever @value{GDBN} gets control. Setting it to
19808 off will take effect the next time the inferior is continued.
19809 Normally, this command has no effect, since when @value{GDBN} has
19810 control, the whole task is suspended. However, if you used @code{set
19811 task pause off} (see above), this command comes in handy to suspend
19812 only the current thread.
19814 @item show thread pause
19815 @kindex show thread@r{, Hurd command}
19816 This command shows the state of current thread suspension.
19818 @item set thread run
19819 This command sets whether the current thread is allowed to run.
19821 @item show thread run
19822 Show whether the current thread is allowed to run.
19824 @item set thread detach-suspend-count
19825 @cindex thread suspend count, @sc{gnu} Hurd
19826 @cindex detach from thread, @sc{gnu} Hurd
19827 This command sets the suspend count @value{GDBN} will leave on a
19828 thread when detaching. This number is relative to the suspend count
19829 found by @value{GDBN} when it notices the thread; use @code{set thread
19830 takeover-suspend-count} to force it to an absolute value.
19832 @item show thread detach-suspend-count
19833 Show the suspend count @value{GDBN} will leave on the thread when
19836 @item set thread exception-port
19837 @itemx set thread excp
19838 Set the thread exception port to which to forward exceptions. This
19839 overrides the port set by @code{set task exception-port} (see above).
19840 @code{set thread excp} is the shorthand alias.
19842 @item set thread takeover-suspend-count
19843 Normally, @value{GDBN}'s thread suspend counts are relative to the
19844 value @value{GDBN} finds when it notices each thread. This command
19845 changes the suspend counts to be absolute instead.
19847 @item set thread default
19848 @itemx show thread default
19849 @cindex thread default settings, @sc{gnu} Hurd
19850 Each of the above @code{set thread} commands has a @code{set thread
19851 default} counterpart (e.g., @code{set thread default pause}, @code{set
19852 thread default exception-port}, etc.). The @code{thread default}
19853 variety of commands sets the default thread properties for all
19854 threads; you can then change the properties of individual threads with
19855 the non-default commands.
19862 @value{GDBN} provides the following commands specific to the Darwin target:
19865 @item set debug darwin @var{num}
19866 @kindex set debug darwin
19867 When set to a non zero value, enables debugging messages specific to
19868 the Darwin support. Higher values produce more verbose output.
19870 @item show debug darwin
19871 @kindex show debug darwin
19872 Show the current state of Darwin messages.
19874 @item set debug mach-o @var{num}
19875 @kindex set debug mach-o
19876 When set to a non zero value, enables debugging messages while
19877 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
19878 file format used on Darwin for object and executable files.) Higher
19879 values produce more verbose output. This is a command to diagnose
19880 problems internal to @value{GDBN} and should not be needed in normal
19883 @item show debug mach-o
19884 @kindex show debug mach-o
19885 Show the current state of Mach-O file messages.
19887 @item set mach-exceptions on
19888 @itemx set mach-exceptions off
19889 @kindex set mach-exceptions
19890 On Darwin, faults are first reported as a Mach exception and are then
19891 mapped to a Posix signal. Use this command to turn on trapping of
19892 Mach exceptions in the inferior. This might be sometimes useful to
19893 better understand the cause of a fault. The default is off.
19895 @item show mach-exceptions
19896 @kindex show mach-exceptions
19897 Show the current state of exceptions trapping.
19902 @section Embedded Operating Systems
19904 This section describes configurations involving the debugging of
19905 embedded operating systems that are available for several different
19909 * VxWorks:: Using @value{GDBN} with VxWorks
19912 @value{GDBN} includes the ability to debug programs running on
19913 various real-time operating systems.
19916 @subsection Using @value{GDBN} with VxWorks
19922 @kindex target vxworks
19923 @item target vxworks @var{machinename}
19924 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
19925 is the target system's machine name or IP address.
19929 On VxWorks, @code{load} links @var{filename} dynamically on the
19930 current target system as well as adding its symbols in @value{GDBN}.
19932 @value{GDBN} enables developers to spawn and debug tasks running on networked
19933 VxWorks targets from a Unix host. Already-running tasks spawned from
19934 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
19935 both the Unix host and on the VxWorks target. The program
19936 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
19937 installed with the name @code{vxgdb}, to distinguish it from a
19938 @value{GDBN} for debugging programs on the host itself.)
19941 @item VxWorks-timeout @var{args}
19942 @kindex vxworks-timeout
19943 All VxWorks-based targets now support the option @code{vxworks-timeout}.
19944 This option is set by the user, and @var{args} represents the number of
19945 seconds @value{GDBN} waits for responses to rpc's. You might use this if
19946 your VxWorks target is a slow software simulator or is on the far side
19947 of a thin network line.
19950 The following information on connecting to VxWorks was current when
19951 this manual was produced; newer releases of VxWorks may use revised
19954 @findex INCLUDE_RDB
19955 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
19956 to include the remote debugging interface routines in the VxWorks
19957 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
19958 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
19959 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
19960 source debugging task @code{tRdbTask} when VxWorks is booted. For more
19961 information on configuring and remaking VxWorks, see the manufacturer's
19963 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
19965 Once you have included @file{rdb.a} in your VxWorks system image and set
19966 your Unix execution search path to find @value{GDBN}, you are ready to
19967 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
19968 @code{vxgdb}, depending on your installation).
19970 @value{GDBN} comes up showing the prompt:
19977 * VxWorks Connection:: Connecting to VxWorks
19978 * VxWorks Download:: VxWorks download
19979 * VxWorks Attach:: Running tasks
19982 @node VxWorks Connection
19983 @subsubsection Connecting to VxWorks
19985 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
19986 network. To connect to a target whose host name is ``@code{tt}'', type:
19989 (vxgdb) target vxworks tt
19993 @value{GDBN} displays messages like these:
19996 Attaching remote machine across net...
20001 @value{GDBN} then attempts to read the symbol tables of any object modules
20002 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
20003 these files by searching the directories listed in the command search
20004 path (@pxref{Environment, ,Your Program's Environment}); if it fails
20005 to find an object file, it displays a message such as:
20008 prog.o: No such file or directory.
20011 When this happens, add the appropriate directory to the search path with
20012 the @value{GDBN} command @code{path}, and execute the @code{target}
20015 @node VxWorks Download
20016 @subsubsection VxWorks Download
20018 @cindex download to VxWorks
20019 If you have connected to the VxWorks target and you want to debug an
20020 object that has not yet been loaded, you can use the @value{GDBN}
20021 @code{load} command to download a file from Unix to VxWorks
20022 incrementally. The object file given as an argument to the @code{load}
20023 command is actually opened twice: first by the VxWorks target in order
20024 to download the code, then by @value{GDBN} in order to read the symbol
20025 table. This can lead to problems if the current working directories on
20026 the two systems differ. If both systems have NFS mounted the same
20027 filesystems, you can avoid these problems by using absolute paths.
20028 Otherwise, it is simplest to set the working directory on both systems
20029 to the directory in which the object file resides, and then to reference
20030 the file by its name, without any path. For instance, a program
20031 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
20032 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
20033 program, type this on VxWorks:
20036 -> cd "@var{vxpath}/vw/demo/rdb"
20040 Then, in @value{GDBN}, type:
20043 (vxgdb) cd @var{hostpath}/vw/demo/rdb
20044 (vxgdb) load prog.o
20047 @value{GDBN} displays a response similar to this:
20050 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
20053 You can also use the @code{load} command to reload an object module
20054 after editing and recompiling the corresponding source file. Note that
20055 this makes @value{GDBN} delete all currently-defined breakpoints,
20056 auto-displays, and convenience variables, and to clear the value
20057 history. (This is necessary in order to preserve the integrity of
20058 debugger's data structures that reference the target system's symbol
20061 @node VxWorks Attach
20062 @subsubsection Running Tasks
20064 @cindex running VxWorks tasks
20065 You can also attach to an existing task using the @code{attach} command as
20069 (vxgdb) attach @var{task}
20073 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
20074 or suspended when you attach to it. Running tasks are suspended at
20075 the time of attachment.
20077 @node Embedded Processors
20078 @section Embedded Processors
20080 This section goes into details specific to particular embedded
20083 @cindex send command to simulator
20084 Whenever a specific embedded processor has a simulator, @value{GDBN}
20085 allows to send an arbitrary command to the simulator.
20088 @item sim @var{command}
20089 @kindex sim@r{, a command}
20090 Send an arbitrary @var{command} string to the simulator. Consult the
20091 documentation for the specific simulator in use for information about
20092 acceptable commands.
20098 * M32R/D:: Renesas M32R/D
20099 * M68K:: Motorola M68K
20100 * MicroBlaze:: Xilinx MicroBlaze
20101 * MIPS Embedded:: MIPS Embedded
20102 * PowerPC Embedded:: PowerPC Embedded
20103 * PA:: HP PA Embedded
20104 * Sparclet:: Tsqware Sparclet
20105 * Sparclite:: Fujitsu Sparclite
20106 * Z8000:: Zilog Z8000
20109 * Super-H:: Renesas Super-H
20118 @item target rdi @var{dev}
20119 ARM Angel monitor, via RDI library interface to ADP protocol. You may
20120 use this target to communicate with both boards running the Angel
20121 monitor, or with the EmbeddedICE JTAG debug device.
20124 @item target rdp @var{dev}
20129 @value{GDBN} provides the following ARM-specific commands:
20132 @item set arm disassembler
20134 This commands selects from a list of disassembly styles. The
20135 @code{"std"} style is the standard style.
20137 @item show arm disassembler
20139 Show the current disassembly style.
20141 @item set arm apcs32
20142 @cindex ARM 32-bit mode
20143 This command toggles ARM operation mode between 32-bit and 26-bit.
20145 @item show arm apcs32
20146 Display the current usage of the ARM 32-bit mode.
20148 @item set arm fpu @var{fputype}
20149 This command sets the ARM floating-point unit (FPU) type. The
20150 argument @var{fputype} can be one of these:
20154 Determine the FPU type by querying the OS ABI.
20156 Software FPU, with mixed-endian doubles on little-endian ARM
20159 GCC-compiled FPA co-processor.
20161 Software FPU with pure-endian doubles.
20167 Show the current type of the FPU.
20170 This command forces @value{GDBN} to use the specified ABI.
20173 Show the currently used ABI.
20175 @item set arm fallback-mode (arm|thumb|auto)
20176 @value{GDBN} uses the symbol table, when available, to determine
20177 whether instructions are ARM or Thumb. This command controls
20178 @value{GDBN}'s default behavior when the symbol table is not
20179 available. The default is @samp{auto}, which causes @value{GDBN} to
20180 use the current execution mode (from the @code{T} bit in the @code{CPSR}
20183 @item show arm fallback-mode
20184 Show the current fallback instruction mode.
20186 @item set arm force-mode (arm|thumb|auto)
20187 This command overrides use of the symbol table to determine whether
20188 instructions are ARM or Thumb. The default is @samp{auto}, which
20189 causes @value{GDBN} to use the symbol table and then the setting
20190 of @samp{set arm fallback-mode}.
20192 @item show arm force-mode
20193 Show the current forced instruction mode.
20195 @item set debug arm
20196 Toggle whether to display ARM-specific debugging messages from the ARM
20197 target support subsystem.
20199 @item show debug arm
20200 Show whether ARM-specific debugging messages are enabled.
20203 The following commands are available when an ARM target is debugged
20204 using the RDI interface:
20207 @item rdilogfile @r{[}@var{file}@r{]}
20209 @cindex ADP (Angel Debugger Protocol) logging
20210 Set the filename for the ADP (Angel Debugger Protocol) packet log.
20211 With an argument, sets the log file to the specified @var{file}. With
20212 no argument, show the current log file name. The default log file is
20215 @item rdilogenable @r{[}@var{arg}@r{]}
20216 @kindex rdilogenable
20217 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
20218 enables logging, with an argument 0 or @code{"no"} disables it. With
20219 no arguments displays the current setting. When logging is enabled,
20220 ADP packets exchanged between @value{GDBN} and the RDI target device
20221 are logged to a file.
20223 @item set rdiromatzero
20224 @kindex set rdiromatzero
20225 @cindex ROM at zero address, RDI
20226 Tell @value{GDBN} whether the target has ROM at address 0. If on,
20227 vector catching is disabled, so that zero address can be used. If off
20228 (the default), vector catching is enabled. For this command to take
20229 effect, it needs to be invoked prior to the @code{target rdi} command.
20231 @item show rdiromatzero
20232 @kindex show rdiromatzero
20233 Show the current setting of ROM at zero address.
20235 @item set rdiheartbeat
20236 @kindex set rdiheartbeat
20237 @cindex RDI heartbeat
20238 Enable or disable RDI heartbeat packets. It is not recommended to
20239 turn on this option, since it confuses ARM and EPI JTAG interface, as
20240 well as the Angel monitor.
20242 @item show rdiheartbeat
20243 @kindex show rdiheartbeat
20244 Show the setting of RDI heartbeat packets.
20248 @item target sim @r{[}@var{simargs}@r{]} @dots{}
20249 The @value{GDBN} ARM simulator accepts the following optional arguments.
20252 @item --swi-support=@var{type}
20253 Tell the simulator which SWI interfaces to support.
20254 @var{type} may be a comma separated list of the following values.
20255 The default value is @code{all}.
20268 @subsection Renesas M32R/D and M32R/SDI
20271 @kindex target m32r
20272 @item target m32r @var{dev}
20273 Renesas M32R/D ROM monitor.
20275 @kindex target m32rsdi
20276 @item target m32rsdi @var{dev}
20277 Renesas M32R SDI server, connected via parallel port to the board.
20280 The following @value{GDBN} commands are specific to the M32R monitor:
20283 @item set download-path @var{path}
20284 @kindex set download-path
20285 @cindex find downloadable @sc{srec} files (M32R)
20286 Set the default path for finding downloadable @sc{srec} files.
20288 @item show download-path
20289 @kindex show download-path
20290 Show the default path for downloadable @sc{srec} files.
20292 @item set board-address @var{addr}
20293 @kindex set board-address
20294 @cindex M32-EVA target board address
20295 Set the IP address for the M32R-EVA target board.
20297 @item show board-address
20298 @kindex show board-address
20299 Show the current IP address of the target board.
20301 @item set server-address @var{addr}
20302 @kindex set server-address
20303 @cindex download server address (M32R)
20304 Set the IP address for the download server, which is the @value{GDBN}'s
20307 @item show server-address
20308 @kindex show server-address
20309 Display the IP address of the download server.
20311 @item upload @r{[}@var{file}@r{]}
20312 @kindex upload@r{, M32R}
20313 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
20314 upload capability. If no @var{file} argument is given, the current
20315 executable file is uploaded.
20317 @item tload @r{[}@var{file}@r{]}
20318 @kindex tload@r{, M32R}
20319 Test the @code{upload} command.
20322 The following commands are available for M32R/SDI:
20327 @cindex reset SDI connection, M32R
20328 This command resets the SDI connection.
20332 This command shows the SDI connection status.
20335 @kindex debug_chaos
20336 @cindex M32R/Chaos debugging
20337 Instructs the remote that M32R/Chaos debugging is to be used.
20339 @item use_debug_dma
20340 @kindex use_debug_dma
20341 Instructs the remote to use the DEBUG_DMA method of accessing memory.
20344 @kindex use_mon_code
20345 Instructs the remote to use the MON_CODE method of accessing memory.
20348 @kindex use_ib_break
20349 Instructs the remote to set breakpoints by IB break.
20351 @item use_dbt_break
20352 @kindex use_dbt_break
20353 Instructs the remote to set breakpoints by DBT.
20359 The Motorola m68k configuration includes ColdFire support, and a
20360 target command for the following ROM monitor.
20364 @kindex target dbug
20365 @item target dbug @var{dev}
20366 dBUG ROM monitor for Motorola ColdFire.
20371 @subsection MicroBlaze
20372 @cindex Xilinx MicroBlaze
20373 @cindex XMD, Xilinx Microprocessor Debugger
20375 The MicroBlaze is a soft-core processor supported on various Xilinx
20376 FPGAs, such as Spartan or Virtex series. Boards with these processors
20377 usually have JTAG ports which connect to a host system running the Xilinx
20378 Embedded Development Kit (EDK) or Software Development Kit (SDK).
20379 This host system is used to download the configuration bitstream to
20380 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
20381 communicates with the target board using the JTAG interface and
20382 presents a @code{gdbserver} interface to the board. By default
20383 @code{xmd} uses port @code{1234}. (While it is possible to change
20384 this default port, it requires the use of undocumented @code{xmd}
20385 commands. Contact Xilinx support if you need to do this.)
20387 Use these GDB commands to connect to the MicroBlaze target processor.
20390 @item target remote :1234
20391 Use this command to connect to the target if you are running @value{GDBN}
20392 on the same system as @code{xmd}.
20394 @item target remote @var{xmd-host}:1234
20395 Use this command to connect to the target if it is connected to @code{xmd}
20396 running on a different system named @var{xmd-host}.
20399 Use this command to download a program to the MicroBlaze target.
20401 @item set debug microblaze @var{n}
20402 Enable MicroBlaze-specific debugging messages if non-zero.
20404 @item show debug microblaze @var{n}
20405 Show MicroBlaze-specific debugging level.
20408 @node MIPS Embedded
20409 @subsection @acronym{MIPS} Embedded
20411 @cindex @acronym{MIPS} boards
20412 @value{GDBN} can use the @acronym{MIPS} remote debugging protocol to talk to a
20413 @acronym{MIPS} board attached to a serial line. This is available when
20414 you configure @value{GDBN} with @samp{--target=mips-elf}.
20417 Use these @value{GDBN} commands to specify the connection to your target board:
20420 @item target mips @var{port}
20421 @kindex target mips @var{port}
20422 To run a program on the board, start up @code{@value{GDBP}} with the
20423 name of your program as the argument. To connect to the board, use the
20424 command @samp{target mips @var{port}}, where @var{port} is the name of
20425 the serial port connected to the board. If the program has not already
20426 been downloaded to the board, you may use the @code{load} command to
20427 download it. You can then use all the usual @value{GDBN} commands.
20429 For example, this sequence connects to the target board through a serial
20430 port, and loads and runs a program called @var{prog} through the
20434 host$ @value{GDBP} @var{prog}
20435 @value{GDBN} is free software and @dots{}
20436 (@value{GDBP}) target mips /dev/ttyb
20437 (@value{GDBP}) load @var{prog}
20441 @item target mips @var{hostname}:@var{portnumber}
20442 On some @value{GDBN} host configurations, you can specify a TCP
20443 connection (for instance, to a serial line managed by a terminal
20444 concentrator) instead of a serial port, using the syntax
20445 @samp{@var{hostname}:@var{portnumber}}.
20447 @item target pmon @var{port}
20448 @kindex target pmon @var{port}
20451 @item target ddb @var{port}
20452 @kindex target ddb @var{port}
20453 NEC's DDB variant of PMON for Vr4300.
20455 @item target lsi @var{port}
20456 @kindex target lsi @var{port}
20457 LSI variant of PMON.
20459 @kindex target r3900
20460 @item target r3900 @var{dev}
20461 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
20463 @kindex target array
20464 @item target array @var{dev}
20465 Array Tech LSI33K RAID controller board.
20471 @value{GDBN} also supports these special commands for @acronym{MIPS} targets:
20474 @item set mipsfpu double
20475 @itemx set mipsfpu single
20476 @itemx set mipsfpu none
20477 @itemx set mipsfpu auto
20478 @itemx show mipsfpu
20479 @kindex set mipsfpu
20480 @kindex show mipsfpu
20481 @cindex @acronym{MIPS} remote floating point
20482 @cindex floating point, @acronym{MIPS} remote
20483 If your target board does not support the @acronym{MIPS} floating point
20484 coprocessor, you should use the command @samp{set mipsfpu none} (if you
20485 need this, you may wish to put the command in your @value{GDBN} init
20486 file). This tells @value{GDBN} how to find the return value of
20487 functions which return floating point values. It also allows
20488 @value{GDBN} to avoid saving the floating point registers when calling
20489 functions on the board. If you are using a floating point coprocessor
20490 with only single precision floating point support, as on the @sc{r4650}
20491 processor, use the command @samp{set mipsfpu single}. The default
20492 double precision floating point coprocessor may be selected using
20493 @samp{set mipsfpu double}.
20495 In previous versions the only choices were double precision or no
20496 floating point, so @samp{set mipsfpu on} will select double precision
20497 and @samp{set mipsfpu off} will select no floating point.
20499 As usual, you can inquire about the @code{mipsfpu} variable with
20500 @samp{show mipsfpu}.
20502 @item set timeout @var{seconds}
20503 @itemx set retransmit-timeout @var{seconds}
20504 @itemx show timeout
20505 @itemx show retransmit-timeout
20506 @cindex @code{timeout}, @acronym{MIPS} protocol
20507 @cindex @code{retransmit-timeout}, @acronym{MIPS} protocol
20508 @kindex set timeout
20509 @kindex show timeout
20510 @kindex set retransmit-timeout
20511 @kindex show retransmit-timeout
20512 You can control the timeout used while waiting for a packet, in the @acronym{MIPS}
20513 remote protocol, with the @code{set timeout @var{seconds}} command. The
20514 default is 5 seconds. Similarly, you can control the timeout used while
20515 waiting for an acknowledgment of a packet with the @code{set
20516 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
20517 You can inspect both values with @code{show timeout} and @code{show
20518 retransmit-timeout}. (These commands are @emph{only} available when
20519 @value{GDBN} is configured for @samp{--target=mips-elf}.)
20521 The timeout set by @code{set timeout} does not apply when @value{GDBN}
20522 is waiting for your program to stop. In that case, @value{GDBN} waits
20523 forever because it has no way of knowing how long the program is going
20524 to run before stopping.
20526 @item set syn-garbage-limit @var{num}
20527 @kindex set syn-garbage-limit@r{, @acronym{MIPS} remote}
20528 @cindex synchronize with remote @acronym{MIPS} target
20529 Limit the maximum number of characters @value{GDBN} should ignore when
20530 it tries to synchronize with the remote target. The default is 10
20531 characters. Setting the limit to -1 means there's no limit.
20533 @item show syn-garbage-limit
20534 @kindex show syn-garbage-limit@r{, @acronym{MIPS} remote}
20535 Show the current limit on the number of characters to ignore when
20536 trying to synchronize with the remote system.
20538 @item set monitor-prompt @var{prompt}
20539 @kindex set monitor-prompt@r{, @acronym{MIPS} remote}
20540 @cindex remote monitor prompt
20541 Tell @value{GDBN} to expect the specified @var{prompt} string from the
20542 remote monitor. The default depends on the target:
20552 @item show monitor-prompt
20553 @kindex show monitor-prompt@r{, @acronym{MIPS} remote}
20554 Show the current strings @value{GDBN} expects as the prompt from the
20557 @item set monitor-warnings
20558 @kindex set monitor-warnings@r{, @acronym{MIPS} remote}
20559 Enable or disable monitor warnings about hardware breakpoints. This
20560 has effect only for the @code{lsi} target. When on, @value{GDBN} will
20561 display warning messages whose codes are returned by the @code{lsi}
20562 PMON monitor for breakpoint commands.
20564 @item show monitor-warnings
20565 @kindex show monitor-warnings@r{, @acronym{MIPS} remote}
20566 Show the current setting of printing monitor warnings.
20568 @item pmon @var{command}
20569 @kindex pmon@r{, @acronym{MIPS} remote}
20570 @cindex send PMON command
20571 This command allows sending an arbitrary @var{command} string to the
20572 monitor. The monitor must be in debug mode for this to work.
20575 @node PowerPC Embedded
20576 @subsection PowerPC Embedded
20578 @cindex DVC register
20579 @value{GDBN} supports using the DVC (Data Value Compare) register to
20580 implement in hardware simple hardware watchpoint conditions of the form:
20583 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
20584 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
20587 The DVC register will be automatically used when @value{GDBN} detects
20588 such pattern in a condition expression, and the created watchpoint uses one
20589 debug register (either the @code{exact-watchpoints} option is on and the
20590 variable is scalar, or the variable has a length of one byte). This feature
20591 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
20594 When running on PowerPC embedded processors, @value{GDBN} automatically uses
20595 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
20596 in which case watchpoints using only one debug register are created when
20597 watching variables of scalar types.
20599 You can create an artificial array to watch an arbitrary memory
20600 region using one of the following commands (@pxref{Expressions}):
20603 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
20604 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
20607 PowerPC embedded processors support masked watchpoints. See the discussion
20608 about the @code{mask} argument in @ref{Set Watchpoints}.
20610 @cindex ranged breakpoint
20611 PowerPC embedded processors support hardware accelerated
20612 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
20613 the inferior whenever it executes an instruction at any address within
20614 the range it specifies. To set a ranged breakpoint in @value{GDBN},
20615 use the @code{break-range} command.
20617 @value{GDBN} provides the following PowerPC-specific commands:
20620 @kindex break-range
20621 @item break-range @var{start-location}, @var{end-location}
20622 Set a breakpoint for an address range.
20623 @var{start-location} and @var{end-location} can specify a function name,
20624 a line number, an offset of lines from the current line or from the start
20625 location, or an address of an instruction (see @ref{Specify Location},
20626 for a list of all the possible ways to specify a @var{location}.)
20627 The breakpoint will stop execution of the inferior whenever it
20628 executes an instruction at any address within the specified range,
20629 (including @var{start-location} and @var{end-location}.)
20631 @kindex set powerpc
20632 @item set powerpc soft-float
20633 @itemx show powerpc soft-float
20634 Force @value{GDBN} to use (or not use) a software floating point calling
20635 convention. By default, @value{GDBN} selects the calling convention based
20636 on the selected architecture and the provided executable file.
20638 @item set powerpc vector-abi
20639 @itemx show powerpc vector-abi
20640 Force @value{GDBN} to use the specified calling convention for vector
20641 arguments and return values. The valid options are @samp{auto};
20642 @samp{generic}, to avoid vector registers even if they are present;
20643 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
20644 registers. By default, @value{GDBN} selects the calling convention
20645 based on the selected architecture and the provided executable file.
20647 @item set powerpc exact-watchpoints
20648 @itemx show powerpc exact-watchpoints
20649 Allow @value{GDBN} to use only one debug register when watching a variable
20650 of scalar type, thus assuming that the variable is accessed through the
20651 address of its first byte.
20653 @kindex target dink32
20654 @item target dink32 @var{dev}
20655 DINK32 ROM monitor.
20657 @kindex target ppcbug
20658 @item target ppcbug @var{dev}
20659 @kindex target ppcbug1
20660 @item target ppcbug1 @var{dev}
20661 PPCBUG ROM monitor for PowerPC.
20664 @item target sds @var{dev}
20665 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
20668 @cindex SDS protocol
20669 The following commands specific to the SDS protocol are supported
20673 @item set sdstimeout @var{nsec}
20674 @kindex set sdstimeout
20675 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
20676 default is 2 seconds.
20678 @item show sdstimeout
20679 @kindex show sdstimeout
20680 Show the current value of the SDS timeout.
20682 @item sds @var{command}
20683 @kindex sds@r{, a command}
20684 Send the specified @var{command} string to the SDS monitor.
20689 @subsection HP PA Embedded
20693 @kindex target op50n
20694 @item target op50n @var{dev}
20695 OP50N monitor, running on an OKI HPPA board.
20697 @kindex target w89k
20698 @item target w89k @var{dev}
20699 W89K monitor, running on a Winbond HPPA board.
20704 @subsection Tsqware Sparclet
20708 @value{GDBN} enables developers to debug tasks running on
20709 Sparclet targets from a Unix host.
20710 @value{GDBN} uses code that runs on
20711 both the Unix host and on the Sparclet target. The program
20712 @code{@value{GDBP}} is installed and executed on the Unix host.
20715 @item remotetimeout @var{args}
20716 @kindex remotetimeout
20717 @value{GDBN} supports the option @code{remotetimeout}.
20718 This option is set by the user, and @var{args} represents the number of
20719 seconds @value{GDBN} waits for responses.
20722 @cindex compiling, on Sparclet
20723 When compiling for debugging, include the options @samp{-g} to get debug
20724 information and @samp{-Ttext} to relocate the program to where you wish to
20725 load it on the target. You may also want to add the options @samp{-n} or
20726 @samp{-N} in order to reduce the size of the sections. Example:
20729 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
20732 You can use @code{objdump} to verify that the addresses are what you intended:
20735 sparclet-aout-objdump --headers --syms prog
20738 @cindex running, on Sparclet
20740 your Unix execution search path to find @value{GDBN}, you are ready to
20741 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
20742 (or @code{sparclet-aout-gdb}, depending on your installation).
20744 @value{GDBN} comes up showing the prompt:
20751 * Sparclet File:: Setting the file to debug
20752 * Sparclet Connection:: Connecting to Sparclet
20753 * Sparclet Download:: Sparclet download
20754 * Sparclet Execution:: Running and debugging
20757 @node Sparclet File
20758 @subsubsection Setting File to Debug
20760 The @value{GDBN} command @code{file} lets you choose with program to debug.
20763 (gdbslet) file prog
20767 @value{GDBN} then attempts to read the symbol table of @file{prog}.
20768 @value{GDBN} locates
20769 the file by searching the directories listed in the command search
20771 If the file was compiled with debug information (option @samp{-g}), source
20772 files will be searched as well.
20773 @value{GDBN} locates
20774 the source files by searching the directories listed in the directory search
20775 path (@pxref{Environment, ,Your Program's Environment}).
20777 to find a file, it displays a message such as:
20780 prog: No such file or directory.
20783 When this happens, add the appropriate directories to the search paths with
20784 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
20785 @code{target} command again.
20787 @node Sparclet Connection
20788 @subsubsection Connecting to Sparclet
20790 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
20791 To connect to a target on serial port ``@code{ttya}'', type:
20794 (gdbslet) target sparclet /dev/ttya
20795 Remote target sparclet connected to /dev/ttya
20796 main () at ../prog.c:3
20800 @value{GDBN} displays messages like these:
20806 @node Sparclet Download
20807 @subsubsection Sparclet Download
20809 @cindex download to Sparclet
20810 Once connected to the Sparclet target,
20811 you can use the @value{GDBN}
20812 @code{load} command to download the file from the host to the target.
20813 The file name and load offset should be given as arguments to the @code{load}
20815 Since the file format is aout, the program must be loaded to the starting
20816 address. You can use @code{objdump} to find out what this value is. The load
20817 offset is an offset which is added to the VMA (virtual memory address)
20818 of each of the file's sections.
20819 For instance, if the program
20820 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
20821 and bss at 0x12010170, in @value{GDBN}, type:
20824 (gdbslet) load prog 0x12010000
20825 Loading section .text, size 0xdb0 vma 0x12010000
20828 If the code is loaded at a different address then what the program was linked
20829 to, you may need to use the @code{section} and @code{add-symbol-file} commands
20830 to tell @value{GDBN} where to map the symbol table.
20832 @node Sparclet Execution
20833 @subsubsection Running and Debugging
20835 @cindex running and debugging Sparclet programs
20836 You can now begin debugging the task using @value{GDBN}'s execution control
20837 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
20838 manual for the list of commands.
20842 Breakpoint 1 at 0x12010000: file prog.c, line 3.
20844 Starting program: prog
20845 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
20846 3 char *symarg = 0;
20848 4 char *execarg = "hello!";
20853 @subsection Fujitsu Sparclite
20857 @kindex target sparclite
20858 @item target sparclite @var{dev}
20859 Fujitsu sparclite boards, used only for the purpose of loading.
20860 You must use an additional command to debug the program.
20861 For example: target remote @var{dev} using @value{GDBN} standard
20867 @subsection Zilog Z8000
20870 @cindex simulator, Z8000
20871 @cindex Zilog Z8000 simulator
20873 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
20876 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
20877 unsegmented variant of the Z8000 architecture) or the Z8001 (the
20878 segmented variant). The simulator recognizes which architecture is
20879 appropriate by inspecting the object code.
20882 @item target sim @var{args}
20884 @kindex target sim@r{, with Z8000}
20885 Debug programs on a simulated CPU. If the simulator supports setup
20886 options, specify them via @var{args}.
20890 After specifying this target, you can debug programs for the simulated
20891 CPU in the same style as programs for your host computer; use the
20892 @code{file} command to load a new program image, the @code{run} command
20893 to run your program, and so on.
20895 As well as making available all the usual machine registers
20896 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
20897 additional items of information as specially named registers:
20902 Counts clock-ticks in the simulator.
20905 Counts instructions run in the simulator.
20908 Execution time in 60ths of a second.
20912 You can refer to these values in @value{GDBN} expressions with the usual
20913 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
20914 conditional breakpoint that suspends only after at least 5000
20915 simulated clock ticks.
20918 @subsection Atmel AVR
20921 When configured for debugging the Atmel AVR, @value{GDBN} supports the
20922 following AVR-specific commands:
20925 @item info io_registers
20926 @kindex info io_registers@r{, AVR}
20927 @cindex I/O registers (Atmel AVR)
20928 This command displays information about the AVR I/O registers. For
20929 each register, @value{GDBN} prints its number and value.
20936 When configured for debugging CRIS, @value{GDBN} provides the
20937 following CRIS-specific commands:
20940 @item set cris-version @var{ver}
20941 @cindex CRIS version
20942 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
20943 The CRIS version affects register names and sizes. This command is useful in
20944 case autodetection of the CRIS version fails.
20946 @item show cris-version
20947 Show the current CRIS version.
20949 @item set cris-dwarf2-cfi
20950 @cindex DWARF-2 CFI and CRIS
20951 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
20952 Change to @samp{off} when using @code{gcc-cris} whose version is below
20955 @item show cris-dwarf2-cfi
20956 Show the current state of using DWARF-2 CFI.
20958 @item set cris-mode @var{mode}
20960 Set the current CRIS mode to @var{mode}. It should only be changed when
20961 debugging in guru mode, in which case it should be set to
20962 @samp{guru} (the default is @samp{normal}).
20964 @item show cris-mode
20965 Show the current CRIS mode.
20969 @subsection Renesas Super-H
20972 For the Renesas Super-H processor, @value{GDBN} provides these
20976 @item set sh calling-convention @var{convention}
20977 @kindex set sh calling-convention
20978 Set the calling-convention used when calling functions from @value{GDBN}.
20979 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
20980 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
20981 convention. If the DWARF-2 information of the called function specifies
20982 that the function follows the Renesas calling convention, the function
20983 is called using the Renesas calling convention. If the calling convention
20984 is set to @samp{renesas}, the Renesas calling convention is always used,
20985 regardless of the DWARF-2 information. This can be used to override the
20986 default of @samp{gcc} if debug information is missing, or the compiler
20987 does not emit the DWARF-2 calling convention entry for a function.
20989 @item show sh calling-convention
20990 @kindex show sh calling-convention
20991 Show the current calling convention setting.
20996 @node Architectures
20997 @section Architectures
20999 This section describes characteristics of architectures that affect
21000 all uses of @value{GDBN} with the architecture, both native and cross.
21007 * HPPA:: HP PA architecture
21008 * SPU:: Cell Broadband Engine SPU architecture
21014 @subsection AArch64
21015 @cindex AArch64 support
21017 When @value{GDBN} is debugging the AArch64 architecture, it provides the
21018 following special commands:
21021 @item set debug aarch64
21022 @kindex set debug aarch64
21023 This command determines whether AArch64 architecture-specific debugging
21024 messages are to be displayed.
21026 @item show debug aarch64
21027 Show whether AArch64 debugging messages are displayed.
21032 @subsection x86 Architecture-specific Issues
21035 @item set struct-convention @var{mode}
21036 @kindex set struct-convention
21037 @cindex struct return convention
21038 @cindex struct/union returned in registers
21039 Set the convention used by the inferior to return @code{struct}s and
21040 @code{union}s from functions to @var{mode}. Possible values of
21041 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
21042 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
21043 are returned on the stack, while @code{"reg"} means that a
21044 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
21045 be returned in a register.
21047 @item show struct-convention
21048 @kindex show struct-convention
21049 Show the current setting of the convention to return @code{struct}s
21056 See the following section.
21059 @subsection @acronym{MIPS}
21061 @cindex stack on Alpha
21062 @cindex stack on @acronym{MIPS}
21063 @cindex Alpha stack
21064 @cindex @acronym{MIPS} stack
21065 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
21066 sometimes requires @value{GDBN} to search backward in the object code to
21067 find the beginning of a function.
21069 @cindex response time, @acronym{MIPS} debugging
21070 To improve response time (especially for embedded applications, where
21071 @value{GDBN} may be restricted to a slow serial line for this search)
21072 you may want to limit the size of this search, using one of these
21076 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
21077 @item set heuristic-fence-post @var{limit}
21078 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
21079 search for the beginning of a function. A value of @var{0} (the
21080 default) means there is no limit. However, except for @var{0}, the
21081 larger the limit the more bytes @code{heuristic-fence-post} must search
21082 and therefore the longer it takes to run. You should only need to use
21083 this command when debugging a stripped executable.
21085 @item show heuristic-fence-post
21086 Display the current limit.
21090 These commands are available @emph{only} when @value{GDBN} is configured
21091 for debugging programs on Alpha or @acronym{MIPS} processors.
21093 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
21097 @item set mips abi @var{arg}
21098 @kindex set mips abi
21099 @cindex set ABI for @acronym{MIPS}
21100 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
21101 values of @var{arg} are:
21105 The default ABI associated with the current binary (this is the
21115 @item show mips abi
21116 @kindex show mips abi
21117 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
21119 @item set mips compression @var{arg}
21120 @kindex set mips compression
21121 @cindex code compression, @acronym{MIPS}
21122 Tell @value{GDBN} which @acronym{MIPS} compressed
21123 @acronym{ISA, Instruction Set Architecture} encoding is used by the
21124 inferior. @value{GDBN} uses this for code disassembly and other
21125 internal interpretation purposes. This setting is only referred to
21126 when no executable has been associated with the debugging session or
21127 the executable does not provide information about the encoding it uses.
21128 Otherwise this setting is automatically updated from information
21129 provided by the executable.
21131 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
21132 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
21133 executables containing @acronym{MIPS16} code frequently are not
21134 identified as such.
21136 This setting is ``sticky''; that is, it retains its value across
21137 debugging sessions until reset either explicitly with this command or
21138 implicitly from an executable.
21140 The compiler and/or assembler typically add symbol table annotations to
21141 identify functions compiled for the @acronym{MIPS16} or
21142 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
21143 are present, @value{GDBN} uses them in preference to the global
21144 compressed @acronym{ISA} encoding setting.
21146 @item show mips compression
21147 @kindex show mips compression
21148 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
21149 @value{GDBN} to debug the inferior.
21152 @itemx show mipsfpu
21153 @xref{MIPS Embedded, set mipsfpu}.
21155 @item set mips mask-address @var{arg}
21156 @kindex set mips mask-address
21157 @cindex @acronym{MIPS} addresses, masking
21158 This command determines whether the most-significant 32 bits of 64-bit
21159 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
21160 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
21161 setting, which lets @value{GDBN} determine the correct value.
21163 @item show mips mask-address
21164 @kindex show mips mask-address
21165 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
21168 @item set remote-mips64-transfers-32bit-regs
21169 @kindex set remote-mips64-transfers-32bit-regs
21170 This command controls compatibility with 64-bit @acronym{MIPS} targets that
21171 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
21172 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
21173 and 64 bits for other registers, set this option to @samp{on}.
21175 @item show remote-mips64-transfers-32bit-regs
21176 @kindex show remote-mips64-transfers-32bit-regs
21177 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
21179 @item set debug mips
21180 @kindex set debug mips
21181 This command turns on and off debugging messages for the @acronym{MIPS}-specific
21182 target code in @value{GDBN}.
21184 @item show debug mips
21185 @kindex show debug mips
21186 Show the current setting of @acronym{MIPS} debugging messages.
21192 @cindex HPPA support
21194 When @value{GDBN} is debugging the HP PA architecture, it provides the
21195 following special commands:
21198 @item set debug hppa
21199 @kindex set debug hppa
21200 This command determines whether HPPA architecture-specific debugging
21201 messages are to be displayed.
21203 @item show debug hppa
21204 Show whether HPPA debugging messages are displayed.
21206 @item maint print unwind @var{address}
21207 @kindex maint print unwind@r{, HPPA}
21208 This command displays the contents of the unwind table entry at the
21209 given @var{address}.
21215 @subsection Cell Broadband Engine SPU architecture
21216 @cindex Cell Broadband Engine
21219 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
21220 it provides the following special commands:
21223 @item info spu event
21225 Display SPU event facility status. Shows current event mask
21226 and pending event status.
21228 @item info spu signal
21229 Display SPU signal notification facility status. Shows pending
21230 signal-control word and signal notification mode of both signal
21231 notification channels.
21233 @item info spu mailbox
21234 Display SPU mailbox facility status. Shows all pending entries,
21235 in order of processing, in each of the SPU Write Outbound,
21236 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
21239 Display MFC DMA status. Shows all pending commands in the MFC
21240 DMA queue. For each entry, opcode, tag, class IDs, effective
21241 and local store addresses and transfer size are shown.
21243 @item info spu proxydma
21244 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
21245 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
21246 and local store addresses and transfer size are shown.
21250 When @value{GDBN} is debugging a combined PowerPC/SPU application
21251 on the Cell Broadband Engine, it provides in addition the following
21255 @item set spu stop-on-load @var{arg}
21257 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
21258 will give control to the user when a new SPE thread enters its @code{main}
21259 function. The default is @code{off}.
21261 @item show spu stop-on-load
21263 Show whether to stop for new SPE threads.
21265 @item set spu auto-flush-cache @var{arg}
21266 Set whether to automatically flush the software-managed cache. When set to
21267 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
21268 cache to be flushed whenever SPE execution stops. This provides a consistent
21269 view of PowerPC memory that is accessed via the cache. If an application
21270 does not use the software-managed cache, this option has no effect.
21272 @item show spu auto-flush-cache
21273 Show whether to automatically flush the software-managed cache.
21278 @subsection PowerPC
21279 @cindex PowerPC architecture
21281 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
21282 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
21283 numbers stored in the floating point registers. These values must be stored
21284 in two consecutive registers, always starting at an even register like
21285 @code{f0} or @code{f2}.
21287 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
21288 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
21289 @code{f2} and @code{f3} for @code{$dl1} and so on.
21291 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
21292 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
21295 @subsection Nios II
21296 @cindex Nios II architecture
21298 When @value{GDBN} is debugging the Nios II architecture,
21299 it provides the following special commands:
21303 @item set debug nios2
21304 @kindex set debug nios2
21305 This command turns on and off debugging messages for the Nios II
21306 target code in @value{GDBN}.
21308 @item show debug nios2
21309 @kindex show debug nios2
21310 Show the current setting of Nios II debugging messages.
21313 @node Controlling GDB
21314 @chapter Controlling @value{GDBN}
21316 You can alter the way @value{GDBN} interacts with you by using the
21317 @code{set} command. For commands controlling how @value{GDBN} displays
21318 data, see @ref{Print Settings, ,Print Settings}. Other settings are
21323 * Editing:: Command editing
21324 * Command History:: Command history
21325 * Screen Size:: Screen size
21326 * Numbers:: Numbers
21327 * ABI:: Configuring the current ABI
21328 * Auto-loading:: Automatically loading associated files
21329 * Messages/Warnings:: Optional warnings and messages
21330 * Debugging Output:: Optional messages about internal happenings
21331 * Other Misc Settings:: Other Miscellaneous Settings
21339 @value{GDBN} indicates its readiness to read a command by printing a string
21340 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
21341 can change the prompt string with the @code{set prompt} command. For
21342 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
21343 the prompt in one of the @value{GDBN} sessions so that you can always tell
21344 which one you are talking to.
21346 @emph{Note:} @code{set prompt} does not add a space for you after the
21347 prompt you set. This allows you to set a prompt which ends in a space
21348 or a prompt that does not.
21352 @item set prompt @var{newprompt}
21353 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
21355 @kindex show prompt
21357 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
21360 Versions of @value{GDBN} that ship with Python scripting enabled have
21361 prompt extensions. The commands for interacting with these extensions
21365 @kindex set extended-prompt
21366 @item set extended-prompt @var{prompt}
21367 Set an extended prompt that allows for substitutions.
21368 @xref{gdb.prompt}, for a list of escape sequences that can be used for
21369 substitution. Any escape sequences specified as part of the prompt
21370 string are replaced with the corresponding strings each time the prompt
21376 set extended-prompt Current working directory: \w (gdb)
21379 Note that when an extended-prompt is set, it takes control of the
21380 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
21382 @kindex show extended-prompt
21383 @item show extended-prompt
21384 Prints the extended prompt. Any escape sequences specified as part of
21385 the prompt string with @code{set extended-prompt}, are replaced with the
21386 corresponding strings each time the prompt is displayed.
21390 @section Command Editing
21392 @cindex command line editing
21394 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
21395 @sc{gnu} library provides consistent behavior for programs which provide a
21396 command line interface to the user. Advantages are @sc{gnu} Emacs-style
21397 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
21398 substitution, and a storage and recall of command history across
21399 debugging sessions.
21401 You may control the behavior of command line editing in @value{GDBN} with the
21402 command @code{set}.
21405 @kindex set editing
21408 @itemx set editing on
21409 Enable command line editing (enabled by default).
21411 @item set editing off
21412 Disable command line editing.
21414 @kindex show editing
21416 Show whether command line editing is enabled.
21419 @ifset SYSTEM_READLINE
21420 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
21422 @ifclear SYSTEM_READLINE
21423 @xref{Command Line Editing},
21425 for more details about the Readline
21426 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
21427 encouraged to read that chapter.
21429 @node Command History
21430 @section Command History
21431 @cindex command history
21433 @value{GDBN} can keep track of the commands you type during your
21434 debugging sessions, so that you can be certain of precisely what
21435 happened. Use these commands to manage the @value{GDBN} command
21438 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
21439 package, to provide the history facility.
21440 @ifset SYSTEM_READLINE
21441 @xref{Using History Interactively, , , history, GNU History Library},
21443 @ifclear SYSTEM_READLINE
21444 @xref{Using History Interactively},
21446 for the detailed description of the History library.
21448 To issue a command to @value{GDBN} without affecting certain aspects of
21449 the state which is seen by users, prefix it with @samp{server }
21450 (@pxref{Server Prefix}). This
21451 means that this command will not affect the command history, nor will it
21452 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
21453 pressed on a line by itself.
21455 @cindex @code{server}, command prefix
21456 The server prefix does not affect the recording of values into the value
21457 history; to print a value without recording it into the value history,
21458 use the @code{output} command instead of the @code{print} command.
21460 Here is the description of @value{GDBN} commands related to command
21464 @cindex history substitution
21465 @cindex history file
21466 @kindex set history filename
21467 @cindex @env{GDBHISTFILE}, environment variable
21468 @item set history filename @var{fname}
21469 Set the name of the @value{GDBN} command history file to @var{fname}.
21470 This is the file where @value{GDBN} reads an initial command history
21471 list, and where it writes the command history from this session when it
21472 exits. You can access this list through history expansion or through
21473 the history command editing characters listed below. This file defaults
21474 to the value of the environment variable @code{GDBHISTFILE}, or to
21475 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
21478 @cindex save command history
21479 @kindex set history save
21480 @item set history save
21481 @itemx set history save on
21482 Record command history in a file, whose name may be specified with the
21483 @code{set history filename} command. By default, this option is disabled.
21485 @item set history save off
21486 Stop recording command history in a file.
21488 @cindex history size
21489 @kindex set history size
21490 @cindex @env{HISTSIZE}, environment variable
21491 @item set history size @var{size}
21492 @itemx set history size unlimited
21493 Set the number of commands which @value{GDBN} keeps in its history list.
21494 This defaults to the value of the environment variable
21495 @code{HISTSIZE}, or to 256 if this variable is not set. If @var{size}
21496 is @code{unlimited}, the number of commands @value{GDBN} keeps in the
21497 history list is unlimited.
21500 History expansion assigns special meaning to the character @kbd{!}.
21501 @ifset SYSTEM_READLINE
21502 @xref{Event Designators, , , history, GNU History Library},
21504 @ifclear SYSTEM_READLINE
21505 @xref{Event Designators},
21509 @cindex history expansion, turn on/off
21510 Since @kbd{!} is also the logical not operator in C, history expansion
21511 is off by default. If you decide to enable history expansion with the
21512 @code{set history expansion on} command, you may sometimes need to
21513 follow @kbd{!} (when it is used as logical not, in an expression) with
21514 a space or a tab to prevent it from being expanded. The readline
21515 history facilities do not attempt substitution on the strings
21516 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
21518 The commands to control history expansion are:
21521 @item set history expansion on
21522 @itemx set history expansion
21523 @kindex set history expansion
21524 Enable history expansion. History expansion is off by default.
21526 @item set history expansion off
21527 Disable history expansion.
21530 @kindex show history
21532 @itemx show history filename
21533 @itemx show history save
21534 @itemx show history size
21535 @itemx show history expansion
21536 These commands display the state of the @value{GDBN} history parameters.
21537 @code{show history} by itself displays all four states.
21542 @kindex show commands
21543 @cindex show last commands
21544 @cindex display command history
21545 @item show commands
21546 Display the last ten commands in the command history.
21548 @item show commands @var{n}
21549 Print ten commands centered on command number @var{n}.
21551 @item show commands +
21552 Print ten commands just after the commands last printed.
21556 @section Screen Size
21557 @cindex size of screen
21558 @cindex pauses in output
21560 Certain commands to @value{GDBN} may produce large amounts of
21561 information output to the screen. To help you read all of it,
21562 @value{GDBN} pauses and asks you for input at the end of each page of
21563 output. Type @key{RET} when you want to continue the output, or @kbd{q}
21564 to discard the remaining output. Also, the screen width setting
21565 determines when to wrap lines of output. Depending on what is being
21566 printed, @value{GDBN} tries to break the line at a readable place,
21567 rather than simply letting it overflow onto the following line.
21569 Normally @value{GDBN} knows the size of the screen from the terminal
21570 driver software. For example, on Unix @value{GDBN} uses the termcap data base
21571 together with the value of the @code{TERM} environment variable and the
21572 @code{stty rows} and @code{stty cols} settings. If this is not correct,
21573 you can override it with the @code{set height} and @code{set
21580 @kindex show height
21581 @item set height @var{lpp}
21582 @itemx set height unlimited
21584 @itemx set width @var{cpl}
21585 @itemx set width unlimited
21587 These @code{set} commands specify a screen height of @var{lpp} lines and
21588 a screen width of @var{cpl} characters. The associated @code{show}
21589 commands display the current settings.
21591 If you specify a height of either @code{unlimited} or zero lines,
21592 @value{GDBN} does not pause during output no matter how long the
21593 output is. This is useful if output is to a file or to an editor
21596 Likewise, you can specify @samp{set width unlimited} or @samp{set
21597 width 0} to prevent @value{GDBN} from wrapping its output.
21599 @item set pagination on
21600 @itemx set pagination off
21601 @kindex set pagination
21602 Turn the output pagination on or off; the default is on. Turning
21603 pagination off is the alternative to @code{set height unlimited}. Note that
21604 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
21605 Options, -batch}) also automatically disables pagination.
21607 @item show pagination
21608 @kindex show pagination
21609 Show the current pagination mode.
21614 @cindex number representation
21615 @cindex entering numbers
21617 You can always enter numbers in octal, decimal, or hexadecimal in
21618 @value{GDBN} by the usual conventions: octal numbers begin with
21619 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
21620 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
21621 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
21622 10; likewise, the default display for numbers---when no particular
21623 format is specified---is base 10. You can change the default base for
21624 both input and output with the commands described below.
21627 @kindex set input-radix
21628 @item set input-radix @var{base}
21629 Set the default base for numeric input. Supported choices
21630 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
21631 specified either unambiguously or using the current input radix; for
21635 set input-radix 012
21636 set input-radix 10.
21637 set input-radix 0xa
21641 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
21642 leaves the input radix unchanged, no matter what it was, since
21643 @samp{10}, being without any leading or trailing signs of its base, is
21644 interpreted in the current radix. Thus, if the current radix is 16,
21645 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
21648 @kindex set output-radix
21649 @item set output-radix @var{base}
21650 Set the default base for numeric display. Supported choices
21651 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
21652 specified either unambiguously or using the current input radix.
21654 @kindex show input-radix
21655 @item show input-radix
21656 Display the current default base for numeric input.
21658 @kindex show output-radix
21659 @item show output-radix
21660 Display the current default base for numeric display.
21662 @item set radix @r{[}@var{base}@r{]}
21666 These commands set and show the default base for both input and output
21667 of numbers. @code{set radix} sets the radix of input and output to
21668 the same base; without an argument, it resets the radix back to its
21669 default value of 10.
21674 @section Configuring the Current ABI
21676 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
21677 application automatically. However, sometimes you need to override its
21678 conclusions. Use these commands to manage @value{GDBN}'s view of the
21684 @cindex Newlib OS ABI and its influence on the longjmp handling
21686 One @value{GDBN} configuration can debug binaries for multiple operating
21687 system targets, either via remote debugging or native emulation.
21688 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
21689 but you can override its conclusion using the @code{set osabi} command.
21690 One example where this is useful is in debugging of binaries which use
21691 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
21692 not have the same identifying marks that the standard C library for your
21695 When @value{GDBN} is debugging the AArch64 architecture, it provides a
21696 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
21697 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
21698 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
21702 Show the OS ABI currently in use.
21705 With no argument, show the list of registered available OS ABI's.
21707 @item set osabi @var{abi}
21708 Set the current OS ABI to @var{abi}.
21711 @cindex float promotion
21713 Generally, the way that an argument of type @code{float} is passed to a
21714 function depends on whether the function is prototyped. For a prototyped
21715 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
21716 according to the architecture's convention for @code{float}. For unprototyped
21717 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
21718 @code{double} and then passed.
21720 Unfortunately, some forms of debug information do not reliably indicate whether
21721 a function is prototyped. If @value{GDBN} calls a function that is not marked
21722 as prototyped, it consults @kbd{set coerce-float-to-double}.
21725 @kindex set coerce-float-to-double
21726 @item set coerce-float-to-double
21727 @itemx set coerce-float-to-double on
21728 Arguments of type @code{float} will be promoted to @code{double} when passed
21729 to an unprototyped function. This is the default setting.
21731 @item set coerce-float-to-double off
21732 Arguments of type @code{float} will be passed directly to unprototyped
21735 @kindex show coerce-float-to-double
21736 @item show coerce-float-to-double
21737 Show the current setting of promoting @code{float} to @code{double}.
21741 @kindex show cp-abi
21742 @value{GDBN} needs to know the ABI used for your program's C@t{++}
21743 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
21744 used to build your application. @value{GDBN} only fully supports
21745 programs with a single C@t{++} ABI; if your program contains code using
21746 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
21747 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
21748 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
21749 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
21750 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
21751 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
21756 Show the C@t{++} ABI currently in use.
21759 With no argument, show the list of supported C@t{++} ABI's.
21761 @item set cp-abi @var{abi}
21762 @itemx set cp-abi auto
21763 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
21767 @section Automatically loading associated files
21768 @cindex auto-loading
21770 @value{GDBN} sometimes reads files with commands and settings automatically,
21771 without being explicitly told so by the user. We call this feature
21772 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
21773 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
21774 results or introduce security risks (e.g., if the file comes from untrusted
21777 Note that loading of these associated files (including the local @file{.gdbinit}
21778 file) requires accordingly configured @code{auto-load safe-path}
21779 (@pxref{Auto-loading safe path}).
21781 For these reasons, @value{GDBN} includes commands and options to let you
21782 control when to auto-load files and which files should be auto-loaded.
21785 @anchor{set auto-load off}
21786 @kindex set auto-load off
21787 @item set auto-load off
21788 Globally disable loading of all auto-loaded files.
21789 You may want to use this command with the @samp{-iex} option
21790 (@pxref{Option -init-eval-command}) such as:
21792 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
21795 Be aware that system init file (@pxref{System-wide configuration})
21796 and init files from your home directory (@pxref{Home Directory Init File})
21797 still get read (as they come from generally trusted directories).
21798 To prevent @value{GDBN} from auto-loading even those init files, use the
21799 @option{-nx} option (@pxref{Mode Options}), in addition to
21800 @code{set auto-load no}.
21802 @anchor{show auto-load}
21803 @kindex show auto-load
21804 @item show auto-load
21805 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
21809 (gdb) show auto-load
21810 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
21811 libthread-db: Auto-loading of inferior specific libthread_db is on.
21812 local-gdbinit: Auto-loading of .gdbinit script from current directory
21814 python-scripts: Auto-loading of Python scripts is on.
21815 safe-path: List of directories from which it is safe to auto-load files
21816 is $debugdir:$datadir/auto-load.
21817 scripts-directory: List of directories from which to load auto-loaded scripts
21818 is $debugdir:$datadir/auto-load.
21821 @anchor{info auto-load}
21822 @kindex info auto-load
21823 @item info auto-load
21824 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
21828 (gdb) info auto-load
21831 Yes /home/user/gdb/gdb-gdb.gdb
21832 libthread-db: No auto-loaded libthread-db.
21833 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
21837 Yes /home/user/gdb/gdb-gdb.py
21841 These are various kinds of files @value{GDBN} can automatically load:
21845 @xref{objfile-gdb.py file}, controlled by @ref{set auto-load python-scripts}.
21847 @xref{objfile-gdb.gdb file}, controlled by @ref{set auto-load gdb-scripts}.
21849 @xref{dotdebug_gdb_scripts section},
21850 controlled by @ref{set auto-load python-scripts}.
21852 @xref{Init File in the Current Directory},
21853 controlled by @ref{set auto-load local-gdbinit}.
21855 @xref{libthread_db.so.1 file}, controlled by @ref{set auto-load libthread-db}.
21858 These are @value{GDBN} control commands for the auto-loading:
21860 @multitable @columnfractions .5 .5
21861 @item @xref{set auto-load off}.
21862 @tab Disable auto-loading globally.
21863 @item @xref{show auto-load}.
21864 @tab Show setting of all kinds of files.
21865 @item @xref{info auto-load}.
21866 @tab Show state of all kinds of files.
21867 @item @xref{set auto-load gdb-scripts}.
21868 @tab Control for @value{GDBN} command scripts.
21869 @item @xref{show auto-load gdb-scripts}.
21870 @tab Show setting of @value{GDBN} command scripts.
21871 @item @xref{info auto-load gdb-scripts}.
21872 @tab Show state of @value{GDBN} command scripts.
21873 @item @xref{set auto-load python-scripts}.
21874 @tab Control for @value{GDBN} Python scripts.
21875 @item @xref{show auto-load python-scripts}.
21876 @tab Show setting of @value{GDBN} Python scripts.
21877 @item @xref{info auto-load python-scripts}.
21878 @tab Show state of @value{GDBN} Python scripts.
21879 @item @xref{set auto-load scripts-directory}.
21880 @tab Control for @value{GDBN} auto-loaded scripts location.
21881 @item @xref{show auto-load scripts-directory}.
21882 @tab Show @value{GDBN} auto-loaded scripts location.
21883 @item @xref{set auto-load local-gdbinit}.
21884 @tab Control for init file in the current directory.
21885 @item @xref{show auto-load local-gdbinit}.
21886 @tab Show setting of init file in the current directory.
21887 @item @xref{info auto-load local-gdbinit}.
21888 @tab Show state of init file in the current directory.
21889 @item @xref{set auto-load libthread-db}.
21890 @tab Control for thread debugging library.
21891 @item @xref{show auto-load libthread-db}.
21892 @tab Show setting of thread debugging library.
21893 @item @xref{info auto-load libthread-db}.
21894 @tab Show state of thread debugging library.
21895 @item @xref{set auto-load safe-path}.
21896 @tab Control directories trusted for automatic loading.
21897 @item @xref{show auto-load safe-path}.
21898 @tab Show directories trusted for automatic loading.
21899 @item @xref{add-auto-load-safe-path}.
21900 @tab Add directory trusted for automatic loading.
21904 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
21905 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
21906 * objfile-gdb.gdb file:: @samp{set/show/info auto-load gdb-script}
21907 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
21908 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
21909 @xref{Python Auto-loading}.
21912 @node Init File in the Current Directory
21913 @subsection Automatically loading init file in the current directory
21914 @cindex auto-loading init file in the current directory
21916 By default, @value{GDBN} reads and executes the canned sequences of commands
21917 from init file (if any) in the current working directory,
21918 see @ref{Init File in the Current Directory during Startup}.
21920 Note that loading of this local @file{.gdbinit} file also requires accordingly
21921 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21924 @anchor{set auto-load local-gdbinit}
21925 @kindex set auto-load local-gdbinit
21926 @item set auto-load local-gdbinit [on|off]
21927 Enable or disable the auto-loading of canned sequences of commands
21928 (@pxref{Sequences}) found in init file in the current directory.
21930 @anchor{show auto-load local-gdbinit}
21931 @kindex show auto-load local-gdbinit
21932 @item show auto-load local-gdbinit
21933 Show whether auto-loading of canned sequences of commands from init file in the
21934 current directory is enabled or disabled.
21936 @anchor{info auto-load local-gdbinit}
21937 @kindex info auto-load local-gdbinit
21938 @item info auto-load local-gdbinit
21939 Print whether canned sequences of commands from init file in the
21940 current directory have been auto-loaded.
21943 @node libthread_db.so.1 file
21944 @subsection Automatically loading thread debugging library
21945 @cindex auto-loading libthread_db.so.1
21947 This feature is currently present only on @sc{gnu}/Linux native hosts.
21949 @value{GDBN} reads in some cases thread debugging library from places specific
21950 to the inferior (@pxref{set libthread-db-search-path}).
21952 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
21953 without checking this @samp{set auto-load libthread-db} switch as system
21954 libraries have to be trusted in general. In all other cases of
21955 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
21956 auto-load libthread-db} is enabled before trying to open such thread debugging
21959 Note that loading of this debugging library also requires accordingly configured
21960 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21963 @anchor{set auto-load libthread-db}
21964 @kindex set auto-load libthread-db
21965 @item set auto-load libthread-db [on|off]
21966 Enable or disable the auto-loading of inferior specific thread debugging library.
21968 @anchor{show auto-load libthread-db}
21969 @kindex show auto-load libthread-db
21970 @item show auto-load libthread-db
21971 Show whether auto-loading of inferior specific thread debugging library is
21972 enabled or disabled.
21974 @anchor{info auto-load libthread-db}
21975 @kindex info auto-load libthread-db
21976 @item info auto-load libthread-db
21977 Print the list of all loaded inferior specific thread debugging libraries and
21978 for each such library print list of inferior @var{pid}s using it.
21981 @node objfile-gdb.gdb file
21982 @subsection The @file{@var{objfile}-gdb.gdb} file
21983 @cindex auto-loading @file{@var{objfile}-gdb.gdb}
21985 @value{GDBN} tries to load an @file{@var{objfile}-gdb.gdb} file containing
21986 canned sequences of commands (@pxref{Sequences}), as long as @samp{set
21987 auto-load gdb-scripts} is set to @samp{on}.
21989 Note that loading of this script file also requires accordingly configured
21990 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21992 For more background refer to the similar Python scripts auto-loading
21993 description (@pxref{objfile-gdb.py file}).
21996 @anchor{set auto-load gdb-scripts}
21997 @kindex set auto-load gdb-scripts
21998 @item set auto-load gdb-scripts [on|off]
21999 Enable or disable the auto-loading of canned sequences of commands scripts.
22001 @anchor{show auto-load gdb-scripts}
22002 @kindex show auto-load gdb-scripts
22003 @item show auto-load gdb-scripts
22004 Show whether auto-loading of canned sequences of commands scripts is enabled or
22007 @anchor{info auto-load gdb-scripts}
22008 @kindex info auto-load gdb-scripts
22009 @cindex print list of auto-loaded canned sequences of commands scripts
22010 @item info auto-load gdb-scripts [@var{regexp}]
22011 Print the list of all canned sequences of commands scripts that @value{GDBN}
22015 If @var{regexp} is supplied only canned sequences of commands scripts with
22016 matching names are printed.
22018 @node Auto-loading safe path
22019 @subsection Security restriction for auto-loading
22020 @cindex auto-loading safe-path
22022 As the files of inferior can come from untrusted source (such as submitted by
22023 an application user) @value{GDBN} does not always load any files automatically.
22024 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
22025 directories trusted for loading files not explicitly requested by user.
22026 Each directory can also be a shell wildcard pattern.
22028 If the path is not set properly you will see a warning and the file will not
22033 Reading symbols from /home/user/gdb/gdb...done.
22034 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
22035 declined by your `auto-load safe-path' set
22036 to "$debugdir:$datadir/auto-load".
22037 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
22038 declined by your `auto-load safe-path' set
22039 to "$debugdir:$datadir/auto-load".
22043 To instruct @value{GDBN} to go ahead and use the init files anyway,
22044 invoke @value{GDBN} like this:
22047 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
22050 The list of trusted directories is controlled by the following commands:
22053 @anchor{set auto-load safe-path}
22054 @kindex set auto-load safe-path
22055 @item set auto-load safe-path @r{[}@var{directories}@r{]}
22056 Set the list of directories (and their subdirectories) trusted for automatic
22057 loading and execution of scripts. You can also enter a specific trusted file.
22058 Each directory can also be a shell wildcard pattern; wildcards do not match
22059 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
22060 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
22061 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
22062 its default value as specified during @value{GDBN} compilation.
22064 The list of directories uses path separator (@samp{:} on GNU and Unix
22065 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
22066 to the @env{PATH} environment variable.
22068 @anchor{show auto-load safe-path}
22069 @kindex show auto-load safe-path
22070 @item show auto-load safe-path
22071 Show the list of directories trusted for automatic loading and execution of
22074 @anchor{add-auto-load-safe-path}
22075 @kindex add-auto-load-safe-path
22076 @item add-auto-load-safe-path
22077 Add an entry (or list of entries) the list of directories trusted for automatic
22078 loading and execution of scripts. Multiple entries may be delimited by the
22079 host platform path separator in use.
22082 This variable defaults to what @code{--with-auto-load-dir} has been configured
22083 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
22084 substitution applies the same as for @ref{set auto-load scripts-directory}.
22085 The default @code{set auto-load safe-path} value can be also overriden by
22086 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
22088 Setting this variable to @file{/} disables this security protection,
22089 corresponding @value{GDBN} configuration option is
22090 @option{--without-auto-load-safe-path}.
22091 This variable is supposed to be set to the system directories writable by the
22092 system superuser only. Users can add their source directories in init files in
22093 their home directories (@pxref{Home Directory Init File}). See also deprecated
22094 init file in the current directory
22095 (@pxref{Init File in the Current Directory during Startup}).
22097 To force @value{GDBN} to load the files it declined to load in the previous
22098 example, you could use one of the following ways:
22101 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
22102 Specify this trusted directory (or a file) as additional component of the list.
22103 You have to specify also any existing directories displayed by
22104 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
22106 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
22107 Specify this directory as in the previous case but just for a single
22108 @value{GDBN} session.
22110 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
22111 Disable auto-loading safety for a single @value{GDBN} session.
22112 This assumes all the files you debug during this @value{GDBN} session will come
22113 from trusted sources.
22115 @item @kbd{./configure --without-auto-load-safe-path}
22116 During compilation of @value{GDBN} you may disable any auto-loading safety.
22117 This assumes all the files you will ever debug with this @value{GDBN} come from
22121 On the other hand you can also explicitly forbid automatic files loading which
22122 also suppresses any such warning messages:
22125 @item @kbd{gdb -iex "set auto-load no" @dots{}}
22126 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
22128 @item @file{~/.gdbinit}: @samp{set auto-load no}
22129 Disable auto-loading globally for the user
22130 (@pxref{Home Directory Init File}). While it is improbable, you could also
22131 use system init file instead (@pxref{System-wide configuration}).
22134 This setting applies to the file names as entered by user. If no entry matches
22135 @value{GDBN} tries as a last resort to also resolve all the file names into
22136 their canonical form (typically resolving symbolic links) and compare the
22137 entries again. @value{GDBN} already canonicalizes most of the filenames on its
22138 own before starting the comparison so a canonical form of directories is
22139 recommended to be entered.
22141 @node Auto-loading verbose mode
22142 @subsection Displaying files tried for auto-load
22143 @cindex auto-loading verbose mode
22145 For better visibility of all the file locations where you can place scripts to
22146 be auto-loaded with inferior --- or to protect yourself against accidental
22147 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
22148 all the files attempted to be loaded. Both existing and non-existing files may
22151 For example the list of directories from which it is safe to auto-load files
22152 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
22153 may not be too obvious while setting it up.
22156 (gdb) set debug auto-load on
22157 (gdb) file ~/src/t/true
22158 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
22159 for objfile "/tmp/true".
22160 auto-load: Updating directories of "/usr:/opt".
22161 auto-load: Using directory "/usr".
22162 auto-load: Using directory "/opt".
22163 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
22164 by your `auto-load safe-path' set to "/usr:/opt".
22168 @anchor{set debug auto-load}
22169 @kindex set debug auto-load
22170 @item set debug auto-load [on|off]
22171 Set whether to print the filenames attempted to be auto-loaded.
22173 @anchor{show debug auto-load}
22174 @kindex show debug auto-load
22175 @item show debug auto-load
22176 Show whether printing of the filenames attempted to be auto-loaded is turned
22180 @node Messages/Warnings
22181 @section Optional Warnings and Messages
22183 @cindex verbose operation
22184 @cindex optional warnings
22185 By default, @value{GDBN} is silent about its inner workings. If you are
22186 running on a slow machine, you may want to use the @code{set verbose}
22187 command. This makes @value{GDBN} tell you when it does a lengthy
22188 internal operation, so you will not think it has crashed.
22190 Currently, the messages controlled by @code{set verbose} are those
22191 which announce that the symbol table for a source file is being read;
22192 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
22195 @kindex set verbose
22196 @item set verbose on
22197 Enables @value{GDBN} output of certain informational messages.
22199 @item set verbose off
22200 Disables @value{GDBN} output of certain informational messages.
22202 @kindex show verbose
22204 Displays whether @code{set verbose} is on or off.
22207 By default, if @value{GDBN} encounters bugs in the symbol table of an
22208 object file, it is silent; but if you are debugging a compiler, you may
22209 find this information useful (@pxref{Symbol Errors, ,Errors Reading
22214 @kindex set complaints
22215 @item set complaints @var{limit}
22216 Permits @value{GDBN} to output @var{limit} complaints about each type of
22217 unusual symbols before becoming silent about the problem. Set
22218 @var{limit} to zero to suppress all complaints; set it to a large number
22219 to prevent complaints from being suppressed.
22221 @kindex show complaints
22222 @item show complaints
22223 Displays how many symbol complaints @value{GDBN} is permitted to produce.
22227 @anchor{confirmation requests}
22228 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
22229 lot of stupid questions to confirm certain commands. For example, if
22230 you try to run a program which is already running:
22234 The program being debugged has been started already.
22235 Start it from the beginning? (y or n)
22238 If you are willing to unflinchingly face the consequences of your own
22239 commands, you can disable this ``feature'':
22243 @kindex set confirm
22245 @cindex confirmation
22246 @cindex stupid questions
22247 @item set confirm off
22248 Disables confirmation requests. Note that running @value{GDBN} with
22249 the @option{--batch} option (@pxref{Mode Options, -batch}) also
22250 automatically disables confirmation requests.
22252 @item set confirm on
22253 Enables confirmation requests (the default).
22255 @kindex show confirm
22257 Displays state of confirmation requests.
22261 @cindex command tracing
22262 If you need to debug user-defined commands or sourced files you may find it
22263 useful to enable @dfn{command tracing}. In this mode each command will be
22264 printed as it is executed, prefixed with one or more @samp{+} symbols, the
22265 quantity denoting the call depth of each command.
22268 @kindex set trace-commands
22269 @cindex command scripts, debugging
22270 @item set trace-commands on
22271 Enable command tracing.
22272 @item set trace-commands off
22273 Disable command tracing.
22274 @item show trace-commands
22275 Display the current state of command tracing.
22278 @node Debugging Output
22279 @section Optional Messages about Internal Happenings
22280 @cindex optional debugging messages
22282 @value{GDBN} has commands that enable optional debugging messages from
22283 various @value{GDBN} subsystems; normally these commands are of
22284 interest to @value{GDBN} maintainers, or when reporting a bug. This
22285 section documents those commands.
22288 @kindex set exec-done-display
22289 @item set exec-done-display
22290 Turns on or off the notification of asynchronous commands'
22291 completion. When on, @value{GDBN} will print a message when an
22292 asynchronous command finishes its execution. The default is off.
22293 @kindex show exec-done-display
22294 @item show exec-done-display
22295 Displays the current setting of asynchronous command completion
22298 @cindex ARM AArch64
22299 @item set debug aarch64
22300 Turns on or off display of debugging messages related to ARM AArch64.
22301 The default is off.
22303 @item show debug aarch64
22304 Displays the current state of displaying debugging messages related to
22306 @cindex gdbarch debugging info
22307 @cindex architecture debugging info
22308 @item set debug arch
22309 Turns on or off display of gdbarch debugging info. The default is off
22310 @item show debug arch
22311 Displays the current state of displaying gdbarch debugging info.
22312 @item set debug aix-solib
22313 @cindex AIX shared library debugging
22314 Control display of debugging messages from the AIX shared library
22315 support module. The default is off.
22316 @item show debug aix-thread
22317 Show the current state of displaying AIX shared library debugging messages.
22318 @item set debug aix-thread
22319 @cindex AIX threads
22320 Display debugging messages about inner workings of the AIX thread
22322 @item show debug aix-thread
22323 Show the current state of AIX thread debugging info display.
22324 @item set debug check-physname
22326 Check the results of the ``physname'' computation. When reading DWARF
22327 debugging information for C@t{++}, @value{GDBN} attempts to compute
22328 each entity's name. @value{GDBN} can do this computation in two
22329 different ways, depending on exactly what information is present.
22330 When enabled, this setting causes @value{GDBN} to compute the names
22331 both ways and display any discrepancies.
22332 @item show debug check-physname
22333 Show the current state of ``physname'' checking.
22334 @item set debug coff-pe-read
22335 @cindex COFF/PE exported symbols
22336 Control display of debugging messages related to reading of COFF/PE
22337 exported symbols. The default is off.
22338 @item show debug coff-pe-read
22339 Displays the current state of displaying debugging messages related to
22340 reading of COFF/PE exported symbols.
22341 @item set debug dwarf2-die
22342 @cindex DWARF2 DIEs
22343 Dump DWARF2 DIEs after they are read in.
22344 The value is the number of nesting levels to print.
22345 A value of zero turns off the display.
22346 @item show debug dwarf2-die
22347 Show the current state of DWARF2 DIE debugging.
22348 @item set debug dwarf2-read
22349 @cindex DWARF2 Reading
22350 Turns on or off display of debugging messages related to reading
22351 DWARF debug info. The default is off.
22352 @item show debug dwarf2-read
22353 Show the current state of DWARF2 reader debugging.
22354 @item set debug displaced
22355 @cindex displaced stepping debugging info
22356 Turns on or off display of @value{GDBN} debugging info for the
22357 displaced stepping support. The default is off.
22358 @item show debug displaced
22359 Displays the current state of displaying @value{GDBN} debugging info
22360 related to displaced stepping.
22361 @item set debug event
22362 @cindex event debugging info
22363 Turns on or off display of @value{GDBN} event debugging info. The
22365 @item show debug event
22366 Displays the current state of displaying @value{GDBN} event debugging
22368 @item set debug expression
22369 @cindex expression debugging info
22370 Turns on or off display of debugging info about @value{GDBN}
22371 expression parsing. The default is off.
22372 @item show debug expression
22373 Displays the current state of displaying debugging info about
22374 @value{GDBN} expression parsing.
22375 @item set debug frame
22376 @cindex frame debugging info
22377 Turns on or off display of @value{GDBN} frame debugging info. The
22379 @item show debug frame
22380 Displays the current state of displaying @value{GDBN} frame debugging
22382 @item set debug gnu-nat
22383 @cindex @sc{gnu}/Hurd debug messages
22384 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
22385 @item show debug gnu-nat
22386 Show the current state of @sc{gnu}/Hurd debugging messages.
22387 @item set debug infrun
22388 @cindex inferior debugging info
22389 Turns on or off display of @value{GDBN} debugging info for running the inferior.
22390 The default is off. @file{infrun.c} contains GDB's runtime state machine used
22391 for implementing operations such as single-stepping the inferior.
22392 @item show debug infrun
22393 Displays the current state of @value{GDBN} inferior debugging.
22394 @item set debug jit
22395 @cindex just-in-time compilation, debugging messages
22396 Turns on or off debugging messages from JIT debug support.
22397 @item show debug jit
22398 Displays the current state of @value{GDBN} JIT debugging.
22399 @item set debug lin-lwp
22400 @cindex @sc{gnu}/Linux LWP debug messages
22401 @cindex Linux lightweight processes
22402 Turns on or off debugging messages from the Linux LWP debug support.
22403 @item show debug lin-lwp
22404 Show the current state of Linux LWP debugging messages.
22405 @item set debug mach-o
22406 @cindex Mach-O symbols processing
22407 Control display of debugging messages related to Mach-O symbols
22408 processing. The default is off.
22409 @item show debug mach-o
22410 Displays the current state of displaying debugging messages related to
22411 reading of COFF/PE exported symbols.
22412 @item set debug notification
22413 @cindex remote async notification debugging info
22414 Turns on or off debugging messages about remote async notification.
22415 The default is off.
22416 @item show debug notification
22417 Displays the current state of remote async notification debugging messages.
22418 @item set debug observer
22419 @cindex observer debugging info
22420 Turns on or off display of @value{GDBN} observer debugging. This
22421 includes info such as the notification of observable events.
22422 @item show debug observer
22423 Displays the current state of observer debugging.
22424 @item set debug overload
22425 @cindex C@t{++} overload debugging info
22426 Turns on or off display of @value{GDBN} C@t{++} overload debugging
22427 info. This includes info such as ranking of functions, etc. The default
22429 @item show debug overload
22430 Displays the current state of displaying @value{GDBN} C@t{++} overload
22432 @cindex expression parser, debugging info
22433 @cindex debug expression parser
22434 @item set debug parser
22435 Turns on or off the display of expression parser debugging output.
22436 Internally, this sets the @code{yydebug} variable in the expression
22437 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
22438 details. The default is off.
22439 @item show debug parser
22440 Show the current state of expression parser debugging.
22441 @cindex packets, reporting on stdout
22442 @cindex serial connections, debugging
22443 @cindex debug remote protocol
22444 @cindex remote protocol debugging
22445 @cindex display remote packets
22446 @item set debug remote
22447 Turns on or off display of reports on all packets sent back and forth across
22448 the serial line to the remote machine. The info is printed on the
22449 @value{GDBN} standard output stream. The default is off.
22450 @item show debug remote
22451 Displays the state of display of remote packets.
22452 @item set debug serial
22453 Turns on or off display of @value{GDBN} serial debugging info. The
22455 @item show debug serial
22456 Displays the current state of displaying @value{GDBN} serial debugging
22458 @item set debug solib-frv
22459 @cindex FR-V shared-library debugging
22460 Turns on or off debugging messages for FR-V shared-library code.
22461 @item show debug solib-frv
22462 Display the current state of FR-V shared-library code debugging
22464 @item set debug symtab-create
22465 @cindex symbol table creation
22466 Turns on or off display of debugging messages related to symbol table creation.
22467 The default is off.
22468 @item show debug symtab-create
22469 Show the current state of symbol table creation debugging.
22470 @item set debug target
22471 @cindex target debugging info
22472 Turns on or off display of @value{GDBN} target debugging info. This info
22473 includes what is going on at the target level of GDB, as it happens. The
22474 default is 0. Set it to 1 to track events, and to 2 to also track the
22475 value of large memory transfers. Changes to this flag do not take effect
22476 until the next time you connect to a target or use the @code{run} command.
22477 @item show debug target
22478 Displays the current state of displaying @value{GDBN} target debugging
22480 @item set debug timestamp
22481 @cindex timestampping debugging info
22482 Turns on or off display of timestamps with @value{GDBN} debugging info.
22483 When enabled, seconds and microseconds are displayed before each debugging
22485 @item show debug timestamp
22486 Displays the current state of displaying timestamps with @value{GDBN}
22488 @item set debugvarobj
22489 @cindex variable object debugging info
22490 Turns on or off display of @value{GDBN} variable object debugging
22491 info. The default is off.
22492 @item show debugvarobj
22493 Displays the current state of displaying @value{GDBN} variable object
22495 @item set debug xml
22496 @cindex XML parser debugging
22497 Turns on or off debugging messages for built-in XML parsers.
22498 @item show debug xml
22499 Displays the current state of XML debugging messages.
22502 @node Other Misc Settings
22503 @section Other Miscellaneous Settings
22504 @cindex miscellaneous settings
22507 @kindex set interactive-mode
22508 @item set interactive-mode
22509 If @code{on}, forces @value{GDBN} to assume that GDB was started
22510 in a terminal. In practice, this means that @value{GDBN} should wait
22511 for the user to answer queries generated by commands entered at
22512 the command prompt. If @code{off}, forces @value{GDBN} to operate
22513 in the opposite mode, and it uses the default answers to all queries.
22514 If @code{auto} (the default), @value{GDBN} tries to determine whether
22515 its standard input is a terminal, and works in interactive-mode if it
22516 is, non-interactively otherwise.
22518 In the vast majority of cases, the debugger should be able to guess
22519 correctly which mode should be used. But this setting can be useful
22520 in certain specific cases, such as running a MinGW @value{GDBN}
22521 inside a cygwin window.
22523 @kindex show interactive-mode
22524 @item show interactive-mode
22525 Displays whether the debugger is operating in interactive mode or not.
22528 @node Extending GDB
22529 @chapter Extending @value{GDBN}
22530 @cindex extending GDB
22532 @value{GDBN} provides three mechanisms for extension. The first is based
22533 on composition of @value{GDBN} commands, the second is based on the
22534 Python scripting language, and the third is for defining new aliases of
22537 To facilitate the use of the first two extensions, @value{GDBN} is capable
22538 of evaluating the contents of a file. When doing so, @value{GDBN}
22539 can recognize which scripting language is being used by looking at
22540 the filename extension. Files with an unrecognized filename extension
22541 are always treated as a @value{GDBN} Command Files.
22542 @xref{Command Files,, Command files}.
22544 You can control how @value{GDBN} evaluates these files with the following
22548 @kindex set script-extension
22549 @kindex show script-extension
22550 @item set script-extension off
22551 All scripts are always evaluated as @value{GDBN} Command Files.
22553 @item set script-extension soft
22554 The debugger determines the scripting language based on filename
22555 extension. If this scripting language is supported, @value{GDBN}
22556 evaluates the script using that language. Otherwise, it evaluates
22557 the file as a @value{GDBN} Command File.
22559 @item set script-extension strict
22560 The debugger determines the scripting language based on filename
22561 extension, and evaluates the script using that language. If the
22562 language is not supported, then the evaluation fails.
22564 @item show script-extension
22565 Display the current value of the @code{script-extension} option.
22570 * Sequences:: Canned Sequences of Commands
22571 * Python:: Scripting @value{GDBN} using Python
22572 * Aliases:: Creating new spellings of existing commands
22576 @section Canned Sequences of Commands
22578 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
22579 Command Lists}), @value{GDBN} provides two ways to store sequences of
22580 commands for execution as a unit: user-defined commands and command
22584 * Define:: How to define your own commands
22585 * Hooks:: Hooks for user-defined commands
22586 * Command Files:: How to write scripts of commands to be stored in a file
22587 * Output:: Commands for controlled output
22591 @subsection User-defined Commands
22593 @cindex user-defined command
22594 @cindex arguments, to user-defined commands
22595 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
22596 which you assign a new name as a command. This is done with the
22597 @code{define} command. User commands may accept up to 10 arguments
22598 separated by whitespace. Arguments are accessed within the user command
22599 via @code{$arg0@dots{}$arg9}. A trivial example:
22603 print $arg0 + $arg1 + $arg2
22608 To execute the command use:
22615 This defines the command @code{adder}, which prints the sum of
22616 its three arguments. Note the arguments are text substitutions, so they may
22617 reference variables, use complex expressions, or even perform inferior
22620 @cindex argument count in user-defined commands
22621 @cindex how many arguments (user-defined commands)
22622 In addition, @code{$argc} may be used to find out how many arguments have
22623 been passed. This expands to a number in the range 0@dots{}10.
22628 print $arg0 + $arg1
22631 print $arg0 + $arg1 + $arg2
22639 @item define @var{commandname}
22640 Define a command named @var{commandname}. If there is already a command
22641 by that name, you are asked to confirm that you want to redefine it.
22642 @var{commandname} may be a bare command name consisting of letters,
22643 numbers, dashes, and underscores. It may also start with any predefined
22644 prefix command. For example, @samp{define target my-target} creates
22645 a user-defined @samp{target my-target} command.
22647 The definition of the command is made up of other @value{GDBN} command lines,
22648 which are given following the @code{define} command. The end of these
22649 commands is marked by a line containing @code{end}.
22652 @kindex end@r{ (user-defined commands)}
22653 @item document @var{commandname}
22654 Document the user-defined command @var{commandname}, so that it can be
22655 accessed by @code{help}. The command @var{commandname} must already be
22656 defined. This command reads lines of documentation just as @code{define}
22657 reads the lines of the command definition, ending with @code{end}.
22658 After the @code{document} command is finished, @code{help} on command
22659 @var{commandname} displays the documentation you have written.
22661 You may use the @code{document} command again to change the
22662 documentation of a command. Redefining the command with @code{define}
22663 does not change the documentation.
22665 @kindex dont-repeat
22666 @cindex don't repeat command
22668 Used inside a user-defined command, this tells @value{GDBN} that this
22669 command should not be repeated when the user hits @key{RET}
22670 (@pxref{Command Syntax, repeat last command}).
22672 @kindex help user-defined
22673 @item help user-defined
22674 List all user-defined commands and all python commands defined in class
22675 COMAND_USER. The first line of the documentation or docstring is
22680 @itemx show user @var{commandname}
22681 Display the @value{GDBN} commands used to define @var{commandname} (but
22682 not its documentation). If no @var{commandname} is given, display the
22683 definitions for all user-defined commands.
22684 This does not work for user-defined python commands.
22686 @cindex infinite recursion in user-defined commands
22687 @kindex show max-user-call-depth
22688 @kindex set max-user-call-depth
22689 @item show max-user-call-depth
22690 @itemx set max-user-call-depth
22691 The value of @code{max-user-call-depth} controls how many recursion
22692 levels are allowed in user-defined commands before @value{GDBN} suspects an
22693 infinite recursion and aborts the command.
22694 This does not apply to user-defined python commands.
22697 In addition to the above commands, user-defined commands frequently
22698 use control flow commands, described in @ref{Command Files}.
22700 When user-defined commands are executed, the
22701 commands of the definition are not printed. An error in any command
22702 stops execution of the user-defined command.
22704 If used interactively, commands that would ask for confirmation proceed
22705 without asking when used inside a user-defined command. Many @value{GDBN}
22706 commands that normally print messages to say what they are doing omit the
22707 messages when used in a user-defined command.
22710 @subsection User-defined Command Hooks
22711 @cindex command hooks
22712 @cindex hooks, for commands
22713 @cindex hooks, pre-command
22716 You may define @dfn{hooks}, which are a special kind of user-defined
22717 command. Whenever you run the command @samp{foo}, if the user-defined
22718 command @samp{hook-foo} exists, it is executed (with no arguments)
22719 before that command.
22721 @cindex hooks, post-command
22723 A hook may also be defined which is run after the command you executed.
22724 Whenever you run the command @samp{foo}, if the user-defined command
22725 @samp{hookpost-foo} exists, it is executed (with no arguments) after
22726 that command. Post-execution hooks may exist simultaneously with
22727 pre-execution hooks, for the same command.
22729 It is valid for a hook to call the command which it hooks. If this
22730 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
22732 @c It would be nice if hookpost could be passed a parameter indicating
22733 @c if the command it hooks executed properly or not. FIXME!
22735 @kindex stop@r{, a pseudo-command}
22736 In addition, a pseudo-command, @samp{stop} exists. Defining
22737 (@samp{hook-stop}) makes the associated commands execute every time
22738 execution stops in your program: before breakpoint commands are run,
22739 displays are printed, or the stack frame is printed.
22741 For example, to ignore @code{SIGALRM} signals while
22742 single-stepping, but treat them normally during normal execution,
22747 handle SIGALRM nopass
22751 handle SIGALRM pass
22754 define hook-continue
22755 handle SIGALRM pass
22759 As a further example, to hook at the beginning and end of the @code{echo}
22760 command, and to add extra text to the beginning and end of the message,
22768 define hookpost-echo
22772 (@value{GDBP}) echo Hello World
22773 <<<---Hello World--->>>
22778 You can define a hook for any single-word command in @value{GDBN}, but
22779 not for command aliases; you should define a hook for the basic command
22780 name, e.g.@: @code{backtrace} rather than @code{bt}.
22781 @c FIXME! So how does Joe User discover whether a command is an alias
22783 You can hook a multi-word command by adding @code{hook-} or
22784 @code{hookpost-} to the last word of the command, e.g.@:
22785 @samp{define target hook-remote} to add a hook to @samp{target remote}.
22787 If an error occurs during the execution of your hook, execution of
22788 @value{GDBN} commands stops and @value{GDBN} issues a prompt
22789 (before the command that you actually typed had a chance to run).
22791 If you try to define a hook which does not match any known command, you
22792 get a warning from the @code{define} command.
22794 @node Command Files
22795 @subsection Command Files
22797 @cindex command files
22798 @cindex scripting commands
22799 A command file for @value{GDBN} is a text file made of lines that are
22800 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
22801 also be included. An empty line in a command file does nothing; it
22802 does not mean to repeat the last command, as it would from the
22805 You can request the execution of a command file with the @code{source}
22806 command. Note that the @code{source} command is also used to evaluate
22807 scripts that are not Command Files. The exact behavior can be configured
22808 using the @code{script-extension} setting.
22809 @xref{Extending GDB,, Extending GDB}.
22813 @cindex execute commands from a file
22814 @item source [-s] [-v] @var{filename}
22815 Execute the command file @var{filename}.
22818 The lines in a command file are generally executed sequentially,
22819 unless the order of execution is changed by one of the
22820 @emph{flow-control commands} described below. The commands are not
22821 printed as they are executed. An error in any command terminates
22822 execution of the command file and control is returned to the console.
22824 @value{GDBN} first searches for @var{filename} in the current directory.
22825 If the file is not found there, and @var{filename} does not specify a
22826 directory, then @value{GDBN} also looks for the file on the source search path
22827 (specified with the @samp{directory} command);
22828 except that @file{$cdir} is not searched because the compilation directory
22829 is not relevant to scripts.
22831 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
22832 on the search path even if @var{filename} specifies a directory.
22833 The search is done by appending @var{filename} to each element of the
22834 search path. So, for example, if @var{filename} is @file{mylib/myscript}
22835 and the search path contains @file{/home/user} then @value{GDBN} will
22836 look for the script @file{/home/user/mylib/myscript}.
22837 The search is also done if @var{filename} is an absolute path.
22838 For example, if @var{filename} is @file{/tmp/myscript} and
22839 the search path contains @file{/home/user} then @value{GDBN} will
22840 look for the script @file{/home/user/tmp/myscript}.
22841 For DOS-like systems, if @var{filename} contains a drive specification,
22842 it is stripped before concatenation. For example, if @var{filename} is
22843 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
22844 will look for the script @file{c:/tmp/myscript}.
22846 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
22847 each command as it is executed. The option must be given before
22848 @var{filename}, and is interpreted as part of the filename anywhere else.
22850 Commands that would ask for confirmation if used interactively proceed
22851 without asking when used in a command file. Many @value{GDBN} commands that
22852 normally print messages to say what they are doing omit the messages
22853 when called from command files.
22855 @value{GDBN} also accepts command input from standard input. In this
22856 mode, normal output goes to standard output and error output goes to
22857 standard error. Errors in a command file supplied on standard input do
22858 not terminate execution of the command file---execution continues with
22862 gdb < cmds > log 2>&1
22865 (The syntax above will vary depending on the shell used.) This example
22866 will execute commands from the file @file{cmds}. All output and errors
22867 would be directed to @file{log}.
22869 Since commands stored on command files tend to be more general than
22870 commands typed interactively, they frequently need to deal with
22871 complicated situations, such as different or unexpected values of
22872 variables and symbols, changes in how the program being debugged is
22873 built, etc. @value{GDBN} provides a set of flow-control commands to
22874 deal with these complexities. Using these commands, you can write
22875 complex scripts that loop over data structures, execute commands
22876 conditionally, etc.
22883 This command allows to include in your script conditionally executed
22884 commands. The @code{if} command takes a single argument, which is an
22885 expression to evaluate. It is followed by a series of commands that
22886 are executed only if the expression is true (its value is nonzero).
22887 There can then optionally be an @code{else} line, followed by a series
22888 of commands that are only executed if the expression was false. The
22889 end of the list is marked by a line containing @code{end}.
22893 This command allows to write loops. Its syntax is similar to
22894 @code{if}: the command takes a single argument, which is an expression
22895 to evaluate, and must be followed by the commands to execute, one per
22896 line, terminated by an @code{end}. These commands are called the
22897 @dfn{body} of the loop. The commands in the body of @code{while} are
22898 executed repeatedly as long as the expression evaluates to true.
22902 This command exits the @code{while} loop in whose body it is included.
22903 Execution of the script continues after that @code{while}s @code{end}
22906 @kindex loop_continue
22907 @item loop_continue
22908 This command skips the execution of the rest of the body of commands
22909 in the @code{while} loop in whose body it is included. Execution
22910 branches to the beginning of the @code{while} loop, where it evaluates
22911 the controlling expression.
22913 @kindex end@r{ (if/else/while commands)}
22915 Terminate the block of commands that are the body of @code{if},
22916 @code{else}, or @code{while} flow-control commands.
22921 @subsection Commands for Controlled Output
22923 During the execution of a command file or a user-defined command, normal
22924 @value{GDBN} output is suppressed; the only output that appears is what is
22925 explicitly printed by the commands in the definition. This section
22926 describes three commands useful for generating exactly the output you
22931 @item echo @var{text}
22932 @c I do not consider backslash-space a standard C escape sequence
22933 @c because it is not in ANSI.
22934 Print @var{text}. Nonprinting characters can be included in
22935 @var{text} using C escape sequences, such as @samp{\n} to print a
22936 newline. @strong{No newline is printed unless you specify one.}
22937 In addition to the standard C escape sequences, a backslash followed
22938 by a space stands for a space. This is useful for displaying a
22939 string with spaces at the beginning or the end, since leading and
22940 trailing spaces are otherwise trimmed from all arguments.
22941 To print @samp{@w{ }and foo =@w{ }}, use the command
22942 @samp{echo \@w{ }and foo = \@w{ }}.
22944 A backslash at the end of @var{text} can be used, as in C, to continue
22945 the command onto subsequent lines. For example,
22948 echo This is some text\n\
22949 which is continued\n\
22950 onto several lines.\n
22953 produces the same output as
22956 echo This is some text\n
22957 echo which is continued\n
22958 echo onto several lines.\n
22962 @item output @var{expression}
22963 Print the value of @var{expression} and nothing but that value: no
22964 newlines, no @samp{$@var{nn} = }. The value is not entered in the
22965 value history either. @xref{Expressions, ,Expressions}, for more information
22968 @item output/@var{fmt} @var{expression}
22969 Print the value of @var{expression} in format @var{fmt}. You can use
22970 the same formats as for @code{print}. @xref{Output Formats,,Output
22971 Formats}, for more information.
22974 @item printf @var{template}, @var{expressions}@dots{}
22975 Print the values of one or more @var{expressions} under the control of
22976 the string @var{template}. To print several values, make
22977 @var{expressions} be a comma-separated list of individual expressions,
22978 which may be either numbers or pointers. Their values are printed as
22979 specified by @var{template}, exactly as a C program would do by
22980 executing the code below:
22983 printf (@var{template}, @var{expressions}@dots{});
22986 As in @code{C} @code{printf}, ordinary characters in @var{template}
22987 are printed verbatim, while @dfn{conversion specification} introduced
22988 by the @samp{%} character cause subsequent @var{expressions} to be
22989 evaluated, their values converted and formatted according to type and
22990 style information encoded in the conversion specifications, and then
22993 For example, you can print two values in hex like this:
22996 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
22999 @code{printf} supports all the standard @code{C} conversion
23000 specifications, including the flags and modifiers between the @samp{%}
23001 character and the conversion letter, with the following exceptions:
23005 The argument-ordering modifiers, such as @samp{2$}, are not supported.
23008 The modifier @samp{*} is not supported for specifying precision or
23012 The @samp{'} flag (for separation of digits into groups according to
23013 @code{LC_NUMERIC'}) is not supported.
23016 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
23020 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
23023 The conversion letters @samp{a} and @samp{A} are not supported.
23027 Note that the @samp{ll} type modifier is supported only if the
23028 underlying @code{C} implementation used to build @value{GDBN} supports
23029 the @code{long long int} type, and the @samp{L} type modifier is
23030 supported only if @code{long double} type is available.
23032 As in @code{C}, @code{printf} supports simple backslash-escape
23033 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
23034 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
23035 single character. Octal and hexadecimal escape sequences are not
23038 Additionally, @code{printf} supports conversion specifications for DFP
23039 (@dfn{Decimal Floating Point}) types using the following length modifiers
23040 together with a floating point specifier.
23045 @samp{H} for printing @code{Decimal32} types.
23048 @samp{D} for printing @code{Decimal64} types.
23051 @samp{DD} for printing @code{Decimal128} types.
23054 If the underlying @code{C} implementation used to build @value{GDBN} has
23055 support for the three length modifiers for DFP types, other modifiers
23056 such as width and precision will also be available for @value{GDBN} to use.
23058 In case there is no such @code{C} support, no additional modifiers will be
23059 available and the value will be printed in the standard way.
23061 Here's an example of printing DFP types using the above conversion letters:
23063 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
23067 @item eval @var{template}, @var{expressions}@dots{}
23068 Convert the values of one or more @var{expressions} under the control of
23069 the string @var{template} to a command line, and call it.
23074 @section Scripting @value{GDBN} using Python
23075 @cindex python scripting
23076 @cindex scripting with python
23078 You can script @value{GDBN} using the @uref{http://www.python.org/,
23079 Python programming language}. This feature is available only if
23080 @value{GDBN} was configured using @option{--with-python}.
23082 @cindex python directory
23083 Python scripts used by @value{GDBN} should be installed in
23084 @file{@var{data-directory}/python}, where @var{data-directory} is
23085 the data directory as determined at @value{GDBN} startup (@pxref{Data Files}).
23086 This directory, known as the @dfn{python directory},
23087 is automatically added to the Python Search Path in order to allow
23088 the Python interpreter to locate all scripts installed at this location.
23090 Additionally, @value{GDBN} commands and convenience functions which
23091 are written in Python and are located in the
23092 @file{@var{data-directory}/python/gdb/command} or
23093 @file{@var{data-directory}/python/gdb/function} directories are
23094 automatically imported when @value{GDBN} starts.
23097 * Python Commands:: Accessing Python from @value{GDBN}.
23098 * Python API:: Accessing @value{GDBN} from Python.
23099 * Python Auto-loading:: Automatically loading Python code.
23100 * Python modules:: Python modules provided by @value{GDBN}.
23103 @node Python Commands
23104 @subsection Python Commands
23105 @cindex python commands
23106 @cindex commands to access python
23108 @value{GDBN} provides two commands for accessing the Python interpreter,
23109 and one related setting:
23112 @kindex python-interactive
23114 @item python-interactive @r{[}@var{command}@r{]}
23115 @itemx pi @r{[}@var{command}@r{]}
23116 Without an argument, the @code{python-interactive} command can be used
23117 to start an interactive Python prompt. To return to @value{GDBN},
23118 type the @code{EOF} character (e.g., @kbd{Ctrl-D} on an empty prompt).
23120 Alternatively, a single-line Python command can be given as an
23121 argument and evaluated. If the command is an expression, the result
23122 will be printed; otherwise, nothing will be printed. For example:
23125 (@value{GDBP}) python-interactive 2 + 3
23131 @item python @r{[}@var{command}@r{]}
23132 @itemx py @r{[}@var{command}@r{]}
23133 The @code{python} command can be used to evaluate Python code.
23135 If given an argument, the @code{python} command will evaluate the
23136 argument as a Python command. For example:
23139 (@value{GDBP}) python print 23
23143 If you do not provide an argument to @code{python}, it will act as a
23144 multi-line command, like @code{define}. In this case, the Python
23145 script is made up of subsequent command lines, given after the
23146 @code{python} command. This command list is terminated using a line
23147 containing @code{end}. For example:
23150 (@value{GDBP}) python
23152 End with a line saying just "end".
23158 @kindex set python print-stack
23159 @item set python print-stack
23160 By default, @value{GDBN} will print only the message component of a
23161 Python exception when an error occurs in a Python script. This can be
23162 controlled using @code{set python print-stack}: if @code{full}, then
23163 full Python stack printing is enabled; if @code{none}, then Python stack
23164 and message printing is disabled; if @code{message}, the default, only
23165 the message component of the error is printed.
23168 It is also possible to execute a Python script from the @value{GDBN}
23172 @item source @file{script-name}
23173 The script name must end with @samp{.py} and @value{GDBN} must be configured
23174 to recognize the script language based on filename extension using
23175 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
23177 @item python execfile ("script-name")
23178 This method is based on the @code{execfile} Python built-in function,
23179 and thus is always available.
23183 @subsection Python API
23185 @cindex programming in python
23187 You can get quick online help for @value{GDBN}'s Python API by issuing
23188 the command @w{@kbd{python help (gdb)}}.
23190 Functions and methods which have two or more optional arguments allow
23191 them to be specified using keyword syntax. This allows passing some
23192 optional arguments while skipping others. Example:
23193 @w{@code{gdb.some_function ('foo', bar = 1, baz = 2)}}.
23196 * Basic Python:: Basic Python Functions.
23197 * Exception Handling:: How Python exceptions are translated.
23198 * Values From Inferior:: Python representation of values.
23199 * Types In Python:: Python representation of types.
23200 * Pretty Printing API:: Pretty-printing values.
23201 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
23202 * Writing a Pretty-Printer:: Writing a Pretty-Printer.
23203 * Type Printing API:: Pretty-printing types.
23204 * Frame Filter API:: Filtering Frames.
23205 * Frame Decorator API:: Decorating Frames.
23206 * Writing a Frame Filter:: Writing a Frame Filter.
23207 * Inferiors In Python:: Python representation of inferiors (processes)
23208 * Events In Python:: Listening for events from @value{GDBN}.
23209 * Threads In Python:: Accessing inferior threads from Python.
23210 * Commands In Python:: Implementing new commands in Python.
23211 * Parameters In Python:: Adding new @value{GDBN} parameters.
23212 * Functions In Python:: Writing new convenience functions.
23213 * Progspaces In Python:: Program spaces.
23214 * Objfiles In Python:: Object files.
23215 * Frames In Python:: Accessing inferior stack frames from Python.
23216 * Blocks In Python:: Accessing blocks from Python.
23217 * Symbols In Python:: Python representation of symbols.
23218 * Symbol Tables In Python:: Python representation of symbol tables.
23219 * Breakpoints In Python:: Manipulating breakpoints using Python.
23220 * Finish Breakpoints in Python:: Setting Breakpoints on function return
23222 * Lazy Strings In Python:: Python representation of lazy strings.
23223 * Architectures In Python:: Python representation of architectures.
23227 @subsubsection Basic Python
23229 @cindex python stdout
23230 @cindex python pagination
23231 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
23232 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
23233 A Python program which outputs to one of these streams may have its
23234 output interrupted by the user (@pxref{Screen Size}). In this
23235 situation, a Python @code{KeyboardInterrupt} exception is thrown.
23237 Some care must be taken when writing Python code to run in
23238 @value{GDBN}. Two things worth noting in particular:
23242 @value{GDBN} install handlers for @code{SIGCHLD} and @code{SIGINT}.
23243 Python code must not override these, or even change the options using
23244 @code{sigaction}. If your program changes the handling of these
23245 signals, @value{GDBN} will most likely stop working correctly. Note
23246 that it is unfortunately common for GUI toolkits to install a
23247 @code{SIGCHLD} handler.
23250 @value{GDBN} takes care to mark its internal file descriptors as
23251 close-on-exec. However, this cannot be done in a thread-safe way on
23252 all platforms. Your Python programs should be aware of this and
23253 should both create new file descriptors with the close-on-exec flag
23254 set and arrange to close unneeded file descriptors before starting a
23258 @cindex python functions
23259 @cindex python module
23261 @value{GDBN} introduces a new Python module, named @code{gdb}. All
23262 methods and classes added by @value{GDBN} are placed in this module.
23263 @value{GDBN} automatically @code{import}s the @code{gdb} module for
23264 use in all scripts evaluated by the @code{python} command.
23266 @findex gdb.PYTHONDIR
23267 @defvar gdb.PYTHONDIR
23268 A string containing the python directory (@pxref{Python}).
23271 @findex gdb.execute
23272 @defun gdb.execute (command @r{[}, from_tty @r{[}, to_string@r{]]})
23273 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
23274 If a GDB exception happens while @var{command} runs, it is
23275 translated as described in @ref{Exception Handling,,Exception Handling}.
23277 @var{from_tty} specifies whether @value{GDBN} ought to consider this
23278 command as having originated from the user invoking it interactively.
23279 It must be a boolean value. If omitted, it defaults to @code{False}.
23281 By default, any output produced by @var{command} is sent to
23282 @value{GDBN}'s standard output. If the @var{to_string} parameter is
23283 @code{True}, then output will be collected by @code{gdb.execute} and
23284 returned as a string. The default is @code{False}, in which case the
23285 return value is @code{None}. If @var{to_string} is @code{True}, the
23286 @value{GDBN} virtual terminal will be temporarily set to unlimited width
23287 and height, and its pagination will be disabled; @pxref{Screen Size}.
23290 @findex gdb.breakpoints
23291 @defun gdb.breakpoints ()
23292 Return a sequence holding all of @value{GDBN}'s breakpoints.
23293 @xref{Breakpoints In Python}, for more information.
23296 @findex gdb.parameter
23297 @defun gdb.parameter (parameter)
23298 Return the value of a @value{GDBN} parameter. @var{parameter} is a
23299 string naming the parameter to look up; @var{parameter} may contain
23300 spaces if the parameter has a multi-part name. For example,
23301 @samp{print object} is a valid parameter name.
23303 If the named parameter does not exist, this function throws a
23304 @code{gdb.error} (@pxref{Exception Handling}). Otherwise, the
23305 parameter's value is converted to a Python value of the appropriate
23306 type, and returned.
23309 @findex gdb.history
23310 @defun gdb.history (number)
23311 Return a value from @value{GDBN}'s value history (@pxref{Value
23312 History}). @var{number} indicates which history element to return.
23313 If @var{number} is negative, then @value{GDBN} will take its absolute value
23314 and count backward from the last element (i.e., the most recent element) to
23315 find the value to return. If @var{number} is zero, then @value{GDBN} will
23316 return the most recent element. If the element specified by @var{number}
23317 doesn't exist in the value history, a @code{gdb.error} exception will be
23320 If no exception is raised, the return value is always an instance of
23321 @code{gdb.Value} (@pxref{Values From Inferior}).
23324 @findex gdb.parse_and_eval
23325 @defun gdb.parse_and_eval (expression)
23326 Parse @var{expression} as an expression in the current language,
23327 evaluate it, and return the result as a @code{gdb.Value}.
23328 @var{expression} must be a string.
23330 This function can be useful when implementing a new command
23331 (@pxref{Commands In Python}), as it provides a way to parse the
23332 command's argument as an expression. It is also useful simply to
23333 compute values, for example, it is the only way to get the value of a
23334 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
23337 @findex gdb.find_pc_line
23338 @defun gdb.find_pc_line (pc)
23339 Return the @code{gdb.Symtab_and_line} object corresponding to the
23340 @var{pc} value. @xref{Symbol Tables In Python}. If an invalid
23341 value of @var{pc} is passed as an argument, then the @code{symtab} and
23342 @code{line} attributes of the returned @code{gdb.Symtab_and_line} object
23343 will be @code{None} and 0 respectively.
23346 @findex gdb.post_event
23347 @defun gdb.post_event (event)
23348 Put @var{event}, a callable object taking no arguments, into
23349 @value{GDBN}'s internal event queue. This callable will be invoked at
23350 some later point, during @value{GDBN}'s event processing. Events
23351 posted using @code{post_event} will be run in the order in which they
23352 were posted; however, there is no way to know when they will be
23353 processed relative to other events inside @value{GDBN}.
23355 @value{GDBN} is not thread-safe. If your Python program uses multiple
23356 threads, you must be careful to only call @value{GDBN}-specific
23357 functions in the main @value{GDBN} thread. @code{post_event} ensures
23361 (@value{GDBP}) python
23365 > def __init__(self, message):
23366 > self.message = message;
23367 > def __call__(self):
23368 > gdb.write(self.message)
23370 >class MyThread1 (threading.Thread):
23372 > gdb.post_event(Writer("Hello "))
23374 >class MyThread2 (threading.Thread):
23376 > gdb.post_event(Writer("World\n"))
23378 >MyThread1().start()
23379 >MyThread2().start()
23381 (@value{GDBP}) Hello World
23386 @defun gdb.write (string @r{[}, stream{]})
23387 Print a string to @value{GDBN}'s paginated output stream. The
23388 optional @var{stream} determines the stream to print to. The default
23389 stream is @value{GDBN}'s standard output stream. Possible stream
23396 @value{GDBN}'s standard output stream.
23401 @value{GDBN}'s standard error stream.
23406 @value{GDBN}'s log stream (@pxref{Logging Output}).
23409 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
23410 call this function and will automatically direct the output to the
23415 @defun gdb.flush ()
23416 Flush the buffer of a @value{GDBN} paginated stream so that the
23417 contents are displayed immediately. @value{GDBN} will flush the
23418 contents of a stream automatically when it encounters a newline in the
23419 buffer. The optional @var{stream} determines the stream to flush. The
23420 default stream is @value{GDBN}'s standard output stream. Possible
23427 @value{GDBN}'s standard output stream.
23432 @value{GDBN}'s standard error stream.
23437 @value{GDBN}'s log stream (@pxref{Logging Output}).
23441 Flushing @code{sys.stdout} or @code{sys.stderr} will automatically
23442 call this function for the relevant stream.
23445 @findex gdb.target_charset
23446 @defun gdb.target_charset ()
23447 Return the name of the current target character set (@pxref{Character
23448 Sets}). This differs from @code{gdb.parameter('target-charset')} in
23449 that @samp{auto} is never returned.
23452 @findex gdb.target_wide_charset
23453 @defun gdb.target_wide_charset ()
23454 Return the name of the current target wide character set
23455 (@pxref{Character Sets}). This differs from
23456 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
23460 @findex gdb.solib_name
23461 @defun gdb.solib_name (address)
23462 Return the name of the shared library holding the given @var{address}
23463 as a string, or @code{None}.
23466 @findex gdb.decode_line
23467 @defun gdb.decode_line @r{[}expression@r{]}
23468 Return locations of the line specified by @var{expression}, or of the
23469 current line if no argument was given. This function returns a Python
23470 tuple containing two elements. The first element contains a string
23471 holding any unparsed section of @var{expression} (or @code{None} if
23472 the expression has been fully parsed). The second element contains
23473 either @code{None} or another tuple that contains all the locations
23474 that match the expression represented as @code{gdb.Symtab_and_line}
23475 objects (@pxref{Symbol Tables In Python}). If @var{expression} is
23476 provided, it is decoded the way that @value{GDBN}'s inbuilt
23477 @code{break} or @code{edit} commands do (@pxref{Specify Location}).
23480 @defun gdb.prompt_hook (current_prompt)
23481 @anchor{prompt_hook}
23483 If @var{prompt_hook} is callable, @value{GDBN} will call the method
23484 assigned to this operation before a prompt is displayed by
23487 The parameter @code{current_prompt} contains the current @value{GDBN}
23488 prompt. This method must return a Python string, or @code{None}. If
23489 a string is returned, the @value{GDBN} prompt will be set to that
23490 string. If @code{None} is returned, @value{GDBN} will continue to use
23491 the current prompt.
23493 Some prompts cannot be substituted in @value{GDBN}. Secondary prompts
23494 such as those used by readline for command input, and annotation
23495 related prompts are prohibited from being changed.
23498 @node Exception Handling
23499 @subsubsection Exception Handling
23500 @cindex python exceptions
23501 @cindex exceptions, python
23503 When executing the @code{python} command, Python exceptions
23504 uncaught within the Python code are translated to calls to
23505 @value{GDBN} error-reporting mechanism. If the command that called
23506 @code{python} does not handle the error, @value{GDBN} will
23507 terminate it and print an error message containing the Python
23508 exception name, the associated value, and the Python call stack
23509 backtrace at the point where the exception was raised. Example:
23512 (@value{GDBP}) python print foo
23513 Traceback (most recent call last):
23514 File "<string>", line 1, in <module>
23515 NameError: name 'foo' is not defined
23518 @value{GDBN} errors that happen in @value{GDBN} commands invoked by
23519 Python code are converted to Python exceptions. The type of the
23520 Python exception depends on the error.
23524 This is the base class for most exceptions generated by @value{GDBN}.
23525 It is derived from @code{RuntimeError}, for compatibility with earlier
23526 versions of @value{GDBN}.
23528 If an error occurring in @value{GDBN} does not fit into some more
23529 specific category, then the generated exception will have this type.
23531 @item gdb.MemoryError
23532 This is a subclass of @code{gdb.error} which is thrown when an
23533 operation tried to access invalid memory in the inferior.
23535 @item KeyboardInterrupt
23536 User interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
23537 prompt) is translated to a Python @code{KeyboardInterrupt} exception.
23540 In all cases, your exception handler will see the @value{GDBN} error
23541 message as its value and the Python call stack backtrace at the Python
23542 statement closest to where the @value{GDBN} error occured as the
23545 @findex gdb.GdbError
23546 When implementing @value{GDBN} commands in Python via @code{gdb.Command},
23547 it is useful to be able to throw an exception that doesn't cause a
23548 traceback to be printed. For example, the user may have invoked the
23549 command incorrectly. Use the @code{gdb.GdbError} exception
23550 to handle this case. Example:
23554 >class HelloWorld (gdb.Command):
23555 > """Greet the whole world."""
23556 > def __init__ (self):
23557 > super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
23558 > def invoke (self, args, from_tty):
23559 > argv = gdb.string_to_argv (args)
23560 > if len (argv) != 0:
23561 > raise gdb.GdbError ("hello-world takes no arguments")
23562 > print "Hello, World!"
23565 (gdb) hello-world 42
23566 hello-world takes no arguments
23569 @node Values From Inferior
23570 @subsubsection Values From Inferior
23571 @cindex values from inferior, with Python
23572 @cindex python, working with values from inferior
23574 @cindex @code{gdb.Value}
23575 @value{GDBN} provides values it obtains from the inferior program in
23576 an object of type @code{gdb.Value}. @value{GDBN} uses this object
23577 for its internal bookkeeping of the inferior's values, and for
23578 fetching values when necessary.
23580 Inferior values that are simple scalars can be used directly in
23581 Python expressions that are valid for the value's data type. Here's
23582 an example for an integer or floating-point value @code{some_val}:
23589 As result of this, @code{bar} will also be a @code{gdb.Value} object
23590 whose values are of the same type as those of @code{some_val}.
23592 Inferior values that are structures or instances of some class can
23593 be accessed using the Python @dfn{dictionary syntax}. For example, if
23594 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
23595 can access its @code{foo} element with:
23598 bar = some_val['foo']
23601 Again, @code{bar} will also be a @code{gdb.Value} object.
23603 A @code{gdb.Value} that represents a function can be executed via
23604 inferior function call. Any arguments provided to the call must match
23605 the function's prototype, and must be provided in the order specified
23608 For example, @code{some_val} is a @code{gdb.Value} instance
23609 representing a function that takes two integers as arguments. To
23610 execute this function, call it like so:
23613 result = some_val (10,20)
23616 Any values returned from a function call will be stored as a
23619 The following attributes are provided:
23621 @defvar Value.address
23622 If this object is addressable, this read-only attribute holds a
23623 @code{gdb.Value} object representing the address. Otherwise,
23624 this attribute holds @code{None}.
23627 @cindex optimized out value in Python
23628 @defvar Value.is_optimized_out
23629 This read-only boolean attribute is true if the compiler optimized out
23630 this value, thus it is not available for fetching from the inferior.
23634 The type of this @code{gdb.Value}. The value of this attribute is a
23635 @code{gdb.Type} object (@pxref{Types In Python}).
23638 @defvar Value.dynamic_type
23639 The dynamic type of this @code{gdb.Value}. This uses C@t{++} run-time
23640 type information (@acronym{RTTI}) to determine the dynamic type of the
23641 value. If this value is of class type, it will return the class in
23642 which the value is embedded, if any. If this value is of pointer or
23643 reference to a class type, it will compute the dynamic type of the
23644 referenced object, and return a pointer or reference to that type,
23645 respectively. In all other cases, it will return the value's static
23648 Note that this feature will only work when debugging a C@t{++} program
23649 that includes @acronym{RTTI} for the object in question. Otherwise,
23650 it will just return the static type of the value as in @kbd{ptype foo}
23651 (@pxref{Symbols, ptype}).
23654 @defvar Value.is_lazy
23655 The value of this read-only boolean attribute is @code{True} if this
23656 @code{gdb.Value} has not yet been fetched from the inferior.
23657 @value{GDBN} does not fetch values until necessary, for efficiency.
23661 myval = gdb.parse_and_eval ('somevar')
23664 The value of @code{somevar} is not fetched at this time. It will be
23665 fetched when the value is needed, or when the @code{fetch_lazy}
23669 The following methods are provided:
23671 @defun Value.__init__ (@var{val})
23672 Many Python values can be converted directly to a @code{gdb.Value} via
23673 this object initializer. Specifically:
23676 @item Python boolean
23677 A Python boolean is converted to the boolean type from the current
23680 @item Python integer
23681 A Python integer is converted to the C @code{long} type for the
23682 current architecture.
23685 A Python long is converted to the C @code{long long} type for the
23686 current architecture.
23689 A Python float is converted to the C @code{double} type for the
23690 current architecture.
23692 @item Python string
23693 A Python string is converted to a target string, using the current
23696 @item @code{gdb.Value}
23697 If @code{val} is a @code{gdb.Value}, then a copy of the value is made.
23699 @item @code{gdb.LazyString}
23700 If @code{val} is a @code{gdb.LazyString} (@pxref{Lazy Strings In
23701 Python}), then the lazy string's @code{value} method is called, and
23702 its result is used.
23706 @defun Value.cast (type)
23707 Return a new instance of @code{gdb.Value} that is the result of
23708 casting this instance to the type described by @var{type}, which must
23709 be a @code{gdb.Type} object. If the cast cannot be performed for some
23710 reason, this method throws an exception.
23713 @defun Value.dereference ()
23714 For pointer data types, this method returns a new @code{gdb.Value} object
23715 whose contents is the object pointed to by the pointer. For example, if
23716 @code{foo} is a C pointer to an @code{int}, declared in your C program as
23723 then you can use the corresponding @code{gdb.Value} to access what
23724 @code{foo} points to like this:
23727 bar = foo.dereference ()
23730 The result @code{bar} will be a @code{gdb.Value} object holding the
23731 value pointed to by @code{foo}.
23733 A similar function @code{Value.referenced_value} exists which also
23734 returns @code{gdb.Value} objects corresonding to the values pointed to
23735 by pointer values (and additionally, values referenced by reference
23736 values). However, the behavior of @code{Value.dereference}
23737 differs from @code{Value.referenced_value} by the fact that the
23738 behavior of @code{Value.dereference} is identical to applying the C
23739 unary operator @code{*} on a given value. For example, consider a
23740 reference to a pointer @code{ptrref}, declared in your C@t{++} program
23744 typedef int *intptr;
23748 intptr &ptrref = ptr;
23751 Though @code{ptrref} is a reference value, one can apply the method
23752 @code{Value.dereference} to the @code{gdb.Value} object corresponding
23753 to it and obtain a @code{gdb.Value} which is identical to that
23754 corresponding to @code{val}. However, if you apply the method
23755 @code{Value.referenced_value}, the result would be a @code{gdb.Value}
23756 object identical to that corresponding to @code{ptr}.
23759 py_ptrref = gdb.parse_and_eval ("ptrref")
23760 py_val = py_ptrref.dereference ()
23761 py_ptr = py_ptrref.referenced_value ()
23764 The @code{gdb.Value} object @code{py_val} is identical to that
23765 corresponding to @code{val}, and @code{py_ptr} is identical to that
23766 corresponding to @code{ptr}. In general, @code{Value.dereference} can
23767 be applied whenever the C unary operator @code{*} can be applied
23768 to the corresponding C value. For those cases where applying both
23769 @code{Value.dereference} and @code{Value.referenced_value} is allowed,
23770 the results obtained need not be identical (as we have seen in the above
23771 example). The results are however identical when applied on
23772 @code{gdb.Value} objects corresponding to pointers (@code{gdb.Value}
23773 objects with type code @code{TYPE_CODE_PTR}) in a C/C@t{++} program.
23776 @defun Value.referenced_value ()
23777 For pointer or reference data types, this method returns a new
23778 @code{gdb.Value} object corresponding to the value referenced by the
23779 pointer/reference value. For pointer data types,
23780 @code{Value.dereference} and @code{Value.referenced_value} produce
23781 identical results. The difference between these methods is that
23782 @code{Value.dereference} cannot get the values referenced by reference
23783 values. For example, consider a reference to an @code{int}, declared
23784 in your C@t{++} program as
23792 then applying @code{Value.dereference} to the @code{gdb.Value} object
23793 corresponding to @code{ref} will result in an error, while applying
23794 @code{Value.referenced_value} will result in a @code{gdb.Value} object
23795 identical to that corresponding to @code{val}.
23798 py_ref = gdb.parse_and_eval ("ref")
23799 er_ref = py_ref.dereference () # Results in error
23800 py_val = py_ref.referenced_value () # Returns the referenced value
23803 The @code{gdb.Value} object @code{py_val} is identical to that
23804 corresponding to @code{val}.
23807 @defun Value.dynamic_cast (type)
23808 Like @code{Value.cast}, but works as if the C@t{++} @code{dynamic_cast}
23809 operator were used. Consult a C@t{++} reference for details.
23812 @defun Value.reinterpret_cast (type)
23813 Like @code{Value.cast}, but works as if the C@t{++} @code{reinterpret_cast}
23814 operator were used. Consult a C@t{++} reference for details.
23817 @defun Value.string (@r{[}encoding@r{[}, errors@r{[}, length@r{]]]})
23818 If this @code{gdb.Value} represents a string, then this method
23819 converts the contents to a Python string. Otherwise, this method will
23820 throw an exception.
23822 Strings are recognized in a language-specific way; whether a given
23823 @code{gdb.Value} represents a string is determined by the current
23826 For C-like languages, a value is a string if it is a pointer to or an
23827 array of characters or ints. The string is assumed to be terminated
23828 by a zero of the appropriate width. However if the optional length
23829 argument is given, the string will be converted to that given length,
23830 ignoring any embedded zeros that the string may contain.
23832 If the optional @var{encoding} argument is given, it must be a string
23833 naming the encoding of the string in the @code{gdb.Value}, such as
23834 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
23835 the same encodings as the corresponding argument to Python's
23836 @code{string.decode} method, and the Python codec machinery will be used
23837 to convert the string. If @var{encoding} is not given, or if
23838 @var{encoding} is the empty string, then either the @code{target-charset}
23839 (@pxref{Character Sets}) will be used, or a language-specific encoding
23840 will be used, if the current language is able to supply one.
23842 The optional @var{errors} argument is the same as the corresponding
23843 argument to Python's @code{string.decode} method.
23845 If the optional @var{length} argument is given, the string will be
23846 fetched and converted to the given length.
23849 @defun Value.lazy_string (@r{[}encoding @r{[}, length@r{]]})
23850 If this @code{gdb.Value} represents a string, then this method
23851 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
23852 In Python}). Otherwise, this method will throw an exception.
23854 If the optional @var{encoding} argument is given, it must be a string
23855 naming the encoding of the @code{gdb.LazyString}. Some examples are:
23856 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
23857 @var{encoding} argument is an encoding that @value{GDBN} does
23858 recognize, @value{GDBN} will raise an error.
23860 When a lazy string is printed, the @value{GDBN} encoding machinery is
23861 used to convert the string during printing. If the optional
23862 @var{encoding} argument is not provided, or is an empty string,
23863 @value{GDBN} will automatically select the encoding most suitable for
23864 the string type. For further information on encoding in @value{GDBN}
23865 please see @ref{Character Sets}.
23867 If the optional @var{length} argument is given, the string will be
23868 fetched and encoded to the length of characters specified. If
23869 the @var{length} argument is not provided, the string will be fetched
23870 and encoded until a null of appropriate width is found.
23873 @defun Value.fetch_lazy ()
23874 If the @code{gdb.Value} object is currently a lazy value
23875 (@code{gdb.Value.is_lazy} is @code{True}), then the value is
23876 fetched from the inferior. Any errors that occur in the process
23877 will produce a Python exception.
23879 If the @code{gdb.Value} object is not a lazy value, this method
23882 This method does not return a value.
23886 @node Types In Python
23887 @subsubsection Types In Python
23888 @cindex types in Python
23889 @cindex Python, working with types
23892 @value{GDBN} represents types from the inferior using the class
23895 The following type-related functions are available in the @code{gdb}
23898 @findex gdb.lookup_type
23899 @defun gdb.lookup_type (name @r{[}, block@r{]})
23900 This function looks up a type by name. @var{name} is the name of the
23901 type to look up. It must be a string.
23903 If @var{block} is given, then @var{name} is looked up in that scope.
23904 Otherwise, it is searched for globally.
23906 Ordinarily, this function will return an instance of @code{gdb.Type}.
23907 If the named type cannot be found, it will throw an exception.
23910 If the type is a structure or class type, or an enum type, the fields
23911 of that type can be accessed using the Python @dfn{dictionary syntax}.
23912 For example, if @code{some_type} is a @code{gdb.Type} instance holding
23913 a structure type, you can access its @code{foo} field with:
23916 bar = some_type['foo']
23919 @code{bar} will be a @code{gdb.Field} object; see below under the
23920 description of the @code{Type.fields} method for a description of the
23921 @code{gdb.Field} class.
23923 An instance of @code{Type} has the following attributes:
23926 The type code for this type. The type code will be one of the
23927 @code{TYPE_CODE_} constants defined below.
23930 @defvar Type.sizeof
23931 The size of this type, in target @code{char} units. Usually, a
23932 target's @code{char} type will be an 8-bit byte. However, on some
23933 unusual platforms, this type may have a different size.
23937 The tag name for this type. The tag name is the name after
23938 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
23939 languages have this concept. If this type has no tag name, then
23940 @code{None} is returned.
23943 The following methods are provided:
23945 @defun Type.fields ()
23946 For structure and union types, this method returns the fields. Range
23947 types have two fields, the minimum and maximum values. Enum types
23948 have one field per enum constant. Function and method types have one
23949 field per parameter. The base types of C@t{++} classes are also
23950 represented as fields. If the type has no fields, or does not fit
23951 into one of these categories, an empty sequence will be returned.
23953 Each field is a @code{gdb.Field} object, with some pre-defined attributes:
23956 This attribute is not available for @code{static} fields (as in
23957 C@t{++} or Java). For non-@code{static} fields, the value is the bit
23958 position of the field. For @code{enum} fields, the value is the
23959 enumeration member's integer representation.
23962 The name of the field, or @code{None} for anonymous fields.
23965 This is @code{True} if the field is artificial, usually meaning that
23966 it was provided by the compiler and not the user. This attribute is
23967 always provided, and is @code{False} if the field is not artificial.
23969 @item is_base_class
23970 This is @code{True} if the field represents a base class of a C@t{++}
23971 structure. This attribute is always provided, and is @code{False}
23972 if the field is not a base class of the type that is the argument of
23973 @code{fields}, or if that type was not a C@t{++} class.
23976 If the field is packed, or is a bitfield, then this will have a
23977 non-zero value, which is the size of the field in bits. Otherwise,
23978 this will be zero; in this case the field's size is given by its type.
23981 The type of the field. This is usually an instance of @code{Type},
23982 but it can be @code{None} in some situations.
23986 @defun Type.array (@var{n1} @r{[}, @var{n2}@r{]})
23987 Return a new @code{gdb.Type} object which represents an array of this
23988 type. If one argument is given, it is the inclusive upper bound of
23989 the array; in this case the lower bound is zero. If two arguments are
23990 given, the first argument is the lower bound of the array, and the
23991 second argument is the upper bound of the array. An array's length
23992 must not be negative, but the bounds can be.
23995 @defun Type.vector (@var{n1} @r{[}, @var{n2}@r{]})
23996 Return a new @code{gdb.Type} object which represents a vector of this
23997 type. If one argument is given, it is the inclusive upper bound of
23998 the vector; in this case the lower bound is zero. If two arguments are
23999 given, the first argument is the lower bound of the vector, and the
24000 second argument is the upper bound of the vector. A vector's length
24001 must not be negative, but the bounds can be.
24003 The difference between an @code{array} and a @code{vector} is that
24004 arrays behave like in C: when used in expressions they decay to a pointer
24005 to the first element whereas vectors are treated as first class values.
24008 @defun Type.const ()
24009 Return a new @code{gdb.Type} object which represents a
24010 @code{const}-qualified variant of this type.
24013 @defun Type.volatile ()
24014 Return a new @code{gdb.Type} object which represents a
24015 @code{volatile}-qualified variant of this type.
24018 @defun Type.unqualified ()
24019 Return a new @code{gdb.Type} object which represents an unqualified
24020 variant of this type. That is, the result is neither @code{const} nor
24024 @defun Type.range ()
24025 Return a Python @code{Tuple} object that contains two elements: the
24026 low bound of the argument type and the high bound of that type. If
24027 the type does not have a range, @value{GDBN} will raise a
24028 @code{gdb.error} exception (@pxref{Exception Handling}).
24031 @defun Type.reference ()
24032 Return a new @code{gdb.Type} object which represents a reference to this
24036 @defun Type.pointer ()
24037 Return a new @code{gdb.Type} object which represents a pointer to this
24041 @defun Type.strip_typedefs ()
24042 Return a new @code{gdb.Type} that represents the real type,
24043 after removing all layers of typedefs.
24046 @defun Type.target ()
24047 Return a new @code{gdb.Type} object which represents the target type
24050 For a pointer type, the target type is the type of the pointed-to
24051 object. For an array type (meaning C-like arrays), the target type is
24052 the type of the elements of the array. For a function or method type,
24053 the target type is the type of the return value. For a complex type,
24054 the target type is the type of the elements. For a typedef, the
24055 target type is the aliased type.
24057 If the type does not have a target, this method will throw an
24061 @defun Type.template_argument (n @r{[}, block@r{]})
24062 If this @code{gdb.Type} is an instantiation of a template, this will
24063 return a new @code{gdb.Type} which represents the type of the
24064 @var{n}th template argument.
24066 If this @code{gdb.Type} is not a template type, this will throw an
24067 exception. Ordinarily, only C@t{++} code will have template types.
24069 If @var{block} is given, then @var{name} is looked up in that scope.
24070 Otherwise, it is searched for globally.
24074 Each type has a code, which indicates what category this type falls
24075 into. The available type categories are represented by constants
24076 defined in the @code{gdb} module:
24079 @findex TYPE_CODE_PTR
24080 @findex gdb.TYPE_CODE_PTR
24081 @item gdb.TYPE_CODE_PTR
24082 The type is a pointer.
24084 @findex TYPE_CODE_ARRAY
24085 @findex gdb.TYPE_CODE_ARRAY
24086 @item gdb.TYPE_CODE_ARRAY
24087 The type is an array.
24089 @findex TYPE_CODE_STRUCT
24090 @findex gdb.TYPE_CODE_STRUCT
24091 @item gdb.TYPE_CODE_STRUCT
24092 The type is a structure.
24094 @findex TYPE_CODE_UNION
24095 @findex gdb.TYPE_CODE_UNION
24096 @item gdb.TYPE_CODE_UNION
24097 The type is a union.
24099 @findex TYPE_CODE_ENUM
24100 @findex gdb.TYPE_CODE_ENUM
24101 @item gdb.TYPE_CODE_ENUM
24102 The type is an enum.
24104 @findex TYPE_CODE_FLAGS
24105 @findex gdb.TYPE_CODE_FLAGS
24106 @item gdb.TYPE_CODE_FLAGS
24107 A bit flags type, used for things such as status registers.
24109 @findex TYPE_CODE_FUNC
24110 @findex gdb.TYPE_CODE_FUNC
24111 @item gdb.TYPE_CODE_FUNC
24112 The type is a function.
24114 @findex TYPE_CODE_INT
24115 @findex gdb.TYPE_CODE_INT
24116 @item gdb.TYPE_CODE_INT
24117 The type is an integer type.
24119 @findex TYPE_CODE_FLT
24120 @findex gdb.TYPE_CODE_FLT
24121 @item gdb.TYPE_CODE_FLT
24122 A floating point type.
24124 @findex TYPE_CODE_VOID
24125 @findex gdb.TYPE_CODE_VOID
24126 @item gdb.TYPE_CODE_VOID
24127 The special type @code{void}.
24129 @findex TYPE_CODE_SET
24130 @findex gdb.TYPE_CODE_SET
24131 @item gdb.TYPE_CODE_SET
24134 @findex TYPE_CODE_RANGE
24135 @findex gdb.TYPE_CODE_RANGE
24136 @item gdb.TYPE_CODE_RANGE
24137 A range type, that is, an integer type with bounds.
24139 @findex TYPE_CODE_STRING
24140 @findex gdb.TYPE_CODE_STRING
24141 @item gdb.TYPE_CODE_STRING
24142 A string type. Note that this is only used for certain languages with
24143 language-defined string types; C strings are not represented this way.
24145 @findex TYPE_CODE_BITSTRING
24146 @findex gdb.TYPE_CODE_BITSTRING
24147 @item gdb.TYPE_CODE_BITSTRING
24148 A string of bits. It is deprecated.
24150 @findex TYPE_CODE_ERROR
24151 @findex gdb.TYPE_CODE_ERROR
24152 @item gdb.TYPE_CODE_ERROR
24153 An unknown or erroneous type.
24155 @findex TYPE_CODE_METHOD
24156 @findex gdb.TYPE_CODE_METHOD
24157 @item gdb.TYPE_CODE_METHOD
24158 A method type, as found in C@t{++} or Java.
24160 @findex TYPE_CODE_METHODPTR
24161 @findex gdb.TYPE_CODE_METHODPTR
24162 @item gdb.TYPE_CODE_METHODPTR
24163 A pointer-to-member-function.
24165 @findex TYPE_CODE_MEMBERPTR
24166 @findex gdb.TYPE_CODE_MEMBERPTR
24167 @item gdb.TYPE_CODE_MEMBERPTR
24168 A pointer-to-member.
24170 @findex TYPE_CODE_REF
24171 @findex gdb.TYPE_CODE_REF
24172 @item gdb.TYPE_CODE_REF
24175 @findex TYPE_CODE_CHAR
24176 @findex gdb.TYPE_CODE_CHAR
24177 @item gdb.TYPE_CODE_CHAR
24180 @findex TYPE_CODE_BOOL
24181 @findex gdb.TYPE_CODE_BOOL
24182 @item gdb.TYPE_CODE_BOOL
24185 @findex TYPE_CODE_COMPLEX
24186 @findex gdb.TYPE_CODE_COMPLEX
24187 @item gdb.TYPE_CODE_COMPLEX
24188 A complex float type.
24190 @findex TYPE_CODE_TYPEDEF
24191 @findex gdb.TYPE_CODE_TYPEDEF
24192 @item gdb.TYPE_CODE_TYPEDEF
24193 A typedef to some other type.
24195 @findex TYPE_CODE_NAMESPACE
24196 @findex gdb.TYPE_CODE_NAMESPACE
24197 @item gdb.TYPE_CODE_NAMESPACE
24198 A C@t{++} namespace.
24200 @findex TYPE_CODE_DECFLOAT
24201 @findex gdb.TYPE_CODE_DECFLOAT
24202 @item gdb.TYPE_CODE_DECFLOAT
24203 A decimal floating point type.
24205 @findex TYPE_CODE_INTERNAL_FUNCTION
24206 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
24207 @item gdb.TYPE_CODE_INTERNAL_FUNCTION
24208 A function internal to @value{GDBN}. This is the type used to represent
24209 convenience functions.
24212 Further support for types is provided in the @code{gdb.types}
24213 Python module (@pxref{gdb.types}).
24215 @node Pretty Printing API
24216 @subsubsection Pretty Printing API
24218 An example output is provided (@pxref{Pretty Printing}).
24220 A pretty-printer is just an object that holds a value and implements a
24221 specific interface, defined here.
24223 @defun pretty_printer.children (self)
24224 @value{GDBN} will call this method on a pretty-printer to compute the
24225 children of the pretty-printer's value.
24227 This method must return an object conforming to the Python iterator
24228 protocol. Each item returned by the iterator must be a tuple holding
24229 two elements. The first element is the ``name'' of the child; the
24230 second element is the child's value. The value can be any Python
24231 object which is convertible to a @value{GDBN} value.
24233 This method is optional. If it does not exist, @value{GDBN} will act
24234 as though the value has no children.
24237 @defun pretty_printer.display_hint (self)
24238 The CLI may call this method and use its result to change the
24239 formatting of a value. The result will also be supplied to an MI
24240 consumer as a @samp{displayhint} attribute of the variable being
24243 This method is optional. If it does exist, this method must return a
24246 Some display hints are predefined by @value{GDBN}:
24250 Indicate that the object being printed is ``array-like''. The CLI
24251 uses this to respect parameters such as @code{set print elements} and
24252 @code{set print array}.
24255 Indicate that the object being printed is ``map-like'', and that the
24256 children of this value can be assumed to alternate between keys and
24260 Indicate that the object being printed is ``string-like''. If the
24261 printer's @code{to_string} method returns a Python string of some
24262 kind, then @value{GDBN} will call its internal language-specific
24263 string-printing function to format the string. For the CLI this means
24264 adding quotation marks, possibly escaping some characters, respecting
24265 @code{set print elements}, and the like.
24269 @defun pretty_printer.to_string (self)
24270 @value{GDBN} will call this method to display the string
24271 representation of the value passed to the object's constructor.
24273 When printing from the CLI, if the @code{to_string} method exists,
24274 then @value{GDBN} will prepend its result to the values returned by
24275 @code{children}. Exactly how this formatting is done is dependent on
24276 the display hint, and may change as more hints are added. Also,
24277 depending on the print settings (@pxref{Print Settings}), the CLI may
24278 print just the result of @code{to_string} in a stack trace, omitting
24279 the result of @code{children}.
24281 If this method returns a string, it is printed verbatim.
24283 Otherwise, if this method returns an instance of @code{gdb.Value},
24284 then @value{GDBN} prints this value. This may result in a call to
24285 another pretty-printer.
24287 If instead the method returns a Python value which is convertible to a
24288 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
24289 the resulting value. Again, this may result in a call to another
24290 pretty-printer. Python scalars (integers, floats, and booleans) and
24291 strings are convertible to @code{gdb.Value}; other types are not.
24293 Finally, if this method returns @code{None} then no further operations
24294 are peformed in this method and nothing is printed.
24296 If the result is not one of these types, an exception is raised.
24299 @value{GDBN} provides a function which can be used to look up the
24300 default pretty-printer for a @code{gdb.Value}:
24302 @findex gdb.default_visualizer
24303 @defun gdb.default_visualizer (value)
24304 This function takes a @code{gdb.Value} object as an argument. If a
24305 pretty-printer for this value exists, then it is returned. If no such
24306 printer exists, then this returns @code{None}.
24309 @node Selecting Pretty-Printers
24310 @subsubsection Selecting Pretty-Printers
24312 The Python list @code{gdb.pretty_printers} contains an array of
24313 functions or callable objects that have been registered via addition
24314 as a pretty-printer. Printers in this list are called @code{global}
24315 printers, they're available when debugging all inferiors.
24316 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
24317 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
24320 Each function on these lists is passed a single @code{gdb.Value}
24321 argument and should return a pretty-printer object conforming to the
24322 interface definition above (@pxref{Pretty Printing API}). If a function
24323 cannot create a pretty-printer for the value, it should return
24326 @value{GDBN} first checks the @code{pretty_printers} attribute of each
24327 @code{gdb.Objfile} in the current program space and iteratively calls
24328 each enabled lookup routine in the list for that @code{gdb.Objfile}
24329 until it receives a pretty-printer object.
24330 If no pretty-printer is found in the objfile lists, @value{GDBN} then
24331 searches the pretty-printer list of the current program space,
24332 calling each enabled function until an object is returned.
24333 After these lists have been exhausted, it tries the global
24334 @code{gdb.pretty_printers} list, again calling each enabled function until an
24335 object is returned.
24337 The order in which the objfiles are searched is not specified. For a
24338 given list, functions are always invoked from the head of the list,
24339 and iterated over sequentially until the end of the list, or a printer
24340 object is returned.
24342 For various reasons a pretty-printer may not work.
24343 For example, the underlying data structure may have changed and
24344 the pretty-printer is out of date.
24346 The consequences of a broken pretty-printer are severe enough that
24347 @value{GDBN} provides support for enabling and disabling individual
24348 printers. For example, if @code{print frame-arguments} is on,
24349 a backtrace can become highly illegible if any argument is printed
24350 with a broken printer.
24352 Pretty-printers are enabled and disabled by attaching an @code{enabled}
24353 attribute to the registered function or callable object. If this attribute
24354 is present and its value is @code{False}, the printer is disabled, otherwise
24355 the printer is enabled.
24357 @node Writing a Pretty-Printer
24358 @subsubsection Writing a Pretty-Printer
24359 @cindex writing a pretty-printer
24361 A pretty-printer consists of two parts: a lookup function to detect
24362 if the type is supported, and the printer itself.
24364 Here is an example showing how a @code{std::string} printer might be
24365 written. @xref{Pretty Printing API}, for details on the API this class
24369 class StdStringPrinter(object):
24370 "Print a std::string"
24372 def __init__(self, val):
24375 def to_string(self):
24376 return self.val['_M_dataplus']['_M_p']
24378 def display_hint(self):
24382 And here is an example showing how a lookup function for the printer
24383 example above might be written.
24386 def str_lookup_function(val):
24387 lookup_tag = val.type.tag
24388 if lookup_tag == None:
24390 regex = re.compile("^std::basic_string<char,.*>$")
24391 if regex.match(lookup_tag):
24392 return StdStringPrinter(val)
24396 The example lookup function extracts the value's type, and attempts to
24397 match it to a type that it can pretty-print. If it is a type the
24398 printer can pretty-print, it will return a printer object. If not, it
24399 returns @code{None}.
24401 We recommend that you put your core pretty-printers into a Python
24402 package. If your pretty-printers are for use with a library, we
24403 further recommend embedding a version number into the package name.
24404 This practice will enable @value{GDBN} to load multiple versions of
24405 your pretty-printers at the same time, because they will have
24408 You should write auto-loaded code (@pxref{Python Auto-loading}) such that it
24409 can be evaluated multiple times without changing its meaning. An
24410 ideal auto-load file will consist solely of @code{import}s of your
24411 printer modules, followed by a call to a register pretty-printers with
24412 the current objfile.
24414 Taken as a whole, this approach will scale nicely to multiple
24415 inferiors, each potentially using a different library version.
24416 Embedding a version number in the Python package name will ensure that
24417 @value{GDBN} is able to load both sets of printers simultaneously.
24418 Then, because the search for pretty-printers is done by objfile, and
24419 because your auto-loaded code took care to register your library's
24420 printers with a specific objfile, @value{GDBN} will find the correct
24421 printers for the specific version of the library used by each
24424 To continue the @code{std::string} example (@pxref{Pretty Printing API}),
24425 this code might appear in @code{gdb.libstdcxx.v6}:
24428 def register_printers(objfile):
24429 objfile.pretty_printers.append(str_lookup_function)
24433 And then the corresponding contents of the auto-load file would be:
24436 import gdb.libstdcxx.v6
24437 gdb.libstdcxx.v6.register_printers(gdb.current_objfile())
24440 The previous example illustrates a basic pretty-printer.
24441 There are a few things that can be improved on.
24442 The printer doesn't have a name, making it hard to identify in a
24443 list of installed printers. The lookup function has a name, but
24444 lookup functions can have arbitrary, even identical, names.
24446 Second, the printer only handles one type, whereas a library typically has
24447 several types. One could install a lookup function for each desired type
24448 in the library, but one could also have a single lookup function recognize
24449 several types. The latter is the conventional way this is handled.
24450 If a pretty-printer can handle multiple data types, then its
24451 @dfn{subprinters} are the printers for the individual data types.
24453 The @code{gdb.printing} module provides a formal way of solving these
24454 problems (@pxref{gdb.printing}).
24455 Here is another example that handles multiple types.
24457 These are the types we are going to pretty-print:
24460 struct foo @{ int a, b; @};
24461 struct bar @{ struct foo x, y; @};
24464 Here are the printers:
24468 """Print a foo object."""
24470 def __init__(self, val):
24473 def to_string(self):
24474 return ("a=<" + str(self.val["a"]) +
24475 "> b=<" + str(self.val["b"]) + ">")
24478 """Print a bar object."""
24480 def __init__(self, val):
24483 def to_string(self):
24484 return ("x=<" + str(self.val["x"]) +
24485 "> y=<" + str(self.val["y"]) + ">")
24488 This example doesn't need a lookup function, that is handled by the
24489 @code{gdb.printing} module. Instead a function is provided to build up
24490 the object that handles the lookup.
24493 import gdb.printing
24495 def build_pretty_printer():
24496 pp = gdb.printing.RegexpCollectionPrettyPrinter(
24498 pp.add_printer('foo', '^foo$', fooPrinter)
24499 pp.add_printer('bar', '^bar$', barPrinter)
24503 And here is the autoload support:
24506 import gdb.printing
24508 gdb.printing.register_pretty_printer(
24509 gdb.current_objfile(),
24510 my_library.build_pretty_printer())
24513 Finally, when this printer is loaded into @value{GDBN}, here is the
24514 corresponding output of @samp{info pretty-printer}:
24517 (gdb) info pretty-printer
24524 @node Type Printing API
24525 @subsubsection Type Printing API
24526 @cindex type printing API for Python
24528 @value{GDBN} provides a way for Python code to customize type display.
24529 This is mainly useful for substituting canonical typedef names for
24532 @cindex type printer
24533 A @dfn{type printer} is just a Python object conforming to a certain
24534 protocol. A simple base class implementing the protocol is provided;
24535 see @ref{gdb.types}. A type printer must supply at least:
24537 @defivar type_printer enabled
24538 A boolean which is True if the printer is enabled, and False
24539 otherwise. This is manipulated by the @code{enable type-printer}
24540 and @code{disable type-printer} commands.
24543 @defivar type_printer name
24544 The name of the type printer. This must be a string. This is used by
24545 the @code{enable type-printer} and @code{disable type-printer}
24549 @defmethod type_printer instantiate (self)
24550 This is called by @value{GDBN} at the start of type-printing. It is
24551 only called if the type printer is enabled. This method must return a
24552 new object that supplies a @code{recognize} method, as described below.
24556 When displaying a type, say via the @code{ptype} command, @value{GDBN}
24557 will compute a list of type recognizers. This is done by iterating
24558 first over the per-objfile type printers (@pxref{Objfiles In Python}),
24559 followed by the per-progspace type printers (@pxref{Progspaces In
24560 Python}), and finally the global type printers.
24562 @value{GDBN} will call the @code{instantiate} method of each enabled
24563 type printer. If this method returns @code{None}, then the result is
24564 ignored; otherwise, it is appended to the list of recognizers.
24566 Then, when @value{GDBN} is going to display a type name, it iterates
24567 over the list of recognizers. For each one, it calls the recognition
24568 function, stopping if the function returns a non-@code{None} value.
24569 The recognition function is defined as:
24571 @defmethod type_recognizer recognize (self, type)
24572 If @var{type} is not recognized, return @code{None}. Otherwise,
24573 return a string which is to be printed as the name of @var{type}.
24574 @var{type} will be an instance of @code{gdb.Type} (@pxref{Types In
24578 @value{GDBN} uses this two-pass approach so that type printers can
24579 efficiently cache information without holding on to it too long. For
24580 example, it can be convenient to look up type information in a type
24581 printer and hold it for a recognizer's lifetime; if a single pass were
24582 done then type printers would have to make use of the event system in
24583 order to avoid holding information that could become stale as the
24586 @node Frame Filter API
24587 @subsubsection Filtering Frames.
24588 @cindex frame filters api
24590 Frame filters are Python objects that manipulate the visibility of a
24591 frame or frames when a backtrace (@pxref{Backtrace}) is printed by
24594 Only commands that print a backtrace, or, in the case of @sc{gdb/mi}
24595 commands (@pxref{GDB/MI}), those that return a collection of frames
24596 are affected. The commands that work with frame filters are:
24598 @code{backtrace} (@pxref{backtrace-command,, The backtrace command}),
24599 @code{-stack-list-frames}
24600 (@pxref{-stack-list-frames,, The -stack-list-frames command}),
24601 @code{-stack-list-variables} (@pxref{-stack-list-variables,, The
24602 -stack-list-variables command}), @code{-stack-list-arguments}
24603 @pxref{-stack-list-arguments,, The -stack-list-arguments command}) and
24604 @code{-stack-list-locals} (@pxref{-stack-list-locals,, The
24605 -stack-list-locals command}).
24607 A frame filter works by taking an iterator as an argument, applying
24608 actions to the contents of that iterator, and returning another
24609 iterator (or, possibly, the same iterator it was provided in the case
24610 where the filter does not perform any operations). Typically, frame
24611 filters utilize tools such as the Python's @code{itertools} module to
24612 work with and create new iterators from the source iterator.
24613 Regardless of how a filter chooses to apply actions, it must not alter
24614 the underlying @value{GDBN} frame or frames, or attempt to alter the
24615 call-stack within @value{GDBN}. This preserves data integrity within
24616 @value{GDBN}. Frame filters are executed on a priority basis and care
24617 should be taken that some frame filters may have been executed before,
24618 and that some frame filters will be executed after.
24620 An important consideration when designing frame filters, and well
24621 worth reflecting upon, is that frame filters should avoid unwinding
24622 the call stack if possible. Some stacks can run very deep, into the
24623 tens of thousands in some cases. To search every frame when a frame
24624 filter executes may be too expensive at that step. The frame filter
24625 cannot know how many frames it has to iterate over, and it may have to
24626 iterate through them all. This ends up duplicating effort as
24627 @value{GDBN} performs this iteration when it prints the frames. If
24628 the filter can defer unwinding frames until frame decorators are
24629 executed, after the last filter has executed, it should. @xref{Frame
24630 Decorator API}, for more information on decorators. Also, there are
24631 examples for both frame decorators and filters in later chapters.
24632 @xref{Writing a Frame Filter}, for more information.
24634 The Python dictionary @code{gdb.frame_filters} contains key/object
24635 pairings that comprise a frame filter. Frame filters in this
24636 dictionary are called @code{global} frame filters, and they are
24637 available when debugging all inferiors. These frame filters must
24638 register with the dictionary directly. In addition to the
24639 @code{global} dictionary, there are other dictionaries that are loaded
24640 with different inferiors via auto-loading (@pxref{Python
24641 Auto-loading}). The two other areas where frame filter dictionaries
24642 can be found are: @code{gdb.Progspace} which contains a
24643 @code{frame_filters} dictionary attribute, and each @code{gdb.Objfile}
24644 object which also contains a @code{frame_filters} dictionary
24647 When a command is executed from @value{GDBN} that is compatible with
24648 frame filters, @value{GDBN} combines the @code{global},
24649 @code{gdb.Progspace} and all @code{gdb.Objfile} dictionaries currently
24650 loaded. All of the @code{gdb.Objfile} dictionaries are combined, as
24651 several frames, and thus several object files, might be in use.
24652 @value{GDBN} then prunes any frame filter whose @code{enabled}
24653 attribute is @code{False}. This pruned list is then sorted according
24654 to the @code{priority} attribute in each filter.
24656 Once the dictionaries are combined, pruned and sorted, @value{GDBN}
24657 creates an iterator which wraps each frame in the call stack in a
24658 @code{FrameDecorator} object, and calls each filter in order. The
24659 output from the previous filter will always be the input to the next
24662 Frame filters have a mandatory interface which each frame filter must
24663 implement, defined here:
24665 @defun FrameFilter.filter (iterator)
24666 @value{GDBN} will call this method on a frame filter when it has
24667 reached the order in the priority list for that filter.
24669 For example, if there are four frame filters:
24680 The order that the frame filters will be called is:
24683 Filter3 -> Filter2 -> Filter1 -> Filter4
24686 Note that the output from @code{Filter3} is passed to the input of
24687 @code{Filter2}, and so on.
24689 This @code{filter} method is passed a Python iterator. This iterator
24690 contains a sequence of frame decorators that wrap each
24691 @code{gdb.Frame}, or a frame decorator that wraps another frame
24692 decorator. The first filter that is executed in the sequence of frame
24693 filters will receive an iterator entirely comprised of default
24694 @code{FrameDecorator} objects. However, after each frame filter is
24695 executed, the previous frame filter may have wrapped some or all of
24696 the frame decorators with their own frame decorator. As frame
24697 decorators must also conform to a mandatory interface, these
24698 decorators can be assumed to act in a uniform manner (@pxref{Frame
24701 This method must return an object conforming to the Python iterator
24702 protocol. Each item in the iterator must be an object conforming to
24703 the frame decorator interface. If a frame filter does not wish to
24704 perform any operations on this iterator, it should return that
24705 iterator untouched.
24707 This method is not optional. If it does not exist, @value{GDBN} will
24708 raise and print an error.
24711 @defvar FrameFilter.name
24712 The @code{name} attribute must be Python string which contains the
24713 name of the filter displayed by @value{GDBN} (@pxref{Frame Filter
24714 Management}). This attribute may contain any combination of letters
24715 or numbers. Care should be taken to ensure that it is unique. This
24716 attribute is mandatory.
24719 @defvar FrameFilter.enabled
24720 The @code{enabled} attribute must be Python boolean. This attribute
24721 indicates to @value{GDBN} whether the frame filter is enabled, and
24722 should be considered when frame filters are executed. If
24723 @code{enabled} is @code{True}, then the frame filter will be executed
24724 when any of the backtrace commands detailed earlier in this chapter
24725 are executed. If @code{enabled} is @code{False}, then the frame
24726 filter will not be executed. This attribute is mandatory.
24729 @defvar FrameFilter.priority
24730 The @code{priority} attribute must be Python integer. This attribute
24731 controls the order of execution in relation to other frame filters.
24732 There are no imposed limits on the range of @code{priority} other than
24733 it must be a valid integer. The higher the @code{priority} attribute,
24734 the sooner the frame filter will be executed in relation to other
24735 frame filters. Although @code{priority} can be negative, it is
24736 recommended practice to assume zero is the lowest priority that a
24737 frame filter can be assigned. Frame filters that have the same
24738 priority are executed in unsorted order in that priority slot. This
24739 attribute is mandatory.
24742 @node Frame Decorator API
24743 @subsubsection Decorating Frames.
24744 @cindex frame decorator api
24746 Frame decorators are sister objects to frame filters (@pxref{Frame
24747 Filter API}). Frame decorators are applied by a frame filter and can
24748 only be used in conjunction with frame filters.
24750 The purpose of a frame decorator is to customize the printed content
24751 of each @code{gdb.Frame} in commands where frame filters are executed.
24752 This concept is called decorating a frame. Frame decorators decorate
24753 a @code{gdb.Frame} with Python code contained within each API call.
24754 This separates the actual data contained in a @code{gdb.Frame} from
24755 the decorated data produced by a frame decorator. This abstraction is
24756 necessary to maintain integrity of the data contained in each
24759 Frame decorators have a mandatory interface, defined below.
24761 @value{GDBN} already contains a frame decorator called
24762 @code{FrameDecorator}. This contains substantial amounts of
24763 boilerplate code to decorate the content of a @code{gdb.Frame}. It is
24764 recommended that other frame decorators inherit and extend this
24765 object, and only to override the methods needed.
24767 @defun FrameDecorator.elided (self)
24769 The @code{elided} method groups frames together in a hierarchical
24770 system. An example would be an interpreter, where multiple low-level
24771 frames make up a single call in the interpreted language. In this
24772 example, the frame filter would elide the low-level frames and present
24773 a single high-level frame, representing the call in the interpreted
24774 language, to the user.
24776 The @code{elided} function must return an iterable and this iterable
24777 must contain the frames that are being elided wrapped in a suitable
24778 frame decorator. If no frames are being elided this function may
24779 return an empty iterable, or @code{None}. Elided frames are indented
24780 from normal frames in a @code{CLI} backtrace, or in the case of
24781 @code{GDB/MI}, are placed in the @code{children} field of the eliding
24784 It is the frame filter's task to also filter out the elided frames from
24785 the source iterator. This will avoid printing the frame twice.
24788 @defun FrameDecorator.function (self)
24790 This method returns the name of the function in the frame that is to
24793 This method must return a Python string describing the function, or
24796 If this function returns @code{None}, @value{GDBN} will not print any
24797 data for this field.
24800 @defun FrameDecorator.address (self)
24802 This method returns the address of the frame that is to be printed.
24804 This method must return a Python numeric integer type of sufficient
24805 size to describe the address of the frame, or @code{None}.
24807 If this function returns a @code{None}, @value{GDBN} will not print
24808 any data for this field.
24811 @defun FrameDecorator.filename (self)
24813 This method returns the filename and path associated with this frame.
24815 This method must return a Python string containing the filename and
24816 the path to the object file backing the frame, or @code{None}.
24818 If this function returns a @code{None}, @value{GDBN} will not print
24819 any data for this field.
24822 @defun FrameDecorator.line (self):
24824 This method returns the line number associated with the current
24825 position within the function addressed by this frame.
24827 This method must return a Python integer type, or @code{None}.
24829 If this function returns a @code{None}, @value{GDBN} will not print
24830 any data for this field.
24833 @defun FrameDecorator.frame_args (self)
24834 @anchor{frame_args}
24836 This method must return an iterable, or @code{None}. Returning an
24837 empty iterable, or @code{None} means frame arguments will not be
24838 printed for this frame. This iterable must contain objects that
24839 implement two methods, described here.
24841 This object must implement a @code{argument} method which takes a
24842 single @code{self} parameter and must return a @code{gdb.Symbol}
24843 (@pxref{Symbols In Python}), or a Python string. The object must also
24844 implement a @code{value} method which takes a single @code{self}
24845 parameter and must return a @code{gdb.Value} (@pxref{Values From
24846 Inferior}), a Python value, or @code{None}. If the @code{value}
24847 method returns @code{None}, and the @code{argument} method returns a
24848 @code{gdb.Symbol}, @value{GDBN} will look-up and print the value of
24849 the @code{gdb.Symbol} automatically.
24854 class SymValueWrapper():
24856 def __init__(self, symbol, value):
24866 class SomeFrameDecorator()
24869 def frame_args(self):
24872 block = self.inferior_frame.block()
24876 # Iterate over all symbols in a block. Only add
24877 # symbols that are arguments.
24879 if not sym.is_argument:
24881 args.append(SymValueWrapper(sym,None))
24883 # Add example synthetic argument.
24884 args.append(SymValueWrapper(``foo'', 42))
24890 @defun FrameDecorator.frame_locals (self)
24892 This method must return an iterable or @code{None}. Returning an
24893 empty iterable, or @code{None} means frame local arguments will not be
24894 printed for this frame.
24896 The object interface, the description of the various strategies for
24897 reading frame locals, and the example are largely similar to those
24898 described in the @code{frame_args} function, (@pxref{frame_args,,The
24899 frame filter frame_args function}). Below is a modified example:
24902 class SomeFrameDecorator()
24905 def frame_locals(self):
24908 block = self.inferior_frame.block()
24912 # Iterate over all symbols in a block. Add all
24913 # symbols, except arguments.
24915 if sym.is_argument:
24917 vars.append(SymValueWrapper(sym,None))
24919 # Add an example of a synthetic local variable.
24920 vars.append(SymValueWrapper(``bar'', 99))
24926 @defun FrameDecorator.inferior_frame (self):
24928 This method must return the underlying @code{gdb.Frame} that this
24929 frame decorator is decorating. @value{GDBN} requires the underlying
24930 frame for internal frame information to determine how to print certain
24931 values when printing a frame.
24934 @node Writing a Frame Filter
24935 @subsubsection Writing a Frame Filter
24936 @cindex writing a frame filter
24938 There are three basic elements that a frame filter must implement: it
24939 must correctly implement the documented interface (@pxref{Frame Filter
24940 API}), it must register itself with @value{GDBN}, and finally, it must
24941 decide if it is to work on the data provided by @value{GDBN}. In all
24942 cases, whether it works on the iterator or not, each frame filter must
24943 return an iterator. A bare-bones frame filter follows the pattern in
24944 the following example.
24949 class FrameFilter():
24951 def __init__(self):
24952 # Frame filter attribute creation.
24954 # 'name' is the name of the filter that GDB will display.
24956 # 'priority' is the priority of the filter relative to other
24959 # 'enabled' is a boolean that indicates whether this filter is
24960 # enabled and should be executed.
24963 self.priority = 100
24964 self.enabled = True
24966 # Register this frame filter with the global frame_filters
24968 gdb.frame_filters[self.name] = self
24970 def filter(self, frame_iter):
24971 # Just return the iterator.
24975 The frame filter in the example above implements the three
24976 requirements for all frame filters. It implements the API, self
24977 registers, and makes a decision on the iterator (in this case, it just
24978 returns the iterator untouched).
24980 The first step is attribute creation and assignment, and as shown in
24981 the comments the filter assigns the following attributes: @code{name},
24982 @code{priority} and whether the filter should be enabled with the
24983 @code{enabled} attribute.
24985 The second step is registering the frame filter with the dictionary or
24986 dictionaries that the frame filter has interest in. As shown in the
24987 comments, this filter just registers itself with the global dictionary
24988 @code{gdb.frame_filters}. As noted earlier, @code{gdb.frame_filters}
24989 is a dictionary that is initialized in the @code{gdb} module when
24990 @value{GDBN} starts. What dictionary a filter registers with is an
24991 important consideration. Generally, if a filter is specific to a set
24992 of code, it should be registered either in the @code{objfile} or
24993 @code{progspace} dictionaries as they are specific to the program
24994 currently loaded in @value{GDBN}. The global dictionary is always
24995 present in @value{GDBN} and is never unloaded. Any filters registered
24996 with the global dictionary will exist until @value{GDBN} exits. To
24997 avoid filters that may conflict, it is generally better to register
24998 frame filters against the dictionaries that more closely align with
24999 the usage of the filter currently in question. @xref{Python
25000 Auto-loading}, for further information on auto-loading Python scripts.
25002 @value{GDBN} takes a hands-off approach to frame filter registration,
25003 therefore it is the frame filter's responsibility to ensure
25004 registration has occurred, and that any exceptions are handled
25005 appropriately. In particular, you may wish to handle exceptions
25006 relating to Python dictionary key uniqueness. It is mandatory that
25007 the dictionary key is the same as frame filter's @code{name}
25008 attribute. When a user manages frame filters (@pxref{Frame Filter
25009 Management}), the names @value{GDBN} will display are those contained
25010 in the @code{name} attribute.
25012 The final step of this example is the implementation of the
25013 @code{filter} method. As shown in the example comments, we define the
25014 @code{filter} method and note that the method must take an iterator,
25015 and also must return an iterator. In this bare-bones example, the
25016 frame filter is not very useful as it just returns the iterator
25017 untouched. However this is a valid operation for frame filters that
25018 have the @code{enabled} attribute set, but decide not to operate on
25021 In the next example, the frame filter operates on all frames and
25022 utilizes a frame decorator to perform some work on the frames.
25023 @xref{Frame Decorator API}, for further information on the frame
25024 decorator interface.
25026 This example works on inlined frames. It highlights frames which are
25027 inlined by tagging them with an ``[inlined]'' tag. By applying a
25028 frame decorator to all frames with the Python @code{itertools imap}
25029 method, the example defers actions to the frame decorator. Frame
25030 decorators are only processed when @value{GDBN} prints the backtrace.
25032 This introduces a new decision making topic: whether to perform
25033 decision making operations at the filtering step, or at the printing
25034 step. In this example's approach, it does not perform any filtering
25035 decisions at the filtering step beyond mapping a frame decorator to
25036 each frame. This allows the actual decision making to be performed
25037 when each frame is printed. This is an important consideration, and
25038 well worth reflecting upon when designing a frame filter. An issue
25039 that frame filters should avoid is unwinding the stack if possible.
25040 Some stacks can run very deep, into the tens of thousands in some
25041 cases. To search every frame to determine if it is inlined ahead of
25042 time may be too expensive at the filtering step. The frame filter
25043 cannot know how many frames it has to iterate over, and it would have
25044 to iterate through them all. This ends up duplicating effort as
25045 @value{GDBN} performs this iteration when it prints the frames.
25047 In this example decision making can be deferred to the printing step.
25048 As each frame is printed, the frame decorator can examine each frame
25049 in turn when @value{GDBN} iterates. From a performance viewpoint,
25050 this is the most appropriate decision to make as it avoids duplicating
25051 the effort that the printing step would undertake anyway. Also, if
25052 there are many frame filters unwinding the stack during filtering, it
25053 can substantially delay the printing of the backtrace which will
25054 result in large memory usage, and a poor user experience.
25057 class InlineFilter():
25059 def __init__(self):
25060 self.name = "InlinedFrameFilter"
25061 self.priority = 100
25062 self.enabled = True
25063 gdb.frame_filters[self.name] = self
25065 def filter(self, frame_iter):
25066 frame_iter = itertools.imap(InlinedFrameDecorator,
25071 This frame filter is somewhat similar to the earlier example, except
25072 that the @code{filter} method applies a frame decorator object called
25073 @code{InlinedFrameDecorator} to each element in the iterator. The
25074 @code{imap} Python method is light-weight. It does not proactively
25075 iterate over the iterator, but rather creates a new iterator which
25076 wraps the existing one.
25078 Below is the frame decorator for this example.
25081 class InlinedFrameDecorator(FrameDecorator):
25083 def __init__(self, fobj):
25084 super(InlinedFrameDecorator, self).__init__(fobj)
25086 def function(self):
25087 frame = fobj.inferior_frame()
25088 name = str(frame.name())
25090 if frame.type() == gdb.INLINE_FRAME:
25091 name = name + " [inlined]"
25096 This frame decorator only defines and overrides the @code{function}
25097 method. It lets the supplied @code{FrameDecorator}, which is shipped
25098 with @value{GDBN}, perform the other work associated with printing
25101 The combination of these two objects create this output from a
25105 #0 0x004004e0 in bar () at inline.c:11
25106 #1 0x00400566 in max [inlined] (b=6, a=12) at inline.c:21
25107 #2 0x00400566 in main () at inline.c:31
25110 So in the case of this example, a frame decorator is applied to all
25111 frames, regardless of whether they may be inlined or not. As
25112 @value{GDBN} iterates over the iterator produced by the frame filters,
25113 @value{GDBN} executes each frame decorator which then makes a decision
25114 on what to print in the @code{function} callback. Using a strategy
25115 like this is a way to defer decisions on the frame content to printing
25118 @subheading Eliding Frames
25120 It might be that the above example is not desirable for representing
25121 inlined frames, and a hierarchical approach may be preferred. If we
25122 want to hierarchically represent frames, the @code{elided} frame
25123 decorator interface might be preferable.
25125 This example approaches the issue with the @code{elided} method. This
25126 example is quite long, but very simplistic. It is out-of-scope for
25127 this section to write a complete example that comprehensively covers
25128 all approaches of finding and printing inlined frames. However, this
25129 example illustrates the approach an author might use.
25131 This example comprises of three sections.
25134 class InlineFrameFilter():
25136 def __init__(self):
25137 self.name = "InlinedFrameFilter"
25138 self.priority = 100
25139 self.enabled = True
25140 gdb.frame_filters[self.name] = self
25142 def filter(self, frame_iter):
25143 return ElidingInlineIterator(frame_iter)
25146 This frame filter is very similar to the other examples. The only
25147 difference is this frame filter is wrapping the iterator provided to
25148 it (@code{frame_iter}) with a custom iterator called
25149 @code{ElidingInlineIterator}. This again defers actions to when
25150 @value{GDBN} prints the backtrace, as the iterator is not traversed
25153 The iterator for this example is as follows. It is in this section of
25154 the example where decisions are made on the content of the backtrace.
25157 class ElidingInlineIterator:
25158 def __init__(self, ii):
25159 self.input_iterator = ii
25161 def __iter__(self):
25165 frame = next(self.input_iterator)
25167 if frame.inferior_frame().type() != gdb.INLINE_FRAME:
25171 eliding_frame = next(self.input_iterator)
25172 except StopIteration:
25174 return ElidingFrameDecorator(eliding_frame, [frame])
25177 This iterator implements the Python iterator protocol. When the
25178 @code{next} function is called (when @value{GDBN} prints each frame),
25179 the iterator checks if this frame decorator, @code{frame}, is wrapping
25180 an inlined frame. If it is not, it returns the existing frame decorator
25181 untouched. If it is wrapping an inlined frame, it assumes that the
25182 inlined frame was contained within the next oldest frame,
25183 @code{eliding_frame}, which it fetches. It then creates and returns a
25184 frame decorator, @code{ElidingFrameDecorator}, which contains both the
25185 elided frame, and the eliding frame.
25188 class ElidingInlineDecorator(FrameDecorator):
25190 def __init__(self, frame, elided_frames):
25191 super(ElidingInlineDecorator, self).__init__(frame)
25193 self.elided_frames = elided_frames
25196 return iter(self.elided_frames)
25199 This frame decorator overrides one function and returns the inlined
25200 frame in the @code{elided} method. As before it lets
25201 @code{FrameDecorator} do the rest of the work involved in printing
25202 this frame. This produces the following output.
25205 #0 0x004004e0 in bar () at inline.c:11
25206 #2 0x00400529 in main () at inline.c:25
25207 #1 0x00400529 in max (b=6, a=12) at inline.c:15
25210 In that output, @code{max} which has been inlined into @code{main} is
25211 printed hierarchically. Another approach would be to combine the
25212 @code{function} method, and the @code{elided} method to both print a
25213 marker in the inlined frame, and also show the hierarchical
25216 @node Inferiors In Python
25217 @subsubsection Inferiors In Python
25218 @cindex inferiors in Python
25220 @findex gdb.Inferior
25221 Programs which are being run under @value{GDBN} are called inferiors
25222 (@pxref{Inferiors and Programs}). Python scripts can access
25223 information about and manipulate inferiors controlled by @value{GDBN}
25224 via objects of the @code{gdb.Inferior} class.
25226 The following inferior-related functions are available in the @code{gdb}
25229 @defun gdb.inferiors ()
25230 Return a tuple containing all inferior objects.
25233 @defun gdb.selected_inferior ()
25234 Return an object representing the current inferior.
25237 A @code{gdb.Inferior} object has the following attributes:
25239 @defvar Inferior.num
25240 ID of inferior, as assigned by GDB.
25243 @defvar Inferior.pid
25244 Process ID of the inferior, as assigned by the underlying operating
25248 @defvar Inferior.was_attached
25249 Boolean signaling whether the inferior was created using `attach', or
25250 started by @value{GDBN} itself.
25253 A @code{gdb.Inferior} object has the following methods:
25255 @defun Inferior.is_valid ()
25256 Returns @code{True} if the @code{gdb.Inferior} object is valid,
25257 @code{False} if not. A @code{gdb.Inferior} object will become invalid
25258 if the inferior no longer exists within @value{GDBN}. All other
25259 @code{gdb.Inferior} methods will throw an exception if it is invalid
25260 at the time the method is called.
25263 @defun Inferior.threads ()
25264 This method returns a tuple holding all the threads which are valid
25265 when it is called. If there are no valid threads, the method will
25266 return an empty tuple.
25269 @findex Inferior.read_memory
25270 @defun Inferior.read_memory (address, length)
25271 Read @var{length} bytes of memory from the inferior, starting at
25272 @var{address}. Returns a buffer object, which behaves much like an array
25273 or a string. It can be modified and given to the
25274 @code{Inferior.write_memory} function. In @code{Python} 3, the return
25275 value is a @code{memoryview} object.
25278 @findex Inferior.write_memory
25279 @defun Inferior.write_memory (address, buffer @r{[}, length@r{]})
25280 Write the contents of @var{buffer} to the inferior, starting at
25281 @var{address}. The @var{buffer} parameter must be a Python object
25282 which supports the buffer protocol, i.e., a string, an array or the
25283 object returned from @code{Inferior.read_memory}. If given, @var{length}
25284 determines the number of bytes from @var{buffer} to be written.
25287 @findex gdb.search_memory
25288 @defun Inferior.search_memory (address, length, pattern)
25289 Search a region of the inferior memory starting at @var{address} with
25290 the given @var{length} using the search pattern supplied in
25291 @var{pattern}. The @var{pattern} parameter must be a Python object
25292 which supports the buffer protocol, i.e., a string, an array or the
25293 object returned from @code{gdb.read_memory}. Returns a Python @code{Long}
25294 containing the address where the pattern was found, or @code{None} if
25295 the pattern could not be found.
25298 @node Events In Python
25299 @subsubsection Events In Python
25300 @cindex inferior events in Python
25302 @value{GDBN} provides a general event facility so that Python code can be
25303 notified of various state changes, particularly changes that occur in
25306 An @dfn{event} is just an object that describes some state change. The
25307 type of the object and its attributes will vary depending on the details
25308 of the change. All the existing events are described below.
25310 In order to be notified of an event, you must register an event handler
25311 with an @dfn{event registry}. An event registry is an object in the
25312 @code{gdb.events} module which dispatches particular events. A registry
25313 provides methods to register and unregister event handlers:
25315 @defun EventRegistry.connect (object)
25316 Add the given callable @var{object} to the registry. This object will be
25317 called when an event corresponding to this registry occurs.
25320 @defun EventRegistry.disconnect (object)
25321 Remove the given @var{object} from the registry. Once removed, the object
25322 will no longer receive notifications of events.
25325 Here is an example:
25328 def exit_handler (event):
25329 print "event type: exit"
25330 print "exit code: %d" % (event.exit_code)
25332 gdb.events.exited.connect (exit_handler)
25335 In the above example we connect our handler @code{exit_handler} to the
25336 registry @code{events.exited}. Once connected, @code{exit_handler} gets
25337 called when the inferior exits. The argument @dfn{event} in this example is
25338 of type @code{gdb.ExitedEvent}. As you can see in the example the
25339 @code{ExitedEvent} object has an attribute which indicates the exit code of
25342 The following is a listing of the event registries that are available and
25343 details of the events they emit:
25348 Emits @code{gdb.ThreadEvent}.
25350 Some events can be thread specific when @value{GDBN} is running in non-stop
25351 mode. When represented in Python, these events all extend
25352 @code{gdb.ThreadEvent}. Note, this event is not emitted directly; instead,
25353 events which are emitted by this or other modules might extend this event.
25354 Examples of these events are @code{gdb.BreakpointEvent} and
25355 @code{gdb.ContinueEvent}.
25357 @defvar ThreadEvent.inferior_thread
25358 In non-stop mode this attribute will be set to the specific thread which was
25359 involved in the emitted event. Otherwise, it will be set to @code{None}.
25362 Emits @code{gdb.ContinueEvent} which extends @code{gdb.ThreadEvent}.
25364 This event indicates that the inferior has been continued after a stop. For
25365 inherited attribute refer to @code{gdb.ThreadEvent} above.
25367 @item events.exited
25368 Emits @code{events.ExitedEvent} which indicates that the inferior has exited.
25369 @code{events.ExitedEvent} has two attributes:
25370 @defvar ExitedEvent.exit_code
25371 An integer representing the exit code, if available, which the inferior
25372 has returned. (The exit code could be unavailable if, for example,
25373 @value{GDBN} detaches from the inferior.) If the exit code is unavailable,
25374 the attribute does not exist.
25376 @defvar ExitedEvent inferior
25377 A reference to the inferior which triggered the @code{exited} event.
25381 Emits @code{gdb.StopEvent} which extends @code{gdb.ThreadEvent}.
25383 Indicates that the inferior has stopped. All events emitted by this registry
25384 extend StopEvent. As a child of @code{gdb.ThreadEvent}, @code{gdb.StopEvent}
25385 will indicate the stopped thread when @value{GDBN} is running in non-stop
25386 mode. Refer to @code{gdb.ThreadEvent} above for more details.
25388 Emits @code{gdb.SignalEvent} which extends @code{gdb.StopEvent}.
25390 This event indicates that the inferior or one of its threads has received as
25391 signal. @code{gdb.SignalEvent} has the following attributes:
25393 @defvar SignalEvent.stop_signal
25394 A string representing the signal received by the inferior. A list of possible
25395 signal values can be obtained by running the command @code{info signals} in
25396 the @value{GDBN} command prompt.
25399 Also emits @code{gdb.BreakpointEvent} which extends @code{gdb.StopEvent}.
25401 @code{gdb.BreakpointEvent} event indicates that one or more breakpoints have
25402 been hit, and has the following attributes:
25404 @defvar BreakpointEvent.breakpoints
25405 A sequence containing references to all the breakpoints (type
25406 @code{gdb.Breakpoint}) that were hit.
25407 @xref{Breakpoints In Python}, for details of the @code{gdb.Breakpoint} object.
25409 @defvar BreakpointEvent.breakpoint
25410 A reference to the first breakpoint that was hit.
25411 This function is maintained for backward compatibility and is now deprecated
25412 in favor of the @code{gdb.BreakpointEvent.breakpoints} attribute.
25415 @item events.new_objfile
25416 Emits @code{gdb.NewObjFileEvent} which indicates that a new object file has
25417 been loaded by @value{GDBN}. @code{gdb.NewObjFileEvent} has one attribute:
25419 @defvar NewObjFileEvent.new_objfile
25420 A reference to the object file (@code{gdb.Objfile}) which has been loaded.
25421 @xref{Objfiles In Python}, for details of the @code{gdb.Objfile} object.
25426 @node Threads In Python
25427 @subsubsection Threads In Python
25428 @cindex threads in python
25430 @findex gdb.InferiorThread
25431 Python scripts can access information about, and manipulate inferior threads
25432 controlled by @value{GDBN}, via objects of the @code{gdb.InferiorThread} class.
25434 The following thread-related functions are available in the @code{gdb}
25437 @findex gdb.selected_thread
25438 @defun gdb.selected_thread ()
25439 This function returns the thread object for the selected thread. If there
25440 is no selected thread, this will return @code{None}.
25443 A @code{gdb.InferiorThread} object has the following attributes:
25445 @defvar InferiorThread.name
25446 The name of the thread. If the user specified a name using
25447 @code{thread name}, then this returns that name. Otherwise, if an
25448 OS-supplied name is available, then it is returned. Otherwise, this
25449 returns @code{None}.
25451 This attribute can be assigned to. The new value must be a string
25452 object, which sets the new name, or @code{None}, which removes any
25453 user-specified thread name.
25456 @defvar InferiorThread.num
25457 ID of the thread, as assigned by GDB.
25460 @defvar InferiorThread.ptid
25461 ID of the thread, as assigned by the operating system. This attribute is a
25462 tuple containing three integers. The first is the Process ID (PID); the second
25463 is the Lightweight Process ID (LWPID), and the third is the Thread ID (TID).
25464 Either the LWPID or TID may be 0, which indicates that the operating system
25465 does not use that identifier.
25468 A @code{gdb.InferiorThread} object has the following methods:
25470 @defun InferiorThread.is_valid ()
25471 Returns @code{True} if the @code{gdb.InferiorThread} object is valid,
25472 @code{False} if not. A @code{gdb.InferiorThread} object will become
25473 invalid if the thread exits, or the inferior that the thread belongs
25474 is deleted. All other @code{gdb.InferiorThread} methods will throw an
25475 exception if it is invalid at the time the method is called.
25478 @defun InferiorThread.switch ()
25479 This changes @value{GDBN}'s currently selected thread to the one represented
25483 @defun InferiorThread.is_stopped ()
25484 Return a Boolean indicating whether the thread is stopped.
25487 @defun InferiorThread.is_running ()
25488 Return a Boolean indicating whether the thread is running.
25491 @defun InferiorThread.is_exited ()
25492 Return a Boolean indicating whether the thread is exited.
25495 @node Commands In Python
25496 @subsubsection Commands In Python
25498 @cindex commands in python
25499 @cindex python commands
25500 You can implement new @value{GDBN} CLI commands in Python. A CLI
25501 command is implemented using an instance of the @code{gdb.Command}
25502 class, most commonly using a subclass.
25504 @defun Command.__init__ (name, @var{command_class} @r{[}, @var{completer_class} @r{[}, @var{prefix}@r{]]})
25505 The object initializer for @code{Command} registers the new command
25506 with @value{GDBN}. This initializer is normally invoked from the
25507 subclass' own @code{__init__} method.
25509 @var{name} is the name of the command. If @var{name} consists of
25510 multiple words, then the initial words are looked for as prefix
25511 commands. In this case, if one of the prefix commands does not exist,
25512 an exception is raised.
25514 There is no support for multi-line commands.
25516 @var{command_class} should be one of the @samp{COMMAND_} constants
25517 defined below. This argument tells @value{GDBN} how to categorize the
25518 new command in the help system.
25520 @var{completer_class} is an optional argument. If given, it should be
25521 one of the @samp{COMPLETE_} constants defined below. This argument
25522 tells @value{GDBN} how to perform completion for this command. If not
25523 given, @value{GDBN} will attempt to complete using the object's
25524 @code{complete} method (see below); if no such method is found, an
25525 error will occur when completion is attempted.
25527 @var{prefix} is an optional argument. If @code{True}, then the new
25528 command is a prefix command; sub-commands of this command may be
25531 The help text for the new command is taken from the Python
25532 documentation string for the command's class, if there is one. If no
25533 documentation string is provided, the default value ``This command is
25534 not documented.'' is used.
25537 @cindex don't repeat Python command
25538 @defun Command.dont_repeat ()
25539 By default, a @value{GDBN} command is repeated when the user enters a
25540 blank line at the command prompt. A command can suppress this
25541 behavior by invoking the @code{dont_repeat} method. This is similar
25542 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
25545 @defun Command.invoke (argument, from_tty)
25546 This method is called by @value{GDBN} when this command is invoked.
25548 @var{argument} is a string. It is the argument to the command, after
25549 leading and trailing whitespace has been stripped.
25551 @var{from_tty} is a boolean argument. When true, this means that the
25552 command was entered by the user at the terminal; when false it means
25553 that the command came from elsewhere.
25555 If this method throws an exception, it is turned into a @value{GDBN}
25556 @code{error} call. Otherwise, the return value is ignored.
25558 @findex gdb.string_to_argv
25559 To break @var{argument} up into an argv-like string use
25560 @code{gdb.string_to_argv}. This function behaves identically to
25561 @value{GDBN}'s internal argument lexer @code{buildargv}.
25562 It is recommended to use this for consistency.
25563 Arguments are separated by spaces and may be quoted.
25567 print gdb.string_to_argv ("1 2\ \\\"3 '4 \"5' \"6 '7\"")
25568 ['1', '2 "3', '4 "5', "6 '7"]
25573 @cindex completion of Python commands
25574 @defun Command.complete (text, word)
25575 This method is called by @value{GDBN} when the user attempts
25576 completion on this command. All forms of completion are handled by
25577 this method, that is, the @key{TAB} and @key{M-?} key bindings
25578 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
25581 The arguments @var{text} and @var{word} are both strings. @var{text}
25582 holds the complete command line up to the cursor's location.
25583 @var{word} holds the last word of the command line; this is computed
25584 using a word-breaking heuristic.
25586 The @code{complete} method can return several values:
25589 If the return value is a sequence, the contents of the sequence are
25590 used as the completions. It is up to @code{complete} to ensure that the
25591 contents actually do complete the word. A zero-length sequence is
25592 allowed, it means that there were no completions available. Only
25593 string elements of the sequence are used; other elements in the
25594 sequence are ignored.
25597 If the return value is one of the @samp{COMPLETE_} constants defined
25598 below, then the corresponding @value{GDBN}-internal completion
25599 function is invoked, and its result is used.
25602 All other results are treated as though there were no available
25607 When a new command is registered, it must be declared as a member of
25608 some general class of commands. This is used to classify top-level
25609 commands in the on-line help system; note that prefix commands are not
25610 listed under their own category but rather that of their top-level
25611 command. The available classifications are represented by constants
25612 defined in the @code{gdb} module:
25615 @findex COMMAND_NONE
25616 @findex gdb.COMMAND_NONE
25617 @item gdb.COMMAND_NONE
25618 The command does not belong to any particular class. A command in
25619 this category will not be displayed in any of the help categories.
25621 @findex COMMAND_RUNNING
25622 @findex gdb.COMMAND_RUNNING
25623 @item gdb.COMMAND_RUNNING
25624 The command is related to running the inferior. For example,
25625 @code{start}, @code{step}, and @code{continue} are in this category.
25626 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
25627 commands in this category.
25629 @findex COMMAND_DATA
25630 @findex gdb.COMMAND_DATA
25631 @item gdb.COMMAND_DATA
25632 The command is related to data or variables. For example,
25633 @code{call}, @code{find}, and @code{print} are in this category. Type
25634 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
25637 @findex COMMAND_STACK
25638 @findex gdb.COMMAND_STACK
25639 @item gdb.COMMAND_STACK
25640 The command has to do with manipulation of the stack. For example,
25641 @code{backtrace}, @code{frame}, and @code{return} are in this
25642 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
25643 list of commands in this category.
25645 @findex COMMAND_FILES
25646 @findex gdb.COMMAND_FILES
25647 @item gdb.COMMAND_FILES
25648 This class is used for file-related commands. For example,
25649 @code{file}, @code{list} and @code{section} are in this category.
25650 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
25651 commands in this category.
25653 @findex COMMAND_SUPPORT
25654 @findex gdb.COMMAND_SUPPORT
25655 @item gdb.COMMAND_SUPPORT
25656 This should be used for ``support facilities'', generally meaning
25657 things that are useful to the user when interacting with @value{GDBN},
25658 but not related to the state of the inferior. For example,
25659 @code{help}, @code{make}, and @code{shell} are in this category. Type
25660 @kbd{help support} at the @value{GDBN} prompt to see a list of
25661 commands in this category.
25663 @findex COMMAND_STATUS
25664 @findex gdb.COMMAND_STATUS
25665 @item gdb.COMMAND_STATUS
25666 The command is an @samp{info}-related command, that is, related to the
25667 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
25668 and @code{show} are in this category. Type @kbd{help status} at the
25669 @value{GDBN} prompt to see a list of commands in this category.
25671 @findex COMMAND_BREAKPOINTS
25672 @findex gdb.COMMAND_BREAKPOINTS
25673 @item gdb.COMMAND_BREAKPOINTS
25674 The command has to do with breakpoints. For example, @code{break},
25675 @code{clear}, and @code{delete} are in this category. Type @kbd{help
25676 breakpoints} at the @value{GDBN} prompt to see a list of commands in
25679 @findex COMMAND_TRACEPOINTS
25680 @findex gdb.COMMAND_TRACEPOINTS
25681 @item gdb.COMMAND_TRACEPOINTS
25682 The command has to do with tracepoints. For example, @code{trace},
25683 @code{actions}, and @code{tfind} are in this category. Type
25684 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
25685 commands in this category.
25687 @findex COMMAND_USER
25688 @findex gdb.COMMAND_USER
25689 @item gdb.COMMAND_USER
25690 The command is a general purpose command for the user, and typically
25691 does not fit in one of the other categories.
25692 Type @kbd{help user-defined} at the @value{GDBN} prompt to see
25693 a list of commands in this category, as well as the list of gdb macros
25694 (@pxref{Sequences}).
25696 @findex COMMAND_OBSCURE
25697 @findex gdb.COMMAND_OBSCURE
25698 @item gdb.COMMAND_OBSCURE
25699 The command is only used in unusual circumstances, or is not of
25700 general interest to users. For example, @code{checkpoint},
25701 @code{fork}, and @code{stop} are in this category. Type @kbd{help
25702 obscure} at the @value{GDBN} prompt to see a list of commands in this
25705 @findex COMMAND_MAINTENANCE
25706 @findex gdb.COMMAND_MAINTENANCE
25707 @item gdb.COMMAND_MAINTENANCE
25708 The command is only useful to @value{GDBN} maintainers. The
25709 @code{maintenance} and @code{flushregs} commands are in this category.
25710 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
25711 commands in this category.
25714 A new command can use a predefined completion function, either by
25715 specifying it via an argument at initialization, or by returning it
25716 from the @code{complete} method. These predefined completion
25717 constants are all defined in the @code{gdb} module:
25720 @findex COMPLETE_NONE
25721 @findex gdb.COMPLETE_NONE
25722 @item gdb.COMPLETE_NONE
25723 This constant means that no completion should be done.
25725 @findex COMPLETE_FILENAME
25726 @findex gdb.COMPLETE_FILENAME
25727 @item gdb.COMPLETE_FILENAME
25728 This constant means that filename completion should be performed.
25730 @findex COMPLETE_LOCATION
25731 @findex gdb.COMPLETE_LOCATION
25732 @item gdb.COMPLETE_LOCATION
25733 This constant means that location completion should be done.
25734 @xref{Specify Location}.
25736 @findex COMPLETE_COMMAND
25737 @findex gdb.COMPLETE_COMMAND
25738 @item gdb.COMPLETE_COMMAND
25739 This constant means that completion should examine @value{GDBN}
25742 @findex COMPLETE_SYMBOL
25743 @findex gdb.COMPLETE_SYMBOL
25744 @item gdb.COMPLETE_SYMBOL
25745 This constant means that completion should be done using symbol names
25749 The following code snippet shows how a trivial CLI command can be
25750 implemented in Python:
25753 class HelloWorld (gdb.Command):
25754 """Greet the whole world."""
25756 def __init__ (self):
25757 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
25759 def invoke (self, arg, from_tty):
25760 print "Hello, World!"
25765 The last line instantiates the class, and is necessary to trigger the
25766 registration of the command with @value{GDBN}. Depending on how the
25767 Python code is read into @value{GDBN}, you may need to import the
25768 @code{gdb} module explicitly.
25770 @node Parameters In Python
25771 @subsubsection Parameters In Python
25773 @cindex parameters in python
25774 @cindex python parameters
25775 @tindex gdb.Parameter
25777 You can implement new @value{GDBN} parameters using Python. A new
25778 parameter is implemented as an instance of the @code{gdb.Parameter}
25781 Parameters are exposed to the user via the @code{set} and
25782 @code{show} commands. @xref{Help}.
25784 There are many parameters that already exist and can be set in
25785 @value{GDBN}. Two examples are: @code{set follow fork} and
25786 @code{set charset}. Setting these parameters influences certain
25787 behavior in @value{GDBN}. Similarly, you can define parameters that
25788 can be used to influence behavior in custom Python scripts and commands.
25790 @defun Parameter.__init__ (name, @var{command-class}, @var{parameter-class} @r{[}, @var{enum-sequence}@r{]})
25791 The object initializer for @code{Parameter} registers the new
25792 parameter with @value{GDBN}. This initializer is normally invoked
25793 from the subclass' own @code{__init__} method.
25795 @var{name} is the name of the new parameter. If @var{name} consists
25796 of multiple words, then the initial words are looked for as prefix
25797 parameters. An example of this can be illustrated with the
25798 @code{set print} set of parameters. If @var{name} is
25799 @code{print foo}, then @code{print} will be searched as the prefix
25800 parameter. In this case the parameter can subsequently be accessed in
25801 @value{GDBN} as @code{set print foo}.
25803 If @var{name} consists of multiple words, and no prefix parameter group
25804 can be found, an exception is raised.
25806 @var{command-class} should be one of the @samp{COMMAND_} constants
25807 (@pxref{Commands In Python}). This argument tells @value{GDBN} how to
25808 categorize the new parameter in the help system.
25810 @var{parameter-class} should be one of the @samp{PARAM_} constants
25811 defined below. This argument tells @value{GDBN} the type of the new
25812 parameter; this information is used for input validation and
25815 If @var{parameter-class} is @code{PARAM_ENUM}, then
25816 @var{enum-sequence} must be a sequence of strings. These strings
25817 represent the possible values for the parameter.
25819 If @var{parameter-class} is not @code{PARAM_ENUM}, then the presence
25820 of a fourth argument will cause an exception to be thrown.
25822 The help text for the new parameter is taken from the Python
25823 documentation string for the parameter's class, if there is one. If
25824 there is no documentation string, a default value is used.
25827 @defvar Parameter.set_doc
25828 If this attribute exists, and is a string, then its value is used as
25829 the help text for this parameter's @code{set} command. The value is
25830 examined when @code{Parameter.__init__} is invoked; subsequent changes
25834 @defvar Parameter.show_doc
25835 If this attribute exists, and is a string, then its value is used as
25836 the help text for this parameter's @code{show} command. The value is
25837 examined when @code{Parameter.__init__} is invoked; subsequent changes
25841 @defvar Parameter.value
25842 The @code{value} attribute holds the underlying value of the
25843 parameter. It can be read and assigned to just as any other
25844 attribute. @value{GDBN} does validation when assignments are made.
25847 There are two methods that should be implemented in any
25848 @code{Parameter} class. These are:
25850 @defun Parameter.get_set_string (self)
25851 @value{GDBN} will call this method when a @var{parameter}'s value has
25852 been changed via the @code{set} API (for example, @kbd{set foo off}).
25853 The @code{value} attribute has already been populated with the new
25854 value and may be used in output. This method must return a string.
25857 @defun Parameter.get_show_string (self, svalue)
25858 @value{GDBN} will call this method when a @var{parameter}'s
25859 @code{show} API has been invoked (for example, @kbd{show foo}). The
25860 argument @code{svalue} receives the string representation of the
25861 current value. This method must return a string.
25864 When a new parameter is defined, its type must be specified. The
25865 available types are represented by constants defined in the @code{gdb}
25869 @findex PARAM_BOOLEAN
25870 @findex gdb.PARAM_BOOLEAN
25871 @item gdb.PARAM_BOOLEAN
25872 The value is a plain boolean. The Python boolean values, @code{True}
25873 and @code{False} are the only valid values.
25875 @findex PARAM_AUTO_BOOLEAN
25876 @findex gdb.PARAM_AUTO_BOOLEAN
25877 @item gdb.PARAM_AUTO_BOOLEAN
25878 The value has three possible states: true, false, and @samp{auto}. In
25879 Python, true and false are represented using boolean constants, and
25880 @samp{auto} is represented using @code{None}.
25882 @findex PARAM_UINTEGER
25883 @findex gdb.PARAM_UINTEGER
25884 @item gdb.PARAM_UINTEGER
25885 The value is an unsigned integer. The value of 0 should be
25886 interpreted to mean ``unlimited''.
25888 @findex PARAM_INTEGER
25889 @findex gdb.PARAM_INTEGER
25890 @item gdb.PARAM_INTEGER
25891 The value is a signed integer. The value of 0 should be interpreted
25892 to mean ``unlimited''.
25894 @findex PARAM_STRING
25895 @findex gdb.PARAM_STRING
25896 @item gdb.PARAM_STRING
25897 The value is a string. When the user modifies the string, any escape
25898 sequences, such as @samp{\t}, @samp{\f}, and octal escapes, are
25899 translated into corresponding characters and encoded into the current
25902 @findex PARAM_STRING_NOESCAPE
25903 @findex gdb.PARAM_STRING_NOESCAPE
25904 @item gdb.PARAM_STRING_NOESCAPE
25905 The value is a string. When the user modifies the string, escapes are
25906 passed through untranslated.
25908 @findex PARAM_OPTIONAL_FILENAME
25909 @findex gdb.PARAM_OPTIONAL_FILENAME
25910 @item gdb.PARAM_OPTIONAL_FILENAME
25911 The value is a either a filename (a string), or @code{None}.
25913 @findex PARAM_FILENAME
25914 @findex gdb.PARAM_FILENAME
25915 @item gdb.PARAM_FILENAME
25916 The value is a filename. This is just like
25917 @code{PARAM_STRING_NOESCAPE}, but uses file names for completion.
25919 @findex PARAM_ZINTEGER
25920 @findex gdb.PARAM_ZINTEGER
25921 @item gdb.PARAM_ZINTEGER
25922 The value is an integer. This is like @code{PARAM_INTEGER}, except 0
25923 is interpreted as itself.
25926 @findex gdb.PARAM_ENUM
25927 @item gdb.PARAM_ENUM
25928 The value is a string, which must be one of a collection string
25929 constants provided when the parameter is created.
25932 @node Functions In Python
25933 @subsubsection Writing new convenience functions
25935 @cindex writing convenience functions
25936 @cindex convenience functions in python
25937 @cindex python convenience functions
25938 @tindex gdb.Function
25940 You can implement new convenience functions (@pxref{Convenience Vars})
25941 in Python. A convenience function is an instance of a subclass of the
25942 class @code{gdb.Function}.
25944 @defun Function.__init__ (name)
25945 The initializer for @code{Function} registers the new function with
25946 @value{GDBN}. The argument @var{name} is the name of the function,
25947 a string. The function will be visible to the user as a convenience
25948 variable of type @code{internal function}, whose name is the same as
25949 the given @var{name}.
25951 The documentation for the new function is taken from the documentation
25952 string for the new class.
25955 @defun Function.invoke (@var{*args})
25956 When a convenience function is evaluated, its arguments are converted
25957 to instances of @code{gdb.Value}, and then the function's
25958 @code{invoke} method is called. Note that @value{GDBN} does not
25959 predetermine the arity of convenience functions. Instead, all
25960 available arguments are passed to @code{invoke}, following the
25961 standard Python calling convention. In particular, a convenience
25962 function can have default values for parameters without ill effect.
25964 The return value of this method is used as its value in the enclosing
25965 expression. If an ordinary Python value is returned, it is converted
25966 to a @code{gdb.Value} following the usual rules.
25969 The following code snippet shows how a trivial convenience function can
25970 be implemented in Python:
25973 class Greet (gdb.Function):
25974 """Return string to greet someone.
25975 Takes a name as argument."""
25977 def __init__ (self):
25978 super (Greet, self).__init__ ("greet")
25980 def invoke (self, name):
25981 return "Hello, %s!" % name.string ()
25986 The last line instantiates the class, and is necessary to trigger the
25987 registration of the function with @value{GDBN}. Depending on how the
25988 Python code is read into @value{GDBN}, you may need to import the
25989 @code{gdb} module explicitly.
25991 Now you can use the function in an expression:
25994 (gdb) print $greet("Bob")
25998 @node Progspaces In Python
25999 @subsubsection Program Spaces In Python
26001 @cindex progspaces in python
26002 @tindex gdb.Progspace
26004 A program space, or @dfn{progspace}, represents a symbolic view
26005 of an address space.
26006 It consists of all of the objfiles of the program.
26007 @xref{Objfiles In Python}.
26008 @xref{Inferiors and Programs, program spaces}, for more details
26009 about program spaces.
26011 The following progspace-related functions are available in the
26014 @findex gdb.current_progspace
26015 @defun gdb.current_progspace ()
26016 This function returns the program space of the currently selected inferior.
26017 @xref{Inferiors and Programs}.
26020 @findex gdb.progspaces
26021 @defun gdb.progspaces ()
26022 Return a sequence of all the progspaces currently known to @value{GDBN}.
26025 Each progspace is represented by an instance of the @code{gdb.Progspace}
26028 @defvar Progspace.filename
26029 The file name of the progspace as a string.
26032 @defvar Progspace.pretty_printers
26033 The @code{pretty_printers} attribute is a list of functions. It is
26034 used to look up pretty-printers. A @code{Value} is passed to each
26035 function in order; if the function returns @code{None}, then the
26036 search continues. Otherwise, the return value should be an object
26037 which is used to format the value. @xref{Pretty Printing API}, for more
26041 @defvar Progspace.type_printers
26042 The @code{type_printers} attribute is a list of type printer objects.
26043 @xref{Type Printing API}, for more information.
26046 @defvar Progspace.frame_filters
26047 The @code{frame_filters} attribute is a dictionary of frame filter
26048 objects. @xref{Frame Filter API}, for more information.
26051 @node Objfiles In Python
26052 @subsubsection Objfiles In Python
26054 @cindex objfiles in python
26055 @tindex gdb.Objfile
26057 @value{GDBN} loads symbols for an inferior from various
26058 symbol-containing files (@pxref{Files}). These include the primary
26059 executable file, any shared libraries used by the inferior, and any
26060 separate debug info files (@pxref{Separate Debug Files}).
26061 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
26063 The following objfile-related functions are available in the
26066 @findex gdb.current_objfile
26067 @defun gdb.current_objfile ()
26068 When auto-loading a Python script (@pxref{Python Auto-loading}), @value{GDBN}
26069 sets the ``current objfile'' to the corresponding objfile. This
26070 function returns the current objfile. If there is no current objfile,
26071 this function returns @code{None}.
26074 @findex gdb.objfiles
26075 @defun gdb.objfiles ()
26076 Return a sequence of all the objfiles current known to @value{GDBN}.
26077 @xref{Objfiles In Python}.
26080 Each objfile is represented by an instance of the @code{gdb.Objfile}
26083 @defvar Objfile.filename
26084 The file name of the objfile as a string.
26087 @defvar Objfile.pretty_printers
26088 The @code{pretty_printers} attribute is a list of functions. It is
26089 used to look up pretty-printers. A @code{Value} is passed to each
26090 function in order; if the function returns @code{None}, then the
26091 search continues. Otherwise, the return value should be an object
26092 which is used to format the value. @xref{Pretty Printing API}, for more
26096 @defvar Objfile.type_printers
26097 The @code{type_printers} attribute is a list of type printer objects.
26098 @xref{Type Printing API}, for more information.
26101 @defvar Objfile.frame_filters
26102 The @code{frame_filters} attribute is a dictionary of frame filter
26103 objects. @xref{Frame Filter API}, for more information.
26106 A @code{gdb.Objfile} object has the following methods:
26108 @defun Objfile.is_valid ()
26109 Returns @code{True} if the @code{gdb.Objfile} object is valid,
26110 @code{False} if not. A @code{gdb.Objfile} object can become invalid
26111 if the object file it refers to is not loaded in @value{GDBN} any
26112 longer. All other @code{gdb.Objfile} methods will throw an exception
26113 if it is invalid at the time the method is called.
26116 @node Frames In Python
26117 @subsubsection Accessing inferior stack frames from Python.
26119 @cindex frames in python
26120 When the debugged program stops, @value{GDBN} is able to analyze its call
26121 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
26122 represents a frame in the stack. A @code{gdb.Frame} object is only valid
26123 while its corresponding frame exists in the inferior's stack. If you try
26124 to use an invalid frame object, @value{GDBN} will throw a @code{gdb.error}
26125 exception (@pxref{Exception Handling}).
26127 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
26131 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
26135 The following frame-related functions are available in the @code{gdb} module:
26137 @findex gdb.selected_frame
26138 @defun gdb.selected_frame ()
26139 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
26142 @findex gdb.newest_frame
26143 @defun gdb.newest_frame ()
26144 Return the newest frame object for the selected thread.
26147 @defun gdb.frame_stop_reason_string (reason)
26148 Return a string explaining the reason why @value{GDBN} stopped unwinding
26149 frames, as expressed by the given @var{reason} code (an integer, see the
26150 @code{unwind_stop_reason} method further down in this section).
26153 A @code{gdb.Frame} object has the following methods:
26155 @defun Frame.is_valid ()
26156 Returns true if the @code{gdb.Frame} object is valid, false if not.
26157 A frame object can become invalid if the frame it refers to doesn't
26158 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
26159 an exception if it is invalid at the time the method is called.
26162 @defun Frame.name ()
26163 Returns the function name of the frame, or @code{None} if it can't be
26167 @defun Frame.architecture ()
26168 Returns the @code{gdb.Architecture} object corresponding to the frame's
26169 architecture. @xref{Architectures In Python}.
26172 @defun Frame.type ()
26173 Returns the type of the frame. The value can be one of:
26175 @item gdb.NORMAL_FRAME
26176 An ordinary stack frame.
26178 @item gdb.DUMMY_FRAME
26179 A fake stack frame that was created by @value{GDBN} when performing an
26180 inferior function call.
26182 @item gdb.INLINE_FRAME
26183 A frame representing an inlined function. The function was inlined
26184 into a @code{gdb.NORMAL_FRAME} that is older than this one.
26186 @item gdb.TAILCALL_FRAME
26187 A frame representing a tail call. @xref{Tail Call Frames}.
26189 @item gdb.SIGTRAMP_FRAME
26190 A signal trampoline frame. This is the frame created by the OS when
26191 it calls into a signal handler.
26193 @item gdb.ARCH_FRAME
26194 A fake stack frame representing a cross-architecture call.
26196 @item gdb.SENTINEL_FRAME
26197 This is like @code{gdb.NORMAL_FRAME}, but it is only used for the
26202 @defun Frame.unwind_stop_reason ()
26203 Return an integer representing the reason why it's not possible to find
26204 more frames toward the outermost frame. Use
26205 @code{gdb.frame_stop_reason_string} to convert the value returned by this
26206 function to a string. The value can be one of:
26209 @item gdb.FRAME_UNWIND_NO_REASON
26210 No particular reason (older frames should be available).
26212 @item gdb.FRAME_UNWIND_NULL_ID
26213 The previous frame's analyzer returns an invalid result.
26215 @item gdb.FRAME_UNWIND_OUTERMOST
26216 This frame is the outermost.
26218 @item gdb.FRAME_UNWIND_UNAVAILABLE
26219 Cannot unwind further, because that would require knowing the
26220 values of registers or memory that have not been collected.
26222 @item gdb.FRAME_UNWIND_INNER_ID
26223 This frame ID looks like it ought to belong to a NEXT frame,
26224 but we got it for a PREV frame. Normally, this is a sign of
26225 unwinder failure. It could also indicate stack corruption.
26227 @item gdb.FRAME_UNWIND_SAME_ID
26228 This frame has the same ID as the previous one. That means
26229 that unwinding further would almost certainly give us another
26230 frame with exactly the same ID, so break the chain. Normally,
26231 this is a sign of unwinder failure. It could also indicate
26234 @item gdb.FRAME_UNWIND_NO_SAVED_PC
26235 The frame unwinder did not find any saved PC, but we needed
26236 one to unwind further.
26238 @item gdb.FRAME_UNWIND_FIRST_ERROR
26239 Any stop reason greater or equal to this value indicates some kind
26240 of error. This special value facilitates writing code that tests
26241 for errors in unwinding in a way that will work correctly even if
26242 the list of the other values is modified in future @value{GDBN}
26243 versions. Using it, you could write:
26245 reason = gdb.selected_frame().unwind_stop_reason ()
26246 reason_str = gdb.frame_stop_reason_string (reason)
26247 if reason >= gdb.FRAME_UNWIND_FIRST_ERROR:
26248 print "An error occured: %s" % reason_str
26255 Returns the frame's resume address.
26258 @defun Frame.block ()
26259 Return the frame's code block. @xref{Blocks In Python}.
26262 @defun Frame.function ()
26263 Return the symbol for the function corresponding to this frame.
26264 @xref{Symbols In Python}.
26267 @defun Frame.older ()
26268 Return the frame that called this frame.
26271 @defun Frame.newer ()
26272 Return the frame called by this frame.
26275 @defun Frame.find_sal ()
26276 Return the frame's symtab and line object.
26277 @xref{Symbol Tables In Python}.
26280 @defun Frame.read_var (variable @r{[}, block@r{]})
26281 Return the value of @var{variable} in this frame. If the optional
26282 argument @var{block} is provided, search for the variable from that
26283 block; otherwise start at the frame's current block (which is
26284 determined by the frame's current program counter). @var{variable}
26285 must be a string or a @code{gdb.Symbol} object. @var{block} must be a
26286 @code{gdb.Block} object.
26289 @defun Frame.select ()
26290 Set this frame to be the selected frame. @xref{Stack, ,Examining the
26294 @node Blocks In Python
26295 @subsubsection Accessing blocks from Python.
26297 @cindex blocks in python
26300 In @value{GDBN}, symbols are stored in blocks. A block corresponds
26301 roughly to a scope in the source code. Blocks are organized
26302 hierarchically, and are represented individually in Python as a
26303 @code{gdb.Block}. Blocks rely on debugging information being
26306 A frame has a block. Please see @ref{Frames In Python}, for a more
26307 in-depth discussion of frames.
26309 The outermost block is known as the @dfn{global block}. The global
26310 block typically holds public global variables and functions.
26312 The block nested just inside the global block is the @dfn{static
26313 block}. The static block typically holds file-scoped variables and
26316 @value{GDBN} provides a method to get a block's superblock, but there
26317 is currently no way to examine the sub-blocks of a block, or to
26318 iterate over all the blocks in a symbol table (@pxref{Symbol Tables In
26321 Here is a short example that should help explain blocks:
26324 /* This is in the global block. */
26327 /* This is in the static block. */
26328 static int file_scope;
26330 /* 'function' is in the global block, and 'argument' is
26331 in a block nested inside of 'function'. */
26332 int function (int argument)
26334 /* 'local' is in a block inside 'function'. It may or may
26335 not be in the same block as 'argument'. */
26339 /* 'inner' is in a block whose superblock is the one holding
26343 /* If this call is expanded by the compiler, you may see
26344 a nested block here whose function is 'inline_function'
26345 and whose superblock is the one holding 'inner'. */
26346 inline_function ();
26351 A @code{gdb.Block} is iterable. The iterator returns the symbols
26352 (@pxref{Symbols In Python}) local to the block. Python programs
26353 should not assume that a specific block object will always contain a
26354 given symbol, since changes in @value{GDBN} features and
26355 infrastructure may cause symbols move across blocks in a symbol
26358 The following block-related functions are available in the @code{gdb}
26361 @findex gdb.block_for_pc
26362 @defun gdb.block_for_pc (pc)
26363 Return the innermost @code{gdb.Block} containing the given @var{pc}
26364 value. If the block cannot be found for the @var{pc} value specified,
26365 the function will return @code{None}.
26368 A @code{gdb.Block} object has the following methods:
26370 @defun Block.is_valid ()
26371 Returns @code{True} if the @code{gdb.Block} object is valid,
26372 @code{False} if not. A block object can become invalid if the block it
26373 refers to doesn't exist anymore in the inferior. All other
26374 @code{gdb.Block} methods will throw an exception if it is invalid at
26375 the time the method is called. The block's validity is also checked
26376 during iteration over symbols of the block.
26379 A @code{gdb.Block} object has the following attributes:
26381 @defvar Block.start
26382 The start address of the block. This attribute is not writable.
26386 The end address of the block. This attribute is not writable.
26389 @defvar Block.function
26390 The name of the block represented as a @code{gdb.Symbol}. If the
26391 block is not named, then this attribute holds @code{None}. This
26392 attribute is not writable.
26394 For ordinary function blocks, the superblock is the static block.
26395 However, you should note that it is possible for a function block to
26396 have a superblock that is not the static block -- for instance this
26397 happens for an inlined function.
26400 @defvar Block.superblock
26401 The block containing this block. If this parent block does not exist,
26402 this attribute holds @code{None}. This attribute is not writable.
26405 @defvar Block.global_block
26406 The global block associated with this block. This attribute is not
26410 @defvar Block.static_block
26411 The static block associated with this block. This attribute is not
26415 @defvar Block.is_global
26416 @code{True} if the @code{gdb.Block} object is a global block,
26417 @code{False} if not. This attribute is not
26421 @defvar Block.is_static
26422 @code{True} if the @code{gdb.Block} object is a static block,
26423 @code{False} if not. This attribute is not writable.
26426 @node Symbols In Python
26427 @subsubsection Python representation of Symbols.
26429 @cindex symbols in python
26432 @value{GDBN} represents every variable, function and type as an
26433 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
26434 Similarly, Python represents these symbols in @value{GDBN} with the
26435 @code{gdb.Symbol} object.
26437 The following symbol-related functions are available in the @code{gdb}
26440 @findex gdb.lookup_symbol
26441 @defun gdb.lookup_symbol (name @r{[}, block @r{[}, domain@r{]]})
26442 This function searches for a symbol by name. The search scope can be
26443 restricted to the parameters defined in the optional domain and block
26446 @var{name} is the name of the symbol. It must be a string. The
26447 optional @var{block} argument restricts the search to symbols visible
26448 in that @var{block}. The @var{block} argument must be a
26449 @code{gdb.Block} object. If omitted, the block for the current frame
26450 is used. The optional @var{domain} argument restricts
26451 the search to the domain type. The @var{domain} argument must be a
26452 domain constant defined in the @code{gdb} module and described later
26455 The result is a tuple of two elements.
26456 The first element is a @code{gdb.Symbol} object or @code{None} if the symbol
26458 If the symbol is found, the second element is @code{True} if the symbol
26459 is a field of a method's object (e.g., @code{this} in C@t{++}),
26460 otherwise it is @code{False}.
26461 If the symbol is not found, the second element is @code{False}.
26464 @findex gdb.lookup_global_symbol
26465 @defun gdb.lookup_global_symbol (name @r{[}, domain@r{]})
26466 This function searches for a global symbol by name.
26467 The search scope can be restricted to by the domain argument.
26469 @var{name} is the name of the symbol. It must be a string.
26470 The optional @var{domain} argument restricts the search to the domain type.
26471 The @var{domain} argument must be a domain constant defined in the @code{gdb}
26472 module and described later in this chapter.
26474 The result is a @code{gdb.Symbol} object or @code{None} if the symbol
26478 A @code{gdb.Symbol} object has the following attributes:
26480 @defvar Symbol.type
26481 The type of the symbol or @code{None} if no type is recorded.
26482 This attribute is represented as a @code{gdb.Type} object.
26483 @xref{Types In Python}. This attribute is not writable.
26486 @defvar Symbol.symtab
26487 The symbol table in which the symbol appears. This attribute is
26488 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
26489 Python}. This attribute is not writable.
26492 @defvar Symbol.line
26493 The line number in the source code at which the symbol was defined.
26494 This is an integer.
26497 @defvar Symbol.name
26498 The name of the symbol as a string. This attribute is not writable.
26501 @defvar Symbol.linkage_name
26502 The name of the symbol, as used by the linker (i.e., may be mangled).
26503 This attribute is not writable.
26506 @defvar Symbol.print_name
26507 The name of the symbol in a form suitable for output. This is either
26508 @code{name} or @code{linkage_name}, depending on whether the user
26509 asked @value{GDBN} to display demangled or mangled names.
26512 @defvar Symbol.addr_class
26513 The address class of the symbol. This classifies how to find the value
26514 of a symbol. Each address class is a constant defined in the
26515 @code{gdb} module and described later in this chapter.
26518 @defvar Symbol.needs_frame
26519 This is @code{True} if evaluating this symbol's value requires a frame
26520 (@pxref{Frames In Python}) and @code{False} otherwise. Typically,
26521 local variables will require a frame, but other symbols will not.
26524 @defvar Symbol.is_argument
26525 @code{True} if the symbol is an argument of a function.
26528 @defvar Symbol.is_constant
26529 @code{True} if the symbol is a constant.
26532 @defvar Symbol.is_function
26533 @code{True} if the symbol is a function or a method.
26536 @defvar Symbol.is_variable
26537 @code{True} if the symbol is a variable.
26540 A @code{gdb.Symbol} object has the following methods:
26542 @defun Symbol.is_valid ()
26543 Returns @code{True} if the @code{gdb.Symbol} object is valid,
26544 @code{False} if not. A @code{gdb.Symbol} object can become invalid if
26545 the symbol it refers to does not exist in @value{GDBN} any longer.
26546 All other @code{gdb.Symbol} methods will throw an exception if it is
26547 invalid at the time the method is called.
26550 @defun Symbol.value (@r{[}frame@r{]})
26551 Compute the value of the symbol, as a @code{gdb.Value}. For
26552 functions, this computes the address of the function, cast to the
26553 appropriate type. If the symbol requires a frame in order to compute
26554 its value, then @var{frame} must be given. If @var{frame} is not
26555 given, or if @var{frame} is invalid, then this method will throw an
26559 The available domain categories in @code{gdb.Symbol} are represented
26560 as constants in the @code{gdb} module:
26563 @findex SYMBOL_UNDEF_DOMAIN
26564 @findex gdb.SYMBOL_UNDEF_DOMAIN
26565 @item gdb.SYMBOL_UNDEF_DOMAIN
26566 This is used when a domain has not been discovered or none of the
26567 following domains apply. This usually indicates an error either
26568 in the symbol information or in @value{GDBN}'s handling of symbols.
26569 @findex SYMBOL_VAR_DOMAIN
26570 @findex gdb.SYMBOL_VAR_DOMAIN
26571 @item gdb.SYMBOL_VAR_DOMAIN
26572 This domain contains variables, function names, typedef names and enum
26574 @findex SYMBOL_STRUCT_DOMAIN
26575 @findex gdb.SYMBOL_STRUCT_DOMAIN
26576 @item gdb.SYMBOL_STRUCT_DOMAIN
26577 This domain holds struct, union and enum type names.
26578 @findex SYMBOL_LABEL_DOMAIN
26579 @findex gdb.SYMBOL_LABEL_DOMAIN
26580 @item gdb.SYMBOL_LABEL_DOMAIN
26581 This domain contains names of labels (for gotos).
26582 @findex SYMBOL_VARIABLES_DOMAIN
26583 @findex gdb.SYMBOL_VARIABLES_DOMAIN
26584 @item gdb.SYMBOL_VARIABLES_DOMAIN
26585 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
26586 contains everything minus functions and types.
26587 @findex SYMBOL_FUNCTIONS_DOMAIN
26588 @findex gdb.SYMBOL_FUNCTIONS_DOMAIN
26589 @item gdb.SYMBOL_FUNCTION_DOMAIN
26590 This domain contains all functions.
26591 @findex SYMBOL_TYPES_DOMAIN
26592 @findex gdb.SYMBOL_TYPES_DOMAIN
26593 @item gdb.SYMBOL_TYPES_DOMAIN
26594 This domain contains all types.
26597 The available address class categories in @code{gdb.Symbol} are represented
26598 as constants in the @code{gdb} module:
26601 @findex SYMBOL_LOC_UNDEF
26602 @findex gdb.SYMBOL_LOC_UNDEF
26603 @item gdb.SYMBOL_LOC_UNDEF
26604 If this is returned by address class, it indicates an error either in
26605 the symbol information or in @value{GDBN}'s handling of symbols.
26606 @findex SYMBOL_LOC_CONST
26607 @findex gdb.SYMBOL_LOC_CONST
26608 @item gdb.SYMBOL_LOC_CONST
26609 Value is constant int.
26610 @findex SYMBOL_LOC_STATIC
26611 @findex gdb.SYMBOL_LOC_STATIC
26612 @item gdb.SYMBOL_LOC_STATIC
26613 Value is at a fixed address.
26614 @findex SYMBOL_LOC_REGISTER
26615 @findex gdb.SYMBOL_LOC_REGISTER
26616 @item gdb.SYMBOL_LOC_REGISTER
26617 Value is in a register.
26618 @findex SYMBOL_LOC_ARG
26619 @findex gdb.SYMBOL_LOC_ARG
26620 @item gdb.SYMBOL_LOC_ARG
26621 Value is an argument. This value is at the offset stored within the
26622 symbol inside the frame's argument list.
26623 @findex SYMBOL_LOC_REF_ARG
26624 @findex gdb.SYMBOL_LOC_REF_ARG
26625 @item gdb.SYMBOL_LOC_REF_ARG
26626 Value address is stored in the frame's argument list. Just like
26627 @code{LOC_ARG} except that the value's address is stored at the
26628 offset, not the value itself.
26629 @findex SYMBOL_LOC_REGPARM_ADDR
26630 @findex gdb.SYMBOL_LOC_REGPARM_ADDR
26631 @item gdb.SYMBOL_LOC_REGPARM_ADDR
26632 Value is a specified register. Just like @code{LOC_REGISTER} except
26633 the register holds the address of the argument instead of the argument
26635 @findex SYMBOL_LOC_LOCAL
26636 @findex gdb.SYMBOL_LOC_LOCAL
26637 @item gdb.SYMBOL_LOC_LOCAL
26638 Value is a local variable.
26639 @findex SYMBOL_LOC_TYPEDEF
26640 @findex gdb.SYMBOL_LOC_TYPEDEF
26641 @item gdb.SYMBOL_LOC_TYPEDEF
26642 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
26644 @findex SYMBOL_LOC_BLOCK
26645 @findex gdb.SYMBOL_LOC_BLOCK
26646 @item gdb.SYMBOL_LOC_BLOCK
26648 @findex SYMBOL_LOC_CONST_BYTES
26649 @findex gdb.SYMBOL_LOC_CONST_BYTES
26650 @item gdb.SYMBOL_LOC_CONST_BYTES
26651 Value is a byte-sequence.
26652 @findex SYMBOL_LOC_UNRESOLVED
26653 @findex gdb.SYMBOL_LOC_UNRESOLVED
26654 @item gdb.SYMBOL_LOC_UNRESOLVED
26655 Value is at a fixed address, but the address of the variable has to be
26656 determined from the minimal symbol table whenever the variable is
26658 @findex SYMBOL_LOC_OPTIMIZED_OUT
26659 @findex gdb.SYMBOL_LOC_OPTIMIZED_OUT
26660 @item gdb.SYMBOL_LOC_OPTIMIZED_OUT
26661 The value does not actually exist in the program.
26662 @findex SYMBOL_LOC_COMPUTED
26663 @findex gdb.SYMBOL_LOC_COMPUTED
26664 @item gdb.SYMBOL_LOC_COMPUTED
26665 The value's address is a computed location.
26668 @node Symbol Tables In Python
26669 @subsubsection Symbol table representation in Python.
26671 @cindex symbol tables in python
26673 @tindex gdb.Symtab_and_line
26675 Access to symbol table data maintained by @value{GDBN} on the inferior
26676 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
26677 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
26678 from the @code{find_sal} method in @code{gdb.Frame} object.
26679 @xref{Frames In Python}.
26681 For more information on @value{GDBN}'s symbol table management, see
26682 @ref{Symbols, ,Examining the Symbol Table}, for more information.
26684 A @code{gdb.Symtab_and_line} object has the following attributes:
26686 @defvar Symtab_and_line.symtab
26687 The symbol table object (@code{gdb.Symtab}) for this frame.
26688 This attribute is not writable.
26691 @defvar Symtab_and_line.pc
26692 Indicates the start of the address range occupied by code for the
26693 current source line. This attribute is not writable.
26696 @defvar Symtab_and_line.last
26697 Indicates the end of the address range occupied by code for the current
26698 source line. This attribute is not writable.
26701 @defvar Symtab_and_line.line
26702 Indicates the current line number for this object. This
26703 attribute is not writable.
26706 A @code{gdb.Symtab_and_line} object has the following methods:
26708 @defun Symtab_and_line.is_valid ()
26709 Returns @code{True} if the @code{gdb.Symtab_and_line} object is valid,
26710 @code{False} if not. A @code{gdb.Symtab_and_line} object can become
26711 invalid if the Symbol table and line object it refers to does not
26712 exist in @value{GDBN} any longer. All other
26713 @code{gdb.Symtab_and_line} methods will throw an exception if it is
26714 invalid at the time the method is called.
26717 A @code{gdb.Symtab} object has the following attributes:
26719 @defvar Symtab.filename
26720 The symbol table's source filename. This attribute is not writable.
26723 @defvar Symtab.objfile
26724 The symbol table's backing object file. @xref{Objfiles In Python}.
26725 This attribute is not writable.
26728 A @code{gdb.Symtab} object has the following methods:
26730 @defun Symtab.is_valid ()
26731 Returns @code{True} if the @code{gdb.Symtab} object is valid,
26732 @code{False} if not. A @code{gdb.Symtab} object can become invalid if
26733 the symbol table it refers to does not exist in @value{GDBN} any
26734 longer. All other @code{gdb.Symtab} methods will throw an exception
26735 if it is invalid at the time the method is called.
26738 @defun Symtab.fullname ()
26739 Return the symbol table's source absolute file name.
26742 @defun Symtab.global_block ()
26743 Return the global block of the underlying symbol table.
26744 @xref{Blocks In Python}.
26747 @defun Symtab.static_block ()
26748 Return the static block of the underlying symbol table.
26749 @xref{Blocks In Python}.
26752 @node Breakpoints In Python
26753 @subsubsection Manipulating breakpoints using Python
26755 @cindex breakpoints in python
26756 @tindex gdb.Breakpoint
26758 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
26761 @defun Breakpoint.__init__ (spec @r{[}, type @r{[}, wp_class @r{[},internal@r{]]]})
26762 Create a new breakpoint. @var{spec} is a string naming the
26763 location of the breakpoint, or an expression that defines a
26764 watchpoint. The contents can be any location recognized by the
26765 @code{break} command, or in the case of a watchpoint, by the @code{watch}
26766 command. The optional @var{type} denotes the breakpoint to create
26767 from the types defined later in this chapter. This argument can be
26768 either: @code{gdb.BP_BREAKPOINT} or @code{gdb.BP_WATCHPOINT}. @var{type}
26769 defaults to @code{gdb.BP_BREAKPOINT}. The optional @var{internal} argument
26770 allows the breakpoint to become invisible to the user. The breakpoint
26771 will neither be reported when created, nor will it be listed in the
26772 output from @code{info breakpoints} (but will be listed with the
26773 @code{maint info breakpoints} command). The optional @var{wp_class}
26774 argument defines the class of watchpoint to create, if @var{type} is
26775 @code{gdb.BP_WATCHPOINT}. If a watchpoint class is not provided, it is
26776 assumed to be a @code{gdb.WP_WRITE} class.
26779 @defun Breakpoint.stop (self)
26780 The @code{gdb.Breakpoint} class can be sub-classed and, in
26781 particular, you may choose to implement the @code{stop} method.
26782 If this method is defined as a sub-class of @code{gdb.Breakpoint},
26783 it will be called when the inferior reaches any location of a
26784 breakpoint which instantiates that sub-class. If the method returns
26785 @code{True}, the inferior will be stopped at the location of the
26786 breakpoint, otherwise the inferior will continue.
26788 If there are multiple breakpoints at the same location with a
26789 @code{stop} method, each one will be called regardless of the
26790 return status of the previous. This ensures that all @code{stop}
26791 methods have a chance to execute at that location. In this scenario
26792 if one of the methods returns @code{True} but the others return
26793 @code{False}, the inferior will still be stopped.
26795 You should not alter the execution state of the inferior (i.e.@:, step,
26796 next, etc.), alter the current frame context (i.e.@:, change the current
26797 active frame), or alter, add or delete any breakpoint. As a general
26798 rule, you should not alter any data within @value{GDBN} or the inferior
26801 Example @code{stop} implementation:
26804 class MyBreakpoint (gdb.Breakpoint):
26806 inf_val = gdb.parse_and_eval("foo")
26813 The available watchpoint types represented by constants are defined in the
26818 @findex gdb.WP_READ
26820 Read only watchpoint.
26823 @findex gdb.WP_WRITE
26825 Write only watchpoint.
26828 @findex gdb.WP_ACCESS
26829 @item gdb.WP_ACCESS
26830 Read/Write watchpoint.
26833 @defun Breakpoint.is_valid ()
26834 Return @code{True} if this @code{Breakpoint} object is valid,
26835 @code{False} otherwise. A @code{Breakpoint} object can become invalid
26836 if the user deletes the breakpoint. In this case, the object still
26837 exists, but the underlying breakpoint does not. In the cases of
26838 watchpoint scope, the watchpoint remains valid even if execution of the
26839 inferior leaves the scope of that watchpoint.
26842 @defun Breakpoint.delete
26843 Permanently deletes the @value{GDBN} breakpoint. This also
26844 invalidates the Python @code{Breakpoint} object. Any further access
26845 to this object's attributes or methods will raise an error.
26848 @defvar Breakpoint.enabled
26849 This attribute is @code{True} if the breakpoint is enabled, and
26850 @code{False} otherwise. This attribute is writable.
26853 @defvar Breakpoint.silent
26854 This attribute is @code{True} if the breakpoint is silent, and
26855 @code{False} otherwise. This attribute is writable.
26857 Note that a breakpoint can also be silent if it has commands and the
26858 first command is @code{silent}. This is not reported by the
26859 @code{silent} attribute.
26862 @defvar Breakpoint.thread
26863 If the breakpoint is thread-specific, this attribute holds the thread
26864 id. If the breakpoint is not thread-specific, this attribute is
26865 @code{None}. This attribute is writable.
26868 @defvar Breakpoint.task
26869 If the breakpoint is Ada task-specific, this attribute holds the Ada task
26870 id. If the breakpoint is not task-specific (or the underlying
26871 language is not Ada), this attribute is @code{None}. This attribute
26875 @defvar Breakpoint.ignore_count
26876 This attribute holds the ignore count for the breakpoint, an integer.
26877 This attribute is writable.
26880 @defvar Breakpoint.number
26881 This attribute holds the breakpoint's number --- the identifier used by
26882 the user to manipulate the breakpoint. This attribute is not writable.
26885 @defvar Breakpoint.type
26886 This attribute holds the breakpoint's type --- the identifier used to
26887 determine the actual breakpoint type or use-case. This attribute is not
26891 @defvar Breakpoint.visible
26892 This attribute tells whether the breakpoint is visible to the user
26893 when set, or when the @samp{info breakpoints} command is run. This
26894 attribute is not writable.
26897 The available types are represented by constants defined in the @code{gdb}
26901 @findex BP_BREAKPOINT
26902 @findex gdb.BP_BREAKPOINT
26903 @item gdb.BP_BREAKPOINT
26904 Normal code breakpoint.
26906 @findex BP_WATCHPOINT
26907 @findex gdb.BP_WATCHPOINT
26908 @item gdb.BP_WATCHPOINT
26909 Watchpoint breakpoint.
26911 @findex BP_HARDWARE_WATCHPOINT
26912 @findex gdb.BP_HARDWARE_WATCHPOINT
26913 @item gdb.BP_HARDWARE_WATCHPOINT
26914 Hardware assisted watchpoint.
26916 @findex BP_READ_WATCHPOINT
26917 @findex gdb.BP_READ_WATCHPOINT
26918 @item gdb.BP_READ_WATCHPOINT
26919 Hardware assisted read watchpoint.
26921 @findex BP_ACCESS_WATCHPOINT
26922 @findex gdb.BP_ACCESS_WATCHPOINT
26923 @item gdb.BP_ACCESS_WATCHPOINT
26924 Hardware assisted access watchpoint.
26927 @defvar Breakpoint.hit_count
26928 This attribute holds the hit count for the breakpoint, an integer.
26929 This attribute is writable, but currently it can only be set to zero.
26932 @defvar Breakpoint.location
26933 This attribute holds the location of the breakpoint, as specified by
26934 the user. It is a string. If the breakpoint does not have a location
26935 (that is, it is a watchpoint) the attribute's value is @code{None}. This
26936 attribute is not writable.
26939 @defvar Breakpoint.expression
26940 This attribute holds a breakpoint expression, as specified by
26941 the user. It is a string. If the breakpoint does not have an
26942 expression (the breakpoint is not a watchpoint) the attribute's value
26943 is @code{None}. This attribute is not writable.
26946 @defvar Breakpoint.condition
26947 This attribute holds the condition of the breakpoint, as specified by
26948 the user. It is a string. If there is no condition, this attribute's
26949 value is @code{None}. This attribute is writable.
26952 @defvar Breakpoint.commands
26953 This attribute holds the commands attached to the breakpoint. If
26954 there are commands, this attribute's value is a string holding all the
26955 commands, separated by newlines. If there are no commands, this
26956 attribute is @code{None}. This attribute is not writable.
26959 @node Finish Breakpoints in Python
26960 @subsubsection Finish Breakpoints
26962 @cindex python finish breakpoints
26963 @tindex gdb.FinishBreakpoint
26965 A finish breakpoint is a temporary breakpoint set at the return address of
26966 a frame, based on the @code{finish} command. @code{gdb.FinishBreakpoint}
26967 extends @code{gdb.Breakpoint}. The underlying breakpoint will be disabled
26968 and deleted when the execution will run out of the breakpoint scope (i.e.@:
26969 @code{Breakpoint.stop} or @code{FinishBreakpoint.out_of_scope} triggered).
26970 Finish breakpoints are thread specific and must be create with the right
26973 @defun FinishBreakpoint.__init__ (@r{[}frame@r{]} @r{[}, internal@r{]})
26974 Create a finish breakpoint at the return address of the @code{gdb.Frame}
26975 object @var{frame}. If @var{frame} is not provided, this defaults to the
26976 newest frame. The optional @var{internal} argument allows the breakpoint to
26977 become invisible to the user. @xref{Breakpoints In Python}, for further
26978 details about this argument.
26981 @defun FinishBreakpoint.out_of_scope (self)
26982 In some circumstances (e.g.@: @code{longjmp}, C@t{++} exceptions, @value{GDBN}
26983 @code{return} command, @dots{}), a function may not properly terminate, and
26984 thus never hit the finish breakpoint. When @value{GDBN} notices such a
26985 situation, the @code{out_of_scope} callback will be triggered.
26987 You may want to sub-class @code{gdb.FinishBreakpoint} and override this
26991 class MyFinishBreakpoint (gdb.FinishBreakpoint)
26993 print "normal finish"
26996 def out_of_scope ():
26997 print "abnormal finish"
27001 @defvar FinishBreakpoint.return_value
27002 When @value{GDBN} is stopped at a finish breakpoint and the frame
27003 used to build the @code{gdb.FinishBreakpoint} object had debug symbols, this
27004 attribute will contain a @code{gdb.Value} object corresponding to the return
27005 value of the function. The value will be @code{None} if the function return
27006 type is @code{void} or if the return value was not computable. This attribute
27010 @node Lazy Strings In Python
27011 @subsubsection Python representation of lazy strings.
27013 @cindex lazy strings in python
27014 @tindex gdb.LazyString
27016 A @dfn{lazy string} is a string whose contents is not retrieved or
27017 encoded until it is needed.
27019 A @code{gdb.LazyString} is represented in @value{GDBN} as an
27020 @code{address} that points to a region of memory, an @code{encoding}
27021 that will be used to encode that region of memory, and a @code{length}
27022 to delimit the region of memory that represents the string. The
27023 difference between a @code{gdb.LazyString} and a string wrapped within
27024 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
27025 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
27026 retrieved and encoded during printing, while a @code{gdb.Value}
27027 wrapping a string is immediately retrieved and encoded on creation.
27029 A @code{gdb.LazyString} object has the following functions:
27031 @defun LazyString.value ()
27032 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
27033 will point to the string in memory, but will lose all the delayed
27034 retrieval, encoding and handling that @value{GDBN} applies to a
27035 @code{gdb.LazyString}.
27038 @defvar LazyString.address
27039 This attribute holds the address of the string. This attribute is not
27043 @defvar LazyString.length
27044 This attribute holds the length of the string in characters. If the
27045 length is -1, then the string will be fetched and encoded up to the
27046 first null of appropriate width. This attribute is not writable.
27049 @defvar LazyString.encoding
27050 This attribute holds the encoding that will be applied to the string
27051 when the string is printed by @value{GDBN}. If the encoding is not
27052 set, or contains an empty string, then @value{GDBN} will select the
27053 most appropriate encoding when the string is printed. This attribute
27057 @defvar LazyString.type
27058 This attribute holds the type that is represented by the lazy string's
27059 type. For a lazy string this will always be a pointer type. To
27060 resolve this to the lazy string's character type, use the type's
27061 @code{target} method. @xref{Types In Python}. This attribute is not
27065 @node Architectures In Python
27066 @subsubsection Python representation of architectures
27067 @cindex Python architectures
27069 @value{GDBN} uses architecture specific parameters and artifacts in a
27070 number of its various computations. An architecture is represented
27071 by an instance of the @code{gdb.Architecture} class.
27073 A @code{gdb.Architecture} class has the following methods:
27075 @defun Architecture.name ()
27076 Return the name (string value) of the architecture.
27079 @defun Architecture.disassemble (@var{start_pc} @r{[}, @var{end_pc} @r{[}, @var{count}@r{]]})
27080 Return a list of disassembled instructions starting from the memory
27081 address @var{start_pc}. The optional arguments @var{end_pc} and
27082 @var{count} determine the number of instructions in the returned list.
27083 If both the optional arguments @var{end_pc} and @var{count} are
27084 specified, then a list of at most @var{count} disassembled instructions
27085 whose start address falls in the closed memory address interval from
27086 @var{start_pc} to @var{end_pc} are returned. If @var{end_pc} is not
27087 specified, but @var{count} is specified, then @var{count} number of
27088 instructions starting from the address @var{start_pc} are returned. If
27089 @var{count} is not specified but @var{end_pc} is specified, then all
27090 instructions whose start address falls in the closed memory address
27091 interval from @var{start_pc} to @var{end_pc} are returned. If neither
27092 @var{end_pc} nor @var{count} are specified, then a single instruction at
27093 @var{start_pc} is returned. For all of these cases, each element of the
27094 returned list is a Python @code{dict} with the following string keys:
27099 The value corresponding to this key is a Python long integer capturing
27100 the memory address of the instruction.
27103 The value corresponding to this key is a string value which represents
27104 the instruction with assembly language mnemonics. The assembly
27105 language flavor used is the same as that specified by the current CLI
27106 variable @code{disassembly-flavor}. @xref{Machine Code}.
27109 The value corresponding to this key is the length (integer value) of the
27110 instruction in bytes.
27115 @node Python Auto-loading
27116 @subsection Python Auto-loading
27117 @cindex Python auto-loading
27119 When a new object file is read (for example, due to the @code{file}
27120 command, or because the inferior has loaded a shared library),
27121 @value{GDBN} will look for Python support scripts in several ways:
27122 @file{@var{objfile}-gdb.py} (@pxref{objfile-gdb.py file})
27123 and @code{.debug_gdb_scripts} section
27124 (@pxref{dotdebug_gdb_scripts section}).
27126 The auto-loading feature is useful for supplying application-specific
27127 debugging commands and scripts.
27129 Auto-loading can be enabled or disabled,
27130 and the list of auto-loaded scripts can be printed.
27133 @anchor{set auto-load python-scripts}
27134 @kindex set auto-load python-scripts
27135 @item set auto-load python-scripts [on|off]
27136 Enable or disable the auto-loading of Python scripts.
27138 @anchor{show auto-load python-scripts}
27139 @kindex show auto-load python-scripts
27140 @item show auto-load python-scripts
27141 Show whether auto-loading of Python scripts is enabled or disabled.
27143 @anchor{info auto-load python-scripts}
27144 @kindex info auto-load python-scripts
27145 @cindex print list of auto-loaded Python scripts
27146 @item info auto-load python-scripts [@var{regexp}]
27147 Print the list of all Python scripts that @value{GDBN} auto-loaded.
27149 Also printed is the list of Python scripts that were mentioned in
27150 the @code{.debug_gdb_scripts} section and were not found
27151 (@pxref{dotdebug_gdb_scripts section}).
27152 This is useful because their names are not printed when @value{GDBN}
27153 tries to load them and fails. There may be many of them, and printing
27154 an error message for each one is problematic.
27156 If @var{regexp} is supplied only Python scripts with matching names are printed.
27161 (gdb) info auto-load python-scripts
27163 Yes py-section-script.py
27164 full name: /tmp/py-section-script.py
27165 No my-foo-pretty-printers.py
27169 When reading an auto-loaded file, @value{GDBN} sets the
27170 @dfn{current objfile}. This is available via the @code{gdb.current_objfile}
27171 function (@pxref{Objfiles In Python}). This can be useful for
27172 registering objfile-specific pretty-printers and frame-filters.
27175 * objfile-gdb.py file:: The @file{@var{objfile}-gdb.py} file
27176 * dotdebug_gdb_scripts section:: The @code{.debug_gdb_scripts} section
27177 * Which flavor to choose?::
27180 @node objfile-gdb.py file
27181 @subsubsection The @file{@var{objfile}-gdb.py} file
27182 @cindex @file{@var{objfile}-gdb.py}
27184 When a new object file is read, @value{GDBN} looks for
27185 a file named @file{@var{objfile}-gdb.py} (we call it @var{script-name} below),
27186 where @var{objfile} is the object file's real name, formed by ensuring
27187 that the file name is absolute, following all symlinks, and resolving
27188 @code{.} and @code{..} components. If this file exists and is
27189 readable, @value{GDBN} will evaluate it as a Python script.
27191 If this file does not exist, then @value{GDBN} will look for
27192 @var{script-name} file in all of the directories as specified below.
27194 Note that loading of this script file also requires accordingly configured
27195 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27197 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
27198 scripts normally according to its @file{.exe} filename. But if no scripts are
27199 found @value{GDBN} also tries script filenames matching the object file without
27200 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
27201 is attempted on any platform. This makes the script filenames compatible
27202 between Unix and MS-Windows hosts.
27205 @anchor{set auto-load scripts-directory}
27206 @kindex set auto-load scripts-directory
27207 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
27208 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
27209 may be delimited by the host platform path separator in use
27210 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
27212 Each entry here needs to be covered also by the security setting
27213 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
27215 @anchor{with-auto-load-dir}
27216 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
27217 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
27218 configuration option @option{--with-auto-load-dir}.
27220 Any reference to @file{$debugdir} will get replaced by
27221 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
27222 reference to @file{$datadir} will get replaced by @var{data-directory} which is
27223 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
27224 @file{$datadir} must be placed as a directory component --- either alone or
27225 delimited by @file{/} or @file{\} directory separators, depending on the host
27228 The list of directories uses path separator (@samp{:} on GNU and Unix
27229 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
27230 to the @env{PATH} environment variable.
27232 @anchor{show auto-load scripts-directory}
27233 @kindex show auto-load scripts-directory
27234 @item show auto-load scripts-directory
27235 Show @value{GDBN} auto-loaded scripts location.
27238 @value{GDBN} does not track which files it has already auto-loaded this way.
27239 @value{GDBN} will load the associated script every time the corresponding
27240 @var{objfile} is opened.
27241 So your @file{-gdb.py} file should be careful to avoid errors if it
27242 is evaluated more than once.
27244 @node dotdebug_gdb_scripts section
27245 @subsubsection The @code{.debug_gdb_scripts} section
27246 @cindex @code{.debug_gdb_scripts} section
27248 For systems using file formats like ELF and COFF,
27249 when @value{GDBN} loads a new object file
27250 it will look for a special section named @samp{.debug_gdb_scripts}.
27251 If this section exists, its contents is a list of names of scripts to load.
27253 @value{GDBN} will look for each specified script file first in the
27254 current directory and then along the source search path
27255 (@pxref{Source Path, ,Specifying Source Directories}),
27256 except that @file{$cdir} is not searched, since the compilation
27257 directory is not relevant to scripts.
27259 Entries can be placed in section @code{.debug_gdb_scripts} with,
27260 for example, this GCC macro:
27263 /* Note: The "MS" section flags are to remove duplicates. */
27264 #define DEFINE_GDB_SCRIPT(script_name) \
27266 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
27268 .asciz \"" script_name "\"\n\
27274 Then one can reference the macro in a header or source file like this:
27277 DEFINE_GDB_SCRIPT ("my-app-scripts.py")
27280 The script name may include directories if desired.
27282 Note that loading of this script file also requires accordingly configured
27283 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27285 If the macro is put in a header, any application or library
27286 using this header will get a reference to the specified script.
27288 @node Which flavor to choose?
27289 @subsubsection Which flavor to choose?
27291 Given the multiple ways of auto-loading Python scripts, it might not always
27292 be clear which one to choose. This section provides some guidance.
27294 Benefits of the @file{-gdb.py} way:
27298 Can be used with file formats that don't support multiple sections.
27301 Ease of finding scripts for public libraries.
27303 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
27304 in the source search path.
27305 For publicly installed libraries, e.g., @file{libstdc++}, there typically
27306 isn't a source directory in which to find the script.
27309 Doesn't require source code additions.
27312 Benefits of the @code{.debug_gdb_scripts} way:
27316 Works with static linking.
27318 Scripts for libraries done the @file{-gdb.py} way require an objfile to
27319 trigger their loading. When an application is statically linked the only
27320 objfile available is the executable, and it is cumbersome to attach all the
27321 scripts from all the input libraries to the executable's @file{-gdb.py} script.
27324 Works with classes that are entirely inlined.
27326 Some classes can be entirely inlined, and thus there may not be an associated
27327 shared library to attach a @file{-gdb.py} script to.
27330 Scripts needn't be copied out of the source tree.
27332 In some circumstances, apps can be built out of large collections of internal
27333 libraries, and the build infrastructure necessary to install the
27334 @file{-gdb.py} scripts in a place where @value{GDBN} can find them is
27335 cumbersome. It may be easier to specify the scripts in the
27336 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
27337 top of the source tree to the source search path.
27340 @node Python modules
27341 @subsection Python modules
27342 @cindex python modules
27344 @value{GDBN} comes with several modules to assist writing Python code.
27347 * gdb.printing:: Building and registering pretty-printers.
27348 * gdb.types:: Utilities for working with types.
27349 * gdb.prompt:: Utilities for prompt value substitution.
27353 @subsubsection gdb.printing
27354 @cindex gdb.printing
27356 This module provides a collection of utilities for working with
27360 @item PrettyPrinter (@var{name}, @var{subprinters}=None)
27361 This class specifies the API that makes @samp{info pretty-printer},
27362 @samp{enable pretty-printer} and @samp{disable pretty-printer} work.
27363 Pretty-printers should generally inherit from this class.
27365 @item SubPrettyPrinter (@var{name})
27366 For printers that handle multiple types, this class specifies the
27367 corresponding API for the subprinters.
27369 @item RegexpCollectionPrettyPrinter (@var{name})
27370 Utility class for handling multiple printers, all recognized via
27371 regular expressions.
27372 @xref{Writing a Pretty-Printer}, for an example.
27374 @item FlagEnumerationPrinter (@var{name})
27375 A pretty-printer which handles printing of @code{enum} values. Unlike
27376 @value{GDBN}'s built-in @code{enum} printing, this printer attempts to
27377 work properly when there is some overlap between the enumeration
27378 constants. @var{name} is the name of the printer and also the name of
27379 the @code{enum} type to look up.
27381 @item register_pretty_printer (@var{obj}, @var{printer}, @var{replace}=False)
27382 Register @var{printer} with the pretty-printer list of @var{obj}.
27383 If @var{replace} is @code{True} then any existing copy of the printer
27384 is replaced. Otherwise a @code{RuntimeError} exception is raised
27385 if a printer with the same name already exists.
27389 @subsubsection gdb.types
27392 This module provides a collection of utilities for working with
27393 @code{gdb.Type} objects.
27396 @item get_basic_type (@var{type})
27397 Return @var{type} with const and volatile qualifiers stripped,
27398 and with typedefs and C@t{++} references converted to the underlying type.
27403 typedef const int const_int;
27405 const_int& foo_ref (foo);
27406 int main () @{ return 0; @}
27413 (gdb) python import gdb.types
27414 (gdb) python foo_ref = gdb.parse_and_eval("foo_ref")
27415 (gdb) python print gdb.types.get_basic_type(foo_ref.type)
27419 @item has_field (@var{type}, @var{field})
27420 Return @code{True} if @var{type}, assumed to be a type with fields
27421 (e.g., a structure or union), has field @var{field}.
27423 @item make_enum_dict (@var{enum_type})
27424 Return a Python @code{dictionary} type produced from @var{enum_type}.
27426 @item deep_items (@var{type})
27427 Returns a Python iterator similar to the standard
27428 @code{gdb.Type.iteritems} method, except that the iterator returned
27429 by @code{deep_items} will recursively traverse anonymous struct or
27430 union fields. For example:
27444 Then in @value{GDBN}:
27446 (@value{GDBP}) python import gdb.types
27447 (@value{GDBP}) python struct_a = gdb.lookup_type("struct A")
27448 (@value{GDBP}) python print struct_a.keys ()
27450 (@value{GDBP}) python print [k for k,v in gdb.types.deep_items(struct_a)]
27451 @{['a', 'b0', 'b1']@}
27454 @item get_type_recognizers ()
27455 Return a list of the enabled type recognizers for the current context.
27456 This is called by @value{GDBN} during the type-printing process
27457 (@pxref{Type Printing API}).
27459 @item apply_type_recognizers (recognizers, type_obj)
27460 Apply the type recognizers, @var{recognizers}, to the type object
27461 @var{type_obj}. If any recognizer returns a string, return that
27462 string. Otherwise, return @code{None}. This is called by
27463 @value{GDBN} during the type-printing process (@pxref{Type Printing
27466 @item register_type_printer (locus, printer)
27467 This is a convenience function to register a type printer.
27468 @var{printer} is the type printer to register. It must implement the
27469 type printer protocol. @var{locus} is either a @code{gdb.Objfile}, in
27470 which case the printer is registered with that objfile; a
27471 @code{gdb.Progspace}, in which case the printer is registered with
27472 that progspace; or @code{None}, in which case the printer is
27473 registered globally.
27476 This is a base class that implements the type printer protocol. Type
27477 printers are encouraged, but not required, to derive from this class.
27478 It defines a constructor:
27480 @defmethod TypePrinter __init__ (self, name)
27481 Initialize the type printer with the given name. The new printer
27482 starts in the enabled state.
27488 @subsubsection gdb.prompt
27491 This module provides a method for prompt value-substitution.
27494 @item substitute_prompt (@var{string})
27495 Return @var{string} with escape sequences substituted by values. Some
27496 escape sequences take arguments. You can specify arguments inside
27497 ``@{@}'' immediately following the escape sequence.
27499 The escape sequences you can pass to this function are:
27503 Substitute a backslash.
27505 Substitute an ESC character.
27507 Substitute the selected frame; an argument names a frame parameter.
27509 Substitute a newline.
27511 Substitute a parameter's value; the argument names the parameter.
27513 Substitute a carriage return.
27515 Substitute the selected thread; an argument names a thread parameter.
27517 Substitute the version of GDB.
27519 Substitute the current working directory.
27521 Begin a sequence of non-printing characters. These sequences are
27522 typically used with the ESC character, and are not counted in the string
27523 length. Example: ``\[\e[0;34m\](gdb)\[\e[0m\]'' will return a
27524 blue-colored ``(gdb)'' prompt where the length is five.
27526 End a sequence of non-printing characters.
27532 substitute_prompt (``frame: \f,
27533 print arguments: \p@{print frame-arguments@}'')
27536 @exdent will return the string:
27539 "frame: main, print arguments: scalars"
27544 @section Creating new spellings of existing commands
27545 @cindex aliases for commands
27547 It is often useful to define alternate spellings of existing commands.
27548 For example, if a new @value{GDBN} command defined in Python has
27549 a long name to type, it is handy to have an abbreviated version of it
27550 that involves less typing.
27552 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
27553 of the @samp{step} command even though it is otherwise an ambiguous
27554 abbreviation of other commands like @samp{set} and @samp{show}.
27556 Aliases are also used to provide shortened or more common versions
27557 of multi-word commands. For example, @value{GDBN} provides the
27558 @samp{tty} alias of the @samp{set inferior-tty} command.
27560 You can define a new alias with the @samp{alias} command.
27565 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
27569 @var{ALIAS} specifies the name of the new alias.
27570 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
27573 @var{COMMAND} specifies the name of an existing command
27574 that is being aliased.
27576 The @samp{-a} option specifies that the new alias is an abbreviation
27577 of the command. Abbreviations are not shown in command
27578 lists displayed by the @samp{help} command.
27580 The @samp{--} option specifies the end of options,
27581 and is useful when @var{ALIAS} begins with a dash.
27583 Here is a simple example showing how to make an abbreviation
27584 of a command so that there is less to type.
27585 Suppose you were tired of typing @samp{disas}, the current
27586 shortest unambiguous abbreviation of the @samp{disassemble} command
27587 and you wanted an even shorter version named @samp{di}.
27588 The following will accomplish this.
27591 (gdb) alias -a di = disas
27594 Note that aliases are different from user-defined commands.
27595 With a user-defined command, you also need to write documentation
27596 for it with the @samp{document} command.
27597 An alias automatically picks up the documentation of the existing command.
27599 Here is an example where we make @samp{elms} an abbreviation of
27600 @samp{elements} in the @samp{set print elements} command.
27601 This is to show that you can make an abbreviation of any part
27605 (gdb) alias -a set print elms = set print elements
27606 (gdb) alias -a show print elms = show print elements
27607 (gdb) set p elms 20
27609 Limit on string chars or array elements to print is 200.
27612 Note that if you are defining an alias of a @samp{set} command,
27613 and you want to have an alias for the corresponding @samp{show}
27614 command, then you need to define the latter separately.
27616 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
27617 @var{ALIAS}, just as they are normally.
27620 (gdb) alias -a set pr elms = set p ele
27623 Finally, here is an example showing the creation of a one word
27624 alias for a more complex command.
27625 This creates alias @samp{spe} of the command @samp{set print elements}.
27628 (gdb) alias spe = set print elements
27633 @chapter Command Interpreters
27634 @cindex command interpreters
27636 @value{GDBN} supports multiple command interpreters, and some command
27637 infrastructure to allow users or user interface writers to switch
27638 between interpreters or run commands in other interpreters.
27640 @value{GDBN} currently supports two command interpreters, the console
27641 interpreter (sometimes called the command-line interpreter or @sc{cli})
27642 and the machine interface interpreter (or @sc{gdb/mi}). This manual
27643 describes both of these interfaces in great detail.
27645 By default, @value{GDBN} will start with the console interpreter.
27646 However, the user may choose to start @value{GDBN} with another
27647 interpreter by specifying the @option{-i} or @option{--interpreter}
27648 startup options. Defined interpreters include:
27652 @cindex console interpreter
27653 The traditional console or command-line interpreter. This is the most often
27654 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
27655 @value{GDBN} will use this interpreter.
27658 @cindex mi interpreter
27659 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
27660 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
27661 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
27665 @cindex mi2 interpreter
27666 The current @sc{gdb/mi} interface.
27669 @cindex mi1 interpreter
27670 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
27674 @cindex invoke another interpreter
27675 The interpreter being used by @value{GDBN} may not be dynamically
27676 switched at runtime. Although possible, this could lead to a very
27677 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
27678 enters the command "interpreter-set console" in a console view,
27679 @value{GDBN} would switch to using the console interpreter, rendering
27680 the IDE inoperable!
27682 @kindex interpreter-exec
27683 Although you may only choose a single interpreter at startup, you may execute
27684 commands in any interpreter from the current interpreter using the appropriate
27685 command. If you are running the console interpreter, simply use the
27686 @code{interpreter-exec} command:
27689 interpreter-exec mi "-data-list-register-names"
27692 @sc{gdb/mi} has a similar command, although it is only available in versions of
27693 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
27696 @chapter @value{GDBN} Text User Interface
27698 @cindex Text User Interface
27701 * TUI Overview:: TUI overview
27702 * TUI Keys:: TUI key bindings
27703 * TUI Single Key Mode:: TUI single key mode
27704 * TUI Commands:: TUI-specific commands
27705 * TUI Configuration:: TUI configuration variables
27708 The @value{GDBN} Text User Interface (TUI) is a terminal
27709 interface which uses the @code{curses} library to show the source
27710 file, the assembly output, the program registers and @value{GDBN}
27711 commands in separate text windows. The TUI mode is supported only
27712 on platforms where a suitable version of the @code{curses} library
27715 The TUI mode is enabled by default when you invoke @value{GDBN} as
27716 @samp{@value{GDBP} -tui}.
27717 You can also switch in and out of TUI mode while @value{GDBN} runs by
27718 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
27719 @xref{TUI Keys, ,TUI Key Bindings}.
27722 @section TUI Overview
27724 In TUI mode, @value{GDBN} can display several text windows:
27728 This window is the @value{GDBN} command window with the @value{GDBN}
27729 prompt and the @value{GDBN} output. The @value{GDBN} input is still
27730 managed using readline.
27733 The source window shows the source file of the program. The current
27734 line and active breakpoints are displayed in this window.
27737 The assembly window shows the disassembly output of the program.
27740 This window shows the processor registers. Registers are highlighted
27741 when their values change.
27744 The source and assembly windows show the current program position
27745 by highlighting the current line and marking it with a @samp{>} marker.
27746 Breakpoints are indicated with two markers. The first marker
27747 indicates the breakpoint type:
27751 Breakpoint which was hit at least once.
27754 Breakpoint which was never hit.
27757 Hardware breakpoint which was hit at least once.
27760 Hardware breakpoint which was never hit.
27763 The second marker indicates whether the breakpoint is enabled or not:
27767 Breakpoint is enabled.
27770 Breakpoint is disabled.
27773 The source, assembly and register windows are updated when the current
27774 thread changes, when the frame changes, or when the program counter
27777 These windows are not all visible at the same time. The command
27778 window is always visible. The others can be arranged in several
27789 source and assembly,
27792 source and registers, or
27795 assembly and registers.
27798 A status line above the command window shows the following information:
27802 Indicates the current @value{GDBN} target.
27803 (@pxref{Targets, ,Specifying a Debugging Target}).
27806 Gives the current process or thread number.
27807 When no process is being debugged, this field is set to @code{No process}.
27810 Gives the current function name for the selected frame.
27811 The name is demangled if demangling is turned on (@pxref{Print Settings}).
27812 When there is no symbol corresponding to the current program counter,
27813 the string @code{??} is displayed.
27816 Indicates the current line number for the selected frame.
27817 When the current line number is not known, the string @code{??} is displayed.
27820 Indicates the current program counter address.
27824 @section TUI Key Bindings
27825 @cindex TUI key bindings
27827 The TUI installs several key bindings in the readline keymaps
27828 @ifset SYSTEM_READLINE
27829 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
27831 @ifclear SYSTEM_READLINE
27832 (@pxref{Command Line Editing}).
27834 The following key bindings are installed for both TUI mode and the
27835 @value{GDBN} standard mode.
27844 Enter or leave the TUI mode. When leaving the TUI mode,
27845 the curses window management stops and @value{GDBN} operates using
27846 its standard mode, writing on the terminal directly. When reentering
27847 the TUI mode, control is given back to the curses windows.
27848 The screen is then refreshed.
27852 Use a TUI layout with only one window. The layout will
27853 either be @samp{source} or @samp{assembly}. When the TUI mode
27854 is not active, it will switch to the TUI mode.
27856 Think of this key binding as the Emacs @kbd{C-x 1} binding.
27860 Use a TUI layout with at least two windows. When the current
27861 layout already has two windows, the next layout with two windows is used.
27862 When a new layout is chosen, one window will always be common to the
27863 previous layout and the new one.
27865 Think of it as the Emacs @kbd{C-x 2} binding.
27869 Change the active window. The TUI associates several key bindings
27870 (like scrolling and arrow keys) with the active window. This command
27871 gives the focus to the next TUI window.
27873 Think of it as the Emacs @kbd{C-x o} binding.
27877 Switch in and out of the TUI SingleKey mode that binds single
27878 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
27881 The following key bindings only work in the TUI mode:
27886 Scroll the active window one page up.
27890 Scroll the active window one page down.
27894 Scroll the active window one line up.
27898 Scroll the active window one line down.
27902 Scroll the active window one column left.
27906 Scroll the active window one column right.
27910 Refresh the screen.
27913 Because the arrow keys scroll the active window in the TUI mode, they
27914 are not available for their normal use by readline unless the command
27915 window has the focus. When another window is active, you must use
27916 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
27917 and @kbd{C-f} to control the command window.
27919 @node TUI Single Key Mode
27920 @section TUI Single Key Mode
27921 @cindex TUI single key mode
27923 The TUI also provides a @dfn{SingleKey} mode, which binds several
27924 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
27925 switch into this mode, where the following key bindings are used:
27928 @kindex c @r{(SingleKey TUI key)}
27932 @kindex d @r{(SingleKey TUI key)}
27936 @kindex f @r{(SingleKey TUI key)}
27940 @kindex n @r{(SingleKey TUI key)}
27944 @kindex q @r{(SingleKey TUI key)}
27946 exit the SingleKey mode.
27948 @kindex r @r{(SingleKey TUI key)}
27952 @kindex s @r{(SingleKey TUI key)}
27956 @kindex u @r{(SingleKey TUI key)}
27960 @kindex v @r{(SingleKey TUI key)}
27964 @kindex w @r{(SingleKey TUI key)}
27969 Other keys temporarily switch to the @value{GDBN} command prompt.
27970 The key that was pressed is inserted in the editing buffer so that
27971 it is possible to type most @value{GDBN} commands without interaction
27972 with the TUI SingleKey mode. Once the command is entered the TUI
27973 SingleKey mode is restored. The only way to permanently leave
27974 this mode is by typing @kbd{q} or @kbd{C-x s}.
27978 @section TUI-specific Commands
27979 @cindex TUI commands
27981 The TUI has specific commands to control the text windows.
27982 These commands are always available, even when @value{GDBN} is not in
27983 the TUI mode. When @value{GDBN} is in the standard mode, most
27984 of these commands will automatically switch to the TUI mode.
27986 Note that if @value{GDBN}'s @code{stdout} is not connected to a
27987 terminal, or @value{GDBN} has been started with the machine interface
27988 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
27989 these commands will fail with an error, because it would not be
27990 possible or desirable to enable curses window management.
27995 List and give the size of all displayed windows.
27999 Display the next layout.
28002 Display the previous layout.
28005 Display the source window only.
28008 Display the assembly window only.
28011 Display the source and assembly window.
28014 Display the register window together with the source or assembly window.
28018 Make the next window active for scrolling.
28021 Make the previous window active for scrolling.
28024 Make the source window active for scrolling.
28027 Make the assembly window active for scrolling.
28030 Make the register window active for scrolling.
28033 Make the command window active for scrolling.
28037 Refresh the screen. This is similar to typing @kbd{C-L}.
28039 @item tui reg float
28041 Show the floating point registers in the register window.
28043 @item tui reg general
28044 Show the general registers in the register window.
28047 Show the next register group. The list of register groups as well as
28048 their order is target specific. The predefined register groups are the
28049 following: @code{general}, @code{float}, @code{system}, @code{vector},
28050 @code{all}, @code{save}, @code{restore}.
28052 @item tui reg system
28053 Show the system registers in the register window.
28057 Update the source window and the current execution point.
28059 @item winheight @var{name} +@var{count}
28060 @itemx winheight @var{name} -@var{count}
28062 Change the height of the window @var{name} by @var{count}
28063 lines. Positive counts increase the height, while negative counts
28066 @item tabset @var{nchars}
28068 Set the width of tab stops to be @var{nchars} characters.
28071 @node TUI Configuration
28072 @section TUI Configuration Variables
28073 @cindex TUI configuration variables
28075 Several configuration variables control the appearance of TUI windows.
28078 @item set tui border-kind @var{kind}
28079 @kindex set tui border-kind
28080 Select the border appearance for the source, assembly and register windows.
28081 The possible values are the following:
28084 Use a space character to draw the border.
28087 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
28090 Use the Alternate Character Set to draw the border. The border is
28091 drawn using character line graphics if the terminal supports them.
28094 @item set tui border-mode @var{mode}
28095 @kindex set tui border-mode
28096 @itemx set tui active-border-mode @var{mode}
28097 @kindex set tui active-border-mode
28098 Select the display attributes for the borders of the inactive windows
28099 or the active window. The @var{mode} can be one of the following:
28102 Use normal attributes to display the border.
28108 Use reverse video mode.
28111 Use half bright mode.
28113 @item half-standout
28114 Use half bright and standout mode.
28117 Use extra bright or bold mode.
28119 @item bold-standout
28120 Use extra bright or bold and standout mode.
28125 @chapter Using @value{GDBN} under @sc{gnu} Emacs
28128 @cindex @sc{gnu} Emacs
28129 A special interface allows you to use @sc{gnu} Emacs to view (and
28130 edit) the source files for the program you are debugging with
28133 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
28134 executable file you want to debug as an argument. This command starts
28135 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
28136 created Emacs buffer.
28137 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
28139 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
28144 All ``terminal'' input and output goes through an Emacs buffer, called
28147 This applies both to @value{GDBN} commands and their output, and to the input
28148 and output done by the program you are debugging.
28150 This is useful because it means that you can copy the text of previous
28151 commands and input them again; you can even use parts of the output
28154 All the facilities of Emacs' Shell mode are available for interacting
28155 with your program. In particular, you can send signals the usual
28156 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
28160 @value{GDBN} displays source code through Emacs.
28162 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
28163 source file for that frame and puts an arrow (@samp{=>}) at the
28164 left margin of the current line. Emacs uses a separate buffer for
28165 source display, and splits the screen to show both your @value{GDBN} session
28168 Explicit @value{GDBN} @code{list} or search commands still produce output as
28169 usual, but you probably have no reason to use them from Emacs.
28172 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
28173 a graphical mode, enabled by default, which provides further buffers
28174 that can control the execution and describe the state of your program.
28175 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
28177 If you specify an absolute file name when prompted for the @kbd{M-x
28178 gdb} argument, then Emacs sets your current working directory to where
28179 your program resides. If you only specify the file name, then Emacs
28180 sets your current working directory to the directory associated
28181 with the previous buffer. In this case, @value{GDBN} may find your
28182 program by searching your environment's @code{PATH} variable, but on
28183 some operating systems it might not find the source. So, although the
28184 @value{GDBN} input and output session proceeds normally, the auxiliary
28185 buffer does not display the current source and line of execution.
28187 The initial working directory of @value{GDBN} is printed on the top
28188 line of the GUD buffer and this serves as a default for the commands
28189 that specify files for @value{GDBN} to operate on. @xref{Files,
28190 ,Commands to Specify Files}.
28192 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
28193 need to call @value{GDBN} by a different name (for example, if you
28194 keep several configurations around, with different names) you can
28195 customize the Emacs variable @code{gud-gdb-command-name} to run the
28198 In the GUD buffer, you can use these special Emacs commands in
28199 addition to the standard Shell mode commands:
28203 Describe the features of Emacs' GUD Mode.
28206 Execute to another source line, like the @value{GDBN} @code{step} command; also
28207 update the display window to show the current file and location.
28210 Execute to next source line in this function, skipping all function
28211 calls, like the @value{GDBN} @code{next} command. Then update the display window
28212 to show the current file and location.
28215 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
28216 display window accordingly.
28219 Execute until exit from the selected stack frame, like the @value{GDBN}
28220 @code{finish} command.
28223 Continue execution of your program, like the @value{GDBN} @code{continue}
28227 Go up the number of frames indicated by the numeric argument
28228 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
28229 like the @value{GDBN} @code{up} command.
28232 Go down the number of frames indicated by the numeric argument, like the
28233 @value{GDBN} @code{down} command.
28236 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
28237 tells @value{GDBN} to set a breakpoint on the source line point is on.
28239 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
28240 separate frame which shows a backtrace when the GUD buffer is current.
28241 Move point to any frame in the stack and type @key{RET} to make it
28242 become the current frame and display the associated source in the
28243 source buffer. Alternatively, click @kbd{Mouse-2} to make the
28244 selected frame become the current one. In graphical mode, the
28245 speedbar displays watch expressions.
28247 If you accidentally delete the source-display buffer, an easy way to get
28248 it back is to type the command @code{f} in the @value{GDBN} buffer, to
28249 request a frame display; when you run under Emacs, this recreates
28250 the source buffer if necessary to show you the context of the current
28253 The source files displayed in Emacs are in ordinary Emacs buffers
28254 which are visiting the source files in the usual way. You can edit
28255 the files with these buffers if you wish; but keep in mind that @value{GDBN}
28256 communicates with Emacs in terms of line numbers. If you add or
28257 delete lines from the text, the line numbers that @value{GDBN} knows cease
28258 to correspond properly with the code.
28260 A more detailed description of Emacs' interaction with @value{GDBN} is
28261 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
28265 @chapter The @sc{gdb/mi} Interface
28267 @unnumberedsec Function and Purpose
28269 @cindex @sc{gdb/mi}, its purpose
28270 @sc{gdb/mi} is a line based machine oriented text interface to
28271 @value{GDBN} and is activated by specifying using the
28272 @option{--interpreter} command line option (@pxref{Mode Options}). It
28273 is specifically intended to support the development of systems which
28274 use the debugger as just one small component of a larger system.
28276 This chapter is a specification of the @sc{gdb/mi} interface. It is written
28277 in the form of a reference manual.
28279 Note that @sc{gdb/mi} is still under construction, so some of the
28280 features described below are incomplete and subject to change
28281 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
28283 @unnumberedsec Notation and Terminology
28285 @cindex notational conventions, for @sc{gdb/mi}
28286 This chapter uses the following notation:
28290 @code{|} separates two alternatives.
28293 @code{[ @var{something} ]} indicates that @var{something} is optional:
28294 it may or may not be given.
28297 @code{( @var{group} )*} means that @var{group} inside the parentheses
28298 may repeat zero or more times.
28301 @code{( @var{group} )+} means that @var{group} inside the parentheses
28302 may repeat one or more times.
28305 @code{"@var{string}"} means a literal @var{string}.
28309 @heading Dependencies
28313 * GDB/MI General Design::
28314 * GDB/MI Command Syntax::
28315 * GDB/MI Compatibility with CLI::
28316 * GDB/MI Development and Front Ends::
28317 * GDB/MI Output Records::
28318 * GDB/MI Simple Examples::
28319 * GDB/MI Command Description Format::
28320 * GDB/MI Breakpoint Commands::
28321 * GDB/MI Catchpoint Commands::
28322 * GDB/MI Program Context::
28323 * GDB/MI Thread Commands::
28324 * GDB/MI Ada Tasking Commands::
28325 * GDB/MI Program Execution::
28326 * GDB/MI Stack Manipulation::
28327 * GDB/MI Variable Objects::
28328 * GDB/MI Data Manipulation::
28329 * GDB/MI Tracepoint Commands::
28330 * GDB/MI Symbol Query::
28331 * GDB/MI File Commands::
28333 * GDB/MI Kod Commands::
28334 * GDB/MI Memory Overlay Commands::
28335 * GDB/MI Signal Handling Commands::
28337 * GDB/MI Target Manipulation::
28338 * GDB/MI File Transfer Commands::
28339 * GDB/MI Miscellaneous Commands::
28342 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28343 @node GDB/MI General Design
28344 @section @sc{gdb/mi} General Design
28345 @cindex GDB/MI General Design
28347 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
28348 parts---commands sent to @value{GDBN}, responses to those commands
28349 and notifications. Each command results in exactly one response,
28350 indicating either successful completion of the command, or an error.
28351 For the commands that do not resume the target, the response contains the
28352 requested information. For the commands that resume the target, the
28353 response only indicates whether the target was successfully resumed.
28354 Notifications is the mechanism for reporting changes in the state of the
28355 target, or in @value{GDBN} state, that cannot conveniently be associated with
28356 a command and reported as part of that command response.
28358 The important examples of notifications are:
28362 Exec notifications. These are used to report changes in
28363 target state---when a target is resumed, or stopped. It would not
28364 be feasible to include this information in response of resuming
28365 commands, because one resume commands can result in multiple events in
28366 different threads. Also, quite some time may pass before any event
28367 happens in the target, while a frontend needs to know whether the resuming
28368 command itself was successfully executed.
28371 Console output, and status notifications. Console output
28372 notifications are used to report output of CLI commands, as well as
28373 diagnostics for other commands. Status notifications are used to
28374 report the progress of a long-running operation. Naturally, including
28375 this information in command response would mean no output is produced
28376 until the command is finished, which is undesirable.
28379 General notifications. Commands may have various side effects on
28380 the @value{GDBN} or target state beyond their official purpose. For example,
28381 a command may change the selected thread. Although such changes can
28382 be included in command response, using notification allows for more
28383 orthogonal frontend design.
28387 There's no guarantee that whenever an MI command reports an error,
28388 @value{GDBN} or the target are in any specific state, and especially,
28389 the state is not reverted to the state before the MI command was
28390 processed. Therefore, whenever an MI command results in an error,
28391 we recommend that the frontend refreshes all the information shown in
28392 the user interface.
28396 * Context management::
28397 * Asynchronous and non-stop modes::
28401 @node Context management
28402 @subsection Context management
28404 In most cases when @value{GDBN} accesses the target, this access is
28405 done in context of a specific thread and frame (@pxref{Frames}).
28406 Often, even when accessing global data, the target requires that a thread
28407 be specified. The CLI interface maintains the selected thread and frame,
28408 and supplies them to target on each command. This is convenient,
28409 because a command line user would not want to specify that information
28410 explicitly on each command, and because user interacts with
28411 @value{GDBN} via a single terminal, so no confusion is possible as
28412 to what thread and frame are the current ones.
28414 In the case of MI, the concept of selected thread and frame is less
28415 useful. First, a frontend can easily remember this information
28416 itself. Second, a graphical frontend can have more than one window,
28417 each one used for debugging a different thread, and the frontend might
28418 want to access additional threads for internal purposes. This
28419 increases the risk that by relying on implicitly selected thread, the
28420 frontend may be operating on a wrong one. Therefore, each MI command
28421 should explicitly specify which thread and frame to operate on. To
28422 make it possible, each MI command accepts the @samp{--thread} and
28423 @samp{--frame} options, the value to each is @value{GDBN} identifier
28424 for thread and frame to operate on.
28426 Usually, each top-level window in a frontend allows the user to select
28427 a thread and a frame, and remembers the user selection for further
28428 operations. However, in some cases @value{GDBN} may suggest that the
28429 current thread be changed. For example, when stopping on a breakpoint
28430 it is reasonable to switch to the thread where breakpoint is hit. For
28431 another example, if the user issues the CLI @samp{thread} command via
28432 the frontend, it is desirable to change the frontend's selected thread to the
28433 one specified by user. @value{GDBN} communicates the suggestion to
28434 change current thread using the @samp{=thread-selected} notification.
28435 No such notification is available for the selected frame at the moment.
28437 Note that historically, MI shares the selected thread with CLI, so
28438 frontends used the @code{-thread-select} to execute commands in the
28439 right context. However, getting this to work right is cumbersome. The
28440 simplest way is for frontend to emit @code{-thread-select} command
28441 before every command. This doubles the number of commands that need
28442 to be sent. The alternative approach is to suppress @code{-thread-select}
28443 if the selected thread in @value{GDBN} is supposed to be identical to the
28444 thread the frontend wants to operate on. However, getting this
28445 optimization right can be tricky. In particular, if the frontend
28446 sends several commands to @value{GDBN}, and one of the commands changes the
28447 selected thread, then the behaviour of subsequent commands will
28448 change. So, a frontend should either wait for response from such
28449 problematic commands, or explicitly add @code{-thread-select} for
28450 all subsequent commands. No frontend is known to do this exactly
28451 right, so it is suggested to just always pass the @samp{--thread} and
28452 @samp{--frame} options.
28454 @node Asynchronous and non-stop modes
28455 @subsection Asynchronous command execution and non-stop mode
28457 On some targets, @value{GDBN} is capable of processing MI commands
28458 even while the target is running. This is called @dfn{asynchronous
28459 command execution} (@pxref{Background Execution}). The frontend may
28460 specify a preferrence for asynchronous execution using the
28461 @code{-gdb-set target-async 1} command, which should be emitted before
28462 either running the executable or attaching to the target. After the
28463 frontend has started the executable or attached to the target, it can
28464 find if asynchronous execution is enabled using the
28465 @code{-list-target-features} command.
28467 Even if @value{GDBN} can accept a command while target is running,
28468 many commands that access the target do not work when the target is
28469 running. Therefore, asynchronous command execution is most useful
28470 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
28471 it is possible to examine the state of one thread, while other threads
28474 When a given thread is running, MI commands that try to access the
28475 target in the context of that thread may not work, or may work only on
28476 some targets. In particular, commands that try to operate on thread's
28477 stack will not work, on any target. Commands that read memory, or
28478 modify breakpoints, may work or not work, depending on the target. Note
28479 that even commands that operate on global state, such as @code{print},
28480 @code{set}, and breakpoint commands, still access the target in the
28481 context of a specific thread, so frontend should try to find a
28482 stopped thread and perform the operation on that thread (using the
28483 @samp{--thread} option).
28485 Which commands will work in the context of a running thread is
28486 highly target dependent. However, the two commands
28487 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
28488 to find the state of a thread, will always work.
28490 @node Thread groups
28491 @subsection Thread groups
28492 @value{GDBN} may be used to debug several processes at the same time.
28493 On some platfroms, @value{GDBN} may support debugging of several
28494 hardware systems, each one having several cores with several different
28495 processes running on each core. This section describes the MI
28496 mechanism to support such debugging scenarios.
28498 The key observation is that regardless of the structure of the
28499 target, MI can have a global list of threads, because most commands that
28500 accept the @samp{--thread} option do not need to know what process that
28501 thread belongs to. Therefore, it is not necessary to introduce
28502 neither additional @samp{--process} option, nor an notion of the
28503 current process in the MI interface. The only strictly new feature
28504 that is required is the ability to find how the threads are grouped
28507 To allow the user to discover such grouping, and to support arbitrary
28508 hierarchy of machines/cores/processes, MI introduces the concept of a
28509 @dfn{thread group}. Thread group is a collection of threads and other
28510 thread groups. A thread group always has a string identifier, a type,
28511 and may have additional attributes specific to the type. A new
28512 command, @code{-list-thread-groups}, returns the list of top-level
28513 thread groups, which correspond to processes that @value{GDBN} is
28514 debugging at the moment. By passing an identifier of a thread group
28515 to the @code{-list-thread-groups} command, it is possible to obtain
28516 the members of specific thread group.
28518 To allow the user to easily discover processes, and other objects, he
28519 wishes to debug, a concept of @dfn{available thread group} is
28520 introduced. Available thread group is an thread group that
28521 @value{GDBN} is not debugging, but that can be attached to, using the
28522 @code{-target-attach} command. The list of available top-level thread
28523 groups can be obtained using @samp{-list-thread-groups --available}.
28524 In general, the content of a thread group may be only retrieved only
28525 after attaching to that thread group.
28527 Thread groups are related to inferiors (@pxref{Inferiors and
28528 Programs}). Each inferior corresponds to a thread group of a special
28529 type @samp{process}, and some additional operations are permitted on
28530 such thread groups.
28532 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28533 @node GDB/MI Command Syntax
28534 @section @sc{gdb/mi} Command Syntax
28537 * GDB/MI Input Syntax::
28538 * GDB/MI Output Syntax::
28541 @node GDB/MI Input Syntax
28542 @subsection @sc{gdb/mi} Input Syntax
28544 @cindex input syntax for @sc{gdb/mi}
28545 @cindex @sc{gdb/mi}, input syntax
28547 @item @var{command} @expansion{}
28548 @code{@var{cli-command} | @var{mi-command}}
28550 @item @var{cli-command} @expansion{}
28551 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
28552 @var{cli-command} is any existing @value{GDBN} CLI command.
28554 @item @var{mi-command} @expansion{}
28555 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
28556 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
28558 @item @var{token} @expansion{}
28559 "any sequence of digits"
28561 @item @var{option} @expansion{}
28562 @code{"-" @var{parameter} [ " " @var{parameter} ]}
28564 @item @var{parameter} @expansion{}
28565 @code{@var{non-blank-sequence} | @var{c-string}}
28567 @item @var{operation} @expansion{}
28568 @emph{any of the operations described in this chapter}
28570 @item @var{non-blank-sequence} @expansion{}
28571 @emph{anything, provided it doesn't contain special characters such as
28572 "-", @var{nl}, """ and of course " "}
28574 @item @var{c-string} @expansion{}
28575 @code{""" @var{seven-bit-iso-c-string-content} """}
28577 @item @var{nl} @expansion{}
28586 The CLI commands are still handled by the @sc{mi} interpreter; their
28587 output is described below.
28590 The @code{@var{token}}, when present, is passed back when the command
28594 Some @sc{mi} commands accept optional arguments as part of the parameter
28595 list. Each option is identified by a leading @samp{-} (dash) and may be
28596 followed by an optional argument parameter. Options occur first in the
28597 parameter list and can be delimited from normal parameters using
28598 @samp{--} (this is useful when some parameters begin with a dash).
28605 We want easy access to the existing CLI syntax (for debugging).
28608 We want it to be easy to spot a @sc{mi} operation.
28611 @node GDB/MI Output Syntax
28612 @subsection @sc{gdb/mi} Output Syntax
28614 @cindex output syntax of @sc{gdb/mi}
28615 @cindex @sc{gdb/mi}, output syntax
28616 The output from @sc{gdb/mi} consists of zero or more out-of-band records
28617 followed, optionally, by a single result record. This result record
28618 is for the most recent command. The sequence of output records is
28619 terminated by @samp{(gdb)}.
28621 If an input command was prefixed with a @code{@var{token}} then the
28622 corresponding output for that command will also be prefixed by that same
28626 @item @var{output} @expansion{}
28627 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
28629 @item @var{result-record} @expansion{}
28630 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
28632 @item @var{out-of-band-record} @expansion{}
28633 @code{@var{async-record} | @var{stream-record}}
28635 @item @var{async-record} @expansion{}
28636 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
28638 @item @var{exec-async-output} @expansion{}
28639 @code{[ @var{token} ] "*" @var{async-output}}
28641 @item @var{status-async-output} @expansion{}
28642 @code{[ @var{token} ] "+" @var{async-output}}
28644 @item @var{notify-async-output} @expansion{}
28645 @code{[ @var{token} ] "=" @var{async-output}}
28647 @item @var{async-output} @expansion{}
28648 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
28650 @item @var{result-class} @expansion{}
28651 @code{"done" | "running" | "connected" | "error" | "exit"}
28653 @item @var{async-class} @expansion{}
28654 @code{"stopped" | @var{others}} (where @var{others} will be added
28655 depending on the needs---this is still in development).
28657 @item @var{result} @expansion{}
28658 @code{ @var{variable} "=" @var{value}}
28660 @item @var{variable} @expansion{}
28661 @code{ @var{string} }
28663 @item @var{value} @expansion{}
28664 @code{ @var{const} | @var{tuple} | @var{list} }
28666 @item @var{const} @expansion{}
28667 @code{@var{c-string}}
28669 @item @var{tuple} @expansion{}
28670 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
28672 @item @var{list} @expansion{}
28673 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
28674 @var{result} ( "," @var{result} )* "]" }
28676 @item @var{stream-record} @expansion{}
28677 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
28679 @item @var{console-stream-output} @expansion{}
28680 @code{"~" @var{c-string}}
28682 @item @var{target-stream-output} @expansion{}
28683 @code{"@@" @var{c-string}}
28685 @item @var{log-stream-output} @expansion{}
28686 @code{"&" @var{c-string}}
28688 @item @var{nl} @expansion{}
28691 @item @var{token} @expansion{}
28692 @emph{any sequence of digits}.
28700 All output sequences end in a single line containing a period.
28703 The @code{@var{token}} is from the corresponding request. Note that
28704 for all async output, while the token is allowed by the grammar and
28705 may be output by future versions of @value{GDBN} for select async
28706 output messages, it is generally omitted. Frontends should treat
28707 all async output as reporting general changes in the state of the
28708 target and there should be no need to associate async output to any
28712 @cindex status output in @sc{gdb/mi}
28713 @var{status-async-output} contains on-going status information about the
28714 progress of a slow operation. It can be discarded. All status output is
28715 prefixed by @samp{+}.
28718 @cindex async output in @sc{gdb/mi}
28719 @var{exec-async-output} contains asynchronous state change on the target
28720 (stopped, started, disappeared). All async output is prefixed by
28724 @cindex notify output in @sc{gdb/mi}
28725 @var{notify-async-output} contains supplementary information that the
28726 client should handle (e.g., a new breakpoint information). All notify
28727 output is prefixed by @samp{=}.
28730 @cindex console output in @sc{gdb/mi}
28731 @var{console-stream-output} is output that should be displayed as is in the
28732 console. It is the textual response to a CLI command. All the console
28733 output is prefixed by @samp{~}.
28736 @cindex target output in @sc{gdb/mi}
28737 @var{target-stream-output} is the output produced by the target program.
28738 All the target output is prefixed by @samp{@@}.
28741 @cindex log output in @sc{gdb/mi}
28742 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
28743 instance messages that should be displayed as part of an error log. All
28744 the log output is prefixed by @samp{&}.
28747 @cindex list output in @sc{gdb/mi}
28748 New @sc{gdb/mi} commands should only output @var{lists} containing
28754 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
28755 details about the various output records.
28757 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28758 @node GDB/MI Compatibility with CLI
28759 @section @sc{gdb/mi} Compatibility with CLI
28761 @cindex compatibility, @sc{gdb/mi} and CLI
28762 @cindex @sc{gdb/mi}, compatibility with CLI
28764 For the developers convenience CLI commands can be entered directly,
28765 but there may be some unexpected behaviour. For example, commands
28766 that query the user will behave as if the user replied yes, breakpoint
28767 command lists are not executed and some CLI commands, such as
28768 @code{if}, @code{when} and @code{define}, prompt for further input with
28769 @samp{>}, which is not valid MI output.
28771 This feature may be removed at some stage in the future and it is
28772 recommended that front ends use the @code{-interpreter-exec} command
28773 (@pxref{-interpreter-exec}).
28775 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28776 @node GDB/MI Development and Front Ends
28777 @section @sc{gdb/mi} Development and Front Ends
28778 @cindex @sc{gdb/mi} development
28780 The application which takes the MI output and presents the state of the
28781 program being debugged to the user is called a @dfn{front end}.
28783 Although @sc{gdb/mi} is still incomplete, it is currently being used
28784 by a variety of front ends to @value{GDBN}. This makes it difficult
28785 to introduce new functionality without breaking existing usage. This
28786 section tries to minimize the problems by describing how the protocol
28789 Some changes in MI need not break a carefully designed front end, and
28790 for these the MI version will remain unchanged. The following is a
28791 list of changes that may occur within one level, so front ends should
28792 parse MI output in a way that can handle them:
28796 New MI commands may be added.
28799 New fields may be added to the output of any MI command.
28802 The range of values for fields with specified values, e.g.,
28803 @code{in_scope} (@pxref{-var-update}) may be extended.
28805 @c The format of field's content e.g type prefix, may change so parse it
28806 @c at your own risk. Yes, in general?
28808 @c The order of fields may change? Shouldn't really matter but it might
28809 @c resolve inconsistencies.
28812 If the changes are likely to break front ends, the MI version level
28813 will be increased by one. This will allow the front end to parse the
28814 output according to the MI version. Apart from mi0, new versions of
28815 @value{GDBN} will not support old versions of MI and it will be the
28816 responsibility of the front end to work with the new one.
28818 @c Starting with mi3, add a new command -mi-version that prints the MI
28821 The best way to avoid unexpected changes in MI that might break your front
28822 end is to make your project known to @value{GDBN} developers and
28823 follow development on @email{gdb@@sourceware.org} and
28824 @email{gdb-patches@@sourceware.org}.
28825 @cindex mailing lists
28827 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28828 @node GDB/MI Output Records
28829 @section @sc{gdb/mi} Output Records
28832 * GDB/MI Result Records::
28833 * GDB/MI Stream Records::
28834 * GDB/MI Async Records::
28835 * GDB/MI Breakpoint Information::
28836 * GDB/MI Frame Information::
28837 * GDB/MI Thread Information::
28838 * GDB/MI Ada Exception Information::
28841 @node GDB/MI Result Records
28842 @subsection @sc{gdb/mi} Result Records
28844 @cindex result records in @sc{gdb/mi}
28845 @cindex @sc{gdb/mi}, result records
28846 In addition to a number of out-of-band notifications, the response to a
28847 @sc{gdb/mi} command includes one of the following result indications:
28851 @item "^done" [ "," @var{results} ]
28852 The synchronous operation was successful, @code{@var{results}} are the return
28857 This result record is equivalent to @samp{^done}. Historically, it
28858 was output instead of @samp{^done} if the command has resumed the
28859 target. This behaviour is maintained for backward compatibility, but
28860 all frontends should treat @samp{^done} and @samp{^running}
28861 identically and rely on the @samp{*running} output record to determine
28862 which threads are resumed.
28866 @value{GDBN} has connected to a remote target.
28868 @item "^error" "," @var{c-string}
28870 The operation failed. The @code{@var{c-string}} contains the corresponding
28875 @value{GDBN} has terminated.
28879 @node GDB/MI Stream Records
28880 @subsection @sc{gdb/mi} Stream Records
28882 @cindex @sc{gdb/mi}, stream records
28883 @cindex stream records in @sc{gdb/mi}
28884 @value{GDBN} internally maintains a number of output streams: the console, the
28885 target, and the log. The output intended for each of these streams is
28886 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
28888 Each stream record begins with a unique @dfn{prefix character} which
28889 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
28890 Syntax}). In addition to the prefix, each stream record contains a
28891 @code{@var{string-output}}. This is either raw text (with an implicit new
28892 line) or a quoted C string (which does not contain an implicit newline).
28895 @item "~" @var{string-output}
28896 The console output stream contains text that should be displayed in the
28897 CLI console window. It contains the textual responses to CLI commands.
28899 @item "@@" @var{string-output}
28900 The target output stream contains any textual output from the running
28901 target. This is only present when GDB's event loop is truly
28902 asynchronous, which is currently only the case for remote targets.
28904 @item "&" @var{string-output}
28905 The log stream contains debugging messages being produced by @value{GDBN}'s
28909 @node GDB/MI Async Records
28910 @subsection @sc{gdb/mi} Async Records
28912 @cindex async records in @sc{gdb/mi}
28913 @cindex @sc{gdb/mi}, async records
28914 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
28915 additional changes that have occurred. Those changes can either be a
28916 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
28917 target activity (e.g., target stopped).
28919 The following is the list of possible async records:
28923 @item *running,thread-id="@var{thread}"
28924 The target is now running. The @var{thread} field tells which
28925 specific thread is now running, and can be @samp{all} if all threads
28926 are running. The frontend should assume that no interaction with a
28927 running thread is possible after this notification is produced.
28928 The frontend should not assume that this notification is output
28929 only once for any command. @value{GDBN} may emit this notification
28930 several times, either for different threads, because it cannot resume
28931 all threads together, or even for a single thread, if the thread must
28932 be stepped though some code before letting it run freely.
28934 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
28935 The target has stopped. The @var{reason} field can have one of the
28939 @item breakpoint-hit
28940 A breakpoint was reached.
28941 @item watchpoint-trigger
28942 A watchpoint was triggered.
28943 @item read-watchpoint-trigger
28944 A read watchpoint was triggered.
28945 @item access-watchpoint-trigger
28946 An access watchpoint was triggered.
28947 @item function-finished
28948 An -exec-finish or similar CLI command was accomplished.
28949 @item location-reached
28950 An -exec-until or similar CLI command was accomplished.
28951 @item watchpoint-scope
28952 A watchpoint has gone out of scope.
28953 @item end-stepping-range
28954 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
28955 similar CLI command was accomplished.
28956 @item exited-signalled
28957 The inferior exited because of a signal.
28959 The inferior exited.
28960 @item exited-normally
28961 The inferior exited normally.
28962 @item signal-received
28963 A signal was received by the inferior.
28965 The inferior has stopped due to a library being loaded or unloaded.
28966 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
28967 set or when a @code{catch load} or @code{catch unload} catchpoint is
28968 in use (@pxref{Set Catchpoints}).
28970 The inferior has forked. This is reported when @code{catch fork}
28971 (@pxref{Set Catchpoints}) has been used.
28973 The inferior has vforked. This is reported in when @code{catch vfork}
28974 (@pxref{Set Catchpoints}) has been used.
28975 @item syscall-entry
28976 The inferior entered a system call. This is reported when @code{catch
28977 syscall} (@pxref{Set Catchpoints}) has been used.
28978 @item syscall-entry
28979 The inferior returned from a system call. This is reported when
28980 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
28982 The inferior called @code{exec}. This is reported when @code{catch exec}
28983 (@pxref{Set Catchpoints}) has been used.
28986 The @var{id} field identifies the thread that directly caused the stop
28987 -- for example by hitting a breakpoint. Depending on whether all-stop
28988 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
28989 stop all threads, or only the thread that directly triggered the stop.
28990 If all threads are stopped, the @var{stopped} field will have the
28991 value of @code{"all"}. Otherwise, the value of the @var{stopped}
28992 field will be a list of thread identifiers. Presently, this list will
28993 always include a single thread, but frontend should be prepared to see
28994 several threads in the list. The @var{core} field reports the
28995 processor core on which the stop event has happened. This field may be absent
28996 if such information is not available.
28998 @item =thread-group-added,id="@var{id}"
28999 @itemx =thread-group-removed,id="@var{id}"
29000 A thread group was either added or removed. The @var{id} field
29001 contains the @value{GDBN} identifier of the thread group. When a thread
29002 group is added, it generally might not be associated with a running
29003 process. When a thread group is removed, its id becomes invalid and
29004 cannot be used in any way.
29006 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
29007 A thread group became associated with a running program,
29008 either because the program was just started or the thread group
29009 was attached to a program. The @var{id} field contains the
29010 @value{GDBN} identifier of the thread group. The @var{pid} field
29011 contains process identifier, specific to the operating system.
29013 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
29014 A thread group is no longer associated with a running program,
29015 either because the program has exited, or because it was detached
29016 from. The @var{id} field contains the @value{GDBN} identifier of the
29017 thread group. @var{code} is the exit code of the inferior; it exists
29018 only when the inferior exited with some code.
29020 @item =thread-created,id="@var{id}",group-id="@var{gid}"
29021 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
29022 A thread either was created, or has exited. The @var{id} field
29023 contains the @value{GDBN} identifier of the thread. The @var{gid}
29024 field identifies the thread group this thread belongs to.
29026 @item =thread-selected,id="@var{id}"
29027 Informs that the selected thread was changed as result of the last
29028 command. This notification is not emitted as result of @code{-thread-select}
29029 command but is emitted whenever an MI command that is not documented
29030 to change the selected thread actually changes it. In particular,
29031 invoking, directly or indirectly (via user-defined command), the CLI
29032 @code{thread} command, will generate this notification.
29034 We suggest that in response to this notification, front ends
29035 highlight the selected thread and cause subsequent commands to apply to
29038 @item =library-loaded,...
29039 Reports that a new library file was loaded by the program. This
29040 notification has 4 fields---@var{id}, @var{target-name},
29041 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
29042 opaque identifier of the library. For remote debugging case,
29043 @var{target-name} and @var{host-name} fields give the name of the
29044 library file on the target, and on the host respectively. For native
29045 debugging, both those fields have the same value. The
29046 @var{symbols-loaded} field is emitted only for backward compatibility
29047 and should not be relied on to convey any useful information. The
29048 @var{thread-group} field, if present, specifies the id of the thread
29049 group in whose context the library was loaded. If the field is
29050 absent, it means the library was loaded in the context of all present
29053 @item =library-unloaded,...
29054 Reports that a library was unloaded by the program. This notification
29055 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
29056 the same meaning as for the @code{=library-loaded} notification.
29057 The @var{thread-group} field, if present, specifies the id of the
29058 thread group in whose context the library was unloaded. If the field is
29059 absent, it means the library was unloaded in the context of all present
29062 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
29063 @itemx =traceframe-changed,end
29064 Reports that the trace frame was changed and its new number is
29065 @var{tfnum}. The number of the tracepoint associated with this trace
29066 frame is @var{tpnum}.
29068 @item =tsv-created,name=@var{name},initial=@var{initial}
29069 Reports that the new trace state variable @var{name} is created with
29070 initial value @var{initial}.
29072 @item =tsv-deleted,name=@var{name}
29073 @itemx =tsv-deleted
29074 Reports that the trace state variable @var{name} is deleted or all
29075 trace state variables are deleted.
29077 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
29078 Reports that the trace state variable @var{name} is modified with
29079 the initial value @var{initial}. The current value @var{current} of
29080 trace state variable is optional and is reported if the current
29081 value of trace state variable is known.
29083 @item =breakpoint-created,bkpt=@{...@}
29084 @itemx =breakpoint-modified,bkpt=@{...@}
29085 @itemx =breakpoint-deleted,id=@var{number}
29086 Reports that a breakpoint was created, modified, or deleted,
29087 respectively. Only user-visible breakpoints are reported to the MI
29090 The @var{bkpt} argument is of the same form as returned by the various
29091 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
29092 @var{number} is the ordinal number of the breakpoint.
29094 Note that if a breakpoint is emitted in the result record of a
29095 command, then it will not also be emitted in an async record.
29097 @item =record-started,thread-group="@var{id}"
29098 @itemx =record-stopped,thread-group="@var{id}"
29099 Execution log recording was either started or stopped on an
29100 inferior. The @var{id} is the @value{GDBN} identifier of the thread
29101 group corresponding to the affected inferior.
29103 @item =cmd-param-changed,param=@var{param},value=@var{value}
29104 Reports that a parameter of the command @code{set @var{param}} is
29105 changed to @var{value}. In the multi-word @code{set} command,
29106 the @var{param} is the whole parameter list to @code{set} command.
29107 For example, In command @code{set check type on}, @var{param}
29108 is @code{check type} and @var{value} is @code{on}.
29110 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
29111 Reports that bytes from @var{addr} to @var{data} + @var{len} were
29112 written in an inferior. The @var{id} is the identifier of the
29113 thread group corresponding to the affected inferior. The optional
29114 @code{type="code"} part is reported if the memory written to holds
29118 @node GDB/MI Breakpoint Information
29119 @subsection @sc{gdb/mi} Breakpoint Information
29121 When @value{GDBN} reports information about a breakpoint, a
29122 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
29127 The breakpoint number. For a breakpoint that represents one location
29128 of a multi-location breakpoint, this will be a dotted pair, like
29132 The type of the breakpoint. For ordinary breakpoints this will be
29133 @samp{breakpoint}, but many values are possible.
29136 If the type of the breakpoint is @samp{catchpoint}, then this
29137 indicates the exact type of catchpoint.
29140 This is the breakpoint disposition---either @samp{del}, meaning that
29141 the breakpoint will be deleted at the next stop, or @samp{keep},
29142 meaning that the breakpoint will not be deleted.
29145 This indicates whether the breakpoint is enabled, in which case the
29146 value is @samp{y}, or disabled, in which case the value is @samp{n}.
29147 Note that this is not the same as the field @code{enable}.
29150 The address of the breakpoint. This may be a hexidecimal number,
29151 giving the address; or the string @samp{<PENDING>}, for a pending
29152 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
29153 multiple locations. This field will not be present if no address can
29154 be determined. For example, a watchpoint does not have an address.
29157 If known, the function in which the breakpoint appears.
29158 If not known, this field is not present.
29161 The name of the source file which contains this function, if known.
29162 If not known, this field is not present.
29165 The full file name of the source file which contains this function, if
29166 known. If not known, this field is not present.
29169 The line number at which this breakpoint appears, if known.
29170 If not known, this field is not present.
29173 If the source file is not known, this field may be provided. If
29174 provided, this holds the address of the breakpoint, possibly followed
29178 If this breakpoint is pending, this field is present and holds the
29179 text used to set the breakpoint, as entered by the user.
29182 Where this breakpoint's condition is evaluated, either @samp{host} or
29186 If this is a thread-specific breakpoint, then this identifies the
29187 thread in which the breakpoint can trigger.
29190 If this breakpoint is restricted to a particular Ada task, then this
29191 field will hold the task identifier.
29194 If the breakpoint is conditional, this is the condition expression.
29197 The ignore count of the breakpoint.
29200 The enable count of the breakpoint.
29202 @item traceframe-usage
29205 @item static-tracepoint-marker-string-id
29206 For a static tracepoint, the name of the static tracepoint marker.
29209 For a masked watchpoint, this is the mask.
29212 A tracepoint's pass count.
29214 @item original-location
29215 The location of the breakpoint as originally specified by the user.
29216 This field is optional.
29219 The number of times the breakpoint has been hit.
29222 This field is only given for tracepoints. This is either @samp{y},
29223 meaning that the tracepoint is installed, or @samp{n}, meaning that it
29227 Some extra data, the exact contents of which are type-dependent.
29231 For example, here is what the output of @code{-break-insert}
29232 (@pxref{GDB/MI Breakpoint Commands}) might be:
29235 -> -break-insert main
29236 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29237 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29238 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29243 @node GDB/MI Frame Information
29244 @subsection @sc{gdb/mi} Frame Information
29246 Response from many MI commands includes an information about stack
29247 frame. This information is a tuple that may have the following
29252 The level of the stack frame. The innermost frame has the level of
29253 zero. This field is always present.
29256 The name of the function corresponding to the frame. This field may
29257 be absent if @value{GDBN} is unable to determine the function name.
29260 The code address for the frame. This field is always present.
29263 The name of the source files that correspond to the frame's code
29264 address. This field may be absent.
29267 The source line corresponding to the frames' code address. This field
29271 The name of the binary file (either executable or shared library) the
29272 corresponds to the frame's code address. This field may be absent.
29276 @node GDB/MI Thread Information
29277 @subsection @sc{gdb/mi} Thread Information
29279 Whenever @value{GDBN} has to report an information about a thread, it
29280 uses a tuple with the following fields:
29284 The numeric id assigned to the thread by @value{GDBN}. This field is
29288 Target-specific string identifying the thread. This field is always present.
29291 Additional information about the thread provided by the target.
29292 It is supposed to be human-readable and not interpreted by the
29293 frontend. This field is optional.
29296 Either @samp{stopped} or @samp{running}, depending on whether the
29297 thread is presently running. This field is always present.
29300 The value of this field is an integer number of the processor core the
29301 thread was last seen on. This field is optional.
29304 @node GDB/MI Ada Exception Information
29305 @subsection @sc{gdb/mi} Ada Exception Information
29307 Whenever a @code{*stopped} record is emitted because the program
29308 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
29309 @value{GDBN} provides the name of the exception that was raised via
29310 the @code{exception-name} field.
29312 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29313 @node GDB/MI Simple Examples
29314 @section Simple Examples of @sc{gdb/mi} Interaction
29315 @cindex @sc{gdb/mi}, simple examples
29317 This subsection presents several simple examples of interaction using
29318 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
29319 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
29320 the output received from @sc{gdb/mi}.
29322 Note the line breaks shown in the examples are here only for
29323 readability, they don't appear in the real output.
29325 @subheading Setting a Breakpoint
29327 Setting a breakpoint generates synchronous output which contains detailed
29328 information of the breakpoint.
29331 -> -break-insert main
29332 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29333 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29334 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29339 @subheading Program Execution
29341 Program execution generates asynchronous records and MI gives the
29342 reason that execution stopped.
29348 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
29349 frame=@{addr="0x08048564",func="main",
29350 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
29351 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
29356 <- *stopped,reason="exited-normally"
29360 @subheading Quitting @value{GDBN}
29362 Quitting @value{GDBN} just prints the result class @samp{^exit}.
29370 Please note that @samp{^exit} is printed immediately, but it might
29371 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
29372 performs necessary cleanups, including killing programs being debugged
29373 or disconnecting from debug hardware, so the frontend should wait till
29374 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
29375 fails to exit in reasonable time.
29377 @subheading A Bad Command
29379 Here's what happens if you pass a non-existent command:
29383 <- ^error,msg="Undefined MI command: rubbish"
29388 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29389 @node GDB/MI Command Description Format
29390 @section @sc{gdb/mi} Command Description Format
29392 The remaining sections describe blocks of commands. Each block of
29393 commands is laid out in a fashion similar to this section.
29395 @subheading Motivation
29397 The motivation for this collection of commands.
29399 @subheading Introduction
29401 A brief introduction to this collection of commands as a whole.
29403 @subheading Commands
29405 For each command in the block, the following is described:
29407 @subsubheading Synopsis
29410 -command @var{args}@dots{}
29413 @subsubheading Result
29415 @subsubheading @value{GDBN} Command
29417 The corresponding @value{GDBN} CLI command(s), if any.
29419 @subsubheading Example
29421 Example(s) formatted for readability. Some of the described commands have
29422 not been implemented yet and these are labeled N.A.@: (not available).
29425 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29426 @node GDB/MI Breakpoint Commands
29427 @section @sc{gdb/mi} Breakpoint Commands
29429 @cindex breakpoint commands for @sc{gdb/mi}
29430 @cindex @sc{gdb/mi}, breakpoint commands
29431 This section documents @sc{gdb/mi} commands for manipulating
29434 @subheading The @code{-break-after} Command
29435 @findex -break-after
29437 @subsubheading Synopsis
29440 -break-after @var{number} @var{count}
29443 The breakpoint number @var{number} is not in effect until it has been
29444 hit @var{count} times. To see how this is reflected in the output of
29445 the @samp{-break-list} command, see the description of the
29446 @samp{-break-list} command below.
29448 @subsubheading @value{GDBN} Command
29450 The corresponding @value{GDBN} command is @samp{ignore}.
29452 @subsubheading Example
29457 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29458 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29459 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29467 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29468 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29469 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29470 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29471 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29472 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29473 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29474 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29475 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29476 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
29481 @subheading The @code{-break-catch} Command
29482 @findex -break-catch
29485 @subheading The @code{-break-commands} Command
29486 @findex -break-commands
29488 @subsubheading Synopsis
29491 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
29494 Specifies the CLI commands that should be executed when breakpoint
29495 @var{number} is hit. The parameters @var{command1} to @var{commandN}
29496 are the commands. If no command is specified, any previously-set
29497 commands are cleared. @xref{Break Commands}. Typical use of this
29498 functionality is tracing a program, that is, printing of values of
29499 some variables whenever breakpoint is hit and then continuing.
29501 @subsubheading @value{GDBN} Command
29503 The corresponding @value{GDBN} command is @samp{commands}.
29505 @subsubheading Example
29510 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29511 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29512 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29515 -break-commands 1 "print v" "continue"
29520 @subheading The @code{-break-condition} Command
29521 @findex -break-condition
29523 @subsubheading Synopsis
29526 -break-condition @var{number} @var{expr}
29529 Breakpoint @var{number} will stop the program only if the condition in
29530 @var{expr} is true. The condition becomes part of the
29531 @samp{-break-list} output (see the description of the @samp{-break-list}
29534 @subsubheading @value{GDBN} Command
29536 The corresponding @value{GDBN} command is @samp{condition}.
29538 @subsubheading Example
29542 -break-condition 1 1
29546 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29547 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29548 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29549 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29550 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29551 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29552 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29553 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29554 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29555 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
29559 @subheading The @code{-break-delete} Command
29560 @findex -break-delete
29562 @subsubheading Synopsis
29565 -break-delete ( @var{breakpoint} )+
29568 Delete the breakpoint(s) whose number(s) are specified in the argument
29569 list. This is obviously reflected in the breakpoint list.
29571 @subsubheading @value{GDBN} Command
29573 The corresponding @value{GDBN} command is @samp{delete}.
29575 @subsubheading Example
29583 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29584 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29585 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29586 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29587 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29588 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29589 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29594 @subheading The @code{-break-disable} Command
29595 @findex -break-disable
29597 @subsubheading Synopsis
29600 -break-disable ( @var{breakpoint} )+
29603 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
29604 break list is now set to @samp{n} for the named @var{breakpoint}(s).
29606 @subsubheading @value{GDBN} Command
29608 The corresponding @value{GDBN} command is @samp{disable}.
29610 @subsubheading Example
29618 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29619 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29620 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29621 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29622 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29623 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29624 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29625 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
29626 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29627 line="5",thread-groups=["i1"],times="0"@}]@}
29631 @subheading The @code{-break-enable} Command
29632 @findex -break-enable
29634 @subsubheading Synopsis
29637 -break-enable ( @var{breakpoint} )+
29640 Enable (previously disabled) @var{breakpoint}(s).
29642 @subsubheading @value{GDBN} Command
29644 The corresponding @value{GDBN} command is @samp{enable}.
29646 @subsubheading Example
29654 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29655 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29656 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29657 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29658 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29659 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29660 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29661 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29662 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29663 line="5",thread-groups=["i1"],times="0"@}]@}
29667 @subheading The @code{-break-info} Command
29668 @findex -break-info
29670 @subsubheading Synopsis
29673 -break-info @var{breakpoint}
29677 Get information about a single breakpoint.
29679 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
29680 Information}, for details on the format of each breakpoint in the
29683 @subsubheading @value{GDBN} Command
29685 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
29687 @subsubheading Example
29690 @subheading The @code{-break-insert} Command
29691 @findex -break-insert
29693 @subsubheading Synopsis
29696 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
29697 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29698 [ -p @var{thread-id} ] [ @var{location} ]
29702 If specified, @var{location}, can be one of:
29709 @item filename:linenum
29710 @item filename:function
29714 The possible optional parameters of this command are:
29718 Insert a temporary breakpoint.
29720 Insert a hardware breakpoint.
29722 If @var{location} cannot be parsed (for example if it
29723 refers to unknown files or functions), create a pending
29724 breakpoint. Without this flag, @value{GDBN} will report
29725 an error, and won't create a breakpoint, if @var{location}
29728 Create a disabled breakpoint.
29730 Create a tracepoint. @xref{Tracepoints}. When this parameter
29731 is used together with @samp{-h}, a fast tracepoint is created.
29732 @item -c @var{condition}
29733 Make the breakpoint conditional on @var{condition}.
29734 @item -i @var{ignore-count}
29735 Initialize the @var{ignore-count}.
29736 @item -p @var{thread-id}
29737 Restrict the breakpoint to the specified @var{thread-id}.
29740 @subsubheading Result
29742 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29743 resulting breakpoint.
29745 Note: this format is open to change.
29746 @c An out-of-band breakpoint instead of part of the result?
29748 @subsubheading @value{GDBN} Command
29750 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
29751 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
29753 @subsubheading Example
29758 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
29759 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
29762 -break-insert -t foo
29763 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
29764 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
29768 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29769 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29770 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29771 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29772 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29773 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29774 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29775 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29776 addr="0x0001072c", func="main",file="recursive2.c",
29777 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
29779 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
29780 addr="0x00010774",func="foo",file="recursive2.c",
29781 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29784 @c -break-insert -r foo.*
29785 @c ~int foo(int, int);
29786 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
29787 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29792 @subheading The @code{-break-list} Command
29793 @findex -break-list
29795 @subsubheading Synopsis
29801 Displays the list of inserted breakpoints, showing the following fields:
29805 number of the breakpoint
29807 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
29809 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
29812 is the breakpoint enabled or no: @samp{y} or @samp{n}
29814 memory location at which the breakpoint is set
29816 logical location of the breakpoint, expressed by function name, file
29818 @item Thread-groups
29819 list of thread groups to which this breakpoint applies
29821 number of times the breakpoint has been hit
29824 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
29825 @code{body} field is an empty list.
29827 @subsubheading @value{GDBN} Command
29829 The corresponding @value{GDBN} command is @samp{info break}.
29831 @subsubheading Example
29836 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29837 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29838 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29839 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29840 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29841 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29842 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29843 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29844 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
29846 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29847 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
29848 line="13",thread-groups=["i1"],times="0"@}]@}
29852 Here's an example of the result when there are no breakpoints:
29857 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29858 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29859 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29860 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29861 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29862 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29863 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29868 @subheading The @code{-break-passcount} Command
29869 @findex -break-passcount
29871 @subsubheading Synopsis
29874 -break-passcount @var{tracepoint-number} @var{passcount}
29877 Set the passcount for tracepoint @var{tracepoint-number} to
29878 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
29879 is not a tracepoint, error is emitted. This corresponds to CLI
29880 command @samp{passcount}.
29882 @subheading The @code{-break-watch} Command
29883 @findex -break-watch
29885 @subsubheading Synopsis
29888 -break-watch [ -a | -r ]
29891 Create a watchpoint. With the @samp{-a} option it will create an
29892 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
29893 read from or on a write to the memory location. With the @samp{-r}
29894 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
29895 trigger only when the memory location is accessed for reading. Without
29896 either of the options, the watchpoint created is a regular watchpoint,
29897 i.e., it will trigger when the memory location is accessed for writing.
29898 @xref{Set Watchpoints, , Setting Watchpoints}.
29900 Note that @samp{-break-list} will report a single list of watchpoints and
29901 breakpoints inserted.
29903 @subsubheading @value{GDBN} Command
29905 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
29908 @subsubheading Example
29910 Setting a watchpoint on a variable in the @code{main} function:
29915 ^done,wpt=@{number="2",exp="x"@}
29920 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
29921 value=@{old="-268439212",new="55"@},
29922 frame=@{func="main",args=[],file="recursive2.c",
29923 fullname="/home/foo/bar/recursive2.c",line="5"@}
29927 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
29928 the program execution twice: first for the variable changing value, then
29929 for the watchpoint going out of scope.
29934 ^done,wpt=@{number="5",exp="C"@}
29939 *stopped,reason="watchpoint-trigger",
29940 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
29941 frame=@{func="callee4",args=[],
29942 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29943 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
29948 *stopped,reason="watchpoint-scope",wpnum="5",
29949 frame=@{func="callee3",args=[@{name="strarg",
29950 value="0x11940 \"A string argument.\""@}],
29951 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29952 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
29956 Listing breakpoints and watchpoints, at different points in the program
29957 execution. Note that once the watchpoint goes out of scope, it is
29963 ^done,wpt=@{number="2",exp="C"@}
29966 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29967 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29968 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29969 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29970 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29971 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29972 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29973 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29974 addr="0x00010734",func="callee4",
29975 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29976 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
29978 bkpt=@{number="2",type="watchpoint",disp="keep",
29979 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
29984 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
29985 value=@{old="-276895068",new="3"@},
29986 frame=@{func="callee4",args=[],
29987 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29988 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
29991 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29992 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29993 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29994 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29995 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29996 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29997 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29998 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29999 addr="0x00010734",func="callee4",
30000 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30001 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
30003 bkpt=@{number="2",type="watchpoint",disp="keep",
30004 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
30008 ^done,reason="watchpoint-scope",wpnum="2",
30009 frame=@{func="callee3",args=[@{name="strarg",
30010 value="0x11940 \"A string argument.\""@}],
30011 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30012 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
30015 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
30016 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30017 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30018 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30019 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30020 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30021 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30022 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30023 addr="0x00010734",func="callee4",
30024 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30025 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30026 thread-groups=["i1"],times="1"@}]@}
30031 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30032 @node GDB/MI Catchpoint Commands
30033 @section @sc{gdb/mi} Catchpoint Commands
30035 This section documents @sc{gdb/mi} commands for manipulating
30038 @subheading The @code{-catch-load} Command
30039 @findex -catch-load
30041 @subsubheading Synopsis
30044 -catch-load [ -t ] [ -d ] @var{regexp}
30047 Add a catchpoint for library load events. If the @samp{-t} option is used,
30048 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
30049 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
30050 in a disabled state. The @samp{regexp} argument is a regular
30051 expression used to match the name of the loaded library.
30054 @subsubheading @value{GDBN} Command
30056 The corresponding @value{GDBN} command is @samp{catch load}.
30058 @subsubheading Example
30061 -catch-load -t foo.so
30062 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
30063 what="load of library matching foo.so",catch-type="load",times="0"@}
30068 @subheading The @code{-catch-unload} Command
30069 @findex -catch-unload
30071 @subsubheading Synopsis
30074 -catch-unload [ -t ] [ -d ] @var{regexp}
30077 Add a catchpoint for library unload events. If the @samp{-t} option is
30078 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
30079 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
30080 created in a disabled state. The @samp{regexp} argument is a regular
30081 expression used to match the name of the unloaded library.
30083 @subsubheading @value{GDBN} Command
30085 The corresponding @value{GDBN} command is @samp{catch unload}.
30087 @subsubheading Example
30090 -catch-unload -d bar.so
30091 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
30092 what="load of library matching bar.so",catch-type="unload",times="0"@}
30097 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30098 @node GDB/MI Program Context
30099 @section @sc{gdb/mi} Program Context
30101 @subheading The @code{-exec-arguments} Command
30102 @findex -exec-arguments
30105 @subsubheading Synopsis
30108 -exec-arguments @var{args}
30111 Set the inferior program arguments, to be used in the next
30114 @subsubheading @value{GDBN} Command
30116 The corresponding @value{GDBN} command is @samp{set args}.
30118 @subsubheading Example
30122 -exec-arguments -v word
30129 @subheading The @code{-exec-show-arguments} Command
30130 @findex -exec-show-arguments
30132 @subsubheading Synopsis
30135 -exec-show-arguments
30138 Print the arguments of the program.
30140 @subsubheading @value{GDBN} Command
30142 The corresponding @value{GDBN} command is @samp{show args}.
30144 @subsubheading Example
30149 @subheading The @code{-environment-cd} Command
30150 @findex -environment-cd
30152 @subsubheading Synopsis
30155 -environment-cd @var{pathdir}
30158 Set @value{GDBN}'s working directory.
30160 @subsubheading @value{GDBN} Command
30162 The corresponding @value{GDBN} command is @samp{cd}.
30164 @subsubheading Example
30168 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30174 @subheading The @code{-environment-directory} Command
30175 @findex -environment-directory
30177 @subsubheading Synopsis
30180 -environment-directory [ -r ] [ @var{pathdir} ]+
30183 Add directories @var{pathdir} to beginning of search path for source files.
30184 If the @samp{-r} option is used, the search path is reset to the default
30185 search path. If directories @var{pathdir} are supplied in addition to the
30186 @samp{-r} option, the search path is first reset and then addition
30188 Multiple directories may be specified, separated by blanks. Specifying
30189 multiple directories in a single command
30190 results in the directories added to the beginning of the
30191 search path in the same order they were presented in the command.
30192 If blanks are needed as
30193 part of a directory name, double-quotes should be used around
30194 the name. In the command output, the path will show up separated
30195 by the system directory-separator character. The directory-separator
30196 character must not be used
30197 in any directory name.
30198 If no directories are specified, the current search path is displayed.
30200 @subsubheading @value{GDBN} Command
30202 The corresponding @value{GDBN} command is @samp{dir}.
30204 @subsubheading Example
30208 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30209 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30211 -environment-directory ""
30212 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30214 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
30215 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
30217 -environment-directory -r
30218 ^done,source-path="$cdir:$cwd"
30223 @subheading The @code{-environment-path} Command
30224 @findex -environment-path
30226 @subsubheading Synopsis
30229 -environment-path [ -r ] [ @var{pathdir} ]+
30232 Add directories @var{pathdir} to beginning of search path for object files.
30233 If the @samp{-r} option is used, the search path is reset to the original
30234 search path that existed at gdb start-up. If directories @var{pathdir} are
30235 supplied in addition to the
30236 @samp{-r} option, the search path is first reset and then addition
30238 Multiple directories may be specified, separated by blanks. Specifying
30239 multiple directories in a single command
30240 results in the directories added to the beginning of the
30241 search path in the same order they were presented in the command.
30242 If blanks are needed as
30243 part of a directory name, double-quotes should be used around
30244 the name. In the command output, the path will show up separated
30245 by the system directory-separator character. The directory-separator
30246 character must not be used
30247 in any directory name.
30248 If no directories are specified, the current path is displayed.
30251 @subsubheading @value{GDBN} Command
30253 The corresponding @value{GDBN} command is @samp{path}.
30255 @subsubheading Example
30260 ^done,path="/usr/bin"
30262 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
30263 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
30265 -environment-path -r /usr/local/bin
30266 ^done,path="/usr/local/bin:/usr/bin"
30271 @subheading The @code{-environment-pwd} Command
30272 @findex -environment-pwd
30274 @subsubheading Synopsis
30280 Show the current working directory.
30282 @subsubheading @value{GDBN} Command
30284 The corresponding @value{GDBN} command is @samp{pwd}.
30286 @subsubheading Example
30291 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
30295 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30296 @node GDB/MI Thread Commands
30297 @section @sc{gdb/mi} Thread Commands
30300 @subheading The @code{-thread-info} Command
30301 @findex -thread-info
30303 @subsubheading Synopsis
30306 -thread-info [ @var{thread-id} ]
30309 Reports information about either a specific thread, if
30310 the @var{thread-id} parameter is present, or about all
30311 threads. When printing information about all threads,
30312 also reports the current thread.
30314 @subsubheading @value{GDBN} Command
30316 The @samp{info thread} command prints the same information
30319 @subsubheading Result
30321 The result is a list of threads. The following attributes are
30322 defined for a given thread:
30326 This field exists only for the current thread. It has the value @samp{*}.
30329 The identifier that @value{GDBN} uses to refer to the thread.
30332 The identifier that the target uses to refer to the thread.
30335 Extra information about the thread, in a target-specific format. This
30339 The name of the thread. If the user specified a name using the
30340 @code{thread name} command, then this name is given. Otherwise, if
30341 @value{GDBN} can extract the thread name from the target, then that
30342 name is given. If @value{GDBN} cannot find the thread name, then this
30346 The stack frame currently executing in the thread.
30349 The thread's state. The @samp{state} field may have the following
30354 The thread is stopped. Frame information is available for stopped
30358 The thread is running. There's no frame information for running
30364 If @value{GDBN} can find the CPU core on which this thread is running,
30365 then this field is the core identifier. This field is optional.
30369 @subsubheading Example
30374 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
30375 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
30376 args=[]@},state="running"@},
30377 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
30378 frame=@{level="0",addr="0x0804891f",func="foo",
30379 args=[@{name="i",value="10"@}],
30380 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
30381 state="running"@}],
30382 current-thread-id="1"
30386 @subheading The @code{-thread-list-ids} Command
30387 @findex -thread-list-ids
30389 @subsubheading Synopsis
30395 Produces a list of the currently known @value{GDBN} thread ids. At the
30396 end of the list it also prints the total number of such threads.
30398 This command is retained for historical reasons, the
30399 @code{-thread-info} command should be used instead.
30401 @subsubheading @value{GDBN} Command
30403 Part of @samp{info threads} supplies the same information.
30405 @subsubheading Example
30410 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30411 current-thread-id="1",number-of-threads="3"
30416 @subheading The @code{-thread-select} Command
30417 @findex -thread-select
30419 @subsubheading Synopsis
30422 -thread-select @var{threadnum}
30425 Make @var{threadnum} the current thread. It prints the number of the new
30426 current thread, and the topmost frame for that thread.
30428 This command is deprecated in favor of explicitly using the
30429 @samp{--thread} option to each command.
30431 @subsubheading @value{GDBN} Command
30433 The corresponding @value{GDBN} command is @samp{thread}.
30435 @subsubheading Example
30442 *stopped,reason="end-stepping-range",thread-id="2",line="187",
30443 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
30447 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30448 number-of-threads="3"
30451 ^done,new-thread-id="3",
30452 frame=@{level="0",func="vprintf",
30453 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
30454 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
30458 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30459 @node GDB/MI Ada Tasking Commands
30460 @section @sc{gdb/mi} Ada Tasking Commands
30462 @subheading The @code{-ada-task-info} Command
30463 @findex -ada-task-info
30465 @subsubheading Synopsis
30468 -ada-task-info [ @var{task-id} ]
30471 Reports information about either a specific Ada task, if the
30472 @var{task-id} parameter is present, or about all Ada tasks.
30474 @subsubheading @value{GDBN} Command
30476 The @samp{info tasks} command prints the same information
30477 about all Ada tasks (@pxref{Ada Tasks}).
30479 @subsubheading Result
30481 The result is a table of Ada tasks. The following columns are
30482 defined for each Ada task:
30486 This field exists only for the current thread. It has the value @samp{*}.
30489 The identifier that @value{GDBN} uses to refer to the Ada task.
30492 The identifier that the target uses to refer to the Ada task.
30495 The identifier of the thread corresponding to the Ada task.
30497 This field should always exist, as Ada tasks are always implemented
30498 on top of a thread. But if @value{GDBN} cannot find this corresponding
30499 thread for any reason, the field is omitted.
30502 This field exists only when the task was created by another task.
30503 In this case, it provides the ID of the parent task.
30506 The base priority of the task.
30509 The current state of the task. For a detailed description of the
30510 possible states, see @ref{Ada Tasks}.
30513 The name of the task.
30517 @subsubheading Example
30521 ^done,tasks=@{nr_rows="3",nr_cols="8",
30522 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
30523 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
30524 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
30525 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
30526 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
30527 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
30528 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
30529 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
30530 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
30531 state="Child Termination Wait",name="main_task"@}]@}
30535 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30536 @node GDB/MI Program Execution
30537 @section @sc{gdb/mi} Program Execution
30539 These are the asynchronous commands which generate the out-of-band
30540 record @samp{*stopped}. Currently @value{GDBN} only really executes
30541 asynchronously with remote targets and this interaction is mimicked in
30544 @subheading The @code{-exec-continue} Command
30545 @findex -exec-continue
30547 @subsubheading Synopsis
30550 -exec-continue [--reverse] [--all|--thread-group N]
30553 Resumes the execution of the inferior program, which will continue
30554 to execute until it reaches a debugger stop event. If the
30555 @samp{--reverse} option is specified, execution resumes in reverse until
30556 it reaches a stop event. Stop events may include
30559 breakpoints or watchpoints
30561 signals or exceptions
30563 the end of the process (or its beginning under @samp{--reverse})
30565 the end or beginning of a replay log if one is being used.
30567 In all-stop mode (@pxref{All-Stop
30568 Mode}), may resume only one thread, or all threads, depending on the
30569 value of the @samp{scheduler-locking} variable. If @samp{--all} is
30570 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
30571 ignored in all-stop mode. If the @samp{--thread-group} options is
30572 specified, then all threads in that thread group are resumed.
30574 @subsubheading @value{GDBN} Command
30576 The corresponding @value{GDBN} corresponding is @samp{continue}.
30578 @subsubheading Example
30585 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
30586 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
30592 @subheading The @code{-exec-finish} Command
30593 @findex -exec-finish
30595 @subsubheading Synopsis
30598 -exec-finish [--reverse]
30601 Resumes the execution of the inferior program until the current
30602 function is exited. Displays the results returned by the function.
30603 If the @samp{--reverse} option is specified, resumes the reverse
30604 execution of the inferior program until the point where current
30605 function was called.
30607 @subsubheading @value{GDBN} Command
30609 The corresponding @value{GDBN} command is @samp{finish}.
30611 @subsubheading Example
30613 Function returning @code{void}.
30620 *stopped,reason="function-finished",frame=@{func="main",args=[],
30621 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
30625 Function returning other than @code{void}. The name of the internal
30626 @value{GDBN} variable storing the result is printed, together with the
30633 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
30634 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
30635 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
30636 gdb-result-var="$1",return-value="0"
30641 @subheading The @code{-exec-interrupt} Command
30642 @findex -exec-interrupt
30644 @subsubheading Synopsis
30647 -exec-interrupt [--all|--thread-group N]
30650 Interrupts the background execution of the target. Note how the token
30651 associated with the stop message is the one for the execution command
30652 that has been interrupted. The token for the interrupt itself only
30653 appears in the @samp{^done} output. If the user is trying to
30654 interrupt a non-running program, an error message will be printed.
30656 Note that when asynchronous execution is enabled, this command is
30657 asynchronous just like other execution commands. That is, first the
30658 @samp{^done} response will be printed, and the target stop will be
30659 reported after that using the @samp{*stopped} notification.
30661 In non-stop mode, only the context thread is interrupted by default.
30662 All threads (in all inferiors) will be interrupted if the
30663 @samp{--all} option is specified. If the @samp{--thread-group}
30664 option is specified, all threads in that group will be interrupted.
30666 @subsubheading @value{GDBN} Command
30668 The corresponding @value{GDBN} command is @samp{interrupt}.
30670 @subsubheading Example
30681 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
30682 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
30683 fullname="/home/foo/bar/try.c",line="13"@}
30688 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
30692 @subheading The @code{-exec-jump} Command
30695 @subsubheading Synopsis
30698 -exec-jump @var{location}
30701 Resumes execution of the inferior program at the location specified by
30702 parameter. @xref{Specify Location}, for a description of the
30703 different forms of @var{location}.
30705 @subsubheading @value{GDBN} Command
30707 The corresponding @value{GDBN} command is @samp{jump}.
30709 @subsubheading Example
30712 -exec-jump foo.c:10
30713 *running,thread-id="all"
30718 @subheading The @code{-exec-next} Command
30721 @subsubheading Synopsis
30724 -exec-next [--reverse]
30727 Resumes execution of the inferior program, stopping when the beginning
30728 of the next source line is reached.
30730 If the @samp{--reverse} option is specified, resumes reverse execution
30731 of the inferior program, stopping at the beginning of the previous
30732 source line. If you issue this command on the first line of a
30733 function, it will take you back to the caller of that function, to the
30734 source line where the function was called.
30737 @subsubheading @value{GDBN} Command
30739 The corresponding @value{GDBN} command is @samp{next}.
30741 @subsubheading Example
30747 *stopped,reason="end-stepping-range",line="8",file="hello.c"
30752 @subheading The @code{-exec-next-instruction} Command
30753 @findex -exec-next-instruction
30755 @subsubheading Synopsis
30758 -exec-next-instruction [--reverse]
30761 Executes one machine instruction. If the instruction is a function
30762 call, continues until the function returns. If the program stops at an
30763 instruction in the middle of a source line, the address will be
30766 If the @samp{--reverse} option is specified, resumes reverse execution
30767 of the inferior program, stopping at the previous instruction. If the
30768 previously executed instruction was a return from another function,
30769 it will continue to execute in reverse until the call to that function
30770 (from the current stack frame) is reached.
30772 @subsubheading @value{GDBN} Command
30774 The corresponding @value{GDBN} command is @samp{nexti}.
30776 @subsubheading Example
30780 -exec-next-instruction
30784 *stopped,reason="end-stepping-range",
30785 addr="0x000100d4",line="5",file="hello.c"
30790 @subheading The @code{-exec-return} Command
30791 @findex -exec-return
30793 @subsubheading Synopsis
30799 Makes current function return immediately. Doesn't execute the inferior.
30800 Displays the new current frame.
30802 @subsubheading @value{GDBN} Command
30804 The corresponding @value{GDBN} command is @samp{return}.
30806 @subsubheading Example
30810 200-break-insert callee4
30811 200^done,bkpt=@{number="1",addr="0x00010734",
30812 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
30817 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30818 frame=@{func="callee4",args=[],
30819 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30820 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
30826 111^done,frame=@{level="0",func="callee3",
30827 args=[@{name="strarg",
30828 value="0x11940 \"A string argument.\""@}],
30829 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30830 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
30835 @subheading The @code{-exec-run} Command
30838 @subsubheading Synopsis
30841 -exec-run [--all | --thread-group N]
30844 Starts execution of the inferior from the beginning. The inferior
30845 executes until either a breakpoint is encountered or the program
30846 exits. In the latter case the output will include an exit code, if
30847 the program has exited exceptionally.
30849 When no option is specified, the current inferior is started. If the
30850 @samp{--thread-group} option is specified, it should refer to a thread
30851 group of type @samp{process}, and that thread group will be started.
30852 If the @samp{--all} option is specified, then all inferiors will be started.
30854 @subsubheading @value{GDBN} Command
30856 The corresponding @value{GDBN} command is @samp{run}.
30858 @subsubheading Examples
30863 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
30868 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30869 frame=@{func="main",args=[],file="recursive2.c",
30870 fullname="/home/foo/bar/recursive2.c",line="4"@}
30875 Program exited normally:
30883 *stopped,reason="exited-normally"
30888 Program exited exceptionally:
30896 *stopped,reason="exited",exit-code="01"
30900 Another way the program can terminate is if it receives a signal such as
30901 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
30905 *stopped,reason="exited-signalled",signal-name="SIGINT",
30906 signal-meaning="Interrupt"
30910 @c @subheading -exec-signal
30913 @subheading The @code{-exec-step} Command
30916 @subsubheading Synopsis
30919 -exec-step [--reverse]
30922 Resumes execution of the inferior program, stopping when the beginning
30923 of the next source line is reached, if the next source line is not a
30924 function call. If it is, stop at the first instruction of the called
30925 function. If the @samp{--reverse} option is specified, resumes reverse
30926 execution of the inferior program, stopping at the beginning of the
30927 previously executed source line.
30929 @subsubheading @value{GDBN} Command
30931 The corresponding @value{GDBN} command is @samp{step}.
30933 @subsubheading Example
30935 Stepping into a function:
30941 *stopped,reason="end-stepping-range",
30942 frame=@{func="foo",args=[@{name="a",value="10"@},
30943 @{name="b",value="0"@}],file="recursive2.c",
30944 fullname="/home/foo/bar/recursive2.c",line="11"@}
30954 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
30959 @subheading The @code{-exec-step-instruction} Command
30960 @findex -exec-step-instruction
30962 @subsubheading Synopsis
30965 -exec-step-instruction [--reverse]
30968 Resumes the inferior which executes one machine instruction. If the
30969 @samp{--reverse} option is specified, resumes reverse execution of the
30970 inferior program, stopping at the previously executed instruction.
30971 The output, once @value{GDBN} has stopped, will vary depending on
30972 whether we have stopped in the middle of a source line or not. In the
30973 former case, the address at which the program stopped will be printed
30976 @subsubheading @value{GDBN} Command
30978 The corresponding @value{GDBN} command is @samp{stepi}.
30980 @subsubheading Example
30984 -exec-step-instruction
30988 *stopped,reason="end-stepping-range",
30989 frame=@{func="foo",args=[],file="try.c",
30990 fullname="/home/foo/bar/try.c",line="10"@}
30992 -exec-step-instruction
30996 *stopped,reason="end-stepping-range",
30997 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
30998 fullname="/home/foo/bar/try.c",line="10"@}
31003 @subheading The @code{-exec-until} Command
31004 @findex -exec-until
31006 @subsubheading Synopsis
31009 -exec-until [ @var{location} ]
31012 Executes the inferior until the @var{location} specified in the
31013 argument is reached. If there is no argument, the inferior executes
31014 until a source line greater than the current one is reached. The
31015 reason for stopping in this case will be @samp{location-reached}.
31017 @subsubheading @value{GDBN} Command
31019 The corresponding @value{GDBN} command is @samp{until}.
31021 @subsubheading Example
31025 -exec-until recursive2.c:6
31029 *stopped,reason="location-reached",frame=@{func="main",args=[],
31030 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
31035 @subheading -file-clear
31036 Is this going away????
31039 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31040 @node GDB/MI Stack Manipulation
31041 @section @sc{gdb/mi} Stack Manipulation Commands
31043 @subheading The @code{-enable-frame-filters} Command
31044 @findex -enable-frame-filters
31047 -enable-frame-filters
31050 @value{GDBN} allows Python-based frame filters to affect the output of
31051 the MI commands relating to stack traces. As there is no way to
31052 implement this in a fully backward-compatible way, a front end must
31053 request that this functionality be enabled.
31055 Once enabled, this feature cannot be disabled.
31057 Note that if Python support has not been compiled into @value{GDBN},
31058 this command will still succeed (and do nothing).
31060 @subheading The @code{-stack-info-frame} Command
31061 @findex -stack-info-frame
31063 @subsubheading Synopsis
31069 Get info on the selected frame.
31071 @subsubheading @value{GDBN} Command
31073 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
31074 (without arguments).
31076 @subsubheading Example
31081 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
31082 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31083 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
31087 @subheading The @code{-stack-info-depth} Command
31088 @findex -stack-info-depth
31090 @subsubheading Synopsis
31093 -stack-info-depth [ @var{max-depth} ]
31096 Return the depth of the stack. If the integer argument @var{max-depth}
31097 is specified, do not count beyond @var{max-depth} frames.
31099 @subsubheading @value{GDBN} Command
31101 There's no equivalent @value{GDBN} command.
31103 @subsubheading Example
31105 For a stack with frame levels 0 through 11:
31112 -stack-info-depth 4
31115 -stack-info-depth 12
31118 -stack-info-depth 11
31121 -stack-info-depth 13
31126 @anchor{-stack-list-arguments}
31127 @subheading The @code{-stack-list-arguments} Command
31128 @findex -stack-list-arguments
31130 @subsubheading Synopsis
31133 -stack-list-arguments [ --no-frame-filters ] @var{print-values}
31134 [ @var{low-frame} @var{high-frame} ]
31137 Display a list of the arguments for the frames between @var{low-frame}
31138 and @var{high-frame} (inclusive). If @var{low-frame} and
31139 @var{high-frame} are not provided, list the arguments for the whole
31140 call stack. If the two arguments are equal, show the single frame
31141 at the corresponding level. It is an error if @var{low-frame} is
31142 larger than the actual number of frames. On the other hand,
31143 @var{high-frame} may be larger than the actual number of frames, in
31144 which case only existing frames will be returned.
31146 If @var{print-values} is 0 or @code{--no-values}, print only the names of
31147 the variables; if it is 1 or @code{--all-values}, print also their
31148 values; and if it is 2 or @code{--simple-values}, print the name,
31149 type and value for simple data types, and the name and type for arrays,
31150 structures and unions. If the option @code{--no-frame-filters} is
31151 supplied, then Python frame filters will not be executed.
31154 Use of this command to obtain arguments in a single frame is
31155 deprecated in favor of the @samp{-stack-list-variables} command.
31157 @subsubheading @value{GDBN} Command
31159 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
31160 @samp{gdb_get_args} command which partially overlaps with the
31161 functionality of @samp{-stack-list-arguments}.
31163 @subsubheading Example
31170 frame=@{level="0",addr="0x00010734",func="callee4",
31171 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31172 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
31173 frame=@{level="1",addr="0x0001076c",func="callee3",
31174 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31175 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
31176 frame=@{level="2",addr="0x0001078c",func="callee2",
31177 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31178 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
31179 frame=@{level="3",addr="0x000107b4",func="callee1",
31180 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31181 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
31182 frame=@{level="4",addr="0x000107e0",func="main",
31183 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31184 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
31186 -stack-list-arguments 0
31189 frame=@{level="0",args=[]@},
31190 frame=@{level="1",args=[name="strarg"]@},
31191 frame=@{level="2",args=[name="intarg",name="strarg"]@},
31192 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
31193 frame=@{level="4",args=[]@}]
31195 -stack-list-arguments 1
31198 frame=@{level="0",args=[]@},
31200 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31201 frame=@{level="2",args=[
31202 @{name="intarg",value="2"@},
31203 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31204 @{frame=@{level="3",args=[
31205 @{name="intarg",value="2"@},
31206 @{name="strarg",value="0x11940 \"A string argument.\""@},
31207 @{name="fltarg",value="3.5"@}]@},
31208 frame=@{level="4",args=[]@}]
31210 -stack-list-arguments 0 2 2
31211 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
31213 -stack-list-arguments 1 2 2
31214 ^done,stack-args=[frame=@{level="2",
31215 args=[@{name="intarg",value="2"@},
31216 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
31220 @c @subheading -stack-list-exception-handlers
31223 @anchor{-stack-list-frames}
31224 @subheading The @code{-stack-list-frames} Command
31225 @findex -stack-list-frames
31227 @subsubheading Synopsis
31230 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
31233 List the frames currently on the stack. For each frame it displays the
31238 The frame number, 0 being the topmost frame, i.e., the innermost function.
31240 The @code{$pc} value for that frame.
31244 File name of the source file where the function lives.
31245 @item @var{fullname}
31246 The full file name of the source file where the function lives.
31248 Line number corresponding to the @code{$pc}.
31250 The shared library where this function is defined. This is only given
31251 if the frame's function is not known.
31254 If invoked without arguments, this command prints a backtrace for the
31255 whole stack. If given two integer arguments, it shows the frames whose
31256 levels are between the two arguments (inclusive). If the two arguments
31257 are equal, it shows the single frame at the corresponding level. It is
31258 an error if @var{low-frame} is larger than the actual number of
31259 frames. On the other hand, @var{high-frame} may be larger than the
31260 actual number of frames, in which case only existing frames will be
31261 returned. If the option @code{--no-frame-filters} is supplied, then
31262 Python frame filters will not be executed.
31264 @subsubheading @value{GDBN} Command
31266 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
31268 @subsubheading Example
31270 Full stack backtrace:
31276 [frame=@{level="0",addr="0x0001076c",func="foo",
31277 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
31278 frame=@{level="1",addr="0x000107a4",func="foo",
31279 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31280 frame=@{level="2",addr="0x000107a4",func="foo",
31281 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31282 frame=@{level="3",addr="0x000107a4",func="foo",
31283 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31284 frame=@{level="4",addr="0x000107a4",func="foo",
31285 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31286 frame=@{level="5",addr="0x000107a4",func="foo",
31287 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31288 frame=@{level="6",addr="0x000107a4",func="foo",
31289 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31290 frame=@{level="7",addr="0x000107a4",func="foo",
31291 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31292 frame=@{level="8",addr="0x000107a4",func="foo",
31293 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31294 frame=@{level="9",addr="0x000107a4",func="foo",
31295 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31296 frame=@{level="10",addr="0x000107a4",func="foo",
31297 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31298 frame=@{level="11",addr="0x00010738",func="main",
31299 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
31303 Show frames between @var{low_frame} and @var{high_frame}:
31307 -stack-list-frames 3 5
31309 [frame=@{level="3",addr="0x000107a4",func="foo",
31310 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31311 frame=@{level="4",addr="0x000107a4",func="foo",
31312 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
31313 frame=@{level="5",addr="0x000107a4",func="foo",
31314 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
31318 Show a single frame:
31322 -stack-list-frames 3 3
31324 [frame=@{level="3",addr="0x000107a4",func="foo",
31325 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
31330 @subheading The @code{-stack-list-locals} Command
31331 @findex -stack-list-locals
31332 @anchor{-stack-list-locals}
31334 @subsubheading Synopsis
31337 -stack-list-locals [ --no-frame-filters ] @var{print-values}
31340 Display the local variable names for the selected frame. If
31341 @var{print-values} is 0 or @code{--no-values}, print only the names of
31342 the variables; if it is 1 or @code{--all-values}, print also their
31343 values; and if it is 2 or @code{--simple-values}, print the name,
31344 type and value for simple data types, and the name and type for arrays,
31345 structures and unions. In this last case, a frontend can immediately
31346 display the value of simple data types and create variable objects for
31347 other data types when the user wishes to explore their values in
31348 more detail. If the option @code{--no-frame-filters} is supplied, then
31349 Python frame filters will not be executed.
31351 This command is deprecated in favor of the
31352 @samp{-stack-list-variables} command.
31354 @subsubheading @value{GDBN} Command
31356 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
31358 @subsubheading Example
31362 -stack-list-locals 0
31363 ^done,locals=[name="A",name="B",name="C"]
31365 -stack-list-locals --all-values
31366 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
31367 @{name="C",value="@{1, 2, 3@}"@}]
31368 -stack-list-locals --simple-values
31369 ^done,locals=[@{name="A",type="int",value="1"@},
31370 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
31374 @anchor{-stack-list-variables}
31375 @subheading The @code{-stack-list-variables} Command
31376 @findex -stack-list-variables
31378 @subsubheading Synopsis
31381 -stack-list-variables [ --no-frame-filters ] @var{print-values}
31384 Display the names of local variables and function arguments for the selected frame. If
31385 @var{print-values} is 0 or @code{--no-values}, print only the names of
31386 the variables; if it is 1 or @code{--all-values}, print also their
31387 values; and if it is 2 or @code{--simple-values}, print the name,
31388 type and value for simple data types, and the name and type for arrays,
31389 structures and unions. If the option @code{--no-frame-filters} is
31390 supplied, then Python frame filters will not be executed.
31392 @subsubheading Example
31396 -stack-list-variables --thread 1 --frame 0 --all-values
31397 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
31402 @subheading The @code{-stack-select-frame} Command
31403 @findex -stack-select-frame
31405 @subsubheading Synopsis
31408 -stack-select-frame @var{framenum}
31411 Change the selected frame. Select a different frame @var{framenum} on
31414 This command in deprecated in favor of passing the @samp{--frame}
31415 option to every command.
31417 @subsubheading @value{GDBN} Command
31419 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
31420 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
31422 @subsubheading Example
31426 -stack-select-frame 2
31431 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31432 @node GDB/MI Variable Objects
31433 @section @sc{gdb/mi} Variable Objects
31437 @subheading Motivation for Variable Objects in @sc{gdb/mi}
31439 For the implementation of a variable debugger window (locals, watched
31440 expressions, etc.), we are proposing the adaptation of the existing code
31441 used by @code{Insight}.
31443 The two main reasons for that are:
31447 It has been proven in practice (it is already on its second generation).
31450 It will shorten development time (needless to say how important it is
31454 The original interface was designed to be used by Tcl code, so it was
31455 slightly changed so it could be used through @sc{gdb/mi}. This section
31456 describes the @sc{gdb/mi} operations that will be available and gives some
31457 hints about their use.
31459 @emph{Note}: In addition to the set of operations described here, we
31460 expect the @sc{gui} implementation of a variable window to require, at
31461 least, the following operations:
31464 @item @code{-gdb-show} @code{output-radix}
31465 @item @code{-stack-list-arguments}
31466 @item @code{-stack-list-locals}
31467 @item @code{-stack-select-frame}
31472 @subheading Introduction to Variable Objects
31474 @cindex variable objects in @sc{gdb/mi}
31476 Variable objects are "object-oriented" MI interface for examining and
31477 changing values of expressions. Unlike some other MI interfaces that
31478 work with expressions, variable objects are specifically designed for
31479 simple and efficient presentation in the frontend. A variable object
31480 is identified by string name. When a variable object is created, the
31481 frontend specifies the expression for that variable object. The
31482 expression can be a simple variable, or it can be an arbitrary complex
31483 expression, and can even involve CPU registers. After creating a
31484 variable object, the frontend can invoke other variable object
31485 operations---for example to obtain or change the value of a variable
31486 object, or to change display format.
31488 Variable objects have hierarchical tree structure. Any variable object
31489 that corresponds to a composite type, such as structure in C, has
31490 a number of child variable objects, for example corresponding to each
31491 element of a structure. A child variable object can itself have
31492 children, recursively. Recursion ends when we reach
31493 leaf variable objects, which always have built-in types. Child variable
31494 objects are created only by explicit request, so if a frontend
31495 is not interested in the children of a particular variable object, no
31496 child will be created.
31498 For a leaf variable object it is possible to obtain its value as a
31499 string, or set the value from a string. String value can be also
31500 obtained for a non-leaf variable object, but it's generally a string
31501 that only indicates the type of the object, and does not list its
31502 contents. Assignment to a non-leaf variable object is not allowed.
31504 A frontend does not need to read the values of all variable objects each time
31505 the program stops. Instead, MI provides an update command that lists all
31506 variable objects whose values has changed since the last update
31507 operation. This considerably reduces the amount of data that must
31508 be transferred to the frontend. As noted above, children variable
31509 objects are created on demand, and only leaf variable objects have a
31510 real value. As result, gdb will read target memory only for leaf
31511 variables that frontend has created.
31513 The automatic update is not always desirable. For example, a frontend
31514 might want to keep a value of some expression for future reference,
31515 and never update it. For another example, fetching memory is
31516 relatively slow for embedded targets, so a frontend might want
31517 to disable automatic update for the variables that are either not
31518 visible on the screen, or ``closed''. This is possible using so
31519 called ``frozen variable objects''. Such variable objects are never
31520 implicitly updated.
31522 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
31523 fixed variable object, the expression is parsed when the variable
31524 object is created, including associating identifiers to specific
31525 variables. The meaning of expression never changes. For a floating
31526 variable object the values of variables whose names appear in the
31527 expressions are re-evaluated every time in the context of the current
31528 frame. Consider this example:
31533 struct work_state state;
31540 If a fixed variable object for the @code{state} variable is created in
31541 this function, and we enter the recursive call, the variable
31542 object will report the value of @code{state} in the top-level
31543 @code{do_work} invocation. On the other hand, a floating variable
31544 object will report the value of @code{state} in the current frame.
31546 If an expression specified when creating a fixed variable object
31547 refers to a local variable, the variable object becomes bound to the
31548 thread and frame in which the variable object is created. When such
31549 variable object is updated, @value{GDBN} makes sure that the
31550 thread/frame combination the variable object is bound to still exists,
31551 and re-evaluates the variable object in context of that thread/frame.
31553 The following is the complete set of @sc{gdb/mi} operations defined to
31554 access this functionality:
31556 @multitable @columnfractions .4 .6
31557 @item @strong{Operation}
31558 @tab @strong{Description}
31560 @item @code{-enable-pretty-printing}
31561 @tab enable Python-based pretty-printing
31562 @item @code{-var-create}
31563 @tab create a variable object
31564 @item @code{-var-delete}
31565 @tab delete the variable object and/or its children
31566 @item @code{-var-set-format}
31567 @tab set the display format of this variable
31568 @item @code{-var-show-format}
31569 @tab show the display format of this variable
31570 @item @code{-var-info-num-children}
31571 @tab tells how many children this object has
31572 @item @code{-var-list-children}
31573 @tab return a list of the object's children
31574 @item @code{-var-info-type}
31575 @tab show the type of this variable object
31576 @item @code{-var-info-expression}
31577 @tab print parent-relative expression that this variable object represents
31578 @item @code{-var-info-path-expression}
31579 @tab print full expression that this variable object represents
31580 @item @code{-var-show-attributes}
31581 @tab is this variable editable? does it exist here?
31582 @item @code{-var-evaluate-expression}
31583 @tab get the value of this variable
31584 @item @code{-var-assign}
31585 @tab set the value of this variable
31586 @item @code{-var-update}
31587 @tab update the variable and its children
31588 @item @code{-var-set-frozen}
31589 @tab set frozeness attribute
31590 @item @code{-var-set-update-range}
31591 @tab set range of children to display on update
31594 In the next subsection we describe each operation in detail and suggest
31595 how it can be used.
31597 @subheading Description And Use of Operations on Variable Objects
31599 @subheading The @code{-enable-pretty-printing} Command
31600 @findex -enable-pretty-printing
31603 -enable-pretty-printing
31606 @value{GDBN} allows Python-based visualizers to affect the output of the
31607 MI variable object commands. However, because there was no way to
31608 implement this in a fully backward-compatible way, a front end must
31609 request that this functionality be enabled.
31611 Once enabled, this feature cannot be disabled.
31613 Note that if Python support has not been compiled into @value{GDBN},
31614 this command will still succeed (and do nothing).
31616 This feature is currently (as of @value{GDBN} 7.0) experimental, and
31617 may work differently in future versions of @value{GDBN}.
31619 @subheading The @code{-var-create} Command
31620 @findex -var-create
31622 @subsubheading Synopsis
31625 -var-create @{@var{name} | "-"@}
31626 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
31629 This operation creates a variable object, which allows the monitoring of
31630 a variable, the result of an expression, a memory cell or a CPU
31633 The @var{name} parameter is the string by which the object can be
31634 referenced. It must be unique. If @samp{-} is specified, the varobj
31635 system will generate a string ``varNNNNNN'' automatically. It will be
31636 unique provided that one does not specify @var{name} of that format.
31637 The command fails if a duplicate name is found.
31639 The frame under which the expression should be evaluated can be
31640 specified by @var{frame-addr}. A @samp{*} indicates that the current
31641 frame should be used. A @samp{@@} indicates that a floating variable
31642 object must be created.
31644 @var{expression} is any expression valid on the current language set (must not
31645 begin with a @samp{*}), or one of the following:
31649 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
31652 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
31655 @samp{$@var{regname}} --- a CPU register name
31658 @cindex dynamic varobj
31659 A varobj's contents may be provided by a Python-based pretty-printer. In this
31660 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
31661 have slightly different semantics in some cases. If the
31662 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
31663 will never create a dynamic varobj. This ensures backward
31664 compatibility for existing clients.
31666 @subsubheading Result
31668 This operation returns attributes of the newly-created varobj. These
31673 The name of the varobj.
31676 The number of children of the varobj. This number is not necessarily
31677 reliable for a dynamic varobj. Instead, you must examine the
31678 @samp{has_more} attribute.
31681 The varobj's scalar value. For a varobj whose type is some sort of
31682 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
31683 will not be interesting.
31686 The varobj's type. This is a string representation of the type, as
31687 would be printed by the @value{GDBN} CLI. If @samp{print object}
31688 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31689 @emph{actual} (derived) type of the object is shown rather than the
31690 @emph{declared} one.
31693 If a variable object is bound to a specific thread, then this is the
31694 thread's identifier.
31697 For a dynamic varobj, this indicates whether there appear to be any
31698 children available. For a non-dynamic varobj, this will be 0.
31701 This attribute will be present and have the value @samp{1} if the
31702 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31703 then this attribute will not be present.
31706 A dynamic varobj can supply a display hint to the front end. The
31707 value comes directly from the Python pretty-printer object's
31708 @code{display_hint} method. @xref{Pretty Printing API}.
31711 Typical output will look like this:
31714 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
31715 has_more="@var{has_more}"
31719 @subheading The @code{-var-delete} Command
31720 @findex -var-delete
31722 @subsubheading Synopsis
31725 -var-delete [ -c ] @var{name}
31728 Deletes a previously created variable object and all of its children.
31729 With the @samp{-c} option, just deletes the children.
31731 Returns an error if the object @var{name} is not found.
31734 @subheading The @code{-var-set-format} Command
31735 @findex -var-set-format
31737 @subsubheading Synopsis
31740 -var-set-format @var{name} @var{format-spec}
31743 Sets the output format for the value of the object @var{name} to be
31746 @anchor{-var-set-format}
31747 The syntax for the @var{format-spec} is as follows:
31750 @var{format-spec} @expansion{}
31751 @{binary | decimal | hexadecimal | octal | natural@}
31754 The natural format is the default format choosen automatically
31755 based on the variable type (like decimal for an @code{int}, hex
31756 for pointers, etc.).
31758 For a variable with children, the format is set only on the
31759 variable itself, and the children are not affected.
31761 @subheading The @code{-var-show-format} Command
31762 @findex -var-show-format
31764 @subsubheading Synopsis
31767 -var-show-format @var{name}
31770 Returns the format used to display the value of the object @var{name}.
31773 @var{format} @expansion{}
31778 @subheading The @code{-var-info-num-children} Command
31779 @findex -var-info-num-children
31781 @subsubheading Synopsis
31784 -var-info-num-children @var{name}
31787 Returns the number of children of a variable object @var{name}:
31793 Note that this number is not completely reliable for a dynamic varobj.
31794 It will return the current number of children, but more children may
31798 @subheading The @code{-var-list-children} Command
31799 @findex -var-list-children
31801 @subsubheading Synopsis
31804 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
31806 @anchor{-var-list-children}
31808 Return a list of the children of the specified variable object and
31809 create variable objects for them, if they do not already exist. With
31810 a single argument or if @var{print-values} has a value of 0 or
31811 @code{--no-values}, print only the names of the variables; if
31812 @var{print-values} is 1 or @code{--all-values}, also print their
31813 values; and if it is 2 or @code{--simple-values} print the name and
31814 value for simple data types and just the name for arrays, structures
31817 @var{from} and @var{to}, if specified, indicate the range of children
31818 to report. If @var{from} or @var{to} is less than zero, the range is
31819 reset and all children will be reported. Otherwise, children starting
31820 at @var{from} (zero-based) and up to and excluding @var{to} will be
31823 If a child range is requested, it will only affect the current call to
31824 @code{-var-list-children}, but not future calls to @code{-var-update}.
31825 For this, you must instead use @code{-var-set-update-range}. The
31826 intent of this approach is to enable a front end to implement any
31827 update approach it likes; for example, scrolling a view may cause the
31828 front end to request more children with @code{-var-list-children}, and
31829 then the front end could call @code{-var-set-update-range} with a
31830 different range to ensure that future updates are restricted to just
31833 For each child the following results are returned:
31838 Name of the variable object created for this child.
31841 The expression to be shown to the user by the front end to designate this child.
31842 For example this may be the name of a structure member.
31844 For a dynamic varobj, this value cannot be used to form an
31845 expression. There is no way to do this at all with a dynamic varobj.
31847 For C/C@t{++} structures there are several pseudo children returned to
31848 designate access qualifiers. For these pseudo children @var{exp} is
31849 @samp{public}, @samp{private}, or @samp{protected}. In this case the
31850 type and value are not present.
31852 A dynamic varobj will not report the access qualifying
31853 pseudo-children, regardless of the language. This information is not
31854 available at all with a dynamic varobj.
31857 Number of children this child has. For a dynamic varobj, this will be
31861 The type of the child. If @samp{print object}
31862 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31863 @emph{actual} (derived) type of the object is shown rather than the
31864 @emph{declared} one.
31867 If values were requested, this is the value.
31870 If this variable object is associated with a thread, this is the thread id.
31871 Otherwise this result is not present.
31874 If the variable object is frozen, this variable will be present with a value of 1.
31877 The result may have its own attributes:
31881 A dynamic varobj can supply a display hint to the front end. The
31882 value comes directly from the Python pretty-printer object's
31883 @code{display_hint} method. @xref{Pretty Printing API}.
31886 This is an integer attribute which is nonzero if there are children
31887 remaining after the end of the selected range.
31890 @subsubheading Example
31894 -var-list-children n
31895 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
31896 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
31898 -var-list-children --all-values n
31899 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
31900 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
31904 @subheading The @code{-var-info-type} Command
31905 @findex -var-info-type
31907 @subsubheading Synopsis
31910 -var-info-type @var{name}
31913 Returns the type of the specified variable @var{name}. The type is
31914 returned as a string in the same format as it is output by the
31918 type=@var{typename}
31922 @subheading The @code{-var-info-expression} Command
31923 @findex -var-info-expression
31925 @subsubheading Synopsis
31928 -var-info-expression @var{name}
31931 Returns a string that is suitable for presenting this
31932 variable object in user interface. The string is generally
31933 not valid expression in the current language, and cannot be evaluated.
31935 For example, if @code{a} is an array, and variable object
31936 @code{A} was created for @code{a}, then we'll get this output:
31939 (gdb) -var-info-expression A.1
31940 ^done,lang="C",exp="1"
31944 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
31946 Note that the output of the @code{-var-list-children} command also
31947 includes those expressions, so the @code{-var-info-expression} command
31950 @subheading The @code{-var-info-path-expression} Command
31951 @findex -var-info-path-expression
31953 @subsubheading Synopsis
31956 -var-info-path-expression @var{name}
31959 Returns an expression that can be evaluated in the current
31960 context and will yield the same value that a variable object has.
31961 Compare this with the @code{-var-info-expression} command, which
31962 result can be used only for UI presentation. Typical use of
31963 the @code{-var-info-path-expression} command is creating a
31964 watchpoint from a variable object.
31966 This command is currently not valid for children of a dynamic varobj,
31967 and will give an error when invoked on one.
31969 For example, suppose @code{C} is a C@t{++} class, derived from class
31970 @code{Base}, and that the @code{Base} class has a member called
31971 @code{m_size}. Assume a variable @code{c} is has the type of
31972 @code{C} and a variable object @code{C} was created for variable
31973 @code{c}. Then, we'll get this output:
31975 (gdb) -var-info-path-expression C.Base.public.m_size
31976 ^done,path_expr=((Base)c).m_size)
31979 @subheading The @code{-var-show-attributes} Command
31980 @findex -var-show-attributes
31982 @subsubheading Synopsis
31985 -var-show-attributes @var{name}
31988 List attributes of the specified variable object @var{name}:
31991 status=@var{attr} [ ( ,@var{attr} )* ]
31995 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
31997 @subheading The @code{-var-evaluate-expression} Command
31998 @findex -var-evaluate-expression
32000 @subsubheading Synopsis
32003 -var-evaluate-expression [-f @var{format-spec}] @var{name}
32006 Evaluates the expression that is represented by the specified variable
32007 object and returns its value as a string. The format of the string
32008 can be specified with the @samp{-f} option. The possible values of
32009 this option are the same as for @code{-var-set-format}
32010 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
32011 the current display format will be used. The current display format
32012 can be changed using the @code{-var-set-format} command.
32018 Note that one must invoke @code{-var-list-children} for a variable
32019 before the value of a child variable can be evaluated.
32021 @subheading The @code{-var-assign} Command
32022 @findex -var-assign
32024 @subsubheading Synopsis
32027 -var-assign @var{name} @var{expression}
32030 Assigns the value of @var{expression} to the variable object specified
32031 by @var{name}. The object must be @samp{editable}. If the variable's
32032 value is altered by the assign, the variable will show up in any
32033 subsequent @code{-var-update} list.
32035 @subsubheading Example
32043 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
32047 @subheading The @code{-var-update} Command
32048 @findex -var-update
32050 @subsubheading Synopsis
32053 -var-update [@var{print-values}] @{@var{name} | "*"@}
32056 Reevaluate the expressions corresponding to the variable object
32057 @var{name} and all its direct and indirect children, and return the
32058 list of variable objects whose values have changed; @var{name} must
32059 be a root variable object. Here, ``changed'' means that the result of
32060 @code{-var-evaluate-expression} before and after the
32061 @code{-var-update} is different. If @samp{*} is used as the variable
32062 object names, all existing variable objects are updated, except
32063 for frozen ones (@pxref{-var-set-frozen}). The option
32064 @var{print-values} determines whether both names and values, or just
32065 names are printed. The possible values of this option are the same
32066 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
32067 recommended to use the @samp{--all-values} option, to reduce the
32068 number of MI commands needed on each program stop.
32070 With the @samp{*} parameter, if a variable object is bound to a
32071 currently running thread, it will not be updated, without any
32074 If @code{-var-set-update-range} was previously used on a varobj, then
32075 only the selected range of children will be reported.
32077 @code{-var-update} reports all the changed varobjs in a tuple named
32080 Each item in the change list is itself a tuple holding:
32084 The name of the varobj.
32087 If values were requested for this update, then this field will be
32088 present and will hold the value of the varobj.
32091 @anchor{-var-update}
32092 This field is a string which may take one of three values:
32096 The variable object's current value is valid.
32099 The variable object does not currently hold a valid value but it may
32100 hold one in the future if its associated expression comes back into
32104 The variable object no longer holds a valid value.
32105 This can occur when the executable file being debugged has changed,
32106 either through recompilation or by using the @value{GDBN} @code{file}
32107 command. The front end should normally choose to delete these variable
32111 In the future new values may be added to this list so the front should
32112 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
32115 This is only present if the varobj is still valid. If the type
32116 changed, then this will be the string @samp{true}; otherwise it will
32119 When a varobj's type changes, its children are also likely to have
32120 become incorrect. Therefore, the varobj's children are automatically
32121 deleted when this attribute is @samp{true}. Also, the varobj's update
32122 range, when set using the @code{-var-set-update-range} command, is
32126 If the varobj's type changed, then this field will be present and will
32129 @item new_num_children
32130 For a dynamic varobj, if the number of children changed, or if the
32131 type changed, this will be the new number of children.
32133 The @samp{numchild} field in other varobj responses is generally not
32134 valid for a dynamic varobj -- it will show the number of children that
32135 @value{GDBN} knows about, but because dynamic varobjs lazily
32136 instantiate their children, this will not reflect the number of
32137 children which may be available.
32139 The @samp{new_num_children} attribute only reports changes to the
32140 number of children known by @value{GDBN}. This is the only way to
32141 detect whether an update has removed children (which necessarily can
32142 only happen at the end of the update range).
32145 The display hint, if any.
32148 This is an integer value, which will be 1 if there are more children
32149 available outside the varobj's update range.
32152 This attribute will be present and have the value @samp{1} if the
32153 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32154 then this attribute will not be present.
32157 If new children were added to a dynamic varobj within the selected
32158 update range (as set by @code{-var-set-update-range}), then they will
32159 be listed in this attribute.
32162 @subsubheading Example
32169 -var-update --all-values var1
32170 ^done,changelist=[@{name="var1",value="3",in_scope="true",
32171 type_changed="false"@}]
32175 @subheading The @code{-var-set-frozen} Command
32176 @findex -var-set-frozen
32177 @anchor{-var-set-frozen}
32179 @subsubheading Synopsis
32182 -var-set-frozen @var{name} @var{flag}
32185 Set the frozenness flag on the variable object @var{name}. The
32186 @var{flag} parameter should be either @samp{1} to make the variable
32187 frozen or @samp{0} to make it unfrozen. If a variable object is
32188 frozen, then neither itself, nor any of its children, are
32189 implicitly updated by @code{-var-update} of
32190 a parent variable or by @code{-var-update *}. Only
32191 @code{-var-update} of the variable itself will update its value and
32192 values of its children. After a variable object is unfrozen, it is
32193 implicitly updated by all subsequent @code{-var-update} operations.
32194 Unfreezing a variable does not update it, only subsequent
32195 @code{-var-update} does.
32197 @subsubheading Example
32201 -var-set-frozen V 1
32206 @subheading The @code{-var-set-update-range} command
32207 @findex -var-set-update-range
32208 @anchor{-var-set-update-range}
32210 @subsubheading Synopsis
32213 -var-set-update-range @var{name} @var{from} @var{to}
32216 Set the range of children to be returned by future invocations of
32217 @code{-var-update}.
32219 @var{from} and @var{to} indicate the range of children to report. If
32220 @var{from} or @var{to} is less than zero, the range is reset and all
32221 children will be reported. Otherwise, children starting at @var{from}
32222 (zero-based) and up to and excluding @var{to} will be reported.
32224 @subsubheading Example
32228 -var-set-update-range V 1 2
32232 @subheading The @code{-var-set-visualizer} command
32233 @findex -var-set-visualizer
32234 @anchor{-var-set-visualizer}
32236 @subsubheading Synopsis
32239 -var-set-visualizer @var{name} @var{visualizer}
32242 Set a visualizer for the variable object @var{name}.
32244 @var{visualizer} is the visualizer to use. The special value
32245 @samp{None} means to disable any visualizer in use.
32247 If not @samp{None}, @var{visualizer} must be a Python expression.
32248 This expression must evaluate to a callable object which accepts a
32249 single argument. @value{GDBN} will call this object with the value of
32250 the varobj @var{name} as an argument (this is done so that the same
32251 Python pretty-printing code can be used for both the CLI and MI).
32252 When called, this object must return an object which conforms to the
32253 pretty-printing interface (@pxref{Pretty Printing API}).
32255 The pre-defined function @code{gdb.default_visualizer} may be used to
32256 select a visualizer by following the built-in process
32257 (@pxref{Selecting Pretty-Printers}). This is done automatically when
32258 a varobj is created, and so ordinarily is not needed.
32260 This feature is only available if Python support is enabled. The MI
32261 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
32262 can be used to check this.
32264 @subsubheading Example
32266 Resetting the visualizer:
32270 -var-set-visualizer V None
32274 Reselecting the default (type-based) visualizer:
32278 -var-set-visualizer V gdb.default_visualizer
32282 Suppose @code{SomeClass} is a visualizer class. A lambda expression
32283 can be used to instantiate this class for a varobj:
32287 -var-set-visualizer V "lambda val: SomeClass()"
32291 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32292 @node GDB/MI Data Manipulation
32293 @section @sc{gdb/mi} Data Manipulation
32295 @cindex data manipulation, in @sc{gdb/mi}
32296 @cindex @sc{gdb/mi}, data manipulation
32297 This section describes the @sc{gdb/mi} commands that manipulate data:
32298 examine memory and registers, evaluate expressions, etc.
32300 @c REMOVED FROM THE INTERFACE.
32301 @c @subheading -data-assign
32302 @c Change the value of a program variable. Plenty of side effects.
32303 @c @subsubheading GDB Command
32305 @c @subsubheading Example
32308 @subheading The @code{-data-disassemble} Command
32309 @findex -data-disassemble
32311 @subsubheading Synopsis
32315 [ -s @var{start-addr} -e @var{end-addr} ]
32316 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
32324 @item @var{start-addr}
32325 is the beginning address (or @code{$pc})
32326 @item @var{end-addr}
32328 @item @var{filename}
32329 is the name of the file to disassemble
32330 @item @var{linenum}
32331 is the line number to disassemble around
32333 is the number of disassembly lines to be produced. If it is -1,
32334 the whole function will be disassembled, in case no @var{end-addr} is
32335 specified. If @var{end-addr} is specified as a non-zero value, and
32336 @var{lines} is lower than the number of disassembly lines between
32337 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
32338 displayed; if @var{lines} is higher than the number of lines between
32339 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
32342 is either 0 (meaning only disassembly), 1 (meaning mixed source and
32343 disassembly), 2 (meaning disassembly with raw opcodes), or 3 (meaning
32344 mixed source and disassembly with raw opcodes).
32347 @subsubheading Result
32349 The result of the @code{-data-disassemble} command will be a list named
32350 @samp{asm_insns}, the contents of this list depend on the @var{mode}
32351 used with the @code{-data-disassemble} command.
32353 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
32358 The address at which this instruction was disassembled.
32361 The name of the function this instruction is within.
32364 The decimal offset in bytes from the start of @samp{func-name}.
32367 The text disassembly for this @samp{address}.
32370 This field is only present for mode 2. This contains the raw opcode
32371 bytes for the @samp{inst} field.
32375 For modes 1 and 3 the @samp{asm_insns} list contains tuples named
32376 @samp{src_and_asm_line}, each of which has the following fields:
32380 The line number within @samp{file}.
32383 The file name from the compilation unit. This might be an absolute
32384 file name or a relative file name depending on the compile command
32388 Absolute file name of @samp{file}. It is converted to a canonical form
32389 using the source file search path
32390 (@pxref{Source Path, ,Specifying Source Directories})
32391 and after resolving all the symbolic links.
32393 If the source file is not found this field will contain the path as
32394 present in the debug information.
32396 @item line_asm_insn
32397 This is a list of tuples containing the disassembly for @samp{line} in
32398 @samp{file}. The fields of each tuple are the same as for
32399 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
32400 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
32405 Note that whatever included in the @samp{inst} field, is not
32406 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
32409 @subsubheading @value{GDBN} Command
32411 The corresponding @value{GDBN} command is @samp{disassemble}.
32413 @subsubheading Example
32415 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
32419 -data-disassemble -s $pc -e "$pc + 20" -- 0
32422 @{address="0x000107c0",func-name="main",offset="4",
32423 inst="mov 2, %o0"@},
32424 @{address="0x000107c4",func-name="main",offset="8",
32425 inst="sethi %hi(0x11800), %o2"@},
32426 @{address="0x000107c8",func-name="main",offset="12",
32427 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
32428 @{address="0x000107cc",func-name="main",offset="16",
32429 inst="sethi %hi(0x11800), %o2"@},
32430 @{address="0x000107d0",func-name="main",offset="20",
32431 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
32435 Disassemble the whole @code{main} function. Line 32 is part of
32439 -data-disassemble -f basics.c -l 32 -- 0
32441 @{address="0x000107bc",func-name="main",offset="0",
32442 inst="save %sp, -112, %sp"@},
32443 @{address="0x000107c0",func-name="main",offset="4",
32444 inst="mov 2, %o0"@},
32445 @{address="0x000107c4",func-name="main",offset="8",
32446 inst="sethi %hi(0x11800), %o2"@},
32448 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
32449 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
32453 Disassemble 3 instructions from the start of @code{main}:
32457 -data-disassemble -f basics.c -l 32 -n 3 -- 0
32459 @{address="0x000107bc",func-name="main",offset="0",
32460 inst="save %sp, -112, %sp"@},
32461 @{address="0x000107c0",func-name="main",offset="4",
32462 inst="mov 2, %o0"@},
32463 @{address="0x000107c4",func-name="main",offset="8",
32464 inst="sethi %hi(0x11800), %o2"@}]
32468 Disassemble 3 instructions from the start of @code{main} in mixed mode:
32472 -data-disassemble -f basics.c -l 32 -n 3 -- 1
32474 src_and_asm_line=@{line="31",
32475 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32476 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32477 line_asm_insn=[@{address="0x000107bc",
32478 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
32479 src_and_asm_line=@{line="32",
32480 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32481 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32482 line_asm_insn=[@{address="0x000107c0",
32483 func-name="main",offset="4",inst="mov 2, %o0"@},
32484 @{address="0x000107c4",func-name="main",offset="8",
32485 inst="sethi %hi(0x11800), %o2"@}]@}]
32490 @subheading The @code{-data-evaluate-expression} Command
32491 @findex -data-evaluate-expression
32493 @subsubheading Synopsis
32496 -data-evaluate-expression @var{expr}
32499 Evaluate @var{expr} as an expression. The expression could contain an
32500 inferior function call. The function call will execute synchronously.
32501 If the expression contains spaces, it must be enclosed in double quotes.
32503 @subsubheading @value{GDBN} Command
32505 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
32506 @samp{call}. In @code{gdbtk} only, there's a corresponding
32507 @samp{gdb_eval} command.
32509 @subsubheading Example
32511 In the following example, the numbers that precede the commands are the
32512 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
32513 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
32517 211-data-evaluate-expression A
32520 311-data-evaluate-expression &A
32521 311^done,value="0xefffeb7c"
32523 411-data-evaluate-expression A+3
32526 511-data-evaluate-expression "A + 3"
32532 @subheading The @code{-data-list-changed-registers} Command
32533 @findex -data-list-changed-registers
32535 @subsubheading Synopsis
32538 -data-list-changed-registers
32541 Display a list of the registers that have changed.
32543 @subsubheading @value{GDBN} Command
32545 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
32546 has the corresponding command @samp{gdb_changed_register_list}.
32548 @subsubheading Example
32550 On a PPC MBX board:
32558 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
32559 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
32562 -data-list-changed-registers
32563 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
32564 "10","11","13","14","15","16","17","18","19","20","21","22","23",
32565 "24","25","26","27","28","30","31","64","65","66","67","69"]
32570 @subheading The @code{-data-list-register-names} Command
32571 @findex -data-list-register-names
32573 @subsubheading Synopsis
32576 -data-list-register-names [ ( @var{regno} )+ ]
32579 Show a list of register names for the current target. If no arguments
32580 are given, it shows a list of the names of all the registers. If
32581 integer numbers are given as arguments, it will print a list of the
32582 names of the registers corresponding to the arguments. To ensure
32583 consistency between a register name and its number, the output list may
32584 include empty register names.
32586 @subsubheading @value{GDBN} Command
32588 @value{GDBN} does not have a command which corresponds to
32589 @samp{-data-list-register-names}. In @code{gdbtk} there is a
32590 corresponding command @samp{gdb_regnames}.
32592 @subsubheading Example
32594 For the PPC MBX board:
32597 -data-list-register-names
32598 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
32599 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
32600 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
32601 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
32602 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
32603 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
32604 "", "pc","ps","cr","lr","ctr","xer"]
32606 -data-list-register-names 1 2 3
32607 ^done,register-names=["r1","r2","r3"]
32611 @subheading The @code{-data-list-register-values} Command
32612 @findex -data-list-register-values
32614 @subsubheading Synopsis
32617 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
32620 Display the registers' contents. @var{fmt} is the format according to
32621 which the registers' contents are to be returned, followed by an optional
32622 list of numbers specifying the registers to display. A missing list of
32623 numbers indicates that the contents of all the registers must be returned.
32625 Allowed formats for @var{fmt} are:
32642 @subsubheading @value{GDBN} Command
32644 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
32645 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
32647 @subsubheading Example
32649 For a PPC MBX board (note: line breaks are for readability only, they
32650 don't appear in the actual output):
32654 -data-list-register-values r 64 65
32655 ^done,register-values=[@{number="64",value="0xfe00a300"@},
32656 @{number="65",value="0x00029002"@}]
32658 -data-list-register-values x
32659 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
32660 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
32661 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
32662 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
32663 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
32664 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
32665 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
32666 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
32667 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
32668 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
32669 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
32670 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
32671 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
32672 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
32673 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
32674 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
32675 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
32676 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
32677 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
32678 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
32679 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
32680 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
32681 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
32682 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
32683 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
32684 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
32685 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
32686 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
32687 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
32688 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
32689 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
32690 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
32691 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
32692 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
32693 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
32694 @{number="69",value="0x20002b03"@}]
32699 @subheading The @code{-data-read-memory} Command
32700 @findex -data-read-memory
32702 This command is deprecated, use @code{-data-read-memory-bytes} instead.
32704 @subsubheading Synopsis
32707 -data-read-memory [ -o @var{byte-offset} ]
32708 @var{address} @var{word-format} @var{word-size}
32709 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
32716 @item @var{address}
32717 An expression specifying the address of the first memory word to be
32718 read. Complex expressions containing embedded white space should be
32719 quoted using the C convention.
32721 @item @var{word-format}
32722 The format to be used to print the memory words. The notation is the
32723 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
32726 @item @var{word-size}
32727 The size of each memory word in bytes.
32729 @item @var{nr-rows}
32730 The number of rows in the output table.
32732 @item @var{nr-cols}
32733 The number of columns in the output table.
32736 If present, indicates that each row should include an @sc{ascii} dump. The
32737 value of @var{aschar} is used as a padding character when a byte is not a
32738 member of the printable @sc{ascii} character set (printable @sc{ascii}
32739 characters are those whose code is between 32 and 126, inclusively).
32741 @item @var{byte-offset}
32742 An offset to add to the @var{address} before fetching memory.
32745 This command displays memory contents as a table of @var{nr-rows} by
32746 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
32747 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
32748 (returned as @samp{total-bytes}). Should less than the requested number
32749 of bytes be returned by the target, the missing words are identified
32750 using @samp{N/A}. The number of bytes read from the target is returned
32751 in @samp{nr-bytes} and the starting address used to read memory in
32754 The address of the next/previous row or page is available in
32755 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
32758 @subsubheading @value{GDBN} Command
32760 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
32761 @samp{gdb_get_mem} memory read command.
32763 @subsubheading Example
32765 Read six bytes of memory starting at @code{bytes+6} but then offset by
32766 @code{-6} bytes. Format as three rows of two columns. One byte per
32767 word. Display each word in hex.
32771 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
32772 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
32773 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
32774 prev-page="0x0000138a",memory=[
32775 @{addr="0x00001390",data=["0x00","0x01"]@},
32776 @{addr="0x00001392",data=["0x02","0x03"]@},
32777 @{addr="0x00001394",data=["0x04","0x05"]@}]
32781 Read two bytes of memory starting at address @code{shorts + 64} and
32782 display as a single word formatted in decimal.
32786 5-data-read-memory shorts+64 d 2 1 1
32787 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
32788 next-row="0x00001512",prev-row="0x0000150e",
32789 next-page="0x00001512",prev-page="0x0000150e",memory=[
32790 @{addr="0x00001510",data=["128"]@}]
32794 Read thirty two bytes of memory starting at @code{bytes+16} and format
32795 as eight rows of four columns. Include a string encoding with @samp{x}
32796 used as the non-printable character.
32800 4-data-read-memory bytes+16 x 1 8 4 x
32801 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
32802 next-row="0x000013c0",prev-row="0x0000139c",
32803 next-page="0x000013c0",prev-page="0x00001380",memory=[
32804 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
32805 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
32806 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
32807 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
32808 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
32809 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
32810 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
32811 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
32815 @subheading The @code{-data-read-memory-bytes} Command
32816 @findex -data-read-memory-bytes
32818 @subsubheading Synopsis
32821 -data-read-memory-bytes [ -o @var{byte-offset} ]
32822 @var{address} @var{count}
32829 @item @var{address}
32830 An expression specifying the address of the first memory word to be
32831 read. Complex expressions containing embedded white space should be
32832 quoted using the C convention.
32835 The number of bytes to read. This should be an integer literal.
32837 @item @var{byte-offset}
32838 The offsets in bytes relative to @var{address} at which to start
32839 reading. This should be an integer literal. This option is provided
32840 so that a frontend is not required to first evaluate address and then
32841 perform address arithmetics itself.
32845 This command attempts to read all accessible memory regions in the
32846 specified range. First, all regions marked as unreadable in the memory
32847 map (if one is defined) will be skipped. @xref{Memory Region
32848 Attributes}. Second, @value{GDBN} will attempt to read the remaining
32849 regions. For each one, if reading full region results in an errors,
32850 @value{GDBN} will try to read a subset of the region.
32852 In general, every single byte in the region may be readable or not,
32853 and the only way to read every readable byte is to try a read at
32854 every address, which is not practical. Therefore, @value{GDBN} will
32855 attempt to read all accessible bytes at either beginning or the end
32856 of the region, using a binary division scheme. This heuristic works
32857 well for reading accross a memory map boundary. Note that if a region
32858 has a readable range that is neither at the beginning or the end,
32859 @value{GDBN} will not read it.
32861 The result record (@pxref{GDB/MI Result Records}) that is output of
32862 the command includes a field named @samp{memory} whose content is a
32863 list of tuples. Each tuple represent a successfully read memory block
32864 and has the following fields:
32868 The start address of the memory block, as hexadecimal literal.
32871 The end address of the memory block, as hexadecimal literal.
32874 The offset of the memory block, as hexadecimal literal, relative to
32875 the start address passed to @code{-data-read-memory-bytes}.
32878 The contents of the memory block, in hex.
32884 @subsubheading @value{GDBN} Command
32886 The corresponding @value{GDBN} command is @samp{x}.
32888 @subsubheading Example
32892 -data-read-memory-bytes &a 10
32893 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
32895 contents="01000000020000000300"@}]
32900 @subheading The @code{-data-write-memory-bytes} Command
32901 @findex -data-write-memory-bytes
32903 @subsubheading Synopsis
32906 -data-write-memory-bytes @var{address} @var{contents}
32907 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
32914 @item @var{address}
32915 An expression specifying the address of the first memory word to be
32916 read. Complex expressions containing embedded white space should be
32917 quoted using the C convention.
32919 @item @var{contents}
32920 The hex-encoded bytes to write.
32923 Optional argument indicating the number of bytes to be written. If @var{count}
32924 is greater than @var{contents}' length, @value{GDBN} will repeatedly
32925 write @var{contents} until it fills @var{count} bytes.
32929 @subsubheading @value{GDBN} Command
32931 There's no corresponding @value{GDBN} command.
32933 @subsubheading Example
32937 -data-write-memory-bytes &a "aabbccdd"
32944 -data-write-memory-bytes &a "aabbccdd" 16e
32949 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32950 @node GDB/MI Tracepoint Commands
32951 @section @sc{gdb/mi} Tracepoint Commands
32953 The commands defined in this section implement MI support for
32954 tracepoints. For detailed introduction, see @ref{Tracepoints}.
32956 @subheading The @code{-trace-find} Command
32957 @findex -trace-find
32959 @subsubheading Synopsis
32962 -trace-find @var{mode} [@var{parameters}@dots{}]
32965 Find a trace frame using criteria defined by @var{mode} and
32966 @var{parameters}. The following table lists permissible
32967 modes and their parameters. For details of operation, see @ref{tfind}.
32972 No parameters are required. Stops examining trace frames.
32975 An integer is required as parameter. Selects tracepoint frame with
32978 @item tracepoint-number
32979 An integer is required as parameter. Finds next
32980 trace frame that corresponds to tracepoint with the specified number.
32983 An address is required as parameter. Finds
32984 next trace frame that corresponds to any tracepoint at the specified
32987 @item pc-inside-range
32988 Two addresses are required as parameters. Finds next trace
32989 frame that corresponds to a tracepoint at an address inside the
32990 specified range. Both bounds are considered to be inside the range.
32992 @item pc-outside-range
32993 Two addresses are required as parameters. Finds
32994 next trace frame that corresponds to a tracepoint at an address outside
32995 the specified range. Both bounds are considered to be inside the range.
32998 Line specification is required as parameter. @xref{Specify Location}.
32999 Finds next trace frame that corresponds to a tracepoint at
33000 the specified location.
33004 If @samp{none} was passed as @var{mode}, the response does not
33005 have fields. Otherwise, the response may have the following fields:
33009 This field has either @samp{0} or @samp{1} as the value, depending
33010 on whether a matching tracepoint was found.
33013 The index of the found traceframe. This field is present iff
33014 the @samp{found} field has value of @samp{1}.
33017 The index of the found tracepoint. This field is present iff
33018 the @samp{found} field has value of @samp{1}.
33021 The information about the frame corresponding to the found trace
33022 frame. This field is present only if a trace frame was found.
33023 @xref{GDB/MI Frame Information}, for description of this field.
33027 @subsubheading @value{GDBN} Command
33029 The corresponding @value{GDBN} command is @samp{tfind}.
33031 @subheading -trace-define-variable
33032 @findex -trace-define-variable
33034 @subsubheading Synopsis
33037 -trace-define-variable @var{name} [ @var{value} ]
33040 Create trace variable @var{name} if it does not exist. If
33041 @var{value} is specified, sets the initial value of the specified
33042 trace variable to that value. Note that the @var{name} should start
33043 with the @samp{$} character.
33045 @subsubheading @value{GDBN} Command
33047 The corresponding @value{GDBN} command is @samp{tvariable}.
33049 @subheading -trace-list-variables
33050 @findex -trace-list-variables
33052 @subsubheading Synopsis
33055 -trace-list-variables
33058 Return a table of all defined trace variables. Each element of the
33059 table has the following fields:
33063 The name of the trace variable. This field is always present.
33066 The initial value. This is a 64-bit signed integer. This
33067 field is always present.
33070 The value the trace variable has at the moment. This is a 64-bit
33071 signed integer. This field is absent iff current value is
33072 not defined, for example if the trace was never run, or is
33077 @subsubheading @value{GDBN} Command
33079 The corresponding @value{GDBN} command is @samp{tvariables}.
33081 @subsubheading Example
33085 -trace-list-variables
33086 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
33087 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
33088 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
33089 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
33090 body=[variable=@{name="$trace_timestamp",initial="0"@}
33091 variable=@{name="$foo",initial="10",current="15"@}]@}
33095 @subheading -trace-save
33096 @findex -trace-save
33098 @subsubheading Synopsis
33101 -trace-save [-r ] @var{filename}
33104 Saves the collected trace data to @var{filename}. Without the
33105 @samp{-r} option, the data is downloaded from the target and saved
33106 in a local file. With the @samp{-r} option the target is asked
33107 to perform the save.
33109 @subsubheading @value{GDBN} Command
33111 The corresponding @value{GDBN} command is @samp{tsave}.
33114 @subheading -trace-start
33115 @findex -trace-start
33117 @subsubheading Synopsis
33123 Starts a tracing experiments. The result of this command does not
33126 @subsubheading @value{GDBN} Command
33128 The corresponding @value{GDBN} command is @samp{tstart}.
33130 @subheading -trace-status
33131 @findex -trace-status
33133 @subsubheading Synopsis
33139 Obtains the status of a tracing experiment. The result may include
33140 the following fields:
33145 May have a value of either @samp{0}, when no tracing operations are
33146 supported, @samp{1}, when all tracing operations are supported, or
33147 @samp{file} when examining trace file. In the latter case, examining
33148 of trace frame is possible but new tracing experiement cannot be
33149 started. This field is always present.
33152 May have a value of either @samp{0} or @samp{1} depending on whether
33153 tracing experiement is in progress on target. This field is present
33154 if @samp{supported} field is not @samp{0}.
33157 Report the reason why the tracing was stopped last time. This field
33158 may be absent iff tracing was never stopped on target yet. The
33159 value of @samp{request} means the tracing was stopped as result of
33160 the @code{-trace-stop} command. The value of @samp{overflow} means
33161 the tracing buffer is full. The value of @samp{disconnection} means
33162 tracing was automatically stopped when @value{GDBN} has disconnected.
33163 The value of @samp{passcount} means tracing was stopped when a
33164 tracepoint was passed a maximal number of times for that tracepoint.
33165 This field is present if @samp{supported} field is not @samp{0}.
33167 @item stopping-tracepoint
33168 The number of tracepoint whose passcount as exceeded. This field is
33169 present iff the @samp{stop-reason} field has the value of
33173 @itemx frames-created
33174 The @samp{frames} field is a count of the total number of trace frames
33175 in the trace buffer, while @samp{frames-created} is the total created
33176 during the run, including ones that were discarded, such as when a
33177 circular trace buffer filled up. Both fields are optional.
33181 These fields tell the current size of the tracing buffer and the
33182 remaining space. These fields are optional.
33185 The value of the circular trace buffer flag. @code{1} means that the
33186 trace buffer is circular and old trace frames will be discarded if
33187 necessary to make room, @code{0} means that the trace buffer is linear
33191 The value of the disconnected tracing flag. @code{1} means that
33192 tracing will continue after @value{GDBN} disconnects, @code{0} means
33193 that the trace run will stop.
33196 The filename of the trace file being examined. This field is
33197 optional, and only present when examining a trace file.
33201 @subsubheading @value{GDBN} Command
33203 The corresponding @value{GDBN} command is @samp{tstatus}.
33205 @subheading -trace-stop
33206 @findex -trace-stop
33208 @subsubheading Synopsis
33214 Stops a tracing experiment. The result of this command has the same
33215 fields as @code{-trace-status}, except that the @samp{supported} and
33216 @samp{running} fields are not output.
33218 @subsubheading @value{GDBN} Command
33220 The corresponding @value{GDBN} command is @samp{tstop}.
33223 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33224 @node GDB/MI Symbol Query
33225 @section @sc{gdb/mi} Symbol Query Commands
33229 @subheading The @code{-symbol-info-address} Command
33230 @findex -symbol-info-address
33232 @subsubheading Synopsis
33235 -symbol-info-address @var{symbol}
33238 Describe where @var{symbol} is stored.
33240 @subsubheading @value{GDBN} Command
33242 The corresponding @value{GDBN} command is @samp{info address}.
33244 @subsubheading Example
33248 @subheading The @code{-symbol-info-file} Command
33249 @findex -symbol-info-file
33251 @subsubheading Synopsis
33257 Show the file for the symbol.
33259 @subsubheading @value{GDBN} Command
33261 There's no equivalent @value{GDBN} command. @code{gdbtk} has
33262 @samp{gdb_find_file}.
33264 @subsubheading Example
33268 @subheading The @code{-symbol-info-function} Command
33269 @findex -symbol-info-function
33271 @subsubheading Synopsis
33274 -symbol-info-function
33277 Show which function the symbol lives in.
33279 @subsubheading @value{GDBN} Command
33281 @samp{gdb_get_function} in @code{gdbtk}.
33283 @subsubheading Example
33287 @subheading The @code{-symbol-info-line} Command
33288 @findex -symbol-info-line
33290 @subsubheading Synopsis
33296 Show the core addresses of the code for a source line.
33298 @subsubheading @value{GDBN} Command
33300 The corresponding @value{GDBN} command is @samp{info line}.
33301 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
33303 @subsubheading Example
33307 @subheading The @code{-symbol-info-symbol} Command
33308 @findex -symbol-info-symbol
33310 @subsubheading Synopsis
33313 -symbol-info-symbol @var{addr}
33316 Describe what symbol is at location @var{addr}.
33318 @subsubheading @value{GDBN} Command
33320 The corresponding @value{GDBN} command is @samp{info symbol}.
33322 @subsubheading Example
33326 @subheading The @code{-symbol-list-functions} Command
33327 @findex -symbol-list-functions
33329 @subsubheading Synopsis
33332 -symbol-list-functions
33335 List the functions in the executable.
33337 @subsubheading @value{GDBN} Command
33339 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
33340 @samp{gdb_search} in @code{gdbtk}.
33342 @subsubheading Example
33347 @subheading The @code{-symbol-list-lines} Command
33348 @findex -symbol-list-lines
33350 @subsubheading Synopsis
33353 -symbol-list-lines @var{filename}
33356 Print the list of lines that contain code and their associated program
33357 addresses for the given source filename. The entries are sorted in
33358 ascending PC order.
33360 @subsubheading @value{GDBN} Command
33362 There is no corresponding @value{GDBN} command.
33364 @subsubheading Example
33367 -symbol-list-lines basics.c
33368 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
33374 @subheading The @code{-symbol-list-types} Command
33375 @findex -symbol-list-types
33377 @subsubheading Synopsis
33383 List all the type names.
33385 @subsubheading @value{GDBN} Command
33387 The corresponding commands are @samp{info types} in @value{GDBN},
33388 @samp{gdb_search} in @code{gdbtk}.
33390 @subsubheading Example
33394 @subheading The @code{-symbol-list-variables} Command
33395 @findex -symbol-list-variables
33397 @subsubheading Synopsis
33400 -symbol-list-variables
33403 List all the global and static variable names.
33405 @subsubheading @value{GDBN} Command
33407 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
33409 @subsubheading Example
33413 @subheading The @code{-symbol-locate} Command
33414 @findex -symbol-locate
33416 @subsubheading Synopsis
33422 @subsubheading @value{GDBN} Command
33424 @samp{gdb_loc} in @code{gdbtk}.
33426 @subsubheading Example
33430 @subheading The @code{-symbol-type} Command
33431 @findex -symbol-type
33433 @subsubheading Synopsis
33436 -symbol-type @var{variable}
33439 Show type of @var{variable}.
33441 @subsubheading @value{GDBN} Command
33443 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
33444 @samp{gdb_obj_variable}.
33446 @subsubheading Example
33451 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33452 @node GDB/MI File Commands
33453 @section @sc{gdb/mi} File Commands
33455 This section describes the GDB/MI commands to specify executable file names
33456 and to read in and obtain symbol table information.
33458 @subheading The @code{-file-exec-and-symbols} Command
33459 @findex -file-exec-and-symbols
33461 @subsubheading Synopsis
33464 -file-exec-and-symbols @var{file}
33467 Specify the executable file to be debugged. This file is the one from
33468 which the symbol table is also read. If no file is specified, the
33469 command clears the executable and symbol information. If breakpoints
33470 are set when using this command with no arguments, @value{GDBN} will produce
33471 error messages. Otherwise, no output is produced, except a completion
33474 @subsubheading @value{GDBN} Command
33476 The corresponding @value{GDBN} command is @samp{file}.
33478 @subsubheading Example
33482 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33488 @subheading The @code{-file-exec-file} Command
33489 @findex -file-exec-file
33491 @subsubheading Synopsis
33494 -file-exec-file @var{file}
33497 Specify the executable file to be debugged. Unlike
33498 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
33499 from this file. If used without argument, @value{GDBN} clears the information
33500 about the executable file. No output is produced, except a completion
33503 @subsubheading @value{GDBN} Command
33505 The corresponding @value{GDBN} command is @samp{exec-file}.
33507 @subsubheading Example
33511 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33518 @subheading The @code{-file-list-exec-sections} Command
33519 @findex -file-list-exec-sections
33521 @subsubheading Synopsis
33524 -file-list-exec-sections
33527 List the sections of the current executable file.
33529 @subsubheading @value{GDBN} Command
33531 The @value{GDBN} command @samp{info file} shows, among the rest, the same
33532 information as this command. @code{gdbtk} has a corresponding command
33533 @samp{gdb_load_info}.
33535 @subsubheading Example
33540 @subheading The @code{-file-list-exec-source-file} Command
33541 @findex -file-list-exec-source-file
33543 @subsubheading Synopsis
33546 -file-list-exec-source-file
33549 List the line number, the current source file, and the absolute path
33550 to the current source file for the current executable. The macro
33551 information field has a value of @samp{1} or @samp{0} depending on
33552 whether or not the file includes preprocessor macro information.
33554 @subsubheading @value{GDBN} Command
33556 The @value{GDBN} equivalent is @samp{info source}
33558 @subsubheading Example
33562 123-file-list-exec-source-file
33563 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
33568 @subheading The @code{-file-list-exec-source-files} Command
33569 @findex -file-list-exec-source-files
33571 @subsubheading Synopsis
33574 -file-list-exec-source-files
33577 List the source files for the current executable.
33579 It will always output both the filename and fullname (absolute file
33580 name) of a source file.
33582 @subsubheading @value{GDBN} Command
33584 The @value{GDBN} equivalent is @samp{info sources}.
33585 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
33587 @subsubheading Example
33590 -file-list-exec-source-files
33592 @{file=foo.c,fullname=/home/foo.c@},
33593 @{file=/home/bar.c,fullname=/home/bar.c@},
33594 @{file=gdb_could_not_find_fullpath.c@}]
33599 @subheading The @code{-file-list-shared-libraries} Command
33600 @findex -file-list-shared-libraries
33602 @subsubheading Synopsis
33605 -file-list-shared-libraries
33608 List the shared libraries in the program.
33610 @subsubheading @value{GDBN} Command
33612 The corresponding @value{GDBN} command is @samp{info shared}.
33614 @subsubheading Example
33618 @subheading The @code{-file-list-symbol-files} Command
33619 @findex -file-list-symbol-files
33621 @subsubheading Synopsis
33624 -file-list-symbol-files
33629 @subsubheading @value{GDBN} Command
33631 The corresponding @value{GDBN} command is @samp{info file} (part of it).
33633 @subsubheading Example
33638 @subheading The @code{-file-symbol-file} Command
33639 @findex -file-symbol-file
33641 @subsubheading Synopsis
33644 -file-symbol-file @var{file}
33647 Read symbol table info from the specified @var{file} argument. When
33648 used without arguments, clears @value{GDBN}'s symbol table info. No output is
33649 produced, except for a completion notification.
33651 @subsubheading @value{GDBN} Command
33653 The corresponding @value{GDBN} command is @samp{symbol-file}.
33655 @subsubheading Example
33659 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33665 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33666 @node GDB/MI Memory Overlay Commands
33667 @section @sc{gdb/mi} Memory Overlay Commands
33669 The memory overlay commands are not implemented.
33671 @c @subheading -overlay-auto
33673 @c @subheading -overlay-list-mapping-state
33675 @c @subheading -overlay-list-overlays
33677 @c @subheading -overlay-map
33679 @c @subheading -overlay-off
33681 @c @subheading -overlay-on
33683 @c @subheading -overlay-unmap
33685 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33686 @node GDB/MI Signal Handling Commands
33687 @section @sc{gdb/mi} Signal Handling Commands
33689 Signal handling commands are not implemented.
33691 @c @subheading -signal-handle
33693 @c @subheading -signal-list-handle-actions
33695 @c @subheading -signal-list-signal-types
33699 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33700 @node GDB/MI Target Manipulation
33701 @section @sc{gdb/mi} Target Manipulation Commands
33704 @subheading The @code{-target-attach} Command
33705 @findex -target-attach
33707 @subsubheading Synopsis
33710 -target-attach @var{pid} | @var{gid} | @var{file}
33713 Attach to a process @var{pid} or a file @var{file} outside of
33714 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
33715 group, the id previously returned by
33716 @samp{-list-thread-groups --available} must be used.
33718 @subsubheading @value{GDBN} Command
33720 The corresponding @value{GDBN} command is @samp{attach}.
33722 @subsubheading Example
33726 =thread-created,id="1"
33727 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
33733 @subheading The @code{-target-compare-sections} Command
33734 @findex -target-compare-sections
33736 @subsubheading Synopsis
33739 -target-compare-sections [ @var{section} ]
33742 Compare data of section @var{section} on target to the exec file.
33743 Without the argument, all sections are compared.
33745 @subsubheading @value{GDBN} Command
33747 The @value{GDBN} equivalent is @samp{compare-sections}.
33749 @subsubheading Example
33754 @subheading The @code{-target-detach} Command
33755 @findex -target-detach
33757 @subsubheading Synopsis
33760 -target-detach [ @var{pid} | @var{gid} ]
33763 Detach from the remote target which normally resumes its execution.
33764 If either @var{pid} or @var{gid} is specified, detaches from either
33765 the specified process, or specified thread group. There's no output.
33767 @subsubheading @value{GDBN} Command
33769 The corresponding @value{GDBN} command is @samp{detach}.
33771 @subsubheading Example
33781 @subheading The @code{-target-disconnect} Command
33782 @findex -target-disconnect
33784 @subsubheading Synopsis
33790 Disconnect from the remote target. There's no output and the target is
33791 generally not resumed.
33793 @subsubheading @value{GDBN} Command
33795 The corresponding @value{GDBN} command is @samp{disconnect}.
33797 @subsubheading Example
33807 @subheading The @code{-target-download} Command
33808 @findex -target-download
33810 @subsubheading Synopsis
33816 Loads the executable onto the remote target.
33817 It prints out an update message every half second, which includes the fields:
33821 The name of the section.
33823 The size of what has been sent so far for that section.
33825 The size of the section.
33827 The total size of what was sent so far (the current and the previous sections).
33829 The size of the overall executable to download.
33833 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
33834 @sc{gdb/mi} Output Syntax}).
33836 In addition, it prints the name and size of the sections, as they are
33837 downloaded. These messages include the following fields:
33841 The name of the section.
33843 The size of the section.
33845 The size of the overall executable to download.
33849 At the end, a summary is printed.
33851 @subsubheading @value{GDBN} Command
33853 The corresponding @value{GDBN} command is @samp{load}.
33855 @subsubheading Example
33857 Note: each status message appears on a single line. Here the messages
33858 have been broken down so that they can fit onto a page.
33863 +download,@{section=".text",section-size="6668",total-size="9880"@}
33864 +download,@{section=".text",section-sent="512",section-size="6668",
33865 total-sent="512",total-size="9880"@}
33866 +download,@{section=".text",section-sent="1024",section-size="6668",
33867 total-sent="1024",total-size="9880"@}
33868 +download,@{section=".text",section-sent="1536",section-size="6668",
33869 total-sent="1536",total-size="9880"@}
33870 +download,@{section=".text",section-sent="2048",section-size="6668",
33871 total-sent="2048",total-size="9880"@}
33872 +download,@{section=".text",section-sent="2560",section-size="6668",
33873 total-sent="2560",total-size="9880"@}
33874 +download,@{section=".text",section-sent="3072",section-size="6668",
33875 total-sent="3072",total-size="9880"@}
33876 +download,@{section=".text",section-sent="3584",section-size="6668",
33877 total-sent="3584",total-size="9880"@}
33878 +download,@{section=".text",section-sent="4096",section-size="6668",
33879 total-sent="4096",total-size="9880"@}
33880 +download,@{section=".text",section-sent="4608",section-size="6668",
33881 total-sent="4608",total-size="9880"@}
33882 +download,@{section=".text",section-sent="5120",section-size="6668",
33883 total-sent="5120",total-size="9880"@}
33884 +download,@{section=".text",section-sent="5632",section-size="6668",
33885 total-sent="5632",total-size="9880"@}
33886 +download,@{section=".text",section-sent="6144",section-size="6668",
33887 total-sent="6144",total-size="9880"@}
33888 +download,@{section=".text",section-sent="6656",section-size="6668",
33889 total-sent="6656",total-size="9880"@}
33890 +download,@{section=".init",section-size="28",total-size="9880"@}
33891 +download,@{section=".fini",section-size="28",total-size="9880"@}
33892 +download,@{section=".data",section-size="3156",total-size="9880"@}
33893 +download,@{section=".data",section-sent="512",section-size="3156",
33894 total-sent="7236",total-size="9880"@}
33895 +download,@{section=".data",section-sent="1024",section-size="3156",
33896 total-sent="7748",total-size="9880"@}
33897 +download,@{section=".data",section-sent="1536",section-size="3156",
33898 total-sent="8260",total-size="9880"@}
33899 +download,@{section=".data",section-sent="2048",section-size="3156",
33900 total-sent="8772",total-size="9880"@}
33901 +download,@{section=".data",section-sent="2560",section-size="3156",
33902 total-sent="9284",total-size="9880"@}
33903 +download,@{section=".data",section-sent="3072",section-size="3156",
33904 total-sent="9796",total-size="9880"@}
33905 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
33912 @subheading The @code{-target-exec-status} Command
33913 @findex -target-exec-status
33915 @subsubheading Synopsis
33918 -target-exec-status
33921 Provide information on the state of the target (whether it is running or
33922 not, for instance).
33924 @subsubheading @value{GDBN} Command
33926 There's no equivalent @value{GDBN} command.
33928 @subsubheading Example
33932 @subheading The @code{-target-list-available-targets} Command
33933 @findex -target-list-available-targets
33935 @subsubheading Synopsis
33938 -target-list-available-targets
33941 List the possible targets to connect to.
33943 @subsubheading @value{GDBN} Command
33945 The corresponding @value{GDBN} command is @samp{help target}.
33947 @subsubheading Example
33951 @subheading The @code{-target-list-current-targets} Command
33952 @findex -target-list-current-targets
33954 @subsubheading Synopsis
33957 -target-list-current-targets
33960 Describe the current target.
33962 @subsubheading @value{GDBN} Command
33964 The corresponding information is printed by @samp{info file} (among
33967 @subsubheading Example
33971 @subheading The @code{-target-list-parameters} Command
33972 @findex -target-list-parameters
33974 @subsubheading Synopsis
33977 -target-list-parameters
33983 @subsubheading @value{GDBN} Command
33987 @subsubheading Example
33991 @subheading The @code{-target-select} Command
33992 @findex -target-select
33994 @subsubheading Synopsis
33997 -target-select @var{type} @var{parameters @dots{}}
34000 Connect @value{GDBN} to the remote target. This command takes two args:
34004 The type of target, for instance @samp{remote}, etc.
34005 @item @var{parameters}
34006 Device names, host names and the like. @xref{Target Commands, ,
34007 Commands for Managing Targets}, for more details.
34010 The output is a connection notification, followed by the address at
34011 which the target program is, in the following form:
34014 ^connected,addr="@var{address}",func="@var{function name}",
34015 args=[@var{arg list}]
34018 @subsubheading @value{GDBN} Command
34020 The corresponding @value{GDBN} command is @samp{target}.
34022 @subsubheading Example
34026 -target-select remote /dev/ttya
34027 ^connected,addr="0xfe00a300",func="??",args=[]
34031 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34032 @node GDB/MI File Transfer Commands
34033 @section @sc{gdb/mi} File Transfer Commands
34036 @subheading The @code{-target-file-put} Command
34037 @findex -target-file-put
34039 @subsubheading Synopsis
34042 -target-file-put @var{hostfile} @var{targetfile}
34045 Copy file @var{hostfile} from the host system (the machine running
34046 @value{GDBN}) to @var{targetfile} on the target system.
34048 @subsubheading @value{GDBN} Command
34050 The corresponding @value{GDBN} command is @samp{remote put}.
34052 @subsubheading Example
34056 -target-file-put localfile remotefile
34062 @subheading The @code{-target-file-get} Command
34063 @findex -target-file-get
34065 @subsubheading Synopsis
34068 -target-file-get @var{targetfile} @var{hostfile}
34071 Copy file @var{targetfile} from the target system to @var{hostfile}
34072 on the host system.
34074 @subsubheading @value{GDBN} Command
34076 The corresponding @value{GDBN} command is @samp{remote get}.
34078 @subsubheading Example
34082 -target-file-get remotefile localfile
34088 @subheading The @code{-target-file-delete} Command
34089 @findex -target-file-delete
34091 @subsubheading Synopsis
34094 -target-file-delete @var{targetfile}
34097 Delete @var{targetfile} from the target system.
34099 @subsubheading @value{GDBN} Command
34101 The corresponding @value{GDBN} command is @samp{remote delete}.
34103 @subsubheading Example
34107 -target-file-delete remotefile
34113 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34114 @node GDB/MI Miscellaneous Commands
34115 @section Miscellaneous @sc{gdb/mi} Commands
34117 @c @subheading -gdb-complete
34119 @subheading The @code{-gdb-exit} Command
34122 @subsubheading Synopsis
34128 Exit @value{GDBN} immediately.
34130 @subsubheading @value{GDBN} Command
34132 Approximately corresponds to @samp{quit}.
34134 @subsubheading Example
34144 @subheading The @code{-exec-abort} Command
34145 @findex -exec-abort
34147 @subsubheading Synopsis
34153 Kill the inferior running program.
34155 @subsubheading @value{GDBN} Command
34157 The corresponding @value{GDBN} command is @samp{kill}.
34159 @subsubheading Example
34164 @subheading The @code{-gdb-set} Command
34167 @subsubheading Synopsis
34173 Set an internal @value{GDBN} variable.
34174 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
34176 @subsubheading @value{GDBN} Command
34178 The corresponding @value{GDBN} command is @samp{set}.
34180 @subsubheading Example
34190 @subheading The @code{-gdb-show} Command
34193 @subsubheading Synopsis
34199 Show the current value of a @value{GDBN} variable.
34201 @subsubheading @value{GDBN} Command
34203 The corresponding @value{GDBN} command is @samp{show}.
34205 @subsubheading Example
34214 @c @subheading -gdb-source
34217 @subheading The @code{-gdb-version} Command
34218 @findex -gdb-version
34220 @subsubheading Synopsis
34226 Show version information for @value{GDBN}. Used mostly in testing.
34228 @subsubheading @value{GDBN} Command
34230 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
34231 default shows this information when you start an interactive session.
34233 @subsubheading Example
34235 @c This example modifies the actual output from GDB to avoid overfull
34241 ~Copyright 2000 Free Software Foundation, Inc.
34242 ~GDB is free software, covered by the GNU General Public License, and
34243 ~you are welcome to change it and/or distribute copies of it under
34244 ~ certain conditions.
34245 ~Type "show copying" to see the conditions.
34246 ~There is absolutely no warranty for GDB. Type "show warranty" for
34248 ~This GDB was configured as
34249 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
34254 @subheading The @code{-list-features} Command
34255 @findex -list-features
34257 Returns a list of particular features of the MI protocol that
34258 this version of gdb implements. A feature can be a command,
34259 or a new field in an output of some command, or even an
34260 important bugfix. While a frontend can sometimes detect presence
34261 of a feature at runtime, it is easier to perform detection at debugger
34264 The command returns a list of strings, with each string naming an
34265 available feature. Each returned string is just a name, it does not
34266 have any internal structure. The list of possible feature names
34272 (gdb) -list-features
34273 ^done,result=["feature1","feature2"]
34276 The current list of features is:
34279 @item frozen-varobjs
34280 Indicates support for the @code{-var-set-frozen} command, as well
34281 as possible presense of the @code{frozen} field in the output
34282 of @code{-varobj-create}.
34283 @item pending-breakpoints
34284 Indicates support for the @option{-f} option to the @code{-break-insert}
34287 Indicates Python scripting support, Python-based
34288 pretty-printing commands, and possible presence of the
34289 @samp{display_hint} field in the output of @code{-var-list-children}
34291 Indicates support for the @code{-thread-info} command.
34292 @item data-read-memory-bytes
34293 Indicates support for the @code{-data-read-memory-bytes} and the
34294 @code{-data-write-memory-bytes} commands.
34295 @item breakpoint-notifications
34296 Indicates that changes to breakpoints and breakpoints created via the
34297 CLI will be announced via async records.
34298 @item ada-task-info
34299 Indicates support for the @code{-ada-task-info} command.
34302 @subheading The @code{-list-target-features} Command
34303 @findex -list-target-features
34305 Returns a list of particular features that are supported by the
34306 target. Those features affect the permitted MI commands, but
34307 unlike the features reported by the @code{-list-features} command, the
34308 features depend on which target GDB is using at the moment. Whenever
34309 a target can change, due to commands such as @code{-target-select},
34310 @code{-target-attach} or @code{-exec-run}, the list of target features
34311 may change, and the frontend should obtain it again.
34315 (gdb) -list-features
34316 ^done,result=["async"]
34319 The current list of features is:
34323 Indicates that the target is capable of asynchronous command
34324 execution, which means that @value{GDBN} will accept further commands
34325 while the target is running.
34328 Indicates that the target is capable of reverse execution.
34329 @xref{Reverse Execution}, for more information.
34333 @subheading The @code{-list-thread-groups} Command
34334 @findex -list-thread-groups
34336 @subheading Synopsis
34339 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
34342 Lists thread groups (@pxref{Thread groups}). When a single thread
34343 group is passed as the argument, lists the children of that group.
34344 When several thread group are passed, lists information about those
34345 thread groups. Without any parameters, lists information about all
34346 top-level thread groups.
34348 Normally, thread groups that are being debugged are reported.
34349 With the @samp{--available} option, @value{GDBN} reports thread groups
34350 available on the target.
34352 The output of this command may have either a @samp{threads} result or
34353 a @samp{groups} result. The @samp{thread} result has a list of tuples
34354 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
34355 Information}). The @samp{groups} result has a list of tuples as value,
34356 each tuple describing a thread group. If top-level groups are
34357 requested (that is, no parameter is passed), or when several groups
34358 are passed, the output always has a @samp{groups} result. The format
34359 of the @samp{group} result is described below.
34361 To reduce the number of roundtrips it's possible to list thread groups
34362 together with their children, by passing the @samp{--recurse} option
34363 and the recursion depth. Presently, only recursion depth of 1 is
34364 permitted. If this option is present, then every reported thread group
34365 will also include its children, either as @samp{group} or
34366 @samp{threads} field.
34368 In general, any combination of option and parameters is permitted, with
34369 the following caveats:
34373 When a single thread group is passed, the output will typically
34374 be the @samp{threads} result. Because threads may not contain
34375 anything, the @samp{recurse} option will be ignored.
34378 When the @samp{--available} option is passed, limited information may
34379 be available. In particular, the list of threads of a process might
34380 be inaccessible. Further, specifying specific thread groups might
34381 not give any performance advantage over listing all thread groups.
34382 The frontend should assume that @samp{-list-thread-groups --available}
34383 is always an expensive operation and cache the results.
34387 The @samp{groups} result is a list of tuples, where each tuple may
34388 have the following fields:
34392 Identifier of the thread group. This field is always present.
34393 The identifier is an opaque string; frontends should not try to
34394 convert it to an integer, even though it might look like one.
34397 The type of the thread group. At present, only @samp{process} is a
34401 The target-specific process identifier. This field is only present
34402 for thread groups of type @samp{process} and only if the process exists.
34405 The number of children this thread group has. This field may be
34406 absent for an available thread group.
34409 This field has a list of tuples as value, each tuple describing a
34410 thread. It may be present if the @samp{--recurse} option is
34411 specified, and it's actually possible to obtain the threads.
34414 This field is a list of integers, each identifying a core that one
34415 thread of the group is running on. This field may be absent if
34416 such information is not available.
34419 The name of the executable file that corresponds to this thread group.
34420 The field is only present for thread groups of type @samp{process},
34421 and only if there is a corresponding executable file.
34425 @subheading Example
34429 -list-thread-groups
34430 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
34431 -list-thread-groups 17
34432 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
34433 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
34434 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
34435 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
34436 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
34437 -list-thread-groups --available
34438 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
34439 -list-thread-groups --available --recurse 1
34440 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34441 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34442 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
34443 -list-thread-groups --available --recurse 1 17 18
34444 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34445 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34446 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
34449 @subheading The @code{-info-os} Command
34452 @subsubheading Synopsis
34455 -info-os [ @var{type} ]
34458 If no argument is supplied, the command returns a table of available
34459 operating-system-specific information types. If one of these types is
34460 supplied as an argument @var{type}, then the command returns a table
34461 of data of that type.
34463 The types of information available depend on the target operating
34466 @subsubheading @value{GDBN} Command
34468 The corresponding @value{GDBN} command is @samp{info os}.
34470 @subsubheading Example
34472 When run on a @sc{gnu}/Linux system, the output will look something
34478 ^done,OSDataTable=@{nr_rows="9",nr_cols="3",
34479 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
34480 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
34481 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
34482 body=[item=@{col0="processes",col1="Listing of all processes",
34483 col2="Processes"@},
34484 item=@{col0="procgroups",col1="Listing of all process groups",
34485 col2="Process groups"@},
34486 item=@{col0="threads",col1="Listing of all threads",
34488 item=@{col0="files",col1="Listing of all file descriptors",
34489 col2="File descriptors"@},
34490 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
34492 item=@{col0="shm",col1="Listing of all shared-memory regions",
34493 col2="Shared-memory regions"@},
34494 item=@{col0="semaphores",col1="Listing of all semaphores",
34495 col2="Semaphores"@},
34496 item=@{col0="msg",col1="Listing of all message queues",
34497 col2="Message queues"@},
34498 item=@{col0="modules",col1="Listing of all loaded kernel modules",
34499 col2="Kernel modules"@}]@}
34502 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
34503 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
34504 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
34505 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
34506 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
34507 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
34508 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
34509 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
34511 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
34512 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
34516 (Note that the MI output here includes a @code{"Title"} column that
34517 does not appear in command-line @code{info os}; this column is useful
34518 for MI clients that want to enumerate the types of data, such as in a
34519 popup menu, but is needless clutter on the command line, and
34520 @code{info os} omits it.)
34522 @subheading The @code{-add-inferior} Command
34523 @findex -add-inferior
34525 @subheading Synopsis
34531 Creates a new inferior (@pxref{Inferiors and Programs}). The created
34532 inferior is not associated with any executable. Such association may
34533 be established with the @samp{-file-exec-and-symbols} command
34534 (@pxref{GDB/MI File Commands}). The command response has a single
34535 field, @samp{thread-group}, whose value is the identifier of the
34536 thread group corresponding to the new inferior.
34538 @subheading Example
34543 ^done,thread-group="i3"
34546 @subheading The @code{-interpreter-exec} Command
34547 @findex -interpreter-exec
34549 @subheading Synopsis
34552 -interpreter-exec @var{interpreter} @var{command}
34554 @anchor{-interpreter-exec}
34556 Execute the specified @var{command} in the given @var{interpreter}.
34558 @subheading @value{GDBN} Command
34560 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
34562 @subheading Example
34566 -interpreter-exec console "break main"
34567 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
34568 &"During symbol reading, bad structure-type format.\n"
34569 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
34574 @subheading The @code{-inferior-tty-set} Command
34575 @findex -inferior-tty-set
34577 @subheading Synopsis
34580 -inferior-tty-set /dev/pts/1
34583 Set terminal for future runs of the program being debugged.
34585 @subheading @value{GDBN} Command
34587 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
34589 @subheading Example
34593 -inferior-tty-set /dev/pts/1
34598 @subheading The @code{-inferior-tty-show} Command
34599 @findex -inferior-tty-show
34601 @subheading Synopsis
34607 Show terminal for future runs of program being debugged.
34609 @subheading @value{GDBN} Command
34611 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
34613 @subheading Example
34617 -inferior-tty-set /dev/pts/1
34621 ^done,inferior_tty_terminal="/dev/pts/1"
34625 @subheading The @code{-enable-timings} Command
34626 @findex -enable-timings
34628 @subheading Synopsis
34631 -enable-timings [yes | no]
34634 Toggle the printing of the wallclock, user and system times for an MI
34635 command as a field in its output. This command is to help frontend
34636 developers optimize the performance of their code. No argument is
34637 equivalent to @samp{yes}.
34639 @subheading @value{GDBN} Command
34643 @subheading Example
34651 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
34652 addr="0x080484ed",func="main",file="myprog.c",
34653 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
34655 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
34663 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
34664 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
34665 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
34666 fullname="/home/nickrob/myprog.c",line="73"@}
34671 @chapter @value{GDBN} Annotations
34673 This chapter describes annotations in @value{GDBN}. Annotations were
34674 designed to interface @value{GDBN} to graphical user interfaces or other
34675 similar programs which want to interact with @value{GDBN} at a
34676 relatively high level.
34678 The annotation mechanism has largely been superseded by @sc{gdb/mi}
34682 This is Edition @value{EDITION}, @value{DATE}.
34686 * Annotations Overview:: What annotations are; the general syntax.
34687 * Server Prefix:: Issuing a command without affecting user state.
34688 * Prompting:: Annotations marking @value{GDBN}'s need for input.
34689 * Errors:: Annotations for error messages.
34690 * Invalidation:: Some annotations describe things now invalid.
34691 * Annotations for Running::
34692 Whether the program is running, how it stopped, etc.
34693 * Source Annotations:: Annotations describing source code.
34696 @node Annotations Overview
34697 @section What is an Annotation?
34698 @cindex annotations
34700 Annotations start with a newline character, two @samp{control-z}
34701 characters, and the name of the annotation. If there is no additional
34702 information associated with this annotation, the name of the annotation
34703 is followed immediately by a newline. If there is additional
34704 information, the name of the annotation is followed by a space, the
34705 additional information, and a newline. The additional information
34706 cannot contain newline characters.
34708 Any output not beginning with a newline and two @samp{control-z}
34709 characters denotes literal output from @value{GDBN}. Currently there is
34710 no need for @value{GDBN} to output a newline followed by two
34711 @samp{control-z} characters, but if there was such a need, the
34712 annotations could be extended with an @samp{escape} annotation which
34713 means those three characters as output.
34715 The annotation @var{level}, which is specified using the
34716 @option{--annotate} command line option (@pxref{Mode Options}), controls
34717 how much information @value{GDBN} prints together with its prompt,
34718 values of expressions, source lines, and other types of output. Level 0
34719 is for no annotations, level 1 is for use when @value{GDBN} is run as a
34720 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
34721 for programs that control @value{GDBN}, and level 2 annotations have
34722 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
34723 Interface, annotate, GDB's Obsolete Annotations}).
34726 @kindex set annotate
34727 @item set annotate @var{level}
34728 The @value{GDBN} command @code{set annotate} sets the level of
34729 annotations to the specified @var{level}.
34731 @item show annotate
34732 @kindex show annotate
34733 Show the current annotation level.
34736 This chapter describes level 3 annotations.
34738 A simple example of starting up @value{GDBN} with annotations is:
34741 $ @kbd{gdb --annotate=3}
34743 Copyright 2003 Free Software Foundation, Inc.
34744 GDB is free software, covered by the GNU General Public License,
34745 and you are welcome to change it and/or distribute copies of it
34746 under certain conditions.
34747 Type "show copying" to see the conditions.
34748 There is absolutely no warranty for GDB. Type "show warranty"
34750 This GDB was configured as "i386-pc-linux-gnu"
34761 Here @samp{quit} is input to @value{GDBN}; the rest is output from
34762 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
34763 denotes a @samp{control-z} character) are annotations; the rest is
34764 output from @value{GDBN}.
34766 @node Server Prefix
34767 @section The Server Prefix
34768 @cindex server prefix
34770 If you prefix a command with @samp{server } then it will not affect
34771 the command history, nor will it affect @value{GDBN}'s notion of which
34772 command to repeat if @key{RET} is pressed on a line by itself. This
34773 means that commands can be run behind a user's back by a front-end in
34774 a transparent manner.
34776 The @code{server } prefix does not affect the recording of values into
34777 the value history; to print a value without recording it into the
34778 value history, use the @code{output} command instead of the
34779 @code{print} command.
34781 Using this prefix also disables confirmation requests
34782 (@pxref{confirmation requests}).
34785 @section Annotation for @value{GDBN} Input
34787 @cindex annotations for prompts
34788 When @value{GDBN} prompts for input, it annotates this fact so it is possible
34789 to know when to send output, when the output from a given command is
34792 Different kinds of input each have a different @dfn{input type}. Each
34793 input type has three annotations: a @code{pre-} annotation, which
34794 denotes the beginning of any prompt which is being output, a plain
34795 annotation, which denotes the end of the prompt, and then a @code{post-}
34796 annotation which denotes the end of any echo which may (or may not) be
34797 associated with the input. For example, the @code{prompt} input type
34798 features the following annotations:
34806 The input types are
34809 @findex pre-prompt annotation
34810 @findex prompt annotation
34811 @findex post-prompt annotation
34813 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
34815 @findex pre-commands annotation
34816 @findex commands annotation
34817 @findex post-commands annotation
34819 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
34820 command. The annotations are repeated for each command which is input.
34822 @findex pre-overload-choice annotation
34823 @findex overload-choice annotation
34824 @findex post-overload-choice annotation
34825 @item overload-choice
34826 When @value{GDBN} wants the user to select between various overloaded functions.
34828 @findex pre-query annotation
34829 @findex query annotation
34830 @findex post-query annotation
34832 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
34834 @findex pre-prompt-for-continue annotation
34835 @findex prompt-for-continue annotation
34836 @findex post-prompt-for-continue annotation
34837 @item prompt-for-continue
34838 When @value{GDBN} is asking the user to press return to continue. Note: Don't
34839 expect this to work well; instead use @code{set height 0} to disable
34840 prompting. This is because the counting of lines is buggy in the
34841 presence of annotations.
34846 @cindex annotations for errors, warnings and interrupts
34848 @findex quit annotation
34853 This annotation occurs right before @value{GDBN} responds to an interrupt.
34855 @findex error annotation
34860 This annotation occurs right before @value{GDBN} responds to an error.
34862 Quit and error annotations indicate that any annotations which @value{GDBN} was
34863 in the middle of may end abruptly. For example, if a
34864 @code{value-history-begin} annotation is followed by a @code{error}, one
34865 cannot expect to receive the matching @code{value-history-end}. One
34866 cannot expect not to receive it either, however; an error annotation
34867 does not necessarily mean that @value{GDBN} is immediately returning all the way
34870 @findex error-begin annotation
34871 A quit or error annotation may be preceded by
34877 Any output between that and the quit or error annotation is the error
34880 Warning messages are not yet annotated.
34881 @c If we want to change that, need to fix warning(), type_error(),
34882 @c range_error(), and possibly other places.
34885 @section Invalidation Notices
34887 @cindex annotations for invalidation messages
34888 The following annotations say that certain pieces of state may have
34892 @findex frames-invalid annotation
34893 @item ^Z^Zframes-invalid
34895 The frames (for example, output from the @code{backtrace} command) may
34898 @findex breakpoints-invalid annotation
34899 @item ^Z^Zbreakpoints-invalid
34901 The breakpoints may have changed. For example, the user just added or
34902 deleted a breakpoint.
34905 @node Annotations for Running
34906 @section Running the Program
34907 @cindex annotations for running programs
34909 @findex starting annotation
34910 @findex stopping annotation
34911 When the program starts executing due to a @value{GDBN} command such as
34912 @code{step} or @code{continue},
34918 is output. When the program stops,
34924 is output. Before the @code{stopped} annotation, a variety of
34925 annotations describe how the program stopped.
34928 @findex exited annotation
34929 @item ^Z^Zexited @var{exit-status}
34930 The program exited, and @var{exit-status} is the exit status (zero for
34931 successful exit, otherwise nonzero).
34933 @findex signalled annotation
34934 @findex signal-name annotation
34935 @findex signal-name-end annotation
34936 @findex signal-string annotation
34937 @findex signal-string-end annotation
34938 @item ^Z^Zsignalled
34939 The program exited with a signal. After the @code{^Z^Zsignalled}, the
34940 annotation continues:
34946 ^Z^Zsignal-name-end
34950 ^Z^Zsignal-string-end
34955 where @var{name} is the name of the signal, such as @code{SIGILL} or
34956 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
34957 as @code{Illegal Instruction} or @code{Segmentation fault}.
34958 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
34959 user's benefit and have no particular format.
34961 @findex signal annotation
34963 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
34964 just saying that the program received the signal, not that it was
34965 terminated with it.
34967 @findex breakpoint annotation
34968 @item ^Z^Zbreakpoint @var{number}
34969 The program hit breakpoint number @var{number}.
34971 @findex watchpoint annotation
34972 @item ^Z^Zwatchpoint @var{number}
34973 The program hit watchpoint number @var{number}.
34976 @node Source Annotations
34977 @section Displaying Source
34978 @cindex annotations for source display
34980 @findex source annotation
34981 The following annotation is used instead of displaying source code:
34984 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
34987 where @var{filename} is an absolute file name indicating which source
34988 file, @var{line} is the line number within that file (where 1 is the
34989 first line in the file), @var{character} is the character position
34990 within the file (where 0 is the first character in the file) (for most
34991 debug formats this will necessarily point to the beginning of a line),
34992 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
34993 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
34994 @var{addr} is the address in the target program associated with the
34995 source which is being displayed. @var{addr} is in the form @samp{0x}
34996 followed by one or more lowercase hex digits (note that this does not
34997 depend on the language).
34999 @node JIT Interface
35000 @chapter JIT Compilation Interface
35001 @cindex just-in-time compilation
35002 @cindex JIT compilation interface
35004 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
35005 interface. A JIT compiler is a program or library that generates native
35006 executable code at runtime and executes it, usually in order to achieve good
35007 performance while maintaining platform independence.
35009 Programs that use JIT compilation are normally difficult to debug because
35010 portions of their code are generated at runtime, instead of being loaded from
35011 object files, which is where @value{GDBN} normally finds the program's symbols
35012 and debug information. In order to debug programs that use JIT compilation,
35013 @value{GDBN} has an interface that allows the program to register in-memory
35014 symbol files with @value{GDBN} at runtime.
35016 If you are using @value{GDBN} to debug a program that uses this interface, then
35017 it should work transparently so long as you have not stripped the binary. If
35018 you are developing a JIT compiler, then the interface is documented in the rest
35019 of this chapter. At this time, the only known client of this interface is the
35022 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
35023 JIT compiler communicates with @value{GDBN} by writing data into a global
35024 variable and calling a fuction at a well-known symbol. When @value{GDBN}
35025 attaches, it reads a linked list of symbol files from the global variable to
35026 find existing code, and puts a breakpoint in the function so that it can find
35027 out about additional code.
35030 * Declarations:: Relevant C struct declarations
35031 * Registering Code:: Steps to register code
35032 * Unregistering Code:: Steps to unregister code
35033 * Custom Debug Info:: Emit debug information in a custom format
35037 @section JIT Declarations
35039 These are the relevant struct declarations that a C program should include to
35040 implement the interface:
35050 struct jit_code_entry
35052 struct jit_code_entry *next_entry;
35053 struct jit_code_entry *prev_entry;
35054 const char *symfile_addr;
35055 uint64_t symfile_size;
35058 struct jit_descriptor
35061 /* This type should be jit_actions_t, but we use uint32_t
35062 to be explicit about the bitwidth. */
35063 uint32_t action_flag;
35064 struct jit_code_entry *relevant_entry;
35065 struct jit_code_entry *first_entry;
35068 /* GDB puts a breakpoint in this function. */
35069 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
35071 /* Make sure to specify the version statically, because the
35072 debugger may check the version before we can set it. */
35073 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
35076 If the JIT is multi-threaded, then it is important that the JIT synchronize any
35077 modifications to this global data properly, which can easily be done by putting
35078 a global mutex around modifications to these structures.
35080 @node Registering Code
35081 @section Registering Code
35083 To register code with @value{GDBN}, the JIT should follow this protocol:
35087 Generate an object file in memory with symbols and other desired debug
35088 information. The file must include the virtual addresses of the sections.
35091 Create a code entry for the file, which gives the start and size of the symbol
35095 Add it to the linked list in the JIT descriptor.
35098 Point the relevant_entry field of the descriptor at the entry.
35101 Set @code{action_flag} to @code{JIT_REGISTER} and call
35102 @code{__jit_debug_register_code}.
35105 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
35106 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
35107 new code. However, the linked list must still be maintained in order to allow
35108 @value{GDBN} to attach to a running process and still find the symbol files.
35110 @node Unregistering Code
35111 @section Unregistering Code
35113 If code is freed, then the JIT should use the following protocol:
35117 Remove the code entry corresponding to the code from the linked list.
35120 Point the @code{relevant_entry} field of the descriptor at the code entry.
35123 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
35124 @code{__jit_debug_register_code}.
35127 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
35128 and the JIT will leak the memory used for the associated symbol files.
35130 @node Custom Debug Info
35131 @section Custom Debug Info
35132 @cindex custom JIT debug info
35133 @cindex JIT debug info reader
35135 Generating debug information in platform-native file formats (like ELF
35136 or COFF) may be an overkill for JIT compilers; especially if all the
35137 debug info is used for is displaying a meaningful backtrace. The
35138 issue can be resolved by having the JIT writers decide on a debug info
35139 format and also provide a reader that parses the debug info generated
35140 by the JIT compiler. This section gives a brief overview on writing
35141 such a parser. More specific details can be found in the source file
35142 @file{gdb/jit-reader.in}, which is also installed as a header at
35143 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
35145 The reader is implemented as a shared object (so this functionality is
35146 not available on platforms which don't allow loading shared objects at
35147 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
35148 @code{jit-reader-unload} are provided, to be used to load and unload
35149 the readers from a preconfigured directory. Once loaded, the shared
35150 object is used the parse the debug information emitted by the JIT
35154 * Using JIT Debug Info Readers:: How to use supplied readers correctly
35155 * Writing JIT Debug Info Readers:: Creating a debug-info reader
35158 @node Using JIT Debug Info Readers
35159 @subsection Using JIT Debug Info Readers
35160 @kindex jit-reader-load
35161 @kindex jit-reader-unload
35163 Readers can be loaded and unloaded using the @code{jit-reader-load}
35164 and @code{jit-reader-unload} commands.
35167 @item jit-reader-load @var{reader}
35168 Load the JIT reader named @var{reader}. @var{reader} is a shared
35169 object specified as either an absolute or a relative file name. In
35170 the latter case, @value{GDBN} will try to load the reader from a
35171 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
35172 system (here @var{libdir} is the system library directory, often
35173 @file{/usr/local/lib}).
35175 Only one reader can be active at a time; trying to load a second
35176 reader when one is already loaded will result in @value{GDBN}
35177 reporting an error. A new JIT reader can be loaded by first unloading
35178 the current one using @code{jit-reader-unload} and then invoking
35179 @code{jit-reader-load}.
35181 @item jit-reader-unload
35182 Unload the currently loaded JIT reader.
35186 @node Writing JIT Debug Info Readers
35187 @subsection Writing JIT Debug Info Readers
35188 @cindex writing JIT debug info readers
35190 As mentioned, a reader is essentially a shared object conforming to a
35191 certain ABI. This ABI is described in @file{jit-reader.h}.
35193 @file{jit-reader.h} defines the structures, macros and functions
35194 required to write a reader. It is installed (along with
35195 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
35196 the system include directory.
35198 Readers need to be released under a GPL compatible license. A reader
35199 can be declared as released under such a license by placing the macro
35200 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
35202 The entry point for readers is the symbol @code{gdb_init_reader},
35203 which is expected to be a function with the prototype
35205 @findex gdb_init_reader
35207 extern struct gdb_reader_funcs *gdb_init_reader (void);
35210 @cindex @code{struct gdb_reader_funcs}
35212 @code{struct gdb_reader_funcs} contains a set of pointers to callback
35213 functions. These functions are executed to read the debug info
35214 generated by the JIT compiler (@code{read}), to unwind stack frames
35215 (@code{unwind}) and to create canonical frame IDs
35216 (@code{get_Frame_id}). It also has a callback that is called when the
35217 reader is being unloaded (@code{destroy}). The struct looks like this
35220 struct gdb_reader_funcs
35222 /* Must be set to GDB_READER_INTERFACE_VERSION. */
35223 int reader_version;
35225 /* For use by the reader. */
35228 gdb_read_debug_info *read;
35229 gdb_unwind_frame *unwind;
35230 gdb_get_frame_id *get_frame_id;
35231 gdb_destroy_reader *destroy;
35235 @cindex @code{struct gdb_symbol_callbacks}
35236 @cindex @code{struct gdb_unwind_callbacks}
35238 The callbacks are provided with another set of callbacks by
35239 @value{GDBN} to do their job. For @code{read}, these callbacks are
35240 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
35241 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
35242 @code{struct gdb_symbol_callbacks} has callbacks to create new object
35243 files and new symbol tables inside those object files. @code{struct
35244 gdb_unwind_callbacks} has callbacks to read registers off the current
35245 frame and to write out the values of the registers in the previous
35246 frame. Both have a callback (@code{target_read}) to read bytes off the
35247 target's address space.
35249 @node In-Process Agent
35250 @chapter In-Process Agent
35251 @cindex debugging agent
35252 The traditional debugging model is conceptually low-speed, but works fine,
35253 because most bugs can be reproduced in debugging-mode execution. However,
35254 as multi-core or many-core processors are becoming mainstream, and
35255 multi-threaded programs become more and more popular, there should be more
35256 and more bugs that only manifest themselves at normal-mode execution, for
35257 example, thread races, because debugger's interference with the program's
35258 timing may conceal the bugs. On the other hand, in some applications,
35259 it is not feasible for the debugger to interrupt the program's execution
35260 long enough for the developer to learn anything helpful about its behavior.
35261 If the program's correctness depends on its real-time behavior, delays
35262 introduced by a debugger might cause the program to fail, even when the
35263 code itself is correct. It is useful to be able to observe the program's
35264 behavior without interrupting it.
35266 Therefore, traditional debugging model is too intrusive to reproduce
35267 some bugs. In order to reduce the interference with the program, we can
35268 reduce the number of operations performed by debugger. The
35269 @dfn{In-Process Agent}, a shared library, is running within the same
35270 process with inferior, and is able to perform some debugging operations
35271 itself. As a result, debugger is only involved when necessary, and
35272 performance of debugging can be improved accordingly. Note that
35273 interference with program can be reduced but can't be removed completely,
35274 because the in-process agent will still stop or slow down the program.
35276 The in-process agent can interpret and execute Agent Expressions
35277 (@pxref{Agent Expressions}) during performing debugging operations. The
35278 agent expressions can be used for different purposes, such as collecting
35279 data in tracepoints, and condition evaluation in breakpoints.
35281 @anchor{Control Agent}
35282 You can control whether the in-process agent is used as an aid for
35283 debugging with the following commands:
35286 @kindex set agent on
35288 Causes the in-process agent to perform some operations on behalf of the
35289 debugger. Just which operations requested by the user will be done
35290 by the in-process agent depends on the its capabilities. For example,
35291 if you request to evaluate breakpoint conditions in the in-process agent,
35292 and the in-process agent has such capability as well, then breakpoint
35293 conditions will be evaluated in the in-process agent.
35295 @kindex set agent off
35296 @item set agent off
35297 Disables execution of debugging operations by the in-process agent. All
35298 of the operations will be performed by @value{GDBN}.
35302 Display the current setting of execution of debugging operations by
35303 the in-process agent.
35307 * In-Process Agent Protocol::
35310 @node In-Process Agent Protocol
35311 @section In-Process Agent Protocol
35312 @cindex in-process agent protocol
35314 The in-process agent is able to communicate with both @value{GDBN} and
35315 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
35316 used for communications between @value{GDBN} or GDBserver and the IPA.
35317 In general, @value{GDBN} or GDBserver sends commands
35318 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
35319 in-process agent replies back with the return result of the command, or
35320 some other information. The data sent to in-process agent is composed
35321 of primitive data types, such as 4-byte or 8-byte type, and composite
35322 types, which are called objects (@pxref{IPA Protocol Objects}).
35325 * IPA Protocol Objects::
35326 * IPA Protocol Commands::
35329 @node IPA Protocol Objects
35330 @subsection IPA Protocol Objects
35331 @cindex ipa protocol objects
35333 The commands sent to and results received from agent may contain some
35334 complex data types called @dfn{objects}.
35336 The in-process agent is running on the same machine with @value{GDBN}
35337 or GDBserver, so it doesn't have to handle as much differences between
35338 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
35339 However, there are still some differences of two ends in two processes:
35343 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
35344 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
35346 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
35347 GDBserver is compiled with one, and in-process agent is compiled with
35351 Here are the IPA Protocol Objects:
35355 agent expression object. It represents an agent expression
35356 (@pxref{Agent Expressions}).
35357 @anchor{agent expression object}
35359 tracepoint action object. It represents a tracepoint action
35360 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
35361 memory, static trace data and to evaluate expression.
35362 @anchor{tracepoint action object}
35364 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
35365 @anchor{tracepoint object}
35369 The following table describes important attributes of each IPA protocol
35372 @multitable @columnfractions .30 .20 .50
35373 @headitem Name @tab Size @tab Description
35374 @item @emph{agent expression object} @tab @tab
35375 @item length @tab 4 @tab length of bytes code
35376 @item byte code @tab @var{length} @tab contents of byte code
35377 @item @emph{tracepoint action for collecting memory} @tab @tab
35378 @item 'M' @tab 1 @tab type of tracepoint action
35379 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
35380 address of the lowest byte to collect, otherwise @var{addr} is the offset
35381 of @var{basereg} for memory collecting.
35382 @item len @tab 8 @tab length of memory for collecting
35383 @item basereg @tab 4 @tab the register number containing the starting
35384 memory address for collecting.
35385 @item @emph{tracepoint action for collecting registers} @tab @tab
35386 @item 'R' @tab 1 @tab type of tracepoint action
35387 @item @emph{tracepoint action for collecting static trace data} @tab @tab
35388 @item 'L' @tab 1 @tab type of tracepoint action
35389 @item @emph{tracepoint action for expression evaluation} @tab @tab
35390 @item 'X' @tab 1 @tab type of tracepoint action
35391 @item agent expression @tab length of @tab @ref{agent expression object}
35392 @item @emph{tracepoint object} @tab @tab
35393 @item number @tab 4 @tab number of tracepoint
35394 @item address @tab 8 @tab address of tracepoint inserted on
35395 @item type @tab 4 @tab type of tracepoint
35396 @item enabled @tab 1 @tab enable or disable of tracepoint
35397 @item step_count @tab 8 @tab step
35398 @item pass_count @tab 8 @tab pass
35399 @item numactions @tab 4 @tab number of tracepoint actions
35400 @item hit count @tab 8 @tab hit count
35401 @item trace frame usage @tab 8 @tab trace frame usage
35402 @item compiled_cond @tab 8 @tab compiled condition
35403 @item orig_size @tab 8 @tab orig size
35404 @item condition @tab 4 if condition is NULL otherwise length of
35405 @ref{agent expression object}
35406 @tab zero if condition is NULL, otherwise is
35407 @ref{agent expression object}
35408 @item actions @tab variable
35409 @tab numactions number of @ref{tracepoint action object}
35412 @node IPA Protocol Commands
35413 @subsection IPA Protocol Commands
35414 @cindex ipa protocol commands
35416 The spaces in each command are delimiters to ease reading this commands
35417 specification. They don't exist in real commands.
35421 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
35422 Installs a new fast tracepoint described by @var{tracepoint_object}
35423 (@pxref{tracepoint object}). @var{gdb_jump_pad_head}, 8-byte long, is the
35424 head of @dfn{jumppad}, which is used to jump to data collection routine
35429 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
35430 @var{target_address} is address of tracepoint in the inferior.
35431 @var{gdb_jump_pad_head} is updated head of jumppad. Both of
35432 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
35433 @var{fjump} contains a sequence of instructions jump to jumppad entry.
35434 @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
35441 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
35442 is about to kill inferiors.
35450 @item probe_marker_at:@var{address}
35451 Asks in-process agent to probe the marker at @var{address}.
35458 @item unprobe_marker_at:@var{address}
35459 Asks in-process agent to unprobe the marker at @var{address}.
35463 @chapter Reporting Bugs in @value{GDBN}
35464 @cindex bugs in @value{GDBN}
35465 @cindex reporting bugs in @value{GDBN}
35467 Your bug reports play an essential role in making @value{GDBN} reliable.
35469 Reporting a bug may help you by bringing a solution to your problem, or it
35470 may not. But in any case the principal function of a bug report is to help
35471 the entire community by making the next version of @value{GDBN} work better. Bug
35472 reports are your contribution to the maintenance of @value{GDBN}.
35474 In order for a bug report to serve its purpose, you must include the
35475 information that enables us to fix the bug.
35478 * Bug Criteria:: Have you found a bug?
35479 * Bug Reporting:: How to report bugs
35483 @section Have You Found a Bug?
35484 @cindex bug criteria
35486 If you are not sure whether you have found a bug, here are some guidelines:
35489 @cindex fatal signal
35490 @cindex debugger crash
35491 @cindex crash of debugger
35493 If the debugger gets a fatal signal, for any input whatever, that is a
35494 @value{GDBN} bug. Reliable debuggers never crash.
35496 @cindex error on valid input
35498 If @value{GDBN} produces an error message for valid input, that is a
35499 bug. (Note that if you're cross debugging, the problem may also be
35500 somewhere in the connection to the target.)
35502 @cindex invalid input
35504 If @value{GDBN} does not produce an error message for invalid input,
35505 that is a bug. However, you should note that your idea of
35506 ``invalid input'' might be our idea of ``an extension'' or ``support
35507 for traditional practice''.
35510 If you are an experienced user of debugging tools, your suggestions
35511 for improvement of @value{GDBN} are welcome in any case.
35514 @node Bug Reporting
35515 @section How to Report Bugs
35516 @cindex bug reports
35517 @cindex @value{GDBN} bugs, reporting
35519 A number of companies and individuals offer support for @sc{gnu} products.
35520 If you obtained @value{GDBN} from a support organization, we recommend you
35521 contact that organization first.
35523 You can find contact information for many support companies and
35524 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
35526 @c should add a web page ref...
35529 @ifset BUGURL_DEFAULT
35530 In any event, we also recommend that you submit bug reports for
35531 @value{GDBN}. The preferred method is to submit them directly using
35532 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
35533 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
35536 @strong{Do not send bug reports to @samp{info-gdb}, or to
35537 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
35538 not want to receive bug reports. Those that do have arranged to receive
35541 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
35542 serves as a repeater. The mailing list and the newsgroup carry exactly
35543 the same messages. Often people think of posting bug reports to the
35544 newsgroup instead of mailing them. This appears to work, but it has one
35545 problem which can be crucial: a newsgroup posting often lacks a mail
35546 path back to the sender. Thus, if we need to ask for more information,
35547 we may be unable to reach you. For this reason, it is better to send
35548 bug reports to the mailing list.
35550 @ifclear BUGURL_DEFAULT
35551 In any event, we also recommend that you submit bug reports for
35552 @value{GDBN} to @value{BUGURL}.
35556 The fundamental principle of reporting bugs usefully is this:
35557 @strong{report all the facts}. If you are not sure whether to state a
35558 fact or leave it out, state it!
35560 Often people omit facts because they think they know what causes the
35561 problem and assume that some details do not matter. Thus, you might
35562 assume that the name of the variable you use in an example does not matter.
35563 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
35564 stray memory reference which happens to fetch from the location where that
35565 name is stored in memory; perhaps, if the name were different, the contents
35566 of that location would fool the debugger into doing the right thing despite
35567 the bug. Play it safe and give a specific, complete example. That is the
35568 easiest thing for you to do, and the most helpful.
35570 Keep in mind that the purpose of a bug report is to enable us to fix the
35571 bug. It may be that the bug has been reported previously, but neither
35572 you nor we can know that unless your bug report is complete and
35575 Sometimes people give a few sketchy facts and ask, ``Does this ring a
35576 bell?'' Those bug reports are useless, and we urge everyone to
35577 @emph{refuse to respond to them} except to chide the sender to report
35580 To enable us to fix the bug, you should include all these things:
35584 The version of @value{GDBN}. @value{GDBN} announces it if you start
35585 with no arguments; you can also print it at any time using @code{show
35588 Without this, we will not know whether there is any point in looking for
35589 the bug in the current version of @value{GDBN}.
35592 The type of machine you are using, and the operating system name and
35596 The details of the @value{GDBN} build-time configuration.
35597 @value{GDBN} shows these details if you invoke it with the
35598 @option{--configuration} command-line option, or if you type
35599 @code{show configuration} at @value{GDBN}'s prompt.
35602 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
35603 ``@value{GCC}--2.8.1''.
35606 What compiler (and its version) was used to compile the program you are
35607 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
35608 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
35609 to get this information; for other compilers, see the documentation for
35613 The command arguments you gave the compiler to compile your example and
35614 observe the bug. For example, did you use @samp{-O}? To guarantee
35615 you will not omit something important, list them all. A copy of the
35616 Makefile (or the output from make) is sufficient.
35618 If we were to try to guess the arguments, we would probably guess wrong
35619 and then we might not encounter the bug.
35622 A complete input script, and all necessary source files, that will
35626 A description of what behavior you observe that you believe is
35627 incorrect. For example, ``It gets a fatal signal.''
35629 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
35630 will certainly notice it. But if the bug is incorrect output, we might
35631 not notice unless it is glaringly wrong. You might as well not give us
35632 a chance to make a mistake.
35634 Even if the problem you experience is a fatal signal, you should still
35635 say so explicitly. Suppose something strange is going on, such as, your
35636 copy of @value{GDBN} is out of synch, or you have encountered a bug in
35637 the C library on your system. (This has happened!) Your copy might
35638 crash and ours would not. If you told us to expect a crash, then when
35639 ours fails to crash, we would know that the bug was not happening for
35640 us. If you had not told us to expect a crash, then we would not be able
35641 to draw any conclusion from our observations.
35644 @cindex recording a session script
35645 To collect all this information, you can use a session recording program
35646 such as @command{script}, which is available on many Unix systems.
35647 Just run your @value{GDBN} session inside @command{script} and then
35648 include the @file{typescript} file with your bug report.
35650 Another way to record a @value{GDBN} session is to run @value{GDBN}
35651 inside Emacs and then save the entire buffer to a file.
35654 If you wish to suggest changes to the @value{GDBN} source, send us context
35655 diffs. If you even discuss something in the @value{GDBN} source, refer to
35656 it by context, not by line number.
35658 The line numbers in our development sources will not match those in your
35659 sources. Your line numbers would convey no useful information to us.
35663 Here are some things that are not necessary:
35667 A description of the envelope of the bug.
35669 Often people who encounter a bug spend a lot of time investigating
35670 which changes to the input file will make the bug go away and which
35671 changes will not affect it.
35673 This is often time consuming and not very useful, because the way we
35674 will find the bug is by running a single example under the debugger
35675 with breakpoints, not by pure deduction from a series of examples.
35676 We recommend that you save your time for something else.
35678 Of course, if you can find a simpler example to report @emph{instead}
35679 of the original one, that is a convenience for us. Errors in the
35680 output will be easier to spot, running under the debugger will take
35681 less time, and so on.
35683 However, simplification is not vital; if you do not want to do this,
35684 report the bug anyway and send us the entire test case you used.
35687 A patch for the bug.
35689 A patch for the bug does help us if it is a good one. But do not omit
35690 the necessary information, such as the test case, on the assumption that
35691 a patch is all we need. We might see problems with your patch and decide
35692 to fix the problem another way, or we might not understand it at all.
35694 Sometimes with a program as complicated as @value{GDBN} it is very hard to
35695 construct an example that will make the program follow a certain path
35696 through the code. If you do not send us the example, we will not be able
35697 to construct one, so we will not be able to verify that the bug is fixed.
35699 And if we cannot understand what bug you are trying to fix, or why your
35700 patch should be an improvement, we will not install it. A test case will
35701 help us to understand.
35704 A guess about what the bug is or what it depends on.
35706 Such guesses are usually wrong. Even we cannot guess right about such
35707 things without first using the debugger to find the facts.
35710 @c The readline documentation is distributed with the readline code
35711 @c and consists of the two following files:
35714 @c Use -I with makeinfo to point to the appropriate directory,
35715 @c environment var TEXINPUTS with TeX.
35716 @ifclear SYSTEM_READLINE
35717 @include rluser.texi
35718 @include hsuser.texi
35722 @appendix In Memoriam
35724 The @value{GDBN} project mourns the loss of the following long-time
35729 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
35730 to Free Software in general. Outside of @value{GDBN}, he was known in
35731 the Amiga world for his series of Fish Disks, and the GeekGadget project.
35733 @item Michael Snyder
35734 Michael was one of the Global Maintainers of the @value{GDBN} project,
35735 with contributions recorded as early as 1996, until 2011. In addition
35736 to his day to day participation, he was a large driving force behind
35737 adding Reverse Debugging to @value{GDBN}.
35740 Beyond their technical contributions to the project, they were also
35741 enjoyable members of the Free Software Community. We will miss them.
35743 @node Formatting Documentation
35744 @appendix Formatting Documentation
35746 @cindex @value{GDBN} reference card
35747 @cindex reference card
35748 The @value{GDBN} 4 release includes an already-formatted reference card, ready
35749 for printing with PostScript or Ghostscript, in the @file{gdb}
35750 subdirectory of the main source directory@footnote{In
35751 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
35752 release.}. If you can use PostScript or Ghostscript with your printer,
35753 you can print the reference card immediately with @file{refcard.ps}.
35755 The release also includes the source for the reference card. You
35756 can format it, using @TeX{}, by typing:
35762 The @value{GDBN} reference card is designed to print in @dfn{landscape}
35763 mode on US ``letter'' size paper;
35764 that is, on a sheet 11 inches wide by 8.5 inches
35765 high. You will need to specify this form of printing as an option to
35766 your @sc{dvi} output program.
35768 @cindex documentation
35770 All the documentation for @value{GDBN} comes as part of the machine-readable
35771 distribution. The documentation is written in Texinfo format, which is
35772 a documentation system that uses a single source file to produce both
35773 on-line information and a printed manual. You can use one of the Info
35774 formatting commands to create the on-line version of the documentation
35775 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
35777 @value{GDBN} includes an already formatted copy of the on-line Info
35778 version of this manual in the @file{gdb} subdirectory. The main Info
35779 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
35780 subordinate files matching @samp{gdb.info*} in the same directory. If
35781 necessary, you can print out these files, or read them with any editor;
35782 but they are easier to read using the @code{info} subsystem in @sc{gnu}
35783 Emacs or the standalone @code{info} program, available as part of the
35784 @sc{gnu} Texinfo distribution.
35786 If you want to format these Info files yourself, you need one of the
35787 Info formatting programs, such as @code{texinfo-format-buffer} or
35790 If you have @code{makeinfo} installed, and are in the top level
35791 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
35792 version @value{GDBVN}), you can make the Info file by typing:
35799 If you want to typeset and print copies of this manual, you need @TeX{},
35800 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
35801 Texinfo definitions file.
35803 @TeX{} is a typesetting program; it does not print files directly, but
35804 produces output files called @sc{dvi} files. To print a typeset
35805 document, you need a program to print @sc{dvi} files. If your system
35806 has @TeX{} installed, chances are it has such a program. The precise
35807 command to use depends on your system; @kbd{lpr -d} is common; another
35808 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
35809 require a file name without any extension or a @samp{.dvi} extension.
35811 @TeX{} also requires a macro definitions file called
35812 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
35813 written in Texinfo format. On its own, @TeX{} cannot either read or
35814 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
35815 and is located in the @file{gdb-@var{version-number}/texinfo}
35818 If you have @TeX{} and a @sc{dvi} printer program installed, you can
35819 typeset and print this manual. First switch to the @file{gdb}
35820 subdirectory of the main source directory (for example, to
35821 @file{gdb-@value{GDBVN}/gdb}) and type:
35827 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
35829 @node Installing GDB
35830 @appendix Installing @value{GDBN}
35831 @cindex installation
35834 * Requirements:: Requirements for building @value{GDBN}
35835 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
35836 * Separate Objdir:: Compiling @value{GDBN} in another directory
35837 * Config Names:: Specifying names for hosts and targets
35838 * Configure Options:: Summary of options for configure
35839 * System-wide configuration:: Having a system-wide init file
35843 @section Requirements for Building @value{GDBN}
35844 @cindex building @value{GDBN}, requirements for
35846 Building @value{GDBN} requires various tools and packages to be available.
35847 Other packages will be used only if they are found.
35849 @heading Tools/Packages Necessary for Building @value{GDBN}
35851 @item ISO C90 compiler
35852 @value{GDBN} is written in ISO C90. It should be buildable with any
35853 working C90 compiler, e.g.@: GCC.
35857 @heading Tools/Packages Optional for Building @value{GDBN}
35861 @value{GDBN} can use the Expat XML parsing library. This library may be
35862 included with your operating system distribution; if it is not, you
35863 can get the latest version from @url{http://expat.sourceforge.net}.
35864 The @file{configure} script will search for this library in several
35865 standard locations; if it is installed in an unusual path, you can
35866 use the @option{--with-libexpat-prefix} option to specify its location.
35872 Remote protocol memory maps (@pxref{Memory Map Format})
35874 Target descriptions (@pxref{Target Descriptions})
35876 Remote shared library lists (@xref{Library List Format},
35877 or alternatively @pxref{Library List Format for SVR4 Targets})
35879 MS-Windows shared libraries (@pxref{Shared Libraries})
35881 Traceframe info (@pxref{Traceframe Info Format})
35883 Branch trace (@pxref{Branch Trace Format})
35887 @cindex compressed debug sections
35888 @value{GDBN} will use the @samp{zlib} library, if available, to read
35889 compressed debug sections. Some linkers, such as GNU gold, are capable
35890 of producing binaries with compressed debug sections. If @value{GDBN}
35891 is compiled with @samp{zlib}, it will be able to read the debug
35892 information in such binaries.
35894 The @samp{zlib} library is likely included with your operating system
35895 distribution; if it is not, you can get the latest version from
35896 @url{http://zlib.net}.
35899 @value{GDBN}'s features related to character sets (@pxref{Character
35900 Sets}) require a functioning @code{iconv} implementation. If you are
35901 on a GNU system, then this is provided by the GNU C Library. Some
35902 other systems also provide a working @code{iconv}.
35904 If @value{GDBN} is using the @code{iconv} program which is installed
35905 in a non-standard place, you will need to tell @value{GDBN} where to find it.
35906 This is done with @option{--with-iconv-bin} which specifies the
35907 directory that contains the @code{iconv} program.
35909 On systems without @code{iconv}, you can install GNU Libiconv. If you
35910 have previously installed Libiconv, you can use the
35911 @option{--with-libiconv-prefix} option to configure.
35913 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
35914 arrange to build Libiconv if a directory named @file{libiconv} appears
35915 in the top-most source directory. If Libiconv is built this way, and
35916 if the operating system does not provide a suitable @code{iconv}
35917 implementation, then the just-built library will automatically be used
35918 by @value{GDBN}. One easy way to set this up is to download GNU
35919 Libiconv, unpack it, and then rename the directory holding the
35920 Libiconv source code to @samp{libiconv}.
35923 @node Running Configure
35924 @section Invoking the @value{GDBN} @file{configure} Script
35925 @cindex configuring @value{GDBN}
35926 @value{GDBN} comes with a @file{configure} script that automates the process
35927 of preparing @value{GDBN} for installation; you can then use @code{make} to
35928 build the @code{gdb} program.
35930 @c irrelevant in info file; it's as current as the code it lives with.
35931 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
35932 look at the @file{README} file in the sources; we may have improved the
35933 installation procedures since publishing this manual.}
35936 The @value{GDBN} distribution includes all the source code you need for
35937 @value{GDBN} in a single directory, whose name is usually composed by
35938 appending the version number to @samp{gdb}.
35940 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
35941 @file{gdb-@value{GDBVN}} directory. That directory contains:
35944 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
35945 script for configuring @value{GDBN} and all its supporting libraries
35947 @item gdb-@value{GDBVN}/gdb
35948 the source specific to @value{GDBN} itself
35950 @item gdb-@value{GDBVN}/bfd
35951 source for the Binary File Descriptor library
35953 @item gdb-@value{GDBVN}/include
35954 @sc{gnu} include files
35956 @item gdb-@value{GDBVN}/libiberty
35957 source for the @samp{-liberty} free software library
35959 @item gdb-@value{GDBVN}/opcodes
35960 source for the library of opcode tables and disassemblers
35962 @item gdb-@value{GDBVN}/readline
35963 source for the @sc{gnu} command-line interface
35965 @item gdb-@value{GDBVN}/glob
35966 source for the @sc{gnu} filename pattern-matching subroutine
35968 @item gdb-@value{GDBVN}/mmalloc
35969 source for the @sc{gnu} memory-mapped malloc package
35972 The simplest way to configure and build @value{GDBN} is to run @file{configure}
35973 from the @file{gdb-@var{version-number}} source directory, which in
35974 this example is the @file{gdb-@value{GDBVN}} directory.
35976 First switch to the @file{gdb-@var{version-number}} source directory
35977 if you are not already in it; then run @file{configure}. Pass the
35978 identifier for the platform on which @value{GDBN} will run as an
35984 cd gdb-@value{GDBVN}
35985 ./configure @var{host}
35990 where @var{host} is an identifier such as @samp{sun4} or
35991 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
35992 (You can often leave off @var{host}; @file{configure} tries to guess the
35993 correct value by examining your system.)
35995 Running @samp{configure @var{host}} and then running @code{make} builds the
35996 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
35997 libraries, then @code{gdb} itself. The configured source files, and the
35998 binaries, are left in the corresponding source directories.
36001 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
36002 system does not recognize this automatically when you run a different
36003 shell, you may need to run @code{sh} on it explicitly:
36006 sh configure @var{host}
36009 If you run @file{configure} from a directory that contains source
36010 directories for multiple libraries or programs, such as the
36011 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
36013 creates configuration files for every directory level underneath (unless
36014 you tell it not to, with the @samp{--norecursion} option).
36016 You should run the @file{configure} script from the top directory in the
36017 source tree, the @file{gdb-@var{version-number}} directory. If you run
36018 @file{configure} from one of the subdirectories, you will configure only
36019 that subdirectory. That is usually not what you want. In particular,
36020 if you run the first @file{configure} from the @file{gdb} subdirectory
36021 of the @file{gdb-@var{version-number}} directory, you will omit the
36022 configuration of @file{bfd}, @file{readline}, and other sibling
36023 directories of the @file{gdb} subdirectory. This leads to build errors
36024 about missing include files such as @file{bfd/bfd.h}.
36026 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
36027 However, you should make sure that the shell on your path (named by
36028 the @samp{SHELL} environment variable) is publicly readable. Remember
36029 that @value{GDBN} uses the shell to start your program---some systems refuse to
36030 let @value{GDBN} debug child processes whose programs are not readable.
36032 @node Separate Objdir
36033 @section Compiling @value{GDBN} in Another Directory
36035 If you want to run @value{GDBN} versions for several host or target machines,
36036 you need a different @code{gdb} compiled for each combination of
36037 host and target. @file{configure} is designed to make this easy by
36038 allowing you to generate each configuration in a separate subdirectory,
36039 rather than in the source directory. If your @code{make} program
36040 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
36041 @code{make} in each of these directories builds the @code{gdb}
36042 program specified there.
36044 To build @code{gdb} in a separate directory, run @file{configure}
36045 with the @samp{--srcdir} option to specify where to find the source.
36046 (You also need to specify a path to find @file{configure}
36047 itself from your working directory. If the path to @file{configure}
36048 would be the same as the argument to @samp{--srcdir}, you can leave out
36049 the @samp{--srcdir} option; it is assumed.)
36051 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
36052 separate directory for a Sun 4 like this:
36056 cd gdb-@value{GDBVN}
36059 ../gdb-@value{GDBVN}/configure sun4
36064 When @file{configure} builds a configuration using a remote source
36065 directory, it creates a tree for the binaries with the same structure
36066 (and using the same names) as the tree under the source directory. In
36067 the example, you'd find the Sun 4 library @file{libiberty.a} in the
36068 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
36069 @file{gdb-sun4/gdb}.
36071 Make sure that your path to the @file{configure} script has just one
36072 instance of @file{gdb} in it. If your path to @file{configure} looks
36073 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
36074 one subdirectory of @value{GDBN}, not the whole package. This leads to
36075 build errors about missing include files such as @file{bfd/bfd.h}.
36077 One popular reason to build several @value{GDBN} configurations in separate
36078 directories is to configure @value{GDBN} for cross-compiling (where
36079 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
36080 programs that run on another machine---the @dfn{target}).
36081 You specify a cross-debugging target by
36082 giving the @samp{--target=@var{target}} option to @file{configure}.
36084 When you run @code{make} to build a program or library, you must run
36085 it in a configured directory---whatever directory you were in when you
36086 called @file{configure} (or one of its subdirectories).
36088 The @code{Makefile} that @file{configure} generates in each source
36089 directory also runs recursively. If you type @code{make} in a source
36090 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
36091 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
36092 will build all the required libraries, and then build GDB.
36094 When you have multiple hosts or targets configured in separate
36095 directories, you can run @code{make} on them in parallel (for example,
36096 if they are NFS-mounted on each of the hosts); they will not interfere
36100 @section Specifying Names for Hosts and Targets
36102 The specifications used for hosts and targets in the @file{configure}
36103 script are based on a three-part naming scheme, but some short predefined
36104 aliases are also supported. The full naming scheme encodes three pieces
36105 of information in the following pattern:
36108 @var{architecture}-@var{vendor}-@var{os}
36111 For example, you can use the alias @code{sun4} as a @var{host} argument,
36112 or as the value for @var{target} in a @code{--target=@var{target}}
36113 option. The equivalent full name is @samp{sparc-sun-sunos4}.
36115 The @file{configure} script accompanying @value{GDBN} does not provide
36116 any query facility to list all supported host and target names or
36117 aliases. @file{configure} calls the Bourne shell script
36118 @code{config.sub} to map abbreviations to full names; you can read the
36119 script, if you wish, or you can use it to test your guesses on
36120 abbreviations---for example:
36123 % sh config.sub i386-linux
36125 % sh config.sub alpha-linux
36126 alpha-unknown-linux-gnu
36127 % sh config.sub hp9k700
36129 % sh config.sub sun4
36130 sparc-sun-sunos4.1.1
36131 % sh config.sub sun3
36132 m68k-sun-sunos4.1.1
36133 % sh config.sub i986v
36134 Invalid configuration `i986v': machine `i986v' not recognized
36138 @code{config.sub} is also distributed in the @value{GDBN} source
36139 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
36141 @node Configure Options
36142 @section @file{configure} Options
36144 Here is a summary of the @file{configure} options and arguments that
36145 are most often useful for building @value{GDBN}. @file{configure} also has
36146 several other options not listed here. @inforef{What Configure
36147 Does,,configure.info}, for a full explanation of @file{configure}.
36150 configure @r{[}--help@r{]}
36151 @r{[}--prefix=@var{dir}@r{]}
36152 @r{[}--exec-prefix=@var{dir}@r{]}
36153 @r{[}--srcdir=@var{dirname}@r{]}
36154 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
36155 @r{[}--target=@var{target}@r{]}
36160 You may introduce options with a single @samp{-} rather than
36161 @samp{--} if you prefer; but you may abbreviate option names if you use
36166 Display a quick summary of how to invoke @file{configure}.
36168 @item --prefix=@var{dir}
36169 Configure the source to install programs and files under directory
36172 @item --exec-prefix=@var{dir}
36173 Configure the source to install programs under directory
36176 @c avoid splitting the warning from the explanation:
36178 @item --srcdir=@var{dirname}
36179 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
36180 @code{make} that implements the @code{VPATH} feature.}@*
36181 Use this option to make configurations in directories separate from the
36182 @value{GDBN} source directories. Among other things, you can use this to
36183 build (or maintain) several configurations simultaneously, in separate
36184 directories. @file{configure} writes configuration-specific files in
36185 the current directory, but arranges for them to use the source in the
36186 directory @var{dirname}. @file{configure} creates directories under
36187 the working directory in parallel to the source directories below
36190 @item --norecursion
36191 Configure only the directory level where @file{configure} is executed; do not
36192 propagate configuration to subdirectories.
36194 @item --target=@var{target}
36195 Configure @value{GDBN} for cross-debugging programs running on the specified
36196 @var{target}. Without this option, @value{GDBN} is configured to debug
36197 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
36199 There is no convenient way to generate a list of all available targets.
36201 @item @var{host} @dots{}
36202 Configure @value{GDBN} to run on the specified @var{host}.
36204 There is no convenient way to generate a list of all available hosts.
36207 There are many other options available as well, but they are generally
36208 needed for special purposes only.
36210 @node System-wide configuration
36211 @section System-wide configuration and settings
36212 @cindex system-wide init file
36214 @value{GDBN} can be configured to have a system-wide init file;
36215 this file will be read and executed at startup (@pxref{Startup, , What
36216 @value{GDBN} does during startup}).
36218 Here is the corresponding configure option:
36221 @item --with-system-gdbinit=@var{file}
36222 Specify that the default location of the system-wide init file is
36226 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
36227 it may be subject to relocation. Two possible cases:
36231 If the default location of this init file contains @file{$prefix},
36232 it will be subject to relocation. Suppose that the configure options
36233 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
36234 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
36235 init file is looked for as @file{$install/etc/gdbinit} instead of
36236 @file{$prefix/etc/gdbinit}.
36239 By contrast, if the default location does not contain the prefix,
36240 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
36241 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
36242 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
36243 wherever @value{GDBN} is installed.
36246 If the configured location of the system-wide init file (as given by the
36247 @option{--with-system-gdbinit} option at configure time) is in the
36248 data-directory (as specified by @option{--with-gdb-datadir} at configure
36249 time) or in one of its subdirectories, then @value{GDBN} will look for the
36250 system-wide init file in the directory specified by the
36251 @option{--data-directory} command-line option.
36252 Note that the system-wide init file is only read once, during @value{GDBN}
36253 initialization. If the data-directory is changed after @value{GDBN} has
36254 started with the @code{set data-directory} command, the file will not be
36257 @node Maintenance Commands
36258 @appendix Maintenance Commands
36259 @cindex maintenance commands
36260 @cindex internal commands
36262 In addition to commands intended for @value{GDBN} users, @value{GDBN}
36263 includes a number of commands intended for @value{GDBN} developers,
36264 that are not documented elsewhere in this manual. These commands are
36265 provided here for reference. (For commands that turn on debugging
36266 messages, see @ref{Debugging Output}.)
36269 @kindex maint agent
36270 @kindex maint agent-eval
36271 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36272 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36273 Translate the given @var{expression} into remote agent bytecodes.
36274 This command is useful for debugging the Agent Expression mechanism
36275 (@pxref{Agent Expressions}). The @samp{agent} version produces an
36276 expression useful for data collection, such as by tracepoints, while
36277 @samp{maint agent-eval} produces an expression that evaluates directly
36278 to a result. For instance, a collection expression for @code{globa +
36279 globb} will include bytecodes to record four bytes of memory at each
36280 of the addresses of @code{globa} and @code{globb}, while discarding
36281 the result of the addition, while an evaluation expression will do the
36282 addition and return the sum.
36283 If @code{-at} is given, generate remote agent bytecode for @var{location}.
36284 If not, generate remote agent bytecode for current frame PC address.
36286 @kindex maint agent-printf
36287 @item maint agent-printf @var{format},@var{expr},...
36288 Translate the given format string and list of argument expressions
36289 into remote agent bytecodes and display them as a disassembled list.
36290 This command is useful for debugging the agent version of dynamic
36291 printf (@pxref{Dynamic Printf}).
36293 @kindex maint info breakpoints
36294 @item @anchor{maint info breakpoints}maint info breakpoints
36295 Using the same format as @samp{info breakpoints}, display both the
36296 breakpoints you've set explicitly, and those @value{GDBN} is using for
36297 internal purposes. Internal breakpoints are shown with negative
36298 breakpoint numbers. The type column identifies what kind of breakpoint
36303 Normal, explicitly set breakpoint.
36306 Normal, explicitly set watchpoint.
36309 Internal breakpoint, used to handle correctly stepping through
36310 @code{longjmp} calls.
36312 @item longjmp resume
36313 Internal breakpoint at the target of a @code{longjmp}.
36316 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
36319 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
36322 Shared library events.
36326 @kindex maint info bfds
36327 @item maint info bfds
36328 This prints information about each @code{bfd} object that is known to
36329 @value{GDBN}. @xref{Top, , BFD, bfd, The Binary File Descriptor Library}.
36331 @kindex set displaced-stepping
36332 @kindex show displaced-stepping
36333 @cindex displaced stepping support
36334 @cindex out-of-line single-stepping
36335 @item set displaced-stepping
36336 @itemx show displaced-stepping
36337 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
36338 if the target supports it. Displaced stepping is a way to single-step
36339 over breakpoints without removing them from the inferior, by executing
36340 an out-of-line copy of the instruction that was originally at the
36341 breakpoint location. It is also known as out-of-line single-stepping.
36344 @item set displaced-stepping on
36345 If the target architecture supports it, @value{GDBN} will use
36346 displaced stepping to step over breakpoints.
36348 @item set displaced-stepping off
36349 @value{GDBN} will not use displaced stepping to step over breakpoints,
36350 even if such is supported by the target architecture.
36352 @cindex non-stop mode, and @samp{set displaced-stepping}
36353 @item set displaced-stepping auto
36354 This is the default mode. @value{GDBN} will use displaced stepping
36355 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
36356 architecture supports displaced stepping.
36359 @kindex maint check-psymtabs
36360 @item maint check-psymtabs
36361 Check the consistency of currently expanded psymtabs versus symtabs.
36362 Use this to check, for example, whether a symbol is in one but not the other.
36364 @kindex maint check-symtabs
36365 @item maint check-symtabs
36366 Check the consistency of currently expanded symtabs.
36368 @kindex maint expand-symtabs
36369 @item maint expand-symtabs [@var{regexp}]
36370 Expand symbol tables.
36371 If @var{regexp} is specified, only expand symbol tables for file
36372 names matching @var{regexp}.
36374 @kindex maint cplus first_component
36375 @item maint cplus first_component @var{name}
36376 Print the first C@t{++} class/namespace component of @var{name}.
36378 @kindex maint cplus namespace
36379 @item maint cplus namespace
36380 Print the list of possible C@t{++} namespaces.
36382 @kindex maint demangle
36383 @item maint demangle @var{name}
36384 Demangle a C@t{++} or Objective-C mangled @var{name}.
36386 @kindex maint deprecate
36387 @kindex maint undeprecate
36388 @cindex deprecated commands
36389 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
36390 @itemx maint undeprecate @var{command}
36391 Deprecate or undeprecate the named @var{command}. Deprecated commands
36392 cause @value{GDBN} to issue a warning when you use them. The optional
36393 argument @var{replacement} says which newer command should be used in
36394 favor of the deprecated one; if it is given, @value{GDBN} will mention
36395 the replacement as part of the warning.
36397 @kindex maint dump-me
36398 @item maint dump-me
36399 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
36400 Cause a fatal signal in the debugger and force it to dump its core.
36401 This is supported only on systems which support aborting a program
36402 with the @code{SIGQUIT} signal.
36404 @kindex maint internal-error
36405 @kindex maint internal-warning
36406 @item maint internal-error @r{[}@var{message-text}@r{]}
36407 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
36408 Cause @value{GDBN} to call the internal function @code{internal_error}
36409 or @code{internal_warning} and hence behave as though an internal error
36410 or internal warning has been detected. In addition to reporting the
36411 internal problem, these functions give the user the opportunity to
36412 either quit @value{GDBN} or create a core file of the current
36413 @value{GDBN} session.
36415 These commands take an optional parameter @var{message-text} that is
36416 used as the text of the error or warning message.
36418 Here's an example of using @code{internal-error}:
36421 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
36422 @dots{}/maint.c:121: internal-error: testing, 1, 2
36423 A problem internal to GDB has been detected. Further
36424 debugging may prove unreliable.
36425 Quit this debugging session? (y or n) @kbd{n}
36426 Create a core file? (y or n) @kbd{n}
36430 @cindex @value{GDBN} internal error
36431 @cindex internal errors, control of @value{GDBN} behavior
36433 @kindex maint set internal-error
36434 @kindex maint show internal-error
36435 @kindex maint set internal-warning
36436 @kindex maint show internal-warning
36437 @item maint set internal-error @var{action} [ask|yes|no]
36438 @itemx maint show internal-error @var{action}
36439 @itemx maint set internal-warning @var{action} [ask|yes|no]
36440 @itemx maint show internal-warning @var{action}
36441 When @value{GDBN} reports an internal problem (error or warning) it
36442 gives the user the opportunity to both quit @value{GDBN} and create a
36443 core file of the current @value{GDBN} session. These commands let you
36444 override the default behaviour for each particular @var{action},
36445 described in the table below.
36449 You can specify that @value{GDBN} should always (yes) or never (no)
36450 quit. The default is to ask the user what to do.
36453 You can specify that @value{GDBN} should always (yes) or never (no)
36454 create a core file. The default is to ask the user what to do.
36457 @kindex maint packet
36458 @item maint packet @var{text}
36459 If @value{GDBN} is talking to an inferior via the serial protocol,
36460 then this command sends the string @var{text} to the inferior, and
36461 displays the response packet. @value{GDBN} supplies the initial
36462 @samp{$} character, the terminating @samp{#} character, and the
36465 @kindex maint print architecture
36466 @item maint print architecture @r{[}@var{file}@r{]}
36467 Print the entire architecture configuration. The optional argument
36468 @var{file} names the file where the output goes.
36470 @kindex maint print c-tdesc
36471 @item maint print c-tdesc
36472 Print the current target description (@pxref{Target Descriptions}) as
36473 a C source file. The created source file can be used in @value{GDBN}
36474 when an XML parser is not available to parse the description.
36476 @kindex maint print dummy-frames
36477 @item maint print dummy-frames
36478 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
36481 (@value{GDBP}) @kbd{b add}
36483 (@value{GDBP}) @kbd{print add(2,3)}
36484 Breakpoint 2, add (a=2, b=3) at @dots{}
36486 The program being debugged stopped while in a function called from GDB.
36488 (@value{GDBP}) @kbd{maint print dummy-frames}
36489 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
36490 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
36491 call_lo=0x01014000 call_hi=0x01014001
36495 Takes an optional file parameter.
36497 @kindex maint print registers
36498 @kindex maint print raw-registers
36499 @kindex maint print cooked-registers
36500 @kindex maint print register-groups
36501 @kindex maint print remote-registers
36502 @item maint print registers @r{[}@var{file}@r{]}
36503 @itemx maint print raw-registers @r{[}@var{file}@r{]}
36504 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
36505 @itemx maint print register-groups @r{[}@var{file}@r{]}
36506 @itemx maint print remote-registers @r{[}@var{file}@r{]}
36507 Print @value{GDBN}'s internal register data structures.
36509 The command @code{maint print raw-registers} includes the contents of
36510 the raw register cache; the command @code{maint print
36511 cooked-registers} includes the (cooked) value of all registers,
36512 including registers which aren't available on the target nor visible
36513 to user; the command @code{maint print register-groups} includes the
36514 groups that each register is a member of; and the command @code{maint
36515 print remote-registers} includes the remote target's register numbers
36516 and offsets in the `G' packets. @xref{Registers,, Registers, gdbint,
36517 @value{GDBN} Internals}.
36519 These commands take an optional parameter, a file name to which to
36520 write the information.
36522 @kindex maint print reggroups
36523 @item maint print reggroups @r{[}@var{file}@r{]}
36524 Print @value{GDBN}'s internal register group data structures. The
36525 optional argument @var{file} tells to what file to write the
36528 The register groups info looks like this:
36531 (@value{GDBP}) @kbd{maint print reggroups}
36544 This command forces @value{GDBN} to flush its internal register cache.
36546 @kindex maint print objfiles
36547 @cindex info for known object files
36548 @item maint print objfiles
36549 Print a dump of all known object files. For each object file, this
36550 command prints its name, address in memory, and all of its psymtabs
36553 @kindex maint print section-scripts
36554 @cindex info for known .debug_gdb_scripts-loaded scripts
36555 @item maint print section-scripts [@var{regexp}]
36556 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
36557 If @var{regexp} is specified, only print scripts loaded by object files
36558 matching @var{regexp}.
36559 For each script, this command prints its name as specified in the objfile,
36560 and the full path if known.
36561 @xref{dotdebug_gdb_scripts section}.
36563 @kindex maint print statistics
36564 @cindex bcache statistics
36565 @item maint print statistics
36566 This command prints, for each object file in the program, various data
36567 about that object file followed by the byte cache (@dfn{bcache})
36568 statistics for the object file. The objfile data includes the number
36569 of minimal, partial, full, and stabs symbols, the number of types
36570 defined by the objfile, the number of as yet unexpanded psym tables,
36571 the number of line tables and string tables, and the amount of memory
36572 used by the various tables. The bcache statistics include the counts,
36573 sizes, and counts of duplicates of all and unique objects, max,
36574 average, and median entry size, total memory used and its overhead and
36575 savings, and various measures of the hash table size and chain
36578 @kindex maint print target-stack
36579 @cindex target stack description
36580 @item maint print target-stack
36581 A @dfn{target} is an interface between the debugger and a particular
36582 kind of file or process. Targets can be stacked in @dfn{strata},
36583 so that more than one target can potentially respond to a request.
36584 In particular, memory accesses will walk down the stack of targets
36585 until they find a target that is interested in handling that particular
36588 This command prints a short description of each layer that was pushed on
36589 the @dfn{target stack}, starting from the top layer down to the bottom one.
36591 @kindex maint print type
36592 @cindex type chain of a data type
36593 @item maint print type @var{expr}
36594 Print the type chain for a type specified by @var{expr}. The argument
36595 can be either a type name or a symbol. If it is a symbol, the type of
36596 that symbol is described. The type chain produced by this command is
36597 a recursive definition of the data type as stored in @value{GDBN}'s
36598 data structures, including its flags and contained types.
36600 @kindex maint set dwarf2 always-disassemble
36601 @kindex maint show dwarf2 always-disassemble
36602 @item maint set dwarf2 always-disassemble
36603 @item maint show dwarf2 always-disassemble
36604 Control the behavior of @code{info address} when using DWARF debugging
36607 The default is @code{off}, which means that @value{GDBN} should try to
36608 describe a variable's location in an easily readable format. When
36609 @code{on}, @value{GDBN} will instead display the DWARF location
36610 expression in an assembly-like format. Note that some locations are
36611 too complex for @value{GDBN} to describe simply; in this case you will
36612 always see the disassembly form.
36614 Here is an example of the resulting disassembly:
36617 (gdb) info addr argc
36618 Symbol "argc" is a complex DWARF expression:
36622 For more information on these expressions, see
36623 @uref{http://www.dwarfstd.org/, the DWARF standard}.
36625 @kindex maint set dwarf2 max-cache-age
36626 @kindex maint show dwarf2 max-cache-age
36627 @item maint set dwarf2 max-cache-age
36628 @itemx maint show dwarf2 max-cache-age
36629 Control the DWARF 2 compilation unit cache.
36631 @cindex DWARF 2 compilation units cache
36632 In object files with inter-compilation-unit references, such as those
36633 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
36634 reader needs to frequently refer to previously read compilation units.
36635 This setting controls how long a compilation unit will remain in the
36636 cache if it is not referenced. A higher limit means that cached
36637 compilation units will be stored in memory longer, and more total
36638 memory will be used. Setting it to zero disables caching, which will
36639 slow down @value{GDBN} startup, but reduce memory consumption.
36641 @kindex maint set profile
36642 @kindex maint show profile
36643 @cindex profiling GDB
36644 @item maint set profile
36645 @itemx maint show profile
36646 Control profiling of @value{GDBN}.
36648 Profiling will be disabled until you use the @samp{maint set profile}
36649 command to enable it. When you enable profiling, the system will begin
36650 collecting timing and execution count data; when you disable profiling or
36651 exit @value{GDBN}, the results will be written to a log file. Remember that
36652 if you use profiling, @value{GDBN} will overwrite the profiling log file
36653 (often called @file{gmon.out}). If you have a record of important profiling
36654 data in a @file{gmon.out} file, be sure to move it to a safe location.
36656 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
36657 compiled with the @samp{-pg} compiler option.
36659 @kindex maint set show-debug-regs
36660 @kindex maint show show-debug-regs
36661 @cindex hardware debug registers
36662 @item maint set show-debug-regs
36663 @itemx maint show show-debug-regs
36664 Control whether to show variables that mirror the hardware debug
36665 registers. Use @code{ON} to enable, @code{OFF} to disable. If
36666 enabled, the debug registers values are shown when @value{GDBN} inserts or
36667 removes a hardware breakpoint or watchpoint, and when the inferior
36668 triggers a hardware-assisted breakpoint or watchpoint.
36670 @kindex maint set show-all-tib
36671 @kindex maint show show-all-tib
36672 @item maint set show-all-tib
36673 @itemx maint show show-all-tib
36674 Control whether to show all non zero areas within a 1k block starting
36675 at thread local base, when using the @samp{info w32 thread-information-block}
36678 @kindex maint set per-command
36679 @kindex maint show per-command
36680 @item maint set per-command
36681 @itemx maint show per-command
36682 @cindex resources used by commands
36684 @value{GDBN} can display the resources used by each command.
36685 This is useful in debugging performance problems.
36688 @item maint set per-command space [on|off]
36689 @itemx maint show per-command space
36690 Enable or disable the printing of the memory used by GDB for each command.
36691 If enabled, @value{GDBN} will display how much memory each command
36692 took, following the command's own output.
36693 This can also be requested by invoking @value{GDBN} with the
36694 @option{--statistics} command-line switch (@pxref{Mode Options}).
36696 @item maint set per-command time [on|off]
36697 @itemx maint show per-command time
36698 Enable or disable the printing of the execution time of @value{GDBN}
36700 If enabled, @value{GDBN} will display how much time it
36701 took to execute each command, following the command's own output.
36702 Both CPU time and wallclock time are printed.
36703 Printing both is useful when trying to determine whether the cost is
36704 CPU or, e.g., disk/network latency.
36705 Note that the CPU time printed is for @value{GDBN} only, it does not include
36706 the execution time of the inferior because there's no mechanism currently
36707 to compute how much time was spent by @value{GDBN} and how much time was
36708 spent by the program been debugged.
36709 This can also be requested by invoking @value{GDBN} with the
36710 @option{--statistics} command-line switch (@pxref{Mode Options}).
36712 @item maint set per-command symtab [on|off]
36713 @itemx maint show per-command symtab
36714 Enable or disable the printing of basic symbol table statistics
36716 If enabled, @value{GDBN} will display the following information:
36720 number of symbol tables
36722 number of primary symbol tables
36724 number of blocks in the blockvector
36728 @kindex maint space
36729 @cindex memory used by commands
36730 @item maint space @var{value}
36731 An alias for @code{maint set per-command space}.
36732 A non-zero value enables it, zero disables it.
36735 @cindex time of command execution
36736 @item maint time @var{value}
36737 An alias for @code{maint set per-command time}.
36738 A non-zero value enables it, zero disables it.
36740 @kindex maint translate-address
36741 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
36742 Find the symbol stored at the location specified by the address
36743 @var{addr} and an optional section name @var{section}. If found,
36744 @value{GDBN} prints the name of the closest symbol and an offset from
36745 the symbol's location to the specified address. This is similar to
36746 the @code{info address} command (@pxref{Symbols}), except that this
36747 command also allows to find symbols in other sections.
36749 If section was not specified, the section in which the symbol was found
36750 is also printed. For dynamically linked executables, the name of
36751 executable or shared library containing the symbol is printed as well.
36755 The following command is useful for non-interactive invocations of
36756 @value{GDBN}, such as in the test suite.
36759 @item set watchdog @var{nsec}
36760 @kindex set watchdog
36761 @cindex watchdog timer
36762 @cindex timeout for commands
36763 Set the maximum number of seconds @value{GDBN} will wait for the
36764 target operation to finish. If this time expires, @value{GDBN}
36765 reports and error and the command is aborted.
36767 @item show watchdog
36768 Show the current setting of the target wait timeout.
36771 @node Remote Protocol
36772 @appendix @value{GDBN} Remote Serial Protocol
36777 * Stop Reply Packets::
36778 * General Query Packets::
36779 * Architecture-Specific Protocol Details::
36780 * Tracepoint Packets::
36781 * Host I/O Packets::
36783 * Notification Packets::
36784 * Remote Non-Stop::
36785 * Packet Acknowledgment::
36787 * File-I/O Remote Protocol Extension::
36788 * Library List Format::
36789 * Library List Format for SVR4 Targets::
36790 * Memory Map Format::
36791 * Thread List Format::
36792 * Traceframe Info Format::
36793 * Branch Trace Format::
36799 There may be occasions when you need to know something about the
36800 protocol---for example, if there is only one serial port to your target
36801 machine, you might want your program to do something special if it
36802 recognizes a packet meant for @value{GDBN}.
36804 In the examples below, @samp{->} and @samp{<-} are used to indicate
36805 transmitted and received data, respectively.
36807 @cindex protocol, @value{GDBN} remote serial
36808 @cindex serial protocol, @value{GDBN} remote
36809 @cindex remote serial protocol
36810 All @value{GDBN} commands and responses (other than acknowledgments
36811 and notifications, see @ref{Notification Packets}) are sent as a
36812 @var{packet}. A @var{packet} is introduced with the character
36813 @samp{$}, the actual @var{packet-data}, and the terminating character
36814 @samp{#} followed by a two-digit @var{checksum}:
36817 @code{$}@var{packet-data}@code{#}@var{checksum}
36821 @cindex checksum, for @value{GDBN} remote
36823 The two-digit @var{checksum} is computed as the modulo 256 sum of all
36824 characters between the leading @samp{$} and the trailing @samp{#} (an
36825 eight bit unsigned checksum).
36827 Implementors should note that prior to @value{GDBN} 5.0 the protocol
36828 specification also included an optional two-digit @var{sequence-id}:
36831 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
36834 @cindex sequence-id, for @value{GDBN} remote
36836 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
36837 has never output @var{sequence-id}s. Stubs that handle packets added
36838 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
36840 When either the host or the target machine receives a packet, the first
36841 response expected is an acknowledgment: either @samp{+} (to indicate
36842 the package was received correctly) or @samp{-} (to request
36846 -> @code{$}@var{packet-data}@code{#}@var{checksum}
36851 The @samp{+}/@samp{-} acknowledgments can be disabled
36852 once a connection is established.
36853 @xref{Packet Acknowledgment}, for details.
36855 The host (@value{GDBN}) sends @var{command}s, and the target (the
36856 debugging stub incorporated in your program) sends a @var{response}. In
36857 the case of step and continue @var{command}s, the response is only sent
36858 when the operation has completed, and the target has again stopped all
36859 threads in all attached processes. This is the default all-stop mode
36860 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
36861 execution mode; see @ref{Remote Non-Stop}, for details.
36863 @var{packet-data} consists of a sequence of characters with the
36864 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
36867 @cindex remote protocol, field separator
36868 Fields within the packet should be separated using @samp{,} @samp{;} or
36869 @samp{:}. Except where otherwise noted all numbers are represented in
36870 @sc{hex} with leading zeros suppressed.
36872 Implementors should note that prior to @value{GDBN} 5.0, the character
36873 @samp{:} could not appear as the third character in a packet (as it
36874 would potentially conflict with the @var{sequence-id}).
36876 @cindex remote protocol, binary data
36877 @anchor{Binary Data}
36878 Binary data in most packets is encoded either as two hexadecimal
36879 digits per byte of binary data. This allowed the traditional remote
36880 protocol to work over connections which were only seven-bit clean.
36881 Some packets designed more recently assume an eight-bit clean
36882 connection, and use a more efficient encoding to send and receive
36885 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
36886 as an escape character. Any escaped byte is transmitted as the escape
36887 character followed by the original character XORed with @code{0x20}.
36888 For example, the byte @code{0x7d} would be transmitted as the two
36889 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
36890 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
36891 @samp{@}}) must always be escaped. Responses sent by the stub
36892 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
36893 is not interpreted as the start of a run-length encoded sequence
36896 Response @var{data} can be run-length encoded to save space.
36897 Run-length encoding replaces runs of identical characters with one
36898 instance of the repeated character, followed by a @samp{*} and a
36899 repeat count. The repeat count is itself sent encoded, to avoid
36900 binary characters in @var{data}: a value of @var{n} is sent as
36901 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
36902 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
36903 code 32) for a repeat count of 3. (This is because run-length
36904 encoding starts to win for counts 3 or more.) Thus, for example,
36905 @samp{0* } is a run-length encoding of ``0000'': the space character
36906 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
36909 The printable characters @samp{#} and @samp{$} or with a numeric value
36910 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
36911 seven repeats (@samp{$}) can be expanded using a repeat count of only
36912 five (@samp{"}). For example, @samp{00000000} can be encoded as
36915 The error response returned for some packets includes a two character
36916 error number. That number is not well defined.
36918 @cindex empty response, for unsupported packets
36919 For any @var{command} not supported by the stub, an empty response
36920 (@samp{$#00}) should be returned. That way it is possible to extend the
36921 protocol. A newer @value{GDBN} can tell if a packet is supported based
36924 At a minimum, a stub is required to support the @samp{g} and @samp{G}
36925 commands for register access, and the @samp{m} and @samp{M} commands
36926 for memory access. Stubs that only control single-threaded targets
36927 can implement run control with the @samp{c} (continue), and @samp{s}
36928 (step) commands. Stubs that support multi-threading targets should
36929 support the @samp{vCont} command. All other commands are optional.
36934 The following table provides a complete list of all currently defined
36935 @var{command}s and their corresponding response @var{data}.
36936 @xref{File-I/O Remote Protocol Extension}, for details about the File
36937 I/O extension of the remote protocol.
36939 Each packet's description has a template showing the packet's overall
36940 syntax, followed by an explanation of the packet's meaning. We
36941 include spaces in some of the templates for clarity; these are not
36942 part of the packet's syntax. No @value{GDBN} packet uses spaces to
36943 separate its components. For example, a template like @samp{foo
36944 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
36945 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
36946 @var{baz}. @value{GDBN} does not transmit a space character between the
36947 @samp{foo} and the @var{bar}, or between the @var{bar} and the
36950 @cindex @var{thread-id}, in remote protocol
36951 @anchor{thread-id syntax}
36952 Several packets and replies include a @var{thread-id} field to identify
36953 a thread. Normally these are positive numbers with a target-specific
36954 interpretation, formatted as big-endian hex strings. A @var{thread-id}
36955 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
36958 In addition, the remote protocol supports a multiprocess feature in
36959 which the @var{thread-id} syntax is extended to optionally include both
36960 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
36961 The @var{pid} (process) and @var{tid} (thread) components each have the
36962 format described above: a positive number with target-specific
36963 interpretation formatted as a big-endian hex string, literal @samp{-1}
36964 to indicate all processes or threads (respectively), or @samp{0} to
36965 indicate an arbitrary process or thread. Specifying just a process, as
36966 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
36967 error to specify all processes but a specific thread, such as
36968 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
36969 for those packets and replies explicitly documented to include a process
36970 ID, rather than a @var{thread-id}.
36972 The multiprocess @var{thread-id} syntax extensions are only used if both
36973 @value{GDBN} and the stub report support for the @samp{multiprocess}
36974 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
36977 Note that all packet forms beginning with an upper- or lower-case
36978 letter, other than those described here, are reserved for future use.
36980 Here are the packet descriptions.
36985 @cindex @samp{!} packet
36986 @anchor{extended mode}
36987 Enable extended mode. In extended mode, the remote server is made
36988 persistent. The @samp{R} packet is used to restart the program being
36994 The remote target both supports and has enabled extended mode.
36998 @cindex @samp{?} packet
36999 Indicate the reason the target halted. The reply is the same as for
37000 step and continue. This packet has a special interpretation when the
37001 target is in non-stop mode; see @ref{Remote Non-Stop}.
37004 @xref{Stop Reply Packets}, for the reply specifications.
37006 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
37007 @cindex @samp{A} packet
37008 Initialized @code{argv[]} array passed into program. @var{arglen}
37009 specifies the number of bytes in the hex encoded byte stream
37010 @var{arg}. See @code{gdbserver} for more details.
37015 The arguments were set.
37021 @cindex @samp{b} packet
37022 (Don't use this packet; its behavior is not well-defined.)
37023 Change the serial line speed to @var{baud}.
37025 JTC: @emph{When does the transport layer state change? When it's
37026 received, or after the ACK is transmitted. In either case, there are
37027 problems if the command or the acknowledgment packet is dropped.}
37029 Stan: @emph{If people really wanted to add something like this, and get
37030 it working for the first time, they ought to modify ser-unix.c to send
37031 some kind of out-of-band message to a specially-setup stub and have the
37032 switch happen "in between" packets, so that from remote protocol's point
37033 of view, nothing actually happened.}
37035 @item B @var{addr},@var{mode}
37036 @cindex @samp{B} packet
37037 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
37038 breakpoint at @var{addr}.
37040 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
37041 (@pxref{insert breakpoint or watchpoint packet}).
37043 @cindex @samp{bc} packet
37046 Backward continue. Execute the target system in reverse. No parameter.
37047 @xref{Reverse Execution}, for more information.
37050 @xref{Stop Reply Packets}, for the reply specifications.
37052 @cindex @samp{bs} packet
37055 Backward single step. Execute one instruction in reverse. No parameter.
37056 @xref{Reverse Execution}, for more information.
37059 @xref{Stop Reply Packets}, for the reply specifications.
37061 @item c @r{[}@var{addr}@r{]}
37062 @cindex @samp{c} packet
37063 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
37064 resume at current address.
37066 This packet is deprecated for multi-threading support. @xref{vCont
37070 @xref{Stop Reply Packets}, for the reply specifications.
37072 @item C @var{sig}@r{[};@var{addr}@r{]}
37073 @cindex @samp{C} packet
37074 Continue with signal @var{sig} (hex signal number). If
37075 @samp{;@var{addr}} is omitted, resume at same address.
37077 This packet is deprecated for multi-threading support. @xref{vCont
37081 @xref{Stop Reply Packets}, for the reply specifications.
37084 @cindex @samp{d} packet
37087 Don't use this packet; instead, define a general set packet
37088 (@pxref{General Query Packets}).
37092 @cindex @samp{D} packet
37093 The first form of the packet is used to detach @value{GDBN} from the
37094 remote system. It is sent to the remote target
37095 before @value{GDBN} disconnects via the @code{detach} command.
37097 The second form, including a process ID, is used when multiprocess
37098 protocol extensions are enabled (@pxref{multiprocess extensions}), to
37099 detach only a specific process. The @var{pid} is specified as a
37100 big-endian hex string.
37110 @item F @var{RC},@var{EE},@var{CF};@var{XX}
37111 @cindex @samp{F} packet
37112 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
37113 This is part of the File-I/O protocol extension. @xref{File-I/O
37114 Remote Protocol Extension}, for the specification.
37117 @anchor{read registers packet}
37118 @cindex @samp{g} packet
37119 Read general registers.
37123 @item @var{XX@dots{}}
37124 Each byte of register data is described by two hex digits. The bytes
37125 with the register are transmitted in target byte order. The size of
37126 each register and their position within the @samp{g} packet are
37127 determined by the @value{GDBN} internal gdbarch functions
37128 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
37129 specification of several standard @samp{g} packets is specified below.
37131 When reading registers from a trace frame (@pxref{Analyze Collected
37132 Data,,Using the Collected Data}), the stub may also return a string of
37133 literal @samp{x}'s in place of the register data digits, to indicate
37134 that the corresponding register has not been collected, thus its value
37135 is unavailable. For example, for an architecture with 4 registers of
37136 4 bytes each, the following reply indicates to @value{GDBN} that
37137 registers 0 and 2 have not been collected, while registers 1 and 3
37138 have been collected, and both have zero value:
37142 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
37149 @item G @var{XX@dots{}}
37150 @cindex @samp{G} packet
37151 Write general registers. @xref{read registers packet}, for a
37152 description of the @var{XX@dots{}} data.
37162 @item H @var{op} @var{thread-id}
37163 @cindex @samp{H} packet
37164 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
37165 @samp{G}, et.al.). @var{op} depends on the operation to be performed:
37166 it should be @samp{c} for step and continue operations (note that this
37167 is deprecated, supporting the @samp{vCont} command is a better
37168 option), @samp{g} for other operations. The thread designator
37169 @var{thread-id} has the format and interpretation described in
37170 @ref{thread-id syntax}.
37181 @c 'H': How restrictive (or permissive) is the thread model. If a
37182 @c thread is selected and stopped, are other threads allowed
37183 @c to continue to execute? As I mentioned above, I think the
37184 @c semantics of each command when a thread is selected must be
37185 @c described. For example:
37187 @c 'g': If the stub supports threads and a specific thread is
37188 @c selected, returns the register block from that thread;
37189 @c otherwise returns current registers.
37191 @c 'G' If the stub supports threads and a specific thread is
37192 @c selected, sets the registers of the register block of
37193 @c that thread; otherwise sets current registers.
37195 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
37196 @anchor{cycle step packet}
37197 @cindex @samp{i} packet
37198 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
37199 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
37200 step starting at that address.
37203 @cindex @samp{I} packet
37204 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
37208 @cindex @samp{k} packet
37211 FIXME: @emph{There is no description of how to operate when a specific
37212 thread context has been selected (i.e.@: does 'k' kill only that
37215 @item m @var{addr},@var{length}
37216 @cindex @samp{m} packet
37217 Read @var{length} bytes of memory starting at address @var{addr}.
37218 Note that @var{addr} may not be aligned to any particular boundary.
37220 The stub need not use any particular size or alignment when gathering
37221 data from memory for the response; even if @var{addr} is word-aligned
37222 and @var{length} is a multiple of the word size, the stub is free to
37223 use byte accesses, or not. For this reason, this packet may not be
37224 suitable for accessing memory-mapped I/O devices.
37225 @cindex alignment of remote memory accesses
37226 @cindex size of remote memory accesses
37227 @cindex memory, alignment and size of remote accesses
37231 @item @var{XX@dots{}}
37232 Memory contents; each byte is transmitted as a two-digit hexadecimal
37233 number. The reply may contain fewer bytes than requested if the
37234 server was able to read only part of the region of memory.
37239 @item M @var{addr},@var{length}:@var{XX@dots{}}
37240 @cindex @samp{M} packet
37241 Write @var{length} bytes of memory starting at address @var{addr}.
37242 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
37243 hexadecimal number.
37250 for an error (this includes the case where only part of the data was
37255 @cindex @samp{p} packet
37256 Read the value of register @var{n}; @var{n} is in hex.
37257 @xref{read registers packet}, for a description of how the returned
37258 register value is encoded.
37262 @item @var{XX@dots{}}
37263 the register's value
37267 Indicating an unrecognized @var{query}.
37270 @item P @var{n@dots{}}=@var{r@dots{}}
37271 @anchor{write register packet}
37272 @cindex @samp{P} packet
37273 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
37274 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
37275 digits for each byte in the register (target byte order).
37285 @item q @var{name} @var{params}@dots{}
37286 @itemx Q @var{name} @var{params}@dots{}
37287 @cindex @samp{q} packet
37288 @cindex @samp{Q} packet
37289 General query (@samp{q}) and set (@samp{Q}). These packets are
37290 described fully in @ref{General Query Packets}.
37293 @cindex @samp{r} packet
37294 Reset the entire system.
37296 Don't use this packet; use the @samp{R} packet instead.
37299 @cindex @samp{R} packet
37300 Restart the program being debugged. @var{XX}, while needed, is ignored.
37301 This packet is only available in extended mode (@pxref{extended mode}).
37303 The @samp{R} packet has no reply.
37305 @item s @r{[}@var{addr}@r{]}
37306 @cindex @samp{s} packet
37307 Single step. @var{addr} is the address at which to resume. If
37308 @var{addr} is omitted, resume at same address.
37310 This packet is deprecated for multi-threading support. @xref{vCont
37314 @xref{Stop Reply Packets}, for the reply specifications.
37316 @item S @var{sig}@r{[};@var{addr}@r{]}
37317 @anchor{step with signal packet}
37318 @cindex @samp{S} packet
37319 Step with signal. This is analogous to the @samp{C} packet, but
37320 requests a single-step, rather than a normal resumption of execution.
37322 This packet is deprecated for multi-threading support. @xref{vCont
37326 @xref{Stop Reply Packets}, for the reply specifications.
37328 @item t @var{addr}:@var{PP},@var{MM}
37329 @cindex @samp{t} packet
37330 Search backwards starting at address @var{addr} for a match with pattern
37331 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
37332 @var{addr} must be at least 3 digits.
37334 @item T @var{thread-id}
37335 @cindex @samp{T} packet
37336 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
37341 thread is still alive
37347 Packets starting with @samp{v} are identified by a multi-letter name,
37348 up to the first @samp{;} or @samp{?} (or the end of the packet).
37350 @item vAttach;@var{pid}
37351 @cindex @samp{vAttach} packet
37352 Attach to a new process with the specified process ID @var{pid}.
37353 The process ID is a
37354 hexadecimal integer identifying the process. In all-stop mode, all
37355 threads in the attached process are stopped; in non-stop mode, it may be
37356 attached without being stopped if that is supported by the target.
37358 @c In non-stop mode, on a successful vAttach, the stub should set the
37359 @c current thread to a thread of the newly-attached process. After
37360 @c attaching, GDB queries for the attached process's thread ID with qC.
37361 @c Also note that, from a user perspective, whether or not the
37362 @c target is stopped on attach in non-stop mode depends on whether you
37363 @c use the foreground or background version of the attach command, not
37364 @c on what vAttach does; GDB does the right thing with respect to either
37365 @c stopping or restarting threads.
37367 This packet is only available in extended mode (@pxref{extended mode}).
37373 @item @r{Any stop packet}
37374 for success in all-stop mode (@pxref{Stop Reply Packets})
37376 for success in non-stop mode (@pxref{Remote Non-Stop})
37379 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
37380 @cindex @samp{vCont} packet
37381 @anchor{vCont packet}
37382 Resume the inferior, specifying different actions for each thread.
37383 If an action is specified with no @var{thread-id}, then it is applied to any
37384 threads that don't have a specific action specified; if no default action is
37385 specified then other threads should remain stopped in all-stop mode and
37386 in their current state in non-stop mode.
37387 Specifying multiple
37388 default actions is an error; specifying no actions is also an error.
37389 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
37391 Currently supported actions are:
37397 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
37401 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
37406 The optional argument @var{addr} normally associated with the
37407 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
37408 not supported in @samp{vCont}.
37410 The @samp{t} action is only relevant in non-stop mode
37411 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
37412 A stop reply should be generated for any affected thread not already stopped.
37413 When a thread is stopped by means of a @samp{t} action,
37414 the corresponding stop reply should indicate that the thread has stopped with
37415 signal @samp{0}, regardless of whether the target uses some other signal
37416 as an implementation detail.
37418 The stub must support @samp{vCont} if it reports support for
37419 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
37420 this case @samp{vCont} actions can be specified to apply to all threads
37421 in a process by using the @samp{p@var{pid}.-1} form of the
37425 @xref{Stop Reply Packets}, for the reply specifications.
37428 @cindex @samp{vCont?} packet
37429 Request a list of actions supported by the @samp{vCont} packet.
37433 @item vCont@r{[};@var{action}@dots{}@r{]}
37434 The @samp{vCont} packet is supported. Each @var{action} is a supported
37435 command in the @samp{vCont} packet.
37437 The @samp{vCont} packet is not supported.
37440 @item vFile:@var{operation}:@var{parameter}@dots{}
37441 @cindex @samp{vFile} packet
37442 Perform a file operation on the target system. For details,
37443 see @ref{Host I/O Packets}.
37445 @item vFlashErase:@var{addr},@var{length}
37446 @cindex @samp{vFlashErase} packet
37447 Direct the stub to erase @var{length} bytes of flash starting at
37448 @var{addr}. The region may enclose any number of flash blocks, but
37449 its start and end must fall on block boundaries, as indicated by the
37450 flash block size appearing in the memory map (@pxref{Memory Map
37451 Format}). @value{GDBN} groups flash memory programming operations
37452 together, and sends a @samp{vFlashDone} request after each group; the
37453 stub is allowed to delay erase operation until the @samp{vFlashDone}
37454 packet is received.
37464 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
37465 @cindex @samp{vFlashWrite} packet
37466 Direct the stub to write data to flash address @var{addr}. The data
37467 is passed in binary form using the same encoding as for the @samp{X}
37468 packet (@pxref{Binary Data}). The memory ranges specified by
37469 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
37470 not overlap, and must appear in order of increasing addresses
37471 (although @samp{vFlashErase} packets for higher addresses may already
37472 have been received; the ordering is guaranteed only between
37473 @samp{vFlashWrite} packets). If a packet writes to an address that was
37474 neither erased by a preceding @samp{vFlashErase} packet nor by some other
37475 target-specific method, the results are unpredictable.
37483 for vFlashWrite addressing non-flash memory
37489 @cindex @samp{vFlashDone} packet
37490 Indicate to the stub that flash programming operation is finished.
37491 The stub is permitted to delay or batch the effects of a group of
37492 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
37493 @samp{vFlashDone} packet is received. The contents of the affected
37494 regions of flash memory are unpredictable until the @samp{vFlashDone}
37495 request is completed.
37497 @item vKill;@var{pid}
37498 @cindex @samp{vKill} packet
37499 Kill the process with the specified process ID. @var{pid} is a
37500 hexadecimal integer identifying the process. This packet is used in
37501 preference to @samp{k} when multiprocess protocol extensions are
37502 supported; see @ref{multiprocess extensions}.
37512 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
37513 @cindex @samp{vRun} packet
37514 Run the program @var{filename}, passing it each @var{argument} on its
37515 command line. The file and arguments are hex-encoded strings. If
37516 @var{filename} is an empty string, the stub may use a default program
37517 (e.g.@: the last program run). The program is created in the stopped
37520 @c FIXME: What about non-stop mode?
37522 This packet is only available in extended mode (@pxref{extended mode}).
37528 @item @r{Any stop packet}
37529 for success (@pxref{Stop Reply Packets})
37533 @cindex @samp{vStopped} packet
37534 @xref{Notification Packets}.
37536 @item X @var{addr},@var{length}:@var{XX@dots{}}
37538 @cindex @samp{X} packet
37539 Write data to memory, where the data is transmitted in binary.
37540 @var{addr} is address, @var{length} is number of bytes,
37541 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
37551 @item z @var{type},@var{addr},@var{kind}
37552 @itemx Z @var{type},@var{addr},@var{kind}
37553 @anchor{insert breakpoint or watchpoint packet}
37554 @cindex @samp{z} packet
37555 @cindex @samp{Z} packets
37556 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
37557 watchpoint starting at address @var{address} of kind @var{kind}.
37559 Each breakpoint and watchpoint packet @var{type} is documented
37562 @emph{Implementation notes: A remote target shall return an empty string
37563 for an unrecognized breakpoint or watchpoint packet @var{type}. A
37564 remote target shall support either both or neither of a given
37565 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
37566 avoid potential problems with duplicate packets, the operations should
37567 be implemented in an idempotent way.}
37569 @item z0,@var{addr},@var{kind}
37570 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
37571 @cindex @samp{z0} packet
37572 @cindex @samp{Z0} packet
37573 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
37574 @var{addr} of type @var{kind}.
37576 A memory breakpoint is implemented by replacing the instruction at
37577 @var{addr} with a software breakpoint or trap instruction. The
37578 @var{kind} is target-specific and typically indicates the size of
37579 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
37580 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
37581 architectures have additional meanings for @var{kind};
37582 @var{cond_list} is an optional list of conditional expressions in bytecode
37583 form that should be evaluated on the target's side. These are the
37584 conditions that should be taken into consideration when deciding if
37585 the breakpoint trigger should be reported back to @var{GDBN}.
37587 The @var{cond_list} parameter is comprised of a series of expressions,
37588 concatenated without separators. Each expression has the following form:
37592 @item X @var{len},@var{expr}
37593 @var{len} is the length of the bytecode expression and @var{expr} is the
37594 actual conditional expression in bytecode form.
37598 The optional @var{cmd_list} parameter introduces commands that may be
37599 run on the target, rather than being reported back to @value{GDBN}.
37600 The parameter starts with a numeric flag @var{persist}; if the flag is
37601 nonzero, then the breakpoint may remain active and the commands
37602 continue to be run even when @value{GDBN} disconnects from the target.
37603 Following this flag is a series of expressions concatenated with no
37604 separators. Each expression has the following form:
37608 @item X @var{len},@var{expr}
37609 @var{len} is the length of the bytecode expression and @var{expr} is the
37610 actual conditional expression in bytecode form.
37614 see @ref{Architecture-Specific Protocol Details}.
37616 @emph{Implementation note: It is possible for a target to copy or move
37617 code that contains memory breakpoints (e.g., when implementing
37618 overlays). The behavior of this packet, in the presence of such a
37619 target, is not defined.}
37631 @item z1,@var{addr},@var{kind}
37632 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}
37633 @cindex @samp{z1} packet
37634 @cindex @samp{Z1} packet
37635 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
37636 address @var{addr}.
37638 A hardware breakpoint is implemented using a mechanism that is not
37639 dependant on being able to modify the target's memory. @var{kind}
37640 and @var{cond_list} have the same meaning as in @samp{Z0} packets.
37642 @emph{Implementation note: A hardware breakpoint is not affected by code
37655 @item z2,@var{addr},@var{kind}
37656 @itemx Z2,@var{addr},@var{kind}
37657 @cindex @samp{z2} packet
37658 @cindex @samp{Z2} packet
37659 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
37660 @var{kind} is interpreted as the number of bytes to watch.
37672 @item z3,@var{addr},@var{kind}
37673 @itemx Z3,@var{addr},@var{kind}
37674 @cindex @samp{z3} packet
37675 @cindex @samp{Z3} packet
37676 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
37677 @var{kind} is interpreted as the number of bytes to watch.
37689 @item z4,@var{addr},@var{kind}
37690 @itemx Z4,@var{addr},@var{kind}
37691 @cindex @samp{z4} packet
37692 @cindex @samp{Z4} packet
37693 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
37694 @var{kind} is interpreted as the number of bytes to watch.
37708 @node Stop Reply Packets
37709 @section Stop Reply Packets
37710 @cindex stop reply packets
37712 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
37713 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
37714 receive any of the below as a reply. Except for @samp{?}
37715 and @samp{vStopped}, that reply is only returned
37716 when the target halts. In the below the exact meaning of @dfn{signal
37717 number} is defined by the header @file{include/gdb/signals.h} in the
37718 @value{GDBN} source code.
37720 As in the description of request packets, we include spaces in the
37721 reply templates for clarity; these are not part of the reply packet's
37722 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
37728 The program received signal number @var{AA} (a two-digit hexadecimal
37729 number). This is equivalent to a @samp{T} response with no
37730 @var{n}:@var{r} pairs.
37732 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
37733 @cindex @samp{T} packet reply
37734 The program received signal number @var{AA} (a two-digit hexadecimal
37735 number). This is equivalent to an @samp{S} response, except that the
37736 @samp{@var{n}:@var{r}} pairs can carry values of important registers
37737 and other information directly in the stop reply packet, reducing
37738 round-trip latency. Single-step and breakpoint traps are reported
37739 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
37743 If @var{n} is a hexadecimal number, it is a register number, and the
37744 corresponding @var{r} gives that register's value. @var{r} is a
37745 series of bytes in target byte order, with each byte given by a
37746 two-digit hex number.
37749 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
37750 the stopped thread, as specified in @ref{thread-id syntax}.
37753 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
37754 the core on which the stop event was detected.
37757 If @var{n} is a recognized @dfn{stop reason}, it describes a more
37758 specific event that stopped the target. The currently defined stop
37759 reasons are listed below. @var{aa} should be @samp{05}, the trap
37760 signal. At most one stop reason should be present.
37763 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
37764 and go on to the next; this allows us to extend the protocol in the
37768 The currently defined stop reasons are:
37774 The packet indicates a watchpoint hit, and @var{r} is the data address, in
37777 @cindex shared library events, remote reply
37779 The packet indicates that the loaded libraries have changed.
37780 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
37781 list of loaded libraries. @var{r} is ignored.
37783 @cindex replay log events, remote reply
37785 The packet indicates that the target cannot continue replaying
37786 logged execution events, because it has reached the end (or the
37787 beginning when executing backward) of the log. The value of @var{r}
37788 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
37789 for more information.
37793 @itemx W @var{AA} ; process:@var{pid}
37794 The process exited, and @var{AA} is the exit status. This is only
37795 applicable to certain targets.
37797 The second form of the response, including the process ID of the exited
37798 process, can be used only when @value{GDBN} has reported support for
37799 multiprocess protocol extensions; see @ref{multiprocess extensions}.
37800 The @var{pid} is formatted as a big-endian hex string.
37803 @itemx X @var{AA} ; process:@var{pid}
37804 The process terminated with signal @var{AA}.
37806 The second form of the response, including the process ID of the
37807 terminated process, can be used only when @value{GDBN} has reported
37808 support for multiprocess protocol extensions; see @ref{multiprocess
37809 extensions}. The @var{pid} is formatted as a big-endian hex string.
37811 @item O @var{XX}@dots{}
37812 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
37813 written as the program's console output. This can happen at any time
37814 while the program is running and the debugger should continue to wait
37815 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
37817 @item F @var{call-id},@var{parameter}@dots{}
37818 @var{call-id} is the identifier which says which host system call should
37819 be called. This is just the name of the function. Translation into the
37820 correct system call is only applicable as it's defined in @value{GDBN}.
37821 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
37824 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
37825 this very system call.
37827 The target replies with this packet when it expects @value{GDBN} to
37828 call a host system call on behalf of the target. @value{GDBN} replies
37829 with an appropriate @samp{F} packet and keeps up waiting for the next
37830 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
37831 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
37832 Protocol Extension}, for more details.
37836 @node General Query Packets
37837 @section General Query Packets
37838 @cindex remote query requests
37840 Packets starting with @samp{q} are @dfn{general query packets};
37841 packets starting with @samp{Q} are @dfn{general set packets}. General
37842 query and set packets are a semi-unified form for retrieving and
37843 sending information to and from the stub.
37845 The initial letter of a query or set packet is followed by a name
37846 indicating what sort of thing the packet applies to. For example,
37847 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
37848 definitions with the stub. These packet names follow some
37853 The name must not contain commas, colons or semicolons.
37855 Most @value{GDBN} query and set packets have a leading upper case
37858 The names of custom vendor packets should use a company prefix, in
37859 lower case, followed by a period. For example, packets designed at
37860 the Acme Corporation might begin with @samp{qacme.foo} (for querying
37861 foos) or @samp{Qacme.bar} (for setting bars).
37864 The name of a query or set packet should be separated from any
37865 parameters by a @samp{:}; the parameters themselves should be
37866 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
37867 full packet name, and check for a separator or the end of the packet,
37868 in case two packet names share a common prefix. New packets should not begin
37869 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
37870 packets predate these conventions, and have arguments without any terminator
37871 for the packet name; we suspect they are in widespread use in places that
37872 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
37873 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
37876 Like the descriptions of the other packets, each description here
37877 has a template showing the packet's overall syntax, followed by an
37878 explanation of the packet's meaning. We include spaces in some of the
37879 templates for clarity; these are not part of the packet's syntax. No
37880 @value{GDBN} packet uses spaces to separate its components.
37882 Here are the currently defined query and set packets:
37888 Turn on or off the agent as a helper to perform some debugging operations
37889 delegated from @value{GDBN} (@pxref{Control Agent}).
37891 @item QAllow:@var{op}:@var{val}@dots{}
37892 @cindex @samp{QAllow} packet
37893 Specify which operations @value{GDBN} expects to request of the
37894 target, as a semicolon-separated list of operation name and value
37895 pairs. Possible values for @var{op} include @samp{WriteReg},
37896 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
37897 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
37898 indicating that @value{GDBN} will not request the operation, or 1,
37899 indicating that it may. (The target can then use this to set up its
37900 own internals optimally, for instance if the debugger never expects to
37901 insert breakpoints, it may not need to install its own trap handler.)
37904 @cindex current thread, remote request
37905 @cindex @samp{qC} packet
37906 Return the current thread ID.
37910 @item QC @var{thread-id}
37911 Where @var{thread-id} is a thread ID as documented in
37912 @ref{thread-id syntax}.
37913 @item @r{(anything else)}
37914 Any other reply implies the old thread ID.
37917 @item qCRC:@var{addr},@var{length}
37918 @cindex CRC of memory block, remote request
37919 @cindex @samp{qCRC} packet
37920 Compute the CRC checksum of a block of memory using CRC-32 defined in
37921 IEEE 802.3. The CRC is computed byte at a time, taking the most
37922 significant bit of each byte first. The initial pattern code
37923 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
37925 @emph{Note:} This is the same CRC used in validating separate debug
37926 files (@pxref{Separate Debug Files, , Debugging Information in Separate
37927 Files}). However the algorithm is slightly different. When validating
37928 separate debug files, the CRC is computed taking the @emph{least}
37929 significant bit of each byte first, and the final result is inverted to
37930 detect trailing zeros.
37935 An error (such as memory fault)
37936 @item C @var{crc32}
37937 The specified memory region's checksum is @var{crc32}.
37940 @item QDisableRandomization:@var{value}
37941 @cindex disable address space randomization, remote request
37942 @cindex @samp{QDisableRandomization} packet
37943 Some target operating systems will randomize the virtual address space
37944 of the inferior process as a security feature, but provide a feature
37945 to disable such randomization, e.g.@: to allow for a more deterministic
37946 debugging experience. On such systems, this packet with a @var{value}
37947 of 1 directs the target to disable address space randomization for
37948 processes subsequently started via @samp{vRun} packets, while a packet
37949 with a @var{value} of 0 tells the target to enable address space
37952 This packet is only available in extended mode (@pxref{extended mode}).
37957 The request succeeded.
37960 An error occurred. @var{nn} are hex digits.
37963 An empty reply indicates that @samp{QDisableRandomization} is not supported
37967 This packet is not probed by default; the remote stub must request it,
37968 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37969 This should only be done on targets that actually support disabling
37970 address space randomization.
37973 @itemx qsThreadInfo
37974 @cindex list active threads, remote request
37975 @cindex @samp{qfThreadInfo} packet
37976 @cindex @samp{qsThreadInfo} packet
37977 Obtain a list of all active thread IDs from the target (OS). Since there
37978 may be too many active threads to fit into one reply packet, this query
37979 works iteratively: it may require more than one query/reply sequence to
37980 obtain the entire list of threads. The first query of the sequence will
37981 be the @samp{qfThreadInfo} query; subsequent queries in the
37982 sequence will be the @samp{qsThreadInfo} query.
37984 NOTE: This packet replaces the @samp{qL} query (see below).
37988 @item m @var{thread-id}
37990 @item m @var{thread-id},@var{thread-id}@dots{}
37991 a comma-separated list of thread IDs
37993 (lower case letter @samp{L}) denotes end of list.
37996 In response to each query, the target will reply with a list of one or
37997 more thread IDs, separated by commas.
37998 @value{GDBN} will respond to each reply with a request for more thread
37999 ids (using the @samp{qs} form of the query), until the target responds
38000 with @samp{l} (lower-case ell, for @dfn{last}).
38001 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
38004 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
38005 @cindex get thread-local storage address, remote request
38006 @cindex @samp{qGetTLSAddr} packet
38007 Fetch the address associated with thread local storage specified
38008 by @var{thread-id}, @var{offset}, and @var{lm}.
38010 @var{thread-id} is the thread ID associated with the
38011 thread for which to fetch the TLS address. @xref{thread-id syntax}.
38013 @var{offset} is the (big endian, hex encoded) offset associated with the
38014 thread local variable. (This offset is obtained from the debug
38015 information associated with the variable.)
38017 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
38018 load module associated with the thread local storage. For example,
38019 a @sc{gnu}/Linux system will pass the link map address of the shared
38020 object associated with the thread local storage under consideration.
38021 Other operating environments may choose to represent the load module
38022 differently, so the precise meaning of this parameter will vary.
38026 @item @var{XX}@dots{}
38027 Hex encoded (big endian) bytes representing the address of the thread
38028 local storage requested.
38031 An error occurred. @var{nn} are hex digits.
38034 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
38037 @item qGetTIBAddr:@var{thread-id}
38038 @cindex get thread information block address
38039 @cindex @samp{qGetTIBAddr} packet
38040 Fetch address of the Windows OS specific Thread Information Block.
38042 @var{thread-id} is the thread ID associated with the thread.
38046 @item @var{XX}@dots{}
38047 Hex encoded (big endian) bytes representing the linear address of the
38048 thread information block.
38051 An error occured. This means that either the thread was not found, or the
38052 address could not be retrieved.
38055 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
38058 @item qL @var{startflag} @var{threadcount} @var{nextthread}
38059 Obtain thread information from RTOS. Where: @var{startflag} (one hex
38060 digit) is one to indicate the first query and zero to indicate a
38061 subsequent query; @var{threadcount} (two hex digits) is the maximum
38062 number of threads the response packet can contain; and @var{nextthread}
38063 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
38064 returned in the response as @var{argthread}.
38066 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
38070 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
38071 Where: @var{count} (two hex digits) is the number of threads being
38072 returned; @var{done} (one hex digit) is zero to indicate more threads
38073 and one indicates no further threads; @var{argthreadid} (eight hex
38074 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
38075 is a sequence of thread IDs from the target. @var{threadid} (eight hex
38076 digits). See @code{remote.c:parse_threadlist_response()}.
38080 @cindex section offsets, remote request
38081 @cindex @samp{qOffsets} packet
38082 Get section offsets that the target used when relocating the downloaded
38087 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
38088 Relocate the @code{Text} section by @var{xxx} from its original address.
38089 Relocate the @code{Data} section by @var{yyy} from its original address.
38090 If the object file format provides segment information (e.g.@: @sc{elf}
38091 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
38092 segments by the supplied offsets.
38094 @emph{Note: while a @code{Bss} offset may be included in the response,
38095 @value{GDBN} ignores this and instead applies the @code{Data} offset
38096 to the @code{Bss} section.}
38098 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
38099 Relocate the first segment of the object file, which conventionally
38100 contains program code, to a starting address of @var{xxx}. If
38101 @samp{DataSeg} is specified, relocate the second segment, which
38102 conventionally contains modifiable data, to a starting address of
38103 @var{yyy}. @value{GDBN} will report an error if the object file
38104 does not contain segment information, or does not contain at least
38105 as many segments as mentioned in the reply. Extra segments are
38106 kept at fixed offsets relative to the last relocated segment.
38109 @item qP @var{mode} @var{thread-id}
38110 @cindex thread information, remote request
38111 @cindex @samp{qP} packet
38112 Returns information on @var{thread-id}. Where: @var{mode} is a hex
38113 encoded 32 bit mode; @var{thread-id} is a thread ID
38114 (@pxref{thread-id syntax}).
38116 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
38119 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
38123 @cindex non-stop mode, remote request
38124 @cindex @samp{QNonStop} packet
38126 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
38127 @xref{Remote Non-Stop}, for more information.
38132 The request succeeded.
38135 An error occurred. @var{nn} are hex digits.
38138 An empty reply indicates that @samp{QNonStop} is not supported by
38142 This packet is not probed by default; the remote stub must request it,
38143 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38144 Use of this packet is controlled by the @code{set non-stop} command;
38145 @pxref{Non-Stop Mode}.
38147 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
38148 @cindex pass signals to inferior, remote request
38149 @cindex @samp{QPassSignals} packet
38150 @anchor{QPassSignals}
38151 Each listed @var{signal} should be passed directly to the inferior process.
38152 Signals are numbered identically to continue packets and stop replies
38153 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
38154 strictly greater than the previous item. These signals do not need to stop
38155 the inferior, or be reported to @value{GDBN}. All other signals should be
38156 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
38157 combine; any earlier @samp{QPassSignals} list is completely replaced by the
38158 new list. This packet improves performance when using @samp{handle
38159 @var{signal} nostop noprint pass}.
38164 The request succeeded.
38167 An error occurred. @var{nn} are hex digits.
38170 An empty reply indicates that @samp{QPassSignals} is not supported by
38174 Use of this packet is controlled by the @code{set remote pass-signals}
38175 command (@pxref{Remote Configuration, set remote pass-signals}).
38176 This packet is not probed by default; the remote stub must request it,
38177 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38179 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
38180 @cindex signals the inferior may see, remote request
38181 @cindex @samp{QProgramSignals} packet
38182 @anchor{QProgramSignals}
38183 Each listed @var{signal} may be delivered to the inferior process.
38184 Others should be silently discarded.
38186 In some cases, the remote stub may need to decide whether to deliver a
38187 signal to the program or not without @value{GDBN} involvement. One
38188 example of that is while detaching --- the program's threads may have
38189 stopped for signals that haven't yet had a chance of being reported to
38190 @value{GDBN}, and so the remote stub can use the signal list specified
38191 by this packet to know whether to deliver or ignore those pending
38194 This does not influence whether to deliver a signal as requested by a
38195 resumption packet (@pxref{vCont packet}).
38197 Signals are numbered identically to continue packets and stop replies
38198 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
38199 strictly greater than the previous item. Multiple
38200 @samp{QProgramSignals} packets do not combine; any earlier
38201 @samp{QProgramSignals} list is completely replaced by the new list.
38206 The request succeeded.
38209 An error occurred. @var{nn} are hex digits.
38212 An empty reply indicates that @samp{QProgramSignals} is not supported
38216 Use of this packet is controlled by the @code{set remote program-signals}
38217 command (@pxref{Remote Configuration, set remote program-signals}).
38218 This packet is not probed by default; the remote stub must request it,
38219 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38221 @item qRcmd,@var{command}
38222 @cindex execute remote command, remote request
38223 @cindex @samp{qRcmd} packet
38224 @var{command} (hex encoded) is passed to the local interpreter for
38225 execution. Invalid commands should be reported using the output
38226 string. Before the final result packet, the target may also respond
38227 with a number of intermediate @samp{O@var{output}} console output
38228 packets. @emph{Implementors should note that providing access to a
38229 stubs's interpreter may have security implications}.
38234 A command response with no output.
38236 A command response with the hex encoded output string @var{OUTPUT}.
38238 Indicate a badly formed request.
38240 An empty reply indicates that @samp{qRcmd} is not recognized.
38243 (Note that the @code{qRcmd} packet's name is separated from the
38244 command by a @samp{,}, not a @samp{:}, contrary to the naming
38245 conventions above. Please don't use this packet as a model for new
38248 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
38249 @cindex searching memory, in remote debugging
38251 @cindex @samp{qSearch:memory} packet
38253 @cindex @samp{qSearch memory} packet
38254 @anchor{qSearch memory}
38255 Search @var{length} bytes at @var{address} for @var{search-pattern}.
38256 @var{address} and @var{length} are encoded in hex.
38257 @var{search-pattern} is a sequence of bytes, hex encoded.
38262 The pattern was not found.
38264 The pattern was found at @var{address}.
38266 A badly formed request or an error was encountered while searching memory.
38268 An empty reply indicates that @samp{qSearch:memory} is not recognized.
38271 @item QStartNoAckMode
38272 @cindex @samp{QStartNoAckMode} packet
38273 @anchor{QStartNoAckMode}
38274 Request that the remote stub disable the normal @samp{+}/@samp{-}
38275 protocol acknowledgments (@pxref{Packet Acknowledgment}).
38280 The stub has switched to no-acknowledgment mode.
38281 @value{GDBN} acknowledges this reponse,
38282 but neither the stub nor @value{GDBN} shall send or expect further
38283 @samp{+}/@samp{-} acknowledgments in the current connection.
38285 An empty reply indicates that the stub does not support no-acknowledgment mode.
38288 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
38289 @cindex supported packets, remote query
38290 @cindex features of the remote protocol
38291 @cindex @samp{qSupported} packet
38292 @anchor{qSupported}
38293 Tell the remote stub about features supported by @value{GDBN}, and
38294 query the stub for features it supports. This packet allows
38295 @value{GDBN} and the remote stub to take advantage of each others'
38296 features. @samp{qSupported} also consolidates multiple feature probes
38297 at startup, to improve @value{GDBN} performance---a single larger
38298 packet performs better than multiple smaller probe packets on
38299 high-latency links. Some features may enable behavior which must not
38300 be on by default, e.g.@: because it would confuse older clients or
38301 stubs. Other features may describe packets which could be
38302 automatically probed for, but are not. These features must be
38303 reported before @value{GDBN} will use them. This ``default
38304 unsupported'' behavior is not appropriate for all packets, but it
38305 helps to keep the initial connection time under control with new
38306 versions of @value{GDBN} which support increasing numbers of packets.
38310 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
38311 The stub supports or does not support each returned @var{stubfeature},
38312 depending on the form of each @var{stubfeature} (see below for the
38315 An empty reply indicates that @samp{qSupported} is not recognized,
38316 or that no features needed to be reported to @value{GDBN}.
38319 The allowed forms for each feature (either a @var{gdbfeature} in the
38320 @samp{qSupported} packet, or a @var{stubfeature} in the response)
38324 @item @var{name}=@var{value}
38325 The remote protocol feature @var{name} is supported, and associated
38326 with the specified @var{value}. The format of @var{value} depends
38327 on the feature, but it must not include a semicolon.
38329 The remote protocol feature @var{name} is supported, and does not
38330 need an associated value.
38332 The remote protocol feature @var{name} is not supported.
38334 The remote protocol feature @var{name} may be supported, and
38335 @value{GDBN} should auto-detect support in some other way when it is
38336 needed. This form will not be used for @var{gdbfeature} notifications,
38337 but may be used for @var{stubfeature} responses.
38340 Whenever the stub receives a @samp{qSupported} request, the
38341 supplied set of @value{GDBN} features should override any previous
38342 request. This allows @value{GDBN} to put the stub in a known
38343 state, even if the stub had previously been communicating with
38344 a different version of @value{GDBN}.
38346 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
38351 This feature indicates whether @value{GDBN} supports multiprocess
38352 extensions to the remote protocol. @value{GDBN} does not use such
38353 extensions unless the stub also reports that it supports them by
38354 including @samp{multiprocess+} in its @samp{qSupported} reply.
38355 @xref{multiprocess extensions}, for details.
38358 This feature indicates that @value{GDBN} supports the XML target
38359 description. If the stub sees @samp{xmlRegisters=} with target
38360 specific strings separated by a comma, it will report register
38364 This feature indicates whether @value{GDBN} supports the
38365 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
38366 instruction reply packet}).
38369 Stubs should ignore any unknown values for
38370 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
38371 packet supports receiving packets of unlimited length (earlier
38372 versions of @value{GDBN} may reject overly long responses). Additional values
38373 for @var{gdbfeature} may be defined in the future to let the stub take
38374 advantage of new features in @value{GDBN}, e.g.@: incompatible
38375 improvements in the remote protocol---the @samp{multiprocess} feature is
38376 an example of such a feature. The stub's reply should be independent
38377 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
38378 describes all the features it supports, and then the stub replies with
38379 all the features it supports.
38381 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
38382 responses, as long as each response uses one of the standard forms.
38384 Some features are flags. A stub which supports a flag feature
38385 should respond with a @samp{+} form response. Other features
38386 require values, and the stub should respond with an @samp{=}
38389 Each feature has a default value, which @value{GDBN} will use if
38390 @samp{qSupported} is not available or if the feature is not mentioned
38391 in the @samp{qSupported} response. The default values are fixed; a
38392 stub is free to omit any feature responses that match the defaults.
38394 Not all features can be probed, but for those which can, the probing
38395 mechanism is useful: in some cases, a stub's internal
38396 architecture may not allow the protocol layer to know some information
38397 about the underlying target in advance. This is especially common in
38398 stubs which may be configured for multiple targets.
38400 These are the currently defined stub features and their properties:
38402 @multitable @columnfractions 0.35 0.2 0.12 0.2
38403 @c NOTE: The first row should be @headitem, but we do not yet require
38404 @c a new enough version of Texinfo (4.7) to use @headitem.
38406 @tab Value Required
38410 @item @samp{PacketSize}
38415 @item @samp{qXfer:auxv:read}
38420 @item @samp{qXfer:btrace:read}
38425 @item @samp{qXfer:features:read}
38430 @item @samp{qXfer:libraries:read}
38435 @item @samp{qXfer:memory-map:read}
38440 @item @samp{qXfer:sdata:read}
38445 @item @samp{qXfer:spu:read}
38450 @item @samp{qXfer:spu:write}
38455 @item @samp{qXfer:siginfo:read}
38460 @item @samp{qXfer:siginfo:write}
38465 @item @samp{qXfer:threads:read}
38470 @item @samp{qXfer:traceframe-info:read}
38475 @item @samp{qXfer:uib:read}
38480 @item @samp{qXfer:fdpic:read}
38485 @item @samp{Qbtrace:off}
38490 @item @samp{Qbtrace:bts}
38495 @item @samp{QNonStop}
38500 @item @samp{QPassSignals}
38505 @item @samp{QStartNoAckMode}
38510 @item @samp{multiprocess}
38515 @item @samp{ConditionalBreakpoints}
38520 @item @samp{ConditionalTracepoints}
38525 @item @samp{ReverseContinue}
38530 @item @samp{ReverseStep}
38535 @item @samp{TracepointSource}
38540 @item @samp{QAgent}
38545 @item @samp{QAllow}
38550 @item @samp{QDisableRandomization}
38555 @item @samp{EnableDisableTracepoints}
38560 @item @samp{QTBuffer:size}
38565 @item @samp{tracenz}
38570 @item @samp{BreakpointCommands}
38577 These are the currently defined stub features, in more detail:
38580 @cindex packet size, remote protocol
38581 @item PacketSize=@var{bytes}
38582 The remote stub can accept packets up to at least @var{bytes} in
38583 length. @value{GDBN} will send packets up to this size for bulk
38584 transfers, and will never send larger packets. This is a limit on the
38585 data characters in the packet, including the frame and checksum.
38586 There is no trailing NUL byte in a remote protocol packet; if the stub
38587 stores packets in a NUL-terminated format, it should allow an extra
38588 byte in its buffer for the NUL. If this stub feature is not supported,
38589 @value{GDBN} guesses based on the size of the @samp{g} packet response.
38591 @item qXfer:auxv:read
38592 The remote stub understands the @samp{qXfer:auxv:read} packet
38593 (@pxref{qXfer auxiliary vector read}).
38595 @item qXfer:btrace:read
38596 The remote stub understands the @samp{qXfer:btrace:read}
38597 packet (@pxref{qXfer btrace read}).
38599 @item qXfer:features:read
38600 The remote stub understands the @samp{qXfer:features:read} packet
38601 (@pxref{qXfer target description read}).
38603 @item qXfer:libraries:read
38604 The remote stub understands the @samp{qXfer:libraries:read} packet
38605 (@pxref{qXfer library list read}).
38607 @item qXfer:libraries-svr4:read
38608 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
38609 (@pxref{qXfer svr4 library list read}).
38611 @item qXfer:memory-map:read
38612 The remote stub understands the @samp{qXfer:memory-map:read} packet
38613 (@pxref{qXfer memory map read}).
38615 @item qXfer:sdata:read
38616 The remote stub understands the @samp{qXfer:sdata:read} packet
38617 (@pxref{qXfer sdata read}).
38619 @item qXfer:spu:read
38620 The remote stub understands the @samp{qXfer:spu:read} packet
38621 (@pxref{qXfer spu read}).
38623 @item qXfer:spu:write
38624 The remote stub understands the @samp{qXfer:spu:write} packet
38625 (@pxref{qXfer spu write}).
38627 @item qXfer:siginfo:read
38628 The remote stub understands the @samp{qXfer:siginfo:read} packet
38629 (@pxref{qXfer siginfo read}).
38631 @item qXfer:siginfo:write
38632 The remote stub understands the @samp{qXfer:siginfo:write} packet
38633 (@pxref{qXfer siginfo write}).
38635 @item qXfer:threads:read
38636 The remote stub understands the @samp{qXfer:threads:read} packet
38637 (@pxref{qXfer threads read}).
38639 @item qXfer:traceframe-info:read
38640 The remote stub understands the @samp{qXfer:traceframe-info:read}
38641 packet (@pxref{qXfer traceframe info read}).
38643 @item qXfer:uib:read
38644 The remote stub understands the @samp{qXfer:uib:read}
38645 packet (@pxref{qXfer unwind info block}).
38647 @item qXfer:fdpic:read
38648 The remote stub understands the @samp{qXfer:fdpic:read}
38649 packet (@pxref{qXfer fdpic loadmap read}).
38652 The remote stub understands the @samp{QNonStop} packet
38653 (@pxref{QNonStop}).
38656 The remote stub understands the @samp{QPassSignals} packet
38657 (@pxref{QPassSignals}).
38659 @item QStartNoAckMode
38660 The remote stub understands the @samp{QStartNoAckMode} packet and
38661 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
38664 @anchor{multiprocess extensions}
38665 @cindex multiprocess extensions, in remote protocol
38666 The remote stub understands the multiprocess extensions to the remote
38667 protocol syntax. The multiprocess extensions affect the syntax of
38668 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
38669 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
38670 replies. Note that reporting this feature indicates support for the
38671 syntactic extensions only, not that the stub necessarily supports
38672 debugging of more than one process at a time. The stub must not use
38673 multiprocess extensions in packet replies unless @value{GDBN} has also
38674 indicated it supports them in its @samp{qSupported} request.
38676 @item qXfer:osdata:read
38677 The remote stub understands the @samp{qXfer:osdata:read} packet
38678 ((@pxref{qXfer osdata read}).
38680 @item ConditionalBreakpoints
38681 The target accepts and implements evaluation of conditional expressions
38682 defined for breakpoints. The target will only report breakpoint triggers
38683 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
38685 @item ConditionalTracepoints
38686 The remote stub accepts and implements conditional expressions defined
38687 for tracepoints (@pxref{Tracepoint Conditions}).
38689 @item ReverseContinue
38690 The remote stub accepts and implements the reverse continue packet
38694 The remote stub accepts and implements the reverse step packet
38697 @item TracepointSource
38698 The remote stub understands the @samp{QTDPsrc} packet that supplies
38699 the source form of tracepoint definitions.
38702 The remote stub understands the @samp{QAgent} packet.
38705 The remote stub understands the @samp{QAllow} packet.
38707 @item QDisableRandomization
38708 The remote stub understands the @samp{QDisableRandomization} packet.
38710 @item StaticTracepoint
38711 @cindex static tracepoints, in remote protocol
38712 The remote stub supports static tracepoints.
38714 @item InstallInTrace
38715 @anchor{install tracepoint in tracing}
38716 The remote stub supports installing tracepoint in tracing.
38718 @item EnableDisableTracepoints
38719 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
38720 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
38721 to be enabled and disabled while a trace experiment is running.
38723 @item QTBuffer:size
38724 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
38725 packet that allows to change the size of the trace buffer.
38728 @cindex string tracing, in remote protocol
38729 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
38730 See @ref{Bytecode Descriptions} for details about the bytecode.
38732 @item BreakpointCommands
38733 @cindex breakpoint commands, in remote protocol
38734 The remote stub supports running a breakpoint's command list itself,
38735 rather than reporting the hit to @value{GDBN}.
38738 The remote stub understands the @samp{Qbtrace:off} packet.
38741 The remote stub understands the @samp{Qbtrace:bts} packet.
38746 @cindex symbol lookup, remote request
38747 @cindex @samp{qSymbol} packet
38748 Notify the target that @value{GDBN} is prepared to serve symbol lookup
38749 requests. Accept requests from the target for the values of symbols.
38754 The target does not need to look up any (more) symbols.
38755 @item qSymbol:@var{sym_name}
38756 The target requests the value of symbol @var{sym_name} (hex encoded).
38757 @value{GDBN} may provide the value by using the
38758 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
38762 @item qSymbol:@var{sym_value}:@var{sym_name}
38763 Set the value of @var{sym_name} to @var{sym_value}.
38765 @var{sym_name} (hex encoded) is the name of a symbol whose value the
38766 target has previously requested.
38768 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
38769 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
38775 The target does not need to look up any (more) symbols.
38776 @item qSymbol:@var{sym_name}
38777 The target requests the value of a new symbol @var{sym_name} (hex
38778 encoded). @value{GDBN} will continue to supply the values of symbols
38779 (if available), until the target ceases to request them.
38784 @itemx QTDisconnected
38791 @itemx qTMinFTPILen
38793 @xref{Tracepoint Packets}.
38795 @item qThreadExtraInfo,@var{thread-id}
38796 @cindex thread attributes info, remote request
38797 @cindex @samp{qThreadExtraInfo} packet
38798 Obtain a printable string description of a thread's attributes from
38799 the target OS. @var{thread-id} is a thread ID;
38800 see @ref{thread-id syntax}. This
38801 string may contain anything that the target OS thinks is interesting
38802 for @value{GDBN} to tell the user about the thread. The string is
38803 displayed in @value{GDBN}'s @code{info threads} display. Some
38804 examples of possible thread extra info strings are @samp{Runnable}, or
38805 @samp{Blocked on Mutex}.
38809 @item @var{XX}@dots{}
38810 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
38811 comprising the printable string containing the extra information about
38812 the thread's attributes.
38815 (Note that the @code{qThreadExtraInfo} packet's name is separated from
38816 the command by a @samp{,}, not a @samp{:}, contrary to the naming
38817 conventions above. Please don't use this packet as a model for new
38836 @xref{Tracepoint Packets}.
38838 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
38839 @cindex read special object, remote request
38840 @cindex @samp{qXfer} packet
38841 @anchor{qXfer read}
38842 Read uninterpreted bytes from the target's special data area
38843 identified by the keyword @var{object}. Request @var{length} bytes
38844 starting at @var{offset} bytes into the data. The content and
38845 encoding of @var{annex} is specific to @var{object}; it can supply
38846 additional details about what data to access.
38848 Here are the specific requests of this form defined so far. All
38849 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
38850 formats, listed below.
38853 @item qXfer:auxv:read::@var{offset},@var{length}
38854 @anchor{qXfer auxiliary vector read}
38855 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
38856 auxiliary vector}. Note @var{annex} must be empty.
38858 This packet is not probed by default; the remote stub must request it,
38859 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38861 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
38862 @anchor{qXfer btrace read}
38864 Return a description of the current branch trace.
38865 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
38866 packet may have one of the following values:
38870 Returns all available branch trace.
38873 Returns all available branch trace if the branch trace changed since
38874 the last read request.
38877 This packet is not probed by default; the remote stub must request it
38878 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38880 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
38881 @anchor{qXfer target description read}
38882 Access the @dfn{target description}. @xref{Target Descriptions}. The
38883 annex specifies which XML document to access. The main description is
38884 always loaded from the @samp{target.xml} annex.
38886 This packet is not probed by default; the remote stub must request it,
38887 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38889 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
38890 @anchor{qXfer library list read}
38891 Access the target's list of loaded libraries. @xref{Library List Format}.
38892 The annex part of the generic @samp{qXfer} packet must be empty
38893 (@pxref{qXfer read}).
38895 Targets which maintain a list of libraries in the program's memory do
38896 not need to implement this packet; it is designed for platforms where
38897 the operating system manages the list of loaded libraries.
38899 This packet is not probed by default; the remote stub must request it,
38900 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38902 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
38903 @anchor{qXfer svr4 library list read}
38904 Access the target's list of loaded libraries when the target is an SVR4
38905 platform. @xref{Library List Format for SVR4 Targets}. The annex part
38906 of the generic @samp{qXfer} packet must be empty (@pxref{qXfer read}).
38908 This packet is optional for better performance on SVR4 targets.
38909 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
38911 This packet is not probed by default; the remote stub must request it,
38912 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38914 @item qXfer:memory-map:read::@var{offset},@var{length}
38915 @anchor{qXfer memory map read}
38916 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
38917 annex part of the generic @samp{qXfer} packet must be empty
38918 (@pxref{qXfer read}).
38920 This packet is not probed by default; the remote stub must request it,
38921 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38923 @item qXfer:sdata:read::@var{offset},@var{length}
38924 @anchor{qXfer sdata read}
38926 Read contents of the extra collected static tracepoint marker
38927 information. The annex part of the generic @samp{qXfer} packet must
38928 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
38931 This packet is not probed by default; the remote stub must request it,
38932 by supplying an appropriate @samp{qSupported} response
38933 (@pxref{qSupported}).
38935 @item qXfer:siginfo:read::@var{offset},@var{length}
38936 @anchor{qXfer siginfo read}
38937 Read contents of the extra signal information on the target
38938 system. The annex part of the generic @samp{qXfer} packet must be
38939 empty (@pxref{qXfer read}).
38941 This packet is not probed by default; the remote stub must request it,
38942 by supplying an appropriate @samp{qSupported} response
38943 (@pxref{qSupported}).
38945 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
38946 @anchor{qXfer spu read}
38947 Read contents of an @code{spufs} file on the target system. The
38948 annex specifies which file to read; it must be of the form
38949 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
38950 in the target process, and @var{name} identifes the @code{spufs} file
38951 in that context to be accessed.
38953 This packet is not probed by default; the remote stub must request it,
38954 by supplying an appropriate @samp{qSupported} response
38955 (@pxref{qSupported}).
38957 @item qXfer:threads:read::@var{offset},@var{length}
38958 @anchor{qXfer threads read}
38959 Access the list of threads on target. @xref{Thread List Format}. The
38960 annex part of the generic @samp{qXfer} packet must be empty
38961 (@pxref{qXfer read}).
38963 This packet is not probed by default; the remote stub must request it,
38964 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38966 @item qXfer:traceframe-info:read::@var{offset},@var{length}
38967 @anchor{qXfer traceframe info read}
38969 Return a description of the current traceframe's contents.
38970 @xref{Traceframe Info Format}. The annex part of the generic
38971 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
38973 This packet is not probed by default; the remote stub must request it,
38974 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38976 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
38977 @anchor{qXfer unwind info block}
38979 Return the unwind information block for @var{pc}. This packet is used
38980 on OpenVMS/ia64 to ask the kernel unwind information.
38982 This packet is not probed by default.
38984 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
38985 @anchor{qXfer fdpic loadmap read}
38986 Read contents of @code{loadmap}s on the target system. The
38987 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
38988 executable @code{loadmap} or interpreter @code{loadmap} to read.
38990 This packet is not probed by default; the remote stub must request it,
38991 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38993 @item qXfer:osdata:read::@var{offset},@var{length}
38994 @anchor{qXfer osdata read}
38995 Access the target's @dfn{operating system information}.
38996 @xref{Operating System Information}.
39003 Data @var{data} (@pxref{Binary Data}) has been read from the
39004 target. There may be more data at a higher address (although
39005 it is permitted to return @samp{m} even for the last valid
39006 block of data, as long as at least one byte of data was read).
39007 @var{data} may have fewer bytes than the @var{length} in the
39011 Data @var{data} (@pxref{Binary Data}) has been read from the target.
39012 There is no more data to be read. @var{data} may have fewer bytes
39013 than the @var{length} in the request.
39016 The @var{offset} in the request is at the end of the data.
39017 There is no more data to be read.
39020 The request was malformed, or @var{annex} was invalid.
39023 The offset was invalid, or there was an error encountered reading the data.
39024 @var{nn} is a hex-encoded @code{errno} value.
39027 An empty reply indicates the @var{object} string was not recognized by
39028 the stub, or that the object does not support reading.
39031 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
39032 @cindex write data into object, remote request
39033 @anchor{qXfer write}
39034 Write uninterpreted bytes into the target's special data area
39035 identified by the keyword @var{object}, starting at @var{offset} bytes
39036 into the data. @var{data}@dots{} is the binary-encoded data
39037 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
39038 is specific to @var{object}; it can supply additional details about what data
39041 Here are the specific requests of this form defined so far. All
39042 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
39043 formats, listed below.
39046 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
39047 @anchor{qXfer siginfo write}
39048 Write @var{data} to the extra signal information on the target system.
39049 The annex part of the generic @samp{qXfer} packet must be
39050 empty (@pxref{qXfer write}).
39052 This packet is not probed by default; the remote stub must request it,
39053 by supplying an appropriate @samp{qSupported} response
39054 (@pxref{qSupported}).
39056 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
39057 @anchor{qXfer spu write}
39058 Write @var{data} to an @code{spufs} file on the target system. The
39059 annex specifies which file to write; it must be of the form
39060 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
39061 in the target process, and @var{name} identifes the @code{spufs} file
39062 in that context to be accessed.
39064 This packet is not probed by default; the remote stub must request it,
39065 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39071 @var{nn} (hex encoded) is the number of bytes written.
39072 This may be fewer bytes than supplied in the request.
39075 The request was malformed, or @var{annex} was invalid.
39078 The offset was invalid, or there was an error encountered writing the data.
39079 @var{nn} is a hex-encoded @code{errno} value.
39082 An empty reply indicates the @var{object} string was not
39083 recognized by the stub, or that the object does not support writing.
39086 @item qXfer:@var{object}:@var{operation}:@dots{}
39087 Requests of this form may be added in the future. When a stub does
39088 not recognize the @var{object} keyword, or its support for
39089 @var{object} does not recognize the @var{operation} keyword, the stub
39090 must respond with an empty packet.
39092 @item qAttached:@var{pid}
39093 @cindex query attached, remote request
39094 @cindex @samp{qAttached} packet
39095 Return an indication of whether the remote server attached to an
39096 existing process or created a new process. When the multiprocess
39097 protocol extensions are supported (@pxref{multiprocess extensions}),
39098 @var{pid} is an integer in hexadecimal format identifying the target
39099 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
39100 the query packet will be simplified as @samp{qAttached}.
39102 This query is used, for example, to know whether the remote process
39103 should be detached or killed when a @value{GDBN} session is ended with
39104 the @code{quit} command.
39109 The remote server attached to an existing process.
39111 The remote server created a new process.
39113 A badly formed request or an error was encountered.
39117 Enable branch tracing for the current thread using bts tracing.
39122 Branch tracing has been enabled.
39124 A badly formed request or an error was encountered.
39128 Disable branch tracing for the current thread.
39133 Branch tracing has been disabled.
39135 A badly formed request or an error was encountered.
39140 @node Architecture-Specific Protocol Details
39141 @section Architecture-Specific Protocol Details
39143 This section describes how the remote protocol is applied to specific
39144 target architectures. Also see @ref{Standard Target Features}, for
39145 details of XML target descriptions for each architecture.
39148 * ARM-Specific Protocol Details::
39149 * MIPS-Specific Protocol Details::
39152 @node ARM-Specific Protocol Details
39153 @subsection @acronym{ARM}-specific Protocol Details
39156 * ARM Breakpoint Kinds::
39159 @node ARM Breakpoint Kinds
39160 @subsubsection @acronym{ARM} Breakpoint Kinds
39161 @cindex breakpoint kinds, @acronym{ARM}
39163 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
39168 16-bit Thumb mode breakpoint.
39171 32-bit Thumb mode (Thumb-2) breakpoint.
39174 32-bit @acronym{ARM} mode breakpoint.
39178 @node MIPS-Specific Protocol Details
39179 @subsection @acronym{MIPS}-specific Protocol Details
39182 * MIPS Register packet Format::
39183 * MIPS Breakpoint Kinds::
39186 @node MIPS Register packet Format
39187 @subsubsection @acronym{MIPS} Register Packet Format
39188 @cindex register packet format, @acronym{MIPS}
39190 The following @code{g}/@code{G} packets have previously been defined.
39191 In the below, some thirty-two bit registers are transferred as
39192 sixty-four bits. Those registers should be zero/sign extended (which?)
39193 to fill the space allocated. Register bytes are transferred in target
39194 byte order. The two nibbles within a register byte are transferred
39195 most-significant -- least-significant.
39200 All registers are transferred as thirty-two bit quantities in the order:
39201 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
39202 registers; fsr; fir; fp.
39205 All registers are transferred as sixty-four bit quantities (including
39206 thirty-two bit registers such as @code{sr}). The ordering is the same
39211 @node MIPS Breakpoint Kinds
39212 @subsubsection @acronym{MIPS} Breakpoint Kinds
39213 @cindex breakpoint kinds, @acronym{MIPS}
39215 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
39220 16-bit @acronym{MIPS16} mode breakpoint.
39223 16-bit @acronym{microMIPS} mode breakpoint.
39226 32-bit standard @acronym{MIPS} mode breakpoint.
39229 32-bit @acronym{microMIPS} mode breakpoint.
39233 @node Tracepoint Packets
39234 @section Tracepoint Packets
39235 @cindex tracepoint packets
39236 @cindex packets, tracepoint
39238 Here we describe the packets @value{GDBN} uses to implement
39239 tracepoints (@pxref{Tracepoints}).
39243 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
39244 @cindex @samp{QTDP} packet
39245 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
39246 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
39247 the tracepoint is disabled. @var{step} is the tracepoint's step
39248 count, and @var{pass} is its pass count. If an @samp{F} is present,
39249 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
39250 the number of bytes that the target should copy elsewhere to make room
39251 for the tracepoint. If an @samp{X} is present, it introduces a
39252 tracepoint condition, which consists of a hexadecimal length, followed
39253 by a comma and hex-encoded bytes, in a manner similar to action
39254 encodings as described below. If the trailing @samp{-} is present,
39255 further @samp{QTDP} packets will follow to specify this tracepoint's
39261 The packet was understood and carried out.
39263 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
39265 The packet was not recognized.
39268 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
39269 Define actions to be taken when a tracepoint is hit. @var{n} and
39270 @var{addr} must be the same as in the initial @samp{QTDP} packet for
39271 this tracepoint. This packet may only be sent immediately after
39272 another @samp{QTDP} packet that ended with a @samp{-}. If the
39273 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
39274 specifying more actions for this tracepoint.
39276 In the series of action packets for a given tracepoint, at most one
39277 can have an @samp{S} before its first @var{action}. If such a packet
39278 is sent, it and the following packets define ``while-stepping''
39279 actions. Any prior packets define ordinary actions --- that is, those
39280 taken when the tracepoint is first hit. If no action packet has an
39281 @samp{S}, then all the packets in the series specify ordinary
39282 tracepoint actions.
39284 The @samp{@var{action}@dots{}} portion of the packet is a series of
39285 actions, concatenated without separators. Each action has one of the
39291 Collect the registers whose bits are set in @var{mask}. @var{mask} is
39292 a hexadecimal number whose @var{i}'th bit is set if register number
39293 @var{i} should be collected. (The least significant bit is numbered
39294 zero.) Note that @var{mask} may be any number of digits long; it may
39295 not fit in a 32-bit word.
39297 @item M @var{basereg},@var{offset},@var{len}
39298 Collect @var{len} bytes of memory starting at the address in register
39299 number @var{basereg}, plus @var{offset}. If @var{basereg} is
39300 @samp{-1}, then the range has a fixed address: @var{offset} is the
39301 address of the lowest byte to collect. The @var{basereg},
39302 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
39303 values (the @samp{-1} value for @var{basereg} is a special case).
39305 @item X @var{len},@var{expr}
39306 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
39307 it directs. @var{expr} is an agent expression, as described in
39308 @ref{Agent Expressions}. Each byte of the expression is encoded as a
39309 two-digit hex number in the packet; @var{len} is the number of bytes
39310 in the expression (and thus one-half the number of hex digits in the
39315 Any number of actions may be packed together in a single @samp{QTDP}
39316 packet, as long as the packet does not exceed the maximum packet
39317 length (400 bytes, for many stubs). There may be only one @samp{R}
39318 action per tracepoint, and it must precede any @samp{M} or @samp{X}
39319 actions. Any registers referred to by @samp{M} and @samp{X} actions
39320 must be collected by a preceding @samp{R} action. (The
39321 ``while-stepping'' actions are treated as if they were attached to a
39322 separate tracepoint, as far as these restrictions are concerned.)
39327 The packet was understood and carried out.
39329 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
39331 The packet was not recognized.
39334 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
39335 @cindex @samp{QTDPsrc} packet
39336 Specify a source string of tracepoint @var{n} at address @var{addr}.
39337 This is useful to get accurate reproduction of the tracepoints
39338 originally downloaded at the beginning of the trace run. @var{type}
39339 is the name of the tracepoint part, such as @samp{cond} for the
39340 tracepoint's conditional expression (see below for a list of types), while
39341 @var{bytes} is the string, encoded in hexadecimal.
39343 @var{start} is the offset of the @var{bytes} within the overall source
39344 string, while @var{slen} is the total length of the source string.
39345 This is intended for handling source strings that are longer than will
39346 fit in a single packet.
39347 @c Add detailed example when this info is moved into a dedicated
39348 @c tracepoint descriptions section.
39350 The available string types are @samp{at} for the location,
39351 @samp{cond} for the conditional, and @samp{cmd} for an action command.
39352 @value{GDBN} sends a separate packet for each command in the action
39353 list, in the same order in which the commands are stored in the list.
39355 The target does not need to do anything with source strings except
39356 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
39359 Although this packet is optional, and @value{GDBN} will only send it
39360 if the target replies with @samp{TracepointSource} @xref{General
39361 Query Packets}, it makes both disconnected tracing and trace files
39362 much easier to use. Otherwise the user must be careful that the
39363 tracepoints in effect while looking at trace frames are identical to
39364 the ones in effect during the trace run; even a small discrepancy
39365 could cause @samp{tdump} not to work, or a particular trace frame not
39368 @item QTDV:@var{n}:@var{value}
39369 @cindex define trace state variable, remote request
39370 @cindex @samp{QTDV} packet
39371 Create a new trace state variable, number @var{n}, with an initial
39372 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
39373 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
39374 the option of not using this packet for initial values of zero; the
39375 target should simply create the trace state variables as they are
39376 mentioned in expressions.
39378 @item QTFrame:@var{n}
39379 @cindex @samp{QTFrame} packet
39380 Select the @var{n}'th tracepoint frame from the buffer, and use the
39381 register and memory contents recorded there to answer subsequent
39382 request packets from @value{GDBN}.
39384 A successful reply from the stub indicates that the stub has found the
39385 requested frame. The response is a series of parts, concatenated
39386 without separators, describing the frame we selected. Each part has
39387 one of the following forms:
39391 The selected frame is number @var{n} in the trace frame buffer;
39392 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
39393 was no frame matching the criteria in the request packet.
39396 The selected trace frame records a hit of tracepoint number @var{t};
39397 @var{t} is a hexadecimal number.
39401 @item QTFrame:pc:@var{addr}
39402 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
39403 currently selected frame whose PC is @var{addr};
39404 @var{addr} is a hexadecimal number.
39406 @item QTFrame:tdp:@var{t}
39407 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
39408 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
39409 is a hexadecimal number.
39411 @item QTFrame:range:@var{start}:@var{end}
39412 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
39413 currently selected frame whose PC is between @var{start} (inclusive)
39414 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
39417 @item QTFrame:outside:@var{start}:@var{end}
39418 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
39419 frame @emph{outside} the given range of addresses (exclusive).
39422 @cindex @samp{qTMinFTPILen} packet
39423 This packet requests the minimum length of instruction at which a fast
39424 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
39425 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
39426 it depends on the target system being able to create trampolines in
39427 the first 64K of memory, which might or might not be possible for that
39428 system. So the reply to this packet will be 4 if it is able to
39435 The minimum instruction length is currently unknown.
39437 The minimum instruction length is @var{length}, where @var{length} is greater
39438 or equal to 1. @var{length} is a hexadecimal number. A reply of 1 means
39439 that a fast tracepoint may be placed on any instruction regardless of size.
39441 An error has occurred.
39443 An empty reply indicates that the request is not supported by the stub.
39447 @cindex @samp{QTStart} packet
39448 Begin the tracepoint experiment. Begin collecting data from
39449 tracepoint hits in the trace frame buffer. This packet supports the
39450 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
39451 instruction reply packet}).
39454 @cindex @samp{QTStop} packet
39455 End the tracepoint experiment. Stop collecting trace frames.
39457 @item QTEnable:@var{n}:@var{addr}
39459 @cindex @samp{QTEnable} packet
39460 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
39461 experiment. If the tracepoint was previously disabled, then collection
39462 of data from it will resume.
39464 @item QTDisable:@var{n}:@var{addr}
39466 @cindex @samp{QTDisable} packet
39467 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
39468 experiment. No more data will be collected from the tracepoint unless
39469 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
39472 @cindex @samp{QTinit} packet
39473 Clear the table of tracepoints, and empty the trace frame buffer.
39475 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
39476 @cindex @samp{QTro} packet
39477 Establish the given ranges of memory as ``transparent''. The stub
39478 will answer requests for these ranges from memory's current contents,
39479 if they were not collected as part of the tracepoint hit.
39481 @value{GDBN} uses this to mark read-only regions of memory, like those
39482 containing program code. Since these areas never change, they should
39483 still have the same contents they did when the tracepoint was hit, so
39484 there's no reason for the stub to refuse to provide their contents.
39486 @item QTDisconnected:@var{value}
39487 @cindex @samp{QTDisconnected} packet
39488 Set the choice to what to do with the tracing run when @value{GDBN}
39489 disconnects from the target. A @var{value} of 1 directs the target to
39490 continue the tracing run, while 0 tells the target to stop tracing if
39491 @value{GDBN} is no longer in the picture.
39494 @cindex @samp{qTStatus} packet
39495 Ask the stub if there is a trace experiment running right now.
39497 The reply has the form:
39501 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
39502 @var{running} is a single digit @code{1} if the trace is presently
39503 running, or @code{0} if not. It is followed by semicolon-separated
39504 optional fields that an agent may use to report additional status.
39508 If the trace is not running, the agent may report any of several
39509 explanations as one of the optional fields:
39514 No trace has been run yet.
39516 @item tstop[:@var{text}]:0
39517 The trace was stopped by a user-originated stop command. The optional
39518 @var{text} field is a user-supplied string supplied as part of the
39519 stop command (for instance, an explanation of why the trace was
39520 stopped manually). It is hex-encoded.
39523 The trace stopped because the trace buffer filled up.
39525 @item tdisconnected:0
39526 The trace stopped because @value{GDBN} disconnected from the target.
39528 @item tpasscount:@var{tpnum}
39529 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
39531 @item terror:@var{text}:@var{tpnum}
39532 The trace stopped because tracepoint @var{tpnum} had an error. The
39533 string @var{text} is available to describe the nature of the error
39534 (for instance, a divide by zero in the condition expression).
39535 @var{text} is hex encoded.
39538 The trace stopped for some other reason.
39542 Additional optional fields supply statistical and other information.
39543 Although not required, they are extremely useful for users monitoring
39544 the progress of a trace run. If a trace has stopped, and these
39545 numbers are reported, they must reflect the state of the just-stopped
39550 @item tframes:@var{n}
39551 The number of trace frames in the buffer.
39553 @item tcreated:@var{n}
39554 The total number of trace frames created during the run. This may
39555 be larger than the trace frame count, if the buffer is circular.
39557 @item tsize:@var{n}
39558 The total size of the trace buffer, in bytes.
39560 @item tfree:@var{n}
39561 The number of bytes still unused in the buffer.
39563 @item circular:@var{n}
39564 The value of the circular trace buffer flag. @code{1} means that the
39565 trace buffer is circular and old trace frames will be discarded if
39566 necessary to make room, @code{0} means that the trace buffer is linear
39569 @item disconn:@var{n}
39570 The value of the disconnected tracing flag. @code{1} means that
39571 tracing will continue after @value{GDBN} disconnects, @code{0} means
39572 that the trace run will stop.
39576 @item qTP:@var{tp}:@var{addr}
39577 @cindex tracepoint status, remote request
39578 @cindex @samp{qTP} packet
39579 Ask the stub for the current state of tracepoint number @var{tp} at
39580 address @var{addr}.
39584 @item V@var{hits}:@var{usage}
39585 The tracepoint has been hit @var{hits} times so far during the trace
39586 run, and accounts for @var{usage} in the trace buffer. Note that
39587 @code{while-stepping} steps are not counted as separate hits, but the
39588 steps' space consumption is added into the usage number.
39592 @item qTV:@var{var}
39593 @cindex trace state variable value, remote request
39594 @cindex @samp{qTV} packet
39595 Ask the stub for the value of the trace state variable number @var{var}.
39600 The value of the variable is @var{value}. This will be the current
39601 value of the variable if the user is examining a running target, or a
39602 saved value if the variable was collected in the trace frame that the
39603 user is looking at. Note that multiple requests may result in
39604 different reply values, such as when requesting values while the
39605 program is running.
39608 The value of the variable is unknown. This would occur, for example,
39609 if the user is examining a trace frame in which the requested variable
39614 @cindex @samp{qTfP} packet
39616 @cindex @samp{qTsP} packet
39617 These packets request data about tracepoints that are being used by
39618 the target. @value{GDBN} sends @code{qTfP} to get the first piece
39619 of data, and multiple @code{qTsP} to get additional pieces. Replies
39620 to these packets generally take the form of the @code{QTDP} packets
39621 that define tracepoints. (FIXME add detailed syntax)
39624 @cindex @samp{qTfV} packet
39626 @cindex @samp{qTsV} packet
39627 These packets request data about trace state variables that are on the
39628 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
39629 and multiple @code{qTsV} to get additional variables. Replies to
39630 these packets follow the syntax of the @code{QTDV} packets that define
39631 trace state variables.
39637 @cindex @samp{qTfSTM} packet
39638 @cindex @samp{qTsSTM} packet
39639 These packets request data about static tracepoint markers that exist
39640 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
39641 first piece of data, and multiple @code{qTsSTM} to get additional
39642 pieces. Replies to these packets take the following form:
39646 @item m @var{address}:@var{id}:@var{extra}
39648 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
39649 a comma-separated list of markers
39651 (lower case letter @samp{L}) denotes end of list.
39653 An error occurred. @var{nn} are hex digits.
39655 An empty reply indicates that the request is not supported by the
39659 @var{address} is encoded in hex.
39660 @var{id} and @var{extra} are strings encoded in hex.
39662 In response to each query, the target will reply with a list of one or
39663 more markers, separated by commas. @value{GDBN} will respond to each
39664 reply with a request for more markers (using the @samp{qs} form of the
39665 query), until the target responds with @samp{l} (lower-case ell, for
39668 @item qTSTMat:@var{address}
39670 @cindex @samp{qTSTMat} packet
39671 This packets requests data about static tracepoint markers in the
39672 target program at @var{address}. Replies to this packet follow the
39673 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
39674 tracepoint markers.
39676 @item QTSave:@var{filename}
39677 @cindex @samp{QTSave} packet
39678 This packet directs the target to save trace data to the file name
39679 @var{filename} in the target's filesystem. @var{filename} is encoded
39680 as a hex string; the interpretation of the file name (relative vs
39681 absolute, wild cards, etc) is up to the target.
39683 @item qTBuffer:@var{offset},@var{len}
39684 @cindex @samp{qTBuffer} packet
39685 Return up to @var{len} bytes of the current contents of trace buffer,
39686 starting at @var{offset}. The trace buffer is treated as if it were
39687 a contiguous collection of traceframes, as per the trace file format.
39688 The reply consists as many hex-encoded bytes as the target can deliver
39689 in a packet; it is not an error to return fewer than were asked for.
39690 A reply consisting of just @code{l} indicates that no bytes are
39693 @item QTBuffer:circular:@var{value}
39694 This packet directs the target to use a circular trace buffer if
39695 @var{value} is 1, or a linear buffer if the value is 0.
39697 @item QTBuffer:size:@var{size}
39698 @anchor{QTBuffer-size}
39699 @cindex @samp{QTBuffer size} packet
39700 This packet directs the target to make the trace buffer be of size
39701 @var{size} if possible. A value of @code{-1} tells the target to
39702 use whatever size it prefers.
39704 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
39705 @cindex @samp{QTNotes} packet
39706 This packet adds optional textual notes to the trace run. Allowable
39707 types include @code{user}, @code{notes}, and @code{tstop}, the
39708 @var{text} fields are arbitrary strings, hex-encoded.
39712 @subsection Relocate instruction reply packet
39713 When installing fast tracepoints in memory, the target may need to
39714 relocate the instruction currently at the tracepoint address to a
39715 different address in memory. For most instructions, a simple copy is
39716 enough, but, for example, call instructions that implicitly push the
39717 return address on the stack, and relative branches or other
39718 PC-relative instructions require offset adjustment, so that the effect
39719 of executing the instruction at a different address is the same as if
39720 it had executed in the original location.
39722 In response to several of the tracepoint packets, the target may also
39723 respond with a number of intermediate @samp{qRelocInsn} request
39724 packets before the final result packet, to have @value{GDBN} handle
39725 this relocation operation. If a packet supports this mechanism, its
39726 documentation will explicitly say so. See for example the above
39727 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
39728 format of the request is:
39731 @item qRelocInsn:@var{from};@var{to}
39733 This requests @value{GDBN} to copy instruction at address @var{from}
39734 to address @var{to}, possibly adjusted so that executing the
39735 instruction at @var{to} has the same effect as executing it at
39736 @var{from}. @value{GDBN} writes the adjusted instruction to target
39737 memory starting at @var{to}.
39742 @item qRelocInsn:@var{adjusted_size}
39743 Informs the stub the relocation is complete. @var{adjusted_size} is
39744 the length in bytes of resulting relocated instruction sequence.
39746 A badly formed request was detected, or an error was encountered while
39747 relocating the instruction.
39750 @node Host I/O Packets
39751 @section Host I/O Packets
39752 @cindex Host I/O, remote protocol
39753 @cindex file transfer, remote protocol
39755 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
39756 operations on the far side of a remote link. For example, Host I/O is
39757 used to upload and download files to a remote target with its own
39758 filesystem. Host I/O uses the same constant values and data structure
39759 layout as the target-initiated File-I/O protocol. However, the
39760 Host I/O packets are structured differently. The target-initiated
39761 protocol relies on target memory to store parameters and buffers.
39762 Host I/O requests are initiated by @value{GDBN}, and the
39763 target's memory is not involved. @xref{File-I/O Remote Protocol
39764 Extension}, for more details on the target-initiated protocol.
39766 The Host I/O request packets all encode a single operation along with
39767 its arguments. They have this format:
39771 @item vFile:@var{operation}: @var{parameter}@dots{}
39772 @var{operation} is the name of the particular request; the target
39773 should compare the entire packet name up to the second colon when checking
39774 for a supported operation. The format of @var{parameter} depends on
39775 the operation. Numbers are always passed in hexadecimal. Negative
39776 numbers have an explicit minus sign (i.e.@: two's complement is not
39777 used). Strings (e.g.@: filenames) are encoded as a series of
39778 hexadecimal bytes. The last argument to a system call may be a
39779 buffer of escaped binary data (@pxref{Binary Data}).
39783 The valid responses to Host I/O packets are:
39787 @item F @var{result} [, @var{errno}] [; @var{attachment}]
39788 @var{result} is the integer value returned by this operation, usually
39789 non-negative for success and -1 for errors. If an error has occured,
39790 @var{errno} will be included in the result. @var{errno} will have a
39791 value defined by the File-I/O protocol (@pxref{Errno Values}). For
39792 operations which return data, @var{attachment} supplies the data as a
39793 binary buffer. Binary buffers in response packets are escaped in the
39794 normal way (@pxref{Binary Data}). See the individual packet
39795 documentation for the interpretation of @var{result} and
39799 An empty response indicates that this operation is not recognized.
39803 These are the supported Host I/O operations:
39806 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
39807 Open a file at @var{pathname} and return a file descriptor for it, or
39808 return -1 if an error occurs. @var{pathname} is a string,
39809 @var{flags} is an integer indicating a mask of open flags
39810 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
39811 of mode bits to use if the file is created (@pxref{mode_t Values}).
39812 @xref{open}, for details of the open flags and mode values.
39814 @item vFile:close: @var{fd}
39815 Close the open file corresponding to @var{fd} and return 0, or
39816 -1 if an error occurs.
39818 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
39819 Read data from the open file corresponding to @var{fd}. Up to
39820 @var{count} bytes will be read from the file, starting at @var{offset}
39821 relative to the start of the file. The target may read fewer bytes;
39822 common reasons include packet size limits and an end-of-file
39823 condition. The number of bytes read is returned. Zero should only be
39824 returned for a successful read at the end of the file, or if
39825 @var{count} was zero.
39827 The data read should be returned as a binary attachment on success.
39828 If zero bytes were read, the response should include an empty binary
39829 attachment (i.e.@: a trailing semicolon). The return value is the
39830 number of target bytes read; the binary attachment may be longer if
39831 some characters were escaped.
39833 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
39834 Write @var{data} (a binary buffer) to the open file corresponding
39835 to @var{fd}. Start the write at @var{offset} from the start of the
39836 file. Unlike many @code{write} system calls, there is no
39837 separate @var{count} argument; the length of @var{data} in the
39838 packet is used. @samp{vFile:write} returns the number of bytes written,
39839 which may be shorter than the length of @var{data}, or -1 if an
39842 @item vFile:unlink: @var{pathname}
39843 Delete the file at @var{pathname} on the target. Return 0,
39844 or -1 if an error occurs. @var{pathname} is a string.
39846 @item vFile:readlink: @var{filename}
39847 Read value of symbolic link @var{filename} on the target. Return
39848 the number of bytes read, or -1 if an error occurs.
39850 The data read should be returned as a binary attachment on success.
39851 If zero bytes were read, the response should include an empty binary
39852 attachment (i.e.@: a trailing semicolon). The return value is the
39853 number of target bytes read; the binary attachment may be longer if
39854 some characters were escaped.
39859 @section Interrupts
39860 @cindex interrupts (remote protocol)
39862 When a program on the remote target is running, @value{GDBN} may
39863 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
39864 a @code{BREAK} followed by @code{g},
39865 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
39867 The precise meaning of @code{BREAK} is defined by the transport
39868 mechanism and may, in fact, be undefined. @value{GDBN} does not
39869 currently define a @code{BREAK} mechanism for any of the network
39870 interfaces except for TCP, in which case @value{GDBN} sends the
39871 @code{telnet} BREAK sequence.
39873 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
39874 transport mechanisms. It is represented by sending the single byte
39875 @code{0x03} without any of the usual packet overhead described in
39876 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
39877 transmitted as part of a packet, it is considered to be packet data
39878 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
39879 (@pxref{X packet}), used for binary downloads, may include an unescaped
39880 @code{0x03} as part of its packet.
39882 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
39883 When Linux kernel receives this sequence from serial port,
39884 it stops execution and connects to gdb.
39886 Stubs are not required to recognize these interrupt mechanisms and the
39887 precise meaning associated with receipt of the interrupt is
39888 implementation defined. If the target supports debugging of multiple
39889 threads and/or processes, it should attempt to interrupt all
39890 currently-executing threads and processes.
39891 If the stub is successful at interrupting the
39892 running program, it should send one of the stop
39893 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
39894 of successfully stopping the program in all-stop mode, and a stop reply
39895 for each stopped thread in non-stop mode.
39896 Interrupts received while the
39897 program is stopped are discarded.
39899 @node Notification Packets
39900 @section Notification Packets
39901 @cindex notification packets
39902 @cindex packets, notification
39904 The @value{GDBN} remote serial protocol includes @dfn{notifications},
39905 packets that require no acknowledgment. Both the GDB and the stub
39906 may send notifications (although the only notifications defined at
39907 present are sent by the stub). Notifications carry information
39908 without incurring the round-trip latency of an acknowledgment, and so
39909 are useful for low-impact communications where occasional packet loss
39912 A notification packet has the form @samp{% @var{data} #
39913 @var{checksum}}, where @var{data} is the content of the notification,
39914 and @var{checksum} is a checksum of @var{data}, computed and formatted
39915 as for ordinary @value{GDBN} packets. A notification's @var{data}
39916 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
39917 receiving a notification, the recipient sends no @samp{+} or @samp{-}
39918 to acknowledge the notification's receipt or to report its corruption.
39920 Every notification's @var{data} begins with a name, which contains no
39921 colon characters, followed by a colon character.
39923 Recipients should silently ignore corrupted notifications and
39924 notifications they do not understand. Recipients should restart
39925 timeout periods on receipt of a well-formed notification, whether or
39926 not they understand it.
39928 Senders should only send the notifications described here when this
39929 protocol description specifies that they are permitted. In the
39930 future, we may extend the protocol to permit existing notifications in
39931 new contexts; this rule helps older senders avoid confusing newer
39934 (Older versions of @value{GDBN} ignore bytes received until they see
39935 the @samp{$} byte that begins an ordinary packet, so new stubs may
39936 transmit notifications without fear of confusing older clients. There
39937 are no notifications defined for @value{GDBN} to send at the moment, but we
39938 assume that most older stubs would ignore them, as well.)
39940 Each notification is comprised of three parts:
39942 @item @var{name}:@var{event}
39943 The notification packet is sent by the side that initiates the
39944 exchange (currently, only the stub does that), with @var{event}
39945 carrying the specific information about the notification.
39946 @var{name} is the name of the notification.
39948 The acknowledge sent by the other side, usually @value{GDBN}, to
39949 acknowledge the exchange and request the event.
39952 The purpose of an asynchronous notification mechanism is to report to
39953 @value{GDBN} that something interesting happened in the remote stub.
39955 The remote stub may send notification @var{name}:@var{event}
39956 at any time, but @value{GDBN} acknowledges the notification when
39957 appropriate. The notification event is pending before @value{GDBN}
39958 acknowledges. Only one notification at a time may be pending; if
39959 additional events occur before @value{GDBN} has acknowledged the
39960 previous notification, they must be queued by the stub for later
39961 synchronous transmission in response to @var{ack} packets from
39962 @value{GDBN}. Because the notification mechanism is unreliable,
39963 the stub is permitted to resend a notification if it believes
39964 @value{GDBN} may not have received it.
39966 Specifically, notifications may appear when @value{GDBN} is not
39967 otherwise reading input from the stub, or when @value{GDBN} is
39968 expecting to read a normal synchronous response or a
39969 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
39970 Notification packets are distinct from any other communication from
39971 the stub so there is no ambiguity.
39973 After receiving a notification, @value{GDBN} shall acknowledge it by
39974 sending a @var{ack} packet as a regular, synchronous request to the
39975 stub. Such acknowledgment is not required to happen immediately, as
39976 @value{GDBN} is permitted to send other, unrelated packets to the
39977 stub first, which the stub should process normally.
39979 Upon receiving a @var{ack} packet, if the stub has other queued
39980 events to report to @value{GDBN}, it shall respond by sending a
39981 normal @var{event}. @value{GDBN} shall then send another @var{ack}
39982 packet to solicit further responses; again, it is permitted to send
39983 other, unrelated packets as well which the stub should process
39986 If the stub receives a @var{ack} packet and there are no additional
39987 @var{event} to report, the stub shall return an @samp{OK} response.
39988 At this point, @value{GDBN} has finished processing a notification
39989 and the stub has completed sending any queued events. @value{GDBN}
39990 won't accept any new notifications until the final @samp{OK} is
39991 received . If further notification events occur, the stub shall send
39992 a new notification, @value{GDBN} shall accept the notification, and
39993 the process shall be repeated.
39995 The process of asynchronous notification can be illustrated by the
39998 <- @code{%%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
40001 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
40003 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
40008 The following notifications are defined:
40009 @multitable @columnfractions 0.12 0.12 0.38 0.38
40018 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
40019 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
40020 for information on how these notifications are acknowledged by
40022 @tab Report an asynchronous stop event in non-stop mode.
40026 @node Remote Non-Stop
40027 @section Remote Protocol Support for Non-Stop Mode
40029 @value{GDBN}'s remote protocol supports non-stop debugging of
40030 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
40031 supports non-stop mode, it should report that to @value{GDBN} by including
40032 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
40034 @value{GDBN} typically sends a @samp{QNonStop} packet only when
40035 establishing a new connection with the stub. Entering non-stop mode
40036 does not alter the state of any currently-running threads, but targets
40037 must stop all threads in any already-attached processes when entering
40038 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
40039 probe the target state after a mode change.
40041 In non-stop mode, when an attached process encounters an event that
40042 would otherwise be reported with a stop reply, it uses the
40043 asynchronous notification mechanism (@pxref{Notification Packets}) to
40044 inform @value{GDBN}. In contrast to all-stop mode, where all threads
40045 in all processes are stopped when a stop reply is sent, in non-stop
40046 mode only the thread reporting the stop event is stopped. That is,
40047 when reporting a @samp{S} or @samp{T} response to indicate completion
40048 of a step operation, hitting a breakpoint, or a fault, only the
40049 affected thread is stopped; any other still-running threads continue
40050 to run. When reporting a @samp{W} or @samp{X} response, all running
40051 threads belonging to other attached processes continue to run.
40053 In non-stop mode, the target shall respond to the @samp{?} packet as
40054 follows. First, any incomplete stop reply notification/@samp{vStopped}
40055 sequence in progress is abandoned. The target must begin a new
40056 sequence reporting stop events for all stopped threads, whether or not
40057 it has previously reported those events to @value{GDBN}. The first
40058 stop reply is sent as a synchronous reply to the @samp{?} packet, and
40059 subsequent stop replies are sent as responses to @samp{vStopped} packets
40060 using the mechanism described above. The target must not send
40061 asynchronous stop reply notifications until the sequence is complete.
40062 If all threads are running when the target receives the @samp{?} packet,
40063 or if the target is not attached to any process, it shall respond
40066 @node Packet Acknowledgment
40067 @section Packet Acknowledgment
40069 @cindex acknowledgment, for @value{GDBN} remote
40070 @cindex packet acknowledgment, for @value{GDBN} remote
40071 By default, when either the host or the target machine receives a packet,
40072 the first response expected is an acknowledgment: either @samp{+} (to indicate
40073 the package was received correctly) or @samp{-} (to request retransmission).
40074 This mechanism allows the @value{GDBN} remote protocol to operate over
40075 unreliable transport mechanisms, such as a serial line.
40077 In cases where the transport mechanism is itself reliable (such as a pipe or
40078 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
40079 It may be desirable to disable them in that case to reduce communication
40080 overhead, or for other reasons. This can be accomplished by means of the
40081 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
40083 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
40084 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
40085 and response format still includes the normal checksum, as described in
40086 @ref{Overview}, but the checksum may be ignored by the receiver.
40088 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
40089 no-acknowledgment mode, it should report that to @value{GDBN}
40090 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
40091 @pxref{qSupported}.
40092 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
40093 disabled via the @code{set remote noack-packet off} command
40094 (@pxref{Remote Configuration}),
40095 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
40096 Only then may the stub actually turn off packet acknowledgments.
40097 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
40098 response, which can be safely ignored by the stub.
40100 Note that @code{set remote noack-packet} command only affects negotiation
40101 between @value{GDBN} and the stub when subsequent connections are made;
40102 it does not affect the protocol acknowledgment state for any current
40104 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
40105 new connection is established,
40106 there is also no protocol request to re-enable the acknowledgments
40107 for the current connection, once disabled.
40112 Example sequence of a target being re-started. Notice how the restart
40113 does not get any direct output:
40118 @emph{target restarts}
40121 <- @code{T001:1234123412341234}
40125 Example sequence of a target being stepped by a single instruction:
40128 -> @code{G1445@dots{}}
40133 <- @code{T001:1234123412341234}
40137 <- @code{1455@dots{}}
40141 @node File-I/O Remote Protocol Extension
40142 @section File-I/O Remote Protocol Extension
40143 @cindex File-I/O remote protocol extension
40146 * File-I/O Overview::
40147 * Protocol Basics::
40148 * The F Request Packet::
40149 * The F Reply Packet::
40150 * The Ctrl-C Message::
40152 * List of Supported Calls::
40153 * Protocol-specific Representation of Datatypes::
40155 * File-I/O Examples::
40158 @node File-I/O Overview
40159 @subsection File-I/O Overview
40160 @cindex file-i/o overview
40162 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
40163 target to use the host's file system and console I/O to perform various
40164 system calls. System calls on the target system are translated into a
40165 remote protocol packet to the host system, which then performs the needed
40166 actions and returns a response packet to the target system.
40167 This simulates file system operations even on targets that lack file systems.
40169 The protocol is defined to be independent of both the host and target systems.
40170 It uses its own internal representation of datatypes and values. Both
40171 @value{GDBN} and the target's @value{GDBN} stub are responsible for
40172 translating the system-dependent value representations into the internal
40173 protocol representations when data is transmitted.
40175 The communication is synchronous. A system call is possible only when
40176 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
40177 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
40178 the target is stopped to allow deterministic access to the target's
40179 memory. Therefore File-I/O is not interruptible by target signals. On
40180 the other hand, it is possible to interrupt File-I/O by a user interrupt
40181 (@samp{Ctrl-C}) within @value{GDBN}.
40183 The target's request to perform a host system call does not finish
40184 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
40185 after finishing the system call, the target returns to continuing the
40186 previous activity (continue, step). No additional continue or step
40187 request from @value{GDBN} is required.
40190 (@value{GDBP}) continue
40191 <- target requests 'system call X'
40192 target is stopped, @value{GDBN} executes system call
40193 -> @value{GDBN} returns result
40194 ... target continues, @value{GDBN} returns to wait for the target
40195 <- target hits breakpoint and sends a Txx packet
40198 The protocol only supports I/O on the console and to regular files on
40199 the host file system. Character or block special devices, pipes,
40200 named pipes, sockets or any other communication method on the host
40201 system are not supported by this protocol.
40203 File I/O is not supported in non-stop mode.
40205 @node Protocol Basics
40206 @subsection Protocol Basics
40207 @cindex protocol basics, file-i/o
40209 The File-I/O protocol uses the @code{F} packet as the request as well
40210 as reply packet. Since a File-I/O system call can only occur when
40211 @value{GDBN} is waiting for a response from the continuing or stepping target,
40212 the File-I/O request is a reply that @value{GDBN} has to expect as a result
40213 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
40214 This @code{F} packet contains all information needed to allow @value{GDBN}
40215 to call the appropriate host system call:
40219 A unique identifier for the requested system call.
40222 All parameters to the system call. Pointers are given as addresses
40223 in the target memory address space. Pointers to strings are given as
40224 pointer/length pair. Numerical values are given as they are.
40225 Numerical control flags are given in a protocol-specific representation.
40229 At this point, @value{GDBN} has to perform the following actions.
40233 If the parameters include pointer values to data needed as input to a
40234 system call, @value{GDBN} requests this data from the target with a
40235 standard @code{m} packet request. This additional communication has to be
40236 expected by the target implementation and is handled as any other @code{m}
40240 @value{GDBN} translates all value from protocol representation to host
40241 representation as needed. Datatypes are coerced into the host types.
40244 @value{GDBN} calls the system call.
40247 It then coerces datatypes back to protocol representation.
40250 If the system call is expected to return data in buffer space specified
40251 by pointer parameters to the call, the data is transmitted to the
40252 target using a @code{M} or @code{X} packet. This packet has to be expected
40253 by the target implementation and is handled as any other @code{M} or @code{X}
40258 Eventually @value{GDBN} replies with another @code{F} packet which contains all
40259 necessary information for the target to continue. This at least contains
40266 @code{errno}, if has been changed by the system call.
40273 After having done the needed type and value coercion, the target continues
40274 the latest continue or step action.
40276 @node The F Request Packet
40277 @subsection The @code{F} Request Packet
40278 @cindex file-i/o request packet
40279 @cindex @code{F} request packet
40281 The @code{F} request packet has the following format:
40284 @item F@var{call-id},@var{parameter@dots{}}
40286 @var{call-id} is the identifier to indicate the host system call to be called.
40287 This is just the name of the function.
40289 @var{parameter@dots{}} are the parameters to the system call.
40290 Parameters are hexadecimal integer values, either the actual values in case
40291 of scalar datatypes, pointers to target buffer space in case of compound
40292 datatypes and unspecified memory areas, or pointer/length pairs in case
40293 of string parameters. These are appended to the @var{call-id} as a
40294 comma-delimited list. All values are transmitted in ASCII
40295 string representation, pointer/length pairs separated by a slash.
40301 @node The F Reply Packet
40302 @subsection The @code{F} Reply Packet
40303 @cindex file-i/o reply packet
40304 @cindex @code{F} reply packet
40306 The @code{F} reply packet has the following format:
40310 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
40312 @var{retcode} is the return code of the system call as hexadecimal value.
40314 @var{errno} is the @code{errno} set by the call, in protocol-specific
40316 This parameter can be omitted if the call was successful.
40318 @var{Ctrl-C flag} is only sent if the user requested a break. In this
40319 case, @var{errno} must be sent as well, even if the call was successful.
40320 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
40327 or, if the call was interrupted before the host call has been performed:
40334 assuming 4 is the protocol-specific representation of @code{EINTR}.
40339 @node The Ctrl-C Message
40340 @subsection The @samp{Ctrl-C} Message
40341 @cindex ctrl-c message, in file-i/o protocol
40343 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
40344 reply packet (@pxref{The F Reply Packet}),
40345 the target should behave as if it had
40346 gotten a break message. The meaning for the target is ``system call
40347 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
40348 (as with a break message) and return to @value{GDBN} with a @code{T02}
40351 It's important for the target to know in which
40352 state the system call was interrupted. There are two possible cases:
40356 The system call hasn't been performed on the host yet.
40359 The system call on the host has been finished.
40363 These two states can be distinguished by the target by the value of the
40364 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
40365 call hasn't been performed. This is equivalent to the @code{EINTR} handling
40366 on POSIX systems. In any other case, the target may presume that the
40367 system call has been finished --- successfully or not --- and should behave
40368 as if the break message arrived right after the system call.
40370 @value{GDBN} must behave reliably. If the system call has not been called
40371 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
40372 @code{errno} in the packet. If the system call on the host has been finished
40373 before the user requests a break, the full action must be finished by
40374 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
40375 The @code{F} packet may only be sent when either nothing has happened
40376 or the full action has been completed.
40379 @subsection Console I/O
40380 @cindex console i/o as part of file-i/o
40382 By default and if not explicitly closed by the target system, the file
40383 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
40384 on the @value{GDBN} console is handled as any other file output operation
40385 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
40386 by @value{GDBN} so that after the target read request from file descriptor
40387 0 all following typing is buffered until either one of the following
40392 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
40394 system call is treated as finished.
40397 The user presses @key{RET}. This is treated as end of input with a trailing
40401 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
40402 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
40406 If the user has typed more characters than fit in the buffer given to
40407 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
40408 either another @code{read(0, @dots{})} is requested by the target, or debugging
40409 is stopped at the user's request.
40412 @node List of Supported Calls
40413 @subsection List of Supported Calls
40414 @cindex list of supported file-i/o calls
40431 @unnumberedsubsubsec open
40432 @cindex open, file-i/o system call
40437 int open(const char *pathname, int flags);
40438 int open(const char *pathname, int flags, mode_t mode);
40442 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
40445 @var{flags} is the bitwise @code{OR} of the following values:
40449 If the file does not exist it will be created. The host
40450 rules apply as far as file ownership and time stamps
40454 When used with @code{O_CREAT}, if the file already exists it is
40455 an error and open() fails.
40458 If the file already exists and the open mode allows
40459 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
40460 truncated to zero length.
40463 The file is opened in append mode.
40466 The file is opened for reading only.
40469 The file is opened for writing only.
40472 The file is opened for reading and writing.
40476 Other bits are silently ignored.
40480 @var{mode} is the bitwise @code{OR} of the following values:
40484 User has read permission.
40487 User has write permission.
40490 Group has read permission.
40493 Group has write permission.
40496 Others have read permission.
40499 Others have write permission.
40503 Other bits are silently ignored.
40506 @item Return value:
40507 @code{open} returns the new file descriptor or -1 if an error
40514 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
40517 @var{pathname} refers to a directory.
40520 The requested access is not allowed.
40523 @var{pathname} was too long.
40526 A directory component in @var{pathname} does not exist.
40529 @var{pathname} refers to a device, pipe, named pipe or socket.
40532 @var{pathname} refers to a file on a read-only filesystem and
40533 write access was requested.
40536 @var{pathname} is an invalid pointer value.
40539 No space on device to create the file.
40542 The process already has the maximum number of files open.
40545 The limit on the total number of files open on the system
40549 The call was interrupted by the user.
40555 @unnumberedsubsubsec close
40556 @cindex close, file-i/o system call
40565 @samp{Fclose,@var{fd}}
40567 @item Return value:
40568 @code{close} returns zero on success, or -1 if an error occurred.
40574 @var{fd} isn't a valid open file descriptor.
40577 The call was interrupted by the user.
40583 @unnumberedsubsubsec read
40584 @cindex read, file-i/o system call
40589 int read(int fd, void *buf, unsigned int count);
40593 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
40595 @item Return value:
40596 On success, the number of bytes read is returned.
40597 Zero indicates end of file. If count is zero, read
40598 returns zero as well. On error, -1 is returned.
40604 @var{fd} is not a valid file descriptor or is not open for
40608 @var{bufptr} is an invalid pointer value.
40611 The call was interrupted by the user.
40617 @unnumberedsubsubsec write
40618 @cindex write, file-i/o system call
40623 int write(int fd, const void *buf, unsigned int count);
40627 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
40629 @item Return value:
40630 On success, the number of bytes written are returned.
40631 Zero indicates nothing was written. On error, -1
40638 @var{fd} is not a valid file descriptor or is not open for
40642 @var{bufptr} is an invalid pointer value.
40645 An attempt was made to write a file that exceeds the
40646 host-specific maximum file size allowed.
40649 No space on device to write the data.
40652 The call was interrupted by the user.
40658 @unnumberedsubsubsec lseek
40659 @cindex lseek, file-i/o system call
40664 long lseek (int fd, long offset, int flag);
40668 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
40670 @var{flag} is one of:
40674 The offset is set to @var{offset} bytes.
40677 The offset is set to its current location plus @var{offset}
40681 The offset is set to the size of the file plus @var{offset}
40685 @item Return value:
40686 On success, the resulting unsigned offset in bytes from
40687 the beginning of the file is returned. Otherwise, a
40688 value of -1 is returned.
40694 @var{fd} is not a valid open file descriptor.
40697 @var{fd} is associated with the @value{GDBN} console.
40700 @var{flag} is not a proper value.
40703 The call was interrupted by the user.
40709 @unnumberedsubsubsec rename
40710 @cindex rename, file-i/o system call
40715 int rename(const char *oldpath, const char *newpath);
40719 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
40721 @item Return value:
40722 On success, zero is returned. On error, -1 is returned.
40728 @var{newpath} is an existing directory, but @var{oldpath} is not a
40732 @var{newpath} is a non-empty directory.
40735 @var{oldpath} or @var{newpath} is a directory that is in use by some
40739 An attempt was made to make a directory a subdirectory
40743 A component used as a directory in @var{oldpath} or new
40744 path is not a directory. Or @var{oldpath} is a directory
40745 and @var{newpath} exists but is not a directory.
40748 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
40751 No access to the file or the path of the file.
40755 @var{oldpath} or @var{newpath} was too long.
40758 A directory component in @var{oldpath} or @var{newpath} does not exist.
40761 The file is on a read-only filesystem.
40764 The device containing the file has no room for the new
40768 The call was interrupted by the user.
40774 @unnumberedsubsubsec unlink
40775 @cindex unlink, file-i/o system call
40780 int unlink(const char *pathname);
40784 @samp{Funlink,@var{pathnameptr}/@var{len}}
40786 @item Return value:
40787 On success, zero is returned. On error, -1 is returned.
40793 No access to the file or the path of the file.
40796 The system does not allow unlinking of directories.
40799 The file @var{pathname} cannot be unlinked because it's
40800 being used by another process.
40803 @var{pathnameptr} is an invalid pointer value.
40806 @var{pathname} was too long.
40809 A directory component in @var{pathname} does not exist.
40812 A component of the path is not a directory.
40815 The file is on a read-only filesystem.
40818 The call was interrupted by the user.
40824 @unnumberedsubsubsec stat/fstat
40825 @cindex fstat, file-i/o system call
40826 @cindex stat, file-i/o system call
40831 int stat(const char *pathname, struct stat *buf);
40832 int fstat(int fd, struct stat *buf);
40836 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
40837 @samp{Ffstat,@var{fd},@var{bufptr}}
40839 @item Return value:
40840 On success, zero is returned. On error, -1 is returned.
40846 @var{fd} is not a valid open file.
40849 A directory component in @var{pathname} does not exist or the
40850 path is an empty string.
40853 A component of the path is not a directory.
40856 @var{pathnameptr} is an invalid pointer value.
40859 No access to the file or the path of the file.
40862 @var{pathname} was too long.
40865 The call was interrupted by the user.
40871 @unnumberedsubsubsec gettimeofday
40872 @cindex gettimeofday, file-i/o system call
40877 int gettimeofday(struct timeval *tv, void *tz);
40881 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
40883 @item Return value:
40884 On success, 0 is returned, -1 otherwise.
40890 @var{tz} is a non-NULL pointer.
40893 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
40899 @unnumberedsubsubsec isatty
40900 @cindex isatty, file-i/o system call
40905 int isatty(int fd);
40909 @samp{Fisatty,@var{fd}}
40911 @item Return value:
40912 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
40918 The call was interrupted by the user.
40923 Note that the @code{isatty} call is treated as a special case: it returns
40924 1 to the target if the file descriptor is attached
40925 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
40926 would require implementing @code{ioctl} and would be more complex than
40931 @unnumberedsubsubsec system
40932 @cindex system, file-i/o system call
40937 int system(const char *command);
40941 @samp{Fsystem,@var{commandptr}/@var{len}}
40943 @item Return value:
40944 If @var{len} is zero, the return value indicates whether a shell is
40945 available. A zero return value indicates a shell is not available.
40946 For non-zero @var{len}, the value returned is -1 on error and the
40947 return status of the command otherwise. Only the exit status of the
40948 command is returned, which is extracted from the host's @code{system}
40949 return value by calling @code{WEXITSTATUS(retval)}. In case
40950 @file{/bin/sh} could not be executed, 127 is returned.
40956 The call was interrupted by the user.
40961 @value{GDBN} takes over the full task of calling the necessary host calls
40962 to perform the @code{system} call. The return value of @code{system} on
40963 the host is simplified before it's returned
40964 to the target. Any termination signal information from the child process
40965 is discarded, and the return value consists
40966 entirely of the exit status of the called command.
40968 Due to security concerns, the @code{system} call is by default refused
40969 by @value{GDBN}. The user has to allow this call explicitly with the
40970 @code{set remote system-call-allowed 1} command.
40973 @item set remote system-call-allowed
40974 @kindex set remote system-call-allowed
40975 Control whether to allow the @code{system} calls in the File I/O
40976 protocol for the remote target. The default is zero (disabled).
40978 @item show remote system-call-allowed
40979 @kindex show remote system-call-allowed
40980 Show whether the @code{system} calls are allowed in the File I/O
40984 @node Protocol-specific Representation of Datatypes
40985 @subsection Protocol-specific Representation of Datatypes
40986 @cindex protocol-specific representation of datatypes, in file-i/o protocol
40989 * Integral Datatypes::
40991 * Memory Transfer::
40996 @node Integral Datatypes
40997 @unnumberedsubsubsec Integral Datatypes
40998 @cindex integral datatypes, in file-i/o protocol
41000 The integral datatypes used in the system calls are @code{int},
41001 @code{unsigned int}, @code{long}, @code{unsigned long},
41002 @code{mode_t}, and @code{time_t}.
41004 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
41005 implemented as 32 bit values in this protocol.
41007 @code{long} and @code{unsigned long} are implemented as 64 bit types.
41009 @xref{Limits}, for corresponding MIN and MAX values (similar to those
41010 in @file{limits.h}) to allow range checking on host and target.
41012 @code{time_t} datatypes are defined as seconds since the Epoch.
41014 All integral datatypes transferred as part of a memory read or write of a
41015 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
41018 @node Pointer Values
41019 @unnumberedsubsubsec Pointer Values
41020 @cindex pointer values, in file-i/o protocol
41022 Pointers to target data are transmitted as they are. An exception
41023 is made for pointers to buffers for which the length isn't
41024 transmitted as part of the function call, namely strings. Strings
41025 are transmitted as a pointer/length pair, both as hex values, e.g.@:
41032 which is a pointer to data of length 18 bytes at position 0x1aaf.
41033 The length is defined as the full string length in bytes, including
41034 the trailing null byte. For example, the string @code{"hello world"}
41035 at address 0x123456 is transmitted as
41041 @node Memory Transfer
41042 @unnumberedsubsubsec Memory Transfer
41043 @cindex memory transfer, in file-i/o protocol
41045 Structured data which is transferred using a memory read or write (for
41046 example, a @code{struct stat}) is expected to be in a protocol-specific format
41047 with all scalar multibyte datatypes being big endian. Translation to
41048 this representation needs to be done both by the target before the @code{F}
41049 packet is sent, and by @value{GDBN} before
41050 it transfers memory to the target. Transferred pointers to structured
41051 data should point to the already-coerced data at any time.
41055 @unnumberedsubsubsec struct stat
41056 @cindex struct stat, in file-i/o protocol
41058 The buffer of type @code{struct stat} used by the target and @value{GDBN}
41059 is defined as follows:
41063 unsigned int st_dev; /* device */
41064 unsigned int st_ino; /* inode */
41065 mode_t st_mode; /* protection */
41066 unsigned int st_nlink; /* number of hard links */
41067 unsigned int st_uid; /* user ID of owner */
41068 unsigned int st_gid; /* group ID of owner */
41069 unsigned int st_rdev; /* device type (if inode device) */
41070 unsigned long st_size; /* total size, in bytes */
41071 unsigned long st_blksize; /* blocksize for filesystem I/O */
41072 unsigned long st_blocks; /* number of blocks allocated */
41073 time_t st_atime; /* time of last access */
41074 time_t st_mtime; /* time of last modification */
41075 time_t st_ctime; /* time of last change */
41079 The integral datatypes conform to the definitions given in the
41080 appropriate section (see @ref{Integral Datatypes}, for details) so this
41081 structure is of size 64 bytes.
41083 The values of several fields have a restricted meaning and/or
41089 A value of 0 represents a file, 1 the console.
41092 No valid meaning for the target. Transmitted unchanged.
41095 Valid mode bits are described in @ref{Constants}. Any other
41096 bits have currently no meaning for the target.
41101 No valid meaning for the target. Transmitted unchanged.
41106 These values have a host and file system dependent
41107 accuracy. Especially on Windows hosts, the file system may not
41108 support exact timing values.
41111 The target gets a @code{struct stat} of the above representation and is
41112 responsible for coercing it to the target representation before
41115 Note that due to size differences between the host, target, and protocol
41116 representations of @code{struct stat} members, these members could eventually
41117 get truncated on the target.
41119 @node struct timeval
41120 @unnumberedsubsubsec struct timeval
41121 @cindex struct timeval, in file-i/o protocol
41123 The buffer of type @code{struct timeval} used by the File-I/O protocol
41124 is defined as follows:
41128 time_t tv_sec; /* second */
41129 long tv_usec; /* microsecond */
41133 The integral datatypes conform to the definitions given in the
41134 appropriate section (see @ref{Integral Datatypes}, for details) so this
41135 structure is of size 8 bytes.
41138 @subsection Constants
41139 @cindex constants, in file-i/o protocol
41141 The following values are used for the constants inside of the
41142 protocol. @value{GDBN} and target are responsible for translating these
41143 values before and after the call as needed.
41154 @unnumberedsubsubsec Open Flags
41155 @cindex open flags, in file-i/o protocol
41157 All values are given in hexadecimal representation.
41169 @node mode_t Values
41170 @unnumberedsubsubsec mode_t Values
41171 @cindex mode_t values, in file-i/o protocol
41173 All values are given in octal representation.
41190 @unnumberedsubsubsec Errno Values
41191 @cindex errno values, in file-i/o protocol
41193 All values are given in decimal representation.
41218 @code{EUNKNOWN} is used as a fallback error value if a host system returns
41219 any error value not in the list of supported error numbers.
41222 @unnumberedsubsubsec Lseek Flags
41223 @cindex lseek flags, in file-i/o protocol
41232 @unnumberedsubsubsec Limits
41233 @cindex limits, in file-i/o protocol
41235 All values are given in decimal representation.
41238 INT_MIN -2147483648
41240 UINT_MAX 4294967295
41241 LONG_MIN -9223372036854775808
41242 LONG_MAX 9223372036854775807
41243 ULONG_MAX 18446744073709551615
41246 @node File-I/O Examples
41247 @subsection File-I/O Examples
41248 @cindex file-i/o examples
41250 Example sequence of a write call, file descriptor 3, buffer is at target
41251 address 0x1234, 6 bytes should be written:
41254 <- @code{Fwrite,3,1234,6}
41255 @emph{request memory read from target}
41258 @emph{return "6 bytes written"}
41262 Example sequence of a read call, file descriptor 3, buffer is at target
41263 address 0x1234, 6 bytes should be read:
41266 <- @code{Fread,3,1234,6}
41267 @emph{request memory write to target}
41268 -> @code{X1234,6:XXXXXX}
41269 @emph{return "6 bytes read"}
41273 Example sequence of a read call, call fails on the host due to invalid
41274 file descriptor (@code{EBADF}):
41277 <- @code{Fread,3,1234,6}
41281 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
41285 <- @code{Fread,3,1234,6}
41290 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
41294 <- @code{Fread,3,1234,6}
41295 -> @code{X1234,6:XXXXXX}
41299 @node Library List Format
41300 @section Library List Format
41301 @cindex library list format, remote protocol
41303 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
41304 same process as your application to manage libraries. In this case,
41305 @value{GDBN} can use the loader's symbol table and normal memory
41306 operations to maintain a list of shared libraries. On other
41307 platforms, the operating system manages loaded libraries.
41308 @value{GDBN} can not retrieve the list of currently loaded libraries
41309 through memory operations, so it uses the @samp{qXfer:libraries:read}
41310 packet (@pxref{qXfer library list read}) instead. The remote stub
41311 queries the target's operating system and reports which libraries
41314 The @samp{qXfer:libraries:read} packet returns an XML document which
41315 lists loaded libraries and their offsets. Each library has an
41316 associated name and one or more segment or section base addresses,
41317 which report where the library was loaded in memory.
41319 For the common case of libraries that are fully linked binaries, the
41320 library should have a list of segments. If the target supports
41321 dynamic linking of a relocatable object file, its library XML element
41322 should instead include a list of allocated sections. The segment or
41323 section bases are start addresses, not relocation offsets; they do not
41324 depend on the library's link-time base addresses.
41326 @value{GDBN} must be linked with the Expat library to support XML
41327 library lists. @xref{Expat}.
41329 A simple memory map, with one loaded library relocated by a single
41330 offset, looks like this:
41334 <library name="/lib/libc.so.6">
41335 <segment address="0x10000000"/>
41340 Another simple memory map, with one loaded library with three
41341 allocated sections (.text, .data, .bss), looks like this:
41345 <library name="sharedlib.o">
41346 <section address="0x10000000"/>
41347 <section address="0x20000000"/>
41348 <section address="0x30000000"/>
41353 The format of a library list is described by this DTD:
41356 <!-- library-list: Root element with versioning -->
41357 <!ELEMENT library-list (library)*>
41358 <!ATTLIST library-list version CDATA #FIXED "1.0">
41359 <!ELEMENT library (segment*, section*)>
41360 <!ATTLIST library name CDATA #REQUIRED>
41361 <!ELEMENT segment EMPTY>
41362 <!ATTLIST segment address CDATA #REQUIRED>
41363 <!ELEMENT section EMPTY>
41364 <!ATTLIST section address CDATA #REQUIRED>
41367 In addition, segments and section descriptors cannot be mixed within a
41368 single library element, and you must supply at least one segment or
41369 section for each library.
41371 @node Library List Format for SVR4 Targets
41372 @section Library List Format for SVR4 Targets
41373 @cindex library list format, remote protocol
41375 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
41376 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
41377 shared libraries. Still a special library list provided by this packet is
41378 more efficient for the @value{GDBN} remote protocol.
41380 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
41381 loaded libraries and their SVR4 linker parameters. For each library on SVR4
41382 target, the following parameters are reported:
41386 @code{name}, the absolute file name from the @code{l_name} field of
41387 @code{struct link_map}.
41389 @code{lm} with address of @code{struct link_map} used for TLS
41390 (Thread Local Storage) access.
41392 @code{l_addr}, the displacement as read from the field @code{l_addr} of
41393 @code{struct link_map}. For prelinked libraries this is not an absolute
41394 memory address. It is a displacement of absolute memory address against
41395 address the file was prelinked to during the library load.
41397 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
41400 Additionally the single @code{main-lm} attribute specifies address of
41401 @code{struct link_map} used for the main executable. This parameter is used
41402 for TLS access and its presence is optional.
41404 @value{GDBN} must be linked with the Expat library to support XML
41405 SVR4 library lists. @xref{Expat}.
41407 A simple memory map, with two loaded libraries (which do not use prelink),
41411 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
41412 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
41414 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
41416 </library-list-svr>
41419 The format of an SVR4 library list is described by this DTD:
41422 <!-- library-list-svr4: Root element with versioning -->
41423 <!ELEMENT library-list-svr4 (library)*>
41424 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
41425 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
41426 <!ELEMENT library EMPTY>
41427 <!ATTLIST library name CDATA #REQUIRED>
41428 <!ATTLIST library lm CDATA #REQUIRED>
41429 <!ATTLIST library l_addr CDATA #REQUIRED>
41430 <!ATTLIST library l_ld CDATA #REQUIRED>
41433 @node Memory Map Format
41434 @section Memory Map Format
41435 @cindex memory map format
41437 To be able to write into flash memory, @value{GDBN} needs to obtain a
41438 memory map from the target. This section describes the format of the
41441 The memory map is obtained using the @samp{qXfer:memory-map:read}
41442 (@pxref{qXfer memory map read}) packet and is an XML document that
41443 lists memory regions.
41445 @value{GDBN} must be linked with the Expat library to support XML
41446 memory maps. @xref{Expat}.
41448 The top-level structure of the document is shown below:
41451 <?xml version="1.0"?>
41452 <!DOCTYPE memory-map
41453 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
41454 "http://sourceware.org/gdb/gdb-memory-map.dtd">
41460 Each region can be either:
41465 A region of RAM starting at @var{addr} and extending for @var{length}
41469 <memory type="ram" start="@var{addr}" length="@var{length}"/>
41474 A region of read-only memory:
41477 <memory type="rom" start="@var{addr}" length="@var{length}"/>
41482 A region of flash memory, with erasure blocks @var{blocksize}
41486 <memory type="flash" start="@var{addr}" length="@var{length}">
41487 <property name="blocksize">@var{blocksize}</property>
41493 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
41494 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
41495 packets to write to addresses in such ranges.
41497 The formal DTD for memory map format is given below:
41500 <!-- ................................................... -->
41501 <!-- Memory Map XML DTD ................................ -->
41502 <!-- File: memory-map.dtd .............................. -->
41503 <!-- .................................... .............. -->
41504 <!-- memory-map.dtd -->
41505 <!-- memory-map: Root element with versioning -->
41506 <!ELEMENT memory-map (memory | property)>
41507 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
41508 <!ELEMENT memory (property)>
41509 <!-- memory: Specifies a memory region,
41510 and its type, or device. -->
41511 <!ATTLIST memory type CDATA #REQUIRED
41512 start CDATA #REQUIRED
41513 length CDATA #REQUIRED
41514 device CDATA #IMPLIED>
41515 <!-- property: Generic attribute tag -->
41516 <!ELEMENT property (#PCDATA | property)*>
41517 <!ATTLIST property name CDATA #REQUIRED>
41520 @node Thread List Format
41521 @section Thread List Format
41522 @cindex thread list format
41524 To efficiently update the list of threads and their attributes,
41525 @value{GDBN} issues the @samp{qXfer:threads:read} packet
41526 (@pxref{qXfer threads read}) and obtains the XML document with
41527 the following structure:
41530 <?xml version="1.0"?>
41532 <thread id="id" core="0">
41533 ... description ...
41538 Each @samp{thread} element must have the @samp{id} attribute that
41539 identifies the thread (@pxref{thread-id syntax}). The
41540 @samp{core} attribute, if present, specifies which processor core
41541 the thread was last executing on. The content of the of @samp{thread}
41542 element is interpreted as human-readable auxilliary information.
41544 @node Traceframe Info Format
41545 @section Traceframe Info Format
41546 @cindex traceframe info format
41548 To be able to know which objects in the inferior can be examined when
41549 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
41550 memory ranges, registers and trace state variables that have been
41551 collected in a traceframe.
41553 This list is obtained using the @samp{qXfer:traceframe-info:read}
41554 (@pxref{qXfer traceframe info read}) packet and is an XML document.
41556 @value{GDBN} must be linked with the Expat library to support XML
41557 traceframe info discovery. @xref{Expat}.
41559 The top-level structure of the document is shown below:
41562 <?xml version="1.0"?>
41563 <!DOCTYPE traceframe-info
41564 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
41565 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
41571 Each traceframe block can be either:
41576 A region of collected memory starting at @var{addr} and extending for
41577 @var{length} bytes from there:
41580 <memory start="@var{addr}" length="@var{length}"/>
41585 The formal DTD for the traceframe info format is given below:
41588 <!ELEMENT traceframe-info (memory)* >
41589 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
41591 <!ELEMENT memory EMPTY>
41592 <!ATTLIST memory start CDATA #REQUIRED
41593 length CDATA #REQUIRED>
41596 @node Branch Trace Format
41597 @section Branch Trace Format
41598 @cindex branch trace format
41600 In order to display the branch trace of an inferior thread,
41601 @value{GDBN} needs to obtain the list of branches. This list is
41602 represented as list of sequential code blocks that are connected via
41603 branches. The code in each block has been executed sequentially.
41605 This list is obtained using the @samp{qXfer:btrace:read}
41606 (@pxref{qXfer btrace read}) packet and is an XML document.
41608 @value{GDBN} must be linked with the Expat library to support XML
41609 traceframe info discovery. @xref{Expat}.
41611 The top-level structure of the document is shown below:
41614 <?xml version="1.0"?>
41616 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
41617 "http://sourceware.org/gdb/gdb-btrace.dtd">
41626 A block of sequentially executed instructions starting at @var{begin}
41627 and ending at @var{end}:
41630 <block begin="@var{begin}" end="@var{end}"/>
41635 The formal DTD for the branch trace format is given below:
41638 <!ELEMENT btrace (block)* >
41639 <!ATTLIST btrace version CDATA #FIXED "1.0">
41641 <!ELEMENT block EMPTY>
41642 <!ATTLIST block begin CDATA #REQUIRED
41643 end CDATA #REQUIRED>
41646 @include agentexpr.texi
41648 @node Target Descriptions
41649 @appendix Target Descriptions
41650 @cindex target descriptions
41652 One of the challenges of using @value{GDBN} to debug embedded systems
41653 is that there are so many minor variants of each processor
41654 architecture in use. It is common practice for vendors to start with
41655 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
41656 and then make changes to adapt it to a particular market niche. Some
41657 architectures have hundreds of variants, available from dozens of
41658 vendors. This leads to a number of problems:
41662 With so many different customized processors, it is difficult for
41663 the @value{GDBN} maintainers to keep up with the changes.
41665 Since individual variants may have short lifetimes or limited
41666 audiences, it may not be worthwhile to carry information about every
41667 variant in the @value{GDBN} source tree.
41669 When @value{GDBN} does support the architecture of the embedded system
41670 at hand, the task of finding the correct architecture name to give the
41671 @command{set architecture} command can be error-prone.
41674 To address these problems, the @value{GDBN} remote protocol allows a
41675 target system to not only identify itself to @value{GDBN}, but to
41676 actually describe its own features. This lets @value{GDBN} support
41677 processor variants it has never seen before --- to the extent that the
41678 descriptions are accurate, and that @value{GDBN} understands them.
41680 @value{GDBN} must be linked with the Expat library to support XML
41681 target descriptions. @xref{Expat}.
41684 * Retrieving Descriptions:: How descriptions are fetched from a target.
41685 * Target Description Format:: The contents of a target description.
41686 * Predefined Target Types:: Standard types available for target
41688 * Standard Target Features:: Features @value{GDBN} knows about.
41691 @node Retrieving Descriptions
41692 @section Retrieving Descriptions
41694 Target descriptions can be read from the target automatically, or
41695 specified by the user manually. The default behavior is to read the
41696 description from the target. @value{GDBN} retrieves it via the remote
41697 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
41698 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
41699 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
41700 XML document, of the form described in @ref{Target Description
41703 Alternatively, you can specify a file to read for the target description.
41704 If a file is set, the target will not be queried. The commands to
41705 specify a file are:
41708 @cindex set tdesc filename
41709 @item set tdesc filename @var{path}
41710 Read the target description from @var{path}.
41712 @cindex unset tdesc filename
41713 @item unset tdesc filename
41714 Do not read the XML target description from a file. @value{GDBN}
41715 will use the description supplied by the current target.
41717 @cindex show tdesc filename
41718 @item show tdesc filename
41719 Show the filename to read for a target description, if any.
41723 @node Target Description Format
41724 @section Target Description Format
41725 @cindex target descriptions, XML format
41727 A target description annex is an @uref{http://www.w3.org/XML/, XML}
41728 document which complies with the Document Type Definition provided in
41729 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
41730 means you can use generally available tools like @command{xmllint} to
41731 check that your feature descriptions are well-formed and valid.
41732 However, to help people unfamiliar with XML write descriptions for
41733 their targets, we also describe the grammar here.
41735 Target descriptions can identify the architecture of the remote target
41736 and (for some architectures) provide information about custom register
41737 sets. They can also identify the OS ABI of the remote target.
41738 @value{GDBN} can use this information to autoconfigure for your
41739 target, or to warn you if you connect to an unsupported target.
41741 Here is a simple target description:
41744 <target version="1.0">
41745 <architecture>i386:x86-64</architecture>
41750 This minimal description only says that the target uses
41751 the x86-64 architecture.
41753 A target description has the following overall form, with [ ] marking
41754 optional elements and @dots{} marking repeatable elements. The elements
41755 are explained further below.
41758 <?xml version="1.0"?>
41759 <!DOCTYPE target SYSTEM "gdb-target.dtd">
41760 <target version="1.0">
41761 @r{[}@var{architecture}@r{]}
41762 @r{[}@var{osabi}@r{]}
41763 @r{[}@var{compatible}@r{]}
41764 @r{[}@var{feature}@dots{}@r{]}
41769 The description is generally insensitive to whitespace and line
41770 breaks, under the usual common-sense rules. The XML version
41771 declaration and document type declaration can generally be omitted
41772 (@value{GDBN} does not require them), but specifying them may be
41773 useful for XML validation tools. The @samp{version} attribute for
41774 @samp{<target>} may also be omitted, but we recommend
41775 including it; if future versions of @value{GDBN} use an incompatible
41776 revision of @file{gdb-target.dtd}, they will detect and report
41777 the version mismatch.
41779 @subsection Inclusion
41780 @cindex target descriptions, inclusion
41783 @cindex <xi:include>
41786 It can sometimes be valuable to split a target description up into
41787 several different annexes, either for organizational purposes, or to
41788 share files between different possible target descriptions. You can
41789 divide a description into multiple files by replacing any element of
41790 the target description with an inclusion directive of the form:
41793 <xi:include href="@var{document}"/>
41797 When @value{GDBN} encounters an element of this form, it will retrieve
41798 the named XML @var{document}, and replace the inclusion directive with
41799 the contents of that document. If the current description was read
41800 using @samp{qXfer}, then so will be the included document;
41801 @var{document} will be interpreted as the name of an annex. If the
41802 current description was read from a file, @value{GDBN} will look for
41803 @var{document} as a file in the same directory where it found the
41804 original description.
41806 @subsection Architecture
41807 @cindex <architecture>
41809 An @samp{<architecture>} element has this form:
41812 <architecture>@var{arch}</architecture>
41815 @var{arch} is one of the architectures from the set accepted by
41816 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
41819 @cindex @code{<osabi>}
41821 This optional field was introduced in @value{GDBN} version 7.0.
41822 Previous versions of @value{GDBN} ignore it.
41824 An @samp{<osabi>} element has this form:
41827 <osabi>@var{abi-name}</osabi>
41830 @var{abi-name} is an OS ABI name from the same selection accepted by
41831 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
41833 @subsection Compatible Architecture
41834 @cindex @code{<compatible>}
41836 This optional field was introduced in @value{GDBN} version 7.0.
41837 Previous versions of @value{GDBN} ignore it.
41839 A @samp{<compatible>} element has this form:
41842 <compatible>@var{arch}</compatible>
41845 @var{arch} is one of the architectures from the set accepted by
41846 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
41848 A @samp{<compatible>} element is used to specify that the target
41849 is able to run binaries in some other than the main target architecture
41850 given by the @samp{<architecture>} element. For example, on the
41851 Cell Broadband Engine, the main architecture is @code{powerpc:common}
41852 or @code{powerpc:common64}, but the system is able to run binaries
41853 in the @code{spu} architecture as well. The way to describe this
41854 capability with @samp{<compatible>} is as follows:
41857 <architecture>powerpc:common</architecture>
41858 <compatible>spu</compatible>
41861 @subsection Features
41864 Each @samp{<feature>} describes some logical portion of the target
41865 system. Features are currently used to describe available CPU
41866 registers and the types of their contents. A @samp{<feature>} element
41870 <feature name="@var{name}">
41871 @r{[}@var{type}@dots{}@r{]}
41877 Each feature's name should be unique within the description. The name
41878 of a feature does not matter unless @value{GDBN} has some special
41879 knowledge of the contents of that feature; if it does, the feature
41880 should have its standard name. @xref{Standard Target Features}.
41884 Any register's value is a collection of bits which @value{GDBN} must
41885 interpret. The default interpretation is a two's complement integer,
41886 but other types can be requested by name in the register description.
41887 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
41888 Target Types}), and the description can define additional composite types.
41890 Each type element must have an @samp{id} attribute, which gives
41891 a unique (within the containing @samp{<feature>}) name to the type.
41892 Types must be defined before they are used.
41895 Some targets offer vector registers, which can be treated as arrays
41896 of scalar elements. These types are written as @samp{<vector>} elements,
41897 specifying the array element type, @var{type}, and the number of elements,
41901 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
41905 If a register's value is usefully viewed in multiple ways, define it
41906 with a union type containing the useful representations. The
41907 @samp{<union>} element contains one or more @samp{<field>} elements,
41908 each of which has a @var{name} and a @var{type}:
41911 <union id="@var{id}">
41912 <field name="@var{name}" type="@var{type}"/>
41918 If a register's value is composed from several separate values, define
41919 it with a structure type. There are two forms of the @samp{<struct>}
41920 element; a @samp{<struct>} element must either contain only bitfields
41921 or contain no bitfields. If the structure contains only bitfields,
41922 its total size in bytes must be specified, each bitfield must have an
41923 explicit start and end, and bitfields are automatically assigned an
41924 integer type. The field's @var{start} should be less than or
41925 equal to its @var{end}, and zero represents the least significant bit.
41928 <struct id="@var{id}" size="@var{size}">
41929 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
41934 If the structure contains no bitfields, then each field has an
41935 explicit type, and no implicit padding is added.
41938 <struct id="@var{id}">
41939 <field name="@var{name}" type="@var{type}"/>
41945 If a register's value is a series of single-bit flags, define it with
41946 a flags type. The @samp{<flags>} element has an explicit @var{size}
41947 and contains one or more @samp{<field>} elements. Each field has a
41948 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
41952 <flags id="@var{id}" size="@var{size}">
41953 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
41958 @subsection Registers
41961 Each register is represented as an element with this form:
41964 <reg name="@var{name}"
41965 bitsize="@var{size}"
41966 @r{[}regnum="@var{num}"@r{]}
41967 @r{[}save-restore="@var{save-restore}"@r{]}
41968 @r{[}type="@var{type}"@r{]}
41969 @r{[}group="@var{group}"@r{]}/>
41973 The components are as follows:
41978 The register's name; it must be unique within the target description.
41981 The register's size, in bits.
41984 The register's number. If omitted, a register's number is one greater
41985 than that of the previous register (either in the current feature or in
41986 a preceding feature); the first register in the target description
41987 defaults to zero. This register number is used to read or write
41988 the register; e.g.@: it is used in the remote @code{p} and @code{P}
41989 packets, and registers appear in the @code{g} and @code{G} packets
41990 in order of increasing register number.
41993 Whether the register should be preserved across inferior function
41994 calls; this must be either @code{yes} or @code{no}. The default is
41995 @code{yes}, which is appropriate for most registers except for
41996 some system control registers; this is not related to the target's
42000 The type of the register. @var{type} may be a predefined type, a type
42001 defined in the current feature, or one of the special types @code{int}
42002 and @code{float}. @code{int} is an integer type of the correct size
42003 for @var{bitsize}, and @code{float} is a floating point type (in the
42004 architecture's normal floating point format) of the correct size for
42005 @var{bitsize}. The default is @code{int}.
42008 The register group to which this register belongs. @var{group} must
42009 be either @code{general}, @code{float}, or @code{vector}. If no
42010 @var{group} is specified, @value{GDBN} will not display the register
42011 in @code{info registers}.
42015 @node Predefined Target Types
42016 @section Predefined Target Types
42017 @cindex target descriptions, predefined types
42019 Type definitions in the self-description can build up composite types
42020 from basic building blocks, but can not define fundamental types. Instead,
42021 standard identifiers are provided by @value{GDBN} for the fundamental
42022 types. The currently supported types are:
42031 Signed integer types holding the specified number of bits.
42038 Unsigned integer types holding the specified number of bits.
42042 Pointers to unspecified code and data. The program counter and
42043 any dedicated return address register may be marked as code
42044 pointers; printing a code pointer converts it into a symbolic
42045 address. The stack pointer and any dedicated address registers
42046 may be marked as data pointers.
42049 Single precision IEEE floating point.
42052 Double precision IEEE floating point.
42055 The 12-byte extended precision format used by ARM FPA registers.
42058 The 10-byte extended precision format used by x87 registers.
42061 32bit @sc{eflags} register used by x86.
42064 32bit @sc{mxcsr} register used by x86.
42068 @node Standard Target Features
42069 @section Standard Target Features
42070 @cindex target descriptions, standard features
42072 A target description must contain either no registers or all the
42073 target's registers. If the description contains no registers, then
42074 @value{GDBN} will assume a default register layout, selected based on
42075 the architecture. If the description contains any registers, the
42076 default layout will not be used; the standard registers must be
42077 described in the target description, in such a way that @value{GDBN}
42078 can recognize them.
42080 This is accomplished by giving specific names to feature elements
42081 which contain standard registers. @value{GDBN} will look for features
42082 with those names and verify that they contain the expected registers;
42083 if any known feature is missing required registers, or if any required
42084 feature is missing, @value{GDBN} will reject the target
42085 description. You can add additional registers to any of the
42086 standard features --- @value{GDBN} will display them just as if
42087 they were added to an unrecognized feature.
42089 This section lists the known features and their expected contents.
42090 Sample XML documents for these features are included in the
42091 @value{GDBN} source tree, in the directory @file{gdb/features}.
42093 Names recognized by @value{GDBN} should include the name of the
42094 company or organization which selected the name, and the overall
42095 architecture to which the feature applies; so e.g.@: the feature
42096 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
42098 The names of registers are not case sensitive for the purpose
42099 of recognizing standard features, but @value{GDBN} will only display
42100 registers using the capitalization used in the description.
42103 * AArch64 Features::
42108 * Nios II Features::
42109 * PowerPC Features::
42114 @node AArch64 Features
42115 @subsection AArch64 Features
42116 @cindex target descriptions, AArch64 features
42118 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
42119 targets. It should contain registers @samp{x0} through @samp{x30},
42120 @samp{sp}, @samp{pc}, and @samp{cpsr}.
42122 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
42123 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
42127 @subsection ARM Features
42128 @cindex target descriptions, ARM features
42130 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
42132 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
42133 @samp{lr}, @samp{pc}, and @samp{cpsr}.
42135 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
42136 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
42137 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
42140 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
42141 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
42143 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
42144 it should contain at least registers @samp{wR0} through @samp{wR15} and
42145 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
42146 @samp{wCSSF}, and @samp{wCASF} registers are optional.
42148 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
42149 should contain at least registers @samp{d0} through @samp{d15}. If
42150 they are present, @samp{d16} through @samp{d31} should also be included.
42151 @value{GDBN} will synthesize the single-precision registers from
42152 halves of the double-precision registers.
42154 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
42155 need to contain registers; it instructs @value{GDBN} to display the
42156 VFP double-precision registers as vectors and to synthesize the
42157 quad-precision registers from pairs of double-precision registers.
42158 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
42159 be present and include 32 double-precision registers.
42161 @node i386 Features
42162 @subsection i386 Features
42163 @cindex target descriptions, i386 features
42165 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
42166 targets. It should describe the following registers:
42170 @samp{eax} through @samp{edi} plus @samp{eip} for i386
42172 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
42174 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
42175 @samp{fs}, @samp{gs}
42177 @samp{st0} through @samp{st7}
42179 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
42180 @samp{foseg}, @samp{fooff} and @samp{fop}
42183 The register sets may be different, depending on the target.
42185 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
42186 describe registers:
42190 @samp{xmm0} through @samp{xmm7} for i386
42192 @samp{xmm0} through @samp{xmm15} for amd64
42197 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
42198 @samp{org.gnu.gdb.i386.sse} feature. It should
42199 describe the upper 128 bits of @sc{ymm} registers:
42203 @samp{ymm0h} through @samp{ymm7h} for i386
42205 @samp{ymm0h} through @samp{ymm15h} for amd64
42208 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
42209 describe a single register, @samp{orig_eax}.
42211 @node MIPS Features
42212 @subsection @acronym{MIPS} Features
42213 @cindex target descriptions, @acronym{MIPS} features
42215 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
42216 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
42217 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
42220 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
42221 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
42222 registers. They may be 32-bit or 64-bit depending on the target.
42224 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
42225 it may be optional in a future version of @value{GDBN}. It should
42226 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
42227 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
42229 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
42230 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
42231 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
42232 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
42234 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
42235 contain a single register, @samp{restart}, which is used by the
42236 Linux kernel to control restartable syscalls.
42238 @node M68K Features
42239 @subsection M68K Features
42240 @cindex target descriptions, M68K features
42243 @item @samp{org.gnu.gdb.m68k.core}
42244 @itemx @samp{org.gnu.gdb.coldfire.core}
42245 @itemx @samp{org.gnu.gdb.fido.core}
42246 One of those features must be always present.
42247 The feature that is present determines which flavor of m68k is
42248 used. The feature that is present should contain registers
42249 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
42250 @samp{sp}, @samp{ps} and @samp{pc}.
42252 @item @samp{org.gnu.gdb.coldfire.fp}
42253 This feature is optional. If present, it should contain registers
42254 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
42258 @node Nios II Features
42259 @subsection Nios II Features
42260 @cindex target descriptions, Nios II features
42262 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
42263 targets. It should contain the 32 core registers (@samp{zero},
42264 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
42265 @samp{pc}, and the 16 control registers (@samp{status} through
42268 @node PowerPC Features
42269 @subsection PowerPC Features
42270 @cindex target descriptions, PowerPC features
42272 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
42273 targets. It should contain registers @samp{r0} through @samp{r31},
42274 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
42275 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
42277 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
42278 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
42280 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
42281 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
42284 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
42285 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
42286 will combine these registers with the floating point registers
42287 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
42288 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
42289 through @samp{vs63}, the set of vector registers for POWER7.
42291 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
42292 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
42293 @samp{spefscr}. SPE targets should provide 32-bit registers in
42294 @samp{org.gnu.gdb.power.core} and provide the upper halves in
42295 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
42296 these to present registers @samp{ev0} through @samp{ev31} to the
42299 @node TIC6x Features
42300 @subsection TMS320C6x Features
42301 @cindex target descriptions, TIC6x features
42302 @cindex target descriptions, TMS320C6x features
42303 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
42304 targets. It should contain registers @samp{A0} through @samp{A15},
42305 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
42307 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
42308 contain registers @samp{A16} through @samp{A31} and @samp{B16}
42309 through @samp{B31}.
42311 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
42312 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
42314 @node Operating System Information
42315 @appendix Operating System Information
42316 @cindex operating system information
42322 Users of @value{GDBN} often wish to obtain information about the state of
42323 the operating system running on the target---for example the list of
42324 processes, or the list of open files. This section describes the
42325 mechanism that makes it possible. This mechanism is similar to the
42326 target features mechanism (@pxref{Target Descriptions}), but focuses
42327 on a different aspect of target.
42329 Operating system information is retrived from the target via the
42330 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
42331 read}). The object name in the request should be @samp{osdata}, and
42332 the @var{annex} identifies the data to be fetched.
42335 @appendixsection Process list
42336 @cindex operating system information, process list
42338 When requesting the process list, the @var{annex} field in the
42339 @samp{qXfer} request should be @samp{processes}. The returned data is
42340 an XML document. The formal syntax of this document is defined in
42341 @file{gdb/features/osdata.dtd}.
42343 An example document is:
42346 <?xml version="1.0"?>
42347 <!DOCTYPE target SYSTEM "osdata.dtd">
42348 <osdata type="processes">
42350 <column name="pid">1</column>
42351 <column name="user">root</column>
42352 <column name="command">/sbin/init</column>
42353 <column name="cores">1,2,3</column>
42358 Each item should include a column whose name is @samp{pid}. The value
42359 of that column should identify the process on the target. The
42360 @samp{user} and @samp{command} columns are optional, and will be
42361 displayed by @value{GDBN}. The @samp{cores} column, if present,
42362 should contain a comma-separated list of cores that this process
42363 is running on. Target may provide additional columns,
42364 which @value{GDBN} currently ignores.
42366 @node Trace File Format
42367 @appendix Trace File Format
42368 @cindex trace file format
42370 The trace file comes in three parts: a header, a textual description
42371 section, and a trace frame section with binary data.
42373 The header has the form @code{\x7fTRACE0\n}. The first byte is
42374 @code{0x7f} so as to indicate that the file contains binary data,
42375 while the @code{0} is a version number that may have different values
42378 The description section consists of multiple lines of @sc{ascii} text
42379 separated by newline characters (@code{0xa}). The lines may include a
42380 variety of optional descriptive or context-setting information, such
42381 as tracepoint definitions or register set size. @value{GDBN} will
42382 ignore any line that it does not recognize. An empty line marks the end
42385 @c FIXME add some specific types of data
42387 The trace frame section consists of a number of consecutive frames.
42388 Each frame begins with a two-byte tracepoint number, followed by a
42389 four-byte size giving the amount of data in the frame. The data in
42390 the frame consists of a number of blocks, each introduced by a
42391 character indicating its type (at least register, memory, and trace
42392 state variable). The data in this section is raw binary, not a
42393 hexadecimal or other encoding; its endianness matches the target's
42396 @c FIXME bi-arch may require endianness/arch info in description section
42399 @item R @var{bytes}
42400 Register block. The number and ordering of bytes matches that of a
42401 @code{g} packet in the remote protocol. Note that these are the
42402 actual bytes, in target order and @value{GDBN} register order, not a
42403 hexadecimal encoding.
42405 @item M @var{address} @var{length} @var{bytes}...
42406 Memory block. This is a contiguous block of memory, at the 8-byte
42407 address @var{address}, with a 2-byte length @var{length}, followed by
42408 @var{length} bytes.
42410 @item V @var{number} @var{value}
42411 Trace state variable block. This records the 8-byte signed value
42412 @var{value} of trace state variable numbered @var{number}.
42416 Future enhancements of the trace file format may include additional types
42419 @node Index Section Format
42420 @appendix @code{.gdb_index} section format
42421 @cindex .gdb_index section format
42422 @cindex index section format
42424 This section documents the index section that is created by @code{save
42425 gdb-index} (@pxref{Index Files}). The index section is
42426 DWARF-specific; some knowledge of DWARF is assumed in this
42429 The mapped index file format is designed to be directly
42430 @code{mmap}able on any architecture. In most cases, a datum is
42431 represented using a little-endian 32-bit integer value, called an
42432 @code{offset_type}. Big endian machines must byte-swap the values
42433 before using them. Exceptions to this rule are noted. The data is
42434 laid out such that alignment is always respected.
42436 A mapped index consists of several areas, laid out in order.
42440 The file header. This is a sequence of values, of @code{offset_type}
42441 unless otherwise noted:
42445 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
42446 Version 4 uses a different hashing function from versions 5 and 6.
42447 Version 6 includes symbols for inlined functions, whereas versions 4
42448 and 5 do not. Version 7 adds attributes to the CU indices in the
42449 symbol table. Version 8 specifies that symbols from DWARF type units
42450 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
42451 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
42453 @value{GDBN} will only read version 4, 5, or 6 indices
42454 by specifying @code{set use-deprecated-index-sections on}.
42455 GDB has a workaround for potentially broken version 7 indices so it is
42456 currently not flagged as deprecated.
42459 The offset, from the start of the file, of the CU list.
42462 The offset, from the start of the file, of the types CU list. Note
42463 that this area can be empty, in which case this offset will be equal
42464 to the next offset.
42467 The offset, from the start of the file, of the address area.
42470 The offset, from the start of the file, of the symbol table.
42473 The offset, from the start of the file, of the constant pool.
42477 The CU list. This is a sequence of pairs of 64-bit little-endian
42478 values, sorted by the CU offset. The first element in each pair is
42479 the offset of a CU in the @code{.debug_info} section. The second
42480 element in each pair is the length of that CU. References to a CU
42481 elsewhere in the map are done using a CU index, which is just the
42482 0-based index into this table. Note that if there are type CUs, then
42483 conceptually CUs and type CUs form a single list for the purposes of
42487 The types CU list. This is a sequence of triplets of 64-bit
42488 little-endian values. In a triplet, the first value is the CU offset,
42489 the second value is the type offset in the CU, and the third value is
42490 the type signature. The types CU list is not sorted.
42493 The address area. The address area consists of a sequence of address
42494 entries. Each address entry has three elements:
42498 The low address. This is a 64-bit little-endian value.
42501 The high address. This is a 64-bit little-endian value. Like
42502 @code{DW_AT_high_pc}, the value is one byte beyond the end.
42505 The CU index. This is an @code{offset_type} value.
42509 The symbol table. This is an open-addressed hash table. The size of
42510 the hash table is always a power of 2.
42512 Each slot in the hash table consists of a pair of @code{offset_type}
42513 values. The first value is the offset of the symbol's name in the
42514 constant pool. The second value is the offset of the CU vector in the
42517 If both values are 0, then this slot in the hash table is empty. This
42518 is ok because while 0 is a valid constant pool index, it cannot be a
42519 valid index for both a string and a CU vector.
42521 The hash value for a table entry is computed by applying an
42522 iterative hash function to the symbol's name. Starting with an
42523 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
42524 the string is incorporated into the hash using the formula depending on the
42529 The formula is @code{r = r * 67 + c - 113}.
42531 @item Versions 5 to 7
42532 The formula is @code{r = r * 67 + tolower (c) - 113}.
42535 The terminating @samp{\0} is not incorporated into the hash.
42537 The step size used in the hash table is computed via
42538 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
42539 value, and @samp{size} is the size of the hash table. The step size
42540 is used to find the next candidate slot when handling a hash
42543 The names of C@t{++} symbols in the hash table are canonicalized. We
42544 don't currently have a simple description of the canonicalization
42545 algorithm; if you intend to create new index sections, you must read
42549 The constant pool. This is simply a bunch of bytes. It is organized
42550 so that alignment is correct: CU vectors are stored first, followed by
42553 A CU vector in the constant pool is a sequence of @code{offset_type}
42554 values. The first value is the number of CU indices in the vector.
42555 Each subsequent value is the index and symbol attributes of a CU in
42556 the CU list. This element in the hash table is used to indicate which
42557 CUs define the symbol and how the symbol is used.
42558 See below for the format of each CU index+attributes entry.
42560 A string in the constant pool is zero-terminated.
42563 Attributes were added to CU index values in @code{.gdb_index} version 7.
42564 If a symbol has multiple uses within a CU then there is one
42565 CU index+attributes value for each use.
42567 The format of each CU index+attributes entry is as follows
42573 This is the index of the CU in the CU list.
42575 These bits are reserved for future purposes and must be zero.
42577 The kind of the symbol in the CU.
42581 This value is reserved and should not be used.
42582 By reserving zero the full @code{offset_type} value is backwards compatible
42583 with previous versions of the index.
42585 The symbol is a type.
42587 The symbol is a variable or an enum value.
42589 The symbol is a function.
42591 Any other kind of symbol.
42593 These values are reserved.
42597 This bit is zero if the value is global and one if it is static.
42599 The determination of whether a symbol is global or static is complicated.
42600 The authorative reference is the file @file{dwarf2read.c} in
42601 @value{GDBN} sources.
42605 This pseudo-code describes the computation of a symbol's kind and
42606 global/static attributes in the index.
42609 is_external = get_attribute (die, DW_AT_external);
42610 language = get_attribute (cu_die, DW_AT_language);
42613 case DW_TAG_typedef:
42614 case DW_TAG_base_type:
42615 case DW_TAG_subrange_type:
42619 case DW_TAG_enumerator:
42621 is_static = (language != CPLUS && language != JAVA);
42623 case DW_TAG_subprogram:
42625 is_static = ! (is_external || language == ADA);
42627 case DW_TAG_constant:
42629 is_static = ! is_external;
42631 case DW_TAG_variable:
42633 is_static = ! is_external;
42635 case DW_TAG_namespace:
42639 case DW_TAG_class_type:
42640 case DW_TAG_interface_type:
42641 case DW_TAG_structure_type:
42642 case DW_TAG_union_type:
42643 case DW_TAG_enumeration_type:
42645 is_static = (language != CPLUS && language != JAVA);
42653 @appendix Manual pages
42657 * gdb man:: The GNU Debugger man page
42658 * gdbserver man:: Remote Server for the GNU Debugger man page
42659 * gcore man:: Generate a core file of a running program
42660 * gdbinit man:: gdbinit scripts
42666 @c man title gdb The GNU Debugger
42668 @c man begin SYNOPSIS gdb
42669 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
42670 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
42671 [@option{-b}@w{ }@var{bps}]
42672 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
42673 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
42674 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
42675 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
42676 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
42679 @c man begin DESCRIPTION gdb
42680 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
42681 going on ``inside'' another program while it executes -- or what another
42682 program was doing at the moment it crashed.
42684 @value{GDBN} can do four main kinds of things (plus other things in support of
42685 these) to help you catch bugs in the act:
42689 Start your program, specifying anything that might affect its behavior.
42692 Make your program stop on specified conditions.
42695 Examine what has happened, when your program has stopped.
42698 Change things in your program, so you can experiment with correcting the
42699 effects of one bug and go on to learn about another.
42702 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
42705 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
42706 commands from the terminal until you tell it to exit with the @value{GDBN}
42707 command @code{quit}. You can get online help from @value{GDBN} itself
42708 by using the command @code{help}.
42710 You can run @code{gdb} with no arguments or options; but the most
42711 usual way to start @value{GDBN} is with one argument or two, specifying an
42712 executable program as the argument:
42718 You can also start with both an executable program and a core file specified:
42724 You can, instead, specify a process ID as a second argument, if you want
42725 to debug a running process:
42733 would attach @value{GDBN} to process @code{1234} (unless you also have a file
42734 named @file{1234}; @value{GDBN} does check for a core file first).
42735 With option @option{-p} you can omit the @var{program} filename.
42737 Here are some of the most frequently needed @value{GDBN} commands:
42739 @c pod2man highlights the right hand side of the @item lines.
42741 @item break [@var{file}:]@var{functiop}
42742 Set a breakpoint at @var{function} (in @var{file}).
42744 @item run [@var{arglist}]
42745 Start your program (with @var{arglist}, if specified).
42748 Backtrace: display the program stack.
42750 @item print @var{expr}
42751 Display the value of an expression.
42754 Continue running your program (after stopping, e.g. at a breakpoint).
42757 Execute next program line (after stopping); step @emph{over} any
42758 function calls in the line.
42760 @item edit [@var{file}:]@var{function}
42761 look at the program line where it is presently stopped.
42763 @item list [@var{file}:]@var{function}
42764 type the text of the program in the vicinity of where it is presently stopped.
42767 Execute next program line (after stopping); step @emph{into} any
42768 function calls in the line.
42770 @item help [@var{name}]
42771 Show information about @value{GDBN} command @var{name}, or general information
42772 about using @value{GDBN}.
42775 Exit from @value{GDBN}.
42779 For full details on @value{GDBN},
42780 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42781 by Richard M. Stallman and Roland H. Pesch. The same text is available online
42782 as the @code{gdb} entry in the @code{info} program.
42786 @c man begin OPTIONS gdb
42787 Any arguments other than options specify an executable
42788 file and core file (or process ID); that is, the first argument
42789 encountered with no
42790 associated option flag is equivalent to a @option{-se} option, and the second,
42791 if any, is equivalent to a @option{-c} option if it's the name of a file.
42793 both long and short forms; both are shown here. The long forms are also
42794 recognized if you truncate them, so long as enough of the option is
42795 present to be unambiguous. (If you prefer, you can flag option
42796 arguments with @option{+} rather than @option{-}, though we illustrate the
42797 more usual convention.)
42799 All the options and command line arguments you give are processed
42800 in sequential order. The order makes a difference when the @option{-x}
42806 List all options, with brief explanations.
42808 @item -symbols=@var{file}
42809 @itemx -s @var{file}
42810 Read symbol table from file @var{file}.
42813 Enable writing into executable and core files.
42815 @item -exec=@var{file}
42816 @itemx -e @var{file}
42817 Use file @var{file} as the executable file to execute when
42818 appropriate, and for examining pure data in conjunction with a core
42821 @item -se=@var{file}
42822 Read symbol table from file @var{file} and use it as the executable
42825 @item -core=@var{file}
42826 @itemx -c @var{file}
42827 Use file @var{file} as a core dump to examine.
42829 @item -command=@var{file}
42830 @itemx -x @var{file}
42831 Execute @value{GDBN} commands from file @var{file}.
42833 @item -ex @var{command}
42834 Execute given @value{GDBN} @var{command}.
42836 @item -directory=@var{directory}
42837 @itemx -d @var{directory}
42838 Add @var{directory} to the path to search for source files.
42841 Do not execute commands from @file{~/.gdbinit}.
42845 Do not execute commands from any @file{.gdbinit} initialization files.
42849 ``Quiet''. Do not print the introductory and copyright messages. These
42850 messages are also suppressed in batch mode.
42853 Run in batch mode. Exit with status @code{0} after processing all the command
42854 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
42855 Exit with nonzero status if an error occurs in executing the @value{GDBN}
42856 commands in the command files.
42858 Batch mode may be useful for running @value{GDBN} as a filter, for example to
42859 download and run a program on another computer; in order to make this
42860 more useful, the message
42863 Program exited normally.
42867 (which is ordinarily issued whenever a program running under @value{GDBN} control
42868 terminates) is not issued when running in batch mode.
42870 @item -cd=@var{directory}
42871 Run @value{GDBN} using @var{directory} as its working directory,
42872 instead of the current directory.
42876 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
42877 @value{GDBN} to output the full file name and line number in a standard,
42878 recognizable fashion each time a stack frame is displayed (which
42879 includes each time the program stops). This recognizable format looks
42880 like two @samp{\032} characters, followed by the file name, line number
42881 and character position separated by colons, and a newline. The
42882 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
42883 characters as a signal to display the source code for the frame.
42886 Set the line speed (baud rate or bits per second) of any serial
42887 interface used by @value{GDBN} for remote debugging.
42889 @item -tty=@var{device}
42890 Run using @var{device} for your program's standard input and output.
42894 @c man begin SEEALSO gdb
42896 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42897 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42898 documentation are properly installed at your site, the command
42905 should give you access to the complete manual.
42907 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42908 Richard M. Stallman and Roland H. Pesch, July 1991.
42912 @node gdbserver man
42913 @heading gdbserver man
42915 @c man title gdbserver Remote Server for the GNU Debugger
42917 @c man begin SYNOPSIS gdbserver
42918 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
42920 gdbserver --attach @var{comm} @var{pid}
42922 gdbserver --multi @var{comm}
42926 @c man begin DESCRIPTION gdbserver
42927 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
42928 than the one which is running the program being debugged.
42931 @subheading Usage (server (target) side)
42934 Usage (server (target) side):
42937 First, you need to have a copy of the program you want to debug put onto
42938 the target system. The program can be stripped to save space if needed, as
42939 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
42940 the @value{GDBN} running on the host system.
42942 To use the server, you log on to the target system, and run the @command{gdbserver}
42943 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
42944 your program, and (c) its arguments. The general syntax is:
42947 target> gdbserver @var{comm} @var{program} [@var{args} ...]
42950 For example, using a serial port, you might say:
42954 @c @file would wrap it as F</dev/com1>.
42955 target> gdbserver /dev/com1 emacs foo.txt
42958 target> gdbserver @file{/dev/com1} emacs foo.txt
42962 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
42963 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
42964 waits patiently for the host @value{GDBN} to communicate with it.
42966 To use a TCP connection, you could say:
42969 target> gdbserver host:2345 emacs foo.txt
42972 This says pretty much the same thing as the last example, except that we are
42973 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
42974 that we are expecting to see a TCP connection from @code{host} to local TCP port
42975 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
42976 want for the port number as long as it does not conflict with any existing TCP
42977 ports on the target system. This same port number must be used in the host
42978 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
42979 you chose a port number that conflicts with another service, @command{gdbserver} will
42980 print an error message and exit.
42982 @command{gdbserver} can also attach to running programs.
42983 This is accomplished via the @option{--attach} argument. The syntax is:
42986 target> gdbserver --attach @var{comm} @var{pid}
42989 @var{pid} is the process ID of a currently running process. It isn't
42990 necessary to point @command{gdbserver} at a binary for the running process.
42992 To start @code{gdbserver} without supplying an initial command to run
42993 or process ID to attach, use the @option{--multi} command line option.
42994 In such case you should connect using @kbd{target extended-remote} to start
42995 the program you want to debug.
42998 target> gdbserver --multi @var{comm}
43002 @subheading Usage (host side)
43008 You need an unstripped copy of the target program on your host system, since
43009 @value{GDBN} needs to examine it's symbol tables and such. Start up @value{GDBN} as you normally
43010 would, with the target program as the first argument. (You may need to use the
43011 @option{--baud} option if the serial line is running at anything except 9600 baud.)
43012 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
43013 new command you need to know about is @code{target remote}
43014 (or @code{target extended-remote}). Its argument is either
43015 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
43016 descriptor. For example:
43020 @c @file would wrap it as F</dev/ttyb>.
43021 (gdb) target remote /dev/ttyb
43024 (gdb) target remote @file{/dev/ttyb}
43029 communicates with the server via serial line @file{/dev/ttyb}, and:
43032 (gdb) target remote the-target:2345
43036 communicates via a TCP connection to port 2345 on host `the-target', where
43037 you previously started up @command{gdbserver} with the same port number. Note that for
43038 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
43039 command, otherwise you may get an error that looks something like
43040 `Connection refused'.
43042 @command{gdbserver} can also debug multiple inferiors at once,
43045 the @value{GDBN} manual in node @code{Inferiors and Programs}
43046 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
43049 @ref{Inferiors and Programs}.
43051 In such case use the @code{extended-remote} @value{GDBN} command variant:
43054 (gdb) target extended-remote the-target:2345
43057 The @command{gdbserver} option @option{--multi} may or may not be used in such
43061 @c man begin OPTIONS gdbserver
43062 There are three different modes for invoking @command{gdbserver}:
43067 Debug a specific program specified by its program name:
43070 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
43073 The @var{comm} parameter specifies how should the server communicate
43074 with @value{GDBN}; it is either a device name (to use a serial line),
43075 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
43076 stdin/stdout of @code{gdbserver}. Specify the name of the program to
43077 debug in @var{prog}. Any remaining arguments will be passed to the
43078 program verbatim. When the program exits, @value{GDBN} will close the
43079 connection, and @code{gdbserver} will exit.
43082 Debug a specific program by specifying the process ID of a running
43086 gdbserver --attach @var{comm} @var{pid}
43089 The @var{comm} parameter is as described above. Supply the process ID
43090 of a running program in @var{pid}; @value{GDBN} will do everything
43091 else. Like with the previous mode, when the process @var{pid} exits,
43092 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
43095 Multi-process mode -- debug more than one program/process:
43098 gdbserver --multi @var{comm}
43101 In this mode, @value{GDBN} can instruct @command{gdbserver} which
43102 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
43103 close the connection when a process being debugged exits, so you can
43104 debug several processes in the same session.
43107 In each of the modes you may specify these options:
43112 List all options, with brief explanations.
43115 This option causes @command{gdbserver} to print its version number and exit.
43118 @command{gdbserver} will attach to a running program. The syntax is:
43121 target> gdbserver --attach @var{comm} @var{pid}
43124 @var{pid} is the process ID of a currently running process. It isn't
43125 necessary to point @command{gdbserver} at a binary for the running process.
43128 To start @code{gdbserver} without supplying an initial command to run
43129 or process ID to attach, use this command line option.
43130 Then you can connect using @kbd{target extended-remote} and start
43131 the program you want to debug. The syntax is:
43134 target> gdbserver --multi @var{comm}
43138 Instruct @code{gdbserver} to display extra status information about the debugging
43140 This option is intended for @code{gdbserver} development and for bug reports to
43143 @item --remote-debug
43144 Instruct @code{gdbserver} to display remote protocol debug output.
43145 This option is intended for @code{gdbserver} development and for bug reports to
43149 Specify a wrapper to launch programs
43150 for debugging. The option should be followed by the name of the
43151 wrapper, then any command-line arguments to pass to the wrapper, then
43152 @kbd{--} indicating the end of the wrapper arguments.
43155 By default, @command{gdbserver} keeps the listening TCP port open, so that
43156 additional connections are possible. However, if you start @code{gdbserver}
43157 with the @option{--once} option, it will stop listening for any further
43158 connection attempts after connecting to the first @value{GDBN} session.
43160 @c --disable-packet is not documented for users.
43162 @c --disable-randomization and --no-disable-randomization are superseded by
43163 @c QDisableRandomization.
43168 @c man begin SEEALSO gdbserver
43170 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
43171 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
43172 documentation are properly installed at your site, the command
43178 should give you access to the complete manual.
43180 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
43181 Richard M. Stallman and Roland H. Pesch, July 1991.
43188 @c man title gcore Generate a core file of a running program
43191 @c man begin SYNOPSIS gcore
43192 gcore [-o @var{filename}] @var{pid}
43196 @c man begin DESCRIPTION gcore
43197 Generate a core dump of a running program with process ID @var{pid}.
43198 Produced file is equivalent to a kernel produced core file as if the process
43199 crashed (and if @kbd{ulimit -c} were used to set up an appropriate core dump
43200 limit). Unlike after a crash, after @command{gcore} the program remains
43201 running without any change.
43204 @c man begin OPTIONS gcore
43206 @item -o @var{filename}
43207 The optional argument
43208 @var{filename} specifies the file name where to put the core dump.
43209 If not specified, the file name defaults to @file{core.@var{pid}},
43210 where @var{pid} is the running program process ID.
43214 @c man begin SEEALSO gcore
43216 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
43217 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
43218 documentation are properly installed at your site, the command
43225 should give you access to the complete manual.
43227 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
43228 Richard M. Stallman and Roland H. Pesch, July 1991.
43235 @c man title gdbinit GDB initialization scripts
43238 @c man begin SYNOPSIS gdbinit
43239 @ifset SYSTEM_GDBINIT
43240 @value{SYSTEM_GDBINIT}
43249 @c man begin DESCRIPTION gdbinit
43250 These files contain @value{GDBN} commands to automatically execute during
43251 @value{GDBN} startup. The lines of contents are canned sequences of commands,
43254 the @value{GDBN} manual in node @code{Sequences}
43255 -- shell command @code{info -f gdb -n Sequences}.
43261 Please read more in
43263 the @value{GDBN} manual in node @code{Startup}
43264 -- shell command @code{info -f gdb -n Startup}.
43271 @ifset SYSTEM_GDBINIT
43272 @item @value{SYSTEM_GDBINIT}
43274 @ifclear SYSTEM_GDBINIT
43275 @item (not enabled with @code{--with-system-gdbinit} during compilation)
43277 System-wide initialization file. It is executed unless user specified
43278 @value{GDBN} option @code{-nx} or @code{-n}.
43281 the @value{GDBN} manual in node @code{System-wide configuration}
43282 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
43285 @ref{System-wide configuration}.
43289 User initialization file. It is executed unless user specified
43290 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
43293 Initialization file for current directory. It may need to be enabled with
43294 @value{GDBN} security command @code{set auto-load local-gdbinit}.
43297 the @value{GDBN} manual in node @code{Init File in the Current Directory}
43298 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
43301 @ref{Init File in the Current Directory}.
43306 @c man begin SEEALSO gdbinit
43308 gdb(1), @code{info -f gdb -n Startup}
43310 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
43311 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
43312 documentation are properly installed at your site, the command
43318 should give you access to the complete manual.
43320 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
43321 Richard M. Stallman and Roland H. Pesch, July 1991.
43327 @node GNU Free Documentation License
43328 @appendix GNU Free Documentation License
43331 @node Concept Index
43332 @unnumbered Concept Index
43336 @node Command and Variable Index
43337 @unnumbered Command, Variable, and Function Index
43342 % I think something like @@colophon should be in texinfo. In the
43344 \long\def\colophon{\hbox to0pt{}\vfill
43345 \centerline{The body of this manual is set in}
43346 \centerline{\fontname\tenrm,}
43347 \centerline{with headings in {\bf\fontname\tenbf}}
43348 \centerline{and examples in {\tt\fontname\tentt}.}
43349 \centerline{{\it\fontname\tenit\/},}
43350 \centerline{{\bf\fontname\tenbf}, and}
43351 \centerline{{\sl\fontname\tensl\/}}
43352 \centerline{are used for emphasis.}\vfill}
43354 % Blame: doc@@cygnus.com, 1991.